key: cord-014533-6qfecv5h
authors: Velasquez, T.; Mackey, G.; Lusk, J.; Kyle, U. G.; Fontenot, T.; Marshall, P.; Shekerdemian, L. S.; Coss-Bu, J. A.; Nishigaki, A.; Yatabe, T.; Tamura, T.; Yamashita, K.; Yokoyama, M.; Ruiz-Rodriguez, J. C.; Encina, B.; Belmonte, R.; Troncoso, I.; Tormos, P.; Riveiro, M.; Baena, J.; Sanchez, A.; Bañeras, J.; Cordón, J.; Duran, N.; Ruiz, A.; Caballero, J.; Nuvials, X.; Riera, J.; Serra, J.; Rutten, A. M. F.; van Ieperen, S. N. M.; Der Kinderen, E. P. H. M.; Van Logten, T.; Kovacikova, L.; Skrak, P.; Zahorec, M.; Kyle, U. G.; Akcan-Arikan, A.; Silva, J. C.; Mackey, G.; Lusk, J.; Goldsworthy, M.; Shekerdemian, L. S.; Coss-Bu, J. A.; Wood, D.; Harrison, D.; Parslow, R.; Davis, P.; Pappachan, J.; Goodwin, S.; Ramnarayan, P.; Chernyshuk, S.; Yemets, H.; Zhovnir, V.; Pulitano’, S. M.; De Rosa, S.; Mancino, A.; Villa, G.; Tosi, F.; Franchi, P.; Conti, G.; Patel, B.; Khine, H.; Shah, A.; Sung, D.; Singer, L.; Haghbin, S.; Inaloo, S.; Serati, Z.; Idei, M.; Nomura, T.; Yamamoto, N.; Sakai, Y.; Yoshida, T.; Matsuda, Y.; Yamaguchi, Y.; Takaki, S.; Yamaguchi, O.; Goto, T.; Longani, N.; Medar, S.; Abdel-Aal, I. R.; El Adawy, A. S.; Mohammed, H. M. E. H.; Mohamed, A. N.; Parry, S. M.; Knight, L. D.; Denehy, L.; De Morton, N.; Baldwin, C. E.; Sani, D.; Kayambu, G.; da Silva, V. Z. M.; Phongpagdi, P.; Puthucheary, Z. A.; Granger, C. L.; Rydingsward, J. E.; Horkan, C. M.; Christopher, K. B.; McWilliams, D.; Jones, C.; Reeves, E.; Atkins, G.; Snelson, C.; Aitken, L. M.; Rattray, J.; Kenardy, J.; Hull, A. M.; Ullman, A.; Le Brocque, R.; Mitchell, M.; Davis, C.; Macfarlane, B.; Azevedo, J. C.; Rocha, L. L.; De Freitas, F. F. M.; Cavalheiro, A. M.; Lucinio, N. M.; Lobato, M. S.; Ebeling, G.; Kraegpoeth, A.; Laerkner, E.; De Brito-Ashurst, I.; White, C.; Gregory, S.; Forni, L. G.; Flowers, E.; Curtis, A.; Wood, C. A.; Siu, K.; Venkatesan, K.; Muhammad, J. B. H.; Ng, L.; Seet, E.; Baptista, N.; Escoval, A.; Tomas, E.; Agrawal, R.; Mathew, R.; Varma, A.; Dima, E.; Charitidou, E.; Perivolioti, E.; Pratikaki, M.; Vrettou, C.; Giannopoulos, A.; Zakynthinos, S.; Routsi, C.; Atchade, E.; Houzé, S.; Jean-Baptiste, S.; Thabut, G.; Genève, C.; Tanaka, S.; Lortat-Jacob, B.; Augustin, P.; Desmard, M.; Montravers, P.; de Molina, F. J. González; Barbadillo, S.; Alejandro, R.; Álvarez-Lerma, F.; Vallés, J.; Catalán, R. M.; Palencia, E.; Jareño, A.; Granada, R. M.; Ignacio, M. L.; Cui, N.; Liu, D.; Wang, H.; Su, L.; Qiu, H.; Li, R.; Jaffal, K.; Rouzé, A.; Poissy, J.; Sendid, B.; Nseir, S.; Paramythiotou, E.; Rizos, M.; Frantzeskaki, F.; Antoniadou, A.; Vourli, S.; Zerva, L.; Armaganidis, A.; Riera, J.; Gottlieb, J.; Greer, M.; Wiesner, O.; Martínez, M.; Acuña, M.; Rello, J.; Welte, T.; Atchade, E.; Mignot, T.; Houzé, S.; Jean-Baptiste, S.; Thabut, G.; Lortat-Jacob, B.; Tanaka, S.; Augustin, P.; Desmard, M.; Montravers, P.; Soussi, S.; Dudoignon, E.; Ferry, A.; Chaussard, M.; Benyamina, M.; Alanio, A.; Touratier, S.; Chaouat, M.; Lafaurie, M.; Mimoun, M.; Mebazaa, A.; Legrand, M.; Sheils, M. A.; Patel, C.; Mohankumar, L.; Akhtar, N.; Noriega, S. K. Pacheco; Aldana, N. Navarrete; León, J. L. Ávila; Baquero, J. Durand; Bernal, F. Fernández; Ahmadnia, E.; Hadley, J. S.; Millar, M.; Hall, D.; Hewitt, H.; Yasuda, H.; Sanui, M.; Komuro, T.; Kawano, S.; Andoh, K.; Yamamoto, H.; Noda, E.; Hatakeyama, J.; Saitou, N.; Okamoto, H.; Kobayashi, A.; Takei, T.; Matsukubo, S.; Rotzel, H. B.; Lázaro, A. Serrano; Prada, D. Aguillón; Gimillo, M. Rodriguez; Barinas, O. Diaz; Cortes, M. L. Blasco; Franco, J. Ferreres; Roca, J. M. Segura; Carratalá, A.; Gonçalves, B.; Turon, R.; Mendes, A.; Miranda, F.; Mata, P. J.; Cavalcanti, D.; Melo, N.; Lacerda, P.; Kurtz, P.; Righy, C.; Rosario, L. E. de la Cruz; Lesmes, S. P. Gómez; Romero, J. C. García; Herrera, A. N. García; Pertuz, E. D. Díaz; Sánchez, M. J. Gómez; Sanz, E. Regidor; Hualde, J. Barado; Hernández, A. Ansotegui; Irazabal, J. M. Guergué; Spatenkova, V.; Bradac, O.; Suchomel, P.; Urli, T.; Lazzeri, E. Heusch; Aspide, R.; Zanello, M.; Perez-Borrero, L.; Garcia-Alvarez, J. M.; Arias-Verdu, M. D.; Aguilar-Alonso, E.; Rivera-Fernandez, R.; Mora-Ordoñez, J.; De La Fuente-Martos, C.; Castillo-Lorente, E.; Guerrero-Lopez, F.; Lesmes, S. P. Gómez; Rosario, L. E. De la Cruz; Pertuz, E. D. Díaz; Hernández, A. Ansotegui; Romero, J. C. García; Sánchez, M. J. Gómez; Herrera, A. N. García; Ramírez, J. Roldán; Sanz, E. Regidor; Hualde, J. Barado; León, J. P. Tirapu; Navarro-Guillamón, L.; Cordovilla-Guardia, S.; Iglesias-Santiago, A.; Guerrero-López, F.; Fernández-Mondéjar, E.; Vidal, A.; Perez, M.; Juez, A.; Arias, N.; Colino, L.; Perez, J. L.; Pérez, H.; Calpe, P.; Alcala, M. A.; Robaglia, D.; Perez, C.; Lan, S. K.; Cunha, M. M.; Moreira, T.; Santos, F.; Lafuente, E.; Fernandes, M. J.; Silva, J. G.; Rosario, L. E. de la Cruz; Lesmes, S. P. Gómez; Herrera, A. N. García; Romero, J. C. García; Pertuz, E. D. Díaz; Sánchez, M. J. Gómez; Sanz, E. Regidor; Echeverría, J. G. Armando; Hernández, A. Ansotegui; Hualde, J. Barado; Podlepich, V.; Sokolova, E.; Alexandrova, E.; Lapteva, K.; Kurtz, P.; Shuinotsuka, C.; Rabello, L.; Vianna, G.; Reis, A.; Cairus, C.; Salluh, J.; Bozza, F.; Torres, J. C. Barrios; Araujo, N. J. Fernández; García-Olivares, P.; Keough, E.; Dalorzo, M.; Tang, L. K.; De Sousa, I.; Díaz, M.; Marcos-Zambrano, L. J.; Guerrero, J. E.; Gomez, S. E. Zamora; Lopez, G. D. Hernandez; Cuellar, A. I. Vazquez; Nieto, O. R. Perez; Gonzalez, J. A. Castanon; Bhasin, D.; Rai, S.; Singh, H.; Gupta, O.; Bhattal, M. K.; Sampley, S.; Sekhri, K.; Nandha, R.; Aliaga, F. A.; Olivares, F.; Appiani, F.; Farias, P.; Alberto, F.; Hernández, A.; Pons, S.; Sonneville, R.; Bouadma, L.; Neuville, M.; Mariotte, E.; Radjou, A.; Lebut, J.; Chemam, S.; Voiriot, G.; Dilly, M. P.; Mourvillier, B.; Dorent, R.; Nataf, P.; Wolff, M.; Timsit, J. F.; Ediboglu, O.; Ataman, S.; Ozkarakas, H.; Kirakli, C.; Vakalos, A.; Avramidis, V.; Obukhova, O.; Kurmukov, I. A.; Kashiya, S.; Golovnya, E.; Baikova, V. N.; Ageeva, T.; Haritydi, T.; Kulaga, E. V.; Rios-Toro, J. J.; Perez-Borrero, L.; Aguilar-Alonso, E.; Arias-Verdu, M. D.; Garcia-Alvarez, J. M.; Lopez-Caler, C.; De La Fuente-Martos, C.; Rodriguez-Fernandez, S.; Sanchez-Orézzoli, M. Gomez; Martin-Gallardo, F.; Nikhilesh, J.; Joshi, V.; Villarreal, E.; Ruiz, J.; Gordon, M.; Quinza, A.; Gimenez, J.; Piñol, M.; Castellanos, A.; Ramirez, P.; Jeon, Y. D.; Jeong, W. Y.; Kim, M. H.; Jeong, I. Y.; Ahn, M. Y.; Ahn, J. Y.; Han, S. H.; Choi, J. Y.; Song, Y. G.; Kim, J. M.; Ku, N. S.; Shah, H.; Kellner, F.; Rezai, F.; Mistry, N.; Yodice, P.; Ovnanian, V.; Fless, K.; Handler, E.; Alejos, R. Martínez; Romeu, J. D. Martí; Antón, D. González; Quinart, A.; Martí, A. Torres; Llaurado-Serra, M.; Lobo-Civico, A.; Ventura-Rosado, A.; Piñol-Tena, A.; Pi-Guerrero, M.; Paños-Espinosa, C.; Peralvo-Bernat, M.; Marine-Vidal, J.; Gonzalez-Engroba, R.; Montesinos-Cerro, N.; Treso-Geira, M.; Valeiras-Valero, A.; Martinez-Reyes, L.; Sandiumenge, A.; Jimenez-Herrera, M. F.; Helyar, S.; Riozzi, P.; Noon, A.; Hallows, G.; Cotton, H.; Keep, J.; Hopkins, P. A.; Taggu, A.; Renuka, S.; Sampath, S.; Rood, P. J. T.; Frenzel, T.; Verhage, R.; Bonn, M.; Pickkers, P.; van der Hoeven, J. G.; van den Boogaard, M.; Corradi, F.; Melnyk, L.; Moggia, F.; Pienovi, R.; Adriano, G.; Brusasco, C.; Mariotti, L.; Lattuada, M.; Bloomer, M. J.; Coombs, M.; Ranse, K.; Endacott, R.; Maertens, B.; Blot, K.; Blot, S.; Amerongen, M. P. van Nieuw; van der Heiden, E. S.; Twisk, J. W. R.; Girbes, A. R. J.; Spijkstra, J. J.; Riozzi, P.; Helyar, S.; Cotton, H.; Hallows, G.; Noon, A.; Bell, C.; Peters, K.; Feehan, A.; Keep, J.; Hopkins, P. A.; Churchill, K.; Hawkins, K.; Brook, R.; Paver, N.; Endacott, R.; Maistry, N.; van Wijk, A.; Rouw, N.; van Galen, T.; Evelein-Brugman, S.; Taggu, A.; Krishna, B.; Sampath, S.; Putzu, A.; Fang, M.; Berto, M. Boscolo; Belletti, A.; Cassina, T.; Cabrini, L.; Mistry, M.; Alhamdi, Y.; Welters, I.; Abrams, S. T.; Toh, C. H.; Han, H. S.; Gil, E. M.; Lee, D. S.; Park, C. M.; Winder-Rhodes, S.; Lotay, R.; Doyle, J.; Ke, M. W.; Huang, W. C.; Chiang, C. H.; Hung, W. T.; Cheng, C. C.; Lin, K. C.; Lin, S. C.; Chiou, K. R.; Wann, S. R.; Shu, C. W.; Kang, P. L.; Mar, G. Y.; Liu, C. P.; Dubó, S.; Aquevedo, A.; Jibaja, M.; Berrutti, D.; Labra, C.; Lagos, R.; García, M. F.; Ramirez, V.; Tobar, M.; Picoita, F.; Peláez, C.; Carpio, D.; Alegría, L.; Hidalgo, C.; Godoy, K.; Bakker, J.; Hernández, G.; Sadamoto, Y.; Katabami, K.; Wada, T.; Ono, Y.; Maekawa, K.; Hayakawa, M.; Sawamura, A.; Gando, S.; Marin-Mateos, H.; Perez-Vela, J. L.; Garcia-Gigorro, R.; Peiretti, M. A. Corres; Lopez-Gude, M. J.; Chacon-Alves, S.; Renes-Carreño, E.; Montejo-González, J. C.; Parlevliet, K. L.; Touw, H. R. W.; Beerepoot, M.; Boer, C.; Elbers, P. W. G.; Tuinman, P. R.; Abdelmonem, S. A.; Helmy, T. A.; El Sayed, I.; Ghazal, S.; Akhlagh, S. H.; Masjedi, M.; Hozhabri, K.; Kamali, E.; Zýková, I.; Paldusová, B.; Sedlák, P.; Morman, D.; Youn, A. M.; Ohta, Y.; Sakuma, M.; Bates, D.; Morimoto, T.; Su, P. L.; Chang, W. Y.; Lin, W. C.; Chen, C. W.; Facchin, F.; Zarantonello, F.; Panciera, G.; De Cassai, A.; Venrdramin, A.; Ballin, A.; Tonetti, T.; Persona, P.; Ori, C.; Del Sorbo, L.; Rossi, S.; Vergani, G.; Cressoni, M.; Chiumello, D.; Chiurazzi, C.; Brioni, M.; Algieri, I.; Tonetti, T.; Guanziroli, M.; Colombo, A.; Tomic, I.; Colombo, A.; Crimella, F.; Carlesso, E.; Gasparovic, V.; Gattinoni, L.; Neto, A. Serpa; Schmidt, M.; Pham, T.; Combes, A.; de Abreu, M. Gama; Pelosi, P.; Schultz, M. J.; Katira, B. H.; Engelberts, D.; Giesinger, R. E.; Ackerley, C.; Yoshida, T.; Zabini, D.; Otulakowski, G.; Post, M.; Kuebler, W. M.; McNamara, P. J.; Kavanagh, B. P.; Pirracchio, R.; Rigon, M. Resche; Carone, M.; Chevret, S.; Annane, D.; Eladawy, S.; El-Hamamsy, M.; Bazan, N.; Elgendy, M.; De Pascale, G.; Vallecoccia, M. S.; Cutuli, S. L.; Di Gravio, V.; Pennisi, M. A.; Conti, G.; Antonelli, M.; Andreis, D. T.; Khaliq, W.; Singer, M.; Hartmann, J.; Harm, S.; Carmona, S. Alcantara; Almudevar, P. Matia; Abellán, A. Naharro; Ramos, J. Veganzones; Pérez, L. Pérez; Valbuena, B. Lobo; Sanz, N. Martínez; Simón, I. Fernández; Arrigo, M.; Feliot, E.; Deye, N.; Cariou, A.; Guidet, B.; Jaber, S.; Leone, M.; Resche-Rigon, M.; Baron, A. Vieillard; Legrand, M.; Gayat, E.; Mebazaa, A.; Balik, M.; Kolnikova, I.; Maly, M.; Waldauf, P.; Tavazzi, G.; Kristof, J.; Herpain, A.; Su, F.; Post, E.; Taccone, F.; Vincent, J. L.; Creteur, J.; Lee, C.; Hatib, F.; Jian, Z.; Buddi, S.; Cannesson, M.; Fileković, S.; Turel, M.; Knafelj, R.; Gorjup, V.; Stanić, R.; Gradišek, P.; Cerović, O.; Mirković, T.; Noč, M.; Tirkkonen, J.; Hellevuo, H.; Olkkola, K. T.; Hoppu, S.; Lin, K. C.; Hung, W. T.; Chiang, C. C.; Huang, W. C.; Juan, W. C.; Lin, S. C.; Cheng, C. C.; Lin, P. H.; Fong, K. Y.; Hou, D. S.; Kang, P. L.; Wann, S. R.; Chen, Y. S.; Mar, G. Y.; Liu, C. P.; Paul, M.; Bougouin, W.; Geri, G.; Dumas, F.; Champigneulle, B.; Legriel, S.; Charpentier, J.; Mira, J. P.; Sandroni, C.; Cariou, A.; Zimmerman, J.; Sullivan, E.; Noursadeghi, M.; Fox, B.; Sampson, D.; McHugh, L.; Yager, T.; Cermelli, S.; Seldon, T.; Bhide, S.; Brandon, R. A.; Brandon, R. B.; Zwaag, J.; Beunders, R.; Pickkers, P.; Kox, M.; Gul, F.; Arslantas, M. K.; Genc, D.; Zibandah, N.; Topcu, L.; Akkoc, T.; Cinel, I.; Greco, E.; Lauretta, M. P.; Andreis, D. T.; Singer, M.; Garcia, I. Palacios; Cordero, M.; Martin, A. Diaz; Pallás, T. Aldabó; Montero, J. Garnacho; Rey, J. Revuelto; Malo, L. Roman; Montoya, A. A. Tanaka; Martinez, A. D. C. Amador; Ayala, L. Y. Delgado; Zepeda, E. Monares; Granillo, J. Franco; Sanchez, J. Aguirre; Alejo, G. Camarena; Cabrera, A. Rugerio; Montenegro, A. Pedraza; Pham, T.; Beduneau, G.; Schortgen, F.; Piquilloud, L.; Zogheib, E.; Jonas, M.; Grelon, F.; Runge, I.; Terzi, N.; Grangé, S.; Barberet, G.; Guitard, P. G.; Frat, J. P.; Constan, A.; Chrétien, J. M.; Mancebo, J.; Mercat, A.; Richard, J. C. M.; Brochard, L.; Soilemezi, E.; Koco, E.; Savvidou, S.; Nouris, C.; Matamis, D.; Di Mussi, R.; Spadaro, S.; Volta, C. A.; Mariani, M.; Colaprico, A.; Antonio, C.; Bruno, F.; Grasso, S.; Rodriguez, A.; Martín-Loeches, I.; Díaz, E.; Masclans, J. R.; Gordo, F.; Solé-Violán, J.; Bodí, M.; Avilés-Jurado, F. X.; Trefler, S.; Magret, M.; Reyes, L. F.; Marín-Corral, J.; Yebenes, J. C.; Esteban, A.; Anzueto, A.; Aliberti, S.; Restrepo, M. I.; Larsson, J. Skytte; Redfors, B.; Ricksten, S. E.; Haines, R.; Powell-Tuck, J.; Leonard, H.; Ostermann, M.; Berthelsen, R. E.; Itenov, T. S.; Perner, A.; Jensen, J. U.; Ibsen, M.; Jensen, A. E. K.; Bestle, M. H.; Bucknall, T.; Dixon, J.; Boa, F.; MacPhee, I.; Philips, B. J.; Doyle, J.; Saadat, F.; Samuels, T.; Huddart, S.; McCormick, B.; DeBrunnar, R.; Preece, J.; Swart, M.; Peden, C.; Richardson, S.; Forni, L.; Kalfon, P.; Baumstarck, K.; Estagnasie, P.; Geantot, M. A.; Berric, A.; Simon, G.; Floccard, B.; Signouret, T.; Boucekine, M.; Fromentin, M.; Nyunga, M.; Sossou, A.; Venot, M.; Robert, R.; Follin, A.; Renault, A.; Garrouste, M.; Collange, O.; Levrat, Q.; Villard, I.; Thévenin, D.; Pottecher, J.; Patrigeon, R. G.; Revel, N.; Vigne, C.; Mimoz, O.; Auquier, P.; Pawar, S.; Jacques, T.; Deshpande, K.; Pusapati, R.; Wood, B.; Pulham, R. A.; Wray, J.; Brown, K.; Pierce, C.; Nadel, S.; Ramnarayan, P.; Azevedo, J. R.; Montenegro, W. S.; Rodrigues, D. P.; Sousa, S. C.; Araujo, V. F.; Leitao, A. L.; Prazeres, P. H.; Mendonca, A. V.; Paula, M. P.; Das Neves, A.; Loudet, C. I.; Busico, M.; Vazquez, D.; Villalba, D.; Lischinsky, A.; Veronesi, M.; Emmerich, M.; Descotte, E.; Juliarena, A.; Bisso, M. Carboni; Grando, M.; Tapia, A.; Camargo, M.; Ulla, D. Villani; Corzo, L.; dos Santos, H. Placido; Ramos, A.; Doglia, J. A.; Estenssoro, E.; Carbonara, M.; Magnoni, S.; Donald, C. L. Mac; Shimony, J. S.; Conte, V.; Triulzi, F.; Stretti, F.; Macrì, M.; Snyder, A. Z.; Stocchetti, N.; Brody, D. L.; Podlepich, V.; Shimanskiy, V.; Savin, I.; Lapteva, K.; Chumaev, A.; Tjepkema-Cloostermans, M. C.; Hofmeijer, J.; Beishuizen, A.; Hom, H.; Blans, M. J.; van Putten, M. J. A. M.; Longhi, L.; Frigeni, B.; Curinga, M.; Mingone, D.; Beretta, S.; Patruno, A.; Gandini, L.; Vargiolu, A.; Ferri, F.; Ceriani, R.; Rottoli, M. R.; Lorini, L.; Citerio, G.; Pifferi, S.; Battistini, M.; Cordolcini, V.; Agarossi, A.; Di Rosso, R.; Ortolano, F.; Stocchetti, N.; Lourido, C. Mora; Cabrera, J. L. Santana; Santana, J. D. Martín; Alzola, L. Melián; del Rosario, C. García; Pérez, H. Rodríguez; Torrent, R. Lorenzo; Eslami, S.; Dalhuisen, A.; Fiks, T.; Schultz, M. J.; Hanna, A. Abu; Spronk, P. E.; Wood, M.; Maslove, D.; Muscedere, J.; Scott, S. H.; Saha, T.; Hamilton, A.; Petsikas, D.; Payne, D.; Boyd, J. G.; Puthucheary, Z. A.; McNelly, A. S.; Rawal, J.; Connolly, B.; McPhail, M. J.; Sidhu, P.; Rowlerson, A.; Moxham, J.; Harridge, S. D.; Hart, N.; Montgomery, H. E.; Jovaisa, T.; Thomas, B.; Gupta, D.; Wijayatilake, D. S.; Shum, H. P.; King, H. S.; Chan, K. C.; Tang, K. B.; Yan, W. W.; Arias, C. Castro; Latorre, J.; De La Rica, A. Suárez; Garrido, E. Maseda; Feijoo, A. Montero; Gancedo, C. Hernández; Tofiño, A. López; Rodríguez, F. Gilsanz; Gemmell, L. K.; Campbell, R.; Doherty, P.; MacKay, A.; Singh, N.; Vitaller, S.; Nagib, H.; Prieto, J.; Del Arco, A.; Zayas, B.; Gomez, C.; Tirumala, S.; Pasha, S. A.; Kumari, B. K.; Martinez-Lopez, P.; Puerto-Morlán, A.; Nuevo-Ortega, P.; Pujol, L. Martinez; Dolset, R. Algarte; González, B. Sánchez; Riera, S. Quintana; Álvarez, J. Trenado; Quintana, S.; Martínez, L.; Algarte, R.; Sánchez, B.; Trenado, J.; Tomas, E.; Brock, N.; Viegas, E.; Filipe, E.; Cottle, D.; Traynor, T.; Martínez, M. V. Trasmonte; Márquez, M. Pérez; Gómez, L. Colino; Martínez, N. Arias; Muñoz, J. M. Milicua; Bellver, B. Quesada; Varea, M. Muñoz; Llorente, M. Á. Alcalá; Calvo, C. Pérez; Hillier, S. D.; Faulds, M. C.; Hendra, H.; Lawrence, N.; Maekawa, K.; Hayakawa, M.; Ono, Y.; Kodate, A.; Sadamoto, Y.; Tominaga, N.; Mizugaki, A.; Murakami, H.; Yoshida, T.; Katabami, K.; Wada, T.; Sawamura, A.; Gando, S.; Silva, S.; Kerhuel, L.; Malagurski, B.; Citerio, G.; Chabanne, R.; Laureys, S.; Puybasset, L.; Nobile, L.; Pognuz, E. R.; Rossetti, A. O.; Verginella, F.; Gaspard, N.; Creteur, J.; Ben-Hamouda, N.; Oddo, M.; Taccone, F. S.; Ono, Y.; Hayakawa, M.; Iijima, H.; Maekawa, K.; Kodate, A.; Sadamoto, Y.; Mizugaki, A.; Murakami, H.; Katabami, K.; Wada, T.; Sawamura, A.; Gando, S.; Kodate, A.; Katabami, K.; Wada, T.; Ono, Y.; Maekawa, K.; Hayakawa, M.; Sawamura, A.; Gando, S.; Andersen, L. W.; Raymond, T.; Berg, R.; Nadkarni, V.; Grossestreuer, A.; Kurth, T.; Donnino, M.; Krüger, A.; Ostadal, P.; Janotka, M.; Vondrakova, D.; Kongpolprom, N.; Cholkraisuwat, J.; Pekkarinen, P. T.; Ristagno, G.; Masson, S.; Latini, R.; Bendel, S.; Ala-Kokko, T.; Varpula, T.; Vaahersalo, J.; Hoppu, S.; Tiainen, M.; Mion, M. M.; Plebani, M.; Pettilä, V.; Skrifvars, M.B.; Son, Y.; Kim, K. S.; Suh, G. J.; Kwon, W. Y.; Ko, J. I.; Park, M. J.; Cavicchi, F. Zama; Iesu, E.; Nobile, L.; Vincent, J. L.; Creteur, J.; Taccone, F. S.; Tanaka, H.; Otani, N.; Ode, S.; Ishimatsu, S.; Martínez, L.; Algarte, R.; Sánchez, B.; Romero, I.; Martínez, F.; Quintana, S.; Trenado, J.; Vondrakova, D.; Ostadal, P.; Kruger, A.; Janotka, M.; Malek, F.; Neuzil, P.; Yeh, Y. C.; Chen, Y. S.; Wang, C. H.; Huang, C. H.; Chao, A.; Lee, C. T.; Lai, C. H.; Chan, W. S.; Cheng, Y. J.; Sun, W. Z.; Kaese, S.; Horstmann, C.; Lebiedz, P.; Mourad, M.; Gaudard, P.; Eliet, J.; Zeroual, N.; Colson, P.; Ostadal, P.; Mlcek, M.; Hrachovina, M.; Kruger, A.; Vondrakova, D.; Janotka, M.; Mates, M.; Hala, P.; Kittnar, O.; Neuzil, P.; Jacky, A.; Rudiger, A.; Spahn, D. R.; Bettex, D. A.; Kara, A.; Akin, S.; Dos reis Miranda, D.; Struijs, A.; Caliskan, K.; van Thiel, R. J.; Dubois, E. A.; de Wilde, W.; Zijlstra, F.; Gommers, D.; Ince, C.; Marca, L.; Xini, A.; Mongkolpun, W.; Cordeiro, C. P. R.; Leite, R. T.; Lheureux, O.; Bader, A.; Rincon, L.; Santacruz, C.; Preiser, J. C.; Chao, A.; Chao, A. S.; Chen, Y. S.; Kim, W.; Ahn, C.; Cho, Y.; Lim, T. H.; Oh, J.; Choi, K. S.; Jang, B. H.; Ha, J. K.; Mecklenburg, A.; Stamm, J.; Soeffker, G.; Kubik, M.; Sydow, K.; Reichenspurner, H.; Kluge, S.; Braune, S.; Bergantino, B.; Ruberto, F.; Magnanimi, E.; Privato, E.; Zullino, V.; Bruno, K.; Pugliese, F.; Sales, G.; Girotto, V.; Vittone, F.; Brazzi, L.; Fritz, C.; Kimmoun, A.; Vanhuyse, F.; Trifan, B.; Orlowski, S.; Albuisson, E.; Tran, N.; Levy, B.; Chhor, V.; Joachim, J.; Follin, A.; Champigneulle, B.; Chatelon, J.; Fave, G.; Mantz, J.; Pirracchio, R.; Diaz, D. Díaz; Villanova, M.; Aguirregabyria, M.; Andrade, G.; López, L.; Palencia, E.; John, G.; Cowan, R.; Hart, R.; Lake, K.; Litchfield, K.; Song, J. W.; Lee, Y. J.; Cho, Y. J.; Choi, S.; Vermeir, P.; Vandijck, D.; Blot, S.; Mariman, A.; Verhaeghe, R.; Deveugele, M.; Vogelaers, D.; Chok, L.; Bachli, E. B.; Bettex, D.; Cottini, S. R.; Keller, E.; Maggiorini, M.; Schuepbach, R.; Fiks, T.; Stiphout, C.; Grevelink, M.; Vaneker, I.; Ruijter, A.; Buise, M.; Spronk, P. E.; Tena, S. Altaba; Barrachina, L. Galarza; Portillo, J. H. Rodriguez; Aznar, G. Pagés; Campos, L. Mateu; Sellés, M. D. Ferrándiz; Tomás, M. Arlandis; Muncharaz, A. Belenguer; Skinner, L.; Monsalvo, S.; Olavarria, E.; Stümpfle, R.; Na, S. J.; Park, J.; Chung, C. R.; Park, C. M.; Suh, G. Y.; Yang, J. H.; Witter, T.; Brousseau, C.; Butler, M. B.; Erdogan, M.; Dougall, P. C. Mac; Green, R. S.; Abbott, T. E. F.; Torrance, H. D. T.; Cron, N.; Vaid, N.; Emmanuel, J.; Siddiqui, S. S.; Prabu, N.; Chaudhari, H. K.; Patil, V. P.; Divatia, J. V.; Solanki, S.; Kulkarni, A. P.; Gutierrez, L. A. Rincon; Bader, A.; Brasseur, A.; Lheureux, O.; Vincent, J. L.; Creteur, J.; Taccone, F. S.; Hempel, D.; Stauffert, N.; Recker, F.; Schröder, T.; Reusch, S.; Schleifer, J.; Breitkreutz, R.; Sjövall, F.; Perner, A.; Møller, M. Hylander; Moraes, R. B.; Borges, F. K.; Guillen, J. A. V.; Zabaletta, W. J. C.; Ruiz-Ramos, J.; Ramirez, P.; Marqués-Miñana, M. R.; Villarreal, E.; Gordon, M.; Sosa, M.; Concha, P.; Castellanos, A.; Menendez, R.; Ramírez, C. Sánchez; Santana, M. Cabrera; Balcázar, L. Caipe; Escalada, S. Hípola; Viera, M. A. Hernández; Vázquez, C. F. Lübbe; Díaz, J. J. Díaz; Campelo, F. Artiles; Monroy, N. Sangil; Santana, P. Saavedra; Santana, S. Ruiz; Gutiérrez-Pizarraya, A.; Garnacho-Montero, J.; Martin, C.; Baumstarck, K.; Leone, M.; Martín-Loeches, I.; Pirracchio, R.; Legrand, M.; Mainardi, J. L.; Mantz, J.; Cholley, B.; Hubbard, A.; Frontera, P. Ruiz; Vega, L. M. Claraco; Miguelena, P. Ruiz de Gopegui; Usón, M. C. Villuendas; López, A. Rezusta; Clemente, E. Aurensanz; Ibañes, P. Gutiérrez; Aguilar, A. L. Ruiz; Palomar, M.; Olaechea, P.; Uriona, S.; Vallverdu, M.; Catalan, M.; Nuvials, X.; Aragon, C.; Lerma, F. Alvarez; Jeon, Y. D.; Jeong, W. Y.; Kim, M. H.; Jeong, I. Y.; Ahn, M. Y.; Ahn, J. Y.; Han, S. H.; Choi, J. Y.; Song, Y. G.; Kim, J. M.; Ku, N. S.; Bassi, G. Li; Xiol, E. Aguilera; Senussi, T.; Idone, F. A.; Motos, A.; Chiurazzi, C.; Travierso, C.; Fernández-Barat, L.; Amaro, R.; Hua, Y.; Ranzani, O. T.; Bobi, Q.; Rigol, M.; Torres, A.; Fernández, I. Fuentes; Soler, E. Andreu; de Vera, A. Pareja Rodríguez; Pastor, E. Escudero; Hernandis, V.; Ros Martínez, J.; Rubio, R. Jara; Torner, M. Miralbés; Brugger, S. Carvalho; Eroles, A. Aragones; Moles, S. Iglesias; Cabello, J. Trujillano; Schoenenberger, J. A.; Casals, X. Nuvials; Vidal, M. Vallverdu; Garrido, B. Balsera; Martinez, M. Palomar; Mirabella, L.; Cotoia, A.; Tullo, L.; Stella, A.; Di Bello, F.; Di Gregorio, A.; Dambrosio, M.; Cinnella, G.; Rosario, L. E. de la Cruz; Lesmes, S. P. Gómez; Romero, J. C. García; Herrera, A. N. García; Pertuz, E. D. Díaz; Sánchez, M. J. Gómez; Sanz, E. Regidor; Hualde, J. Barado; Hernández, A. Ansotegui; Ramirez, J. Roldán; Takahashi, H.; Kazutoshi, F.; Okada, Y.; Oobayashi, W.; Naito, T.; Baidya, D. K.; Maitra, S.; Anand, R. K.; Ray, B. R.; Arora, M. K.; Ruffini, C.; Rota, L.; Corona, A.; Sesana, G.; Ravasi, S.; Catena, E.; Naumann, D. N.; Mellis, C.; Husheer, S. L.; Bishop, J.; Midwinter, M. J.; Hutchings, S.; Corradi, F.; Brusasco, C.; Manca, T.; Ramelli, A.; Lattuada, M.; Nicolini, F.; Gherli, T.; Vezzani, A.; Young, A.; Carmona, A. Fernández; Santiago, A. Iglesias; Guillamon, L. Navarro; Delgado, M. J. García; Delgado-Amaya, M.; Curiel-Balsera, E.; Rivera-Romero, L.; Castillo-Lorente, E.; Carrero-Gómez, F.; Aguayo-DeHoyos, E.; Healey, A. J.; Cameron, C.; Jiao, L.R.; Stümpfle, R.; Pérez, A.; Martin, S.; del Moral, O. Lopez; Toval, S.; Rico, J.; Aldecoa, C.; Oguzhan, K.; Demirkiran, O.; Kirman, M.; Bozbay, S.; Kosuk, M. E.; Asyralyyeva, G.; Dilek, M.; Duzgun, M.; Telli, S.; Aydin, M.; Yilmazer, F.; Hodgson, L. E.; Dimitrov, B. D.; Stubbs, C.; Forni, L. G.; Venn, R.; Vedage, D.; Shawaf, S.; Naran, P.; Sirisena, N.; Kinnear, J.; Dimitrov, B. D.; Hodgson, L. E.; Stubbs, C.; Forni, L. G.; Venn, R.; Londoño, J. Gonzalez; Cardenas, C. Lorencio; Ginés, A. Sánchez; Gubianas, C. Murcia; Sánchez, E. Clapes; Sirvent, J. M.; Panafidina, V.; Shlyk, I.; Ilyina, V.; Judickas, S.; Kezyte, G.; Urbanaviciute, I.; Serpytis, M.; Gaizauskas, E.; Sipylaite, J.; Sprung, C. L.; Munteanu, G.; Morales, R. C.; Kasdan, H.; Volker, T.; Reiter, A.; Cohen, Y.; Himmel, Y.; Meissonnier, J.; Banderas-Bravo, M. E.; Gómez-Jiménez, C.; García-Martínez, M. V.; Martínez-Carmona, J. F.; Fernández-Ortega, J. F.; O‘Dwyer, M. J.; Starczewska, M.; Wilks, M.; Vincent, J. L.; Torsvik, M.; Gustad, L. T.; Bangstad, I. L.; Vinje, L. J.; Damås, J. K.; Solligård, E.; Mehl, A.; Tsunoda, M.; Kang, M.; Saito, M.; Saito, N.; Akizuki, N.; Namiki, M.; Takeda, M.; Yuzawa, J.; Yaguchi, A.; Frantzeskaki, F.; Tsirigotis, P.; Chondropoulos, S.; Paramythiotou, E.; Theodorakopoulou, M.; Stamouli, M.; Gkirkas, K.; Dimopoulou, I. K.; Makiko, S.; Tsunoda, M.; Kang, M.; Yuzawa, J.; Akiduki, N.; Namiki, M.; Takeda, M.; Yaguchi, A.; Preau, S.; Ambler, M.; Sigurta, A.; Saeed, S.; Singer, M.; Jochmans, S.; Chelly, J.; Vong, L. V. P.; Sy, O.; Serbource-Goguel, J.; Rolin, N.; Weyer, C. M.; Abdallah, R. I.; Adrie, C.; Vinsonneau, C.; Monchi, M.; Mayr, U.; Huber, W.; Karsten, E.; Lahmer, T.; Thies, P.; Henschel, B.; Fischer, G.; Schmid, R. M.; Ediboglu, O.; Ataman, S.; Naz, I.; Yaman, G.; Kirakli, C.; Su, P. L.; Kou, P. S.; Lin, W. C.; Chen, C. W.; Lozano, J. A. Benítez; Sánchez, P. Carmona; Francioni, J. E. Barrueco; Ferrón, F. Ruiz; Simón, J. M. Serrano; Riad, Z.; Mezidi, M.; Aublanc, M.; Perinel, S.; Lissonde, F.; Louf-Durier, A.; Yonis, H.; Tapponnier, R.; Richard, J. C.; Louis, B.; Guérin, C.; Mezidi, M.; Yonis, H.; Aublanc, M.; Lissonde, F.; Louf-Durier, A.; Perinel, S.; Tapponnier, R.; Richard, J. C.; Guérin, C.; Marmanidou, K.; Oikonomou, M.; Nouris, C.; Loizou, C.; Soilemezi, E.; Matamis, D.; Somhorst, P.; Gommers, D.; Hayashi, K.; Hirayama, T.; Yumoto, T.; Tsukahara, K.; Iida, A.; Nosaka, N.; Sato, K.; Ugawa, T.; Nakao, A.; Ujike, Y.; Hirohata, S.; Mojoli, F.; Torriglia, F.; Giannantonio, M.; Orlando, A.; Bianzina, S.; Tavazzi, G.; Mongodi, S.; Pozzi, M.; Iotti, G. A.; Braschi, A.; Jansen, D.; Gadgil, S.; Doorduin, J.; Roesthuis, L.; van der Hoeven, J. G.; Heunks, L. M. A.; Chen, G. Q.; Sun, X. M.; He, X.; Yang, Y. L.; Shi, Z. H.; Xu, M.; Zhou, J. X.; Pereira, S. M.; Tucci, M. R.; Tonelotto, B. F. F.; Simoes, C. M.; Morais, C. C. A.; Pompeo, M. S.; Kay, F. U.; Amato, M. B. P.; Vieira, J. E.; Suzuki, S.; Mihara, Y.; Hikasa, Y.; Okahara, S.; Morimatsu, H.; Kwon, H. M.; Moon, Y. J.; Lee, S. H.; Jung, K. W.; Shin, W. J.; Jun, I. G.; Song, J. G.; Hwang, G. S.; Lee, S.; Moon, Y. J.; Kwon, H. M.; Jung, K.; Shin, W. J.; Jun, I. G.; Song, J. G.; Hwang, G. S.; Ramelli, A.; Manca, T.; Corradi, F.; Brusasco, C.; Nicolini, F.; Gherli, T.; Brianti, R.; Fanzaghi, P.; Vezzani, A.; Tudor, B. A.; Klaus, D. A.; Lebherz-Eichinger, D.; Lechner, C.; Schwarz, C.; Bodingbauer, M.; Seemann, R.; Kaczirek, K.; Fleischmann, E.; Roth, G. A.; Krenn, C. G.; Malyshev, A.; Sergey, S.; Yamaguchi, Y.; Nomura, T.; Yoshitake, E.; Idei, M.; Yoshida, T.; Takaki, S.; Yamaguchi, O.; Kaneko, M.; Goto, T.; Tencé, N.; Zaien, I.; Wolf, M.; Trouiller, P.; Jacobs, F. M.; Kelly, J. M.; Veigas, P.; Hollands, S.; Min, A.; Rizoli, S.; Robles, C. M. Coronado; de Oca Sandoval, M. A. Montes; Tarabrin, O.; Gavrychenko, D.; Mazurenko, G.; Tarabrin, P.; Garcia, I. Palacios; Martin, A. Diaz; Mendez, M. Casado; orden, V. Arellano; Noval, R. Leal; McCue, C.; Gemmell, L.; MacKay, A.; Luján, J.; Villa, P.; Llorente, B.; Molina, R.; Alcázar, L.; Juanas, C. Arenillas; Rogero, S.; Pascual, T.; Cambronero, J. A.; Almudévar, P. Matía; Domínguez, J. Palamidessi; Carmona, S. Alcántara; Castañeda, D. Palacios; Abellán, A. Naharro; Lucendo, A. Pérez; Pérez, L. Pérez; Rivas, R. Fernández; Sanz, N. Martínez; Ramos, J. Veganzones; Villamizar, P. Rodríguez; Javadpour, S.; Kalani, N.; Amininejad, T.; Jamali, S.; Sobhanian, S.; Laurent, A.; Bonnet, M.; Rigal, R.; Aslanian, P.; Hebert, P.; Capellier, G.; Contreras, M. R. Diaz; Mejías, C. Rodriguez; Ruiz, F. C. Santiago; Lombardo, M. Duro; Perez, J. Castaño; de Hoyos, E. Aguayo; Estella, A.; Viciana, R.; Fontaiña, L. Perez; Rico, T.; Madueño, V. Perez; Recuerda, M.; Fernández, L.; Sandiumenge, A.; Bonet, S.; Mazo, C.; Rubiera, M.; Ruiz-Rodríguez, J. C.; Gracia, R. M.; Espinel, E.; Pont, T.; Kotsopoulos, A.; Jansen, N.; Abdo, W. F.; Gopcevic, A.; Gavranovic, Z.; Vucic, M.; Glogoski, M. Zlatic; Penavic, L. Videc; Horvat, A.; Martin-Villen, L.; Egea-Guerero, J. J.; Revuelto-Rey, J.; Aldabo-Pallas, T.; Correa-Chamorro, E.; Gallego-Corpa, A. I.; Granados, P. Ruiz del Portal-Ruiz; Faivre, V.; Wildenberg, L.; Huot, B.; Lukaszewicz, A. C.; Simsir, M.; Mengelle, C.; Payen, D.; Sanz, N. Martinez; Valbuena, B. Lobo; de la Fuente, M. Valdivia; Almudena, P. Matía; Pérez, L. Pérez; Carmona, S. Alcántara; Abellán, A. Navarro; Simón, I. Fernández; Muñoz, J. J. Rubio; Ramos, J. Veganzones; Carmona, S. Alcantara; Almudevar, P. Matia; Abellan, A. Naharro; Lucendo, M. A. Perez; Perez, L. Perez; Dominguez, J. Palamidessi; Rivas, R. Fernandez; Villamizar, P. Rodriguez; Wee, S.; Ong, C.; Lau, Y. H.; Wong, Y.; Banderas-Bravo, M. E.; Olea-Jiménez, V.; Mora-Ordóñez, J. M.; Gómez-Jiménez, C.; Muñoz-Muñoz, J. L.; Vallejo-Báez, J.; Daga-Ruiz, D.; Lebrón-Gallardo, M.; Rialp, G.; Raurich, J. M.; Morán, I.; Martín, M. C.; Heras, G.; Mas, A.; Vallverdú, I.; Hraiech, S.; Bourenne, J.; Guervilly, C.; Forel, J. M.; Adda, M.; Sylla, P.; Mouaci, A.; Gainnier, M.; Papazian, L.; Bauer, P. R.; Kumbamu, A.; Wilson, M. E.; Pannu, J. K.; Egginton, J. S.; Kashyap, R.; Gajic, O.; Yoshihiro, S.; Sakuraya, M.; Hayakawa, M.; Hirata, A.; Kawamura, N.; Tsutui, T.; Yoshida, K.; Hashimoto, Y.; Chang, C. H.; Hu, H. C.; Chiu, L. C.; Hung, C. Y.; Li, S. H.; Kao, K. C.; Sibley, S.; Drover, J.; D’Arsigny, C.; Parker, C.; Howes, D.; Moffatt, S.; Erb, J.; Ilan, R.; Messenger, D.; Ball, I.; Boyd, J. G.; Harrison, M.; Ridi, S.; Muscedere, J.; Andrade, A. H.; Costa, R. C.; Souza, V. A.; Gonzalez, V.; Amorim, V.; Rolla, F.; Filho, C. A. C. Abreu; Miranda, R.; Atchasiri, S.; Buranavanich, P.; Wathanawatthu, T.; Suwanpasu, S.; Bureau, C.; Rolland-Debord, C.; Poitou, T.; Clavel, M.; Perbet, S.; Terzi, N.; Kouatchet, A.; Similowski, T.; Demoule, A.; Diaz, P.; Nunes, J.; Escórcio, S.; Silva, G.; Chaves, S.; Jardim, M.; Câmara, M.; Fernandes, N.; Duarte, R.; Jardim, J. J.; Pereira, C. A.; Nóbrega, J. J.; Chen, C. M.; Lai, C. C.; Cheng, K. C.; Chou, W.; Lee, S. J.; Cha, Y. S.; Lee, W. Y.; Onodera, M.; Nakataki, E.; Oto, J.; Imanaka, H.; Nishimura, M.; Khadjibaev, A.; Sabirov, D.; Rosstalnaya, A.; Akalaev, R.; Parpibaev, F.; Antonucci, E.; Rossini, P.; Gandolfi, S.; Montini, E.; Orlando, S.; van Nes, M.; Karachi, F.; Hanekom, S.; Andrade, A. H.; Pereira, U. V.; Filho, C. A. C. Abreu; Costa, R. C.; Parkin, M. S. W.; Moore, M.; Andrade, A. H.; Costa, R. C.; Carvalho, K. V. Silva; Filho, C. A. C. Abreu; Min, H. J.; Kim, H. J.; Lee, D. S.; Choi, Y. Y.; Lee, E. Y.; Song, I.; Kim, D. J.; E, Y. Y.; Kim, J. W.; Park, J. S.; Cho, Y. J.; Lee, J. H.; Suh, J. W.; Jo, Y. H.; Kim, K. S.; Lee, Y. J.; Ferrero-Calleja, J.; Merino-Vega, D.; González-Jiménez, A. I.; Sigcha, M. Sigcha; Hernández-Tejedor, A.; Martin-Vivas, A.; Gabán-Díez, Á.; Luna, R. Ruiz-de; De la Calle-Pedrosa, N.; Temprano-Gómez, I.; Afonso-Rivero, D.; Pellin-Ariño, J. I.; Algora-Weber, A.; Fumis, R. R. L.; Ferraz, A. B.; Junior, J. M. Vieira; Kirca, H.; Cakin, O.; Unal, M.; Mutlu, H.; Ramazanoglu, A.; Cengiz, M.; Nicolini, E. A.; Pelisson, F. G. F.; Nunes, R. S.; da Silva, S. L.; Carreira, M. M.; Bellissimo-Rodrigues, F.; Ferez, M. A.; Basile-Filho, A.; Chao, H. C.; Chen, C. M.; Chen, L.; Hravnak, M.; Clermont, G.; Pinsky, M.; Dubrawski, A.; Varas, J. Luján; Montero, R. Molina; Sánchez-Elvira, L. Alcázar; Díaz, P. Villa; Delgado, C. Pintado; Ruiz, B. Llorente; Guerrero, A. Pardo; Galache, J. A. Cambronero; Sherif, H.; Hassanin, H.; El Hossainy, R.; Samy, W.; Ly, H.; David, H.; Burtin, P.; Charpentier, C.; Barral, M.; Courant, P.; Fournel, E.; Gaide-Chevronnay, L.; Durand, M.; Albaladejo, P.; Payen, J. F.; Chavanon, O.; Ortiz, A. Blandino; Pozzebon, S.; Lheureux, O.; Brasseur, A.; Vincent, J. L.; Creteur, J.; Taccone, F. S.; Fumagalli, F.; Scala, S.; Affatato, R.; De Maglie, M.; Zani, D.; Novelli, D.; Marra, C.; Luciani, A.; De Zani, D.; Luini, M.; Letizia, T.; Pravettoni, D.; Staszewsky, L.; Masson, S.; Belloli, A.; Di Giancamillo, M.; Scanziani, E.; Latini, R.; Ristagno, G.; Kye, Y. C.; Suh, G. J.; Kwon, W. Y.; Kim, K. S.; Yu, K. M.; Babini, G.; Ristagno, G.; Grassi, L.; Fumagalli, F.; Bendel, S.; De Maglie, M.; Affatato, R.; Masson, S.; Latini, R.; Scanziani, E.; Reinikainen, M.; Skrifvars, M.; Kappler, F.; Blobner, M.; Schaller, S. J.; Roasio, A.; Costanzo, E.; Cardellino, S.; Iesu, E.; Cavicchi, F. Zama; Fontana, V.; Nobile, L.; Vincent, J. L.; Creteur, J.; Taccone, F. S.; Park, M.; You, K. M.; Suh, G. J.; Kwon, W. Y.; Ko, S. B.; Kim, K. S.; Xini, A.; Marca, L.; Lheureux, O.; Brasseur, A.; Vincent, J. L.; Creteur, J.; Taccone, F. S.; Beane, A.; Thilakasiri, M. C. K. T.; De Silva, A. P.; Stephens, T.; Sigera, C. S.; Athapattu, P.; Jayasinghe, S.; Padeniya, A.; Haniffa, R.; Santiago, A. Iglesias; Sáez, V. Chica; Ruiz-Ruano, R. de la Chica; González, A. Sánchez; Kunze-Szikszay, N.; Wand, S.; Klapsing, P.; Wetz, A.; Heyne, T.; Schwerdtfeger, K.; Troeltzsch, M.; Bauer, M.; Quintel, M.; Moerer, O.; Cook, D. J.; Rutherford, W. B.; Scales, D. C.; Adhikari, N. K.; Cuthbertson, B. H.; Suzuki, T.; Takei, T.; Fushimi, K.; Iwamoto, M.; Nakagawa, S.; Mendsaikhan, N.; Begzjav, T.; Lundeg, G.; Dünser, M. W.; Romero, D. González; Cabrera, J. L. Santana; Santana, J. D. Martín; Padilla, Y. Santana; Pérez, H. Rodríguez; Torrent, R. Lorenzo; Kleinpell, R.; Chouris, I.; Radu, V.; Stougianni, M.; Lavrentieva, A.; Lagonidis, D.; Price, R. D. T.; Day, A.; Arora, N.; Henderson, M. A.; Hickey, S.; Costa, M. I. Almeida; Carvalho, J. P.; Gomes, A. A.; Mergulhão, P. J.; Chan, K. K. C.; Shum, H. P.; Yan, W. W.; Maghsoudi, B.; Tabei, S. H.; Masjedi, M.; Sabetian, G.; Tabatabaei, H. R.; Akbarzadeh, A.; Saigal, S.; Pakhare, A.; Joshi, R.; Pattnaik, S. K.; Ray, B.; Rousseau, A. F.; Michel, L.; Bawin, M.; Cavalier, E.; Reginster, J. Y.; Damas, P.; Bruyere, O.; Zhou, J. C.; Cauwenberghs, H.; De Backer, A.; Neels, H.; Deblier, I.; Berghmans, J.; Himpe, D.; Barea-Mendoza, J. A.; Portillo, I. Prieto; Fernández, M. Valiente; Gigorro, R. Garcia; Vela, J. L. Perez; Mateos, H. Marín; Alves, S. Chacón; Varas, G. Morales; Rodriguez-Biendicho, A.; Carreño, E. Renes; González, J. C. Montejo; Yang, J. S.; Chiang, C. H.; Hung, W. T.; Huang, W. C.; Cheng, C. C.; Lin, K. C.; Lin, S. C.; Chiou, K. R.; Wann, S. R.; Lin, K. L.; Kang, P. L.; Mar, G. Y.; Liu, C. P.; Zhou, J. C.; Choi, Y. J.; Yoon, S. Z.; Gordillo-Brenes, A.; Fernandez-Zamora, M. D.; Perez-Borrero, L.; Arias-Verdu, M. D.; Aguilar-Alonso, E.; Herruzo-Aviles, A.; Garcia-Delgado, M.; Hinojosa-Perez, R.; Curiel-Balsera, E.; Rivera-Fernandez, R.; Lesmes, S. P. Gómez; Rosario, L. E. De la Cruz; Hernández, A. Ansotegui; Herrera, A. N. García; Sanz, E. Regidor; Sánchez, M. J. Gómez; Hualde, J. Barado; Pascual, O. Agudo; León, J. P. Tirapu; Irazabal, J. M. Guergue; Pérez, A. González; Fernández, P. Alvarez; Amor, L. Lopéz; Albaiceta, G. Muñiz; Lesmes, S. P. Gómez; Rosario, L. E. De la Cruz; Hernández, A. Ansotegui; Sanz, E. Regidor; Sánchez, M. J. Gómez; Calvo, S. Aldunate; Herrera, A. N. García; Hualde, J. Barado; Pascual, O. Agudo; León, J. P. Tirapu; Corona, A.; Ruffini, C.; Spazzadeschi, A.; Marrazzo, F.; Gandola, A.; Sciurti, R.; Savi, C.; Catena, E.; Ke, M. W.; Cheng, C. C.; Huang, W. C.; Chiang, C. H.; Hung, W. T.; Lin, K. C.; Lin, S. C.; Wann, S. R.; Chiou, K. R.; Tseng, C. J.; Kang, P. L.; Mar, G. Y.; Liu, C. P.; Bertini, P.; De Sanctis, F.; Guarracino, F.; Bertini, P.; Baldassarri, R.; Guarracino, F.; Buitinck, S. H.; van der Voort, P. H. J.; Oto, J.; Nakataki, E.; Tsunano, Y.; Izawa, M.; Tane, N.; Onodera, M.; Nishimura, M.; Ghosh, S.; Gupta, A.; De Gasperi, A.; Mazza, E.; Limuti, R.; Prosperi, M.; Bissenova, N.; Yergaliyeva, A.; Talan, L.; Yılmaz, G.; Güven, G.; Yoruk, F.; Altıntas, N. D.; Mukherjee, D. N.; Agarwal, L. K.; Mandal, K.; Palomar, M.; Balsera, B.; Vallverdu, M.; Martinez, M.; Garcia, M.; Castellana, D.; Lopez, R.; Barcenilla, F.; Kaminsky, G. E.; Carreño, R.; Escribá, A.; Fuentes, M.; Gálvez, V.; Del Olmo, R.; Nieto, B.; Vaquerizo, C.; Alvarez, J.; De la Torre, M. A.; Torres, E.; Bogossian, E.; Nouer, S. Aranha; Salgado, D. Ribeiro; Brugger, S. Carvalho; Jiménez, G. Jiménez; Torner, M. Miralbés; Vidal, M. Vallverdú; Garrido, B. Balsera; Casals, X. Nuvials; Gaite, F. Barcenilla; Cabello, J. Trujillano; Martínez, M. Palomar; Doganci, M.; Izdes, S.; Besevli, S. Guzeldag; Alkan, A.; Kayaaslan, B.; Ramírez, C. Sánchez; Balcázar, L. Caipe; Santana, M. Cabrera; Viera, M. A. Hernández; Escalada, S. Hípola; Vázquez, C. F. Lübbe; Penichet, S. M. Marrero; Campelo, F. Artiles; López, M. A. De La Cal; Santana, P. Saavedra; Santana, S. Ruíz; Repessé, X.; Artiguenave, M.; Paktoris-Papine, S.; Espinasse, F.; Dinh, A.; El Sayed, F.; Charron, C.; Géri, G.; Vieillard-Baron, A.; Marmanidou, K.; Oikonomou, M.; Nouris, C.; Dimitroulakis, K.; Soilemezi, E.; Matamis, D.; Ferré, A.; Guillot, M.; Teboul, J. L.; Lichtenstein, D.; Mézière, G.; Richard, C.; Monnet, X.; Pham, T.; Beduneau, G.; Schortgen, F.; Piquilloud, L.; Zogheib, E.; Jonas, M.; Grelon, F.; Runge, I.; Terzi, N.; Grangé, S.; Barberet, G.; Guitard, P. G.; Frat, J. P.; Constan, A.; Chrétien, J. M.; Mancebo, J.; Mercat, A.; Richard, J. C. M.; Brochard, L.; Prīdāne, S.; Sabeļņikovs, O.; Mojoli, F.; Orlando, A.; Bianchi, I.; Torriglia, F.; Bianzina, S.; Pozzi, M.; Iotti, G. A.; Braschi, A.; Beduneau, G.; Pham, T.; Schortgen, F.; Piquilloud, L.; Zogheib, E.; Jonas, M.; Grelon, F.; Runge, I.; Terzi, N.; Grangé, S.; Barberet, G.; Guitard, P. G.; Frat, J. P.; Constan, A.; Chrétien, J. M.; Mancebo, J.; Mercat, A.; Richard, J. C. M.; Brochard, L.; Kondili, E.; Psarologakis, C.; Kokkini, S.; Amargianitakis, V.; Babalis, D.; Chytas, A.; Chouvarda, I.; Vaporidi, K.; Georgopoulos, D.; Trapp, O.; Kalenka, A.; Mojoli, F.; Orlando, A.; Bianchi, I.; Torriglia, F.; Bianzina, S.; Pozzi, M.; Iotti, G. A.; Braschi, A.; Lozano, J. A. Benítez; Sánchez, P. Carmona; Francioni, J. E. Barrueco; Ferrón, F. Ruiz; Simón, J. M. Serrano; Spadaro, S.; Karbing, D. S.; Gioia, A.; Moro, F.; Corte, F. Dalla; Mauri, T.; Volta, C. A.; Rees, S. E.; Petrova, M. V.; Mohan, R.; Butrov, A. V.; Beeharry, S. D.; Vatsik, M. V.; Sakieva, F. I.; Gobert, F.; Yonis, H.; Tapponnier, R.; Fernandez, R.; Labaune, M. A.; Burle, J. F.; Barbier, J.; Vincent, B.; Cleyet, M.; Richard, J. C.; Guérin, C.; Shinotsuka, C. Righy; Creteur, J.; Taccone, F. S.; Törnblom, S.; Nisula, S.; Vaara, S.; Poukkanen, M.; Andersson, S.; Pettilä, V.; Pesonen, E.; Xie, Z.; Liao, X.; Kang, Y.; Zhang, J.; Kubota, K.; Egi, M.; Mizobuchi, S.; Hegazy, S.; El-Keraie, A.; El Sayed, E.; El Hamid, M. Abd; Rodrigues, N. J.; Pereira, M.; Godinho, I.; Gameiro, J.; Neves, M.; Gouveia, J.; e Silva, Z. Costa; Lopes, J. A.; Mckinlay, J.; Kostalas, M.; Kooner, G.; Dudas, G.; Horton, A.; Kerr, C.; Karanjia, N.; Creagh-Brown, B.; Forni, L.; Yamazaki, A.; Ganuza, M. Sanz; Molina, J. A. Martinez; Martinez, F. Hidalgo; Freile, M. T. Chiquito; Fernandez, N. Garcia; Travieso, P. Medrano; Bandert, A.; Frithiof, R.; Lipcsey, M.; Smekal, D.; Schlaepfer, P.; Durovray, J. D.; Plouhinec, V.; Chiappa, C.; Bellomo, R.; Schneider, A. G.; Mitchell, S.; Durrant, J.; Street, H.; Dunthorne, E.; Shears, J.; Caballero, C. Hernandez; Hutchison, R.; Schwarze, S.; Ghabina, S.; Thompson, E.; Prowle, J. R.; Kirwan, C. J.; Gonzalez, C. A.; Pinto, J. L.; Orozco, V.; Patiño, J. A.; Garcia, P. K.; Contreras, K. M.; Rodriguez, P.; Echeverri, J. E.
title: ESICM LIVES 2016: part three: Milan, Italy. 1–5 October 2016
date: 2016-09-29
journal: Intensive Care Med Exp
DOI: 10.1186/s40635-016-0100-7
sha: 
doc_id: 14533
cord_uid: 6qfecv5h

nan

months is better than Cole's formula. Therefore, we conducted a retrospective analysis to investigate our hypothesis. Methods: The ethics committee of our hospital approved this retrospective study. We included consecutive patients aged < 6 years who underwent tracheal intubation under general anesthesia in our hospital from August 2013 to October 2015. We collected the following data from the anesthesia records: age in months, height, weight, type of a tracheal tube, and ID and OD of tracheal tube. Patients who were intubated using a cuffed tracheal tube or had incomplete data were excluded. We developed a regression formula for calculating ID and OD based on age in months and calculated the coefficient of determination R 2 by using a regression analyses. A difference of 0.4 mm in the actual and predicted tube size was considered clinically permissible. Then, we compared the rate of a clinical permissible estimation of the Cole's formula and our new formulas used by multiple comparison analysis and a p value less than 0.05 was considered statistically significant. Results: A total of 207 pediatric patients received general anesthesia during the study period. Of these, 67 patients were excluded because they did not meet the inclusion criteria. Finally, we included 140 patients for this analysis. The regression formula for predicting ID by based on age in months was ID = 0.019 × age in months + 3.48, and the coefficient of determination R 2 was 0.54. The regression formula for predicting OD based on age in months was OD = 0.024 × age in months + 5.21, and coefficient of determination R 2 was 0.558. The rate of a clinical permissible estimation of our ID and OD formulas were significantly higher than that of the Cole's formula (61 %, 69 % and 43 %, respectively; p < 0.01). Conclusions: Our results showed that the prediction of ID based on age in months is more useful than that using Cole's formula. In addition, estimation of OD based on age in months might be more rational because OD varies according to the type of the tracheal tube used. These results should be confirmed in a future prospective study.

Introduction: Survival among out-of-hospital cardiac arrest (CA) relies primarily on bystanders and their knowledge of basic life support (BLS) manouvers [1] . Many medical societies and organizations recommend teaching BLS at schools as part of the educative program [2] ; being this a reality in North European countries, but not yet an education standard issue in others including Spain. Moreover, less is written about the perception of CA and cardiopulmonary resuscitation (CPR) among the general population, and even less in school age. Objectives: Describe the perception and knowledge about CA and CPR among a teenager school population in Barcelona, Spain. Methods: Prospective, descriptive study carried out between 2007-2009 and 2012-2015 among teenagers school population, based on surveys before and after BLS -CPR classes. During this period , 17th classes were held, in 3 different schools in Barcelona. Before attending the class , each pupil was asked to answer a survey with questions related to previous knowledge of sudden death, CA, and CPR, and their attitude towards them. The class consisted on a threehour theorical and practical instruction based on the European Resuscitation Council guidelines, adapted for laypersons. Practices were held with an instructor (ratio instructor:pupil 1: [6] [7] [8] , with the Little Anne mannequins (Laerdal®). After the class, a new survey (post intervention) was distributed, with questions related to the new concepts and skills learnt, the attitude toward CA and CPR. Results: We have instructed 561 pupils (14.02 (±0.79) years, 48.2 % female). The 87.8 % had heard about sudden death and CA before the class. Regarding starting CPR: 40 % said they were not capable of doing it, and 51.2 % suggested they would be able to do CPR but in a wrong manner. In a CA scene 58.9 % would contact the emergency service and start CPR, 27.4 % would call and wait, and 11.4 % would only do CPR. After attending the classes 98.6 % declared had understood the theorical concepts and practical skilles taught; 95.4 % would changed positively their attitude towards CPR; and 97.4 % would be prone to start maneuvers. Conclusions: CPR and CA remain a well known issue among teenager population in Barcelona, as long as being an interesting topic. Nevertheless they do not feel capable of starting maneuvers. The concepts taught during the class were easy to learn , and after the intervention the majority were prone to start CPR. This population is adequate to teach CPR.

Introduction: To meet the need of patients family members and staff we started to guide visiting children at our adult ICU in the St Elisabeth hospital (EZ) in Tilburg 3 years ago. To do so we developed a guidance leaflet for parents with practical instructions and information. Additionally, practical advice is given, such as what to say to the child and what to expect when visiting. The leaflet is subdivided in developmental stages. Furthermore we developed a book "mees op bezoek", in which a child visits his father at the ICU. Pictures show what children can expect, which helps prepare the child for visiting at home. An instruction box is present at the ICU with ICU materials such as an iv catheter, a pulse oximetry or a tracheal tube. These materials give children a tactile experience of the ICU. The box is divided in two parts; the second part contains guidance materials for when a patient may die. Pedagogical staff are available to support parents, children and staff. If there are more profound problems a referral to our children's psychologist is possible. We made some improvements to our waiting area to make it more appealing to children. We instructed and educated our nurses and doctors on how to use these materials and how to guide children. We recently merged with the Twee steden Hospital in Tilburg (TSZ), in this hospital there was no program to guide children. With the merger we also wanted to introduce our "Child as a visitor program" at the ICU on location TSZ. We wanted to know if there were differences of opinion between the nursing staff on guidance of children.

Introduction: Critically ill children are underfed early in their Pediatric Intensive Care Unit (PICU) stay and this may contribute to worse outcomes. Acute Kidney Injury (AKI) occurs in 10 % of all PICU admissions and the risk of acute and chronic malnutrition is high in these patients with AKI, and the presence of malnutrition in the context of AKI has been associated with more severe clinical deterioration and organ dysfunction. Critically ill children with AKI are at high risk of underfeeding. Objectives: To evaluate the effects of protein feeding on the resolution of AKI. Methods: This is a retrospective study of critically ill children admitted from 10/2012-12/2013 to the PICU. Patients with a diagnosis of end stage renal disease requiring renal replacement therapy or had received a kidney transplant were excluded. Nutritional status assessed by weight and height WHO z-scores after admission and caloric and protein intakes calculated from I.V. fluids and parenteral and enteral nutrition for the first 8 days of admission. Energy and protein needs estimated by Schofield and A.S.P.E.N., respectively. AKI was defined by pRIFLE (creatinine only) and persistent AKI was defined as patients who did not resolve their AKI during the first eight days of PICU stay. Introduction: Sedation and analgesia are important components of postoperative management of neonates who underwent cardiac surgery. Excessive or inadequate sedation may have a significant adverse effect on patient outcome. Objectives. We aimed to determine which drug regimen would be most effective with less side-effect and better outcome. Methods: From March 2012 till March 2014 we conducted a randomized controlled prospective study in 60 neonates with congenital heart disease who underwent Arterial Switch Operation in our clinic. Inclusion criteria: 1) gestational age more than 36 weeks, 2) birth weight over 2500 g, 3) age -up to 28 days, 4) absence of concomitant diseases and surgical complications. Patients were randomized into 2 cohorts: 30 patients (50 %) were given infusion of dexmedetomidine with morphine boluses (study group) and 30 patients (50 %) were randomized to the standard regimen -infusion of morphine with diazepam boluses (control group). Results: In both groups there were no differences in pre-and intraoperative indexes, duration of mechanical ventilation, sympathomimetic support, and time of infusion of dexmedetomidine/ morphine. In study group time of ICU stay -93.5 h -was significantly shorter than in control group -120 h (p-0.02). Onset of peristalsis and start of feeding in study group was earlier than in control group -1st vs 2-d day (p-0.007) and 2-d vs 2.5-day (p-0.035), respectively. In the control group there were more patients who had complicated feeding (start after 3-d day, bloating or vomiting) -11(37 %) vs 3(10 %) in the study group. We did not observe any decrease of mean blood pressure and heart rate in the study group as it could be expected. Conclusion: Use of dexmedetomidine with morphine hydrochloride boluses for postoperative sedation and analgesia is effective and facilitates feeding process in neonates, leads to earlier onset of peristalsis and start of feeding, decreasing ICU stay.

Impact of positive end expiratory pressure on cerebral hemodynamic in paediatric patients with post-traumatic brain swelling treated by surgical decompression S.M. Pulitano' 1 , S. De Rosa 1,2 , A. Mancino 1 , G. Villa 2,3 , F. Tosi 1 , P. Franchi 1 , G. Conti 1 Introduction: Asthma exacerbation is one of the most common diagnoses seen in the pediatric ED. Several adult randomized controlled trials have shown that administration of high concentration oxygen leads to rise in carbon dioxide and increases admission rates. However, there are no studies in the pediatric population comparing the effects of high concentration oxygen versus titrated oxygen therapy in asthma exacerbation. Objectives: We evaluated the effects of transcutaneous carbon dioxide (tPaCO 2 ) in high concentration oxygen therapy versus titrated oxygen therapy to maintain saturation between 92 to 95 % in pediatric patients with acute asthma exacerbation in the ED. Methods: Children 2 to 18 years with previously diagnosed asthma with moderate to severe asthma exacerbation (asthma score > 5) were randomized to high concentration oxygen therapy (100 % oxygen via face mask at >4 L/min.) or titrated oxygen therapy (titrated up from 21 % via a blender continuously) to maintain saturations between 92 to 95 % while receiving their nebulized treatments. Exclusion criteria included disorders with hypercapnic respiratory failure, unconscious patient, history of congenital heart disease, pregnancy, history of smoking or using sedatives and depressants. Asthma therapy was provided per the ED physician. Asthma score, tPaCO 2 , PEFR (age >7 years) were measured at the start of the study and every 20 minutes for the first hour then every 30 minutes until disposition decision. The primary outcome was increase in tPaCO 2 with high concentration oxygen therapy. Secondary outcome included rate of admission to the hospital. Results: 73 patients were enrolled with mean age of 8.6 years. 60 % were males and 72 % had poorly controlled asthma with mean asthma score of 7.6. There were 36 patients enrolled in the high concentration oxygen group (HCOT) and 37 patients in the titrated oxygen group (TOT). The 0 minute tPaCO 2 were not statistically different(35.6 ± 3.8 HCOT v. 37.4 ± 4.4 TOT,p = 0.07); whereas, the 20 minutes tPaCO 2 was statistically different(40 ± 3.8 HCOT v. 37.5 ± 5.1 TOT, p = 0.02). The 60 minutes tPaCO 2 was 39.2 ± 4.6 HCOT v. 35.5 ± 4.3 TOT, p = 0.0009. At 20 minutes, 89 % of the patients had a rise in tPaCO 2 in HCOT v. 30 % in the TOT(p = < 0.0001), and at 60 minutes 78 % had a rise in tPaCO2 in HCOT v. 16 % in the TOT(p = < 0.0001). The asthma score was similar in the two groups at 0 minute (7.8 ± 1.4 HCOT v. 7.4 ± 1.3 TOT, p = 0.23); whereas, the 60 minutes asthma score was lower in the TOT(4.7 ± 1.5 HCOT v. 3.7 ± 1.3 TOT, p = 0.002). The rate of admission to the hospital was 36.1 % in HCOT v. 24.3 % in the TOT. Conclusions: High concentration oxygen therapy in pediatric asthma exacerbation leads to significantly higher carbon dioxide levels. It also causes rise in carbon dioxide from the baseline which increases the asthma scores and rate of admission.

Introduction: In critically ill patients, temperature measurement is a routine important care task and can lead to important decisions. Rectal temperature and bladder temperature are now used as a continuous body temperature measuring method in the pediatric intensive care, but these practices have several disadvantages including the patient´s discomfort, the risk of organ injury and the inaccurate measuring caused by the sensor position. A new temperature monitoring system 3M TM Spot-On TM (SpotOn) is a non-invasive zero-heat-flux thermometer designed to estimate core body temperature from the skin surface. Although the usefulness and accuracy of SpotOn system in adult patients have been demonstrated, there are no reports on pediatric intensive care patients. Objectives: The aim of this study was to evaluate the effectiveness of a new temperature measurement system attached to the forehead, and compare it to rectal temperature sensors in terms of correlation and accuracy. Methods: Pediatric patients weighing less than 10Kg, who were managed in our ICU during the period from February 2015 to March 2016, were enrolled in this study. Core temperature was measured and recorded at every minute from the both thermistor of a rectal thermal probe and with SpotOn in these patients. The data when the forehead sensor or rectal probe was taken out for nursing care was excluded from statistical analysis. Results: 53495 sets of data of 26 children (Mean BW 5630 g) were examined retrospectively. In all patients, SpotOn showed higher than the rectal temperatures. The SpotOn temperature was analyzed to be 0.82 degrees (95 % limits of agreement of ± 0.51) higher temperature than the rectal one with a moderate correlation(r = 0.73). Discussion and conclusion: Rectal temperature measurement is the gold standard method for pediatric patients in ICU despite several complications of rectal injury. Our children´s study demonstrated the slightly higher temperature in the SpotOn than rectal temperature with a substantial correlation. One possible explanation could be that the abundance of brain blood flow of children affected the results. Our study concluded that SpotOn system could be used as a highly reliable noninvasive core body temperature measurement for small pediatric patients.

Introduction: Viral bronchiolitis (VB) remains one of the leading causes of hospitalization in early childhood. Despite the heavy burden of VB on the healthcare system, little is known about the incidence of Acute Respiratory Distress Syndrome (ARDS) in this cohort of patients. In 2015, the Pediatric Acute Lung Injury Consensus Conference (PALICC) published guidelines for the definition, management and research in pediatric ARDS (PARDS) (1) . Objectives: To study the incidence and prevalence of PARDS in VB and to study the association between PARDS and specific PICU outcomes such as incidence of mechanical ventilation, noninvasive ventilator settings length of PICU stay in this group of patients. Methods: This is a retrospective single center observational cohort study that examined children 0-2 years of age admitted to the PICU with VB and respiratory failure (RF) from 2011-2014. PALICC criteria were applied to define PARDS. Clinical and demographic data was collected. Patients with a diagnosis of congenital heart disease or pre-existing chronic lung disease were excluded. Data was expressed as median with IQR ranges. Test of bivariate association were performed using Mann Whitney U test and chi square test. A two tailed p value of ≤ 0.05 was used to denote statistical significance. Results: Out of 1700 patients with RF, 330 with VB met study criteria. Eighty of these 330 (24 %) patients admitted for VB met the criteria for PARDS or at risk for PARDS. Out of these 80 patients, 25 (31 %) met criteria for PARDS and 55 (69 %) met criteria for "at risk of PARDS". Median age was 5 (2,11) months and the median weight was 6.9 (5.3, 9.5) kgs. Most common etiology for VB was respiratory syncytial virus (RSV) 68 % followed by Rhinovirus (20 %). There was no statistically significant difference in age, weight, and etiology of VB in patients with PARDS and those "at risk of PARDS." Patients with PARDS had longer hospital and PICU length of stay (LOS) and more likely to receive diuretics compared to those "at risk for PARDS" (16 (10, 21) Vs 8 (6, 10.5) , p = 0.0001; 10 (7, 13) Vs 3 (2, 4.5), p < 0.0001; and 66 % vs 33 %, p = 0.02 respectively). Nineteen (19/25, 76 %) patients with PARDS received invasive mechanical ventilation with a median duration of ventilation of 6 (1, 10) days. Conclusions: Almost a quarter of children with VB developed PARDS or were at risk of PARDS. The presence of PARDS in children with VB was significantly associated with longer PICU and Hospital LOS compared to those "at risk of PARDS". Children with VB are a high risk group for the development of PARDS.

Introduction: Mean platelet volume(MPV) seems to be a marker of platelet activation and may be related to severity of illness. 1 Changes in MPV and platelet count(PLC)could be used for disease prognosis and mortality in ICU patients. 2 We hypothesized that MPV changes and PLC could be used as prognostic tools in pediatric surgical intensive care units(PSICU). Objectives: To study the association between MPV changes and mortality and morbidity in PSICU. Also to study the relation between PLC and PSICU mortality and morbidity. Methods: This descriptive observational study was conducted on consecutive 100 pediatric surgical patients who admitted to PSICUs at Cairo University Hospitals starting from 1/6-1/12/2015.After approval by research ethics committee,informed consents were obtained from parents and pediatric cases aged from 1 month-18 years and stayed for > 48 h were enrolled.MPV and PLC were obtained and recorded at baseline(preoperative values),on the day of ICU admission(day 0),1 st ,2 nd ,3 rd ,5 th and 7 th days.To measure daily MPV changes; (ΔMPV) was constructed and computed where ΔMPV = ([MPVday(X) − MPVday (0)]/MPVday(0) × 100 %. Pediatric Index of Mortality(PIM)score was calculated on day 0 and the Pediatric Logistic Organ Dysfunction(PELOD)Score was recorded daily. Results: Patients who developed ICU complications (fever, sepsis, pneumonia, required mechanical ventilation, needed vasopressors or blood transfusion); showed higher ΔMPV compared to non complicated cases (Fig. 2) . This association was statistically significant on days 2 (p value = 0.035),3(p value < 0.001), 5 (p value < 0.001) and 7(p value = 0.017) of ICU stay but it´s insignificant on day1(p value =0.691).According to receiver operating characteristics(ROC) curve analysis, the sensitivity of ΔMPV to detect complications on day 2 was 57.2 % but its specificity on day 2 was 76.6 %.Patients who developed ICU complications showed lower PLC compared to non complicated cases (Fig. 3) .This association was statistically significant on days1(p value < 0.001),2(p value < 0.001) and 3(p value < 0.001) but it was insignificant on day 0(p value =0.237 ),5(p value =0.861) and 7(p value =0.247). On other hand, the sensitivity of PLC to detect complications day1 was 81.4 % but the specificity was 71.9 %, while the sensitivity of PLC to detect complications day 2 was 81.1 % but the specificity day 2 was 100 %. Conclusions: MPV dynamics and PLC have prognostic roles and could be used in determining several complications in critically ill pediatric surgical patients. PLC is a more specific and sensitive tool to detect complications than mean MPV dynamics. Introduction: Limited information exists regarding the association between functional status at ICU admission at and outcomes.

Objectives: We hypothesized that initial functional status assessment as well the amount of physical therapy delivered would be associated with outcomes in ICU survivors. Methods: We performed a retrospective cohort study in one Boston teaching hospital on 2,828 adults who received critical care from 1997 to 2011 and survived hospitalization. All patients had a formal evaluation by a physical therapist in the week prior to ICU admission and at hospital discharge. The exposure of interest was functional status determined by a licensed physical therapist based on the functional mobility sub scales of the Functional Independence Measure. All patients received physical therapy to improve functional performance. The primary outcome was 90-day all-cause mortality. We used logistic regression to describe how 90-day mortality differed with functional status at ICU admission. Negative binomial regression was utilized to describe how functional status at hospital discharge differed with functional status at ICU admission, the extent of physical therapy received and hospital length of stay. Results: The cohort was 52 % male, 22 % non-white and had a mean age of 64.1 years. 10 % of the cohort had sepsis, 7 % had acute kidney injury, 32 % had respiratory failure and 53 % were surgical cases. The median [IQR] hospital length of stay was 8 [4, 14] days. The 90-day mortality rate was 14.6 %. Functional status at ICU admission was robustly associated with 90-day mortality.

In a logistic regression model adjusted for age, gender, race, surgical patient type, Deyo-Charlson index, acute organ failure, sepsis, length of stay and the extent of physical therapy received, the second lowest and lowest quartiles of functional status at ICU admission was associated with a 1. 8 In critically ill patients, decreased functional status at ICU admission is associated with increased 90-day mortality. Increased intensity of physical therapy is associated with improved mortality outcomes. Both functional status at ICU admission and the intensity of physical therapy contribute to functional status determined at hospital discharge.

Introduction: Patients admitted to critical care are shown to lose significant muscle mass, with the degree of muscle loss as high as 20 % in the first week for those in multi organ failure (Puthucheary, 2013) . Early rehabilitation has been demonstrated as a safe and effective method of improving functional status at the point of critical care discharge and reducing both ICU and hospital length of stay (McWilliams et al., 2015) , although the specific impact of this on muscle mass has not been reported.

Objectives: This study aimed to analyse the impact of enhanced physiotherapy incorporating early mobilisation on the rate of muscle decline for patients admitted to critical care. Methods: Patients admitted to a large UK teaching hospital during the trial period and ventilated for ≥ 5 days were included in the study. Patients were randomised to either enhanced physiotherapy or standard care groups as part of a larger RCT. Baseline measurements were taken on the day of recruitment and then repeated at critical care discharge. Muscle mass was measured using ultrasound to calculate cross sectional area of quadriceps and biceps. To ensure validity , 2 measures were taken and the average of these taken as the final value. All scans were reviewed for agreement by 2 therapists trained in muscle ultrasound.

Results: 40 patients were included in the analysis. Patients in the enhanced physiotherapy group mobilised earlier and achieved a higher level of mobility at the point of critical care discharge (see Table 2 .) All subjects demonstrated a reduction in muscle mass of both quadriceps and biceps over the course of their critical care stay. The extent of muscle loss was however lower in those receiving the enhanced physiotherapy, although this did not reach statistical significance (Quads 3 % vs 13 %. p = 0.14317; Biceps 6 % vs 13 %, p = 0.12033).

Conclusions: A programme of enhanced physiotherapy appeared to be associated with a lower rate of muscle loss in both biceps and quadriceps in comparison to standard care. An appropriately powered RCT is required to assess these findings.

Introduction: Survivors of critical illness experience a range of impairments after intensive care, including physical, cognitive and psychological compromise. The provision of information using a diary to describe the intensive care unit (ICU) experience is one strategy that has been proposed to improve psychological health.

Objectives: The purpose of this study was to explore similarities and differences in patients' and relatives' perceptions of information containing strategies, including ICU diaries, to assist recovery after critical illness.

Methods: An exploratory mixed-methods study was undertaken in an Australian tertiary hospital with general ICU patients admitted for ≥3 days and their relatives. Semi-structured interviews were conducted 3-5 months after ICU discharge. Transcripts were analysed using content analysis. Results: Twenty-two patients and 19 relatives consented to participation and completed interviews prior to reaching data saturation. Patients were usually male (63 %) and aged 52 ± 14 years. Patients raised similar themes to relatives, although with diverse opinions. Themes of wanting to have a diary kept and considering they would find a diary helpful were consistent across a majority of participants, although with a minority expressing a desire to 'move on' and would not have liked a diary kept. Differences between patients and relatives arose in the areas of the purpose, content, ownership and timing of delivery of a diary. Patients viewed the diary as a therapeutic tool while relatives considered it as an information sharing mechanism, including as a mechanism to demonstrate to the patient 'how sick he really was' and 'what he put us through'. Possibly as a result of these differences, patients considered that ownership of the diary rested with them while some relatives envisaged shared ownership. Patients were more likely to note that the diary should not be provided to them until some weeks after ICU while relatives considered an early time point soon after ICU discharge to be appropriate. Patients were more likely to raise concerns about the potential negative impact of information sharing strategies including diaries and return visits to the ICU. Conclusions: Patients and relatives expressed common themes related to information sharing strategies after ICU, but with some important differences. Differences in purpose, content, ownership and timing of delivery of a diary suggest there is a need to consider whether the same intervention meets the needs of both groups of stakeholders. Introduction: In the Intensive Care Unit (ICU) several patients are disturbed in their cerebral function due to their critical illness and medication, leading to discomfort, agitation, restlessness, pain and delirium. Rocking Chair Mobilization Therapy (RCMT) is a chair with good seating comfort which gives rhythmic movements. Rocking chair studies have shown concrete results to improve patient satisfaction, balance and well-being in patients who suffered from dementia (1) . However, no studies have evaluated the value and the effect of RCMT for critically ill patients in the ICU.

Objectives: The purpose of the study was to evaluate whether RCMT could be used in the rehabilitation of critically ill patients in the intensive care. The focus was to explore the impact of RCMT on critically ill patients comfort, pain, agitation and delirium. Methods: The evaluation took place in a medical/surgical ICU in Denmark in the period from May to July 2015. Patients ≥ 18 years, who were physically stable and had the ability to be mobilized to chair could participate in the evaluation. The RCMT session lasted 20 minutes.Each session with RCMT was evaluated by registration of patient consciousness (Richmond Agitation and Sedation Scale (RASS)), pain (numeric rating scale (NRS) 0-10 or by Critical-Care Pain Observation Tool (CPOT)), delirium (CAM-ICU) before and after the session. Patient comfort was assessed by the patients as well as by the nurses during the session.

Results: 47 sessions with RCMT were evaluated. 24 males and 23 females, age between 49 and 88 years, participated in the evaluation.

The results showed a decrease in patient agitation level and an increase in patient consiousness. Patients´with RASS > 0 decreased from 18 before the session to 6 after the session. Patients with RASS ≤ −1 decreased from 5 before the session to 3 after the session. A decrease in delirium where 11 patients were assessed CAM-ICU positive before the session and 4 patients after the session. A decrease in pain where six patients scored NRS > 3 before the session compared to one patient after the session and 9 patients had CPOT scores > 2 before the session compared to 5 patients after the session. Assessment and evaluation of comfort by patients themselves and by the nurses, who cared for the particular patient, showed that RCMT was associated with a high degree of patient relaxation and comfort.

Conclusions: Promising results gives reason to recommend RCMT for critically ill patients in the ICU, as an alternative holistic nonpharmacological intervention to stimulate patients´bodily awareness and enhance patient comfort and rehabilitation. Introduction: Critical illness and immobility in the Intensive care unit (ICU) lead to a loss of muscle mass and reduced exercise capacity for many years following hospital discharge. [1] Nutritional management of the critically ill is challenging and most nutritional studies are focused in this period. Nutritional recommendations are for a high protein diet to minimise muscle breakdown and support protein synthesis during rehabilitation. Nevertheless, during the rehabilitation period little is known of patients' protein intake and physical functioning. Objective: To investigate physical functioning, frailty and dietary protein intake after 6 months of ICU discharge. Method: Our ICU is recognised as a therapy rehabilitation centre and the only ICU member of the UK Rehabilitation Outcomes (UK-ROC). Patients cognitive and physical functioning is assessed as part of their rehabilitation therapy with the Functional Independence Measure (FIM) score [2] . The FIM contains 18 items on motor (13) and cognitive (5) functions that are scored on a 7-point ordinal scale based on the amount of assistance a person requires to perform specific activities. The FIM scores on ICU discharge and also on return to the rehabilitation clinic after 6 months were assessed. In addition, frailty was assessed based on a scale ranging from very fit to very severely frail, terminally ill [3] and patients were asked to complete a protein food frequency questionnaire. Results: Twenty patients were assessed. Data are reported as mean and (standard deviation). Patients were male 66 % and 53.4 years (33.5). Paired t tests of the changes in FIM scores from discharge showed significant increments; 9.7 ± 11.4 (P = 0.05) and 24.7 ± 23.4 (P < 0.04) for motor and cognitive scales respectively. Nevertheless, patients reported that they were "vulnerable to moderately frail" in the frailty scale. Dietary intake was also inadequate with a protein intake of 0.83 g/kg (1.15 ). Objectives: Physiotherapy rehabilitation is a recognised component of ICU care. The Intensive Care Society -Core Standards 2 recommends that rehabilitation is 'at a level that enables the patient to meet their rehabilitation goals for as long as they…are able to tolerate it'. In order to investigate and measure the terms 'tolerate' and 'level', physiological measurements and their relationship with self-perceived exertion and tolerance were analysed. Methods: The project was registered with Guy's & St.Thomas' NHS Foundation Trust, Clinical Audit Group, (Project No. 4701). A convenience sample, of ICU patients undergoing active physiotherapy led rehabilitation, were observed between July and September 2014. A Modified Exertion Scale was used to measure patients' perceived effort. Patients also rated tolerance of the session using a Tolerability Scale, created based on the exertion scale. Sessions were timed, heart rate, blood pressure and oxygen saturation were monitored and the cardiovascular impact of the session measured using Heart Rate Reserve (HRR).

Results: Nine rehabilitation sessions were observed; mean length of 17 minutes (range 9-28). Minimum target HRR (>30 %) was achieved, but not sustained, by 3 patients, while 1 peaked within a normal target HRR (40-85 %). Of the 9 patients, 6 were able to use the tolerability scale and 7 the exertion scale. There did not appear to be a relationship between HRR and either perceived scale measurements. There did appear to be a link between perceived exertion and perceived tolerability with 5 of the 6 patients scoring within 3 points.

Conclusions: Reported perception of exertion and physiological markers could both indicate the 'level' patients are working at. We were able to measure effects of rehabilitation on heart rate. The majority of patients were able to use exertion and tolerance scales. However, the change in heart rate was not great enough to suggest a training effect, despite their exertion scores implying high effort levels. To fulfil the ICU society recommendations, a good understanding is needed of how hard patients are working during rehabilitation. Further research is needed to determine why there may be disparity between heart rate and patient-reported measures of exertion; and if either is a useful guide for exercise prescription with ICU patients.

Introduction: Critically ill patients are at risk of developing deconditioning, muscle atrophy and functional impairments long after hospital discharge. There is evidence demonstrating benefits of mobilization in critically ill patients -improved functional outcome and reduced ICU and hospital length of stay. However, there is limited information about how these advances are translated to clinical practice.

Objectives: To obtain a baseline data on patients who are eligible for mobilisation in ICU and how many of these patients are optimally mobilised in ICU. This would enable us to undertake a clinical practice improvement project (CPIP) using the Plan-Do-Study-Act (PDSA) implementation strategy to optimise mobilisation in at least 85 % of all eligible ICU patients. Methods: Setting. 14-bedded intensivist led closed surgical ICU. The mobilisation team composed of physiotherapists, bedside nurses and respiratory therapists who worked along with an intensivist. Prospective audit conducted to collect data on the patients who met the eligibility criteria of mobilisation over a 3-month period. CPIP team Results: Our audit revealed that at baseline, only 44 % of all eligible patients were optimally mobilised. RCA revealed a total of 21 barriers and through multi-voting and pareto-charting, we identified the top 3 barriers to change. Key barriers identified were: 1. Mobility not being a part of the daily review routine 2. Staff were unsure of the eligibility criteria 3. Lack of knowledge the benefits of optimal mobilisation in the critically ill. The team proposed following strategies to overcome the barriers: 1. Combined ICU multi-disciplinary handover rounds with the lead consultant asking the question "Can this patient be mobilised?" for every patient reviewed. 2. Providing a bedside decision-making algorithm on eligibility criteria, displayed within visibility of staff's work area. 3. Undertake sharing session with ground staff on the importance and benefits of optimising mobility of the critically ill.

Conclusions: Our audit revealed that less than half of eligible patients received early mobilisation. Our CPIP -a quality improvement initiative identified barriers in translating knowledge into clinical practice. Through various tools of CPIP, we identified the key barriers and strategies to overcome these barriers and thereby achieving the goal of optimising mobilisation in ICU patients.

Introduction: Bed rest and immobility during critical illness may result in profound physical deconditioning. The multidisciplinary team in intensive care includes physiotherapists, who are responsible for performing diagnoses and procedures for critically ill patients, such as ventilation, respiratory monitoring and assessments of musculoskeletal, neurological, metabolic and cardiovascular diseases, and for the prevention and treatment of the effects of prolonged immobility. Objectives: To evaluate the influence of physiotherapy on quality indicators in the intensive care unit of the Sagrada Esperança Clinic in Luanda, Angola.

Methods: A retrospective before-after study was designed to assess some quality indicators within the intensive care unit between July and September 2013, where there were no physiotherapists specially trained for respiratory care, and from January to March 2014, where the physiotherapists integrated a multidisciplinary team. The quality indicators analyzed were: the average duration of mechanical ventilation, prevalence of ventilator associated pneumonia and the rate of ventilated patients with non-invasive ventilation. The study population comprised 62 patients for 2013 and 71 for 2014. In this study the patientsć ategorization was made by age, sex, pathology and also according with the patient classification systems SAPS 3 and SOFA. The statistical analysis used the systems SPSS version 22 for a 5 % significance level. Results: The results obtained after analyzing the two homogeneous groups according to age, gender, type of admission and severity influencing the physiotherapy care in ICU quality indicators, in the Sagrada Esperança clinic, highlights the decrease of the average number of days with mechanical ventilation but it is not observed a significant relation between physical therapy and this indicator (p = 0:06). Furthermore, it is also observed a decrease ventilator associated pneumonia, and a significant relation between this indicator and the respiratory physiotherapy. Last, there is a strong relation between the increase on the number of patients without invasive ventilation and physiotherapy (p = 0.017). Conclusions: In this study it is demonstrated the respiratory therapy influences in some quality indicators, namely regarding the reduction of ventilation associated pneumonia and the promotion of non-invasive ventilation in the ICU of the CSE. Introduction: The incidence of candidemia has increased in ICU patients (1) . In addition, there are differences in epidemiology among different countries. We have previously shown an increased proportion of nonalbicans Candida species in our ICU (2) . Objectives: To identify the variables associated with candidemia due to non-albicans Candida species, as well as with fluconazole-resistant strains in a multidisciplinary ICU. Methods: All ICU patients with candidemia were prospectively studied over two time periods (2005-2008 and 2012-2015) . Demographics, illness severity, clinical and laboratory variables were recorded. SOFA score value on ICU admission subtracted from the value on the day of candidemia occurrence was defined as Delta SOFA. Patients with C. albicans candidemia were compared to those with non-albicans candidemia. Also, patients with fluconazole-resistant candidemia were compared to those without fluconazole resistance.

Results: Among 143 patients with ICU-acquired candidemia, in 55 patients candidemia was due to C. albicans and in 88 patients to non-albicans species. C. parapsilosis was the most common (46 %) followed by C. albicans (38 %). The median time from ICU admission to candidemia onset was 12 and 23 days for C. albicans and nonalbicans respectively, p = 0.02. Similarly, the median time for candidemia due to fluconazole sensitive isolate was 14 days and 32 days for fluconazole resistant, p < 0.001. Resistance to fluconazole was 9 % and 51 % in C. albicans and in non-albicans species respectively, p < 0.001).Presence of shock on candidemia day (OR 6.85; CI: 2.2-25, p = 0.001) and the Delta SOFA score (OR 0.74; CI: 0.60-0.89, p = 0.002) were independently associated with candidemia due to C. albicans. Independent risk factors for fluconazole resistant isolates were the length of ICU stay before the development of candidemia (OR 1.03; CI: 1.01-1.05, p = 0.003) and the presence of shock on candidemia day (OR 0.23; CI: 0.07-0.64, p = 0.006). Previous fluconazole exposure (10 patients) was not associated with fluconazole resistance.

Conclusions: This study confirms the predominance of non-albicans Candida species, in our ICU patients with candidemia, with high prevalence of fluconazole resistance. Early onset of candidemia and the presence of shock were most likely due to C. albicans whereas late onset was associated with fluconazole-resistant non-albicans species. These findings may be of value for empiric antifungal treatment selection.

Introduction: Invasive Aspergillus infections are well-known complications of immunocompromised states, chronic obstructive pulmonary disease and haematopoietic stem cell transplant. Bacterial coinfection is well described in influenza literature but there is scarce data on invasive aspergillosis complicated severe influenza infection.

Objectives: The aim of this study is to describe the clinical and demographic characteristics of patients with Aspergillus isolation in severe influenza A(H1 N1) pneumonia. Methods: Prospective, observational, multicenter study conducted in 148 Spanish ICUs from 2009 to 2015. All individuals with severe primary influenza A(H1 N1) pneumonia requiring invasive mechanical ventilation were included in the study. Influenza A(H1 N1) patients without coinfection were compared with those with Aspergillus isolation in respiratory samples. All serotypes were confirmed using RT-PCR at ICU admission. Patients´demographic, clinical, radiologic features, laboratory values, ICU and hospital length of stay (LOS) and outcomes were recorded. Discrete variables are expressed as counts (percentage) and continuous variables as medians with 25th to 75th interquartile range (IQR Conclusions: The mortality rate was significantly higher in H1 N1 patients with Aspergillus isolation in respiratory samples. Diagnosis of invasive aspergillosis in critically ill patients in the post-influenza era must be re-evaluated. Clinical studies should be conducted in order to know the clinical significance of Aspergillus isolation in respiratory samples in intubated patients with primary influenza A(H1 N1) pneumonia. Methods: This prospective, monocentric, study performed over 12months assessed the value of a twice weekly dosage of fungal biomakers (Candida serology IgA, IgM, IgG, ß-D-Glucan (BDG) and mannan antigens) in ICU patients after LT. Proven/probable/possible infection was defined according to the EORTC/MSG criteria. Colonisation was defined by presence of Candida sp in respiratory samples without any sign of invasive infection. The study was approved by the Ethical Committee. Results are presented as means.

Results: We analysed 39 ICU patients after LT. 77 % had a Candida sp colonisation while an invasive infection was proven in 4 (10 %) patients. Candida albicans was cultured from 72 % of the pulmonary samples. 50 % of the invasive infections were related to C. glabrata. Results of biomarkers dosages are presented in the table. Positive Candida IgG serology was observed in 45 % of the cases. Mortality rate at 6-months after LT was 71 % in the immunised patients versus 26 % in nonimmunised patients. An invasive candidiasis (IC) was present in 14 % of the immunised patients versus 10 % in non-immunised patients. At least one BDG dosage was positive in 60 % of the cases. BDG dosage value decreased after surgery, reaching a non-significant value after the 4th day. In proven IC, BDG measurements reached concentrations >500 pg/ml, 16 days before initiation of antifungal treatment. No patient had positive mannan antigen measurement.

Conclusions: A twice weekly dosage of BDG seems to be useful in the decision making process for early initiation of antifungal therapy in LT patients. The cutoff for a significant value of BDG needs to be defined. Pre-transplantation assessment of Candida IgG serology could help to identify patients at risk of post-operative fungal infection. Introduction: The antifungal (AF) therapy strategy (pre-emptive vs culture based treatment) in intensive care unit is a matter of debate [1] . The necessity to not delay the initiation of the AF in invasive candidiasis (IC) must be balanced with the cost and risk of selecting resistant pathogens when AF are prescribed too widely. Burn patients are at risk of IC because of the frequent use of antibiotics and immunodeficiency. Objectives: To evaluate our antifungal (AF) therapy strategy in suspected or proven IC in terms of prognosis and risk factors of IC. Methods: Observational, descriptive, retrospective study conducted from June 2012 to September 2015 in the Saint Louis Hospital Burn Unit. Inclusion criteria: patients treated with pre-emptive (severe sepsis or septic shock with Candida Sp colonization) or curative (proven, PIC) AF. The outcome was the PIC (candidemia and/or positive peritoneal sample). Clinical characteristics, organ supports, AF treatments and outcome were collected and compared between PIC and suspected IC (SIC). The results are presented in median (IQR) or n (%). Results: 715 patients were admitted during the study period including 184 with a total body surface area (TBSA) >20 %. 44 treated with AF including 14 PIC (32 %). In those 44 patients: age 55 (43-67), TBSA (43-67), SAPSII 41 (29-53), ABSI 10 (8-12) and SOFA 10 (5-15). Renal replacement therapy 22 (50 %), mechanical ventilation 36 (82 %), parenteral nutrition 7 (16 %). Inhospital mortality = 60 % (68 % SIC vs 32 % PIC, p = 0.33). 3 patients with PIC (21 %) were treated before the IC diagnosis (2 because of filamentous infection before the PIC). The delay between admission and AF treatment initiation was 16 days. Patients characteristics, organs supports were not significantly different between PIC and SIC at the treatment initiation except for the SAPSII (PIC 43 (29-57) VS SIC 37 (26-49), p = 0.03). 30 patients (68 %) received an echinocandin as a first-line treatment. 6 (5.5-6.5) sites were monitored for Candida colonization the week before treatment initiation. Patients with PIC had higher colonization index than those with SIC (55 % vs 33 %, p = 0.01) and a candida score significantly higher (4 vs 3 (0.5-3.5) respectively, p = 0.03). A semiquantitative estimation of the fungal inoculum had no predictive value. Conclusions: In this study, the majority of PIC were treated after diagnosis confirmation. Only 1/30 (3 %) patient treated preemptively did declare a PIC. The outcome was not different when the treatment was initiated after confirmation. The results of this study highlight the difficulty to identify patients at highest risk of IC, and question the strategy of preemptive treatment in this population. Objectives: We wanted to determine whether PCT guided antibiotic rationalisation could reduce fungal colonisation and antifungal usage. Methods: We undertook a retrospective observational study at a nine bedded ICU department in the United Kingdom. We collected data on all patients admitted to the unit in the year prior and post the introduction of PCT guided rationalisation of antibiotics. We used the pharmacy database to assess the use of antibiotics, correcting for changes in costs over this time. We used the microbiology database to assess the rate of patients colonising fungal species and those requiring treatment.

Results: Since the introduction of PCT, the average expenditure on antibiotics per ICU admission fell 35.11 % (p 0.01). The rate of ICU patients colonised with a fungal species fell from 32.3 % to 14.89 % (p < 0.0001). The incidence of patient's prescribed systemic antifungal therapy fell from 14.58 % to 4.26 % (p < 0.0001).

Conclusions: We demonstrated a significant reduction in patients colonised with fungal species and those requiring anti-fungal therapy since introducing PCT guided rationalisation of antibiotics. A prospective randomised controlled trial is required to assess whether this equates to improved patient outcome. 5 ), median 28 days. C. albicans was the most commonly isolated species (sp) (34,6 %), Candidemia was the most common diagnosed infection. The sp isolated in blood cultures (BC) were: 23,1 % C. albicans, 21,1 % C. haemulonii, 14.4 % C. parasilopsis. We had 2 + BC for Trichosporon asahii. Catheter -related infection by C. albicans, C. parasilopsis, and candida haemulonii was diagnosed in 6 patients. Positive Urine samples were found mostly for C. albicans 11.5 %. The most frequently factors associated with fungal infection were: > than 7 days in the ICU 90.4 %, urinary catheter 90.4 %, broad-spectrum antibiotic exposure 88.5 %, indwelling central venous catheter (CVC) 82.7 %, feeding tube 82.7 %, total parenteral nutrition 63.5 %, Invasive mechanical ventilation 63.5 %. To highlight an association with Acinetobacter baumannii in 44.2 % of our patients. Doctors chose Fluconazole in 88.9 % as a first line of therapy. An antibiogram was performed and the susceptibility was confirmed. ICU mortality rate, 26.9 %. Conclusions: In our environment, C. albicans continues to be the species that causes the largest number of invasive candidiasis. Prolonged stay in the ICU is an important risk factor to develop fungal infections. Even with the particular features of a burn patient, their complexity, and the negative impact of each infection; Fluconazole keeps having an important role in the treatment as a first line.

The effect of introduction of daily chlorhexidine bathing on healthcare-associated infections and acquisition of multi-drug resistant organisms E. Ahmadnia 1 Introduction: It has been suggested that daily bathing with chlorhexidine impregnated cloths may significantly reduce the acquisition of multi-drug resistant organisms (MDROs), incidence of central line associated bloodstream infections (CLABSIs), and the development of intensive care unit (ICU) acquired bloodstream infections [1] . However, more recent data have failed to support daily bathing of critically ill patients with chlorhexidine for these purposes [2] . Objectives: To determine if the implementation of a daily chlorhexidine bathing regimen affects acquisition rates of MDROs, the incidence of CLABSIs, and ICU bacteraemias.

Methods: A quality improvement project was conducted at a 44 bedded adult critical care unit within a UK University Hospital (incorporating major trauma, medical, and surgical patients). During the 1 year control period (December 2013 to November 2014), all patients were bathed using soap and water. During the subsequent intervention period (December 2014 to November 2015), all patients were bathed using 2 % chlorhexidine impregnated cloths (Clinell, GAMA Healthcare). The acquisition of MDROs, incidence of CLABSIs and ICU bacteraemias were recorded during these periods (6 months pre-and 6 months post-chlorhexidine for CLABSIs, one year for the other outcomes).

Results: The study covered 34317 patient bed days (16887 pre-and 17430 post-introduction of chlorhexidine bathing). There were an identical number of MDRO acquisitions in each group (290), giving rise to an MDRO acquisition rate per 1000 bed days of 17.17 in the control group compared to 16.63 in the chlorhexidine group (P = 0.70). CLABSI incidence per 1000 bed days was higher in the control group compared to the chlorhexidine group (9.51 vs 7.90; P = 0.26). The incidence of significant bacteraemias per 1000 bed days was similar in the the two groups (5.57 before and 5.45 during chlorhexidine bathing; P = 0.88), but the incidence of bacteraemias due to skin commensals per 1000 bed days was lower in the chlorhexidine group (7.22 vs 5.45; P = 0.04).

Conclusions: At our large University Hospital ICU with a heterogeneous patient population, the introduction of routine daily chlorhexidine-impregnated cloth bathing appears to significantly reduce the incidence of bacteraemias due to skin commensals and demonstrates a non-significant reduction in CLABSIs. Given the uncertainties surrounding diagnosis in the ICU, the effect seen may be of benefit in reducing the use of antibiotics to cover for these skin commensals -both in terms of antibiotic stewardship and health economics.

Introduction: The current CDC guideline published in 2011 for the prevention of intravascular catheter-related infections recommends skin preparation with a greater than 0.5 % chlorhexidine with alcohol solution before CVCs or ACs placement and with dressing changes, which was changed from 2 % chlorhexidine recommended in the 2002 guideline. However, few studies investigated the superiority of 1 % CHG over either 0.5 % CHG or 10 % PVI for the prevention of catheter colonization as CDC guideline recommends.

Objectives: Efficacy comparison of three antiseptic solutions [10 % aqueous povidone-iodine (PVI), and 0.5 % and 1.0 % alcoholic chlorhexidine gluconate (CHG)] for preventing intravascular catheter colonization.

This was a open-label, multicenter, prospective, randomized controlled trial conducted at 15 ICUs in Japan. The intravascular catheters included central venous catheters (CVCs) and arterial catheters (ACs). Patients aged >18 years of age undergoing CVC and AC insertion in ICU were randomized to receive one of three antiseptic preparations pre-insertion. Catheters were removed when no longer necessary or if catheter-related infection was suspected. After catheter removal, distal tips were cultured using semi-quantitative/ quantitative techniques. Catheter colonization and catheter-related bloodstream infection (CRBSI) incidences were compared. Results: While a total of 1132 catheters were randomized, several catheters were excluded due to withdraw of their informed consent and lack of cultured catheters after randomization, and 796 (70 %) catheters were included in the full analysis (0.5 % CHG n = 261, 1.0 % CHG n = 278, and 10 % PVI n = 257). The median catheterization duration was 3.8 days (95 % CI: 2.0-6.7 days); no significant intergroup differences were observed (p = 0.43). Catheter-tip colonization incidence (per 1000 catheter days) was 10.5, 3.9, and 3.7 events in 10 % PVI, 1 % CHG, and 0.5 % CHG groups, respectively (p = 0.04). Catheter colonization risk was significantly higher in the 10 % PVI group. No significant intergroup differences CRBSI probability were observed Introduction: Spontaneous intracerebral hemorrhage (ICH) is the most fatal stroke subtype worldwide caused by spontaneous vascular rupture due to hypertension or amyloid angiopathy. An accurate prediction of ICH outcome would assist both families and physicians to decide therapies and monitorization at an early stage.

Objectives: To evaluate the relationship between the hematoma volume and location with mortality and functional outcome in patients with spontaneous ICH. Methods: We performed a prospective observational study, included patients admitted in ICU with spontaneous ICH. We determined hematoma volume at admission with Kothari modified formula (AxBxC/2) and divided them in two groups according the location as infratentorial or supratentorial. We collected GCS, SOFA, APACHE II and GRAEB at admission, medical history and complications during the first week in the ICU. We established modified Rankin Scale (mRS: poor outcome >2) and Glasgow Outcome Scale (GOS, poor outcome < 4) at ICU discharge. We used %, mean (SD) and median (minimal/maximum). T-Student and χ2 (p < 0.05) were used for the univariable analysis. We conducted a multivariable analysis for mortality with binary logistic regression (95 % CI, OR) p < 0,05. ROC curve was determined for the volume of hematoma associated with mortality (IC 95 % p < 0.05). Results: We enrolled 120 patients. 68 % were men, mean age 62 (±12.6) years. Global mortality was 34.2 %. 83.3 % were supratentorial and 16.7 % infratentorial. Mean APACHE II 14 (±6, 5) and GCS 10.4 (±4,1) and median SOFA 4 (0-14) and hematoma volume 21,35 cc 3 . There were no significant differences between the two groups (infra and supratentorial) except ICH volume (p 0.000) and length of stay (LOS)-ICU (p 0.021). In the univariable analysis worse outcome with mRS was related with the volume of the hematoma (p 0,03) but not with GOS (p 0,29). Variables associated with mortality: GCS (p 0.000), APACHE II (p 0.000), GRAEB (p 0.001), SOFA (p 0.000), LOS-ICU (p 0.004) and ICH volume (p 0.000). After the multivariable analysis we determined hematoma volume was an independent risk factor for mortality (OR 1,032; 95 % CI 1,019-1,046; p 0.000). According the location we obtained a significantly association with mortality in the supratentorial group (p 0,000). We performed a ROC curve of this group and obtained an AUC 0,718 (95 % CI 0,603-0,834; p 0.000) with cutoff point of 20,2 cc 3 . Conclusions: Hematoma volume and LOS-ICU are greater in supratentorial ICH. The hematoma volume is associated with a worse outcome at ICU discharge and a supratentorial ICH volume above 20.2 cc 3 is related to higher risk of mortality. Introduction: Aneurysmal subarachnoid hemorrhage (SAH) is an acute cerebrovascular event, which leads to devastating consequences, high mortality and is an important cause of neurologic disability among survivors. Incidence is reported between 2 to 16/100000 and mortality rates vary widely, ranging from 8 to 67 % among different authors. Many complications associated with SAH, such as delayed cerebral ischemia or hydrocephalus, also play a role in the poor functional outcome in survivors. Paulo Niemeyer State Brain Institute is a reference and high-volume center for SAH, located in Rio de Janeiro, Brazil, receiving patients from all over the state.

Objectives: The aim of the study was to describe the characteristics of patients with SAH admitted to the ICU, as part of a large prospective ongoing study, and to evaluate the factors associated with outcome.

Methods: From July 2015 to March 2016, every patient admitted to the ICU with aneurysmal SAH, 18 years and older was enrolled in the study. Data were collected prospectively during hospital stay. The primary endpoint was mortality and dichotomized functional outcome, (poor outcome defined as Modified Rankin Scale 4-6) at hospital discharge.

Results: A total of 53 patients were included. The median age was 56 (47-66), 43 patients (81 %) were female. Demographic characteristics are presented in Tables 8 and 9 . Twenty-nine patients (55 %) were treated by clipping, and 13 patients (25 %) were hydrocephalic and needed an EVD. An intracranial pressure monitor was inserted in 11 patients (21 %). Nine patients (17 %) developed sepsis or septic shock during ICU stay and pneumonia was present in 14 (26 %) patients. Rebleeding was diagnosed in 8 patients (15 %), vasospasm was present in 17 (32 %) patients, post-surgical deterioration was diagnosed in 16 (30 %) patients and 13 (25 %) patients developed DCI.

Twenty-two (42 %) patients were mechanically ventilated. Hospital mortality was 11 % (6 patients); and 26 patients had unfavorable (49 %). In univariate analysis, factors most frequently seen in patients with unfavorable outcome were rebleeding (69 % vs 0 %, p = 0.002), vasospasm (46 % vs 19 %, p = 0.031), post-surgical neurological deterioration (46 % vs 15 %, p = 0.013), DCI (42 % vs 7 %, p = 0.003) and pneumonia (42 % vs 11 %, p = 0.01). Although not statistically significant, there was a trend towards the association between sepsis/septic shock (27 % vs 7 %, p = 0.076) and unfavorable outcome. Conclusions: SAH is associated with high morbidity. Neurological complications such as rebleeding, vasospasm, post-surgical neurological deterioration and DCI, as well as clinical complications (eg. pneumonia) were associated with unfavorable outcomes. Therapeutic interventions to prevent neurological and systemic complications may have an impact on clinical outcomes. Introduction: The management of patient into the ICU after been submitted to a CNS resection is an important challenge. Surgery is indicated for diagnosis, to reduce tumor bulk and to manage raised intracranial pressure. Primary brain tumors are classified based on their cellular origin and histologic appearance. The most common malignant brain tumor is glioblastoma multiforme, this group have a poor prognosis.

Objectives: The goal was to make a descriptive analysis about the evolution of patient submitted in the ICU for postoperative control following a surgical resection of intracranial tumors. Methods: A retrospective and observational study was conducted on all elective consecutive surgical procedures for tumor resection admitted into the ICU. We analyzed variables related with the tumor, predisposing pathology, surgical data and evolution in the ICU. We considered as an unfavorable evolution the death into the first month after the intervention or the decrease in two points or more of the Canadian´s scale score (CSS). Is a comparative study analyzed by Student´s t-test, ANOVA of one factor and Pearson´s chi-square test. Comparative study expressed by: mean difference, relative risk and confidence intervals at 95 %. Results: We analyzed 271 patients over of 5 years (2011-2015) . Of the total, 31.3 % are high-grade gliomas, 9.6 % low grade gliomas, 32,7 % meningiomas , 8.5 % metastasis and 14.2 % other type of tumors. Average age is 54.6 years (SD 14.8), it is significantly lower in the low-grade gliomas, and in the group of other tumor types compared to other groups. 50.6 % are men , the most common in men (63.5 %) and meningiomas and other tumors in women (60.9 % and 62.5 % respectively) gliomas. 88.7 % are supratentorial location. Average size is 29.9 mL (SD 28.2) . The average score in the preoperative Karnofsky scale is 73.2 (SD 12.9) . The average income APACHE is 6.9 points (SD 4.5 ). An unfavorable evolution is observed in 14.0 % of patients (6.6 % per patient died and 7.4 % decline in the CSS) after one month , with no differences between different types of tumors. The percentage of deaths in the first month is higher in those undergoing surgery for metastasis (20.8 %, RR 4.0, CI 1.5 to 10.2) . Mortality at two years of intervention is 31.5 %, being higher in sifnificativamente undergoing metastasis (75 %; relative risk 5.7, CI 3.6 to 9.0) and high-grade gliomas (63.6 %; relative risk 4.6, CI 3.1 to 7.4).

Conclusions: Patients undergoing brain tumors have a significant risk of poor outcome , which is significantly higher in metastatic patients from the first month of intervention and in patients undergoing high-grade gliomas at two years.

Introduction: Nosocomial infection (NI) is still an issue in neuroritical care. Objectives: We analysed NI in a preventive multimodal protocol in patients with acute brain disease. Method: We performed a 10-year prospective observational cohort study in 3464 patients (pts) with acute brain disease admitted to an eight-bed adult neuro-intensive care unit (NICU). We defined our preventive multimodal protocol as: 1) keeping a hygienic and epidemiological regime including isolation of pts with multi-drug resistant bacteria 2) correct antibiotic policy, and 3) regular microbiological screening. There were 198 (5.7 %; wound 2.1 %, respiratory 1.8 %, urinary 1.0 %, bloodstream 0.6 % and other 0.2 %) pts with NI. We compared NI group pts with the control group of 3266 pts and searching predictors of NI in univariete analysis.

We did not find differences in age (p = 0.416), male (p = 0.716), weight (p = 0.423) or body mass index (p = 0.966), but there were more stroke pts and fewer tumour pts (p < 0.001). NI pts stayed in NICU longer (mean 15.3 vs 4.8, p < 0.001), on admission had lower Glasgow Coma Scale (mean 11.5 vs 13.1, p < 0.001), higher Therapeutic Intervention Scoring System (TISS, p < 0.001), Acute Physiology and Chronic Health Evaluation II (p < 0.001), and CRP (p < 0.001); in the NICU they had higher CRP (p < 0.001) and NICU mortality (p < 0.001); on discharge they had worse Glasgow Outcome Scale (p < 0.001) and higher TISS sums (p < 0.001). NI pts had more accesses, which were strong predictors of NI: artery (odds ratio [ Conclusions: Our study confirmed that nosocomial infection is associated with worse outcome and higher cost, and that accesses are still risk factors in a preventive multimodal protocol.

The predictive value of emergency triage codes on the outcome of aneurysmal subarachnoid hemorrhage Introduction: Outcome of patients with aneurysmal subarachnoid hemorrhage (SAH) was associated in different studies with different variables (baseline illness severity, physical status, treatments, complications), but the relationship between outcome and triage assessment in the emergency setting has never been evaluated. Emergency triage in Italy is carried out with color codes: red (immediate life-saving intervention needed), yellow (urgent intervention needed), green (delayed intervention is sufficient), white (not urgent).

Objectives: To study the relationship between triage severity codes assigned to patients with SAH in an Italian emergency setting and the outcome expressed as modified Rankin Score (mRS) at hospital discharge (good outcome for mRS ≤ 3, poor outcome for mRS > 3).

Methods: A retrospective clinical study included 52 patients with aneurysmatic SAH admitted to emergency departments of Bologna catchment area, and then to intensive care unit (ICU), from January 2014 to January 2015. Aneurysm coiling or clipping was performed after neuroradiological diagnosis and clinical stabilization, excluding patients too ill to benefit. Intensive care treatment was carried out according to current practical guidelines. Demographic, clinical and interventional data, complications, severity scores and outcome scores were recorded. The following parameters were considered in univariate analysis: age, sex, clinical condition on arrival in the emergency department (triage code, GCS, WFSN scale, vomiting and seizures) aneurysm clipping or coiling and other neurosurgical interventions, hydrocephalus, vasospasm, cerebral infarction (CT scan), fever, sepsis, acute respiratory failure with P/F ≤ 200, cardiovascular complications (hypotension requiring vasopressor therapy, acute cardiomyopathy, arrhythmias requiring treatment); the outcome variable was modified Rankin Score >3 at hospital discharge. Results: Poor outcome (mRS > 3) was observed in 33 % of triage green codes, 53 % of yellow codes, 75 % of red codes. The univariate analysis showed the statistically significant (p < 0.05) association with mRS > 3 for the following variables: triage red code, WFSN scale >2, acute respiratory failure, cardiovascular complications, sepsis. On logistic regression analysis, the red code assigned in the emergency department, cardiovascular complications and sepsis were associated with poor outcome.

Conclusions: The severity of general clinical conditions after subarachnoid hemorrhage needing immediate life-saving intervention, feature labelled "red code" in the emergency triage, was associated with poor outcome (mRS > 3), while the other triage codes did not show any significant correlation with outcome. Cardiovascular complications and sepsis during hospital stay were other variables associated with mRS > 3.

Evaluation of intracerebral hemorrhage (ICH) score in patients admitted in intensive care by supratentorial brain hemorrhage L. Perez-Borrero 1 Introduction: Intracerebral hemorrhage is a stroke subtype with high mortality and significant disability among survivors.

Objective: To evaluate in our area the Intracerebral Hemorrage (ICH) score in patients with spontaneous supratentorial brain hemorrhage.

Methods: Multicenter prospective observational study in three hospitals in Andalusia (Spain). We studied all patients with supratentorial brain hemorrhage admitted to the Regional Hospital of Malaga (between  2006 to Introduction: Within the clinical importance of the SAH, there are factors described in the scientific literature that speak of an unfavorable evolution of the disease. Our hypothesis is based on trying to demonstrate if only one therapeutic intervention could alter the significance of these factors. Objectives: Analyze the sociodemographic, laboratory findings, clinical and radiological factors that influence prognosis at 6 months in discharged aneurysmal SAH patients treated with endovascular intervention. Methods: We performed a retrospective longitudinal observational study of all patients who were diagnosed with an aneurysmal SAH in ICU services of 2 hospitals between March 1st 2012 and November 10th 2015. They were treated by endovascular intervention. After being discharged from ICU and after 6 months of neurologic follow-up. Patients were divided into two groups, one formed by those who presented a favorable outcome (EF) and the other by those who didn´t (ED). The variables studied were age, sex, HBP, DM, smoking and dyslipidemia. At the time of admission PO2, PCO2, leukocytosis, hyperglycemia and hypertension was determined as well as sodium, magnesium and chlorine plasma levels.

The clinical status of patients on admission was assessed using the Hunt-Hess and WFNS scales. The severity of SAH was determined by CT using the Fischer scale. The aneurysm was located by four vessel angiography. The time between the SAH clinic presentation and the procedure was recorded, as well as if aneurysmal occlusion was complete or not. As for the complications, we took into account the presence of fever, hydrocephalus, vasospasm and infarction. Results: For the study, 39 patients who underwent acute endovascular SAH treatment using coils, were selected. Female sex was the predominant sex 80 % Vs 75.8 % between ED and EF, respectively. The age group most frequently found was between 45 and 65 years (60 % for ED and 62 % for EF). Logistic regression analysis determined as associated with a worse outcome factor: hyperglycemia on admission(OR 8.94, 95 % CI 1.76-45.3, p = 0.007), clinical status on admission determinated by Hunt-Hess (OR 14.58 CI 2.62-81 95 %, p = 0.0018) and WFNS scales (OR 5.75, 95 % CI 1.21-27.13, p = 0.02). The presence of fever on admission also has proven to be a poor prognostic factor (OR 7.33 95 % CI 1.48-36, p = 0.01). Conclusions: Clinical factors for aneurysmal SAH patients treated with endovascular procedure that have shown relation with the clinical outcome at six months are: poor clinical grade on admission, hyperglycemia and fever. These data are similar to those found in the literature and support the idea that the therapeutic decision (surgical or endovascular) is not the determining factor for the evolution of these patients, however, the ones mentioned above could be. Background: Conflicting results have been obtained by studies attempting to assess the risks of ischemic stroke in patients with venous thromboembolism, while the long-term risk of stroke in survivors of venous thromboembolism remains unexplored. Objective: We evaluated whether the risk of ischemic stroke in patients hospitalized with venous thromboembolism is higher when compared to the general population. Methods: One million patients from National Health Insurance beneficiaries in Taiwan were sampled. There were 2,145 patients who had been hospitalized with diagnosis of venous thromboembolism and 727,607 unexposed subjects. All adult patients were followed from 1 January 2005 to 31 December 2013 to evaluate if ischemic stroke was diagnosed. Cox regression models were applied to compare the hazards adjusted for potential confounders. Results: After controlling for age, gender, urbanization level, socioeconomic status, diabetes, hypertension, coronary artery disease, hyperlipidemia, history of alcohol intoxication, malignancies, congestive heart failure, atrial fibrillation, smoking, peripheral artery disease and Charlson Comorbidity Index, the adjusted hazard ratio of ischemic stroke was significantly increased in patients with venous thromboembolism (2.47; 95 % CI, 2.16-2.83). A subgroup analysis based on patients who survived longer than 12 months in the cohort also revealed higher hazard ratio in the patients with venous thromboembolism. (1.32; 95 % CI, 1.05-1.66). Conclusion: The possible risk of ischemic stroke is significantly higher in patients hospitalized with venous thromboembolism than in the general population. Introduction: Status epilepticus (SE) is a common neurological emergency with considerable associated health-care costs, morbidity and mortality. 1,2 SE is defined as a prolonged seizure or multiple seizures with incomplete return to baseline. 1, 2 The overall mortality of SE is around 20 % with convulsive status epilepticus representing about 45-74 % of all cases. 1 Status Epilepticus Severity Score (STESS) is a prognostic score that relies on four outcome measures (age, history of seizures, seizure type and extent of consciousness impairment) determined before treatment institution that ranges between 0 and 6. 3 Objective: Evaluation of STESS as a prognostic measure of functional impairment, neurologic motor deficits and 28-day mortality. Methods: Retrospective observational study of patients with SE admitted at a general intensive care unit (ICU) from 2009 to 2015. Age, gender, SAPS II/III, type of SE, length of stay, number of anti-epileptic drugs, duration of SE, functional impairment, neurologic motor deficits and 28-day mortality were collected through the ICU informatics database -PICIS®. Data is presented as mean ± SD and we used logistic regression to correlate STESS with study variables. Statistical analysis was performed using XStat 2016®. Results: Sample included 29 patients, 72,6 % male, age 46,6 ± 12,5 years, SAPS II 42,3 ± 13,1, SAPS III 55,7 ± 16,6, ICU length of stay 4,4 ± 2,6 days and hospital length of stay 16,2 ± 13,3 days. Convulsive SE represented 79,3 % of cases. STESS score`s mean was 2,6 ± 0,8. SE lasted more than 1 day in 48,3 %. Electroencephalogram was performed in 51,6 % of the patients. 48,1 % of the patients needed two or more anti-epileptic drug for SE. At hospital discharge 20,7 % had functional impairment and 13,8 % had neurologic motor deficits. Mortality was 6,8 % at 28 days. There was a correlation between STESS and mortality (OR = 10,4; ROC = 0,963), functional impairment (OR = 1,93; ROC = 0,667) and neurologic motor deficits (OR = 4,32; ROC = 0,820). The number of antiepileptic drugs and SE duration had no correlation significance. Conclusions: We found an excellent correlation between STESS and mortality in our study. Besides this, we also found this score to be a good prognostic tool for functional impairment and neurologic motor deficits. We consider our main limitations the sample size and lower mortality. Despite we recommend using STESS as an outcome predictor. Introduction: In order to determine optimal airway protection measures in early postoperative period after fossa posterior surgery (PFS), it is important to carry out a prognosis of neurological dynamics based on the preoperative neurological exam. We have designed Neurological Evaluation Scale (NES). Objectives: Our study was aimed at determining the potential of NES to predict brain stem deterioration in early postoperative period after PFS based on the assessment of the preoperative neurological status. Methods: The prospective study was carried out during the period from December 2013 to June 2014 and included 182 patients operated for fossa posterior tumors (FPT). To be included in the study, patients had to be over 18 years old and operated for fossa posterior non infiltrative paraxial tumors. We examined all patients before and after the operation, immediately after the extubation in ICU. NES provided complex neurological assessment with an emphasis on the brain stem function. Postoperative NES points were subtracted from the preoperative points -AB-criterion (ABc). Positive ABc corresponds to intensification of neurological deterioration. Negative or zero ABc corresponds to neurological improvement. All neurological symptoms were grouped in 7 NES blocks according to their relation to CNS. Results: We divided all patients in two groups depending on their ABc, which revealed that the patients with positive ABc had reliability less NES points before operation compared to the patients with negative or zero ABc. We found out the frequency of occurrence of each NES block for inclusion in the full neurologic status. We discovered that caudal stem affection occurred more frequently in the patients with more NES points. We evaluated the probability of neurological impairment or regression of neurological symptoms depending on ABc with sensitivity 90,7 % and specificity 95 %. We created a prognostic model, which could predict the discharge from clinic outcome on the basis of the NES blocks points assigned during the early postoperative period. Conclusions: We revealed neurological features of postoperative period in patient after FPS. Our data could predict neurological outcomes, and be useful in optimization tactic of airway protection. and those who died (5.9 ± 3; 6.5 ± 2 and ± 5.6 25 ± 2 vs 42; 29.4 ± 52 and 23.1 ± 43 respectively) but difference was not significant. An inverse correlation between inflammatory biomarkers (PCT, CRP and IL-6) and IgM EndoCAb was detected. IL6 showed a higher correlation, but without statistically significant differences. ICU mortality rate was 8 %.

Conclusion: IgM EndoCAb were detected in septic shock caused by GPB, it could be explain by a bacterial translocation. Patients with major endotoxaemia have higher consumption of antibodies and therefore lower levels of IgM EndoCAb that is associated with a worse prognosis.

The relationship between the neutrophil/lymphocyte ratio and mortality in the severe sepsis patients Y. Conclusions: SP compliance is below recommended but the mean HOBE reaches the lower limit of the recommendation. The factors affecting SP compliance differ according to the method used for data collection and include other factors than patient's clinical condition. Politics targeting to increase its compliance should address various areas of care such as team and professionals, resources and equipment and re-consider clinical indications for SP.

The project was funded by the 14°National Award of Nursing Research from Marques de Valdecilla Hospital (Spain). gained to collect staff opinions and anonymised patient data. 90 ED, 30 OR and 110 ICU professionals were surveyed following a pilot to determine options, ranking and scoring criteria a priori where needed. Anonymous patient data from 36 intubated patients who were cared for in the ED, OR and ICU within their first 24 hours were collected. This included physiological observations and supportive care standards around A, B and C. Results: The most striking differences in staff opinion involved the preferential use of artificial colloid-based fluid resuscitation in sepsis (17 % ED staff; 50 % OR staff; 8 % critical care staff); the value and significance of recording End Tidal CO2 (4 % ED staff; 97 % OR staff; 69 % critical care staff); and the potential preferential use of flowdirected fluid boluses rather than pressure-directed fluid boluses in critically ill patients (22 % ED staff; 90 % anaesthetic staff; 75 % critical care staff). When observing supportive care standards the largest differences were in the use of stress ulcer prophylaxis (only prescribed in critical care); patient positioning (head-up: 94 % patients in CC; 6 % in ED); the recording of sedation level (0 % ED; 6 % anaesthetic; 78 % critical care) and the recording of ventilator parameterstidal volume, peak pressure and ET-CO2-(3 % ED; 69 % anaesthetic; 100 % critical care) . Finally, in respect to patient pathophysiology, all groups were under ventilated and over oxygenated. Mean arterial pressure was most divergent from baseline in the ED. However, changes in pathophysiology were related to interventions (fluid boluses, analgesia, surgical interventions, inotropes, pressors) rather than location. Despite the divergent views regarding the relative value of flow monitoring, observed fluid boluses were predominantly triggered by pressure changes in all three locations. Conclusions: Differences in staff attitudes; application of standards and patient pathophysiology were identified between care locations. The influence of variation in resources and professional composition of teams (nurses:doctors) on these results requires further work. It remains uncertain whether more uniform approaches would improve patient outcomes.

Lung comet score (LCS) for evaluation of extravascular lung water (EVLW) in intensive care unit (ICU) patients undergoing renal replacement therapies (RRT) A. Taggu Methods: A prospective observational study was conducted on 450 patients in ICU needing RRT. Exclusion criteria were Age < 18 years, pregnant, amputees, cardiac pacemakers, pre-existing lung diseases and ascites. Lung comet score as per validated technique , BIA measurements and baseline data were collected pre and post dialysis. Lung comet score and other covariates were fitted into a regression model using BIA as the standard test. Based on BIA delta hydration relative (HS rel), patients were divided into normohydration and hyperhydration using a cut-off of 15 %. Results: A linear regression model in predialysis state showed that only LCS could significantly predict lung water (const 8.79, Coef.0.203, p value 0.00 ). In the postdialysis state LCS perfectly predicted lung water (const 9.98, Coef.,0.261; p value 0.00 ).Bland Altman plots showed good agreement between LCS and hydration status (BIA) pre and post dialysis. The LCS >15 nearly perfectly predicted hydration status in both pre and post dialysis states. Conclusions: Lung comet score is a good surrogate of EVLW and reliably predicts reflects hydration status pre and post dialysis in ICU patients. Introduction: Treatment withdrawal in intensive care is common (1) . Whilst significant research attention has focused on how treatment is withdrawn and what information is communicated to families (2) Introduction: Micro-aspiration of subglottic secretions is considered a major pathogenic mechanism of endotracheal tube-associated pneumonia (ETAP), either postoperative pneumonia or ventilatorassociated pneumonia. Endotracheal tubes (ETs) with taper-shaped cuffs have been proposed to provide a better seal of the extraluminal airway, thereby preventing micro-aspiration and possibly ETAP. Objectives: To perform a systematic review and meta-analysis to assess the efficacy of ETs with taper-shaped cuffs in the prevention of ETAP. Methods: A systematic search of MEDLINE, EMBASE and CENTRAL/ CCTR was conducted in March 2016. Eligible trials were randomized controlled clinical trials (RCTs) comparing taper-shaped cuffs with standard, cylindrical-shaped cuffs in intubated patients. All studies reporting the incidence of ETAP were included. Inclusion of trials was irrespective of publication status, date of publication or language. Random-effects meta-analysis calculated the risk ratio (RR) and 95 % confidence interval (CI) for the incidence of ETAP between both groups using the Mantel-Haenszel method. Results: Three RCTs, given a total of 855 patients, met the inclusion criteria. One trial was published as a conference abstract only (1), while the others were published in full (2, 3) . None of the trials was blinded for the intervention. 440 patients were allocated to the intervention arm and 415 to the control arm. 87 ETAP episodes occurred in the intervention group and 83 in the control group. The pooled RR for the incidence of ETAP was 1.01 (95 % CI, 0.78-1.31; z = 0.11 p = 0.91). Conclusions: The use of endotracheal tubes with taper-shaped cuffs did not show to reduce the incidence of ETAP. However, the number of available studies is small, and there is an inherent risk of bias due to the unblinded designs. Background: Understaffing of ICU's can have serious adverse consequences both for patients and for nurses, and therefore it is important to have an adequate number of nurses on the ward. Nurses however are in short demand and resources are scarce. Being able to predict the nursing workload for a certain group of patients may help to allocate nursing capacity as efficiently as possible and thus to reduce costs, without endangering the patients safety and nurses' health. NAS is a validated tool for the measurement of nursing workload in an intensive care unit. Goal: This study was conducted to investigate whether it is possible to predict the nursing workload for a homogeneous group of patients, admitted after an in or out of hospital cardiac arrest and to assess the effects of baseline characteristics, vital parameters and admittance time on this workload. Method: We performed a retrospective analysis of NAS scores of all IHCA and OHCA patients admitted to our ICU from October 2012 until September 2015 during the first 48 hours of stay. The NAS was recorded per patient per nursing shift. We furthermore recorded patient characteristics and vital parameters. Results: During this period 386 patients, 273 males and 113 females, were admitted to the ICU after cardiac arrest. The mean age at admission was 63.4 years (SD = 14.8). The mean NAS at admission was 71.4 (SD = 22.4). Patients admitted in the evening shift had a significantly higher NAS compared to patients admitted in the night shift (74.3; SD = 21.2 vs 67.4; SD = 22.3 (p = 0.037)), but no significant difference was found with the day shift (mean NAS day shift: 71.7; SD = 23.3). After admission the workload decreased in all patients by a mean of 18.3 points (SD = 23.9; p < 0.001). A higher SOFA score, a higher PEEP and a lower pH at admittance resulted in a higher NAS score on average over time (p < 0.001). Conclusions: The nursing workload at admission of patients after cardiac arrest is fairly predictable, with no clinically significant difference between shifts , necessitating a nurse-to-patient ratio of at least 1:1. After the first shift it is almost always possible to decrease the nurseto-patient ratio to 1:2. The NAS was influenced by severity of illness. These results can be used to assess the needed nursing staff for the treatment of these patients for the first days after admittance. Introduction: Major trauma and severe sepsis are both leading causes of admission to the resuscitation rooms in emergency departments across the world. Despite obvious differences in precipitating mechanism, there are surprising similarities between subsequent pathophysiology: both disorders lead to disorders of the macrocirculation, microcirculation and host inflammatory response (1, 2) . Objectives: Here we compare the baseline epidemiology, pathophysiology, operational and clinical management of intubated resuscitation room patients with these two critical illness syndromes (Major Trauma/Septic shock). The results will be used to facilitate the design and planning of a study to test the feasibility/effectiveness of advanced monitoring systems (thromboelastography, oesophageal doppler flow monitoring, echocardiography, and microcirculatory monitoring) in the resuscitation room management of critically ill patients with these conditions. Methods: Institutional approval was gained to collect anonymised patient data over a 6-month period from a mixture of written and electronic records. Where appropriate, significance was tested by Mann Whitney U (Sigmaplot 11.0). Results: 162 patients, intubated pre-hospital or in ED resus, were identified with trauma or sepsis diagnoses. Trauma patients were commoner (n = 128; 79 %) and more likely to be intubated prehospital (107/128; 83.6 % vs 27/34; 79 %). Lactate profiles were similar in the two groups at start and end of resus episode ( Figure 18 ). Patients with major trauma were more hypertensive but equally tachycardic when compared with patients with severe sepsis/septic shock. Patients with septic shock/severe sepsis (median 202 minutes vs 128 minutes) spent longer in resuscitation room, but received less documented consultant-level review (90.6 % vs 20.6 %). Imaging of major trauma patients was with CT (128/128; 100 %) and ultrasound (44/128; 34.3 %) in contrast to septic patients (11/34 CT; 32.4 %; 2/34, 1 % US). ICU and hospital mortality was higher in patients with sepsis (38.2 % vs 14.8 %), but death in resus only occurred in the trauma population. Only two patients, both with facial trauma, would have had a relative contraindication to the proposed advanced monitoring. Conclusions: This novel preliminary work has highlighted some important differences between the epidemiology, outcomes, pathophysiology and clinical/operational management of intubated patients with severe sepsis versus major trauma. These will influence the conduct and outcome measures of any trial of advanced monitoring in this setting. However, contraindications to any of the advanced monitoring technologies being considered were rare and no obvious barriers to the planned study of advanced monitoring were identified. Introduction: In our 15-bedded GICU, demand for beds has increased while recruitment of ICU trained nurses has decreased. There is enduring evidence of links between workload and stress [1, 2] , with high levels of burnout reported in ICU nurses [3] . There has been a shift towards measuring what makes people positive and engaged [4] , rather than why people reach the extreme state of burn out. It is important to understand factors that affect work engagement to develop strategies that enhance nurse retention and improve the quality of ICU patient care. Objectives: To examine the impact of an education initiative for novice ICU nurses on work engagement for the ICU nursing staff and organisational resource use. Methods: A pre -post design was used to collect data from all ICU nurses at the start of the education programme and at 6 months following he intervention. Work engagement was measured using the self-report 17 item Utrecht Work Engagement Scale (UWES) [4] with an open question to capture staff experiences. Organisational impact was measured using levels of sick leave, % staff turnover, use of agency nurses and staff recruitment. Results: Fifty three ICU nurses completed the pre-intervention survey (61 % response) and 42 completed the post-intervention survey (46 % response). Respondents had reasonable years of ICU experience (Mean7.57, SD 8.1) and time in current post (Mean6.21, SD 8.2). Internal consistency for the UWES was high (alpha .91). Levels of work engagement (Mean [SD]) increased (3.94, [0.64] vs 4.03 [0.69]) but did not reach significance and remained in the 'average' band as judged by the scale authors [4] . When examined by senior and junior nurses, the increase was similar. Organisational measures showed decrease in sick leave, turnover, agency use and increase in recruitment of experienced ICU nurses. Qualitative feedback was positive, with perceptions of improvement in unit morale due to time being invested in the individual and reduced stress and workload for shift leaders. Conclusions: Providing education for the newest ICU recruits can have benefits for the whole ICU team. However, it is important to examine how work engagement might be further improved.

The incidence of silent aspiration on intensive care N. Maistry Royal Brompton and Harefield NHS Foundation Trust, Rehabilitation and Therapies, London, United Kingdom Intensive Care Medicine Experimental 2016, 4(Suppl 1):A872 Introduction: The incidence of Dysphagia on intensive care is an area of growing research. Dysphagia is associated with aspiration pneumonia and increased ICU bed days. In general, Speech and Language Therapy (SLT), makes recommendations based on the results of a clinical bedside swallowing evaluation, despite the unreliability of this method 1. This is largely due to the difficulty accessing gold standard assessment methods such as Videofluoroscopy (VF) and Fibreoptic Endoscopic Evaluation of Swallowing (FEES)3. Referral for these assessment methods are based on a local defined criteria. This study evaluates the incidence of silent aspiration identified by VF and FEES in a 20 bedded tertiary Cardio-Respiratory Intensive Care Unit. Objectives: To determine the incidence of silent aspiration, defined as "aspiration before, during, or after swallowing in the absence of cough or visible signs of choking and distress 2," in ICU patients assessed by VF or FEES between July 2014 and June 2015. Method: Data was retrospectively reviewed for 12 month period from all ICU referrals made to SLT for swallowing evaluation. All patients received a clinical bedside swallowing evaluation. The results are presented as percentages and counts for patients receiving VF and FEES that silently aspirated. Results: A total of 97 patients were referred for swallowing assessment and 35 % (34/97) had a VF or FEES. There were 27 males and ages were 44.1 ± 15.7 years. In this group, 22 patients had Videofluoroscopic assessments and 12 patients had FEES. In the VF group 77 % silently aspirated whilst in the FEES group the values were 83 %. 27 patients (79 %) silently aspirated during objective assessment, impacting on how and when oral feeding was commenced. Conclusion: This study suggests that silent aspiration is highly prevalent in this population group. Consequently, VF and FEES should be part of standard routine assessment in the management of critically ill patients.

Delerium related incidents at the ICU and nursing aspects A. van Introduction: At the ICU of VU University Medical Center (VUmc) nurses are frequently confronted with delirium 1,2 . Delirium is known to be present in 60-80 % of mechanical ventilated patients and 20-50 % in non-ventilated patients. Immediate consequences are falling incidents or for patients to remove tubes and IV lines that are necessary for treatment. In literature, this is stated as a result of treatment, but often data is missing. Consequences of removal are increased risk of complications 3 , prolonged mechanical ventilation, LOS and increased morbidity/mortality 4 . Objectives: To measure the frequency of removing tubes, lines and falling incidents related to delirium. Methods: A multidisciplinary focus group was formed (2014) in order to properly diagnose, prevent and/or treat delirium due to the high prevalence. The first steps were increasing awareness and implementing the CAM-ICU score. To clarify delirium-related incidents a one year period was set in which the dedicated senior nurse informed and trained the nursing staff regarding delirium and potential risks. To register delirium-related incidents a modified report button was built in the EPR (Metavision, IMD Soft) and used beside the regular incident reporting system 5 . Results: After one year, 52 individual patient incidents were reported concerning falling or tube or IV line removal. This included 14 gastric tubes, 15 airway tubes, 14 IV/CVC/arterial lines, 5 other lines and 4 fall incidents. In 83 % of the cases the patient was diagnosed with delirium. 4 out of 5 patients received medication or were fixated before the incident despite a 40 % CAM-ICU registration rate. Because the focus group doubted about underreporting 10 nurses were interviewed if the results corresponded with their experience. They were unanimous that there was hardly any underreporting. Discussion: Despite therapy or fixation delirium-related incidents occur on a weekly basis at our ICU, causes harm and increases nursing workload. Although the incidence rate is presumed to be low, there is no feeling of satisfaction. Further improvement is necessary due to the high risks for the patient. Therefore, we need to be able to diagnose incidents faster so we can start treatment sooner. Although the CAM-ICU score was implemented, compliance is insufficient. Increasing compliance is the first step to further improvement. The follow-up question is whether delirium-related injury can be reduced when CAM-ICU compliance improves. Second step is to investigate the effectiveness of our fixation protocol.

Prospective study to determine the predictors of extubation success A. Taggu Introduction: Timely extubation is crucial in critically ill patients. Traditional indices like rapid shallow breathing index are considered as accurate during the spontaneous breathing trial. Multiple other proposed parameters like diaphragm thickness, fluid balance and cardiac indices have been shown to predict succesful extubation in the recent years. Objectives: To assess the reliability of the parameters in predicting successful extubation.

Methods: A prospective observational study done on 220 adult patients eligible for extubation as decided by the attending intensivists. Exclusion criteria: Pregnant and tracheostomised patients. Along with baseline parameters, following measurements were taken pre and post extubation.

1. Cardiac parameters including Simpsons method for ejection fraction, E/A, E/e' (lateral) for diastolic function, TAPSE and TAD for Right ventricular function. All recordings were taken just before extubation and within six hours post extubation. 2. Just before extubation,high frequency linear ultrasound probe was used to measure the right sided DT at the zone of apposition (ZOA) between 8 th to 10 th intercostal spaces in mid-axillary line.The change in DT fraction(Δdtfrac_pre%) was calculated as DT(end-inspiration)-DT(end-expiration)/DT (end-expiration)x100.RSBI was simultaneously recorded. 3. Fluid balance 24 hours were recorded. Figure 20) . We found no significant difference on mortality when limiting the results just to trial employing CVVH at hemofiltration rate lower or higher then 50 ml kg −1 h −1 .

Conclusions: Blood purification with CVVH might be associated with a significant reduction in mortality when performed in patients with sepsis or ARDS. This is the first meta-analysis suggesting beneficial effects of CVVH on mortality and we could suppose that the beneficial effects of CVVH in these inflammatory conditions could arise from the immunomodulatory properties of hemofiltration. Further high-quality randomized controlled trials adequate powered for mortality are needed to clarify the impact of CVVH on these inflammatory conditions.

The authors declare no support or funding and no potential conflict of interest. We defined extreme hyperbilirubinemia as a state of total bilirubin above 20 mg/dl and selected all patients whose serum total bilirubin increased above 20 mg/dl at least once during their stay in the intensive care unit. We investigated the overall clinical course of the patients and compared the differences between one group with normalization of total bilirubin (recovery group) and the other group without normalization (non-recovery group). Furthermore, we evaluated the association between prognosis and various clinical factors, including the peak total bilirubin levels, increasing rate of total bilirubin (Vi), results of laboratory analyses related to hepatic function, and clinical features at the time of extreme hyperbilirubinemia. These data were analyzed using Chi-square test and Cox and logistic regression analyses. Introduction: Propofol is widely used in critical care sedation due to its pharmacological properties which allow serial neurological examination (1) .Hypo tension is a common side effect of propofol infusion, which affect patient outcome. Introduction: Propofol is a common intravenous drug used during anesthetic induction and sedation because of its rapid onset and short duration. Its downfall, however, is that patients experience injection pain so severe that they recall induction as the most painful part of the sedation process. Among numerous reports in efforts to decrease propofol injection pain, the most effective combination of drug and non-drug intervention evaluated through a quantitative systematic review revealed to be pretreatment with 0.5 mg/kg lidocaine in combination with a tourniquet for venous occlusion. The majority of these reports conclude that a single method is insufficient in eliminating propofol injection pain. Objectives: We evaluated the effect of heated carrier fluids (40°C) in decreasing propofol injection pain. Methods: A randomized controlled clinical trial was conducted in 90 patients (ASA 1 or 2), ages 18 to 65. Patients were allocated into 3 groups (n = 30) each. Group W received 200 ml of heated carrier fluids for 20 minutes prior to 2 mg/kg propofol injection; group L received 200 ml of heated carrier fluids for 20 minutes prior to lidocaine pretreatment and 2 mg/kg propofol injection: and group C (control group) received 200 ml of room temperature fluids prior to 2 mg/kg propofol injection. Propofol injection pain was evaluated using the verbal pain score (VPS). Results: Group W and L showed significant reduction in the incidence and severity of injection pain compared to group C (P < 0.05). VPS was significantly lower in group W (p = 0.018) and L (p = 0.036) compared to group C. There was no statistical difference between group W and group L (p = 0.3). There was statistically significant difference in mean blood pressures measured after 2 mg/kg propofol injection among groups. Conclusions: Both heated carrier fluids and combination of lidocaine pretreatment effectively reduced propofol injection pain. Objectives: We conducted a prospective cohort study to clarify the epidemiology and the nature of AEs in surgical inpatients in Japan.

Methods: The Japan adverse EvenT (JET) study was a prospective cohort study which had evaluated AEs and medical errors (MEs) at 2 tertiary care hospitals. The 38 medical and surgical wards were stratified according to hospital and whether they were medical or surgical wards, and study wards were randomly selected. Intensive care units (ICUs) were all included. We included all adult patients aged > = 15 years old who were admitted to any of the selected 23 study wards (10 medical, 11 surgical, and 2 ICUs) over a 2-month period.

The primary outcome of this study was the epidemiology and the nature of AEs and MEs in the patients who had operation during the study period. Trained nurses placed at each participating hospital reviewed all charts daily on weekends, along with laboratories, incident reports, and prescription queries to collect any potential event.

They also collected the characteristics of the patients in the cohort. Independent physician reviewers evaluated all potential events and classified to whether they were AEs or MEs, as well as to their classification, severity and preventability. Introduction: Whereas the importance of low tidal volume to avoid ventilator-induced lung injury (VILI) in patients with ARDS is well known, several uncertainties still exist regarding how to set positive end-expiratory pressure (PEEP). Many approaches have been considered, but no one showed a clear effectiveness in terms of outcome. Recently a ventilator strategy using esophageal pressure to estimate the transpulmonary pressure has been proposed by Talmor and colleagues 1. Although they found an improvement in arterial oxygenation, it was not explored whether the increase in oxygenation was due to lung recruitment. Objectives: The aim of this study was to assess whether the PEEP set to maintain a positive end-expiratory transpulmonary pressure (P L ) is associated with an increase in lung recruitment estimated by lung ultrasound score (LUS) 2 . Methods: 12 patients with moderate and severe ARDS were enrolled. For the first 2 hours, PEEP was set according to the Acute Respiratory Distress Syndrome Network standard-of-care recommendations (phase A). It was then adjusted according to measurements of esophageal pressure for the following 2 hours (phase B) to maintain a positive P L at the end of expiration. The primary end point was the improvement in lung recruitment assessed with lung ultrasound. [2] No data are available on the relationship between opening pressures and disease severity. Objectives: To describe lung recruitment as a function of the transpulmonary pressure in mild, moderate and severe ARDS. Methods: ARDS patients underwent a low-dose end-expiratory CT scan at PEEP 5 cmH 2 O and three end-inspiratory CT scans at the plateau pressures reached starting from PEEP 5 cmH 2 O, 30 cmH 2 O and 45 cmH 2 O. In each of the CT slices, lung profiles were manually delineated, excluding hilar structures. Thereafter, quantitative analysis of CT scan images was performed and the gas and tissue fractions were computed. We defined the recruitability as the difference of not inflated tissue between 45 and 5 cmH 2 O, that we arbitrarily assumed to be the "full recruitment". [3] The grams of recruited tissue were computed across the pressure intervals at which the CT scan were performed, as the differences of not aerated tissue. Airway and esophageal pressures were continuously measured and transpulmonary pressure was computed as: Driving airway pressure (cmH 2 O) -(esophageal plateau pressure (cmH 2 O) -esophageal end-expiratory pressure at PEEP (cmH 2 O) [1] . Results: Thirty-three patients were studied, 5 with mild, 10 with moderate and 18 with severe ARDS, according to the Berlin definition. [2] As reported in the Table 20 and Fig. 23 , the amount of tissue which can be opened between 30 and 45 cmH 2 O was 8 %, 17 % and 37 % respectively in mild, moderate and severe ARDS). Mild ARDS patients nearly completed recruitment at approximately 15 cmH 2 O transpulmonary pressure while in moderate and severe ARDS recruitment continues up to 26 cmH 2 O transpulmonary pressure. Conclusions: At the clinically recommended plateau pressure of 30 cmH 2 O, in severe ARDS, up to 1/3 of the lung tissue recruitable at 45 cmH 2 O, stays always closed. Beyond contributing to the gas exchange impairment (depending on the perfusion), these "always" collapsed regions may also act as stress risers at their interface with aerated regions, though they are theoretically protected from the mechanical ventilation. Introduction: Extracorporeal membrane oxygenation (ECMO) is a rescue therapy for patients with acute respiratory distress syndrome (ARDS) by providing additional oxygenation, and removing carbon dioxide thus permitting less injurious mechanical ventilation settings that have been shown to protect the lungs from additional injury. Objectives: To evaluate associations between distinct ventilator settings during ECMO, and outcome of ARDS patients. Methods: Individual patient data analysis of observational studies in adult ARDS patients receiving ECMO for refractory hypoxemia. Multilevel multivariable logistic regression models and Cox-proportional hazards models were used to determine which settings and parameters had an independent association with the primary endpoint all-cause mortality. Results: Nine studies with 545 patients were selected ( Figure 24 ). Initiation of ECMO was accompanied by significant decreases in tidal volume, positive end-expiratory pressure (PEEP), plateau pressure (Pplat), and driving pressure (ΔP = Pplat -PEEP), respiratory rate and minute volume (Figure 25 ), and resulted in higher PaO 2 to FiO 2 ratios, higher arterial pH and lower PaCO 2 ( Figure  26 ). Higher age, lower body mass index, and higher lactate were associated with all-cause mortality after multivariable adjustment. ΔP, both before and during the first three days of ECMO, demonstrated an independent association with all-cause mortality ( Conclusions: In this series of ARDS patients receiving ECMO for refractory hypoxia, ΔP and FiO 2 were the only ventilatory variables that had an independent association with outcome. These findings indicate the potential for improvement in the management of patients with ARDS undergoing ECMO. Lungs were analysed for wet-to-dry ratio, BAL protein, static compliance, SpO 2 and histology. To detect the timing of injury, rats received Evans Blue dye (EBD-30 mg/kg IV) at the initiation and were euthanized immediately before lung deflation or at 2, 5, 10 or 20 min afterwards (4/group). Terminal BAL analysed for EBD absorbance. Ultrastructural impact was studied by electron microscopy on lungs sampled from rats euthanized before deflation, and at 1 and 5 min after deflation. Hemodynamic data was obtained by ECHO performed at baseline (PEEP 3 cmH 2 O), immediately before and after deflation, and at 30mins after deflation. RV pressure was measured with a Millar catheter.

Results: Wet-to-dry ratio (6.1 ± 0.6 vs 4.6 ± 0.4; P = 0.00) and BAL protein (3.9 ± 4.0 vs 1.5 ± 0.7; P = 0.18) was higher; and static compliance (0.48 ± 0.97 vs 0.82 ± 0.2; P = 0.00) and SpO 2 (67 ± 23.5 vs 91 ± 4.4; P = 0.04) were lower in Intervention vs Control. Histology revealed collapse, hemorrhage and neutrophil accumulation in the intervention group. BAL Evans Blue demonstrated that microvascular leak was absent before deflation and was maximal by 10 min of deflation. Ultrastructural analysis showed that sustained inflation caused minimal swelling of epithelium and endothelium before deflation; deflation resulted in major cellular and interstitial edema, and endothelial injury. Hemodynamic data showed that RV and LV were under-filled during inflation. Upon deflation, RV output, pulmonary vascular resistance, RV systolic transmural and diastolic pressures increased precipitously. RV/LV ratio increased progressively.

Conclusion: Sudden deflation after sustained inflation with PEEP causes protein leak, inflammation, hypoxemia, reduced compliance, endothelial injury and RV failure. The mechanism appears to be endothelial injury resulting in microvascular leakage, pulmonary hypertension and RV failure. Significance: Deflation injury may be an important entity to prevent when using sustained inflation manoeuvres and may explain -in partwhy several important RCTs in ARDS have been negative.

Low dose steroids reduce short term mortality in septic shock patients: results of an individual patient data meta-analysis R. Introduction: Previous research has suggested that the use of low dose steroids may be beneficial during septic shock. However subsequent inconsistent results explain the lack of consensus amongst doctors around the world about whether treatment with low dose steroids does improve the overall recovery and survival in patients with septic shock. We hypothetize that the lack of consistent evidence on the effect of low-dose steroids on short term mortality may be related to underpower. treated for septic shock.

Objectives: The primary objective of the present study was to estimate the effect of three different therapeutic regimens (hydrocortisone alone, hydrocortisone plus fludrocortisone, neither hydrocortisone nor fludrocortisone) on 28-day mortality in patients treated for septic shock using an individual patient data meta-analysis. Methods: Individual patient data meta-analysis including the 3 major recent randomized controlled trials comparing early lowdose short course hydrocortisone and fludrocortisone to placebo (GER-inf (1)), hydrocortisone alone to placebo (CORTICUS (2)) or hydrocortisone to hydrocortisone and fludrocortisone (COIITSS (3)) in septic shock patients. The primary outcome measure was all cause 28-day mortality. Secondary outcomes measures were 90day mortality, resolution of organ dysfunction (as measured by the time to reach a Sequential Organ Failure Assessement score < 8), time to vasopressor and mechanical ventilation discontinuation, intensive care unit and hospital lengths of stay as well as the rate of superinfection. Treatment effect on the primary outcome was quantified using relative risk and estimated using targeted maximum likelihood estimation.

Results: A total of 1,307 patients were enrolled in the 3 trials. When compared to the placebo, hydrocortisone + fludrocortisone significantly reduced 28-day mortality (RR = 0.73, 95%CI = 0.67-0.79, p < 0.001). Hydrocortisone + fludrocortisone was also superior when compared to the placebo and hydrocortisone pooled together (RR = 0.86, 95%CI = 0.79-0.94, p = 0.002). Hydrocortisone + fludrocortisone significantly decreased 28-day mortality (RR = 0.75, 95%CI = 0.67-0.85, p < 0.001) in the nonresponders, while it was associated with an increase in 28-day mortality in the responders (RR = 1.17, 95%CI = 1.06-1.29, p = 0.002) ( Figure 30 ). Hydrocortisone + fludrocortisone was also superior when considering secondary outcomes such as vasopressor discontinuation or lengths of stay.

Conclusions: In this individual patient data meta-analysis including the 3 major randomized controlled trials on the subject, we found that an early short course of low-dose hydrocortisone and fludrocortisone decreases 28-day mortality and improves recovery from organ failure in septic shock patients non responding to a corticotropin stimulation test. Introduction: Statin therapy during intensive care unit (ICU) stay has been associated with a reduction in all-cause hospital mortality in some studies. This association was especially noted in septic patients. However, potential benefit needs to be validated in randomized, controlled trials.

Objectives: The purpose of this study was to compare the effect of simvastatin plus standard therapy on mortality and total ICU length of stay (LOS) to that of standard therapy alone in critically ill septic patients.

Methods: A prospective randomized, open label, controlled pilot clinical trial was conducted on patients diagnosed with sepsis/severe sepsis as defined by the American College of Chest Physicians (ACCP). Hundred patients met the study criteria and were randomized into two groups; a standard group who received standard treatment and simvastatin group who received the standard treatment plus 40 mg simvastatin. Primary outcomes were 28 days ICU mortality and total ICU LOS. Plasma C-reactive protein (CRP), total creatine kinase (CK) and liver enzymes [alanine aminotransferase (ALT) and aspartate aminotransferase (AST)] were measured as secondary outcome measures.

Results: A total of 72 patients completed the study. Simvastatin was well tolerated, with no increase in adverse events between the two groups. Total ICU LOS was significantly lower in the simvastatin group. However, the number of patients with 28 days ICU mortality in the simvastatin group was lower compared to standard group; but survival failed to reach statistical significance. Similarly, plasma Creactive protein failed to reach statistical significance between the two groups Conclusions: Treatment with simvastatin 40 mg in patients with sepsis/severe sepsis is safe and associated with an improvement in number of deaths and ICU LOS but without subsequent improvement in survival.

The use of anapnoguard 100 system in intubated critically ill patients a randomized controlled study Introduction: The AnapnoGuard 100 system (AG) (Hospitech Respiration LTD., Petach-Tikva, Israel) is an innovative respiratory guard system that continuously monitors and controls the cuff pressure by measurements of CO2 levels above the cuff, and allowing simultaneous rinsing and aspiration of subglottic secretions.

Objectives: To determine the safety and clinical efficacy of AG 100 system compared with usual care in critically ill patients. Methods: Prospective, single centre, open-label, randomized, controlled feasibility and safety trial. Sixty patients, without pneumonia, were randomized to be intubated with the AG tube and connected to the system (n = 30) or with a conventional tube (n = 30) combined with subglottic secretion drainage and manually control of tracheal cuff pressure (P cuff ). Primary outcome was the rate of adverse events. Other outcomes included the rate of mechanical complications, the level of ICU staff satisfaction, the incidence of ventilator-associated pneumonia (VAP), the quality of P cuff control, and the amount of SS drained. Results: Out of 60 patients enrolled in the study, 56 were included in the analysis (28 per each group). Both groups were similar at randomization in demographic characteristics, ICU admission diagnosis, main comorbidities and severity of illness. No device-related adverse events occurred in any of the two groups. No differences were detected using AG system vs conventional tubes in terms of post- Introduction: During sepsis, intrinsic stress responses may become maladaptive and contribute to poor outcomes. Targeted intervention with β-blockade to 'de-stress' such patients may be beneficial. We developed a 72-h rodent model of fluid-resuscitated faecal peritonitis in which mortality (occurring between 18 and 42 h) can be predicted at 6 h by a low stroke volume (AUROC 0.87), and where survivors are clinically improving by study end. [1] Objectives: To investigate the impact of β-blockade on outcomes in predicted survivors and nonsurvivors of faecal peritonitis. Methods: Instrumented, fluid resuscitated, male Wistar rats (300-400 g) had sepsis induced by intraperitoneal injection of faecal slurry (8.5 ml/kg). At 6 h, under brief isoflurane sedation, echocardiography was performed to differentiate predicted survivors from nonsurvivors based on a stroke volume cut-off of 0.20 ml. Rats in each prognostic group were then randomised to receive either esmolol (500 μg/kg over 1 min followed by 75 μg/kg/min infusion) or matching placebo (0.9 % NaCl) until 24 h. Animals were observed for up to 72 h, and time of death was recorded. The study was powered to detect a mortality reduction in predicted nonsurvivors from 90 % to 45 % with esmolol, with a power of 0.80 and type-1 error of 0.05. Results: 64 rats were randomised after prognostication to receive either esmolol or placebo. At 6 h, predicted survivors and nonsurvivors were clinically indistinguishable (both groups appeared only mildly unwell), though predicted nonsurvivors (stroke volume < 0.20 ml) had lower cardiac output (86 ± 9 vs. 130 ± 25 ml/min), higher heart rate (487 ± 36 vs. 459 ± 24 bpm) and blood pressure (131 ± 12 vs. 125 ± 10 mmHg) and more haemoconcentration (haemoglobin 16.9 ± 1.6 vs. 15.0 ± 1.3 g/dl) (all p < 0.05). Survival was significantly improved by esmolol in predicted nonsurvivors (p = 0.004), but worsened in predicted survivors (p = 0.06).

Conclusions: Mortality was approximately halved in predicted nonsurvivors by esmolol, but doubled in predicted survivors. Early prognostication appears key in identifying the subset(s) of animals (and, potentially, patients) who might benefit from additional treatment, while avoiding iatrogenic harm in those that would naturally survive. Mechanisms by which esmolol impact upon mortality are under investigation. Introduction: Endotoxins (lipopolysaccharides, LPS) have become interesting targets in extracorporeal therapies. LPS is a major constituent of the outer cell wall of gram-negative bacteria and strongly triggers inflammatory responses in humans at concentrations as low as 1 ng/kg body weight. Although the elimination of LPS is promising for the supportive therapy of sepsis and liver failure, endotoxin neutralization using endotoxin adsorbents is controversial.

Objectives: We could recently show that endotoxin inactivation by low-dose Polymyxin B (PMB; 200 ng/ml) could be applied for endotoxin inactivation in blood [1] . Aim of this study was to establish an adsorbent-based system which combines constant PMB release for endotoxin inactivation and effective cytokine adsorption during extracorporeal treatment. Methods: We established an adsorbent-based PMB release system which ensures a constant PMB level in plasma during extracorporeal therapies. A polystyrene-divinylbenzene based cytokine adsorbent (CG161c) with nanostructured pores was coated with a defined amount of PMB by hydrophobic interactions. The endotoxin inactivation and cytokine adsorption was tested in an in vitro model using fresh donated blood which was stimulated with 1 ng/ml lipopolysaccaride from E. coli. Results: In plasma or blood an equilibration between the free and bound form of PMB will lead to a constant PMB level in plasma. The PMB release experiments in plasma clearly show that the adsorption and desorption is a function of the ratio PMB concentration: adsorbent surface. Furthermore the PMB release depends on the protein concentration of the plasma. It makes a big difference whether the PMB coated adsorbent is used in plasma or in fractionated plasma where the hydrophobicity is much lower. The experiments suggest that the PMB coating of the CG161c adsorbent doesn´t influence the cytokine removal which can take place in parallel. The ability of LPS inactivation by the PMB coated CG161c adsorbents was similar to PMB which was infused directly into the plasma. Conclusions: Our in vitro model shows that the combination of cytokine removal and controlled PMB release by the same adsorbent results in a strong suppression of inflammatory effects in blood. Objective: Management of hemodynamically stable pulmonary embolism (PE) with right ventricular (RV) dysfunction is still controversial. The objective of our study is to evaluate the effectiveness of local intraarterial thrombolysis (LIT) in this group of patients and analyze its complications. Patients and methods: Prospective study (January 2008-December 2105). Patients included had been diagnosed of PE by computed tomography (CT), were hemodynamically stable [systolic arterial pressure (SAP) > 90 mmHg] and had a clinical suspicion of RV dysfunction (biventricular quotient in CT > 1 or elevated levels of troponin I), that was confirmed afterwards by the presence of at least one of the following findings in the echocardiogram: subjective alteration of RV contractility, RV basal diameter (four chamber view) > 40 mm, tricuspid annular plane systolic excursion (TAPSE) < 15 mm or estimated systolic pulmonary artery pressure (SPAP) > 30 mmHg. LIT was done with a urokinase infusion (bolus dose of 200.000 UI followed by a perfusion of 100.000 UI/h) administered thru a pulmonary artery catheter, placed with radiological guidance, using an antecubital puncture. Patients received simultaneous systemic anticoagulation with unfractionated heparin. After 48-72 h of treatment, and before ending the urokinase infusion, a radiological control was done using angiography or CT. Within the seven days after LIT, patients underwent a follow-up echocardiogram. Statistical analysis was performed with Student´s T test for parametric paired data, Wilcoxon´s Test for non parametric and Stuart-Maxwell for qualitative values. Results: Eighty-seven patients were included and their general data are detailed in Fig. 32 . Mean treatment time was 56,3 ± 15,5 h. Ninety percent of patients experienced a radiological improvement (50.5 % a complete/almost complete resolution and, 37.8 % a significant improvement). Only 5,7 % didn´t improve radiologically. The evolutions of the different RV parameters studied are shown in Fig. 33 . Minimum fibrinogen and platelet values where 251,9 ± 91,8 mg/dl and 134 x 10 3 ± 44.6 x 10 3 cells/mm 3 . Eighteen patients (20,7 %) suffered form hemorrhagic complications that, in 13 cases, where puncture site hematomas and, in six occasions (6,9 %) required an early interruption of the treatment. Three patients (3,4 %) received a blood cell transfusion of ≤ 2 blood units. Mean ICU and hospital stays where 4 ± 1,5 and 14 ± 9,2 days. All patients survived. Conclusion: In our group of patients, LIT rapidly improved the function and decreased the hemodynamic strain of the RV, while being associated with a low incidence of major complications. Introduction: Atrial fibrillation (AFib) is associated with higher shortterm mortality in critical illness, but it is still uncertain whether AFib independently contributes to unfavorable outcome.

Objectives: The aim of this study was to test the hypothesis that AFib during critical illness is independently associated with increased in-hospital and long-term risk of death.

Methods: The FROG-ICU study was a prospective, observational, multicenter cohort study designed to investigate outcome of critically ill patients. Heart rhythm was assessed at inclusion and during ICU stay with digital ECG recordings. Among patients who had any AFib during ICU stay, newonset and recurrent AFib were diagnosed in patients without and with previous history of AFib, respectively. Primary endpoints were in-hospital and 1-year mortality. Covariate adjusted logistic regression models and Cox proportional hazards models were used to evaluate the association between AFib and in-hospital mortality or 1-year mortality, respectively. In-hospital mortality was adjusted for 7 independent covariates (age, gender, Simplified Acute Physiology Score (SAPS II), treatment with inotropes or vasopressors, serum lactate level, high-sensitive troponin I, B-type natriuretic peptide), 1-year mortality was adjusted for 8 covariates (age, gender, SAPS II, history of congestive heart failure, treatment with inotropes or vasopressors, serum lactate level, C-reactive protein and serum creatinine).

Results: The study included 2087 critically ill patients. The study population consisted of 1841 patients for whom data about heart rhythm during ICU stay was available. AFib occurred in 343 patients (19 %). Newonset AFib (n = 212) had higher in-hospital mortality (47 %) compared to no AFib (23 %, P < 0.001) or recurrent AFib (34 %, P = 0.032). Newonset AFib showed increased in-hospital risk of death after multivariable adjustment compared to no AFib ( Introduction: The incidence of the supraventricular arrhythmias is increased in septic shock patient, and it is associated with worse short and long term prognosis.

Objective: To test that propafenon could be a feasible antiarrhythmic in the absence of contraindications. Methods: Patients with septic shock who received antiarrhythmic drugs for supraventricular arrhythmias were included over 24 months. The patients were divided into the three groups according to antiarrhythmic agent: amiodarone (Group1), propafenon (Group2) and metoprolol (Group3). In the first 24 h the type of arrhythmia, dosages, cardioversion rates, demographic, haemodynamic, laboratory parameters were recorded. Mortality was compared between the groups and between the cardioverted vs those remaining in acute and chronic arrhythmias. clinical studies; presumably due to an impairment of myocardial oxygenation and ventricular filling. A randomised control trial of heart rate (HR) control in septic shock showed an increase of survival for the patients receiving esmolol 1 . An animal study observed a similar improvement of survival and an increase in left ventricular (LV) contractility when esmolol was associated with norepinephrine (NE) 2 . However beta-blockers therapy in sepsis is still debated considering its negative inotropic side effect. Ivabradine, a pure bradycardic agent, blocking selectively the If channels in the sinus node, could represent a safer option for HR control.

Objectives: Compare the hemodynamic tolerance of HR control either with intravenous (IV) ivabradine or esmolol perfusion, in a large animal model of septic shock. Methods: We used a closed chest swine model of fecal peritonitis. Analgesia and sedation were provided by sufentanil and sevoflurane. Hemodynamic monitoring included arterial blood pressure (ABP); continuous cardiac output (CCO); LV maximum rate of pressure (dP/dTmax) and LV elastance (E-LV); mixed venous oxygen saturation (SVO2) and arterial lactate (Lac). After the development of septic shock, fluid resuscitation was started and animals were randomised in 3 groups of 6 pigs: ivabradine (IVB), esmolol (ESM) or control. Ivabradine was administered with an IV bolus of 0,1 mg/kg that could be repeated at 0,3 mg/kg, aiming an HR between 80 and 90 beats per minute (BPM). Continuous IV perfusion of esmolol was started at 1 mg/kg/h and adapted to reach the same HR range. After 5 hours of HR control, a fixed dose of 0,3 mcg/kg/ min NE was introduced in all groups. Results: All animals developed an hyperdynamic distributive shock, including tachycardia above 110 bpm. HR control between 80 and 90 bpm was successful in both IVB and ESM groups. IVB administration didn't affect ABP, CCO, dP/dTmax, E-LV, SVO2 or Lac. ESM perfusion tended to decrease ABP, CCO and SVO2; E-LV and Lac were unaffected but dP/ dTmax decreased markedly. Under NE perfusion, E-LV was similar in all groups but dP/dTmax was lower in ESM group.

Conclusions: In septic shock, HR control with an IV administration of ivabradine doesn´t alter global organs perfusion and cardiac function. Esmolol perfusion, in order to achieve the same goal, reduces LV dP/dTmax and didn´t enhance LV contractility in association with NE. Introduction: Patients in critical care settings are often at risk of developing hypotension, which can lead to poor outcomes such as increased morbidity and mortality. Current hemodynamic parameters for monitoring such hypotension often exhibit pronounced changes only when the hypotensive event is already occurring or when it is too late. We have developed a hypotension probability indicator (HPI™) to predict hypotensive episodes based on machine learning techniques. The HPI™ model was trained on~3000 ICU and OR patients. The objective of this study is two-fold: 1) To test the accuracy of HPI™ to predict events on a completely independent test data set of ICU patients, not used in the development of the algorithm; and 2) To compare timing of interventions in response to an event to the timing of detection of an event by HPI™.

Methods: Data used in this study came from the MIMIC II MIT Research Database. Arterial pressure waveforms of 326 patients were analyzed for HPI™ and then tested for event detection and prediction accuracy. All features of the HPI™ as well as other hemodynamic parameters for comparison were calculated using FloTrac (Edwards Lifesciences, Irvine, CA). A hypotensive event was defined as any time period where MAP < 65 mmHg for at least 1 minute. An ROC analysis was performed to assess AUC, sensitivity, and specificity of the HPI™ to identify an event during the event, and 5, 10, and 15 minutes prior to the start of event. Next, clinical records of the 326 patients were reviewed for any drug or fluid interventions during start of event to 5 minutes after an event and the elapsed time from start of event to intervention time was calculated. A drug or fluid intervention was defined as any bolus or IV infusion start. In addition, the time at which HPI™ probability of event > 0.85 prior to the start of an event was also calculated for comparison. Data are presented in median [25-75 th percentiles]. Conclusion: In conclusion, HPI™ can accurately detect an event up to 15 minutes prior. HPI™ may serve as a useful addition in the care of critically ill patients by potentially facilitating earlier intervention either in response to an event or serve as a decision support and direct a physician's attention to potential oncoming events when HPI™ is high. This statement is valid for both in-hospital as well as out-of-hospital cardiac arrest. Regardless of the location of the cardiac arrest, there are at least four factors that appear to be of major importance for survival. The first is the time from collapse to delivery of treatment; the second is the quality of cardiopulmonary resuscitation (CPR); the third is the patient's co-morbidity and the fourth is the aetiology of the CA and the presenting rhythm. The present study will focus on the first three parts of the chain of survival, time from collapse to call/CPR/defibrillation. Objectives: To describe the number of survivors following inhospital cardiac arrest (IHCA) in Sweden during one year and, based on estimations and assumptions, calculate the potential number of additional lives saved following improvements in the chain of survival. There was a strong inverse relation between delay to call for the rescue team and delay to treatment and survival. If delay from collapse to a/call and, b/start of CPR were reduced to < 1 minute in patients with a longer delay than that and if c/time from collapse to defibrillation was reduced to < 3 minutes among those with a longer delay than that: a/57; b/32; and c/35 further lives could potentially be saved. We speculate that about 100 additional lives (one per 200 hospital beds each year) could theoretically be saved by improved adherence to guidelines regarding the first three components in the chain of survival in Swedish hospitals yearly.

Conclusions: In 2014, approximately 900 patients (four per 100 hospital beds) were successfully resuscitated following IHCA in Sweden.

There was a strong negative relation between collapse and call for rescue team/CPR/defibrillation and 30-day survival. With reduced delay times a further 100 lives (one per 200 hospital beds) could theoretically be saved each year in Sweden.

The study was supported by grants from the Laerdal Foundation of Acute Medicine in Norway (JH) and the Scientific Council of Halland (FH).

Prophylactic versus clinically-driven antibiotics in comatose survivors of out-of-hospital cardiac arrest -a pilot study S. Results: Proportion of patients on antibiotics was significantly greater from day 1 to 5 in prophylactic group while there was no difference on days 6 to 7. Peak C-reactive protein in prophylactic group was significantly smaller (186 ± 61 vs. 229 ± 60 mg/L; p = 0.04). There was no difference in peak white blood cell count (14.6 ± 6.6 vs. 16.6 ± 6.2; p = 0.24), procalcitonin (4.02 ± 10.12 vs. 4.84 ± 8.5 microg/L; p = 0.80) and CD 64. Except for positive mini BAL on day 3 (7 % vs. 42 %; p < 0.01), there was no significant impact on other microbiological samples and X-ray signs of pneumonia (50 % in each group). Use vasopressors/inotropes (93 % in each groups), duration of mechanical ventilation (5.4 ± 3.7 vs. 5.2 ± 3.1 days), tracheal intubation (6.5 ± 4.6 vs. 5.9 ± 4.3 days), ICU stay (7.7 ± 5.2 vs. 6.9 ± 4.5 days), survival (73 % vs. 73 %) and survival with good neurological outcome (50 % vs. 40 %) were also comparable. Conclusions: Tracheobronchial aspiration was documented in more than a quarter of comatose survivors of OHCA using bronchoscopy on admission. In the absence of aspiration, prophylactic antibiotics reduced peak CRP and the incidence of positive mini-BAL on day 3 and had no significant impact on other Introduction: Survival to discharge after in-hospital cardiac arrest (IHCA) is poor (10 − 20 %) and has not improved despite developments in modern medicine. 1 Data on the aetiology of in-hospital cardiac arrests is very limited, and conducted studies include IHCA patients resuscitated in emergency departments, intensive care units and high dependency units. Objectives: To determine the underlying causes of IHCAs occurring on general wards and investigate, whether the aetiology is independently associated with six months survival. Methods: A prospective observational study between 2009-2011 in a Finnish university hospital. We included all adult IHCA patients on general wards who were attended by ICU´s medical emergency team. Definite aetiology was determined from the autopsy records and medical records. No autopsies were conducted solely for study purposes. The local Ethics Committee approved the study protocol (Approval no: R08116). Results: The cohort consisted of 279 patients, of which 185 (66 %) were male. Median age of the patients was 72 (64, 80) years. Altogether 178 (64 %) IHCAs were monitored/witnessed, first rhythm was shockable in 42 (15 %) cases and 53 (19 %) patients survived six months. Autopsy was conducted in 153 (55 %) cases. Aetiology was determined as cardiac in 141 events, 73 of which were due to acute myocardial infarction and 26 due to acute myocardial ischaemia without infarction. Congestive heart failure was the third most prevalent reason in cardiac sub cohort (16). Altogether 138 IHCAs were considered non-cardiac; most common causes were pneumonia (39), exsanguination (16), pulmonary embolism (12) and peritonitis (11). Cardiac IHCAs were more commonly preceded by subjective symptoms (e.g. chest pain, respiratory distress) than non-cardiac IHCAs (47 % vs. 32 %, p = 0.022), while objective vital dysfunctions preceded IHCAs as often in both sub cohorts (40 % vs. 44 %, p = 0.448). In a multivariate logistic regression model monitored/witnessed event, shockable primary rhythm and low age-adjusted Charlson comorbidity index score were factors independently associated with 180-day survival, but the aetiology (cardiac vs. non-cardiac) was not. Conclusions: Aetiology of IHCAs on general wards is cardiac in 50 % of the events. Ischaemic reasons for IHCAs were twice as common as shockable primary rhythms in this study. Subjective symptoms and objective vital dysfunctions often precede general ward IHCAs. However, neither the aetiology nor the presence of antecedents, but low comorbidity, observed arrest and shockable primary rhythm are factors associated with a favorable outcome.

Reducing in-hospital cardiac arrest by implementation of innovative early warning information system in a tertiary medical center Introduction: In-hospital cardiac arrest (IHCA) is a common and high-risk issue with less than 20 % surviving to hospital discharge. Most patients show signs of clinical deterioration in the hours before IHCA. As a result, the development of vital signbased early warning system was designed to detect early signs of clinical deterioration before IHCA attack in order to trigger early intensive care. Objectives: In this study, we investigate the impact of the implementation of an innovative early warning information system on the rate of IHCA and survival rate in IHCA patients. Methods: A multidisciplinary team among intensivists, cardiologists, emergency physicians, and nursing staffs in a tertiary medical center was organized since May 2015. The key interventions include automatic national early warning score (NEWS) calculating information system, nurses and physicians computer-based reminding alarm if NEWS ≥ 7 or more than highest scores among previous 3 measurements, real time early warning screen saver and electric board, in service education and early warning monitor team. All patients admitted between January 2013 and January 2016 were enrolled. Total 143,450 patients were divided into three groups: pre-interventional group from Jan 2013 to April 2015 (n = 107,437), Interventional group from May to June 2015 (n = 7,923) and post-interventional group from July 2015 to Jan 2016 (n = 28,090). The definition of In-hospital cardiac arrest is the number of In-hospital cardiac arrest per thousand admitted patients. We compared the rates of IHCA, 48 hours survival rate and discharge survival rate in IHCA patients among these 3 groups.

Results: The rate of In-hospital cardiac arrest improved from 2.53‰ in pre-interventional group, to 2.15‰ in interventional group and to 1.32‰ in post-interventional group (p < 0.05). The 48 hours survival rate in IHCA patients increased from 34.4 % in pre-interventional group, to 41.2 % in interventional group and to 45.4 % in postinterventional group (p < 0.05). The discharge survival rate in IHCA patients also increased from 15.6 % in pre-interventional group, to 29.4 % in interventional group and to 36.4 % in post-interventional group (p < 0.05). Conclusions: The study demonstrated that implementation of early warning information system and innovative strategies could attenuate the rate of IHCA, 48 hours survival rate and discharge survival rate in IHCA patients. Introduction: Although prolonged unconsciousness after cardiac arrest (CA) is a sign of poor neurological outcome, limited evidence shows that a late recovery may occur in a minority of patients. Objectives: We investigated the prevalence and the predictive factors of delayed awakening in comatose CA survivors treated with targeted temperature management (TTM). Methods: Retrospective analysis of the Parisian Region Out-of-Hospital CA Registry (2008-2013). In adult comatose CA survivors treated with TTM, sedated with midazolam and fentanyl, time to awakening was measured starting from discontinuation of sedation at the end of rewarming. Awakening was defined as delayed when it occurred after more than 48 h. Results: A total of 326 patients (71 % male, mean age 59 ± 16 years) were included, among whom 194 awoke. Delayed awakening occurred in 56/194 (29 %) patients, at a median time of 93 h (IQR 70-117) from discontinuation of sedation. In 5/56 (9 %) late awakeners, pupillary reflex and motor response were both absent 48 h after sedation discontinuation. In multivariate analysis, age over 59 years (OR 2.1, 95 % CI 1.0-4.3), postresuscitation shock (OR 2.6 [1.3-5.2]), and renal insufficiency at admission (OR 3.1 [1.4-6.8]) were associated with significantly higher rates of delayed awakening. Conclusions: Delayed awakening is common among patients recovering from coma after CA. Renal insufficiency, older age, and postresuscitation shock were independent predictors of delayed awakening. Presence of unfavorable neurological signs at 48 h after rewarming from TTM and discontinuation of sedation did not rule out recovery of consciousness in late awakeners. Grant acknowledgment None Note: This abstract has been previously published and is available at [1] . It is included here as a complete record of the abstracts from the conference. Introduction: Viral infections play a key role in preventable deaths of children globally, and can be antecedents to bacterial pneumonia and sepsis. Diagnosis of viral infection is often problematic due to non-specific clinical presentation. We developed a host immune response gene expression signature to distinguish systemic inflammation due to viral infection vs. bacterial or noninfectious causes. Objectives: To define and validate the host immune response gene expression signature against multiple independent datasets. Methods: Four public GEO datasets describing transcriptomic responses to viral infection were used to identify biomarkers, ranked by AUC, which could separate affected from unaffected subjects. Biomarkers that also responded (AUC > 0.80) to non-viral causes of systemic inflammation were removed. Remaining biomarkers were then ranked for performance in 10 other GEO transcriptomic datasets for viral infection; those with mean AUC >0.75 were retained. Next, a greedy search was applied to the merged (4 + 10) viral GEO datasets to identify the best combinations of biomarkers for discrimination of viral infection. The signature was then validated using independent datasets. Results: A 4-gene signature (comprised of ISG15, IL16, OASL, ADGRE5) had AUC 0.94 across the merged (4 + 10) viral GEO datasets. This signature was validated in 11 additional GEO datasets covering a wide variety of viral pathogens including a time-course study of respiratory syncytial virus (RSV) in children (Fig 36) , and in two independent datasets of our own: adults from the emergency department (Fig 37 Introduction: Using a tourniquet to temporary cut off blood supply to the arm (Remote Ischemic Preconditioning -RIPC) has been shown to result in myocardial protection and reduced incidence of AKI in patients undergoing cardiac surgery. However, a recently performed large multi-center trial in CABG patients showed no beneficial effects on clinically relevant endpoints [1] . Animal studies have shown an`early window of protection' in the 1-2 hours after RIPC as well as a`late window of protection`12-24 hours after RIPC. Several mechanisms have been suggested to mediate the protective effects of RIPC, of which attenuation of the immune response is an important candidate, although this has hitherto also only been shown in animal studies [2] . Objectives: To determine the effect of single and repeated RIPC, thereby investigating both the early and late windows of protection, on the inflammatory response during endotoxemia, a standardized, controlled model of systemic inflammation in humans in vivo. Methods: We performed a randomized controlled study in 30 healthy non-smoking male volunteers. Subjects were assigned to either the single-dose RIPC group, multiple-dose RIPC group, or the control group (n = 10 per group). The single-dose RIPC group received 1 dose of RIPC, consisting of 4 cycles of 5-minute ischemia of the arm followed by 5 minutes of reperfusion just before administration of 2 ng/kg lipopolysaccharide (LPS). The multiple-dose RIPC group received one dose of RIPC per day on the 6 days before the endotoxemia experiment day, and 1 dose just before LPS administration. Results: LPS administration resulted in a typical increase in body temperature, flu-like symptoms, and hemodynamic changes, with no differences between groups. Administration of LPS resulted in a sharp increase in plasma levels of the proinflammatory cytokines TNF-α, IL-6, and IL-8 as well as the antiinflammatory cytokine IL-10. No differences in plasma levels of these cytokines were observed between the different groups ( Figure 39 ). Conclusions: In the present study, we demonstrate that RIPC does not affect the in vivo inflammatory response induced by administration of endotoxin in humans. These results implicate that RIPC does not exert direct anti-inflammatory effects and that the previously observed protective effects are mediated through other mechanisms. Furthermore, the absence of immunomodulatory effects of RIPC in the present study tempers expectations of using RIPC as an immunomodulatory treatment strategy in patients. Introduction: Sepsis-induced immune alterations are associated with secondary infections and increased risk of death (1). Mesenchymal Stem Cells (MSCs) have been described as a novel therapeutic strategy for the treatment of diseases related to inflammation and tissue injury with their potent modulatory effects on immune system (2) . Objectives: In this study, we evaluated the immune-modulatory effects of Human Dental Follicle Mesenchymal Stem Cells (HD-MSCs) on lymphocytes which are isolated from peripheral blood samples of sepsis and septic shock patients.

Methods: According to the International Sepsis Definitions Conference (3), patients divided into two groups as sepsis (Group I, n = 10) and septic shock (Group II, n = 10). Peripheral Blood Mononuclear Cells (PBMCs) were isolated from venous blood samples of Group I, Group II and healthy subjects named as Group III, n = 10. Anti-CD3/CD28 PBMCs were co-cultured with DF-MSCs, IFN-g stimulated DF-MSCs and with no MSCs about 72 hour. CD4 + CD25 + FoxP3+ T cells levels (Treg), lymphocyte proliferation and apoptosis were evaluated with the flow cytometry.

Results: DF-MSCs and IFN-g induced DF-MSCs cultures significantly supressed proliferation in sepsis group when compare to septic shock group(p < 0,005).

Conclusions: MSCs demonstrate their effects on immune system by increasing the number and activity of regulatory T cells (Treg) (4) .In our study, MSCs suppressed lymphocyte proliferation and apoptosis but increased the rate of Treg cells in sepsis cocultures. This effect was more obvious with IFN -g stimulation. These responses were not seen in septic shock patients´blood samples and might be explained with anergy. Our findings revealed that DF-MSCs application has immunoregulatory effects in sepsis. This approach opened a new area to work how will MSCs be used to reduce organ dysfunctions and mortality in the clinical practice. Introduction: Inhibition of mitochondrial Complex I is described in human and animal sepsis. 1,2 This may be responsible, at least in part, for the decrease in mitochondrial functionality seen in sepsis. We have recently demonstrated that the mitochondrial uncoupling agent, dinitrophenol (DNP) failed to increase body temperature and oxygen consumption (VO 2 ) in septic rats, as was seen in healthy controls. This suggests that uncoupling is active in sepsis and can contribute to fever. We further postulated that the blunted effects of DNP in sepsis may be related in part to upstream mitochondrial inhibition.

Objectives: To determine if complex I inhibition by metformin in healthy rats can prevent the increment in temperature and oxygen consumption (VO 2 ) by DNP, and thus mimic the pattern seen in sepsis.

Methods: VO 2 was measured in awake, cannulated male Wistar rats (approx 300 g body weight) in metabolic cages (Oxymax, Columbus Instruments). Sepsis was induced with an intraperitoneal injection of faecal slurry at time 0. Sham control animals received no slurry. Fluid resuscitation (10 ml/kg/h crystalloid) was started at 2 hours and continued throughout the whole experiment. Half the septic and sham animals were treated with an IV infusion of metformin (186 mg/kg) between hours 2-6. At 6 and 24 hours, all animals received iv DNP (30 mg/kg). Arterial blood gases, echocardiography and core temperature were measured at times 0, 6 and 8, and 24 and 26 hours (i.e. before and after the two doses of DNP). Mean arterial pressure was recorded continuously. Wilcoxon Rank Sum test was used to compare groups and two-way ANOVA to compare changes in continuous variables from baseline between groups. p values < 0.05 were considered statistically significant.

Results: Pretreatment with metformin completely prevented the increase in temperature and VO 2 induced by DNP in sham animals at 6 hours and reflected that seen in non-metformin treated septic rats ( Figure 43 ). The reduction in myocardial contractility (stroke volume and Vmax) seen in the septic animals treated with DNP was prevented by Complex I inhibition at 24 h. Metformin was metabolically well tolerated, with no increase in blood lactate.

Conclusions: Inhibiting complex I with metformin prevents the uncoupling effect of DNP in sham animals. This mimics the pattern seen in septic animals and confirms that both Complex I inhibition and pre-existing mitochondrial uncoupling could be active in septic rats. Objectives: The inflammasome is a multiprotein complex that stimulates cytokines release such as interleukin-1β (IL-1β) and IL-18, involved in the inflammatory response. Our aim is to quantify the state of activation of the inflammasome complex in septic patients, as well as to study possible differences in the cytokines levels in sepsis and septic shock, its temporary evolution, and its prognostic value. Methods: Prospective study including patients admitted to the ICU with sepsis or septic shock during 15 months. On days 1, 3 and 7, IL 1-β serum levels and real-time expression of NLRP3 inflammasome (nucleotide-binding oligomerization domain, leucine rich repeat domain containing protein and Pyrin) were determined by Elisa and Real time-PCR respectively. Demographic variables, severity scores on ICU admission (APACHE II and SOFA), sepsis focus and mortality were collected. Statistical analysis: T-Student, Kruskal-Wallis and U-Mann-Whitney test as appropriate.

Results: There were included 31 patients (severe sepsis 14 and septic shock 18). Overall mortality was 29 % (9 patients). The levels of IL-1β on day 1 (16.5 ± 2.6 vs 14.1 ± 2.4 pg/mL; p < 0.05) and NLRP3inflammasome (8.5 ± 1.3 vs 7.5 ± 1 mRNA arbitrary units; p < 0.05) were significantly higher in septic shock patients than in sepsis, with no differences in the following days set (3 and 7). The IL-1β and NLRP3 inflammasome levels decreased significantly on days 3 and 7 compared to first day (p < 0.001), without differences between survivors and deceased patients.

Conclusions: In septic patients, inflammasome activation complex occurs, with higher levels detected in septic shock. Decreased levels of IL-1β and NLRP3 inflammasome in septic process have been observed during evolution, actually without relation with mortality. Introduction: According the consensus conference on weaning from mechanical ventilation, intubated patients should pass a spontaneous breathing trial (SBT) to assess their readiness to be extubated. Objectives: To characterize patients who are extubated without any SBT and to compare them to patients who had at least 1 SBT during their weaning period. Methods: The prospective multicentre observational WIND (Weaning accordIng New Definition) study was performed from April to August 2013. Ventilation and weaning modalities were daily assessed until discharge in all intubated patients admitted to the participating ICUs. We defined 1) weaning attempt (WA) as a spontaneous breathing trial (SBT) or an extubation (with or without SBT), 2) successful weaning as an extubation without death or invasive mechanical ventilation within 7 days. Variables are presented as mean ± standard deviation, median [interquartile range] or number (percentage). Comparisons were made using Chi2 test, exact Fisher tests, Student t-test or Wilcoxon rank sum test as appropriate. All statistical tests were two-sided and P value ≤ 0.05 were considered significant. Results: Among the 2729 patients included, 2051 patients had at least 1 WA comprising 1669 patients whose first WA was a SBT and 382 who had another type of first WA. These 382 patients with no SBT had a total of 454 WA: 252 (55.6 %) planned extubation without SBT, 124 self-extubations (27.3 %), 30 WA while tracheostomized (6.6 %) and 48 SBT after their first WA (10.6 %). The majority of patients with self-extubation had a successful weaning not requiring reintubation (75.8 %). Almost a quarter (n = 95) of the patients who were extubated without any SBT had a decision of withholding or withdrawing invasive mechanical ventilation, representing 89.7 % (N = 78) of the 87 deceased patients. We then excluded patients with a decision of limitation and patients with a self extubation to compare patients who had a planned extubation with or without SBT as first WA (Table 22) . Patients with no SBT were younger, less severe and were more often admitted for unplanned surgery: they had an easier weaning with a lower (but non significative) rate of reintubation, a shorter duration of invasive mechanical ventilation and a shorter length of stay in the ICU. Conclusion: Patients who are extubated without SBT seem to belong to three different groups: self-extubation, terminal extubation and patients in whom physicians anticipate an uneventful weaning and extubation. Among the patients with a planned extubation and without any limitation decision, clinical judgment regarding weanability appears to be effective as this group of patients had a good outcome with a low reintubation rate.

This study benefited of a grant of the non-profit Association Départementale des Insuffisants Respiratoires (ADIR) of the Haute Normandie, France Introduction: Decrease in diaphragmatic maximal relaxation rate (MRR) occurs early in the process of diaphragmatic fatigue and well before the diaphragm fails as a force generator; its measurement would, therefore, be especially valuable in ICU patients during a weaning trial. However, the use of oesophageal pressure catheters for that purpose impedes wide clinical use. On the contrary, M-mode sonography, allows non-invasive, real-time measurement of the speed of the diaphragmatic motion.

Objective: Purpose of our study was to investigate a possible correlation between diaphragmatic MRR traditionally acquired with transdiaphragmatic pressure (Pdi) catheters (MRR-Pdi) and an ECHO equivalent MRR (MRR-ECHO) acquired during different breathing conditions. Methods: The slope of MRR was measured from the initial steepest part of the descending Pdi curve simultaneously with the slope of the initial steepest descending part of diaphragmatic excursion with M-mode sonography. The protocol entrained four consecutive stages: i) breathing spontaneously during T-piece trial, ii) breathing spontaneously with performance of sniff-like maneuvers, iii) breathing with resistances of 40cmH 2 O/L, and iv) breathing with resistances of 40cmH 2 O/L with performance of sniff-like maneuvers. Statistical comparisons between slope recordings from the two methods were performed with Pearson correlation, while Bland and Altman plots were obtained in order to demonstrate reliable agreement between methods at each different breathing condition.

Results: A total of 512 separate breaths during the four previously reported breathing conditions from six ICU patients were recorded. Table 23 summarizes the slopes measured from MRR-Pdi and MRR-ECHO as means ± standard deviations (SD), and their linear correlations with p values. Statistical significant correlations were observed in all four stages; i) Pearson correlation coefficient r = 0.474, p < 0.001, R 2 = 0.2247, ii) r = 0.834, p < 0.001, R 2 = 0.6950, iii) r = 0.653, p < 0.001, R 2 = 0.4269, and iv) r = 0.794, p < 0.001, R 2 = 0.6302. Bland and Altman plots demonstrating differences of measurements against means, as well as confidence intervals (means of differences ± 2SD) were obtained for each breathing condition. Graph 1 represents the Bland and Altman plot for spontaneous breathing with sniff-like maneuvers without resistances. High R 2 indexes, indicating high agreement between the two methods were noted: i) 0.9915, ii) 0.9986, iii) 0.9941, and iv) 0.9990.

The results of our study suggest a statistical significant correlation and reliability between diaphragmatic MRR measured from Pdi tracings and the assumed diaphragmatic relaxation rate calculated from simultaneous M-mode sonographic recordings. Clinical studies are required to confirm the potential of this non-invasive index of diaphragmatic MRR to be used as a predictor for weaning success.

Grant acknolwedgement None declared. Introduction: High flow nasal cannula oxygen therapy (HF OXY) has been recently shown to decrease re-intubation rate, as compared with low flow oxygen therapy (LF OXY). [1] , [2] Objectives: To assess the effects of HF OXY as compared with LF OXY on diaphragmatic electrical activity (Eadi), respiratory rate (RR), tidal volume (VT) and gas exchange in the post extubation period. Our hypothesis was that HF OXY, as compared with LF OXY, would improve gas exchange and decrease EAdi. Methods: 10 patients underwent a crossover study immediately after extubation. Each patient was submitted to three consecutive steps of 1 hour each, according to an ON-OFF design: 1) HF OXY; 2) LF OXY; 3) HF OXY. Oxygen fraction was maintained stable throughout the study. The Eadi was continuously monitored through Eadi cathether (Maquet, Solna Sweden). The heated and humidified HF OXY was delivered through nasal cannula at flow rates of 50-70 L/min, (F&P, Auckland New Zealand).

Results: RR remained similar throughout the study, VT was significantly higher during the LF OXY step as compared with the HF OXY steps. Oxygenation significantly improved during the HF period, whereas PaCO 2 remained unchanged throughout the study (Table  24 ). EAdi was significantly higher during LF OXY ( Figure 45 ) Conclusions: Since the EAdi is correlated to work of breathing, our physiological data suggest that HF OXY significantly reduces WOB while improving oxygenation in the post extubation period. Further studies are required to define if diaphragm unloading may explain the favourable results of HF OXY in clinical trials. Introduction: I Non-invasive mechanical ventilation (NIV) has been seen to play a major role in decreasing intubation rates in patients with severe exacerbation of chronic obstructive pulmonary disease and congestive heart failure. Unsuccessful NIV has been found to be independently associated with increased mortality in patients with ARF. The NIV failure and their impact on mortality in patients with inlfuenza infection is unknown. Objectives:

1) to describe non-invasive ventilation failure (NIVf) rate, 2) to identify risk factor for NIVf using CHAID (Chi-square Automatic Interaction Detection) and 3) to determine if NIVf is associated with ICU-mortality.

Methods: Secondary analysis in 1,898 patients with influenza requiring mechanical ventilation(MV). Three groups were considered: 1) patients with NIV who failed (Group A); 2) patients with NIV who succeeded (Group B); and 3) patients with invasive MV (Group C). Cox analysis was used to assess survival. Risk factors for NIVf were obtained using CHAID. Conclusions: NIV failure is frequent and independently associated with ICU-mortality in patients with influenza. CHAID analysis might be a promising tool to assist in clinical decision-making. Introduction: Acute kidney injury (AKI) after liver transplantation is a common complication with an incidence of approximately 50 % [1] , resulting in high morbidity and mortality. To increase the possibilities to prevent or treat AKI after liver transplantation, it is essential to increase the knowledge on changes in renal physiology after liver transplantation.

Objectives: The aim of this study was to gain insights into renal perfusion, filtration and oxygenation in the immediate postoperative period in patients undergoing liver transplantation and to compare these data to those obtained from a group of patients undergoing major surgery with no postoperative renal impairment. Methods: Informed consent was obtained preoperatively from twelve patients with normal renal function accepted for liver transplantation. Glomerular filtration rate (GFR) was measured preoperatively by plasma clearance of Cr-EDTA. The patients were studied after liver transplantation in the ICU in the immediate postoperative period, sedated and mechanically ventilated. Systemic haemodynamics and renal variables where obtained during two 30-min periods. Renal blood flow (RBF) and GFR were measured by the renal vein retrograde thermodilution technique and by renal extraction of Cr-EDTA (=filtration fraction, FF), respectively. Arterial (a) and renal vein (rv) blood samples were taken for measurements of arterial (CaO 2 ) and renal vein (CrvO 2 ) oxygen contents. Renal oxygen consumption [RVO 2 = RBF x (CaO 2 -CrvO 2 )], renal oxygen delivery (RDO 2 = RBF x CaO 2 ) and renal oxygen extraction [RO 2 Ex = (CaO 2 -CrvO 2 )/CaO 2 )] were calculated. Sixty-three patients undergoing uneventful cardiac surgery with no postoperative renal impairment served as controls.

Results: Cardiac index (65 %) and systemic oxygen delivery index (62 %) were higher and systemic vascular resistance index was lower (−38 %) in the liver transplant group compared to controls (p < 0.001). RBF was 17 % higher and renal vascular resistance was 16 % lower compared to controls (p < 0.05). In the liver transplanted group, GFR was 35 % lower compared to the preoperative value (p = 0.016), accompanied by a 41 % increase in serum creatinine (p < 0.05). After surgery, when compared to controls, GFR and FF was 23 % and 40 % lower, respectively (p < 0.05, p < 0.01), and RVO 2 and RO 2 Ex were 42 % and 24 % higher, respectively, in the liver transplanted patients (p < 0.01, p < 0.05).

Conclusions: Despite the hyperdynamic systemic circulation, GFR is considerably reduced immediately after liver transplantation, most likely caused by a post-glomerular renal vasodilation decreasing upstream glomerular filtration pressure. Renal oxygenation is impaired after liver transplantation due to the high RVO 2 , which was not met by a proportional increase in RDO 2 . Introduction: Acute kidney injury is common in critically ill patients and associated with increased short and long-term mortality. Most published studies have focussed on patients with severe AKI. Little is known about the long-term outcome of patients with less severe AKI.

Our objective was to determine the outcome of patients with different stages of AKI at 5 and 7 years after admission to the Intensive Care Unit (ICU).

We retrospectively analysed the data of all adult patients admitted to a multi-disciplinary ICU in a teaching hospital in the UK between March 2004 -May 2009. Patients with chronic dialysis dependent renal failure were excluded. Patients were categorised according to their maximum stage of AKI during stay in ICU as defined by the serum creatinine criteria of the KDIGO classification. APACHE II and SOFA scores were used to describe severity of illness on admission to ICU. In patients with >1 admission to ICU, we only included the first admission in the analysis.

Results: Data of 3094 adult patients were analysed of whom 53 % had AKI during their stay in ICU.

Patients with any degree of AKI had a higher mortality at 5 and 7 years but they were also sicker on admission to ICU.

Conclusions: Any stage of AKI during critical illness is associated with an increased risk of mortality at 5 and 7 years. Mortality is highest in patients with AKI II and III. More work is necessary to explore the relationship between AKI and long-term outcome and to identify independent risk factors for mortality. Introduction and objective: Observational studies of intensive care unit (ICU) patients with acute kidney injury have shown a negative correlation between accumulation of fluids and survival [1] . It is unknown whether rapid removal of accumulated fluids is feasible and beneficial. Therefore we wish to perform a pilot trial of forced fluid removal vs. standard care in critically ill patients with high-risk acute kidney injury and severe fluid overload.

Methods: The FFAKI-trial is a pilot, multicenter, randomized clinical trial recruiting adult intensive care patients with high-risk acute kidney injury and fluid overload defined as > 10 % of ideal bodyweight. To reduce the signal-to-noise ratio we only wish to include patients with a high baseline risk of persistent renal failure. Baseline risk will be calculated using a newly developed model, the renal recovery score (RRS), to predict the chance of recovering renal function within 28 days. In-and exclusion criteria are shown in Tables 26 and  27 .

Patients are randomized to either forced fluid removal or standard care for the entire ICU stay. Forced fluid removal is done by infusion of furosemide and/or fluid removal with continuous renal replacement therapy. The fluid removal rate is adjusted 3 times daily to achieve a therapeutic goal of net negative fluid balance ≥ 1 ml/kg/h. Physiologic tolerance to fluid removal is continually evaluated according to predefined criteria of hypoperfusion: Lactate ≥ 4 mmol/l, mean arterial pressure < 50 mmHg or mottling beyond the edge of the kneecaps. In case of hypoperfusion, fluid removal is suspended until all criteria have been resolved for a minimum of 1 hour. The flow chart for the experimental FFAKI-treatment is seen in Figure 46 , 47, 48. The primary outcome is cumulative fluid balance 5 days after randomization. By inclusion of 50 patients we are able to detect a difference of 5.9 L between groups (α = 0.05 and β = 0.80 Renal Recovery Score ≤ 60 %.

Fluid overload defined as a positive fluid balance ≥ 10 % of ideal body weight.

Able to undergo randomization within 12 hours of fulfilling the other inclusion criteria Introduction: Enhanced recovery pathways have been a focus for patient optimisation of morbidiy and mortality in the post-operative patient. Significant mortality improvement was seen following the implementation of the emergency laparotomy pathway quality improvement care (ELPQuiC) bundle with an adjusted risk of death from 15.6 % to 9.6 % (1). The first national emergency laparotomy audit (NELA) has since been published demonstrating a 30-day mortality of 11 % and recommending access to pathways that identify need to escalate care (2) . However acute kidney injury (AKI) in critically unwell patients remains a major source of mortality, of up to 60 %, and morbidity (3). It is not yet clear whether enhanced recovery pathways, specifically those that utilise early goal directed therapy, affect the incidence of AKI.

Objectives: To determine if there was a difference in incidence of combined AKI pre and post implementation of an enhanced recovery protocol, one that had already demonstrated a significant mortality benefit.

Methods: A subgroup analysis of the data gathered via the ELPQuiC bundle was performed (1). We obtained buy-in from the 4 participating centres and requested an extrapolation of values from their raw data. If required further data was obtained via the hospital's electronic path system. All data was reviewed by a second investigator.

We defined the baseline creatinine as the best available preoperative creatinine from the past 1 year. The data recorded included creatinine at baseline, post-op, worse recorded creatinine between day1 and day30, MAKE 30, P-Possum and 30-day mortality data. CKD stage was identified via MDRD equation with age, gender and baseline creatinine. Patients with AKI were stratified according to KIDGO stages of AKIN. Primary outcome was the incidence of AKI in each of combined pre and post ELQUIC patient population. Secondary outcome included the stage specific incidence of AKI.

Results: There was no significant difference between the cumulative incidence of AKIN pre and post ELQUIC implementation on day 1 post-op (18.4 % vs 19.8 %, P = 0.6920) or day 30 post-op (14.3 % vs 8.6 %, P = 0.0647).

Conclusion: This multi-centre cohort subgroup analysis demonstrates that the implementation of a quality improvement care bundle does not affect the incidence of AKI. This is in contrast to the clear mortality benefit that such a care bundle has provided and provides stimulus to discover what factors may yet improve AKI, and so further improve these patients outcome. Introduction: It is now well documented that critically ill patients are exposed to stressful conditions and experience discomforts from multiple sources. Improved identification of the discomforts of patients in intensive care units (ICUs) may have implications for managing their care, including consideration of ethical issues, and may assist clinicians in choosing the most appropriate interventions.

Objectives: The primary objective of this study was to assess the effectiveness of a multicomponent program (MCP) of discomfort reduction in critically ill patients. The secondary objectives were to assess the sustainability of the impact of the program and the potential seasonality effect.

Methods: We conducted a multicenter, cluster-randomized, controlled, single (patient)-blind study involving 34 French adult ICUs. The experimental intervention was the implementation of the MCP including the following steps: identification of discomforts, immediate feedback to the healthcare team, and implementation of targeted interventions under control of local champions who received monthly feedback and organized monthly meetings with their healthcare team. All ICUs started with a 1-month period with no intervention, and then they were randomized to one of two groups: 17 ICUs with MCP implemented during a 6-month period (experimental group) and 17 ICUs without any programm during the same period (control group). To assess the sustainabilty of the impact of the MCP, the study was completed with a second 6-month period during which the MCP was no longer applied in the experimental group. The primary endpoint was the monthly overall score of self-reported discomfort from the French 16-item questionnaire on discomforts in ICU patients (IPREA) 1 (range from 0 to 100, the lowest possible level of discomfort to the highest). The secondary endpoints were the scores of each item of IPREA.

Results: At the end of the 6-month period, taking into account the clustering design, the monthly overall discomfort score was lower in the experimental group ( Parents were asked to consent to being contacted 12 months after discharge, at which point they were asked to complete the PedsQL, a generic measure of quality of life. The PedsQL enables a total score, physical health summary score and psychosocial health summary score to be calculated, with possible scores ranging from 0-100 and higher scores equating to better quality of life.

Results: Parents of 60 children aged 1-16.5 years (median age: 2.1 years; 37 (62 %) males), the majority of whom had had an emergency PICU admission due to sepsis (n = 25, 42 %) or respiratory problems (n = 18; 30 %), completed the PedsQL 12 months after discharge from PICU. For the group overall the total score was 78.07 (SD 20.89), physical health summary score was 81.18 (SD 21.93) and psychosocial health summary score 77.60 (SD 21.63). Babies aged 13-23 months (n = 25) had total scores (m = 85.80, SD =12.39) comparable to those of healthy norms (m = 85.55, SD = 8.74). However older children in all age groups had lower total scores than healthy norms. Whilst 25 % (6/25) of babies had scores of more than one standard deviation below the score of healthy norms, which is recognised as being of clinical significance, this rose to 54 % (7/15) of children aged 5-18 years and 60 % (12/20) of children aged 2-4 years.

Of note is that 6 children (30 %) aged 2-4 years had been admitted to PICU for reasons related to trauma or neurological concerns whereas no child aged 13-24 months had been admitted for those reasons.

Conclusions: Children who have had an emergency admission to PICU are at risk for impaired quality of life 12 months after discharge. The risk appears to be greater for children of 2 years and older which is likely to be at least partly attributable to the underlying reason for their admission. Evaluating quality of life outcomes in the longer term after PICU discharge is warranted and identification of potential risk factors will enable interventions to be targeted to optimise outcomes after an emergency admission to PICU. Introduction: Cognitive dysfunction is an important long-term complication of critical illness associated with reduced quality of life, increase in healthcare costs and institutionalization. Delirium, an acute form of brain dysfunction that is common during critical illness has been shown to be associated with long-term cognitive dysfunction(1).

Objectives: The aim of this prospective cohort study was to estimate the prevalence and severity of cognitive dysfunction in survivors of critical illness and to evaluate if delirium duration is an independent determinant of the severity of cognitive dysfunction.

Methods: Included were all adult patients admitted to a 45-bed medical surgical ICU over a 12-month period(from March 2014 to February 2015).We excluded patients with preexisting cognitive dysfunction; those that in the evaluation by the psychologist on admission to the ICU had evidence of impaired cognition through the Mini Mental State Examination and patients who could not be reliably assessed for delirium owing to blindness, deafness or language deficit and patients for whom informed consent could not be obtained. After at least 3 months of hospital discharge patients were assessed for cognition using a validated battery of tests including:1)the Digit Span, forward and backward;2) the Rey Auditory Verbal Learning Test (RAVLT);3) the Clock Drawing Test (CDT);4) the Verbal Fluency Test; and the Mini Mental State Examination. We classified patients as having mild or moderate impairment if they had either two cognitive test scores 1.5 standard deviation (SD) below the mean or one cognitive test score 2 SD below the mean; we classified patients as having severe cognitive impairment if they had 3 or more cognitive test scores 1.5 SD below the mean or two or more cognitive test scores 2 SD below the mean. Results: Enrolled in the clinical trial were 724 patients and 667 patients were eligible for the cohort (Fig. 52) . Four hundred and thirteen patients were tested 11 (03-18) months after discharge. Table 29 shows demographic and clinical data of these patients.

Cognitive impairment was identified in 206(49.9 %) patients; 120 (29.1 %) had mild or moderate and 86(20 %) severe cognitive dysfunction (Table 30) . Eleven(34.3 %) patients with delirium for 3 days or more presented severe cognitive dysfunction. In logistic regression analysis the duration of delirium for 3 days or more was not an independent predictor of cognitive dysfunction(p = 0.76).

Conclusions: This investigation in an unselected population of critically ill medical and surgical patients demonstrates that cognitive dysfunction is a frequent and severe long-term complication in survivors of critical illness. On the other hand, unlike other studies we couldn't demonstrate that the duration of delirium is an independent determinant of cognitive impairment. Table 32 Positive determinants of the evolution of the EQ-index were time and admission Glasgow score (p 0.032 and 0.005 respectively) while age, duration of MV and weakness were negatively associated (p 0.000, 0.014 and 0.000) ]. EQ-EVA paralleled EQ-index changes. Conclusions: After ICU discharge, patients suffered frequent longterm consequences that negatively affect their HRQOL. Alterations in mobility, daily activities and personal care exhibited the greatest deterioration. Prevalence of pain, anxiety and depression was high even before ICU admission, aggravated after 1-year post-discharge (40 % of patients) Duration of VM was the only intra-ICU variable that affected HRQOL. Pre-ICU conditions as age and the extent of neurological injury and, after ICU, time and weakness, were also independent determinants.

The present study was supported by the Argentinian Society of Critical Care (SATI) Introduction: Diffuse axonal injury (DAI) is a common event following traumatic brain injury (TBI), which is likely related to worst long term outcome. Diffusion tensor imaging (DTI), a magnetic resonance imaging (MRI) technique that investigates white matter integrity, is recognized as a useful tool to quantify DAI extent in TBI and possibly predict outcome. Few studies explored whole brain longitudinal changes of DTI-derived parameters in single subjects following TBI. Methods: Patients with severe TBI underwent brain MRI including DTI (32 directions, b = 1000, voxel size 2x2x2) 2-3 weeks and 1 year after trauma. 31 age-matched healthy controls underwent the same DTI protocol. We used region of interest (ROI) automated analysis (www.mristudio.org) covering the entire brain to quantify white matter integrity. The ROI fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) were extracted. Abnormalities were defined as DTI values more than 2 standard deviations below or above the mean values of controls for each ROI.

Results: 14 TBI patients with a median age of 34 (IQR 20-38) and a median GCS score of 5 (IQR 4-7) were included. 11 had diffuse injury according to Marshall classification. Regions with increased MD and reduced FA were more than expected in both early and late scan (p < 0.001 binomial test), while AD and RD abnormalities were less common. More than 50 % of the patients had increased MD in the early scan in the frontobasal girae, corona radiata and thalami; in late scans MD abnormalities were larger and more diffuse, affecting also all frontal and temporal girae and corpus callosum. FA was frequently reduced in the corpus callosum, internal capsule and fronto-basal girae in early scan, while in late phase reductions were similar but more widespread, also including the central girae, cerebellum and inferior longitudinal fascicles. The number of regions with abnormal MD increased over time (p < 0.01 Mann-Whitney), whereas for FA it was not statistically different.

An inverse correlation between the number of ROI with altered MD at early scan and outcome evaluated with GOSE was found (p < 0.05, Spearman r).

The present results indicate that early alterations of mean diffusivity and fractional anisotropy persist or worsen (for MD) at 1 year after TBI, suggesting an ongoing loss of white matter integrity and gliosis. The more frequently affected regions were the frontal girae, corpus callosum, corona radiata, inferior longitudinal fascicles and cerebellum. The number of ROI with early abnormal mean diffusivity is inversely correlated with outcome. All patients underwent SSEP, AEP and TMS the day before operation. After operation, all patients were delivered to ICU intubated and mechanically ventilated. 111 patients demonstrated full recovery from anesthesia with regaining consciousness, passed spontaneous breathing test (SBT) and gained 6 points on CST without deficiency. These patients had none or low level of dysphagia and were successfully extubated after operation. These patients formed 1st group. 21 patients had a neurogenic dysphagia and formed 2nd group. We performed SSEP, AEP and TMS on all patients immediately after admission to ICU. Results: We revealed no clinical or electrophysiological points that could have predicted neurogenic dysphagia before operation. In our research, we found the EP values which were different for the first group and for the second group. The AEP and TMS data were not informative. We found instrumental the SSEP values that reflected perioperative CCT dynamics, lat p45, amp n13, AUC n11-n13, AUC n13-n18. These SSEP values were used to create a prognostic rule through logistic regression and ROC-curves. As a result, we were able to predict neurogenic dysphagia in early Introduction: EEG monitoring during the first 24 hours robustly contributes to the prediction of either poor or good outcome in comatose patients after cardiac arrest [1] . Quantitative EEG (qEEG) measures can be useful to visualize evolution of the EEG over hours.

We recently proposed the Cerebral Recovery Index (CRI), an index based on a combination of five qEEG measures grading the severity of hypoxic brain damage on a scale from zero to one to facilitate prognostication [2] . Objectives: To evaluate the prognostic accuracy of a revised CRI, after optimalization by the use of a random forest classifier instead of a manually chosen feature combination and the addition of four qEEG measures, resuscitation parameters and patient characteristics. Methods: In this prospective cohort study, 283 consecutive comatose patients after cardiac arrest were included in two intensive care units. Continuous EEG was recorded during the first three days. Outcome at 6 months was dichotomized as good (CPC 1-2) or poor (CPC 3-5). Nine qEEG measures were extracted: alpha to delta ratio, signal power, Shannon entropy, delta coherence, regularity, the number of burst/min, mean and max burst correlation, and fraction of burst correlation >0.8. These measures were combined with patient characteristics and resuscitation data, including sex, age, initial heart rhythm, in-versus out-of-hospital-cardiac-arrest, and presumed cause of cardiac arrest. Patients were randomly divided over a training and a validation set of respectively 143 and 140 patients. Within the training set, a random forest classifier was fitted for each hour after cardiac arrest. Based on results in the test set, two thresholds were chosen: one for predicting poor neurological outcome and one for predicting good neurological outcome. Subsequently, the revised CRI was evaluated in the validation set.

Results: Poor outcome could reliable be predicted with the revised CRI (with 100 % specificity) in the validation set with a sensitivity of 67 and 37 % at respectively 12 and 24 hours after cardiac arrest. Good neurological outcome could be predicted with a sensitivity of 60 and 62 % at a specificity of 94 and 93 %.

Conclusions: Here we show that a combination of qEEG and clinical measures, extracted and combined by a random forest classifier, provides reliable, objective prognostic information. This revised CRI can be used for the prediction of both poor and good neurological outcome, thereby poor outcome can be reliable predicted (without false positives) with relatively high sensitivity. The revised CRI is expressed as a single index between 0 and 1, which can be used in real time at the bedside, even by professionals who are not trained in EEG interpretation. Introduction: Continuous electroencephalography (cEEG) allows real-time monitoring critically-ill patients neurophysiology and to detect non-clinical seizures in comatose patients, delayed cerebral ischemia after subarachnoid haemorrhage, and guide therapies for status epilepticus. The application of cEEG is still limited because it requires awkward analysis by experienced neurophysiologists of huge amount of EEG tracings. Quantitative EEG (qEEG) techniques, i.e.amplitude integrated EEG (aEEG) and Density Spectra Array (DSA), have been developed to simplify the complexity of EEG interpretation, to allow rapid evaluation of cerebral background electrical activity and the power spectrum of the EEG frequencies derived from raw data EEG. These developments offer the potentiality to transform an instrument interpreted by neurophysiologist afterwards in a monitoring tool useful to ICU staff. Objectives: To test the hypothesis that EEG-nonexpert neurointensivists can obtain real-time reliable information from qEEG after training under the supervision of an in-house neurophysiologist. To describe the implementation of qEEG monitoring in 2 neurointensive care units. Methods: The implementation occurred in sequential phases. cEEG was recorded using 8 surface electrodes according to the international 10-20 system, on a bipolar longitudinal montage in patients with brain injury. qEEG-naïve neurointensivists, after a short training from a neurophysiologist followed by daily supervision for the study period, were subjected to a baseline test evaluating aEEG and DSA traces. Each panel consisted of raw EEG data and 3 qEEG tools: the color density spectral array (DSA), amplitude integrated EEG (aEEG) and the burst suppression rate (BSR). After this evaluation, daily qEEG evaluation was performed by the neurointensivists and reviewed by the neurophysiologist. Results: From July 2015 to April 2016 we monitored 77 patients (57 ± 17 years, 34 male) admitted for brain trauma (41 %), stroke Introduction: It has been noted the importance of job satisfaction in healthcare services and the consequences resulting therefrom, such as increasing the quality of care services provided and satisfaction of their users.

Objectives: To develop a model of influence of human resource management directed to the quality management and organizational excellence in the organizational results, from the perspective of healthcare staff. Methods: We carried out a research study, of a transversal nature, whose study population were a total of 248 (12,9 % physicians, 49,2 % nurses and 37,9 % nurse assistants) ICU staff. A personal questionnaire was used to measure, through Likert scales of 7 points, the application of human resource practices of high commitment (HR), the quality of service provided to the patient (QUALITY), the satisfaction with the capacity of the service (CAPACITY), the personal satisfaction with the work done (SATISFACTION) and the affective commitment with the organization (COMMITMENT).

Results: The measure models of these five constructs were validated by confirmatory factorial analysis, whose results were satisfactory. The measurement model of HR is a second order construct which is Introduction: In the period between 2007-2010, a successful implementation project was finished aimed at strict blood glucose level (BGL) regulation in the Intensive Care Unit (ICU) [1] . We hypothesized that glucose control would afterwards slack and that implementing other measures to modify behavior would be required to regain adequate glucose control. Methods: A prospective study was performed in a 12-bed mixed medical-surgical ICU of a university affiliated teaching hospital. All BGL values were extracted from the ICU database in 4 years following the implementation project until December 2014. Following the project, BGL targets were set at a range of 80-135 mg/dl, Nurses' instructions for keeping BGL values in target were not changed. After 3.5 years, an automated warning system was implemented in the patient data management system that triggered a centrally placed monitor with feedback about the need for obtaining a BGL value, based on the actual value compared to the previous one. The primary outcome measure was mean BGL. Secondary endpoints were sampling frequency, BGL within predefined targets, incidences of severe hypoglycemia, and hyperglycemia. The analysis was restricted to patients with at least two blood glucose measurements. These indicators were analyzed over the course of time using the XMR control chart, a tool belonging to Statistical Process Control.

Results: Data of 3760 patient admissions were evaluated, which corresponded to 117,080 BGL measurements. The BGL sampling interval (Figure 56 ), mean BGL and percentage of severe hypoglycemia all increased after introducing nurses' instruction and decreased significantly after monitoring feedback (p < 0.05). Percentage of severe hypoglycemia events, which is associated tosafety, decreased with some delay after nurses' instruction and remained unchanged (0.18 % on average) and stable after introducing monitoring feedback. Percentage of "in range" measurements of both normoglycemia (80-110) and protocol recommended (80-135) decreased after nurses' instructions and then increased after feedback monitoring. Mean of per patient's standard deviation as a measurement of variability remained unchanged and stable after nurses' instruction and even decreased after monitoring feedback.

Conclusion: even after successful implementation of a BGL control system, behavior changed within 3 months with inherent worsening of BGL control. An automated warning monitor in a central location was able to restore BGL control in the ICU.

Using Objective: The overall objective of this research program is to use the KINARM to define the neurocognitive phenotype of ICU survivors (i.e. required invasive mechanical ventilation and/or vasoactive agents for hemodynamic support). This group is compared to healthy age-and gender-matched controls, as well as 3 active control groups. These active control groups were patients 1) pre-and 2) post-cardiac surgery, and 3) patients postcardiac arrest.

Methods: Participants performed 7 tasks on the KINARM that ranged from simple sensorimotor tasks to more complex executive tasks. For each task, 6-12 performance metrics were recorded. These metrics were compared to a normative database of age-and gendermatched controls and z-scores were generated. A composite score for each task was generated using a score derived from Maholanobis distance, with increasing scores representing worse performance. Cluster analysis was applied to these performance metrics using Euclidian distance. (Fig. 58) .

Conclusions: Serial MLT measurements significantly underestimate muscle wasting in critical illness and are not related to development of muscle weakness. In comparison, changes in RF CSA reflect changes in 'gold standard' methods of assessing muscle mass, and are related to loss of muscle mass and function in critically ill patients. There is significant evidence that electronic prescribing can significantly reduce the errors, however implementation of it is a long term project and is not feasible in attempt to improve medicines safety over short period of time. Therefore we aimed to improve safety of a current paper based system. Multidisciplinary intervention was chosen as this approach has been previously demonstrated to reduce medication errors on ICU 2 .

Objectives: Evaluate effect of multidisciplinary intervention to improve medicines safety. Methods: Over the course of 12 months following interventions were introduced: development and implementation of new ICU specific IV infusion chart, prescription checks during nursing handover, introduction of daily pharmacy handover and on-site feedback, additional medicines training for current staff and new medicines safety induction module for new-starters. Outcome data was based on monthly spot audits carried out by pharmacy staff. Comparison is made between Quarter 2 and Quarter 4 after the start of intervention. Chi-square test was used to compare the two datasets.

Results: There were 1847 prescriptions analysed in Q2 and 2293 in Q4. We observed a five-fold reduction in prescription validity errors from 0.97 % to 0.17 % (p < 0.001). And nearly ten-fold reduction in administration of medicines against non-valid prescriptions from 0.38 % to 0.4 % (p < 0.001). Pre-printed ICU specific IV infusion chart eliminated errors related to variable dilutions, choice of diluent, incorrect or inconsistent infusion rates. Month-by-month trends are presented in Figure 59 . Conclusion: Multidisciplinary intervention has resulted in significant improvement in medicines safety. Introduction: The concept of frailty has been defined as a multidimensional syndrome characterised by the loss of physical and cognitive reserve that predisposes to adverse events. The prevalence of frailty amongst the critically ill is unknown, however it is probably increasing. This audit aimed to look retrospectively at our admissions to Intensive Care, to categorise them into frail or non frail, and evaluate how frailty correlated with ICU length of stay and mortality Methods: A retrospective case note review of all patients admitted to Intensive Care over a six month period in the Victoria Infirmary and then Queen Elizabeth University hospital in Glasgow. Classification of frail or non-frail was done using a combination of the Clinical Frailty Score (CFS) and Edmonton frailty scale. [1, 2] . Once classified into frail and non-frail we looked at ICU outcome, length of stay, APACHE, weight on admission, lowest albumin and admission haemoglobin and compared the frail population to the non-frail population.

Results: Two hundred and eighty four patients were admitted to Intensive Care in this time period. Of those, 102 were over the age of 65 years. Of the 102 patients, 68patients were deemed to be frail, and 34 were deemed to be non-frail using the CFS. Approximately 40 % of the patients admitted to Intensive Care are over the age of 65. There was no significant difference found in mortality, ICU length of stay or hospital stay, APACHE or weight between the two groups.

[see Table 35 ] Conclusions: We know that the utilisation of intensive care resources by older people is rising. Our data shows that almost 40 % of those admitted to ICU are over the age of 65. Interestingly, there is no significant difference between the non frail and frail groups of patients admitted to intensive care. This may be because of small sample size. The length of stay of the frail patient is shorter and this may be because as intensivists we are better at treatment limitation in this group of patients. No difference in overall mortality suggests that the patients we deem suitable for intensive care who are frail do as well as the non-frail cohort as the selection process for admission has been adequate. Patients deemed to be frail are more likely to be dependant on care if they survive, with 50 % requiring some sort of support on discharge.

Most studies show that frailty is associated with increased mortality so it is indeed interesting that this audit has shown no difference between the two groups. Figure 60 shows a significant increase in admissions among the elder groups along the five-year periods. The severity scores increased significantly as shown in Figure 61 (p < 0.001). ICU lenght stay also decreased significantly (Table 36) Introduction: In our Intensive Care Medicine (ICU) department we used a database (GESPAC) with uniform and quality data for all admissions from 2000 to 2014, which allows us to study the evolution of severity scales and the clinical activity by age and type of patient.

Objectives: To describe the effectiveness of severity scales used in our ICU over 15 years by age and type of patient.

Methods: A retrospective, single-center and descriptive study was conducted from 2000 to 2014. All patients admitted consecutively were included. Patients with lenght of stay less than 24 hours were excluded.The severity scales we analyzed were MPM0, MPM24, SAPS2, APACHEIII. Patients were divided in 4 groups of age by quartiles (<50 years, 51-64 years, 65-74 years, > 74 years). The type of patient was classified in medical and urgent or scheduled surgery. We used descriptive statistics. Qualitative variables are expressed as percentages and quantitative variables are expressed as means and standard deviations (± SD) and ROC curves for the analysis of discrimination. We used SPSS v19. Results: We included 7396 patients, 4955 were men (67 %), mean age was 61.6 years (SD 16.6). ICU mortality was 11.7 %. In Figure 63 we show 4 curve ROC corresponding to the 4 severity scales for all patients, MPM0 has a significantly worse discrimination respect to the other 3 scales. MPM 24, SAPS 2 and APACHE III have a similar behavior.

In Table 37 we show the severity scales effectiveness by age groups. In Table 38 we show the severity scales effectiveness by type of patients.

We observed a decrease of effectiveness of severity scales over time, however this effectiveness remains optimal in all the severity scales except for MPM0. Introduction: We implemented a Critical Care EPR using the Quadramed system on the 30th Sept 2015. Our objective was to evaluate whether the EPR had improved the quality of our documention and the responsiveness of our notes. Methods: We evaluated the patient record from 21 hospital days prior to the implentation of the EPR and 21 hospital days 6 months after using and refining the system. Results: The proportion of completed nursing risk assessments did not change after implementation of the EPR. They depend on the user to shedule their completion. Safety checks for arterial and CVC lines were well established and changed little. There was an improvement in the percentage of shift checks completed when they were automatically sheduled.

The system provides a date, time and audit trail for each entry. The user traceability in the medical notes increased. The presence of the author´s name improved from 88 % to 100 %, the date from 96 % to 100 %, the time from 76 % to 100 %, and signature from 77 % to 100 %. Legibility improved from 89 % to 100 %. The proportion of entries with a contact number dropped from 23 % to 18 %. The nursing care plans in the paper notes were better completed than the medical notes, but still improved. The presence of the nurse´s name increased from 97 % to 100 %, the date from 94 % to 100 %, the time from 95 % to 100 % and the signature from 95 to 100 %. Legibility was 100 % in both groups.

The Quadramed system provides automatic calculations of early warning scores and fluid balances.The more complicated the calculation, the greater the improvement.

Integration of data: · The increase in data points that cross populate is: +350 · The allergy advice populates ALL the sheets compared to an average of 1 on paper (excluding the drug chart). · The average number of scheduled events (that instruct staff to perform functions) has gone from 5 to 35. Conclusions:

1. The largest improvement came in the accessibility of the notes. They can now be accessed within one minute from any PC in the Trust. Previously a standard time to deliver notes was two days, reducing to one day in an emergency. 2. Correct filing of the EPR notes and the search facility reduced the average time to complete the audit by 16 minutes per patient. 3. The user audit trail and traceability improved in both medical and nursing paperwork, more so in the former. This is explained by a baseline of lower documentation standards in the medical group. 4. The typed out notes are now legible. 5. There was a large improvement in the quality of data calculations that are now up to 100 %. 6. There was a large increase in the number of scheduled events, but this these only lead to an improvement in documentation when they were automatically scheduled by the computer.

There was no improvement when user scheduling was required. We noticed very high levels of satisfaction regarding the professional care (frequency of communication, physician skill and competence, understanding information, honesty and facilities of getting information) and overall with care. Satisfaction was even higher when we considered the usefulness of the ECP. Every respondents supported it as a complement to daily information but it was only supposed to replace verbal information in 11,8 %. 17 % did not access the website because of sufficient verbal information or cultural or age-related difficulties. The access was mostly via computer (47,1 %) followed by smarthphone (23,5 %). Particularly desired were daily updates, an established timetable and more detailed information. There were no statistically significant differences in the need of web access among families living near the hospital and not or prior experience with ICU familiar admission.

Conclusions: ECP appears to decrease the level of anxiety of families, improves perceived quality and can help to combine patient care with their work and personal responsibilities without replacing the daily evolution provided by physicians. Introduction: Neuroimaging shows promise for determining early prognosis after cardiac arrest (CA). Nevertheless, conventional MRI sequences, as T1-weightened sequences, are currently considered not precise enough to detect brain structural anomalies in this context, and therefore are supposed to be unable to accurately predict outcome 1 .

Objectives: We hypothesize that the combined use of cortical thickness measurement and subcortical grey matter volumetry could provide an early and accurate in vivo assessment of the structural impact of cardiac arrest (CA), and therefore could be used for longterm neuroprognostication in this setting.

Methods: Prospective study undertaken in five Intensive Critical Care Units affiliated to the University in Toulouse (France), Paris (France), Clermont-Ferrand (France), Liège (Belgium) and Monza (Italy). High-resolution anatomical T1-weighted images were acquired in 157 anoxic coma patients (16 +/− 8 days after CA) and 70 matched controls. Patients were followed up one year after CA. Cortical thickness was computed on the whole cortical ribbon and deep grey matter volumetry was performed after automatic segmentation 2 . Brain morphometric data was employed to create multivariate predictive models using learning machine techniques ( Figure 67) . Results: Patients displayed significantly extensive cortical and subcortical brain volumes atrophy compared to controls. A dissociated vulnerability to anoxic insult was observed: subcortical volumes were related to CA duration and cortical thickness values were linked to the time to MRI acquisition ( Figure 68 ) The accuracy of a predictive classifier, encompassing cortical and subcortical components has a significant discriminative power (AUC = 0.87).

The anatomical regions which volume changes were significantly related to patient's outcome were: frontal cortex, posterior cingulate cortex, thalamus, putamen, pallidum, caudate, hippocampus and brainstem ( Figure 69 ) Conclusions: These findings are consistent with the hypothesis of pathological disconnection within a striatopallidal-thalamo-cortical mesocircuit induced by CA 4 and pave the way for the use of combined brain quantitative morphometry in this setting.

Clinical and electrophysiological correlates of absent somatosensory evoked potentials after post-anoxic brain damage: a multicentre cohort study Conclusions: Our data confirm that bilateral absence of N20 reflects severe post-anoxic cerebral damage and therefore frequently correlates with concordant clinical and EEG signs of poor outcome. However, our study also identified a subset of patients with discordant signs, in whom clinical examination and/or EEG were reactive despite bilaterally absent N20. Our findings raise further questions on outcome prognostication after CA and underline the importance of multimodal assessment in this setting.

The response time threshold for predicting favorable neurological outcomes in patients with bystander-witnessed out-of-hospital cardiac arrest Introduction: It is well established that the period of time between when a call in made to emergency medical service (EMS) to the point when EMS arrive at the scene (i.e., the response time) affects the survival outcomes in out-of-hospital cardiac arrest (OHCA) patients. However, the relationship between response time and favorable neurological outcomes remains unclear. We therefore aimed to determine a response time threshold in bystander-witnessed OHCA patients that is associated with positive neurological outcomes and to assess the relationship between the neurological outcomes and response time in OHCA patient.

Methods: This study was a retrospective, observational analysis of data from 204,277 episodes of bystander-witnessed OHCA between 2006 and 2012 in Japan. We used classification and regression trees (CARTs) and receiver operating characteristic (ROC) curve analysis to determine the threshold of response time associated with favorable neurological outcomes (Cerebral Performance Category 1 or 2) one month after cardiac arrest. Results: Both CARTs and ROC analyses indicated that a threshold of 6.5 min was associated with improved neurological outcomes in all bystander-witnessed OHCA events from cardiac origin. Furthermore, bystander cardiopulmonary resuscitation (CPR) prolonged the threshold of response time by 1 min (to 7.5 min). The adjusted odds ratios for favorable neurological outcomes in OHCA patients who received care within ≤ 6.5 min was 1.935 (95 % confidential interval: 1.834-2.041, P < 0.001).

Conclusions: A response time ≤ 6.5 min was closely associated with favorable neurological outcomes in all bystander-witnessed OHCA patients. Bystander CPR prolonged the response time threshold by 1 min. Methods: Patients with established out-of-hospital cardiac arrest (OHCA) who underwent cardiopulmonary resuscitation with subsequent return of spontaneous circulation were retrospectively enrolled. Two hundred and eight DIC patients diagnosed by the Japanese Association for Acute Medicine (JAAM) DIC criteria were divided into two subgroups with hyperfibrinolysis (73) and without hyperfibrinolysis (135). The definition of hyperfibrinolysis was made by a FDP level > 100 μg/mL. Platelet count, global markers of coagulation and fibrinolysis were measured 4 times after admission to emergency department (T1, 0-6; T2, 6-12; T3, 12-18; T4, 18-24 hr). The outcome measure was the hospital all-cause mortality. Results: Patients with hyperfibrinolysis had higher DIC, SIRS, and sequential organ failure assessment (SOFA) scores associated with higher prevalence of MODS, leading to a higher mortality rate of 68.5 % in comparison to patients without hyperfibrinolysis (41.7 %).

Stepwise logistic regression analyses confirmed that DIC, SOFA scores, and lactate levels are independent predictors of patient death. Hyperfibrinolysis also predicted patient death. Tissue hypoperfusion (as indicated by lactate level) is a main determinant of hyperfibrinolysis. Receiver operating characteristic curves showed a significant discriminative performance of DIC scores for patient death. Kaplan-Meier curves showed that DIC, especially DIC with hyperfibrinolysis, significantly affected patient death. Conclusions: DIC with the fibrinolytic phenotype during the early phase of post-CPR more frequently results in SIRS and MODS, and affects the outcome of OHCA patients. Hypoxia/ischemia during cardiac arrest and CPR are considered to be the cause of increased fibrin(ogen)olysis.

The association between tracheal intubation during pediatric inhospital cardiac arrest and survival L.W. Andersen 1,2 , T. Introduction: Tracheal intubation is common during pediatric inhospital cardiac arrest, although the relationship between intubation during cardiac arrest and outcomes is unknown. Objective: To determine if intubation during pediatric in-hospital cardiac arrest is associated with improved outcomes. Methods: This was an observational study of prospectively collected data from United States hospitals participating in the Get With the Guidelines -Resuscitation registry. We included pediatric patients (age < 18 years) with index in-hospital cardiac arrest. We excluded patients who were receiving invasive mechanical ventilation and/or had an invasive airway in place at the time chest compressions were initiated. The exposure was tracheal intubation during the cardiac arrest. The primary outcome was survival to hospital discharge. Secondary outcomes included return of spontaneous circulation and neurological outcome. A favorable neurological outcome was defined as a score of 1-2 on the pediatric cerebral performance category score. Patients being intubated at any given minute (from 0 to 15 minutes) were matched with patients at risk of being intubated within the same minute (i.e. still receiving resuscitation) based on a timedependent propensity score calculated from multiple patient, event, and hospital characteristics. Modified Poisson regression with adjustment for matching and clustering were then performed to obtain risk ratios. Results: 2294 patients were included. Of these, 1555 (68 %) were intubated during the cardiac arrest. In the time-dependent propensity score-matched cohort (n = 2270), survival was lower in those intubated compared to those not intubated during cardiac .02], p = 0.08) between those intubated and not intubated during cardiac arrest. The association between intubation and decreased survival remained in the majority of our sensitivity and subgroup analyses Conclusions: Tracheal intubation during in-hospital pediatric cardiac arrest was associated with decreased survival to hospital discharge. These findings challenge the present resuscitation paradigm for pediatric in-hospital cardiac arrest. Introduction: Substantial proportion of patients who suffered cardiac arrest do not respond to conventional cardiopulmonary resuscitation. Recently, extracorporeal cardiopulmonary resuscitation (ECPR) has been introduced as a potentially life-saving procedure in refractory cardiac arrest. Objectives. The aim of our study was to evaluate the relation between ECPR survival, lactate levels and blood pH. Methods: Eligible patients for this analysis had to undergo ECPR after at least ten minutes of unsuccessful cardiopulmonary resuscitation with a minimum of three defibrillation attempts. For extracorporeal life support (ECLS) we used Cardiohelp system (Maquet, Germany) or Levitronix CentriMag blood pump (Levitronix, USA). LUCAS II system (PhysioControl, Sweden) was used for chest compressions during ECLS insertion and cannulas were placed with percutaneous puncture under fluoroscopy or ultrasound control. Blood lactate and pH levels measured before ECLS insertion and after 24 hours were used for this study. Results: We analyzed data from 29 patients treated with ECPR for refractory cardiac arrest. The mean age of our patients was 57 years (31-81). Out-of-hospital cardiac arrest occurred in 16 patients, 13 patients suffered from in-hospital arrest. Thirty-day mortality in our group was 57 % and 32 % of patients recovered with good neurological outcome. Percutaneous coronary intervention was performed in 18 (67 %) patients. Baseline value of lactate was 11.52 ± 5.42 mmol/l, initial pH 6.97 ± 0.21. In comparison with survivors, patients who died had significantly higher initial lactate levels (15.05 ± 1.56 vs. 10.01 ± 1.03; P < 0.05) and lower baseline pH (6.87 ± 0.06 vs 7.04 ± 0.04; P < 0.05). Moreover, survivors had significantly lower lactate levels after 24 hours. Conclusions: ECPR represents virtually the last chance to survive refractory cardiac arrest. The levels of blood lactate and pH are significantly associated with clinical outcomes of ECPR. Introduction: Post-cardiac arrest survivors treated with therapeutic hypothermia (TH) remain comatose after rewarming. In contrast to survivors without TH, neurological prognostication is imprecise due to a persistent sedative effect [1] . Objectives: We aimed to evaluate clinical signs and findings that could predict neurological recovery and determined the optimal time for prognosis. Methods: We retrospectively reviewed database of 51 post-arrest patients treated with TH in our hospital from 2006 to 2014. Cerebral performance category (CPC), neurological signs and findings in EEG and brain CT were evaluated. Neurological recovery was scored as favorable neurological outcome, namely normal cerebral function(CPC1) and moderate disability(CPC2) or unfavorable neurological outcome, namely severe disability(CPC3), vegetative state(CPC4) and death(CPC5). Neurological signs and findings in EEG and brain CT, which possibly predicted neurological recovery, and the optimal time to evaluate neurological status were analyzed. Results: TH was performed in 51 post-arrest patients. Approximately 53 % (27/51) of TH-patients survived and 33 % of the survivors had favorable neurological outcome. Findings predicting unfavorable outcome at discharge were lack of pupillary response and/or Gag reflex after rewarming, and the absence of at least one of the brainstem reflexes, no eye opening or motor response worse than pain withdrawal (M ≤ 3) on the seventh day. (Table  43 ) Myoclonus and seizure could not be used to indicate poor prognosis. One of 17 survivors with myoclonus had full recovery and 50 % of the survivors with seizures regained consciousness upon discharge. Findings of EEG and brain CT showed that the patients with burst-suppression EEG pattern or brain swelling became vegetative or died, but the prognostic values of these findings were inconclusive. Conclusions: Our study showed that the simple neurological signs helped predict short-term neurological prognosis of comatose survivors undergoing TH. The most reliable signs which determined unfavorable outcome were the lack of the pupillary light response and Gag reflex. The optimal time to assess prognosis was either at 48 to 72 hours or 7 days after return of spontaneous circulation. Physicians can use these neurological signs to evaluate the prognosis of postcardic arrest survivors treated with TH. Objectives: Procalcitonin (PCT) and presepsin are biomarkers associated with severe infections. We asked, if they could be used to reflect the severity of the post-cardiac arrest syndrome and to predict poor outcome. A significantly greater increase in procalcitonin from admission to 24 h was observed in patients with eventual poor outcome compared to those with a favorable one (p < 0.001). Presepsin levels were on average constantly higher in patients with poor outcome but did not show any statistically significant changes in repeated measures analysis of variance.

Conclusions: Plasma procalcitonin may be a useful tool for the evaluation of long-term outcome of out-of hospital cardiac arrest patients at the ICU. On the contrary, presepsin did not provide clinically relevant additional predictive value in the study setting. Introduction: Prognosis of cardiac arrest survivor is mainly determined by ischemic brain injury. Post-cardiac arrest state is characterized by elevated circulating cytokines and hemodynamic instability, called as a sepsis-like syndrome. In many critical ill diseases such as acute pancreatitis and sepsis, a low serum level of high-density lipoprotein (HDL) and apolipoprotein A-1 (ApoA1) were associated with poor outcomes. Objectives: In this study, we examined whether a serum level of HDL and ApoA1 at intensive care unit (ICU) admission is associated with a neurologic outcomes in cardiac arrest survivors.

Methods: This study was a retrospective observational study conducted in a single tertiary urban hospital ICU. All admitted patients following cardiac arrest were screened during from March 2013 to December 2015. Patients younger than 15 years and without admission lipid panel were excluded. Neurologic outcome was determined by hospital discharge cerebral performance categories (CPC). Good neurologic outcome was defined as CPC 1 and 2. Note: This abstract has been previously published and is available at [2] . It is included here as a complete record of the abstracts from the conference. We analyzed all patients admitted on hospital ward that were assisted by the IHCA team. Patients admitted less than 24 hours on Ward and patients not eligible for resuscitation were excluded. Demographic data (age and gender) were collected. We analyzed the type of patient (medical or surgical), the schedule in which the IHCA happens (weekdays from 8 hours am to 8 hours pm and the rest, every day from 8 hours pm to 8 hours am, weekend and holidays), IHCA witnessed, the IHCA team time reaction, IHCA established or not at IHCA team´s arrival, return of spontaneous circulation (ROSC) and Hospital mortality. Statistics:Qualitative variables are expressed as percentages and compared using the X2-test; quantitative ones are expressed as means and standard deviations (± S.D), and analyzed using Student´s t-test. Multivariate logistic regression was performed, with hospital mortality as the dependent variable. The level of significance was placed at p < 0.05. The statistical analysis was performed using specific software ( IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp). Results: 120 patients were assisted by the IHCA team and 85 patients were included. In Figure 72 we described the characteristics of the study population. The beginning of cardiopulmonary resuscitation (CPR) maneuvers were immediate on ward, according to IHCA protocol. The arrival of IHCA team was less than 5 minutes in all cases. Table 45 shows an assosciation between hospital mortality and the capability of anticipation of IHCA situations (schedule, witness, pre-cardiac arrest …). Table 46 shows the persistence in the multivariate analysis of the relationship of these factors with the hospital mortality. Conclusions: The number of activations of IHCA team is remarkable, mortality of these patients is very high despite being patients on ward without a bad expected outcome. The improvement in the factors associated with the capability of anticipation of IHCA situations (schedule, witness, pre-cardiac arrest …) could lead to an improvement in the prognosis of IHCA. , and vessels with diameter smaller than 20 μm were defined as small vessel. Serum level of endothelial cell specific molecule-1(endocan) was measured at specific time points. The hemodynamic parameters, the inotropic equivalent score, and prognosis of the patients were recorded. Results: 20 patients were iinvestigated in this preliminary report. They were equally divided into two groups (Survival and Nonsurvival) according to 28-day mortality. The baseline patient characteristics were not significantly between the two groups. The perfused small vessel density and proportion of perfused vessels at 24 h were higher in the Survival group than in the Non-survival group.

The endocan level were higher in the Non-survival group than in the Survival group, but the difference was not significant.

Conclusions: Our results revealed that the perfused small vessel density were higher in the Survival group than in the Non-survival group. It encourages further studies to investigate whether aiming to improve microcirculation can improve outcomes in patient with venoarterial ECMO life support system. Demographic data, sedation and vasopressor dose were recorded. Enteral feeding was started as soon as possible. The cumulative GRV was recorded up to a maximum of 7 days, with a cut-off value of 250 ml used to define intolerance to enteral feeding (IEF). All data is presented as median(p25-p75.) Statistical analysis was performed by STATS13.0 with Mann-Whitney U test and Chi-square test. Results: Data from 95 patients were recorded. Baseline demographic data were similar in the 3 groups. The average GRV and the doses of midazolam were the highest in patients with VV ECMO, while the number of days with IEF and the doses of NA were the highest in the VA ECMO group (Table 49) . Overall, GRV and number of days with IEF tended to be higher in survivors (n = 63, GRV 112 (0-360), days with IEF 3(2-5)) than in non-survivors (n = 33, GRV 364 (100-488, days with IEF 4(3-6)). Conclusion: Early enteral feeding is feasible during ECMO, in spite of impairments of gastrointestinal function potentially related to sedation and/or vasopressor treatment.

Extracorporeal membrane oxygenation for refractory cardiogenic shock in patients with peripartum cardiomyopathy A. Chao 1 Results: Six patients with confirmed PPCM were found. Two (33 %) patients died of neurological consequences (cerebral infarct and hypoxic encephalopathy) and their left ventricular (LV) ejection fraction remained about 20 %. One patient underwent heart transplantation. The other three patients weaned off ECMO and their LV function began to improve on day 3. They were discharged uneventfully. Conclusions: ECMO can provide an effective and simple treatment for critical PPCM with a satisfactory result. Patients supported by ECMO whose heart function did not begin to recover on day 3 and had neurological complications had a poor prognosis. Introduction: The use of venoarterial extracorporeal membrane oxygenation (va-ECMO) for prolonged cardiopulmonary resuscitation (CPR) and severe cardiogenic shock after CPR has widely increased (1). Bleeding complications, due to necessary therapeutic anticoagulation and CPR/SIRS induced coagulopathy are common (2) . Targeted temperature management (TTM) has shown positive effects on neurological outcome after CPR. Although optimal target temperature is not exactly known, TTM remains a recommended approach in patients after CPR (3). Objectives: To determine the incidence of bleeding complications in patients after CPR, who are on va-ECMO and treated with TTM (target temperature 34°C) simultaneously. Methods: We conducted a retrospective observational study from Jan 2009 to Dec 2015 and extracted relevant clinical data from electronically medical records. Outcomes of interest were 28d-mortality and incidence of bleeding complications within 36 hrs of CPR. Demographic data, (anti-)coagulation status and need for transfusion were also analyzed. Results: A total of 258 patients received va-ECMO during the study period of which 183 patients (70,9 %) underwent CPR before ECMO. Of these, 44 patients (24 %) were treated with TTM. The median age was 49 yrs (range 22-87 yrs) and 31 patients were male (71 %). SOFA score on admission was 15 (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) . Patients received CPR mainly because of either acute myocardial ischemia (MI) (19; 43 %) or malignant dysrhythmias not attributable to acute MI (11; 25 %). ECMO implantation was performed within 3 hrs (0.5-26 hrs) of CPR and ECMO duration was 92.5 hrs (6-648 hrs). TTM was implemented within 2 hrs (0-7 hrs) of CPR and the duration of simultaneous treatment with va-ECMO and TTM was 21.3 hrs (2-48 hrs). 40 Introduction: During initial resuscitation of patients with shock, last consensus recommend to target a mean arterial pressure of at least 65 mmHg. However, there is no recommendation for the mean arterial pressure target in in the particular setting of Extra Corporeal Cardio Pulmonary Resuscitation patients in the first hours following a refractory cardiac arrest.

Objectives: Therefore, we conducted an experimental study to assess the effects of two different levels of mean arterial pressure for macrocirculatory, microcirculatory and metabolic functions. Methods: Randomized animal study in university research laboratory. In fourteen male pigs, a myocardial infarction was induced by a surgical ligature of the inter-ventricular coronary artery, triggering a refractory ventricular fibrillation. After twenty minutes of standardised cardiopulmonary resuscitation, Extra Corporeal Life Support was initiated to restore the circulatory flow. Then, animals were randomly allocated to a high mean arterial pressure group (High MAP, 80-85 mmHg) or to a standard mean arterial pressure group (Standard-MAP, 65-70 mmHg). Evaluations at baseline, just before and six hours after ECLS initiation were focused on 1) lactate, 2) amount of fluid infused and 3) microcirculatory parameters (Sidestream Darkfield Imaging, renal and liver functions). Results: The two groups were similar at baseline and also at time of ECSL initiation including for the lactate levels ( Objectives: The goal of our study was to assess the factors associated with percutaneous cannulation success or failure during CA.

Methods: This was a prospective observational monocentric study conducted between 2014 May and 2016 February including all consecutive patients with CA (no return to spontaneous circulation after 30 minutes of cardio-pulmonary resuscitation) and an indication of ECLS (low-flow below 90 minutes). Femoro-femoral cannulation (17Fr for arterial cannula and 19Fr for venous cannula, Maquet®) was performed using the Seldinger technique under ultrasound (US) guidance. Patient characteristics, physician's a priori about cannulation conditions (ranging from 1 (expected very easy) to 5 (expected very difficult)) and US measures of femoral vessels diameter were recorded. The primary endpoint was the time to ECLS initiation (ICU admission -ECLS running) and was analyzed using a stepwise multivariable linear regression. As a secondary analysis, we also explored the differences between the patients with a time to ECLS initiation < 20 min and all others (>20 minutes or cannulation failure Conclusions: Our data suggests that the level of care patients receive does not affect baby being with its mother. However, level 3 patients are admitted to a general critical care unit and not captured in this review. The results also suggest that maternal well-being is affected by and/or affects whether baby is with her, however the 95 % confidence intervals overlap. The above data does not take into account babies being in Special Care Baby Units and therefore not able to join their mothers at their bedside (this data will be recorded in the future). When this data is recorded it is expected that a high proportion of babies who are able to be at their mother's bedside during AMC will be there.

The Introduction: High levels of job satisfaction are associated with decreased turnover intention, burnout incidence and absenteeism among health care professionals. Moreover, turnover and burnout negatively impact on quality of care and healthcare costs. As the intensive care unit (ICU) represents a highly complex and stressful environment, prevention of conflicts among team members as well as improvement of communication and job satisfaction can as such reduce burnout risk.

Objectives: This study explores the relationship between communication -and job satisfaction and the impact on burnout and turnover intention among ICU nurses. 1 %, 182/303 ). An average job satisfaction of 7.66/10 was found. 5.33 % (16/300) had a score ≤ 5 on job satisfaction, indicating significant dissatisfaction. ICU nurses were most satisfied with the trust received from their supervisor (76.6 %) and least with the information about accomplishments and/or failures of the organization (21.8 %). 49.5 %, (150/290) had a low, 39.6 % (120/290) an average and only 6.6 % (20/290) a high turnover intention. 3 % of the ICU nurses had an indication for burnout. 23.7 % of the nurses had a low experience of their personal accomplishment. Conclusions: In this survey ICU nurses had a reasonable job satisfaction. They are most satisfied with the trust received from supervisors. Despite a low indication for burnout risk, a quarter of ICU nurses report low personal accomplishment. This may represent a particular focus for both preventive and interventional actions, which should preferably be developed through and in conjunction with the supervising staff.

Introduction: In 2013 the Swiss Diagnosis Related Groups ((Swiss)-DRG)) was implemented in Intensive Care Units (ICU). Its impact on hospitalizations has not yet been thoroughly examined. We compared the number of ICU admissions according to clinical severity and referring institution, screened whether implementation of SwissDRG affected admission policy, ICU length-of-stay (LOS) or ICU mortality. Methods: Retrospective single center cohort study conducted at the University Hospital Zurich, Switzerland between January 2009 and end of 2013. Demographic and clinical data were retrieved from a quality assurance data base. Conclusions: DRG introduction had not affected ICU admissions policy, except for an increase of in-house patients with a low clinical severity of disease. DRG had neither affect ICU mortality nor ICU LOS.

Interactive gaming as part of mobilisation programs is feasible in the ICU, but specific explanation about the usefulness of these games to patients is crucial for improving motivation and engagement T. Introduction: In recent years, light sedation has gained attention as part of standard daily care in the intensive care unit (ICU). Consequently, patients are increasingly engaged in their rehabilitation process. Particularly early mobilization is associated with shorter time on the ventilator, shorter ICU length of stay and better survival [1] . Interactive gaming may be a challenging way of engaging the patient in his own rehabilitation program. Few data are available for the use of these interactive games in the ICU envirnoment as part of daily routine physiotherapy, although one study showed that it was safe [2] . We developed a trolley with a Wii (TM) device that can be easily used when the patient is mobilized in a chair. We hypothesized that this would be associated with increased motivation to participate in interactive gaming by our patients. Methods: The Wii device was used with 4 different games. Participating patients were offered to play 2 games of their own choice as part of the mobilisation program to improve their strength and coordination. No extensive explanation about the potential usefulness of these interactive games was given to the patients. After finishing the games, a specific survey was administered addressing motivation and affects on mental health. Scores were obtained using a Likert scale (range 1-7). Results are shown as median and interquartile range {P 25 -P 75 ] Results: At the time of abstract submission, 13 participating patients had finished a cycle of 2 games. Some of the patients liked to use the Wii device, particularly because a choice in games made it more interesting to use. Other patients, however, felt they were required to participate. Tennis, bowling and boxing were most frequently used. The use of the Wii was programmed in the daily mobilization schedule together with a physiotherapist, or just with the attending ICU nurse. In general, patients were not that enthusiastic about the Wii-games (median score 2 [1-2]), were not convinced that playing these games improved their well-being (median score 2 [1] [2] [3] [4] ), and most felt that they did not have a choice but to participate (median score 6 [5] [6] [7] ). Conclusion: Interactive gaming with the Wii-device is feasible in ICU patients. However, thorough explanation of the potential usefulness of these games is required to engage and motivate patients to participate. Methods: Descriptive and retrospective study. We include all patients admitted in ICU during 2012 (previous to RRT establishment) and 2015 (year of RRT establishment), and who were discharged to the ward. We analysed the ICU and hospital mortality in both groups. Results: In 2012 we admitted 1296 patients (17,3 % from the ward and 82,7 % from the other places like Emergency, other ICU or other hospitals), with a ICU mortality of 10,4 % and hospital mortality of 15,55 %, with a hospital mortality after ICU discharge of 5,69 %. In 2015 we admitted 1307 patients (14,6 % from the ward and 85,4 % from the other places), with a ICU mortality of 12,6 % and hospital mortality of 15,68 %, and a hospital mortality after ICU discharge of 3,5 % (p = 0.03). Conclusions: After the first year of RRT establishment in our hospital, we appreciate that the continuation of patients after ICU discharge, decrease the hospital mortality ( 5,69 % versus 3,5 %, p = 0.03). The number of patients admitted in ICU from the ward decrease in 2015 (14,6 % versus 17,3 %), maybe because we did a previous assessment of this patients, with a stabilization in the ward and avoiding the ICU admission. Objectives: To identify outcomes and prognostic factors in HM patients admitted over 5-years to a general intensive care unit in a specialist haematology centre. (Figure 76 ). Patients with 3-4 organ failures had significantly worse outcomes than those with 1-2 organ failures (p < 0.0001) OR 13.4 [6.7 -25.0, CI 95 %]. Median APA-CHE II, SOFA and SAPS II scores were 24, 10 and 48, respectively. Median APACHE II (p < 0.05), SOFA (p < 0.04) and SAPS II (p < 0.001) scores were greater in those ventilated for 1-4 days vs. ≥5 days. Conclusions: APACHE II scores and mortality were greater than described in similar HM populations. Given the severity of critical illness in our cohort, we suggest that admission to ICU earlier in the acute illness may improve outcomes. Poorer outcomes were observed in those with >2 organ failures and in ventilated patients. The survival of 1/3 rd of patients on ICU for ≥5 days to hospital discharge suggests that 5-day trials of ICU in HM patients are unlikely to reliably distinguish between survivors and non-survivors. Background: Dedicated intensive care unit (ICU) physician staffing was associated with a reduction of ICU mortality in the general medical and surgical ICU. However, limited data were available on the role of a cardiac intensivist in the cardiac intensive care unit (CCU). We compared the clinical outcomes in adult patients admitted to CCU before and after implementing the cardiac intensivist-directed care. Methods: We enrolled 2,923 consecutive patients admitted to a CCU at Samsung Medical Center, from January 2012 to December 2015. In January 2013, CCU was changed from a low-intensity staffing model to highintensity staffing model which managed by a dedicated cardiac intensivist. We divided eligible patients into low-intensity group (n = 616) and highintensity group (n = 1,815). The primary outcome was CCU mortality. Results: High-intensity group had significantly lower CCU (8.9 % vs 4.1 %; p < 0.001) and hospital (10.9 % vs 6.1 %; p < 0.001) mortality compared to the low-intensity group. The decrease in CCU (54.5 % vs 22.5 %; p = 0.001) and hospital (57.6 % vs 29.4 %; p = 0.003) mortality in high-intensity group were consistent in 135 (low-intensity group 33, high-intensity group 102) patient with profound cardiogenic shock treated with extracorporeal membrane oxygenation. Kaplan-Meier survival curve showed significant higher cumulative survival rates in highintensity group at 1 year follow-up (Log rank test, p < 0.001). CCU (7.6 % vs 5.8 %; p = 0.101) and hospital re-admission rate (7.5 % vs 6.0 %; p = 0.179) were decreased as well after conversion to highintensity although these results were not statistically significant.

Conclusions: Dedicated cardiac intensivist was associated with reductions of CCU mortality in patient with cardiovascular disease requiring critical care. Introduction: Opioids are commonly given to alleviate pain and distress in patients admitted to the intensive care unit (ICU) patients or undergoing major surgery. Previous studies have shown that patients who are already taking opioids prior to surgery or ICU admission are more likely to experience an extended duration of opioid use postoperatively or post-discharge (1). However, it is unknown whether patterns of opioid usage differ between patients who are admitted to the ICU and those undergoing a surgical procedure.

Objectives: The objective of this study was to describe opioid use in critically ill patients before and after ICU admission and to compare it with preoperative and postoperative opioid use in a surgical population. Methods: Retrospective review and comparison of adult patients admitted to the ICU or undergoing surgery at a tertiary care center between January 1, 2006 and December 31, 2008. We divided the populations based on their degree of opioid use into "non-user", "intermittent", and "chronic" opioid users as previously described (1). We assessed opioid use at 3 months prior to ICU admission or surgery, at discharge, and monthly for 6 months thereafter. Patients admitted to ICU who had surgery were categorized under the ICU population. To assess for risk of monthly chronic opioid use, a Cox-Proportional Hazards model was postulated that allowed for recurrent events to account for patients irregularly requiring opioids over the course of the study period. The model showed that the risk of chronic opioid use was 52.2 times greater for those with prior chronic opioid use compared to patients who were non-users. There was no difference in risk of chronic opioid use between the ICU and surgery group. Conclusions: Our findings suggest that ICU and surgical patients have similar risk of prolonged chronic opioid use post-discharge. Chronic opioid use prior to ICU admission or surgery is the strongest predictor of chronic opioid usage at and after discharge.

A single-centre cohort study of national early warning score (NEWS) and blood gas derived biomarkers in patients with acute medical illness Introduction: Empirical combination antibiotic therapy for treatment of severe sepsis is a matter of debate. The proposed rationale for using a combination of two or more different antimicrobials is several fold. First, it allows for a broader empirical coverage with a higher likelihood of targeting the causative organism. Second, it may decrease the development of resistance to the antibiotics used. Third, a combination of active drugs potentially cause a synergistic effect increasing the efficacy of bacterial eradication. The Surviving Sepsis Campaign recommends combination therapy in some patient populations and certain type of infections but the quality of the evidence supporting empirical combination antibiotic therapy is weak and does not include high quality randomised clinical trials (RCTs). Objectives: To assess benefits and harms of empirical mono-vs. combination antibiotic therapy in adult patients with severe sepsis in the intensive care unit (ICU). Methods: We performed a systematic review according to the Cochrane Collaboration methodology, including meta-analysis, risk of bias assessment and trial sequential analysis (TSA). We included RCTs assessing empirical mono-antibiotic therapy versus a combination of two or more antibiotics in adult ICU patients with severe sepsis. We exclusively assessed patient-important outcomes, including mortality. Two reviewers independently evaluated studies for inclusion, extracted data, and assessed risk of bias. Risk ratios (RRs) with 95 % confidence intervals (CIs) were estimated and the risk of random errors was assessed by TSA.

Results: Thirteen RCTs (n = 2,633) were included; all were judged as having high risk of bias. There was no difference in mortality (RR 1.11, 95 % CI 0.95 -1.29; p = 0.19) or in any other patient-important outcomes between mono-vs. combination therapy. In TSA of mortality, the Z-curve reached the futility area, indicating that a 20 % relative risk difference in mortality may be excluded between the two groups. For the other outcomes, TSA indicated lack of data and high risk of random errors.

Conclusions: This systematic review of RCTs with meta-analysis and TSA demonstrated no differences in mortality or other patientimportant outcomes between empirical mono-vs. combination antibiotic therapy in adult ICU patients with severe sepsis. The quantity and quality of data was low without firm evidence for benefit or harm of combination therapy. Introduction: De-escalation antibiotic in sepsis is associated with reduced costs and bacterial resistance. However, often it is not done. Objectives: We designed this study with the primary objective to evaluate the prevalence of de-escalation in patients with severe sepsis or septic shock in an academic public hospital in south Brazil. Secondarily we evaluated antibiotic adequacy and cultures positivity. Methods: We analyzed prevalence of de-escalation, antibiotic adequacy and culture positivity in severe sepsis and septic shock patients in an Intensive Care Unit. Results: Of the 224 patients included, de-escalation could have been performed in 29 % of cases (66 patients), but was implemented in only 19 % of cases (44 patients). Among patients who received deescalation, half was for antimicrobial spectrum narrowing. The mortality was not different between patients with or without de-escalation (56.8 % versus 56.1 %, p = 0.999). Empirical antimicrobial therapy was adequate in 89 % of cases. Pathogens were isolate in 30 % of all cultures and 26.3 % of blood cultures. Conclusion: The rate of empiric antibiotic adequacy was high, reflecting active institutional policy of monitoring the epidemiological profile and institutional protocols of antimicrobial use. However, the antimicrobial de-escalation could have been higher than reported. De-escalation did not impact mortality. There are few data in the literature regarding the care of severe sepsis patients in developing countries. This data can contribute to adequate treatment in this scenario. Introduction: Despite recent advances, appropriate initial amikacin dose in critically ill patients is still challenging. Relationship between pharmacokinetic/pharmacodynamic (pk/pD) parameter peak concentration (Cmax)/minimum inhibitory concentration (MIC) in critically ill patients is not clear. Objectives: We assessed the impact of amikacin pharmacokinetic and pharmacodynamic parameters on clinical and microbiological outcome in these patients. Methods: Observational prospective study. Adult patients (>18 years) admitted to an intensive care unit (ICU) with a gram negative documented infection and treatment with amikacin were included (Study period: September 2014 -April 2015). Amikacin blood samples were taken 24 to 72 hours after treatment started. Amikacin concentration were determined using Indiko® (Thermo Fisher Scientific), and drug adjustment were based on the recommendations given by the Pharmacokinetics Unit (Pharmacy Service). Clinical response, defined as sign and symptoms presented at the moment of infection diagnosis (fever, chest radiography alteration, infection biomarkers elevation and hemodynamic instability), was evaluated. Ji-square and U-Mann Whitney test were used to compare results between treatment responders and notresponders. . Mean initial dose was 1150 mg (SD: 45,9)/day, equivalent to 15.8 (0.6) mg/kg/day. With that dose, 40 patients (83,7 %) reached a Cmax/MIC value higher than 8. Final treatment response was higher for those patients with amikacin Cmax/MIC value >8 (61,0 % vs 12,5 %; p = 0.028). No significant differences were reached in early treatment response (Initial 72 h) (57,5 % vs 37,5 %;p = 0,403) or 30 days mortality (19,5 % vs 37,5 %;p = 0,265). Cmax/MIC values was not associated with toxicity-related treatment discontinuation (17,4 vs 12,5; p = 0,581). Conclusions: Initial Cmax/MIC value is associated with clinical response in those patients treated with amikacin. High initial amikacin dose may be necessary to optimize pk/pD parameters. Method: In a 30 bed mixed ICU from October 1, 2011 to September 30, 2015 nosocomial infections (pneumonia, urinary tract infections, catheter-related bacteremia (CRB) and secondary nosocomial bacteremia) were prospectively collected. ENVIN-HELICS diagnostic criteria were applied. Etiology, inflammatory response to infection, antibiotic treatment (ATB T) and treatment modifications according to culture results, were analyzed. SDD was applied to all admitted patients requiring endotracheal intubation over 48 hours. For each groups categorical variables were summarized as frequencies and percentages and number in means and standard deviations (SD) or median with interquartile ranges (IQR).Percentages were compared, as appropriate, with the Fisher´s exact test or X 2 test and medians with the Wilcoxon test for independent samples. For those variables that were associated with DE in the univariate analysis were entered into a logistic multidimensional analysis. The model obtained was expressed by p-values and odd-ratios, which were estimated by confidence intervals at 95 %. A hypothesis test was considered statistically significant when p-value was less than .05. Results: Ninety patients (34,8 %) had ATB DE and 168 did not. There were no significant differences in demographics or type of admission in both groups (Fig. 79) . Mortality was lower in patients receiving DE antibiotic (ATB) (23,3 %, p: 0.013). In the multivariate analysis, ICU mortality and urinary tract infection were the only variables found significant (Fig. 80 ) DE was performed in 53 out of 105 (50,4 %) with CRB and in 45 out of 103 (43,6 %) who had nosocomial pneumonia. The ATB T was inadequate in 58 out of 364 infections (15,9 %). Targeted therapy was performed in 136 out of 258 patients (52,7 %) and in 162 out of 364 infections, at least once occasion (44,5 %). Finally, 199 ATB were targeted prescribed.

In all studied patients with DE, this was performed in 72 patients once, in 13 patients twice and in 3 patients three times. The number of antibiotics used was 668 and ATB DE was performed in 104 occasions. Frequency of ATB used and of theirs DE is shown in Fig. 81 Of note, meropenem was DE in 14,4 %. Conclusions: Patients who received ATB DE compared to those that did not had a significant lower ICU mortality. The factors independently associated to DE were ICU mortality and urinary tract infection. Inadequate ATB T in our ICU occurred in 15.9 % of nosocomial infections. ATB DE was performed in 90 patients. Targeted therapy was applied to 44,5 % of infections. The most commonly used antibiotics were meropenem (12,3 %), levofloxacin (12.2 %) and piperacillin-tazobactam (12,05 %). Meropenem, was DE in 14,4 %. Introduction: Antimicrobial prescription represents a major challenge for clinicians in the daily practice especially in certain difficult clinical scenarios. Thus, in critically ill septic patients, prompt and adequate antimicrobial therapy reduces morbidity and mortality Objectives: We set out to assess the impact on in-hospital of antibiotic de-escalation in patients admitted to the ICU with severe sepsis or septic shock. Methods: Collaborative study enrolling patients admitted to the ICU with severe sepsis or septic shock from two different cohorts. The first one, a Spanish prospective and observational cohort and the second one, a multicenter non-blinded, randomized and non inferiority trial conducted in France. Severity was estimated by the use of the predicted mortality rate at ICU admissionfor every included patientby implementing the likelihood of death logit formule defined according to the APACHE II and SAPS II scores criteria and taking this cuantititative variable into account as a confounder factor in the regression model. De-escalation was defined as discontinuation of an antimicrobial agent or change of antibiotic to one with a narrower spectrum once culture results were available. To control for confounding variables we performed a multivariatebinomial logistic regression analysis adjusted by Wald test. Results: Nine hundred and one patients with severe sepsis or septic shock at ICU admission were treated empirically with broadspectrum antibiotics. Eight hundred and seventeen patients were evaluated (84 died before cultures were available). De-escalation was applied in 274 patients (33.5 %). We found no differences in hospital long of stay between de-escalation group comparedto those who did not received it. We also found a significant lower hospital mortality in de-escalation group in front of the others (25.9 vs. 43.1 %; p < 0.001). By multivariate analysis (adjusted by severity scores-APACHE and SAPS), factors independently associated with in-hospital mortality were age (Odds-Ratio [OR] 1.02; 95 % confidence interval [CI] 1.01-1.03), and SOFA score at ICU admission (OR 1.17; 95 % CI 1.11-1.23), whereas de-escalation therapy was a protective factor (OR 0.50; 95 % CI 0.35 -0.71) as well as urinary focus (OR 0.23; 95 % CI 0.12-0.43). Analysis of the 646 patients with etiological diagnosis revealed that the factors associated with mortality were age and SOFA and conversely de-escalation therapy was a protective factor (OR 0.45; 95 % CI 0.30-0.66) Conclusions: De-escalation therapy for septic critically ill patients is a safe strategy associated with a lower mortality. Efforts to increase the frequency of this strategy are indeed justified. Introduction: At the population level, both vancomycin and aminoglycosides are known to be nephrotoxic. The risk of nephrotoxicity might even be higher when combining the 2 agents together (1). Nonetheless, in septic patients, the benefit in terms of sepsis control may outweight the risk of nephrotoxicity. Thus, being able to appraise the risk/benefit ratio at the patient level would be of great interest to better tailor individual treatment. Objectives: Capitalizing on recent statistical innovations in personnalized medicine, our goal was to develop a patient-centered estimation of the impact of the association vancomycin/aminoglycosides on the kidney function. Methods: Our data come from a cohort study performed between 2000 and 2010 in the departments of anesthesia, critical care and cardiovascular surgery at a French teaching hospital. This study included all consecutive patients operated for an acute endocarditis (2). The primary endpoint was postoperative evolution of the kidney function as evaluated by the AKIN score (Stade 1: elevation in serum creatinine (SrCr) ⩾ 26.2 μmol/L or ⩾ 1,5xbaseline; Stade 2: elevation in SrCr ⩾ 2xbaseline; Stade 3: elevation in SrCr ⩾ 3xbaseline or creatinine ⩾ 354 μmol/L with increase >44 μmol/L or need for RRT). The impact of vancomycin/aminoglycosides on kidney function was estimated using targeted maximum likelihood estimation on a risk difference (RD) scale. The association between patient characteristics and the individual effect of the drugs on the kidney function was estimated using conditional recursive partitioning (cTREE). Results: 202 patients were included in the study. Their baseline characteristics are described in Table 53 . At a population level, we confirmed the strong association between vancomycin + aminoglycosides and the risk of kidney dysfunction (RD = 0.54, 95%CI: 0.47-0.61, p < 0.001). However, at the patient level, this effect was very variable and could be predicted based on patients characteristics (r 2 = 0.74) (Figure 82 ). Conclusions: The individual impact of vancomycin + aminoglycosides on kidney function may be very different than the overall effect at the population level. Innovant statistical approaches may be used to identify patients in whom this drug combination is safe, and others in whom it may seriously threaten kidney function. Introduction: Catheter-related bloodstream infection (CRBSI) is one of the most frequent nosocomial infections in critically ill patients, resulting in a significant increase of morbidity and mortality. That is why it is essential to detect CRBSI precociously so that optimal treatment can be initiated as soon as possible.

Objective: Our objective is to evaluate the effectiveness of rapid non-invasive tests that may allow clinicians to detect colonisation of the central venous catheter (CVC) as a source of bloodstream infection in critically ill patients, permitting catheter withdrawal and the initiation of early goal-directed antibiotic therapy.

Methods: Over the course of eight months, we selected for evaluation those critically ill patients admitted to our ICU who developed fever (>38°C) without source, to whom the clinician in charge decided to withdraw the CVC. Before extraction, we obtained a skin smear at the insertion site, in addition to a catheter-hubs smear. We sent both smears, along with the catheter tip and blood cultures, to the microbiology laboratory. These results indicate that our test has a negative predictive value of 100 %. To evaluate the degree of association between our test and the gold standard, we calculated the contingency coefficient, which resulted in 0,65 out of a maximum 0,7(p < 0,001). This shows the strong validity of the combination of skin and catheter-hub cultures as a diagnostic method for CVC colonisation.

Conclusions: The combination of skin and catheter-hub cultures is a rapid and very effective method for detecting the colonisation of CVC as a possible source of bloodstream infection in critically ill patients. While a negative result, which can be obtained within 24 h, would prevent the need for CVC withdrawal, a positive result would not only enable the removal of the likely source of infection, but it would also allow for, if the intensivist deemed it necessary, the initiation of early goal-directed antibiotic therapy. This would mean, in more than 60 % of cases, starting optimal treatment at least 48 hours prior to the diagnosis of CRBSI from blood cultures. Results: Introduction: Piperacillin/tazobactam (PTZ) is a β-lactam-β-lactamase inhibitor combination with a broad spectrum of antibacterial activity. βlactams are time-dependent antibiotics and their effectiveness is in association with the duration of free drug concentrations over the minimum inhibitory concentration (t > MIC) of organisms. Prolonged infusion has a pharmacokinetic (PK) advantage compared to intermittent bolus dosing, a continuous infusion lower dose of 8 g PTZ may be as effective as a higher dose of intermittent bolus PTZ. Objectives: In this study we intend to evaluate the continuous infusion of 8/1 grams of PTZ along with a loading dose of 4/0,5 grams in patients admitted in a tertiary ICU. Methods: Between October 2015 and December 2015 eight patients had piperacillin plasma concentration monitored during treatment with continuous PTZ infusion in a monocentric prospective observational study. Patients received a loading dose of 4/0,5 grams of PTZ followed by infusion of 8/1 grams, reconstituted in 90 ml Sodium Chloride 0,9 % and transferred to Braun Space Infusion System®. The pump had a flow rate of ± 4 ml/h. Blood was always extracted from the contralateral arm to the infusion, over 60 hours at predetermined times. Serum piperacillin/tazobactam concentrations were determined using an HPLC method (2). After extraction, samples (99 μL) were injected into a XBRIDGE C18 column (Waters, Spain) and were scanned by an UV detector at 220 nm with gradient elution. Mobile phase was composed by acetonitrile and a solution of tetrabutylammonium bisulfate (10 g/ L). Penicillin G was used as internal standard (Sigma Aldrich, Spain). Results: 8 patients have been examined (4 men and 4 women). The average age was 53 ± 22, the weight was 81 ± 22 kg, the creatinine clearance 122 ± 57 mL/min and the Apache II score 21,7 ± 7,3. The mean concentrations of PTZ in serum are represented in the next figures. Conclusions: In this ICU patient group, our results suggest that continuous infusion of PTZ at 8/1 g per day is sufficient to obtain therapeutic plasma-concentrations in critical care patients with infections caused by PTZ sensitive bacteria with a MIC lower than 8 mg/dl. However, in our group there were three patients with levels of 9 mg/dl, which are not sufficient for bacteria with MIC lower than 16 mg/dl; these three patients are neurocritical. In conclusion, further studies in this are needed, especially studies regarding the association between piperacillin therapeutic drug monitoring and clinical outcome. Introduction: Ultrasound guided internal jugular vein cannulation is recommended technique in current anaesthesia and intensive care practice. However, classic short axis view has inherent problem of needle visualization during venous access. In contrast, medial oblique view may enhance needle visibility during venipuncture and decrease overlap between IJV and carotid artery and thereby increase the safety of US guided IJV cannulation (1) (2) . Objectives: To compare the safety and efficacy of medial oblique view and in-plane technique as compared to short axis view and out-of-plane technique during US guided IJV cannulation. Methods: Two hundred patients aged between 18-50 yrs of either sex and American Society of Anesthesiologists' physical status I-II who were undergoing any surgery under general anaesthesia requiring an internal jugular vein cannulation, enrolled for this prospective randomized controlled trial. Three patients were excluded due to US machine malfunction. In patients of group M, IJV cannulation was performed with medial oblique probe position and in plane approach. In patients belonging to group S, IJV cannulation was done in out of plane approach with the US probe in short axis position. Primary outcome was needle and guide-wire visibility during procedure. Results: Needle visibility (entire needle tract and needle tip) was significantly higher during IJV puncture in medial oblique probe position (68 of 98 patients in group M versus 40 of 99 patients in group S; p = 0.00002). Guide wire visibility during insertion was also higher when medial oblique probe position was used (59 of 98 in group M versus 34 out of 99; p = 0.00013). First insertion success rate for IJV puncture, incidence of posterior wall of IJV puncture and time to cannulation were similar both the groups.

No serious complications such as carotid artery puncture, haematoma formation and pneumothorax were reported. Conclusions: Medial oblique view may increase safety of US guided IJV cannulation in comparison to short axis view by increasing needle visibility during puncture. Resuscitation "philosophy" shifted from being heart oriented towards brain oriented 2 since its delay may increase neurological deficits 1 . Recent studies are highlighting the role of the Bystander-CPR as a critical variable affecting OCHA neurological outcome. 3, 4 Objectives: Two step prospective interventional observational study to assess the role of the Bystander-CPR in affecting neurological outcome in our OHCA population. Introduction: There is a clear rationale for monitoring microcirculatory behaviour during shock since it is the anatomical location of oxygen and substrate exchange, and may not correspond to global haemodynamics. And yet despite over a decade of research and technological advances such monitoring has not reached clinical bedside utility. Analysis of the data is performed offline and too time consuming for clinical use. There is an urgent need for a system to assess the microcirculation at the bedside. We present a novel 5-point grading system (the Point-Of-carE Microcirculation (POEM) scoring system) that can be used at the bedside (using sublingual microcirculatory monitoring).

Objectives: To assess the inter-user variability of the novel POEM scoring system amongst doctors and nurses who may use such technology for clinical practice, and to benchmark POEM scores against traditional offline computer analysis. Methods: The POEM score is an ordinal scale from 1 (worst) to 5 (best), and calculated based on assessment of 4 individual video clips. Online calculator found at: www.POEMscore.com. Thirtytwo naïve study participants from two UK teaching hospitals (Birmingham and London) participated in a standardised 1-hour interactive training session in how to assign POEM scores based on microcirculatory video clips from sublingual incident dark field (IDF) videomicroscopy imaging. They were then asked to assign scores for 5 different video sequences (each of varying clinical status, played in a random order). They were blinded to clinical status. Inter-user consistency and agreement were assessed using intra-class correlation coefficient (ICC) analysis. Blinded expert POEM scores were also validated against offline computer analysis of the same clips using traditional microcirculatory parameters, and the time taken to assign each was recorded.

Results: Raters showed good inter-rater consistency (ICC 0.83, 95 % CI 0.626, 0.976) and agreement (ICC 0.815, 95 % CI 0.602, 0.974) for assigned POEM scores. Expert POEM scores correlated well with offline analysis but took far less time to assign (mean times of 2 minutes versus 44 minutes; p < 0.001).

Conclusions: A new 5-point ordinal scale of microcirculatory function has been tested amongst 'front line' emergency physicians and nurses at two large UK teaching hospitals, and has minimal inter-user variability, even after just 1 hour of training. POEM scores take a matter of minutes to assign, and correspond well to computer-analysis variables. We present for the first time a bedside microcirculatory grading system that is quick, reliable, and gives potentially meaningful clinical parameters that might guide resuscitation. Prospective randomised trials utilising goal directed therapy using the POEM score are required to test its reallife clinical utility. Introduction: Vocal cord palsy is a known postoperative complication following cardiothoracic surgery. 1, 2. Although the incidence is relatively low its existence cannot be ignored and thus its identification necessary in order to avoid any further complications and maintain patient wellbeing. This study aims to look at the incidence of vocal cord palsy following cardiothoracic surgery in a tertiary referral centre and highlight the importance of the speech and language therapist's role in working with this cohort. Objectives: To measure the incidence of vocal cord palsy post cardiac and thoracic surgery and to identify the consequent effects. Methodology. A retrospective analysis, within a tertiary cardiothoracic centre. Data for all patients who underwent either a cardiac or thoracic surgical procedure between December 2015 and April 2016 and were referred to speech and language therapy (SLT) was collected. Vocal cord palsy was identified by Fibreoptic Endoscopic Evaluation of Swallowing (FEES) or Bronchoscopy. Results: A total of 25 patients were seen by the SLT. Patients with vocal cord palsy were identified by FEES and Bronchoscopy; 83 % and 17 % respectively. Six patients assessed presented with vocal cord palsy; patients were post cardiac surgery (4/6) and post thoracic surgery (2/6). The consequence of vocal cord palsy was dysphonia in all the patients and dysphagia in two thirds of patients. The median duration that patients experienced dysphagia was 12 days (range 4-38 days) and dysphonia was 21 days (range 11-38). Conclusion: This review highlights the high prevalence of vocal cord palsy post cardiothoracic surgery. Vocal cord palsy led to high levels of dysphagia and dysphonia. Early identification of these is imperative to ensure patient safety and optimise recovery and quality of life. 7) . 20.4 % were over 75 years old at the moment of the surgery. Figure 86 shows previous cardiac clinical history Figure 87 shows other relevant comorbidities. It was first surgery in 90.5 % of the patients. 68,7 % of the cases were valvular surgery (with or without CABG); 21,4 % were isolated CABG; 5,1 % thoracic aortic surgery; 5,4 % were other surgeries (congenital disease surgery, post AMI complications, pericardiectomy). Tables 61 and 62 summarizes the results of the application of SF-36 questionnaire in the 1382 patients interviewed in the follow up In the comparative analysis we found worse QOL in women than men (p = 0,0001), and a negative correlation between age and QOL (p = 0,0001). Women in our study were significantly older than men (p = 0,0001). We found no differences between the type of surgery and the postoperative QOL, or between surgery or extracorporeal circulation duration and QOL. There was a relationship between NYHA degree during the follow up and the SF 36 Health score Prognostic scores showed an inverse relationship with QOL, but with a low correlation; Pearson coefficient −0,22 (Euroscore), −0,28 (SAPS 3).

Conclusions: In our study, involving 11 hospitals in Andalusia (South Spain), 71,5 % of cardiac surgery patients didn´t show any activity limitation or only a slight limitation before 3 years of follow up. Perceived quality of life decreases as age increases or worsens the functional status of the patients. Conclusion: Concsultation for the severely ill cases in the wards take very much time of the intensivits. It was seen that consultation request was mostly emergent and due to respiratory problems and sepsis. Maybe we need another system like rapid response team for decrease the insivists work and decrease the mortality amd morbidity. The limitation of the beds in ICU is one of the most important problem, and there must be more empty beds for inhospital emergencies.

An external validation study of the qSOFA score to predict inhospital mortality in medical patients with infection and derivation of a new enhanced score using automatically available variables: news-hazard L. Introduction: Sepsis has recently been redefined as´life-threatening organ dysfunction caused by a dysregulated host response to infec-tion´. It is one of the leading causes of mortality internationally. Earlier identification of sepsis means more timely management, reduced length of hospital admission, and prevention of septic shock; ultimately reducing sepsis associated mortality. (1) There is currently no standard diagnostic test for sepsis. Distinguishing sepsis from alternative, uncomplicated infections is pertinent to ensuring an appropriate clinical approach. Systemic inflammatory response syndrome (SIRS) criteria have been used since 1992 to define sepsis (Sepsis 1.0). However they have been found to lack sensitivity or specificity. The quick sequential organ failure assessment score (qSOFA) is an emerging initial assessment method that uses three simple bedside criteria to measure organ dysfunction; altered mental status, respiratory rate ≥22 and systolic blood pressure ≤100 mmHg. The aim is to facilitate earlier recognition of sepsis outside of ITU by prompting the clinician to think, and adequately screen for sepsis. In 14 cases, the diagnosis of pneumococcal meningitis by culture or pneumococcal CRP in CSF was confirmed. In all cases, pneumococcus antigen was positive. Therefore, there were no false negatives. In 59 cases pneumococcal meningitis was not diagnosed, being the final diagnosis a non-pneumococcal bacterial meningitis or another pathology. In all cases, pneumococcus antigen was negative. Therefore, there were no false positives. In conclusion, in our sample of 72 patients, the sensitivity and specificity of the test for S. pneumoniae antigen in CSF was 100 %. The PPV and NPV were also 100 %.

Conclusions: In our series, the sensitivity and specificity of the test for S. pneumoniae antigen in CSF by immunochromatography ( BinaxNOW® test ) was 100 %. The VPP and VPN were also 100 %. These results are similar to those reported previously in the literature. Knowing the reliability of this fast, simple and inexpensive test, will allow to remove unnecessary isolation and to establish a more specific treatment and a better prognoses of the disease.

Evaluation of sensitivity and specificity of different criteria using for diagnosis of burn sepsis and without these (group №2, n = 16). Logistic regression was performed to identify the independent factors for the prediction of early death (7 days and less). We examined sensitivity and specificity with area under the Receiver Operating Characteristic Curve (ROC AUC).

Results: There were no significant differences between two groups for demographics, burn size, inhalation injuries. Fatal outcome came early in the group №2 (mean ICU length of stay 7,5 days vs 10 days, p < 0,05). Organ dysfunction at day 4 was significantly higher in the group №2 (mean SOFA 3,4 vs 2,0; p < 0,05). There were no significant differences in ABA, FSBI and CDBS between two groups, and the highest AUC were for CDBS by the day 6 (AUC 0,687 95 % CI 0,517-0,796 vs 0,567 95 % CI 0,469-0,719 and 0,504 95 % CI 0,315-0,702 for CDBS, ABA and FSBI, respectively). SIRS criteria were significantly higher in the group №2, AUC was 0,745 (95 % CI 0,412-0,889) but specificity and sensitivity was too low (for 2 SIRS criteria is 64,9 % and 12 %, respectively). Independent factors for early death include: more than 20 % immature neutrophils at the day 3, SOFA more than 4 by the day 4, thrombocytopenia less than 100 by the day 4. Conclusion: Patients without specific pathomorphological signs of sepsis have more severe organ dysfunction, greater number signs of systemic inflammation and earlier fatal outcome. Diagnostic model of sepsis by Chinese experts has more sensitivity and specificity for diagnosis of burn sepsis confirmed by autopsy. Immature neutrophils count, thrombocytopenia and SOFA score are stronger risk factors for early death.

The value of neutrophil to lymphocyte count ratio in diagnosing blood-stream infection S. Involved microorganisms: S. Pneumoniae and N. Meningitidis in communitary ABM, S. Epidermidis and Gram-negative bacilli in nosocomial ABM. When comparing both groups, only we observed difference in the C-reactive protein at admission (98 ± 99,5 in nosocomial ABM vs 177,6 ± 71,1 in communitary, p 0,05). There were differences with other variables but were not statiscally significant. So, we observed a higher mortality in nosocomial ABM group but the difference was not statistically significant (p 0,08). Global ICU mortality was 19 % (4) and hospital mortality was 30,4 % (7). Conclusions: The demographic and bacteriological profiles of patients with Acute Bacterial Meningitis have changed in the last years mainly due to the expansion of neurosurgical procedures. Still has a high morbidity and mortality.

The detection of microbial DNA but not cultured bacteria is associated with increased mortality in patients with suspected severe sepsis -a European multi-centre observational study M. Introduction: Sepsis is a leading cause of worldwide mortality. Blood culture results poorly discriminate the mortality risk in critically ill patients with sepsis. Here we aimed to determine whether the detection of microbial DNA in the blood stream of patients with suspected sepsis was associated with mortality.

We performed an analysis of data collected during the Rapid Diagnosis of Infections in the Critically Ill (RADICAL) study (1) . Patients were considered eligible for this study if they developed suspected sepsis and were either in or were referred for treatment to one of nine intensive care units (ICUs) in six European countries.

When initial blood cultures were taken for clinical indications an additional blood sample was obtained for a culture-independent polymerase chain reaction/electrospray ionization-mass spectrometry (PCR/ESI-MS) assay. The results of the PCR/ESI-MS test were not communicated to the treating clinicians. Results: Of the 529 patients analysed in the original study outcome data, blood culture results and PCR/ESI-MS results were available for 439 patients (Table 63) Objectives: To evaluate if a "bundle" consisting of a SIRS and organ failure (SOF)-triage, flow chart response and alert system, and a SIRS/ sepsis training course for all wards nurses improved clinical observations, lead to fewer patients developing severe sepsis, decreased length of stay in the high-level care (LOS) and increased survival. Methods: A before and after intervention study in one emergency and community hospital within the Mid-Norway sepsis study catchment area. All patients with confirmed blood stream infection (BSI) and evidence of sepsis have been prospectively registered continuously since 1994. The severity of sepsis, observation frequency of vital signs, treatment data, LOS and mortality were retrospectively registered from the patients' medical journals until end 2013.

Results: The pre-intervention group was patients with confirmed BSI from Jan 2008 to Dec 2010 (n = 472) whilst the postinterventions group was recruited between Nov 2011 to Dec 2013 (n = 409). The nurses' observation frequency of vital signs increased in BSI patients with and without severe organ failure comparing these periods. The post-intervention group had, in average, 2.4 days shorter LOS. Patients admitted without severe organ failure in the post-intervention group had a lower probability of developing severe organ failure (0.7, 95 % CI 0.4-0.9) than the pre-intervention group. Adjusted for differences in disease severity the post-intervention group also had higher odds of surviving 30 days (OR 2.7, 95 % CI 1.6-4.6).

Conclusion: A sepsis specific triage-, flow chart alert and treatment system was an effective tool to increase ward nurses recognition and early treatment of patients with confirmed BSI. In addition to increased survival, the shorter LOS is important from a hospital perspective in term of resource utilization.

This study was supported by the liaison committee between Nord-Trøndelag Hospital Trust and Nord University Introduction: The Endotoxin Activity Assay (EAA™; Spectral Diagnostics Inc., Toronto, Canada) is a rapid in vitro diagnostic test of the neutrophil's reaction to endotoxin and reflects the endotoxemia 1) . Recently, EAA is used to confirm endotoxemia such as in the EUPHRATES (Evaluating the Use of Polymyxin B Hemoperfusion in a Randomized controlled trial of Adults Treated for Endotoxemia and Septic shock) trial study in North America 2) . However, EAA has not been routinely used to diagnose sepsis, yet.

Objectives: Our hypothesis is EAA is useful to diagnosis for new definition's sepsis due to Gram-negative infection.

Methods: The present study is a single-center retrospective observational analysis. Of all adult patients in whom EAA was measured at our medico-surgical ICU from July 2008 to July 2013, patients with new definition's sepsis in 2016 were included in this study. New definition's sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection which is identified with total SOFA score of 2 or greater 3) . Patients were divided into two groups, 1) with Gram-negative organisms in some cultures and 2) with no Gram-negative organisms in any cultures. Age, sex, body temperature (BT), WBC, CRP, procalcitonin (PCT), SOFA score, and EAA values were compared between two groups. Values are expressed as mean ± SD. Data was analysed by chi-square test and unpaired Students t-test. P values less than 0.05 were considered significant.

Results: Five hundred and twenty seven patients (330 men and 197 women; mean age 66.2 ± 17.3 years) were studied. There were 1) 251 patients with Gram-negative infection and 2) 276 patients with no Gram-negative infection. EAA values and SOFA score were statistically significant differences between Gram-negative infection and no Gram-negative infection (0.40 ± 0.18 vs. 0.37 ± 0.16, p = 0.03, 8.6 ± 4.3 vs. 7.3 ± 4.0, p < 0.0005, respectively). PCT was different but did not reach to statistically differences between two groups (17.5 ± 38.3 vs. 11.8 ± 28.9, p = 0.06) and age, sex, BT, WBC were no significant differences between two groups. Patients and methods: Mechanically ventilated patients with severe sepsis/septic shock, treated in ICU, were included in this prospective observational study. Exclusion criteria were mechanical ventilation for more than 72 hours prior ICU admission and brain death. Clinical and laboratory data were recorded, on a daily basis. Thrombocytopenia was defined as a PLT below 150X10 3 /μl. Thrombocytopenia was considered as mild (100X10 3 /μl ≤ PLT < 150X10 3 /μl), moderate (50X10 3 /μl ≤ PLT < 100X10 3 /μl), or severe (PLT < 50X10 3 /μl) depending on PLT counts. Serum levels of IFNγ, IL-8, ICAM, VCAM, and soluble urokinase plasminogen activation receptor (SuPAR) were estimated by using Luminex xMAP technology.

Results: Fifty-six out of 105 (53 %) patients enrolled in the study were thrombocytopenic at the time of admission in ICU. The overall incidence of thrombocytopenia during ICU hospitalization was 75 %, while mild, moderate, and severe thrombocytopenia developed in 18 (17 %), 15 (14 %), and in 46 (44 %) respectively. Patients with severe thrombocytopenia had higher APACHE score, higher serum ICAM, IL-8 and SUPAR levels, higher incidence of bacteremia and higher probability to present with septic shock as compared with patients with normal platelet counts. Moreover, severe thrombocytopenia was associated with statistically significantly higher hospital mortality. Patients with severe thrombocytopenia showed significant higher serum ICAM (p < 0.001), IL-8 (p = 0.013), and SUPAR (p < 0.001) levels respectively, as compared to patients with normal platelets count, or patients with mild or moderate thrombocytopenia. In multivariate analysis, higher APA-CHE score, thrombocytopenia, and higher serum SUPAR levels were statistically significantly associated with a higher risk of ICU mortality. Enrolled patients were stratified in different groups according to their APACHE II score (APACHE II > 18), PLT counts (PLT ≤ 128.000) and serum SUPAR levels (SUPAR > 8.1). In multivariate analysis, this new scoring system remained the only and most significant factor associated with statistically significantly increased ICU mortality [OR = 8.41, (95 % CI, 4.19 to 16.9), p < 0.0001].

Conclusion: Severity of thrombocytopenia in severe sepsis and septic shock parallels the severity of inflammation and subsequent endothelial dysfunction and is associated with higher mortality.

The and IL-17α, were measured and compared between immediately before the first PMX-DHP therapy, before the second PMX-DHP therapy and after PMX-DHP therapy. Human Cytokine ELISA Plate Array I (Chemiluminescence) are used the plate which 31 of cytokine capture antibodies are coated on 31 wells respectively. Each cytokine value is shown as relative light units of luminescence. Values were expressed as mean ± SD. Data were analyzed by Wilcoxon signed-ranks test. A p < .05 was considered as statistically significant.

Results: All results were provided in about 5 hours after starting this assay. One measurement of all 31 cytokines costed $340.0. TNF-α and MCP-1 values were significantly decreased between immediately before and after the second PMX-DHP therapy (987.7 ± 1194.0 vs. 360.9 ± 426.8, 590.7 ± 486.8 vs. 360.6 ± 368.3, p < .05, respectively). IL-8, IL-12 and MCP-1 were also significantly decreased between before the second and after PMX-DHP therapy (668.6 ± 463.6 vs. 324.9 ± 330.7, 623.7 ± 478.5 vs. 292.3 ± 264.9, 476.1 ± 412.0 vs. 360.6 ± 368.3, p < .05, respectively). There were no statistically significant differences between before and after PMX-DHP therapies in other cytokines.

The present study has some limitations because of a retrospective analysis and numbers of patients. However, Human Cytokine ELISA Plate Array I is a fast and low-cost assay compared with the previous ELISA method. And several cytokines are evaluated with one sample at the same time. This assay could be useful especially for the clinical research because small volumes of sample allow to several cytokines' information at the bedside. Introduction: Critical illness-acquired myopathy in rats is characterized by homogeneous muscle atrophy (1). Conversely, histological abnormalities are heterogeneous: oxidative muscles show patchy alterations (myofascitis, necrosis), while glycolytic types demonstrate normal patterns. Akt and mTOR are key proteins of the anabolic pathway, leading to myocyte growth when activated. Conversely, AMPK and FoxO3 are key proteins of the catabolic pathway, leading to myocyte atrophy when activated. Whether anabolic or catabolic pathway activation is dependent on skeletal muscle type (i.e. oxidative and glycolytic) during critical illness is unknown. Objectives: To characterize activation of the anabolic and catabolic signalling pathways in a long-term rat peritonitis model by skeletal muscle type. Methods: Male Wistar rats were followed for up to 2 weeks after intraperitoneal injection of the yeast cell wall constituent, zymosan or n-saline. Soleus (oxidative, slow twitch muscle), and gastrocnemius (mixed glycolytic-oxidative, fast twitch muscle) were harvested from both zymosan and control groups at 2, 7 and 14 days after the insult. Expression of phospho-(p-) and total proteins were assessed by Western blots. Expression of Akt, p-Akt (p-threonine 308, active form), mTOR, p-mTOR (p-serine 2448, active form), AMPK, p-AMPK (pthreonine 172, active form), FoxO3, and p-FoxO3 (p-threonine 32, inactive form) were assessed at all time points. Results: Weight loss was not statistically different in soleus versus gastrocnemius in the zymosan group (−26 ± 11 % versus −16 ± 9 %, p = 0.17) at day 2. Gastrocnemius displayed a decrease in p-Akt at day 2, and an increase of p-Akt and p-FoxO3 at day 14. Soleus displayed an increase of p-Akt, p-AMPK, and p-FoxO3 at day 2, and an increase of p-AMPK, and p-FoxO3 at day 7. Results are detailed in the Table 64 .

Conclusions: In a rodent model of long-term peritonitis, both oxidative and glycolytic muscles display little change in the anabolic signalling pathway. AMPK (an autophagy activator) is activated while FoxO3, (an autophagy and ubiquitin-proteasome system activator) is inhibited up until day 7 in oxidative but not glycolytic muscle. Introduction: Esophageal pressure (P es ) guided setting of PEEP has been described in ARDS patients 1 either to avoid expiratory alveolar collapse 2 or to promote maximum inspiratory recruitment 3 . The proportion of ARDS patients that may benefit from maximum recruitment strategy and its effects regarding dead space (V D /V T ), shunt, driving pressure (DP), transpulmonary driving pressure (TPDP) and expiratory transpulmonary pressure (TPP exp ) remain unclear. Methods: We included moderate and severe ARDS patients under mechanical ventilation and paralyzed, in the first 12 hours after reaching ARDS criteria. Patients were monitored with esophageal balloon catheter and ventilated with EXPRESS study settings for 1 hour after recruitment maneuver. Then PEEP was modified to obtain an inspiratory transpulmonary pressure (TPP insp-P ) based on P es between 20 and 25 cmH 2 Higher PEEP in Group B led to higher plateau pressure and TPP insp-P , positivation of TPP exp without increase in V D /V T (p = 0.97), shunt (p = 0.84), DP (p = 0.16), TPDP (p = 0.09) or oxygen stretch index (p = 1). However agreement between TPP insp-P and TPP insp calculated from respiratory motion equation and chest wall elastance (TPP insp-E ) was weak with Band-Altman bias (TPP insp-E -TPP insp-P ) = 4.5 ± 8.9 [95%CI −13;22]. Conclusions: P es measurement in moderate to severe ARDS patients distinguishes 2 groups of patients in whom PEEP appears to be taylorized without side-effects. However physiologic studies should assess reliability of transpulmonary measurement based on either P es or chest wall elastance. Introduction: To optimize mechanical ventilation different targets are used including tidal volume (TV), peak and mean airway pressure and PEEP. However, prevention of alveolar collapse not only depends on intra-pulmonary, but also on the extra-pulmonary pressure (EPP). EPP can be estimated by measuring esophageal pressure (EP). Ventilator strategies aiming at optimized trans-pulmonary pressure TPP (difference between intra-and extrapulmonary pressure: TPP = TIP-TEP) have been shown to improve outcome. TPP-guided ventilator setting might be useful in patients with liver cirrhosis and ascites. However, the impact of paracentesis on TPP is poorly investigated.

Objectives: To investigate the impact of high volume paracentesis (HVP; ≥3000 mL) on TPP and on other parameters of pulmonary and circulatory function.

Methods: Analysis of 23 HVP-procedures in 11 patients ventilated with the AVEA Viasys ventilator (CareFusion, USA) capable to measure EP via an esophageal tube. Haemodynamic monitoring with the PiCCO-2-device (Pulsion Medical Systems SE, Feldkirchen, Germany) was available during 17 measurements. Intra-abdominal pressure IAP was determined by intra-peritoneal (IAP_P) and intra-vesical (IAP_V) pressure measurement. High grade esophageal varices had been excluded endoscopically before measurement of EP. Statistics: SPSS 23.

Results: 5 male, 6 female patients, aetiology of cirrhosis alcoholic (n = 7), viral (2) and cryptogenic (2) . Age 51 ± 15 years, APACHE-II 30 ± 9, SOFA 13 ± 4, MELD 25 ± 10. Paracentesis of 4826 ± 1276 mL resulted in marked increases in inspiratory (17.9 ± 8.9 vs. 5.4 ± 13.4; p < 0.001) and expiratory (−3.0 ± 4.7 vs. -15.9 ± 10.9 cmH 2 O; p < 0.001) TPP. In parallel inspiratory (2.4 ± 8.7 vs. 14.1 ± 14.5 cmH 2 O; p < 0.001) and expiratory (12.4 ± 6.0 vs. 24.9x ± 11.3cmH 2 O; p < 0.001) EP significantly decreased. Paracentesis resulted in decreases in IAP_P (5.2 ± 2.3 vs. 11.7 ± 2.0 mmHg; p < 0.001), IAP_V Conclusions: Paracentesis markedly increases inspiratory and expiratory TPP in parallel with a decrease in IAP. Increased IAP before paracentesis resulted in markedly decreased inspiratory and endexpiratory TPP despite ventilation according to the ARDSnet guidelines. To avoid decreased end-expiratory TPP and alveolar collapse in patients with increased IAP, paracenteses and/or higher PEEP-setting should be used. IAP and its changes markedly confound CVP, but neither GEDVI nor CI. Introduction: If the proportional assist ventilation(PAV) level is known, then muscular effort can be estimated from the difference between peak airway pressure and PEEP (△P) during PAV. Namely, P mus, peak, aw = (P aw, peak -PEEP) X (100 -Gain)/Gain. Pressure time product estimated from airway (PTP aw ) = P mus, peak, aw X inspiratory time/2 X respiratory rate [1] .

Objectives: Validation of this hypothesis by using the esophageal pressure time product calculation. Methods: Eleven mechanically ventilated patients who received esophageal pressure monitoring under PAV were enrolled. Patients were randomly assigned to seven PAV assist levels (20-80 %, PAV20 means 20 % PAV Gain) for 15 minutes. Maximal muscular pressure (P mus, peak, es and P mus, peak, aw ) and pressure time product (PTP es and PTP aw ) estimated from △P and esophageal pressure were determined from the last minute of each PAV level.

Results: PAV significantly reduced the breathing efforts of patients with increasing PAV gain (PTP es 197.7 ± 75.8 at PAV20 vs. 71.2 ± 47.0 cmH 2 O•sec/min at PAV80, PTP es, PEEPi 262.7 ± 95.6 at PAV20 vs. 101.2 ± 66.4 cmH 2 O•sec/min at PAV80, p < 0.0001). P mus, peak, aw overestimate P mus, peak, es in PAV of low gain (PAV20) and underestimate in PAV of moderate to high gain (from PAV40 to PAV80). Linear regression analysis revealed that the slope PTP es, br (PTP es per breath)/P mus, peak,es for PTP es, PEEPi is 0.84 (r 2 = 0.8939), for PTP es is 0.70 (r 2 = 0.8760), and PTP aw, br (PTP aw per breath)/P mus, peak, aw for PTP aw is 0.56 (r 2 = 0.9351).

Conclusions: Adjustments should be made when extrapolating PTP aw into PTP es . An additional 25 % should be added when extrapolating PTP es from P mus, peak, aw and an additional 50 % should be added when extrapolating PTP es, PEEPi from P mus, peak, aw , assuming P mus, peak, es and P mus, peak, aw are equal.

Introduction: Airway occlusion pressure is a noninvasive measure of motor neural output. If the airway is occluded, the change in pressure in the pleural space and at the airway open, both are equivalent Objectives: We studied the similarity of effort and work of breathing measure with Pesophageal (Peso) at regular cycles, versus inadvertent airway pressure (Paw) occluded during end-expiration (Paw_occl). Methods: Esophageal, airway pressure and airway flow, sampling 278 Hz, were registered in 10 patients during weaning time, with levels of sedation Ramsay 2, at Pressure support ventilation (PSV) with differents levels of assistance (High:15-23 cmH2O, medium: 10-14 cmH2O, low: 5-9 cmH2O). Respiratory effort was quantified using pressure-time product (PTP/min) with esophageal and occluded cycle (Figure 94 ), and WOB_Occlusion, using the occlusion pressure with the flow of preceding, not occluded, cycles ( Figure  95 ). The work of breathing esophageal referent (WOB_Peso, j/l)) is obtained from integral of Peso versus differential of volume. Also we are calculated Δ Peso and Δ Paw_occl as additional parameter. For all data, the Bland-Altman analysis and linear regression was applied. The results are expressed as mean ± SD, the comparison was made by t-test.

Results: A total of 69 paired measures were obtained. The mean comparison of the respiratory effort and work did not show showed statistical differences for all data, except for low assistance (Table 66) . A good correlation between both measures methods was observed for PTP and WOB (R = 0,91 and 0,87; respectively). The mean bias was for PTP and WOB: 42,64 (±58,95) cmH2O*sec/min and 0,07 (±0,29) j/l, respectively; and the 95 % limits of agreement were −72,91 to 158,19cmH2O*sec/min and −0,50 to 0,64 j/L, respectively; this indicates wide dispersion. Conclusions: Airway occlusion pressure is a noninvasive procedure that could be useful to assess the effort and work of breathing patients during mechanical ventilation. Introduction: Chest wall elastance (Ecw) is thought to increase in prone (P) as compared to supine (S) position in ARDS patients (1) (2) (3) . This makes respiratory system elastance (Ers) not reflecting lung elastance (E L) . Little is known about the changes of Ecw, E L and lung resistance (R L ) when moving the patient from the supine to the prone position via the lateral (L) position (4) (5) .

Objectives: The goal of present study was to measure Ecw, E L and R L in ARDS patients in S, L and P position during the proning procedure. Methods: ARDS patients intubated, sedated and paralyzed with an indication of P positioning were included. Mechanical ventilation was delivered in volume controlled mode with constant flow inflation. End-inspiratory pause of 0.100 sec was set during the breathing cycles. Ventilator settings were unaltered during the procedure. Airway and esophageal pressures and airflow were continuously measured during 30 minutes in S, then during 1 minute in L and 30 minutes in P. The side for the lateralization was that selected by routine practice (in the opposite side from central venous line). Prone positioning was performed manually by 3 caregivers. Ecw, E L and R L were obtained by fitting the first order equation model to flow and pressures signals. Values are expressed as mean ± SD. Results: Fifteen patients (8 males) of 66 ± 12 years, SAPS 2 41 ± 12 and SOFA 7 ± 2 were included 2 ± 3 days after ARDS criteria (12 moderate and 3 severe) were met. Tidal volume averaged 6 ± 0.6 ml/kg predicted body weight, PEEP 11 ± 3 cmH2O, FiO2 73 ± 15 %, PaO2/FiO2 ratio 116 ± 19 mmHg. The cause of ARDS was pneumonia in 9 cases, undetermined in 6 cases. Side positioning was the right in 8 and the left in 7 patients. The results are shown in the Table 67 .

Conclusions: During prone positioning in ARDS patients, as compared to S we observed an increased R L and E L in L and increased Ecw in P. Introduction: Driving pressure of respiratory system (ΔP) is defined as the difference between plateau pressure (Pplat) and total positive end-expiratory pressure (PEEPtot) measured after endinspiratory and end-expiratory occlusion, respectively, on airway pressure signal. ΔP has recently been shown as a strong predictor of mortality in patients with acute respiratory distress syndrome (ARDS) (1) . Most of the studies involved in this demonstration measured Pplat 0.5 sec after onset of endinspiratory pause according to the ARMA trial (2) and used PEEP set on the ventilator (PEEPvent) instead of the PEEPtot. This does not take into account slow decay of airway pressure after endinspiratory occlusion and instrinsic PEEP, respectively. Objectives: The aim of the study was to compare ΔP when Pplat was measured at different times after end-inspiratory pause and whether PEEPtot or PEEPvent were used. Our hypothesis was that ΔP was *P < 0.05 versus supine **P < 0.01 versus supine higher with Pplat measured at 0.5 and PEEPvent than with any other combinations of Pplat and PEEPtot. Methods: A retrospective analysis of patients with ARDS in whom respiratory mechanics was measured. Most of the patients had recordings at two levels of PEEPvent. Data were analyzed with Acqknowledge software. Pplat was measured at 0.5, 1 and 2 seconds after end-inspiratory pause. PEEPtot was measured after a 3-sec end-expiratory pause. The low-PEEP and high-PEEP measures pertained to PEEPvent ≤10cmH20 and PEEPvent > 10 cmH20, respectively. The primary outcome was the comparison of ΔP calculated as Pplat 0.5 sec -PEEPvent : ΔP reference versus ΔP computed as Pplat 2 sec -PEEPtot: ΔP physiologic . The values are expressed as mean ± SD and are compared by using signed rank test for paired values, ANOVA for repeated measures and Bland-Altman. Results: Twenty-three patients were analyzed. ΔP reference was significatively higher than ΔP physiologic in low and high PEEP groups: 13.9(±5.6) vs 10.1(±3.5) cmH2O (p < 10 −5 ) and 13.5(±4. Introduction: End Expiratory Lung Volume (EELV) is reduced in ventilated patients especially in patients with acute respiratory failure (ARF). Recruitment maneuvers and different levels of PEEP are used to restore EELV and to improve oxygenation. However; no matter how useful is the value of EELV in clinical practice his bedside measurement at baseline, at different levels of PEEP and after recruitment maneuvers is cumbersome. We measured the EELV with two techniques and we noted the technical problems and pitfalls. Methods: We measure the EELV in patients with ARF at two levels of PEEP (8 and 16 cm H2O) with two techniques. With a dilutional Nitrogen wash in and wash out method using the Carescape R860 ventilator (GE) and by measuring the expired lung volume after a sudden release of PEEP to zero PEEP (ZEEP). Specifically after 10 min of stabilization at each level of PEEP we performed an expiratory hold and we decrease the ventilator frequency to zero to obtain sufficient time for a complete expiration. Therefore, by releasing the expiratory hold we permit the return of the lung to his passive lung volume at ZEEP(passive FRC).The expired volume was calculated by the integration of the expiratory flow. We call this volume ΔEELV. Correlation and Agreement Fig. 96 (abstract A1040) . Variation of Driving presure at low PEEP between the two methods with the Bland and Altmann analysis were performed. Results: 22 mechanically ventilated patients were studied. At ZEEP EELV was low (1276 ml) 43 % of the predicted. Good correlation was found between the two methods (1637 ± 556 and 1816 ± 521 ml respectively) when EELV was measured at 8 cmH2O of PEEP(R = O.92, p < 0.0001). On the contrary at 16 cmH2O of PEEP a wider variability and less agreement was noticed between the EELV values(2040 ± 703 vs. 2411 ± 597, R = 0.86, p < 0.001). Technical problems with Carescape were spontaneous breathing attempts or asynchrony in the patient-ventilator interactions leading to instable VO2-VCO2 measurements. On the contrary with the release of PEEP method high expired volumes from PEEP to ZEEP induced high expired flow exceeding the 1lit/sec, thus affecting the linear sensitivity of the ventilator's pneumotachograph.

Conclusions: The measurement of the EELV remains a precious parameter for the ventilatory management of the ICU patients but his measurement is still far away to be accurate by both techniques at bedside.

Low end-expiratory trans-pulmonary pressure is associated with lung collapse P. Somhorst, D. Introduction: The open lung concept aims to reduce lung injury due to cyclic opening and closing of alveoli. 1 Finding the 'optimal' PEEP to maintain an open lung proofs difficult and patient-specific. 2 We hypothesize that targeting positive trans-pulmonary pressure (Ptp) at end of expiration (PtpEE) may prevent collapse. We used Electrical Impedance Tomography (EIT) clinically to optimize PEEP, visualizing over-distention and collapse. 3 Objectives: To show the association between collapse and low PtpEE, as visualized by EIT. Methods: We retrospectively analyzed data of ten patients with acute respiratory distress syndrome (ARDS) who underwent measurement of the Ptp and EIT due to clinical considerations. A PEEP trial was performed to identify optimal ventilator settings. Esophageal pressure (Pes) was measured using endo-esophageal pressure balloons (Cooper Surgical, Germany, or Sidam, Italy). The Ptp was calculated as the continuous difference between the airway pressure and Pes. EIT was measured at the 5th/6th intercostal space (Dräger, Germany) and analyzed using specialized software (Dräger, Germany). Collapse is defined as a local decrease in ventilation after a reduction in PEEP; over-distention is a local decrease in ventilation after an increase in PEEP. 4 Results: Collapse was associated with a lower PtpEE at lower PEEP levels. For most patients (8/10), collapse occurred when PtpEE was ≤ 2 cmH2O. Collapse was also present at a PEEP level of 15 cmH2O. Inversely, we showed that collapse and over-distention can occur simultaneously (Fig. 101) . The PtpEE was strongly correlated to the PEEP level (r2 = 0.93, p < 0.001; corrected for each individual patient). The overall regression is shown as a dashed line. Conclusions: For most ARDS patients, collapse did not occur when PtpEE was above +2 cmH2O. In addition, PEEP increase in order to prevent collapse may induce over-distention due to heterogeneity in the ventilation distribution in each patient. Introduction: Medical experts recommend keeping plateau pressure below 30 cm H 2 O to avoid ventilator-induced lung injury in patients with acute respiratory distress syndrome (ARDS). Transpulmonary pressure (Ptp), the difference between alveolar and pleural pressure, has been measured as a surrogate for plateau pressure for lung protective strategies. However, placement of an esophageal balloon catheter is required to measure esophageal pressure. Objectives: We investigated the relationship between Ptp and ventilator waveform parameters according to the strength of spontaneous breathing effort. Methods: Eight patients (four patients with ARDS and four with non-ARDS) mechanically ventilated with AVEA® were included in this study. An esophageal balloon catheter (AVEA® SmartCath® esophageal balloon) was placed to measure esophageal pressure. We evaluated the relationship between ΔPtp (difference between inspiratory and end expiratory Ptp) and peak inspiratory flow, or Ti ratio (percentage of time until peak negative esophageal pressure to total inspiratory time). Spontaneous breathing effort was categorized as strong or weak and was analyzed with inspiratory waveform using VOXR data management software. Results: Although there was no significant relationship between ΔPtp and peak inspiratory flow (R = 0.450, P = 0.080), a significant correlation was found between ΔPtp and Ti ratio (R = 0.720, P = 0.002). Median ΔPtp and Ti ratio were significantly higher in patients with strong spontaneous breathing effort compared with those with weak breathing effort (24.3 vs. 13.4 cmH2O, P = 0.012, and 0.50 vs. 0.21, P = 0.008, respectively). In patients with strong spontaneous breathing effort, median ΔPtp was higher in the ARDS group compared with the non-ARDS group (15.3 vs. 3.5 cmH2O, P = 0.029).

Conclusions: Measuring Ti ratio and ventilator waveform parameters may be helpful to estimate Ptp. Appropriate sedatives, analgesics, or muscle relaxants may be required to limit Ptp in cases of higher Ti ratio in patients with ARDS.

In vivo calibration of the esophageal balloon catheter: a simplified procedure F. Introduction: A calibration procedure has been recently proposed to obtain reliable esophageal pressure (Pes) measurements in mechanically ventilated patients [1] . This procedure helps optimizing esophageal balloon filling and removing esophageal artifacts, but is timeconsuming.

Objectives: To test accuracy of a simplified procedure, designed according to average values of esophageal elastance (Ees) and minimum appropriate filling volume (Vmin) previously observed [1] . Methods: In 15 patients under Pressure Controlled Ventilation, 35 pairs of end-expiratory and end-inspiratory calibrated Pes values (Pes,cal) were obtained with the standard procedure, consisting in measure of Ees and detection of Vmin and Vbest (filling volume associated with the largest tidal swings of Pes): Pes,cal = Pes -Ees * (Vbest -Vmin). "Simplified" calibrated Pes values (S-Pes,cal) were also obtained with a simplified procedure based on detection of Vbest and on the assumptions that Ees = 1 cmH 2 O/ml and Vmin = 1 ml: S-Pes,cal = Pes -(Vbest -1). We used the Nutrivent catheter (Sidam, Italy), equipped with an esophageal balloon that is 10 cm long and has a 10 ml nominal volume. Results: In the 35 conditions tested, Vmin was 1.4 ± 0.5 ml, Vbest 4.2 ± 1.9 ml and Ees 1.2 ± 0.3 cmH 2 O/ml. At optimal filling volume (Vbest), difference between Pes and Pes,cal was 3.1 ± 1.9 cmH 2 O (range 0.0-8.0). S-Pes,cal strictly correlated with Pes,cal (R 2 = 0.97; p < 0.0001); difference between S-Pes,cal and Pes,cal was −0.1 ± 0.9 cmH 2 O (Figure 102) . Conclusions: When optimal filling of the esophageal balloon is adopted in mechanically ventilated patients, absolute values of Pes are affected by significant esophageal artifact. A simplified calibration procedure seems to be adequately accurate in removing this artifact and suitable for clinical use. Introduction: Mechanical ventilation unloads the inspiratory muscles in case of high work of breathing to prevent development of muscle injury and patient discomfort. On the other hand, over-assist is associated with disuse atrophy and patient-ventilator asynchrony. Two indices for assessing respiratory muscle effort have recently been published. Patient-ventilator breath contribution (PVBC) index provides an estimation for the percentage of the total work of breathing performed by the patient related to the total work of breathing (patient + ventilator). Neuromuscular efficiency (NME) or pressureelectrical activity index (PEi) expresses how much pressure the inspiratory muscles generate (Pmus) for each microvolt of diaphragm electrical activity (EAdi).

Objectives:

The aim of the current study was to assess the repeatability of both PVBC and NME and investigate how these indices changes in time in ventilated ICU patients. Methods: We included 31 mechanically ventilated adult ICU patients with a dedicated naso-gastric feeding tube for assessing diaphragmatic EMG activity (EAdi catheter). PVBC and NME were calculated at inclusion, after 12, 24 and 72 hrs and repeated 5 times each with a 1-minute interval. PVBC was calculated by (tidal volume no assist/ EAdi no assist)/(tidal volume assist/Edi assist). NME was calculated by measuring change in airway pressure divided by amplitude of the electrical activity during end expiratory occlusion (delta Paw/Edi). Results: The repeatability coefficient (RC) of PVBC and NME was 20 % and 0.34 cm H 2 O/uV respectively. Median PVBC at T = 0 was 76 % and decreased until 71 % at T = 72. In the same period, the mean NAVA level decreased (0.88 until 0.73) and mean EAdi peak increased (14.2 until 18.9uV). Five patients had a PVBC index >100 %; four of them had a calculated pressure support (mean EAdi peak x NAVA level) < 5cmH 2 O. The median NME was 0. Conclusions: We showed a repeatability of 20 % for PVBC and 0.34cmH 2 O/uV for NME. This means that the absolute difference between two repeated measurements lies between this value with a probability of 95 %. For example, with a calculated PVBC of 70 % it is expected that 95 % of the subsequent measurements will be between 50-90 %. NME was much more heterogeneous which indicates that neuromechanical coupling changes during ICU stay in an unpredictable manner. Pressure developed by the diaphragm in our patients appears within physiological limits.

Use of sigmoid regression for determining the optimal balloon volume in esophageal pressure monitoring: a bench and clinical feasibility study Introduction: Esophageal pressure (P ES ), which has been used as a substitute for pleural pressure, is commonly measured by catheter with air-filled balloon. The accuracy of measurement depends on the proper balloon volume (V B ). Assessment of optimal V B is difficult in clinical settings because the surrounding pressure of the balloon cannot be directly measured. In the present study, we introduced a sigmoid fitting method for determining the optimal V B .

Objectives: To assess the accuracy of optimal V B measured by sigmoid fitting and to evaluate the feasibility of this method in clinical practice.

Methods: Six randomly selected esophageal balloon catheters (Cooper catheter, Cooper Surgical, USA) were tested in a bench model with the lung and the pleural cavity during simulated mechanical ventilation. The balloon was progressively inflated in 0.5 mL increments from 0 to 2.5 mL, and pressure in the balloon pressure (P B ) and in the pleural cavity (P C ) were measured. Balloon transmural pressure (P TM ) was calculated as P B -P C . Balloon pressure-volume was fitted by a sigmoid regression: V B = a/[1 + e -(P-b)/c ], where a = the vertical distance of the upper asymptote, b = the pressure at the midpoint between zero and a, and c = the pressure range with the greatest volume change ( Figure 103A shows a sample curve). The optimal V B was predicted by zero P TM and zero (P B -b). Bland-Altman´s analysis was used to assess the accuracy of the optimal V B predicted by P TM and P B . The balloon catheter was introduced into lower third of esophagus in 10 patients with mechanical ventilation, and the balloon was inflated as the same sequence as that in the bench study. P ES and V B were also fitted by the sigmoid regression ( Figure 103B shows a sample curve) and the optimal V B was predicted by zero (P ES -b). At each V B , dynamic occlusion test was performed, and ratio of changes in P ES and airway pressure (ΔP ES /ΔP AW ) was calculated.

Results: In the bench study, the best-fit coefficient R 2 of sigmoid regression ranged from 0.918 to > 0.999 with a median (interquartile range, IQR) of 0.989 (0.972, 0.998). The natural logarithmically transformed bias (and lower to upper limit of agreement) in optimal V B predicted by P TM and P B was −0.001 (−0.012 to 0.010). In the clinical study, 20 V B tests were performed. R 2 of sigmoid regression ranged from 0.770 to >0.999 with a median (IQR) of 0.986 (0.972, 0.995). The optimal V B was 1.2 (1.2, 1.2) mL. The b value (R = 0.847, p < 0.001) and predicted optimal V B (R = 0.348, p = 0.028) significantly correlated with respective P ES measured at the V B with the best ΔP ES /ΔP AW ratio.

Conclusions: In the determination of balloon pressure-volume response, the performance of nonlinear sigmoid fitting was excellent in Introduction: The use of more physiological tidal volumes (6-8 mL/ Kg of ideal body weight) during general anesthesia can minimize the risk of lung injury but may be associated with increased atelectasis. A recent meta-analysis has suggested that high driving pressure and PEEP level changes that result in an increase of driving pressure are associated with more postoperative pulmonary complications (1).

There is no consensus, however, on how to tailor the level of PEEP to best suit each patient. Objectives: Our primary objective is to evaluate the variability of PEEP titrated by EIT in healthy patients submitted to elective abdominal surgery. Our secondary objective is to compare the consequences on lung mechanics and on the formation of atelectasis during abdominal surgery in two groups: titrated PEEP or PEEP of 4cmH 2 O.

Methods: Forty patients will be allocated into two groups: laparoscopic (n = 20) or open surgery (n = 20). After induction of anesthesia and neuromuscular blockade, and before insufflation of abdominal cavity, all patients will be submitted to a recruitment maneuver (RM) in pressure-controlled ventilation mode for two minutes followed by a decremental PEEP titration starting at PEEP of 20 and diminished in steps of 2 cmH 2 O. Optimal PEEP is defined as that with the best compromise of atelectasis and overdistention as measured by EIT. Patients in each subgroup will be randomized to one of two ventilatory strategies during intraoperative period:

(1) PEEP chosen by the PEEP titration procedure (titrated PEEP);

(2) PEEP set at 4 cm of H 2 O (PEEP4). A chest CT will be performed one hour after extubation. A density range of −200 to +100 Hounsfield units (HU) was used to define atelectasis.

Results: Thirty nine patients have been recruited. The median of titrated PEEP was 12 (IQ 10-14) (Table 68) . A weak correlation between BMI and titrated PEEP (R 2 = 0.37) is shown in figure 104 . Lung compliance was significantly lower and driving pressure was significantly higher at baseline, with PEEP = 4 and before RM, when compared to same measures using titrated PEEP during PEEP titration (Table 69) . During surgery, compliance (p < 0,01) and driving pressure (p < 0,01) were also significantly different between PEEP4 and titrated-PEEP group ( figure 105 ). Lung collapse evaluated through lung CT after extubation presented less non-aerated lung tissue in patients submitted to mechanical ventilation under EIT-titrated PEEP.

Conclusions: In this sample of 39 patients, the individualized value of PEEP titrated by EIT had a great variability. PEEP titrated by EIT was able to reduce both lung collapse and driving pressure. Introduction: Lung protective ventilation strategies could improve clinical outcomes in patients undergoing surgery. These strategies did not include specific goals for oxygenation. There is increasing recognition of potential harmful effect of hyperoxia in critically ill patients. However, little is known about current oxygen management during surgery.

Objectives: To describe current oxygen administration during general anesthesia in Japanese hospitals. Methods: A multicenter cross-sectional study was conducted. We screened all consecutive adult patients (≥16 years) who received general anesthesia from 14 to 18 September 2015 or from 9 to 13 November 2015 at the participating hospitals (each participating hospital could choose whichever was more convenient). Ventilator settings and the corresponding vital signs were collected 1 hour after the induction of general anesthesia. We investigated the prevalence and risk factors for excess oxygen exposure ( 11.6 ± 2.5 9.9 ± 3.3 9.2 ± 2.5 9 ± 1.9

Driving Pressure with Titrated PEEP (cmH 2 O) 5.3 ± 0.7 5.2 ± 0.7 6 ± 1.3 6.5 ± 1.1 Fig. 105 (abstract A1047) . Driving pressure Background:. Acute kidney injury (AKI) is common and is associated with significant morbidity and mortality after liver transplantation (LT). Although the creatinine value is highly specific to estimate renal dysfunction, an inadequate sensitivity of creatinine level is demonstrated, particularly in early stage AKI. Cystatin C is founded to be a stronger predictor of the risk of cardiovascular events and death than creatinine. We aimed to determine whether pretransplant serum levels of cystatin C predict 30-day major cardiovascular events (MACE) and all-cause mortality in LT recipients with normal serum creatinine values. Methods: Between May 2010 and October 2015, 1187 consecutive LT recipients (mean age: 53 years; 75 % male; 97 % living-donor LT) who have pretransplant creatinine level < 1.4 mg/dL were retrospectively evaluated. The 30-day MACE was a composite of troponin I > 0.2 pg/mL, arrhythmias, congestive heart failure, death, cerebrovascular accidents.

Results: There was a 19.1 % 30-day MACE event and 5.2 % of LT recipients were dead during a median of 2.1 years follow-up. Mean values of cystatin C and creatinine were 0.91 ± 0.37 mg/dL and 0.77 ± 0.21 mg/dL, respectively. The risk for a 30-day MACE event increased significantly with increasing quartiles of cystatin C; hazard ratios ranged from 1.13 to 2.73 for the highest versus the lowest quartile (P < 0.001 for trend). The Kaplan-Meier curves showed that the highest quartile (cystatin C > 1.0 mg/dL) had a significantly worse survival rate than the lowest quartile (cystatin C < 0.7 mg/dL) (logrank P = 0.023). However, pretransplant creatinine level showed neither increasing MACE event rate nor worse survival rate with increasing quartiles of creatinine values (P = 0.094 for trends, log-rank P = 0.082, respectively). Conclusions: Our results demonstrate that pretransplant cystatin C levels were significantly and progressively associated with 30-day MACE and all-cause mortality in LT recipients with normal serum creatinine values, in contrast, the creatinine levels were not significant and gradual predictor of adverse clinical outcomes. were decreased (P < 0.001), and prevalence of postoperative AKI was increased (Q4: 44 %, Q3: 52 %, Q2: 57 %, Q1: 63 %, respectively, P < 0.001). Odds ratios for AKI ranged from 1.39 to 2.16 for the highest versus the lowest quartile (P < 0.001 for trend). On the multivariate logistic analysis, low MAP was an independent risk factor of the postoperative AKI (P < 0.001), after adjusting factors of age, sex, body mass index, diabetes, hypertension, creatinine, QTc interval, MELD score, B-type natriuretic peptide, beta blocker uses, intraoperative red blood cell uses, postreperfusion syndrome, and cyclosporine uses.

Conclusions: Our results demonstrate that pretransplant low MAP was significantly and progressively associated with the postoperative AKI in LT recipients with normal serum creatinine values, therefore, our findings may assist in determining the optimal perioperative management of patients to prevent postoperative AKI.

Introduction: Patients undergoing cardiac surgery often develop, in the post operative period, pulmonary impairment and abnormalities gas exchanges (1) . Lung ultrasound (LUS) examination may detect main pulmonary abnormalities at the bedside of the patient (2). To increase bronchial drainage and help lung reaeration, physiotherapy treatment is daily applied starting from the first day after cardiac surgery.

Objectives: Our study was to evaluate if physiotherapy treatment was able to induce changing in lung ultrasound pattern in the postoperative patients. Compared total loss of aeration, before and after treatment, we identified a significant increase of rearation after physiotherapy (p = 0,046) evaluated with Wilcoxon test. Conclusions: Our results confirm an elevate rate of loss of aereation in patients after cardiac surgery. Physiotherapy may induce increase of reareation when evaluated with LUS even thought it is not able to reduce consolidation. Introduction: The therapy of malignant liver diseases has changed over the last 100 years. During this period the frequency of liver resection has increased with great improvement in morbidity, mortality and long-term survival. [1] Thereby, the duration of liver transection and the amount of perioperative blood loss are of great importance for postoperative recovery time and therefore they are measures for choosing the optimal resection method. [2] Furthermore, the release of cytokines, chemokines, and stress hormones correlates with postoperative infection and organ dysfunction [3] . To minimize cell damage and limit apoptotic cell death the so called heat shock response is initialized by various body cells as countermeasure to increased stress levels [4] . Moreover, Pittet et al. showed a positive correlation between the small heat shock protein 27 (HSP72) serum levels and survival after severe trauma. [5] Objectives: Measurement of HSP27 could give an insight about pathological mechanism and their counter regulations of the liver. Furthermore the HSP27 serum level should be correlated with the transection speed of the two resection methods CUSA and Stabler. . Immediately after collection, samples were aliquoted, snap frozen and stored at −80°C until further analyzation. To quantify HSP27 in serum commercially available ELISA kits from R&D (DuoSet IC) have been used according to the manufacturer's protocol. Furthermore the duration of transection and the resection surface expressed as cm 2 /s were recorded. Results: During surgery a significant increase in HSP27 levels was detected in patients undergoing Stabler hepatectomy or CUSA resection (n = 40, p < 0.05). During postoperative ICU stay, HSP27 concentrations decline to levels comparable before surgery. The transection speed was significant faster in patients undergoing stapler resection compared to the CUSA method (p < 0.0001). The mean length of ICU stay after liver resection was in both groups 2 days. Conclusions: Our data show increased levels in serum of HSP27, which might reflect the body's countermeasure to increased systemic stress levels during hepatectomy. Moreover the HSP27 levels are in both groups equal high during surgery even though the resection conducted with the Stabler is significant faster than CUSA. Introduction: Performing laparoscopic surgery using carboxipneumoperitoneum usually accompanied with a moderate increase of the concentration of carbon dioxide at the end of expiration, as well as higher peak airway pressure that easily manages to compensate by the correction of ventilation parameters. In the postoperative period marked a fairly long recovery of baseline respiratory function associated not only with the post-operative pain, but with the restriction of the lung as a result of intraabdominal hypertension. Objectives: Assess the impact of prolonged pneumoperitoneum during laparoscopic surgery on respiratory function and to follow the dynamic of its rehabilitation.

The study included 94 patients (42 men and 52 women) in the age of 60.3 years (min 32, max 86), operated in Moscow municipal hospital №4. The volume of surgical procedures: gastric resection (n = 18), gastrectomy (n = 5), pancreatoduodenal resection (n = 8), hemicolectomy (n = 39), resection of the sigmoid colon (n = 18), anterior resection of rectum (n = 6). Depending on the surgical access patients were divided into two study groups: 1st -basic -(n = 33) group -laparoscopic procedures, 2nd -control -(n = 61) group of traditional laparotomy. All patients were under equal anesthesia during surgery: combined general anesthesia (sevoflurane + fentanil) and epidural infusion of 0.2 % ropivacaine solution, as well as myoplegia; postoperative multimodal analgesia: nonsteroidal anti-inflammatory drugs, antispasmodics, epidural analgesia. A study of respiratory function was carried out in four stages: 1 -before surgery, 2 -2nd, 3 -6th, 4-11th day after surgery. Results: In patients of both groups to the second stage of study determined a significant reduction of volume parameters of respiratory function (VC, FVC, FEV, FEV1, MEF, MVV etc.). For example VC decreased in patients of group 1 by 41 % against the initial values, and 52 % of patients in group 2 (dynamics presented in the diagram). Similarly changes in VC there is a decrease of all volume parameters: FEV1 for the second phase decreased by 39.1 % in group 1 and 48 % in group 2; MEF decreased by 61.5 % and 74.3 % in the first and second groups, respectively. However, in addition to a statistically smaller decrease in the absolute values of volumetric parameters of respiratory function in the 1st group, we found them more intense recovery. Conclusions: Reducing the volume indicators of respiratory function after extensive laparoscopic surgery is less than after similar in volume laparotomy. Recovery of acquired restrictive respiratory disorders is more intense and after laparoscopic surgery. At the same time in either group studies we have not observed a complete rehabilitation the initial levels of respiratory parameters, even after 11 days after surgery. Introduction: Critically ill patients sometimes need laparoscopic surgery. It has been reported that steeped head-down position could increase intracranial pressure during robotic surgery. But we don´t know whether mild Trendelenburg position and carbon dioxide pneumoperitoneum cause intracranial hypertension. We conducted a prospective observational study. Objectives: The aim of our study was to investigate the change of optic nerve sheath diameter (ONSD) in head-down position during carbon dioxide pneumoperitoneum. Methods: We included patients scheduled to undergo laparoscopic gynecological surgery. Exclusion criteria were ocular disease and central nervous system diseases. ONSD were measured 3 mm sagittal behind the globe We assessed ONSD after tracheal intubation (T baseline), after pneumoperitoneum and Trendelenburg position (T 0 ) and every 30 minutes (T 30 , T 60 , T 90 , T 120 ). Anesthetic management were standardized. Results: Twenty seven patients were enrolled in this study. Four patients were excluded from analysis because it was difficult for us to measure ONSD. The degree of head-down angle was 13.8 ± 2.61. ONSD is significantly higher than baseline after pneumoperitoneum and Trendelenburg position ( Figure 109 ). Conclusions: Carbon dioxide pneumoperitoneum and Trendelenburg position increased intracranial pressure even if the head-down angle was mild. Introduction: Monitoring the anticoagulant effect of unfractionated heparin (UFH) is mandatory. This monitoring can be done by the mean of the activated partial thromboplastin time (aPTT) or by anti-Xa levels measurements. Compared with anti-Xa levels testing, aPTT is more frequently impacted by preanalytic variables and biologic factors (increased levels of acute phase reactants, consumption coagulopathy) often encountered among critically ill patients. We studied the agreement of both tests results in unselected critically ill patients. Objectives: To study the agreement of both tests results in unselected critically ill patients. Methods: aPTT and anti-Xa levels were simultaneously monitored in patients treated by continuous intravenous infusion of UFH. Blood samples were drawn into sodium citrate tubes (Greiner Bio-One SAS, France). aPTT was measured with TriniCLOT Automated aPTT reagent (Tcoag, Ireland) and anti-Xa levels with BIO-PHEN HEPARIN (LRT) (HYPHEN Biomed, France) . An aPTT of 2-3 times the control and anti-Xa levels between 0.3-0.7 IU/ml were defined as therapeutic. Results: Forty-four patients (mean age 71.13 ± 15.7 years; mean SAP-SII 39.1 ± 14.4) were included. Reasons for admission were medical in 31, surgical in 13. The indications for UFH therapy were atrial fibrillation (26), venous thromboembolism/pulmonary embolism (13), thrombophilia (2), acute coronary syndrome (1), arterial thrombosis (2). Paired measurements of aPTT and anti-Xa were performed on Page 540 of 607 353 samples. Linear regression analysis was used to evaluate the relationship between aPTT and anti-Xa. The correlation between aPTT and anti-Xa levels was low (r = 0.495 ) Concordant aPTT and anti-Xa values were observed in 213 (60.3 %) data pairs. aPTT was discordantly high in 96 (29.9 %) data pairs and discordantly low in 44 (12.5 %) ones. Considering anti-Xa as gold standard, monitoring anticoagulation treatment by aPTT leads to a high risk of misdosing. aPTT is frequently impacted by biologic factors. Although less commonly, anti-Xa levels can also be influenced by biological cofounders. Poor correlation between aPTT and anti-Xa could result from lterations in FII and FVIII activity.

Conclusions: Use of aPTT and anti-Xa levels to guide heparin therapy may lead to different estimates of UFH concentration in the same patient.

Both aPTT and anti-Xa have limitations when used for UFH monitoring and may not accurately assess anticoagulant status. Further investigation (using thromboelastometry or thrombin generation assays) could be useful to determine the optimal anticoagulation testing protocol in critically ill patients.

Note: This abstract has been previously published and is available at [1] . It is included here as a complete record of the abstracts from the conference. Introduction: The severe capillary leak-induced respiratory and renal failure limit large-volume resuscitation with crystalloids and blood components. The combined use of low volumes of crystalloids and "damage control resuscitation" (DCR), a blood product resuscitation goal of a 1:1:1 ratio of packed red blood cells (PRBC), fresh frozen plasma (FFP) has recently been applied to obstetric patients in hemorrhagic shock. Another important consideration is the association of FFP with the risk of transfusion-related acute lung injury (TRALI), a major cause of death after transfusion. This risk is not present with the use of prothrombin complex concentrate (PCC) as the antibodies responsible for TRALI are removed during the manufacturing processes. Methods: Our research involved 51 patients with massive bleeding after Cesarean section. Patients were divided into 2 groups: 1 st group contained 10 patients as a treatment of massive bleeding with coagulopathy was scheduled PCC in a dose of 1 ml/kg (25 IU/kg), packed red blood cells (PRBC). 2 nd group (41 patients) received fresh frozen plasma (FFP) in a dose of 20 ml/kg and PRBC. Basic infusiontransfusion therapy was administered according to the protocols of hemorrhagic shock treatment in obstetrics. Evaluation of the functional state of the hemostasis system was carried out using lowfrequency pyezoelectric thromboelastography (LPTEG) on admission to hospital and every 2 hours after the patient´s admission until normalization of hemostasis state.

Results: According to LPTEG indicators obstetric patients with massive bleeding has a statistically significant abnormality in all parts of hemostatic system: platelet aggregation -Intensity of contact coagulation (ICC), the coagulation -Intensity of coagulation drive (ICD), clot maximum density (MA) and fibrinolytic activity -Index of retraction and clot lysis (IRCL Introduction: Both anaemia and transfusion of red cells (as defined by WHO criteria) (1) have been associated with adverse outcomes, and the potential for anaemia to be a marker of a greater disease burden is frequently raised in discussion. Cohort studies of patients aged >65 years demonstrate that anaemia is associated with increased mortality (2) . Anaemia is also associated with a variety of morbidities in older people, being linked with an increase in hospitalisation, poorer physiological, physical and cognitive function, development of Alzheimer´s and Parkinson´s diseases, depression, falls and hip fracture rates. We aimed to investigate whether anaemia was associated with adverse outcomes, increased lengths of stay and increased overall mortality in our ICU cohort. We also thought it would be interesting to know if there was a difference in haemoglobin level depending on the specialty in which the patients were admitted -hereby defining the physiology of their anaemic process. Methods: We conducted a retrospective review of all patients over the age of 65 years that were admitted to the Victoria Infirmary, Glasgow between 01/01/2005 and 31/12/2014 using the Wardwatcher national database. We looked at admitting specialty (Medicine or Surgical), haemoglobin at admission, length of stay and hospital mortality.

Results: 1248 patients were included in the analysis, however full data set was available for 934 patients. The patients were more predominantly male with similar numbers in the medical and surgical groups. Medical patients were slightly younger, but with higher physiology scores and mortality. There was no statistical difference between length of stay in Intensive Care between the two groups. Medical patients had a higher admission haemoglobin but this did not trend with outcome or length of stay. (see Table 72 ) Conclusions: Interestingly, and not as expected, it seems that admission haemoglobin to Intensive Care is not associated with outcome in the elderly population. It is noteworthy that discussion continues in the literature regarding the definition of anaemia in this age group as the population used to generate the WHO criteria did not include any over 65´s. Admission haemoglobin levels did not seem to correlate with APACHE, length of stay or outcome. However, the medical patients with more likely chronic anaemic state had higher APACHE-II scores, were younger with higher mortality than the surgical admission who were older, had better outcomes but were more significantly anaemic on admission to Intensive Care.

Introduction: Because of the substancial morbidity and mortality provoked by massive bleeding, a protocol to guide treatment of this event in each hospital is required.

Objectives: The aim of this study was to determine whether implementation of the massive transfusion protocol (MTP) was associated with a change in clinical practice or mortality. Conclusions: The number of patients is greater in post-MTP group and APACHE score is lower in the same group since we are warned of these patients at an early stage. There were no differences in clinical practice regarding the administration of blood and hemostatic products.No change in mortality could be documented using the protocol. We have not found any statistically differences probably in part due to the sample size.

Introduction: Discontinuation of life sustaining treatments (LST) is an accepted approach for certain ICU patients. There are different ways of limiting LST, and while terminal extubation (TE) is one of them, it may lead to dyspnoea and respiratory distress, which can be regarded as morally troublesome. Table 74 shows the characteristics of the patients included.

Conclusions: Denial of admission to an intensive care unit due to LST decisions was associated with a high morbidity and mortality. Mortality, APACHE II and Charlson index were significantly lower in the group of patients refused admission to an ICU with a non invasive treatment recommendation. Introduction: Donation after circulatory death (DCD), refers to the procurement of organs from patients whose death is diagnosed and confirmed after circulatory arrest. In the Netherlands the timeframe for DCD to proceed is set at two hours. A considerable number of potential donors after circulatory death are lost because they do not die within the specified timeframe after withdrawal of life-supporting treatment (WLST). Identification of those dying within 2 hours after WSLT results in efficient utilization of the organ procurement teams, hospital resources and above all fulfillment of family expectations.

Objectives: The aim of this study is to determine factors predicting time to cardiac circulatory death after WSLT within 2 hours. Methods: In this single-center study we retrospectively evaluated 92 potential and actual DCD III donors. Patients younger than 16 years of age, and clinically brain dead patients in whom relatives requested a DCD procedure, were excluded. Univariate logistic regression analyses were performed to establish the effect of different predictors.

Results: Only 20 (32 %) converted into actual donor partly due to the fact that cardiac death did not occur within 2 hours. Univariate analysis showed an association between the following predicitors and death within Introduction: Family refusal of organ donation from DBD (donors after brain death) is a limiting factor of the whole donation process and plays an important role in shortage of organs available for transplantation. Although Croatia is a state with presumed consent when it comes to DBD (donors after brain death) organ donation, family is always informed about the possibility of organ donation after it is verified that the deceased is not registered in the non-donor registry.

If the family objects organ donation, their decision is always respected.

Objectives Conclusions: Main reason for refusal of organ donation in our hospital is unknown wish or opposition of the deceased person. No family refused donation due to fear of organ trafficking which is an encouraging fact. Although refusal rate in our hospital is 22 %, which is higher than Croatian average of 13 %, we could not clearly identify contributing factors. We also could not confirm the hypothesis that additional education of transplant coordinators lowers refusal rate. A more detailed prospective evaluation is needed in order to further reduce refusal rate in our hospital.

Impact of brain injury (BI) and BI with brain death ( The course was aimed to Spanish health professionals (nurses and doctors), with a total duration of 10.5 hours. It consisted of a small theoretical introduction followed by several workshops, which included: donor after circulatory death management protocol through high fidelity simulator, family interview, preservation and perfusion procedures with extracorporeal membrane oxygenation in animal models.

The students sat a pre-course self-assessment 10-question test to evaluate their knowledge about cDCD. At the end of the course they filled out a survey, offering their opinion on different sections: content, usefulness, documentation and educational support, organization, duration and overall assessment. The score ranged between one, the most negative value, to 10. Average score was analyzed.

A survey was sent to students working in different hospitals to evaluate the impact of the course in the cDCD programs at their hospitals. Results: 366 students completed the course, their characteristics are in Table 76 . 364 students did the pre-course test, with an average score of 8.64 points. 364 filled out the survey, results plotted in Table  77 .

Feedback trough the after-course survey was received from 56 students. 48 % of the students worked at hospitals without a cDCD program, established after completing the course. 89 % of these students considered that the course contributed to the development and implementation of cDCD program. All professionals who worked at centers where there was already a cDCD program felt that the course contributed to its improvement. Conclusions: Despite the fact that there was a high knowledge on the subject among the students, they showed interest and enjoyed the course. The course had a high impact because it helped improving and developing cDCD programs in several hospitals. We believe that this course, based on high-fidelity simulaton training, has been one of the factors that has promoted controlled donation after circulatory death in Spain in the last years Introduction: The demand for donor lungs thoroughly exceeds the supply.This situation and the application of a strict group of selection criteria, has made donor lung shortage a major problem. To overcome this scarcity, some studies have examined the possibility of using lungs from older donors with mixed results [1] [2] [3] [4] Objective: Using as a starting point the favourable clinical evolution of a recipient of a 85 year old lung, we reviewed all of our donors from the last 6 years that were dismissed strictly because advance age (>60 years) Methods: Retrospective (Feb 10-Jan 16) and descriptive study. All donors of a Spanish tertiary hospital were analysed. We selected all patients excluded for donation and reviewed their contraindication to serve as lung donors. Demographic data and comorbidities, reason for ICU admission, ICU length of stay, respiratory parameters, days of mechanical ventilation, respiratory cultures and antimicrobial therapy were collected. Results: During the period studied we identified 133 potential donors that translated into 90 real donors and only 12 of them (13,3 %) served as lung donors. After analysing the 78 (86,6 %) patients that were not considered as lung donors, we identified 5 (5,5 %) that had been dismissed strictly because advance age. All 5 patients studied were brain death donors. Demographic data and comorbidities, reason for ICU admission and ICU length of stay are detailed on Figure 111 .Last mean pO 2 /FiO 2 recorded was 325,2 ± 115 (PEEP 5,2 ± 0,44) and last mean pCO 2 recorded was 44,3 ± 10,7 mmHg. Median days of mechanical ventilation were 2 (1-8 We recorded the main characteristics of the donors, the most important periods in the process and the evolution of liver receptors admitted to the ICU. Results: 13 patients were admitted. The mean age was 58,69 ± 7,64. Most were male (12/13). Reason for ICU admission more frequent: haemorrhagic stroke (8) . The ICU stay until LIST decision was 6,07 ± 3,32 days. The cares at the end of life (LIST) were performed in the first four patients in the UCI, and another patients in the operating room, intervening in all of them intensivists who had participated in the previous treatment. The time from extubation to significant hypoperfusion of organs (SBP < 60 mmHg) was 9 minutes, the time to cardiac arrest was 13,3 minutes, and to the beginning of the cold perfusión was 21,7 minutes. 7 liver transplants were performed without complications in the ICU, and the ICU stay was 2 days (1, 6). The higher ALT level was 2553 ± 1679,53. 6 of the 7 liver transplants are well and with functioning organ today (one died in hospital ward unexpected cardiac arrest). 21 kidneys were obtained from these donors.

The Maastricht III donors provide valid organs for transplantation and the intensivists play an important role both in the detection as in the development of care at the end of life. The first transplants had long functional prolonged warm ischemia, which has been reflected in graft function, but the performing of LIST in operating room, the ultrafast extractions and the presence of the receptor in the hospital are improving the viability of organs, so the results of the last donors are better. Thus, Maastricht III donors must be considered today as an additional source of organs for transplantation. (1), cisatracurium was administered at a constant and high posology without monitoring the depth of neuromuscular blockade.

Objectives: To assess if the monitoring of the train-of four (TOF) and the management of cisatracurium posology by nurses according to an algorithm can ensure an effective neuromuscular paralysis and allow to decrease cisatracurium consumption during ARDS Methods: We conducted a prospective study in 2 medical ICUs. All the patients with a PaO2/FiO2 < 120 for more than 6 hours and requiring a continuous perfusion of cisatracurium were included. Neuromuscular blockade was monitored by a TOF at the adductor pollicis. Nurses followed an algorithm of adaptation of cisatracurium posology depending on the TOF with an aim of 0/4. The initial posology was based on the maximal doses recommended in anesthesiology and on the patient ideal body weight. This posology was increased and a bolus done each time that the TOF was >0. The interruption of NMBAs was decided by physicians. The initial and final posology and the daily consumption of cisatracurium, the need of performing boli, the results of TOF and the occurrence of adverse events such as patient/ventilator asynchrony were noted. Effective cisatracurium consumption was compared to the theoretical posology that would have been administered if the patient had been treated according to the ACURASYS study protocol (i.e. 37,5 mg/h). We also evaluated the economic impact of the reduction of cisatracurium consumption. Introduction: Sepsis-associated acute respiratory failure is frequent, occurs early and is associated with significant mortality. With the increasing use of noninvasive techniques, timing of intubation can vary and may lead to a difference in outcome.

Objectives: The objectives of this study were 1) to draw on practitioners' current practice and perspectives to understand and identify practice variation in intubation and 2) to develop an explanatory theoretical model that demonstrates the relationship of various factors contributing to practice variance. Methods: Between March and July 2015, using a grounded theory approach, we conducted semi-structured interviews with providers involved in intubation and audio recorded them. The interview guide focused on clinicians' perspectives on and practices of intubation in patients with sepsis and impending respiratory failure.

Results: Eighteen interviews were conducted with intensivists, fellows, nurse-practitioners, respiratory therapists and registered nurses. Intubation perspective and practice varied dependent on three domains: patient's characteristics, clinician's characteristics, and organizational structure. Patient factors included nature of acute illness, underlying comorbidities, clinical presentation, and patient's values. Clinician factors included background, training, experience and practice style. System factors included of standardized policies and protocols, hierarchy and team dynamics. Although most clinicians agreed that intubation is needed in case of persistent respiratory distress, altered mental status, or shock, they disagreed on when to initiate it. In different contexts, intubation could be considered as preemptive (prophylactic), therapeutic ('just in time'), and as a rescue. Assessment, reassessment, and time-limited trial off noninvasive techniques matter. Based on these results, we propose a model regarding intubation in sepsis consisting of the steps in the decisional process, a classification of the categories of timing of intubation, and decisional context factors that impact the timing of intubation.

Conclusions: In patients with sepsis-associated acute respiratory failure, variability of intubation was a natural phenomenon and appeared case-driven. Intubation timing should be adjusted based on explicit consideration of each patient situation, their fitness, the cadence and trajectory of their respiratory failure, the team's proficiency in providing noninvasive and invasive ventilator support, and emphasis on clear, frequent closed-loop communication of the treatment plan and rationale within the entire critical care team. (1) . Objectives: To determine the efficacy of rhTM in septic patients with severe respiratory failure. Methods: We performed sub-analysis of a retrospective observational study (Japan Septic Disseminated Intravascular Coagulation study, J-SEPTIC DIC study), which was conducted in 42 intensive care units in Japan. Among 3195 septic patients enrolled in this original trial, we selected septic patients with severe respiratory failure and compared patients based on rhTM treatment (rhTM group and control group). Propensity score analysis was performed between two groups. Outcome was the number of ventilator free days. Results: 1180 patients (rhTM, n = 356, control, n = 824) were analyzed in this trial. After adjusting for baseline imbalances by propensity score analysis, VFDs increased significantly in rhTM group (rhTM group: 11.9 ± 10.3 days vs. control group: 10.3 ± 10.4 days, P = 0.03). Conclusions: In this analysis, rhTM improved outcomes in septic patients with severe respiratory failure. We need further evaluation. Results: During the study period, totally 64 ARDS patients with pathologic diagnosis of DAD were eligible for analysis. These 64 patients were divided as mild (n = 14, 21.9 %), moderate (n = 36, 56.2 %) and severe ARDS (n = 14, 21.9 %) by Berlin definition and the hospital mortality rate were not significantly different between these three groups (64.3 %, 69.4 % and 85.7 %, p = 0.4). According to the etiology, these 64 DAD patients were divided into known etiology group (n = 47, 73.4 %) and unknown etiology group (n = 17, 26.6 %), and the hospital mortality rate had no significant difference (72.3 % vs 70.6 %, p = 0.890). The known etiology group had higher percentage of male and lower PaO2/FiO2 ratio than unknown etiology group (72.3 % vs 41.2 %, p = 0.022; 202.3 ± 84.4 vs 146.3 ± 60.0, p = 0.005). The multivariable logistic regression revealed sequential organ failure assessment (SOFA) score at the time of open lung biopsy was the only predictor of hospital mortality (odds ratio 1.413, 95 % confidence interval 1.127-1.772; p = 0.03). In terms of glucocorticoid treatment, there was no significant difference in glucocorticoid use, timing from ARDS to glucocorticoid use, dose and duration between survival and nonsurvival patients. Conclusions: For the ARDS patients with DAD, SOFA score was the predictor of hospital mortality but glucocorticoid treatment did not improve the survival rate. Introduction: Ventilator associated pneumonia (VAP) is a known complication of mechanical ventilation. Aspiration of oropharyngeal secretions results in infection that leads to significant morbidity, mortality and cost 1 . Use of sub-glottic secretion drainage (SSD) devices have been shown to decrease both the incidence of VAP and intensive care unit (ICU) days 2,3 . There have been safety concerns associated with use of SSD devices and herniation of tracheal tissue into the suction port 4 . A study in sheep showed significant tracheal injury associated with continuous suction 5 . Human studies have shown conflicting results regarding the risk of tracheal injury 6, 7 . Objectives: To determine the risk of tracheal injury using an SSD device versus a standard endotracheal tube. Methods: 57 patients undergoing tracheostomy in the ICU were enrolled in the study. Patients were intubated in the ICU, operating or emergency room, pre-hospital, or referring hospitals. Intubation conditions and duration of intubation were documented. At the time of tracheostomy, a bronchoscopy was performed and the presence and degree of tracheal injury were noted. Patients were followed to hospital discharge and decannulation, otolaryngology consults, and discharge or death were recorded. Results: 41 patients were intubated with a Malinckrodt EVAC SSD device and 16 were intubated with a standard endotracheal tube. 33 patients were found to have a tracheal injury ranging from mild erythema to severe ulceration; 23/41 (56 %) in the EVAC group and 10/ 16 (63 %) in the standard group (RR -0.89; 95 % CI 0.56 to 1.43). 5/ 41 (12 %) patients were reported to have injury at the site of the suction port; 2 were reported to have mild edema and erythema and 3 had mild to moderate ulceration. Of the patients with tracheal ulceration at the suction catheter port, 4 were decannulated successfully without further complication and 1 patient died prior to termination of mechanical ventilation. Conclusions: There was no significant difference in the risk of tracheal injury with SSD devices compared to standard endotracheal tubes. The degree of injury was similar in both groups. A small Introduction: VAP rates in Brazil are higher than those listed in Europe and EUA.No Municipal Hospital Moyses Deutsch since 2009 implemented the protocol, applied the five strategies to reduce VAP, however we maintained VAP density with little reduction and we never zero target . Objectives.Objective of the study was to examine the effect of Healthcare Improvement ventilation bundle Institute in addition to focusing on three strategies : Oral decontamination with chlorhexidine (ODC), the head elevation and awakening daily in the incidence of VAP in a unit intensive care. Methods: The study was conducted in a 20-bed, medical-surgical ICU. Criteria for nosocomial pneumonia are those from the CDC. Strategy was to implement the IHI's ventilator bundle , focused and optimized in the first three The goals were the ICU team adhesion of 80 % achieved in six month after bundle implementation and 98 % after one year of follow up. These measures included five strategies to prevent ventilator-associated pneumonia: 1-45°elevation of the head of the bed, 2-adequate sedation level (RASS −1 a −2 ), 3-oral decontamination with chlorhexidine 0.12 %, 4-DVT/PE prevention and 5-peptic ulcer prophylaxis . From January 2012 on, the ICU nursing staff and ICT performed a daily checklist in order to observe the five issues accomplishment. If any item was found to be inadequate it was promptly corrected. Results: In January and December 2012 , adhesion to the whole bundle was 68 % and 100 % respectively. VAP density was proportionally lower to bundle adhesion in the same period, 20 per 1000 ventilation/day and respectively. In 2015 we achieved zero VAP in both semesters.

Conclusions: Initial VAP rates were extremely high even for Brazilian benchmarks. Although we could not implement expensive technologies like continuous aspiration of subglottic secretions, ICU team and ICT efforts were crucial for satisfactory results, as well the administrative board support, which turned this issue an institutional priority. Our goals are to reduce even more, implementing ''ventilator bundle-getting to zero'' program, maintaining a continuum effort to sustain these results. Introduction: Ventilators-Associated pneumonia (VAP) and its prevention is a significant concern for ventilated patients in the acute care. Objectives: To determine if the knowledge and awareness of "ventilator bundle" helped in the prevention of ventilator associated pneumonia in the patients admitted to hospital. Methods: A prospective observational study that evaluated VAP rates from August through October 2015 were evaluated. All the adult medical patients who were intubated and ventilated in 8 medical wards from August through October in the year 2015 were included in the study. During the period of June to July 2015 the staff nurses were educated and made aware about the problem of VAP and the use of ventilator bundle in helping to prevent this VAP. Patients who expired within 24 hrs of admission, who were transferred to intensive care unit within 48 hrs, and those who were diagnosed with pulmonary embolism or metastasis were excluded from this study. Intervention. The concept of "ventilator bundle' was introduced after educating the nursing staff and the medical personnel through group discussions."CHULA Ventilator bundle" is a package of evidence -based interventions that include:

(1) Clean equipment and environment;

(2) Hand Hygiene and Elevation of patient's head of bed to 30-45 degrees; (3) Use 0.12 % chlorhexidine as a part of oral care every 4 hour; (4) Labor over weaning and extubation each day; (5) Aspiration precaution protocol.

Measurement. Demographic data was collected from the patient data files. VAP was diagnosed when it met the (clinical non-invasive) diagnostic criteria. Incidence of VAP and protocol compliant were calculated. Results: A total of 147 were on mechanical ventilator for a vary period of 0-90 days. Average age was 71.48 ± 15.92 with 59.5 % of male. Introducing the concept of "CHULA Ventilator Bundle to Prevent Ventilators-Associated pneumonia" significantly reduced the VAP rate per 1000 ventilator days from 12 % to 0 % in the medical group (8 medical wards). It significantly reduced the incidence of oral cavity problem (5.72 ± 1.352 vs. 5.23 ± 0.88, p = < 0.001). Ventilator Bundle compliance was 60.14 %. Conclusions: However, Ventilator Bundle compliance was less than 80 %, introducing the concept of "CHULA ventilator bundle" helped us to reduce the incidence of VAP and the incidence of oral cavity problem. Grant acknowledgment Quality Improvement Center, King Chulalongkorn Memorial Hospital Introduction: Patient-ventilator asynchrony is a mismatch between patient and ventilator inspiratory and expiratory time. It is associated with prolonged duration of mechanical ventilation (MV), increased need for tracheostomy and increased mortality. Five main patterns of asynchrony are described, without universal agreement on definition. Studies on patient ventilator asynchrony have quantified asynchrony at heterogeneous time points and during periods of various durations. In addition, most of these studies were of single centre type.

Objectives: The aim of the present study was to evaluate the factors associated with and the prognosis impact of asynchrony, according to two methods of quantification: visual inspection of airway flow and pressure signal and a computerized method integrating electromyographic activity of the diaphragm (EAdi) as a maker of patient inspiratory time at the early phase of weaning.

Methods: Ancillary study of a multicentre, randomized controlled trial comparing neurally adjusted ventilator assist to pressure support ventilation at early phase of weaning. Airway flow, pressure and EAdi were recorded during 20 minutes 12, 24, 36 and 48 hours following inclusion. Asynchrony were quantified according to two methods: 1) "flow-and-pressure" based on the visual inspection of the flow and pressure signals 2) "EAdi-based" with analysis of the EAdi signal in addition to the flow and pressure signals. Asynchrony index (AI) was calculated as the number of asynchronous breaths divided by the total number breaths multiplied by 100.

Results: 103 patients mechanically ventilated for 5 days (3-9) were included, 72 men (68 %), aged 66 (60-77) years, SAPS II 44 (35-59), 62 % were mechanically ventilated for de novo hypoxemic respiratory failure. Prevalence of ineffective efforts was higher with flow-and-pressure method than with EAdi-based method. Auto-triggering, doubletriggering, premature and late cycling were more frequently observed with EAdi-based method than with flow-and-pressure method. AI and the total prevalence of asynchrony were significantly lower with the flow-and-pressure method than with the EAdi-based method (Table 78) . No significant difference in term of gender, age, SAPS 2, Charlson score or length of MV prior to inclusion was observed with severe asynchrony (AI >10 %) Severe asynchrony was not associated with difference in term of hospital length of stay, duration of MV and day-28 mortality. ICU length of stay determined by the flow-and-pressure method was shorter in patients with AI ≥ 10 % (18 (14-28) vs 9 (9-12), p = 0.02).

Conclusions: The prevalence of patient ventilator asynchrony varies according the methods and definitions used to quantify asynchrony, which suggests the need for a consensus statement in asynchrony's definition. Patient ventilator asynchrony was not associated with a poorer outcome. Introduction: Tracheostomy is a frequent procedure in intensive care units, in the US over the past 2 decades utilization rose substantially, driven by surgical patients [1] . The optimal timing for tracheostomy in critically ill patients remains a topic of debate.

Objectives: To analyse tracheostomy utilization and trends in an intensive care unit (ICU) and to determine the impact of tracheostomy timing (early vs late) in critically ill patients on duration of mechanical ventilation, ICU stay, overall hospital stay and mortality.

Methods: Retrospective study including all critically ill patients who underwent tracheostomy in an ICU from 2005 to 2015. The sample was stratified in two groups, according to time of invasive mechanical ventilation until tracheostomy: early tracheostomy (≤ 10 days) and late (> 10 days). Results: Over the study period a total of 688 tracheostomies were performed, representing 15,3 % of the admissions in the ICU. Tracheostomy was more common in medical patients (67.4 %). Mean time until tracheostomy was 9 days. There was no tendency in tracheostomy rates and timing over the years. Early and late tracheostomy groups did not differ significantly by gender, age, SOFA score and type of admission. In the early tracheostomy group there was a statistically significant reduction in the length of invasive mechanical ventilation (10 days vs 25 days, p < 0.001) and ICU stay (13 days vs 28 days, p < 0.001), with impact in ICU and hospital mortality.

Conclusions: Early tracheostomy was associated with reduction in invasive mechanical ventilation days and ICU stay, with possible implications in long term morbidity and health care costs. Reinforcing that tracheostomy timing should be considered in the decision process, when evaluating risks and benefits. Introduction: Development of critical care medicine has been decreasing mortality of critical illness. However, 60-80 % of survivors suffer functional impairment or ICU-acquired weakness (ICU-AW). In order to address interventions in ICU-AW, it is essential to know when ICU-AW developed in addition to its incidence and risk factors. Objectives: To assess the onset of ICU-AW and its incidence and risk factors in the ICU of Tokushima University Hospital. Methods: Prospective observational study. Critically ill adults were enrolled when they were mechanically ventilated at least 4 days. Patients younger than 20 years old, with neuromuscular diseases, central nervous system disorders, and pregnancy were excluded. After we determined feasibility of communication, medical research council (MRC) sum score was measured as soon as possible. When MRC score was less than 48, we diagnosed patient as ICU-AW. Basic profiles, underlying diseases, APACHE II score, administration of neuromuscular blocking agents (NMBA) and corticosteroids, and laboratory data were recorded. Introduction: Severe traumatic brain injury (STBI) remains as the most significant medical and social problem due to high prevalence and mortality, primarily among young and employable population. The leading problem of intensive care of STBI is the prevention and elimination of intracranial hypertension (ICH). One of the methods of ICH elimination is mechanical ventilation as a component of complex therapy. Among the various methods and modes of mechanical ventilation high-frequency jet ventilation (HFJV) is particularly distinguished, which is enduring "the second birth". In HFV transpulmonary pressure and the pressure in airways is much lower than one during traditional methods, the negative pressure in pleural cavity is maintained during inspiration phase and spontaneous breathing.

Objectives: Comparative assessment of efficacy of different modes of mechanical ventilation in patients with STBI. Methods: We studied the cerebral perfusion during various modes of mechanical ventilation in 70 patients with STBI. Mean age was 32 ± 5. The general status in admission was severe, Glasgow coma score was 7 ± 3. All patients had traditional intensive care with different modes of respiratory support: controlled mechanical ventilation -CMV (n = 10); synchronized intermittent mandatory ventilation -SIMV (n = 20); HFJV (n = 40). The efficacy of all modes were assessed by arterial blood gases analysis (SaO2 -96-99 %, pCO2 -34.7-35.2 mmHg). Intracranial pressure were measured invasively and was 20 ± 5 mmHg. All patients regularly had clinical and neurological examination, control of laboratory tests (common blood count, arterial blood gases, arteriovenous gradient of O2 (AVDO2) and oxygen saturation in jugular vein (SjO2). Cerebral hemodynamics was studied by transcranial dopplerography. The registered parameters were mean linear velocity of cerebral blood flow (Vm), pulsatile index (Pi) and overshoot coefficient (OC).

Results: There were significant differences in parameters of cerebral hemodynamics in various modes of respiratory support: CMV: Vm -51. Most patients found mobilisation to be a positive experience and the beginning of their recovery. However, mobilization was described as a difficult component of the care, mainly due to pain, tiredness and dizziness. Almost all patients commented on the benefits of participation in physiotherapy, which was verified by physical improvements and progression in their abilities. Although most improvements discussed were physical, two patients also described the psychological benefits that occurred in the sessions. They reported that the physiotherapists 'built them up' and encouraged them. One patient described a mind shift that occurred once she had mobilised out of the bed. She described it as being able to see what she was capable of. It was described as a precious and muchneeded service, without which some patients felt they may not have survived or recovered as quickly. Methods: This retrospective study was performed in a 12-bed medical ICU in Spain from 2011 to 2015. All patients admitted to the ICU during this period were included in the study. CCI patients were defined as those with more than 60 days of ICU stay. Data were collected in 3 ways: review of a prospectively elaborated database, review of electronic records, and telephone survey evaluating the functional status of survivors, one year after their discharge from the ICU. Results: During the study period, 46 patients (9 females −20 %) were considered CCI. The characteristics of these patients are shown in Table 82 . All the studied patients needed prolonged mechanical ventilation (median 66 days), defined as >6 hours/day of ventilator support for > 21 consecutive days. The follow-up period is drawn in Figure 115 . The in-ICU mortality was 28 %. In the first year, 17 patients (37 %) were alive. Most patients improved their quality of life over a year, with approximately 40 % of them displaying some help for dressing or to performing the transfer of themselves. Symptoms of anxiety and depression improved during the first year, being present in up to 20 % (of the patients), but in 6 % if we refer to the presence of nightmares or hallucinations. 41 % these patients were transferred (discharge) to a rehabilitation center and 47 % needed hospital readmission within the follow-up period. Conclusions: For CCI patients in-hospital mortality rate is still high after discharge from the ICU. However, more than one third of them are alive one year after their Hospital stay and in an almost independent condition. Efforts focused on early specific therapeutic strategies after ICU admission to prevent the progress of the acute disease towards chronic critical illness and to improve the outcome must be explored. [1] [2] [3] [4] . At ICU, 40.9 % had delirium, 40.9 % needed blood transfusion and 18.2 % renal replacement therapy. Patients worsened in all parameters of the five dimensions of the EQ-5D after 30-days: The extreme problems level increased in the mobility dimension from 12.1 % at ICU to 42.1 %, self-care from 25.8 % to 52.6 %; usual activities from 19.7 % to 68.4 %; pain/discomfort from 7.6 to 13.2 and anxiety/depression from 4.5 % to 7.9 %. The dependence observed in the KATZ index worsened in 30 days when 28.8 % of patients were dependents before ICU admission increasing to 44.7 % after 30 days. About family members, 42.4 % were spouses and 39.4 offspring, their mean of age was 55.89 ± 13.69 years and 84.8 % had previous experience of ICU. We observed that they presented more symptoms of anxiety (50 %) and depression (27.3 %) at ICU when compared 30 days after (21.1 %) and (13.2 %), symptoms of anxiety and depression respectively. Conclusions: The most common eligibility conditions of CCI were sepsis followed by mechanical ventilation. We observed a great mortality on 30 days and among survivors a worsen quality of life with more dependence in their Activities of Daily Living. We also observed that family members suffered more while in ICU stay. Introduction: Tracheostomy is a favored alternative option for providing prolonged mechanical ventilation and safety airway used for more than 50 years. Despite its numerous advantages, tracheostomy may have severe complications as being an invasive method for presenting respiratory tract patency. Besides, the tracheostomized patients usually have prolonged ICU stay, high mortality and morbidity arise from concomitant comorbidities. Objectives: The aim of the study was to evaluate the frequency, patient characteristics, complications and the prognosis related with our percutaneous tracheostomy practice. Methods: Hospital electronic records and ICU files of the patients with percutaneous tracheostomies performed in our 34 bed anesthesiology ICU were evaluated between January 2010 and December 2014. Ethic consent was obtained from local ethic committee. The patients who were discharged with home type mechanical ventilator or their relatives were contacted by phone for getting information about their health status or related complications. (Tables 83 and 84) .

Conclusions: The blood serum sodium levels at admission, especially hypernatremia, may also be used as an independent predictor of outcome in the surgical critically ill patients.

Introduction: WHO estimates that the worldwide dengue fever incidence is about tens of thousands of cases every year. As Taiwan is situated in the high risk subtropical region, dengue fever has virtually become a seasonal infectious disease. Climate warming, demographic movement and the higher probability of increase in intermittent rainfall in recent years have added many factors unfavorable to dengue fever prevention. Years of prevalence and the emergence of different types have also caused the risk of mortality for dengue fever to become relatively high.Of the total 37,224 confirmed dengue fever cases in 2015, there were 174 deaths (with a mortality rate of 4.7 per thousand), marking the largest outbreak over nearly one decade in Taiwan.

Objectives: Analysis was conducted on the 93 confirmed severe cases of dengue fever or dengue hemorrhagic fever reported to this hospital over the period between July 20 and September 30, 2015 in terms of gender, age, history of chronic diseases, warning signs and diagnostic criteria for severe conditions. Methods: Retrospective case study was also conducted to identify risk factors in dengue fever and dengue hemorrhagic fever as well as predictors of death among dengue fever cases for statistical analysis.

Results: According to the results, those susceptible to infection concentrated on older people aged over 65 (with an average age of 68); in total 73 cases had chronic diseases (with an average rate of 78.5 %), among which hypertension and diabetes constituted the majorities; and based on symptoms, fever accounted for 83.51 % while gastrointestinal bleeding was the most common at 38.7 %. Of the 93 cases, there were 18 deaths, with an average APACHE II score of 17.73 and an average mortality rate of 19.8 %.

Conclusions: This study shows that patients with chronic diseases aged over 65 will have 10 times higher risk of death if infected with dengue hemorrhagic fever. It is therefore suggested that older people aged over 65 and patients with chronic diseases who are infected with dengue hemorrhagic fever must be closely monitored in clinical practice to pinpoint the best time for treatment and effectively reduce mortality rates. To sum up, effective use of knowledge about risk factors and prognostic factors in dengue hemorrhagic fever can help epidemic prevention organizations to focus their limited resources on high risk groups and increase the effectiveness of prevention.

Cardiorespiratory instability risk escalation patterns: an association study with risk factors and length of stay L. Chen (Fig. 116) . 66 % of them belong to "late onset" types whose risk escalated ≤30 minutes before CRI onset, but with different initial RR levels (low, medium and high). 19 % of patients belonged to "early onset" type with gradual escalating risk starting about 3 hours before overt CRI, and 15 % falling into a "persistently high" type. The mean RR during the first 4 hours of SDU stay are 0.9, 1.15 and 1.35 for "late onset" types; 1.10 for "early onset" type, and 1.55 for "persistently high" type, comparing with baseline RR of 1.0 for CRI negative patients. The mean RR derived in the first 4-hours after admission is strongly associated with risk escalation patterns observed (p-value < 0.001), specifically, patients of "persistently high" type were more likely to have higher mean risk levels at SDU admission . Risk escalation patterns were not significantly associated with age, CCI or SDU LOS. However, they are significantly associated with hospital LOS (p = 0.002).

Conclusions: There is potential "risk stratification value" of VS collected during initial hours of SDU stay in predicting the CRI risk escalation patterns later on, which may in turn predict hospital LOS. These insights may guide monitoring resource allocation for CRI management. 1-18.3 ] mmol/L and most of patients were on vasopressors therapy. Coronary angiography was performed in 28/57 (49 %) patients with a cardiac cause; continuous renal replacement therapy was initiated in 27 out of the 52 patients (52 %) developing acute kidney injury during the ICU stay. 14 patients showed a full neurological recovery during the ICU stay (23 %) but only 12 were still alive with intact neurological function at 3 months (20 %); 6/57 after OHCA (11 %) and 6/23 (26 %) after IHCA. Eight patients (13 %) with irreversible brain damage had organ function suitable for donation and 4 were eventually explanted. Conclusions: ECPR provided acceptable survival rate with good neurologic recovery in refractory cardiac arrest. These patients underwent several additional therapeutic interventions, which, in case of irreversible brain damage, could stabilize extra-cerebral organ function and potentially provide some available organs for donation.

Post-resuscitation treatment with inhaled argon improves outcome even after a prolonged untreated cardiac arrest in a porcine model Introduction: After the initial success of cardiopulmonary resuscitation (CPR), the majority of patients die, mainly due to postresuscitation (PR) cardiac failure and ischemic brain damage. Inhaled argon has shown neuroprotective effects in a porcine model of cardiac arrest (CA) of short duration. Objectives: To investigate the effect of post-resuscitation treatment with inhaled argon on outcome in a preclinical porcine model of prolonged untreated CA and CPR. We hypothesized that argon would ameliorate post-resuscitation neurologic dysfunction. Methods: The left anterior descending coronary artery was occluded in 24 pigs (39 ± 2 kg), and ventricular fibrillation (VF) was induced. After 12 min of untreated VF, CPR, including mechanical chest compression, ventilation and adrenaline administration, was performed for 5 min prior to defibrillation. Following successful resuscitation, animals were subjected to 4 hr ventilation with (a) 70 % argon -30 % O 2 (n = 10) or (b) 70 % N 2 -30 % O 2 (n = 10). Hemodynamics were continuously monitored and systolic myocardial function (i.e. ejection fraction (EF), shortening fraction (SF)) was assessed by echocardiography. Serial blood samples were obtained for blood gas, serum neuron specific enolase (NSE) and plasma high sensitive cardiac troponin T (hs-cTnT) assays. Animals were observed up to 96 hr for assessment of survival and neurological recovery (Cerebral Performance Categories (CPC) scale). Results: Twenty animals were successfully resuscitated and enrolled in the study (Table 87) . Ventilation with argon did not have any detrimental effects on respiratory gas exchange during the 4 hr ventilation (Table 87) . Animals receiving argon showed a significantly lower heart rate and higher mean arterial pressure and stroke volume compared to controls during the 4 hr of observation (Table 87) . Animals treated with argon presented also a significantly better recovery of systolic myocardial function, as represented by the higher SF at 96 hr compared to controls (Table 87) . Nine of the 10 resuscitated animals in the argon group survived for 96 h in comparison to 6 out of 10 in the control group. Animals treated with argon presented a significantly better neurological recovery (CPC 1.7 ± 1.3) in contrast to animals in the control group (3.4 ± 1.6, Figure 121 ). Lower circulating levels of hs-cTnT (median: 1332 ng/mL vs. 8015 ng/mL, p < 0.05) and NSE (median 8.5 ng/mL vs. 21.2, p not significant) were observed in the animals ventilated with argon compared to controls. Conclusions: In this severe model of CA, post-resuscitation treatment with argon allowed for improved hemodynamics, myocardial function and neurologic recovery, without detrimental effects on respiratory gas exchanges.

Munich, Germany, which is staffed with physicians working at a university hospital in the specialities anaesthesia or surgery.

(2) Test if there is a difference between specialists and residents in pain treatment of trauma patients.

Methods: After Ethics Committee approval, retrospective analysis of the protocols of our prehospital emergency service location in Munich, Germany of 2014-2015. Statistical calculation was done using logistic regressions with STATA14 (College Station, TX, USA). Results: 1178 documented trauma cases. 242 trauma cases could be assessed for frequency of oligoanalgesia, which was present in 39 of these cases (see Figure 123 , dashed frames), leading to an relative frequency of 16 % of cases. There was no difference in frequency between residents and specialists (Table 89) . Relatively more trauma cases where handed by specialists, while documentation of pain was better in residents (Table 89) . Documentation of pain, however, was insufficient, since pain assessment at hospital admission was documented in 38 % of possible cases of oligoanalgesia only.

Conclusions: Frequency of oligoanalgesia in trauma patients seems to vary in different systems, since it was much lower in Munich compared to Switzerland (16 % vs. 43 %, respectively). There are several possible explanations: Data from Swizerland was from an air resuce service while our data is from a ground based system. Second, in our system possibility of treatment by a specialist was much higher (83 % residents in Switzerland). Third, documentation in our system was inadequate. Theoretically, frequency of oligoanalgesia could increase up to 35 % if all cases without adequate pain documentation were counted as oligoanalgetic. To assess appropriate numbers improvement in documentation is essential. Of the 39 attempted resuscitations 34 were immediately unsuccessful, 5 resulted in ROSC (3 sent to ICU for post-resuscitation care, whilst 2 remained on the ward). At 24 hours 2 (both in ICU) were still alive. Defibrillation was attempted in 5 cases. Intubation was attempted on 5 occasions. In 5 (13 %) of the resuscitation attempts CPR was the only intervention reported while 27 (69 %) received more than 1 vial of adrenaline, or defibrillation, and or intubation. Interviewees reported that in 25 (64 %) of these patients they were 'not at all' or only a 'little bit surprised' by the patient having a cardiac arrest (Fig 125) . They further described the chances of a successful outcome as 'unlikely or very unlikely 61 % of the time and likely or very likely only 10.3 % of the time (Fig 126) .

Conclusions: Perspectives of junior doctors interviewed suggest many cardiac arrests were not a surprise and that the probability of ROSC following attempted resuscitation was unlikely. There is high incidence of patients receiving CPR attempts before death in hospitals across Sri Lanka with DNAR practices remaining uncommon. Outcomes remain poor, with ROSC after cardiac arrest being 12.8 % and survival at 24 hrs 5.1 %. Of the 34 unsuccessful resuscitation attempts, defibrillation and or repeated adrenaline was reported in 67.6 % of cases.

Introduction: Pro-coagulatoric effects after cardiac arrest and consecutively appearing microthromboses have been considered major contributors to morbidity and mortality after CPR [1] . In contrast, recently published data suggest that 35-53 % of patients after out-ofhospital cardiac arrest (OHCA) present with hyperfibrinolysis during and after CPR [2;3] . The interpretation of these inconsistent observations remains unclear and complicated, because of methodological differences and lacking analytical approach in the underlying studies. Fibrinolytic activation might be the physiological reaction to restore perfusion after hypoperfusion due to microthromboses. This leads to the question, if the duration of no-flow (time without chest compressions) after cardiac arrest influences the level of coagulation activation and subsequent fibrinolysis during CPR.

Objective: To investigate the influence of a delayed onset of CPR on the extent of fibrinolysis and the function of the coagulation system measured by rotational thrombelastometry (ROTEM). Methods: After approval of the local authorities (Nds. LAVES, approval G13-1088) cardiac arrest was induced in 10 anaesthetized female Göttingen Minipigs via rapid ventricular overpacing resulting in ventricular fibrillation (VF). In order to simulate a BLS-CPR in 5 animals (CPR-Group), chest compressions (CC) and ventilation were started after 5 min of VF (30:2-ratio, FiO 2 = 0.21). In order to simulate consecutive ALS-CPR, continuous CC (100 min −1 ) and ventilations (15 min −1 , FiO 2 = 1.0) were started 5 minutes later.

No CPR was started in the remaining 5 pigs (Non-CPR-Group). Blood samples for a complete ROTEM analysis (ROTEM delta® analyzer, Tem Int. GmbH, Munich, Germany) and laboratory analyses were taken before induction of VF (baseline) and 10, 20 and 30 min after VF. All parameters were investigated for normal distribution (Shapiro-Wilks-Test). Statistical significance of differences (p < 0.05) was investigated using the unpaired t-test (normal distributed parameters) and the Mann-Whitney-U-Test (notnormal distributed parameters). Results: Figure 127 summarizes laboratory and ROTEM results. In no group maximum lysis increased significantly after cardiac arrest ( Figure 128 ). Maximum clot firmness (MCF) in FIBTEM analyses decreased significantly in both groups (Figure 129 ), but plasma fibrinogen levels (measured using the Clauss method) remained stable. Introduction: Perception and knowledge of hospital staff involved in an emergency evacuation of hospitalized patients is usually low. This is especially remarked in an ICU due to the complexity of moving patients who depend on invasive monitoring and organ support due to acute illness. Objectives: To analyze the differences between different members of ICU staff about their perception and knowledge of self-protection and evacuation plans. Methods: A quantitative, descriptive and cross-sectional study was carried out by a fully structured and self-administered survey in 12 public and private ICU staff through a total sample of 434 participants. They were asked to complete a questionnaire about their perception and knowledge of self-protection and evacuation plans in the ICU. This study pretended to analyze the differences between participants, taking into account their demographic and occupational characteristics and their level of satisfaction and commitment to their jobs.

Results: On a rating scale from 1 to 7, ICUs workers perceive that their preparation and knowledge were too low to meet a possible emergency that could require an evacuation, although they were aware of the need to make an update. However, they state that the different ICUs where they work do not have these plans, and, consequently, they do not feel prepared to act in an emergency situation, even though they think this type of situation may occur. The significant differences (p < 0,000) were observed when levels of satisfaction and commitment to their jobs were high.

The results of this study showed that there is a need for more knowledge in the area of emergency training. This should be the basis for the development of educational programs and also promoting awareness of ICU staff on self-protection and evacuations plans.

Integrating nurse practitioners and physician assistants in the ICU: results of a national survey R. Kleinpell Rush University Medical Center, Chicago, United States Intensive Care Medicine Experimental 2016, 4(Suppl 1):A1122

Introduction: An increasing number of intensive care units (ICUs) are integrating advanced practice providers (APP) including nurse practitioners (NPs) and physician assistants (PAs) to meet workforce demands to care for acute and critically ill patients. Although these roles are established ones, limited information is known about the specific care models used in ICU settings. This information is crucial in objectively evaluating the effectiveness of APP roles.

Objectives: To address this gap, a national survey was conducted targeting NPs and PAs, including those working in ICU settings.

Methods: A web-based survey was used to assess 4 domains including role components (i.e. direct care management; care coordination; performing procedures; education; quality assurance; research); role responsibilities (i.e. practice autonomy, prescriptive authority, credentialing and privileging delineations) unit-level organization (i.e. physician staffing models, components of the multidisciplinary care team); and hospital organization (i.e. academic status, bed size, location, payer-mix).

Results: A total of 1851 APPs responded to the national survey including 1350 NPs and 491 PAs. The respondents reported working in a variety of settings including hospitals, clinics, urgent care centers and specialty practice sites. A total of 636 reported working in an ICU setting. Of these, 286 reported 24/7 coverage of ACNPs and PAs in the ICU. Main role components included patient care management as part of the multiprofessional ICU team; teaching to patients, families and healthcare staff; involvement in quality improvement and research initiatives and administrative components such as committee work. Specific aspects include conducting history & physical exams, ordering and interpreting diagnostic test/labs; providing care coordination, performing specialty procedures such as wound care or other specialty care. Major areas of impact that were identified included continuity of care, improving evidence based practice care, reducing hospital length of stay, preventing hospital readmissions and promoting patient, family and staff satisfaction.

The results of the study provide information from a large national sample of NPs and PAs that identifies the comprehensive care components of the role as well as areas of impact, highlighting the value of APP care. Globally, this information can be useful to other countries who are considering use of NPs and PAs in the ICU.

national health care resources. With 470 ICU beds, the reserves of the system are often overwhelmed. The responsibility for a rational management and distribution of these costly resources burdens the admitting Intensivist. Intensivists not using protocols expressed a strong desire (79 %) to introduce protocol based criteria for admission. Conclusions: The most important factors influencing decisions about admission (or refusal of admission) in the ICU are bed availability and prognosis of the underlying disease. Socioeconomic and religious criteria are clearly of marginal significance. Drug abuse and severe psychiatric disease do not emerge as compelling causes of biased decisions. It appears that the Intensivist's decisions are largely individualized, as application of admission protocols is limited among the ICUs. However, the responses documented in this survey strongly indicate that introduction of such protocols would be welcome by a majority of our colleagues.

Conclusions: It is essential that non-cardiac surgery should be delivered in the most appropriate clinical setting. In Scotland, adults with moderate to complex congenital cardiac disease are managed by the Scottish Adult Congenital Cardiac Service (SACCS), based at the Golden Jubilee National Hospital (GJNH), near Glasgow. [3] Existing guidelines have established when patients should have elective non-cardiac surgery performed at GJNH. However, many surgical specialties are not routinely available at GJNH, the bed occupancy rate is high and with an increasing SACCS population there is a need for appropriate patients to receive optimal care at their base hospital. Additionally, urgent and emergency non-cardiac surgery ought to be performed at the base hospital. While nitric oxide is a core cardiac therapy we have shown that it is scarce in Scotland and unfamiliar to many ICUs. There is a need for a national discourse and consensus to ensure that nitric oxide is more widely available as part of a bundle of optimal cardiac critical care. This should include education, material resources, clinical guidelines and perhaps cardiac critical care outreach services to support general ICUs.

Introduction: Scarcity of Intensive Care Unit (ICU) beds has long been a problem. Among other things, it increases the work load of Emergency Department (ED), contributing to its crowding and probably to worst care, jeopardizing outcomes. Despite the plausibility of this premise, studies aren't consensual about the impact on outcome of delayed ICU admission from ED. Hospital de São João is a Portuguese tertiary care center. ED receives around 150 000 adult admissions per year, and is spatially organized according to Manchester triage priorities. Emergency Room (ER) receives patients from the street and all patients from other areas of the ED that need critical care. It is staffed by trained personnel and is equipped with level III ICU material. Intensive care department is composed by 37 level III and 28 level II ICU beds Objectives: Assess if there's a link between time spent in ED and outcome of patients admitted to level II and/or III ICU beds.

Methods: This is a retrospective study analysing older than 18 years old patients admitted to ICU from ED from 1 st January to 31 st December 2014. We excluded patients transferred from other hospitals. Demographic and clinical data was collected from records. We selected hospital outcome (dead, alive, transferred), hospital length of stay, ICU length of stay, vital status at 28 and 90 days after admission and ED and ER duration as outcomes. Simplified Acute Physiologic Score (SAPS) II and Sequential Organ Failure Assessment (SOFA) were calculated by considering the worst values in the first 24 hours of hospital admission. We performed a descriptive analysis, with median and interquartile ranges presented for continuous variables and proportions for categorical variables. For analysis of subgroups we did a chisquare or Mann-Whitney test. Statistical analyses were done on IBM-SPSS (version 20). A p-value of < 0,05 was considered significant.

Results: 147239 adults were assisted in ED in this period, with a median length of stay of 434 minutes. 500 were admitted to ICU beds, which accounts for 0,34 % of all adults cared for in ED. Around 66 % of patients admitted in ICU were treated in the ER at some point of their ED care. Patients admitted to ICU stayed around 470 minutes in ED. The more severe the disease, the least time spent (p = 0,001). Patients treated in ER were significantly more likely to be admitted quickly in ICU (p < 0,001). Taking in consideration the time spent in the ED, we found an opposite relation with global outcome, meaning that patients staying longer periods in ED had lower ICU mortality and lower length of stay in ICU. There was no association with hospital mortality.

Conclusions: Time spent in ED had no negative impact on outcome. However, given the fact that the majority of patients admitted to ICU beds were cared in a devoted area with trained staff and full level III equipment, we hypothesize that what might impact the most on outcome is provision of early critical care.

Determination of ICU bed requirement using resampling K.K. Introduction: Planning for ICU-bed provision, with a statistical confidence level, required the average number of critically-ill patients, their average ICU length of stay (LOS), and the fluctuation/variance of these two parameters. The actual ICU bed occupancy would under-estimates the variance, as ICU could never exceed its full capacity. With an under-estimate, the predicted ICU bed requirement would be inaccurate, with a tendency of under estimation. Objectives: Estimate the bed requirement to cover 97.5 % of time, by resampling of admission/discharge entries in 2014, for the two busiest ICUs in Hong Kong (~1600 admissions/year each) Methods: We assumed that the chance of an ICU admission was identical in a period of four weeks before and after a certain date. Based on this assumption, a computer simulation of ICU admissions was performed as if the year 2014 happened again. In brief, we pooled patients admitted on a particular date in 2014, and those admitted on the same day of week in the previous four and subsequent four weeks. Then patients were randomly selected from the pool to simulate ICU admission on that particular date. A mechanism (not described here) was in place to handle the public holidays. The hourly ICU occupancy was calculated using the actual ICU LOS of the selected patient. Re-sampling for the whole year was repeated 200 times to provide the estimates required.

Results: The actual hourly medians of ICU occupancy were 82 % and 77 %. They were close to that obtained using resampling (77 % and 77 %). As predicted, the distributions of the actual occupancy were skewed to left, indicating a negative bias on the variance estimates. The observed standard deviations of the two ICUs' occupancy were 11.5 % and 8.6 % respectively. After resampling, the distributions became more symmetrical, and had higher standard deviations of 18.9 % and 16.9 % (both p = 0.000). The 97.5 percentile occupancy in reality were 95 % and 92 %, while that from resampling were significantly higher at 118 % and 112 % (both P = 0.000). This corresponded to three or four additional ICU beds in each ICU. Conclusions: In conclusion, using a simple and conservative assumption, resampling could provide valuable insight for ICU bed planning.

Introduction: The number of available intensive care unit (ICU) beds are limited while the request for the beds are high. Thus rationing the admission to ICU is necessary especially in developing countries where the resources are limited. Also models are needed to estimate and re-estimate regularly, shortage in the number of ICU beds in any hospital. In the current study we tried to design a model for estimating shortage in the number of intensive care beds in a developing country tertiary university hospital after an initial Delphi consensus study.

Objectives: Designing a model for estimating shortage in the number of ICU beds in a hospital. Methods: Initially the standard indications for ICU triage were extracted from the literature. Four intensivists were served as steering committee and the initial questionnaire were further prioritized by 22 experts with three rounded Delphi method and formed a standardized checklist for ICU triage. Indications were considered as critical, important, and all indications. Then a cross-sectional study being performed during a 1-month period from August to September 2013 for all admissions to Nemazee hospital, a tertiary healthcare center affiliated to Shiraz University of Medical Sciences. Cardiac, transplantation and pediatrics patients were excluded from the study, as to be studied separately. The checklist were filled every day by an observing physician and any indications for ICU admission were marked in the questionnaire. Decision making for requesting ICU admission were performed by the specialized physicians of each ward regardless of the results of completed checklists.The results were further assessed according to the mentioned criteria and the reliability and viability was calculated. Finally assuming that there was no available ICU bed-days, the required ICU bed-days were compared with the total ICU bed-days of the hospitals, to estimate the shortage of ICU beds. Results: Totally 893 patients were admitted and studied.The required bed-day regarding critical indications, important indications, and all indications for ICU admission was 561, 1083, and 1950. By comparing the required bed-days with available bed-days of the hospital, 30 beds were calculated as shortage of ICU beds.

The results of the current study indicate that our center has deficiency in the number of ICU beds. It seems that a checklist is not only useful for prioritizing patients but also it is useful for estimating the required number of the ICU beds.The actual number of shortage is greater as three group of patients were not included.

Transfer delay from intensive care unit: retrospective analytical study in an Indian tertiary care hospital S. K Introduction: There lies scarcity of Intensive Care Unit (ICU) beds in every tertiary care hospitals, and on top of it delayed transfer of patients from ICU to wards is further increasing the burdensome. Numerous factors affect in making delayed transfer, which in itself is a risk factor for patient related morbidity and mortality, especially the after hour transfers.

Objectives: The aim of the study was to analyze the hours of transfer delay and their effect on readmission rates in the ICU. Methods: We conducted a retrospective study of patients transfer from our ICU to the wards over last one year (Jan-Dec'2015).Data collected from the ICU database by the secretarial staff during the study period and divided into following categories of transfer delays:

Results: There were 3362 patients admitted to our ICU during the study period of which 2475 patients were shifted to the wards. The average delay in shifting was around 6.5 hours (2-10.5 hrs).Delayed transfer of more than 8 hrs was found in 64 % patients and the percentage of after-hours transfer was 43 % of the total transfers. There were 16 readmissions into the ICU within 48 hrs of shift out among patients transferred in after hours as against 3 in patients transferred during routine hours.

Conclusions: Prevalence of delayed discharge from ICU was significant, especially the after hour discharges, which has got an impact on readmission rate as well. Discharge delay should be considered as an important quality indicator for critically ill patients to decrease the morbidity and mortality in ICU patients. Further studies are warranted to identify factors associated with delayed discharge. Introduction: Critical illness (CI) and stay in an intensive care unit (ICU) are known to induce physical and functional changes. Bone is often forgotten in survivors. Limited published data reported an altered bone metabolism in case of prolonged ICU stay [1] and a decreased in bone mineral density (BMD) in the year following ICU admission [2] . Clinical impact of these changes is still not well described.

Objectives: Our retrospective study aimed to assess incidence of any new bone fractures (BF) two years after a severe CI. Methods: Patients admitted in our ICU during 2013 were screened. Adults >18 years (y) old with an ICU length of stay (LOS) >7 days (d) were included. Lost to follow-up were considered exclusion criteria. Patients who died in ICU or who died during the follow-up period (FUP) with an ICU LOS ≤ 7d were also excluded. Demographic data, medical Objectives: Pulmonary arterial hypertension (PAH) is associated with reductions in health-related quality of life (HRQL). The patient care still played an important role in improvement of HRQL, even though more drug therapy was identified in recent decade.. In this study, we investigated to provide quality care for patients with pulmonary arterial hypertension via multidisciplinary care model. Methods: A multidisciplinary team was organized.in a tertiary medical center, including intensivists, cardiologists, pulmonologists, cardiac surgeons, rheumatologists, chest surgeons, rehabilitation physicians, psychologist, pharmacologists, hospice care physicians, nutritionist, social workers and nursing staffs. The key interventions include home based rehabilitation therapy, 24 hours hot line care, PAH care nurse training program, hospice care information and consultation, phychological care and autogenic training, prompt PAH referral system, social care connections, on-line self PAH risk assessment system, on-line and innovative mobile apps patient instructions, facebook patient care group and ourdoor PAH patient education program. The PAH patients were divided into three groups: pre-interventional group from May to Dec 2013, Interventional group from Jan to June 2014 and post-interventional group from July 2014 to Feb 2016. HRQL was measured using the Short Form 36 Health Survey (SF-36) in all enrolled subjects. Results: The average physical compartment scale of SF-36, including physical functioning; role limitations due to physical health, pain and general health improved from 49 ± 30 in pre-interventional group, to 52 ± 34 in interventional group and to 72 ± 22 in post-interventional group (p < 0.05). The average mental compartment scale of SF-36 , including role limitations due to emotional problems, energy/fatigue, emotional well-being and social functioning, improved from 54 ± 30 in pre-interventional group, to 56 ± 31 in interventional group and to 71 ± 17 in post-interventional group (p < 0.05). Conclusions: The study demonstrated multidisciplinary care model could improve HRQL of patients with pulmonary arterial hypertension.

Blood pressure management with urapidil for patients with aortic dissection is associated with less esmolol usage than nicardipine J.-C. Zhou Sir Run Run Shaw Hospital, Intensive Care Medicine, Hanghzou, China Intensive Care Medicine Experimental 2016, 4(Suppl 1):A1136 Introduction: Acute aortic disease is a common but challenging entity in clinical practice. Titration the blood pressure and heart rate to a target level is of paramount importance in the acute phase regardless of whether the patient will undergo a surgery or not eventually. In addition to the initially intravenous β-blockers, parenteral infusion of nicardipine and urapidil are the most common used antihypertensive therapy currently in mainland China. However, few empirical data was available with respect to the different effect on patients' outcome of the two antihypertensive strategies, especially given the deleterious reflex tachycardia of vasodilators which may increase force of ventricular contraction and potentially worsen aortic disease. Objectives: To evaluate the difference of the abovementioned two antihypertensive strategies on the outcome of patients with aortic disease. Methods: All patients with new diagnosed aortic diseases presented to our hospitals from January 1, 2013 to June 30, 2014 were retrospectively reviewed. The antihypertensive strategies and their association with patients' outcomes were evaluated with logistics regression. Results: A total of 120 patients with new diagnosed aortic disease were included in the study. Of them, 47 patients received urapidil while 73 patients received nicardipine antihypertensive therapy. Patients with nicardipine were more quickly to reach the target blood pressure level than those treated with urapidil (median 18 vs 35 mins, P = 0.024). After adjustment for patient demographics, comorbidity, involved extend of aorta, interventional strategies, antihypertensive therapy with nicardipine (with urapidil as reference) for patients with aortic disease was significantly associated with high esmolol cost (OR: 6.2, 95%CI: 1.8-21.6, P = 0.004) and longer ICU length of stay (OR: 3.9, 95%CI: 1.5-10.3, P = 0.006). However, there was no significant correlation between nicardipine use and ICU mortality (OR: 0.3; 95%CI, 0.1-1.4, P = 0.123). Conclusions: Although nicardipine achieved the target blood pressure level more quickly than urapidil for patients with aortic disease, it was associated with more esmolol use and longer ICU length of stay.

Introduction: Postoperative bleeding is one of the most common complications of cardiac surgery. Excessive perioperative bleeding continues to complicate cardiac surgery with cardio-pulmonary bypass (CPB) in spite of improvements in extracorporeal oxygenation and surgical techniques. Even bleeding after cardiac surgery has variable causes, we thought the applying ISTH scoring system may be able to predict the postoperative excessive blood loss in patients after cardiac surgery with CPB. Objectives: The aim of present study was to examine the effectiveness of International Society on Thrombosis and Hemostasis (ISTH) scoring system in patients with cardiac surgery. Methods: The medical records of patients undergoing elective cardiac surgery using CPB between Mar. 2010 and Feb. 2014 were retrospectively reviewed. These demographic and clinical characteristics, perioperative laboratory findings, and postoperative complications were assessed using computerized databases from our institution. ISTH score was calculated in ICU and Patients were divided with overt DIC group and non-overt DIC group. Results: Among 384 patients with cardiac surgery, 70 patients with Overt DIC Group (n = 20) or Non-overt DIC Group (n = 50) were enrolled. Mean DIC scores at ICU admission was 5.35 ± 0.59 (Overt DIC Group) and 2.66 ± 1.29 (Non-overt DIC Group) and overt DIC was induced in 29 % (20/70). Overt DIC group had much more EBL for 24 hrs (P = 0.006) and maintained longer time of intubation time (P = 0.005) Conclusions: In spite of limitation of retrospective design, management using ISTH score in patients after cardiac surgery seems to be helpful for prediction of the post-CPB excessive blood loss and prolonged tracheal intubation duration. renal failure, 65.6 % vs 20 %, p ≤ 0,001; respiratory failure, 40.6 % vs 13.5 %, P ≤ 0,001; Mechanical > 48 h 67.7 % vs 13.7 %, P ≤ 0.001 ventilation. The variables that reached statistical significance in the multivariate analysis as predictors of mortality were APACHE II OR 1.13 (95 % CI 1.1-1.2), P = 0.001, Euroscore OR 1.4 (95 % CI 1,2-1.6), P < 0.001; acute respiratory failure OR 3.4 (95 % CI 1.4 to 8.2), P = 0.008; acute renal failure OR 3.1 (95 % CI 1.3-7.5), postoperative bleeding OR 4.9 (95 % CI 2.1 to 11.9), P < 0.001 Conclusions: Mortality in this group is similar to other series, being patients with more comorbidities, with the highest score in the Euro-SCORE and APACHE II and more often subjected to mixed surgery. The Euroscore, APACHE II, respiratory failure, renal failure and postoperative bleeding, predict higher mortality.

Methods: We measured rEa as ratio of pulmonary pressure at the dicrotic notch (dyPAP) and stroke volume (SV) [2] and rEes as ratio of the difference between mean pulmonary artery pressure (mPAP) and wedge pressure (PCWP) and end systolic volume (rESV) (mPAP-PCWP/rESV) [3] after the induction of anaesthesia (T0) via pulmonary artery catheter (SwanGanz 774 F45 and Vigilance II Monitor by Edwards Lifesciences), after weaning from CPB (T1) and 4 h after in ICU(T2) in 4 patients. Results: Measure of rVAC has been demonstrated feasible in all four patients undergoing cardiac surgery. As expected all the patients were found uncoupled (rVAC > 1) before surgery, immediately after weaning from CPB rVAC worsen and in ICU it was restored to the basal. Conclusions: In this preliminary analysis we demonstrated the feasibility of measuring rVAC in critical patients undergoing cardiac surgery, to our knowledge this is the first report in this field. As expected rVAC is very much influenced by CPB although further investigation is needed to confirm the utility of this technique to monitor the right heart in such patients.

Introduction: Detection of tissue hypoperfusion is paramount in the management of VA ECMO. Arterial to pulmonary artery CO 2 difference has been demonstrated to be an early marker of hypoperfusion in the shock patient [1] and during hypothermic cardiopulmonary bypass [2] , Objectives: In this report we investigated the accuracy and feasibility of mixed venous to arterial CO 2 difference as an early marker of perfusion mismatch during VA ECMO. Methods: In a patient treated with VA ECMO for refractory cardiac arrest due to acute myocarditis we performed serial measurements of pulmonary artery to arterial CO 2 difference as well as SVO 2 , MAP, urine output and lactate level. Results: During reduced perfusion periods, assessed by elevated lactacidemia (>2 mmol/l) we observed high > 7 CO 2 difference which is concordant to literature [3] . During 15 episodes of reduced systemic perfusion, demonstrated by increase of serum lactic acid we were able to early detect hemodynamic derangement (avg 16 minutes) by identifying elevated (>6 mmHg) CO 2 difference. Conclusions: This case report underlines the importance of pulmonary artery to arterial CO 2 difference as an early marker of hypoperfusion if compared to lactate level in the intensive care unit. To our knowledge this is the first report on venous to arterial carbon dioxide difference in VA ECMO. Further investigation is needed to confirm those preliminary results. Introduction: Many studies have shown clinical benefits from SDD for critically ill patients. However, there is still doubt concerning the emergence of antimicrobial resistance in the long term. Previously no evidence to support this view was found but long-term effects of SDD on antimicrobial resistance on the unit level is understudied. 1, 2 Objectives: To determine the incidence of antimicrobial resistance in aerobic gram-negative potentially pathogenic micro-organisms (AGNBs) to the components of SDD and frequently used i.v. antibiotics on ICU-level over a 20 year period with unchanged antibiotic policy.

Methods: This is a single-center observational cohort study in a Dutch 20-bed adult intensive care unit in a teaching hospital. All consecutive patients admitted to the ICU between January 1994 and December 2013 were included when at least one culture was taken during ICU-admission. Data on all cultures taken during ICU stay were collected from the hospital database. Susceptibility testing was performed following the guidelines of the 'clinical and laboratory standards institute' (CSLI) until 2011 and 'the European committee on antimicrobial susceptibility testing' (EUCAST) from 2011 until 2013. Incidence rates of antimicrobial resistance to tobramycin, ciprofloxacin, polymyxin B or cefotaxime were calculated per year. Only ICUacquired resistant pathogens were selected by excluding resistant pathogens in cultures taken on day 1-day 3. Patients at risk were defined as all admissions with a length of stay longer than 2 days. Differences between the incidence in the first and last year of the study were tested using Chi-square test. Results: Data of 127.830 cultures was analyzed containing 15.557 AGNBs. The number of admissions with a length of stay more than 2 days was 6781. In 237 admissions newly acquired resistance to cefotaxime was found, in 117 to polymxin B, 183 to tobramycin and 180 to ciprofloxacin. Figure 136 presents incidence rates per year. In 1994-1996 date of discharge to the ward was unknown and therefore incidence rates could not be calculated for these years but absolute numbers were comparably low. There was no significant difference in incidence of ICU acquired resistance in cefotaxime (χ 2 = 0.14, p = 0.71), polymyxin B (χ 2 = 0.71, p = 0.40), tobramycin (χ 2 = 1.89, p = 0.17) and ciprofloxacin (χ 2 = 0.28, p = 0.59) between 1997 and 2013. Conclusions: The incidence of newly acquired AGNB resistant to cefotaxime, polymyxin B, tobramycin and ciprofloxacin continues to be low during a 20 year unchanged antimicrobial policy of SDD. The increase in resistance in the society may impact these numbers and should be studied. Results: Overall, ICU and H mortality rates were 94 and 87 %. 17.7 % of the pts (55/310) became infected in the early postoperative period. (ICU and H mortality rate 21 and 29 % respectively). CRPK infections were present in 8 pts (2.5 % of the entire series, 14.5 % of the infected pts). Sarcopenia (50 % vs 10 %, p = 0.0045) and MELD (29 + 9 vs 19 + 8, p = 0.0022) were significant preoperative risk factors. ICU and H mortality rates were 37 % and 75 % in CRPK pts, 19 % and 25 % in non -CRKP infected pts respectively : while ICU mortality was not different (p = 0.65), H mortality was significantly higher in CRKP pts (p = 0.0056 ICU vs H, CI 7.3 -74.6). If compared to non -CRKP pts , CRKP pts were more often in septic shock (75 % vs 34 %, p = 0.031) and more frequently underwent CRRT (75 % vs 36 % p = 0.0404). Intraabdominal infections were largely represented (80 %) among CRKP pts. Blood loss and transfusion needs, Early gratt dysfunction and reOLT were more represented in infected vs non infected pts. However, no differences were found when CRKP and non-CRKP transplanted pts were compared. Conclusions: CRKP infections are on the rise also in Italy. Post OLT mortality is high and strategies able to control CRKP are urgently needed to be implemented. Introduction: The prevalence of antibiotic-resistant pathogens in ICU conditions makes it difficult to treat these infections, and treatment becomes impossible in some cases. Acinetobacter baumannii is important infectious agent ICU patients, which effective antibiotic therapy is currently limited. Objectives: We aimed to determine the range of A.baumannii associated infections among ICU patients, to summarize the level of resistance to antimicrobial drugs, and provide an overview of strategies to prevent the spread of resistance. Methods: A prospective microbiological study of the prevalence and antibiotic resistance of A.baumannii strains isolated from adult ICU patients hospitalized to the tertiary hospital after cardiac surgery from 2010 to 2014. Results: A total of 781isolates from ICU patients were included to the study. 52.3 % of the isolated strains (409) were Gram-negative, among which 20.9 % (164) of A.baumannii isolates. Strains of A. baumannii showed a high level of resistance to the III generation cephalosporins (96.9 % to ceftazidime, 97.6 % to cefotaxime, 93.5 % to ceftriaxone). Resistance to carbapenems was at 88 %. Investigation of antimicrobial activity of ciprofloxacin showed the resistance in 96.8 % of strains, to levofloxacin -88.9 %. The lowest level of resistance recorded to doxycycline -24.5 % and polymyxin -4.8 %. Conclusions: Rapid microbiological diagnostics (including the results of antibiotic resistance), strict adherence to infection control, the appointment of an effective regime of antibiotic therapy, optimization schemes appointment of antibiotics, all of which are the most important priorities for the effective fight against A. baumannii associated infections in ICU patients. In order to reduce the emergence and spread of drug-resistant strains in the ICU, it is strongly recommended to carry out microbiological monitoring and optimization of the use of antibiotics in each hospital. Therefore local resistance surveillance programs have the greatest value in the development of appropriate therapeutic recommendations for specific types of patients and infections.

Introduction: Acinetobacter spp. are opportunistic, nosocomial pathogens that may colonize the surfaces in intensive care units. Their tendency to harbor multi-drug resistance and to develop resistance mechanisms to commonly available drugs make their treatment a challenge. Carbapenem resistance, and newly reported colistin resistance has led to a search for new treatment options. There are in vitro studies which report synergistic effect with rifampicin in combination therapies. Objectives: We aimed to present and discuss the results of our patients who were infected with either panresistant (5 patients) or only tigecycline susceptible (5 patients) acinetobacter spp. and were treated with rifampicin combination regimens. Methods: Patients reported to be infected with colistin resistant acinetobacter spp. and treated with rifampicin combination regimens upon decision of the responsible teams were traced from the intensive care unit (ICU) records between the years 2014 and 2016 retrospectively. Their demographic data, liver function tests, ICU and hospital outcomes were recorded. Results: There were a total of 10 patients, 6 were women. Mean age was 69.5. In 8 patients pulmonary site was the source. Nine patients had positive blood cultures. Mean SOFA score at the start of therapy was 14.5; all were intubated, and 5 (50 %) were on vasopressor therapy. Combination regimens comprised of at least 3 antibiotics and all regimens included rifampicin and tigecycline. At the end of first week, mean SOFA score was 12.6. Of these 7 (70 %) survived to hospital discharge. Patients who were lost had higher initial and follow-up SOFA scores. Initial and follow-up liver enzymes and renal function tests were similar to their basal values in patients who survived; unlike the patients who were lost. When lost patients were re-evaluated: the first patient had irreversible lung fibrosis due to bleomycine; in the second patient; combination treatment was delayed until 9 days after the cultures were performed; the third patient had been admitted to ICU with acute renal failure and acute respiratory distress syndrome, after autologous stem cell transplantation for multiple myeloma. Conclusions: When the importance of accurate antibiotic choice is taken into account for treatment success; rifampicin combinations may be considered as an appropriate treatment option for infections caused by colistin resistant acinetobacter strains.

Introduction: Carbapenem resistant enterobacteriacae (CRE) emerged in recent years as one of the most challenging group of antibiotic resistant pathogens. Polymyxins are considered as the last resort for the treatment of infections with carbapenem resistant gram negative bacilli (GNB). Inadequateor extensive use of colistin leads to emergence of colistin resistance, increasing mortality and morbidity and necessitating prudent use of alternative antibiotics. Fosfomycin, a phosponic acid derivative which acts by disrupting bacterial cell wall synthesis, is a broad spectrum antibiotic. It is available as sodium/disodium formulation for intravenous use and is showing promising result against multi drug resistant(MDR)/Pan Drug Resistant (PDR) pathogens. Methods: A total of eight colistin resistant (MIC ≥ 4) GNB were isolated from ICU patients with nosocomial MDR infections during a period of one year. All eight isolates were Klebsiella pneumonia. Among these isolates five were from blood and three from endotracheal aspirate. All the isolates were sensitive to fosfomycin in vitro. All of these patients had multiple co-morbidities with recent history of colistin exposure. Intravenous fosfomycin was given as a combination therapy. Results: Among the five bacterimic patients, three recovered completely from sepsis. One patient took discharge against medical advice and the only one bacterimic patient who died during the course of therapy was later on diagnosed to have azole resistant fungemia as super infection. The patient with ventilator associated pneumonia also responded well after initiation of fosfomycin therapy. Average duration of antibiotic therapy in all these cases was ten days. Conclusions: Based on the evidence of clinical experience and available studies, intravenous fosfomycin therapy may be considered as the last option for the treatment of MDR GNB infection where there is documented colistin resistance and where there is literally no other choice of antibiotic therapy. The success of the therapy is encouraging in selected group of patients. Further research on intravenous fosfomycin use specially against MDR pathogens and on the effectiveness and safety of the drug in the treatment of patients with such infections may be warranted. Introduction: Patients at the Intensive care units have an increased risk of infection due to their underlying diseases or conditions, impaired immunity, and exposure to multiple invasive procedures (surgery, mechanical ventilation, central venous catheters, artherial catheters, urinary tract catheters). Multidrug-resistant organisms infection has become a public health problem and has been associated with increased morbidity, mortality, and costs. Objectives: To analyze the principal features of postsurgical patients with colonization or infection by multidrug-resistant organisms, acquired before the admission at the Intensive Care Unit. Methods: Retrospective observational study, descriptive, case series, collected from 04/01/2014 to 29/11/2015 in a 400-bed hospital, with a 10-bed polyvalent Intensive Care Unit in Fuenlabrada, Madrid, Spain. The Hospital is attached to Resistance Zero Project, with screening at admission and every week in all patients (pharyngeal, rectal, nasal, wounds and bronchial suction). 18 postsurgical patients have been identified with multidrugresistant organisms isolation in screening in the first 24 hours of admission to the Intensive Care Unit. Studied Variables: age, sex, adjusted Charlson comorbidity index, Barthel index, APACHE II, SAPS 3, days of hospitalization prior to ICU, days of antibiotic treatment administered before ICU, previous days of parenteral nutrition, prealbumin, surgical wound infection, multigrug-resistant organisms identified sample. Statistical analysis: SPSS20. Categorical in frequencies and percentages, mean and standard deviation or median and interquartile range. Analysis Kolmogorov Smirnov, Shapiro Wlik and QQPlot to normality. Confidence intervals (CI) 95 % by T Student for normal variables, Boot stramp to not normal. Results: In 20 months we identified 18 postsurgical patients with multidrug-resistant organisms in screening at the admission or before the admission to the Intensive Care Unit. Multidrug-resistant organisms identified: Pseudomonas aeruginosa 6 (33.3 %), ESBL enterobacteriaceae 6 (33.3 %), MRSA 3 (16.7 %), Stenotrophomonas maltophilia 1 (5.6 %). Isolated on: surgical wound 7 (38.6 %), bronchial suction 2 (11.1 %), peritoneal fluid 2 (11.1 %), exudates monitoring 2 (11.1 %), blood 1 (5.6 %). Antibiotic therapy: Carbapenem 8 (44.4 %), Piperacilina-tazobactam 6 (33.3 %). Conclusions: In our study the risk of prior acquisition of multidrugresistant organisms at the admission to the Intensive Care Unit in postsurgical patients was characterized by long hospital stay, high comorbidity and dependence, malnutrition, prolonged use of broadspectrum antibiotic, parenteral nutrition and surgical wound infection. Introduction: The emergence and dissemination of Klebsiella pneumoniae carbapenemase (KPC) is of great concern. Outbreaks have been reported in different types of Intensive Care Units (ICU). In Brazil, there have been reports of KPC since 2006. We recently experienced a large outbreak at our hospital. Risk factors for KPC colonization and outcome of ICU patients are still to be determined. Objectives: To study the differences between patients who acquired from those who did not acquired KPC during their stay in the ICU, focusing on risk factors and outcomes. Introduction: Patients with a prolonged weaning represent a small part of the total ICU population but this prolonged state has many implications on their later recovery and can highly impact health expenditures. Objectives: To better characterize patients with prolonged weaning and assess factors associated with their survival. Methods: The prospective multicentre observational WIND (Weaning accordIng New Definition) study was performed from April to August 2013. Ventilation and weaning modalities were daily assessed until discharge in all intubated patients admitted to the participating ICUs. We defined 1) weaning attempt (WA) as a spontaneous breathing trial (SBT) or an extubation (with or without SBT), 2) successful weaning as an extubation without death or invasive mechanical ventilation within 7 days. We considered patients as having a prolonged weaning if weaning was not terminated at 7 days following their first WA. Conclusions: In this multicentre international prospective cohort, 11.5 % of the patients entering the weaning process had a prolonged weaning with a high mortality rate of 29.8 %. The only baseline factor associated with death were previous immunodeficiency and chronic cardiac failure. These patients highly impact the ICU workload as they receive mechanical ventilation for a median duration of 19 days and their median length of stay in the ICU is 25 days. Patients with a prolonged weaning spend a long ICU time after the end of weaning without mechanical ventilation raising the issue of the need for specialized units. Introduction: Ineffective efforts (IE), defined as the inability of patient's inspiratory effort to trigger a ventilator-delivered breath, is a commonly encountered asynchrony, and has been reported to adversely affect patient outcome1,2,3. The incidence of IE depends on several factors, including patient population, ventilator settings, and the observation period, which in most studies so far was limited2,3.

Objectives: Aim of this study was to investigate the incidence of ineffective efforts, using continuous recordings, in critically ill patients mechanically ventilated only on assisted mode and their potential effects on patient outcome. Methods: 110 adult critically ill patients hospitalized in the ICU of the University Hospital of Heraklion on mechanical ventilation for >12 h were enrolled. Patients were studied when they were on assisted ventilation for >1 hour and expected to remain on assisted ventilation for the next 24 hours. Patients were studied again on the 3rd and 6th day if they remained on assisted ventilation. Continuous 24 h measurements were obtained using a monitor validated to identify ineffective efforts (PVI monitor)4. The output of PVI monitor data was processed before analysis to optimize data quality and re-sampled to a time-series with the number of IEs calculated in uniform intervals of 30secs while preserving the total number and duration of IEs5. The IE index was calculated as previously described. Because IE occurred in clusters, the concept of IE event was introduced, to describe variable periods of time containing IE > 10 % of breaths. IE events were characterized by their duration and power (number of IE)5. Introduction: The neural timing during mechanical ventilation can be obtained from conventional airway flow tracing, or invasive esophageal and gastric signal; however, it is difficult clinical practice and could be imprecise. The first derivative of airway flow signal show line segments with distinctly different slopes and with welldefined the inflections points, therefore this closely indicate the respiratory times, it can be calculated easily.

Objectives: To evaluate the accuracy of the derivative of the flow signal (DF) as method for measurement of the respiratory times compared with esophageal-gastric signals. Introduction: Providing appropriate levels of pressure support (PS) at the bedside is challenging. Physicians should avoid both over-support, which increases the risk of lung trauma, muscle atrophy and prolonged weaning; and under-support, which increases the risk of patient discomfort and respiratory muscle fatigue. The latter can be determined by the using the Tension Time index of the inspiratory muscles (TTi es ) derived from measurement of esophageal pressure. TTi es values higher than 0.18 indicate fatiguing patient effort. The Beacon Caresystem (Mermaid Care, Denmark) advises on level of PS using physiological models of lung mechanics, pulmonary gas exchange, respiratory drive, acid-base status and muscle function; along with clinical preference functions quantifying the risk of muscle atrophy, patient stress, and lung trauma. Mathematical models are tuned to measurements allowing advice to be patient specific.

Objectives: This study investigates the variation of TTi es and other indices of respiratory muscle function induced by an increase/decrease of the level of PS, and whether the consequent advice proposed by the Beacon System results in appropriate patient effort.

Methods: Ten patients with acute respiratory failure residing in an ICU in Ferrara, Italy, have currently been included for this analysis. An esophageal balloon was inserted and its correct position determined by the occlusion test. The advice of the Beacon system was followed for an hour from states of over-and under-support defined as 150 % and 50 % of baseline PS. The level of PEEP was kept constant throughout the study. Data were analysed in terms of TTi es and esophageal pressure developed in the first 100 ms of an occluded inspiration (P 0.1 ) Results: The baseline TTi es values of 0.10 ± 0.05 were consistent with absence of fatiguing effort in all patients but one. As expected, reducing/increasing the level of PS resulted in TTi es and P 0.1 increase or decrease, respectively. In 3 patients the reduction of PS was associated with impending muscle fatigue. The levels of PS proposed by the Beacon system resulted in TTi es of 0.13 ± 0.01, slightly higher than obtained by the treating physician, but always below the values indicating muscle fatigue, a part from the patient in which the TTi es indicated fatigue at baseline. Of note, this new value of TTies was not associated with a significant variation P 0.1 , which implies that the proposed level of PS was not associated with an increased respiratory drive or higher transpulmonary pressure Conclusion: These initial results indicate that Beacon Caresystem responds appropriately to over-and under-support avoiding muscle fatigue and excessive P0.1.

The use of vo2 level changes as a predictor for weaning success in the mechanically ventilated patients Introduction: Experimentally, hyperchloraemia may induce vasoconstriction of the renal afferent arterioles and tubular dysfunction, potentially resulting in acute kidney injury (AKI). The clinical implications of these findings are not well established, especially in septic patients.

Objectives: To investigate whether chloride serum and urinary concentrations as well as chloride load, output, balance and urinary anion gap are associated with the development of AKI in septic patients. Methods: Retrospective analysis of an institutional database including all patients admitted to the Intensive Care Unit (ICU) for severe sepsis and septic shock from January 2011 to June 2015. Inclusion criteria were length of stay in the ICU ≥ 72 hours and complete data available on serum and urinary samples for at least 3 days. Patients were excluded if they had anuria on ICU admission, continuous bladder irrigation, if they were on hemodialysis (of recent onset and chronic) and if they were kidney-transplanted. Demographics and data on outcome were also analysed from the database. We collected chloride levels on daily blood (BCl) and urinary (UCl) analyses; chloride load (CL) was calculated by considering the amount of chloride present in the IV fluids administered daily to the patient, while chloride balance (CB) was calculated as: CL -CO, where CO is chloride output (UCl * daily urine output). Creatinine clearance (CrCl) was calculated on 24-hr urinary collection. AKI was defined according to standard criteria. Conclusions: Most of septic patients developed AKI and this complication was associated with a significant reduction in renal chloride elimination. The impact of such findings on the management of fluid therapy in this setting remains to be further evaluated. Introduction: Plasma interleukin 6 (IL-6) is associated with acute kidney injury (AKI) in sepsis. IL-6 receptor (IL-6R) is not expressed in the kidney. Circulating IL-6 in a complex with soluble IL-6R (sIL-6R) activates ubiquitously expressed transmembrane signal transducing glycoprotein 130 on renal epithelial cells. This IL-6 trans-signaling is associated with mortality in experimental sepsis.

Objectives: To study IL-6 trans-signaling in patients with sepsis in a clinical intensive care setting. Methods: In 190 septic patients showing first organ failure at intensive care unit (ICU) admission ±24 hours, we measured plasma IL-6 and sIL-6R at admission and 24 hours later. Our primary endpoint was AKI during the first five ICU days by KDIGO criteria. Mann-Whitney's, Spearman's correlation and Chi Square tests were used. Results: Plasma IL-6 was significantly higher in patients with AKI at 0 h (p = 0.001) and 24 h (p < 0.001). Plasma IL-6 correlated with KDIGO stage at 0 h (R = 0.214, p = 0.003) and 24 h (R = 0.361, p < 0.001). Plasma sIL-6R did not differ between AKI and non-AKI groups. Using cut-off values of 328 pg/ml of IL-6 and 53491 pg/ml of sIL-6R at 0 h (detected by Youden method), the combination of low IL-6 and low sIL6R was associated with non-AKI (p < 0.001).

Conclusions: Combination of low IL-6 and low sIL-6R in plasma is associated with decreased incidence of AKI, suggesting that IL-6 transsignaling contributes to septic AKI.

18.5 % for the non-AKI group (OR = 3.11; 95 % CI, 1.72 -5.62, p < 0.001). Multivariate analysis indicated that the BPV was well associated with AKI (adjusted OR = 1.18; 95 % CI, 1.08 -1.28, p < 0.001) while the mean blood pressure was not (adjusted OR = 0.99; 95 % CI, 0.95 -1.02, p = 0.384).

Conclusions: Elevated blood pressure variability is associated with increased risk of AKI in septic patients. This understanding may be helpful to develop requirement for stabilising blood pressure in the BP management of septic patients.

Introduction: Acute kidney injury (AKI) is a frequent and serious complication of sepsis in intensive care units (ICU). According to acute kidney injury criteria (AKIN), the most current diagnostic criteria for AKI is an abrupt (within 48 hrs.) reduction in kidney function currently defined as an absolute increase in serum creatinine of more than or equal to 0.3 mg/dl, or1.5 fold from baseline or a reduction in urine output (documented oliguria of < 0.5 ml/kg per hr. for >6 hr.). By time of occurrence of these criteria actual kidney insult has occurred & probably this leads to late intervention for kidney protection &/or renal replacement therapy (RRT). So early prediction of AKI by using biomarkers like urinary angiotensinogen could help patients to benefit from a quicker and more appropriate therapy. Urinary angiotensinogen appears quite promising due to its reported correlation with the intrarenal angiotensinogen and Angiotensin II levels which play a major role in molecular mechanisms of AKI.

Objectives: The aim of this work was to evaluate the role of urinary angiotensinogen as a possible predictor of AKI in patients with severe sepsis. Methods: The study was carried on 100 adult patients who were admitted to the Department of Critical Care Medicine, at the Alexandria Main University Hospital and who suffered from severe sepsis. Patients were categorized into two groups according to AKI development; non AKI group which consisted of 30 patients (Group I), and AKI group which consisted of 70 patients (Group II). Patients were excluded if they have chronic kidney disease, already started RRT, received angiotensin convertase enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARBS), or septically shocked. Urinary angiotensinogen and creatnine were withdrawn once from each patient on the day of admission to calculate Urinary angiotensinogen/creatinine ratio (uAnCR, ng/mg). AKIN staging was assessted daily for seven days.

Results: there was a significant difference between the two studied groups regarding uAnCR Ratio on admission (p < 0.001), whereas this Introduction: Acute pancreatitis with organ dysfunction is termed severe acute pancreatitis (SAP) and complex SAP if local complications develop (such as infected pseudocyst). We receive tertiary referrals of complex SAP patients to our unit, who often have multiple CT scans. Muscle wasting is known to occur in critically ill patients (1) and can be quantified by measurement of the cross-sectional area (CSA) of para-spinal muscles at the third lumbar vertebral level on CT imaging. AKI is one of the most common causes of death in SAP patients (2) and is a risk factor for developing CKD (3). KDIGO guidelines suggest using creatinine changes to detect AKI (4) but creatinine changes may be inaccurate in the presence of musclewasting (myopenia) (3).

Objectives: To utilise measurements of L3 para-spinal muscle CSA (L3MCSA) from complex pancreatitis patients between April 2008-December 2014 and compare these to changes in plasma creatinine during their ICU stay. Methods: Patients were identified from our ICU patient database (WardWatcher software) and additional clinical details including creatinine/eGFR level on CT-scan days, were acquired from electronic databases. Images were exported from our PACS system as DICOM files and analysed using ImageJ software (REF) in duplicate by two independent users, average values were used. For patients who had no renal-replacement therapy (RRT), between-scan L3MCSA and creatinine change were paired and analysis was with Excel (MS) and GraphPad (PRISM).

Results: 45 patients met inclusion criteria. 21 patients had ≥ 2 CT scans in ICU, enabling serial estimation of L3MCSA. 8/21 (38.1 %) patients did not have RRT in ICU. There was no statistically significant difference in overall (start to end of ICU) % change of L3MCSA between patients who did/did not have RRT. There was also no correlation between overall (start-to-end of ICU) % creatinine change and % change/day L3MCSA: r = −0.14, p = 0.75. For between-scan data (n = 23): The median (IQR) % creatinine change/scan was −17.1 % (−9.0 to −44) and the % L3MCSA change/ scan was −5.75 % (−1.8 to −9.8). However, there was no correlation between % L3MCSA change and % creatinine change between scans ( r = −0.14, p = 0.62).

Conclusions: L3MCSA (relating to lean muscle mass) was shown to decrease in complex severe acute pancreatitis (SAP) patients. However, there was no correlation with change in L3MCSA and change in creatinine. This suggests that normal/stable creatinine values may be falsely reassuring in the context of muscle mass loss (myopenia) and ongoing AKI could be under-diagnosed. Acknowledging myopenia and interpreting creatinine value in context is therefore vital.

Introduction: Continuous renal replacement therapy (CRRT) is the most common therapy in critical ill patients with acute renal failure, having circuit coagulation as the most frequent complication. The CRRT circuit requires careful anticoagulation to avoid coagulation and bleeding complications. Critically ill patients with acquired antithrombin (AT) deficiency, may have a shorter filter lifespan.

Objectives: Evaluate the relation between the modification of AT levels from baseline and circuit survival during CCRT. We would like to determine the existence of an AT critical level, related to the risk of the clotting filter. Methods: We started an observational study with prospective data collection in a university hospital. From October 2013 to April 2015, 61 patients were included, with 122 filters in total. We measured the level of AT activity at the beginning (basal AT), daily, and at the moment of circuit coagulation. We divided the patients in two groups depending in their AT´s basal level (<60 % or > 60 %). Then, we observed the percentage of change in AT from baseline, and we divided the patients in tertiles to obtain three comparable groups. The main outcome measure was filter lifespan of first circuit and the correlation with AT´s levels. Results: Low AT´s basal level (<60 %) has significant association with longer filter life span (p = 0.009). We obtained three groups according to a percentage changes of ±8 % in AT from baseline. One group declined the AT´s basal level (8 % decrease), other had little changes (between 8 % decrease and 8 % increase) and the last one had an increase (8 % increase). The group which presented the highest percentual increase showed the largest median survival time to circuit coagulation (37 hours; 95 % CI: 26-48). We observed a significant association (p = 0.008) between the greater percentage change in AT from baseline, and a larger time intervals to circuit coagulation.

Conclusions: The circuit lifespan shows a narrow correlation with evolution of AT´s levels since the start of CRRT until filter clotting. AT measurement should be considered an essential factor during CRRT. Calcium supplementation was required with 34 filters (34 %) in 18 patients (60 %). In these patients, the median supplementary calcium dose (in addition to replacement fluid Ca) was 0.5 mmol/hr (0.023-1.39). 11 of those 18 were initiated with calcium with only 5 requiring further calcium in the next filter and 5 did not (1 patient who started on calcium only used 1 filter). One patient in the citrate group was discontinued for alkalosis. No patients were discontinued for hypocalcaemia. Conclusions: Post dilution RCA, using replacement fluid which contains calcium, in patients with a relative contraindication to heparin, reduces need for post filter calcium supplementation and provides acceptable filter life.

Mortality risk factors in continuous renal replacement therapy in a university hospital from Colombia C. Introduction: Acute kidney injury (AKI) occurs in more than 50 % of critically ill patients, 23 % need renal replacement therapy, preferring continuous therapies. However mortality seems not to change with this technology. The research available focus on the right time to start therapy, but only evaluating renal dysfunction characteristics.

Objectives: To identify mortality risk factors at the start of continuous renal replacement therapy (CRRT) for acute kidney injury and early mortality risk factors in this patients. Methods: A cohort study was performed in patients over 18 years old with AKI who required CRRT in the intensive care unit of a university hospital in Bogota Colombia between 2009 and 2014. The CRRT was provided with Aquarius® Edwards® technology, polyethersulfone membrane of 1.2 and 1.4 m2 (Aquamax®) and replacement fluids with lactate (Premixed®). Modality selection were guided by the hospital guideline. Sample size calculation was estimated selecting 10 cases (death) for each variable associated with mortality. A description of demographic and clinical variables was performed, bivariate analysis with mortality and early death defined as death within 24 hours of onset of CRRT, and finally we proceed to perform a multivariate prediction analysis. We considered statistically significant p value < 0.05. Results: A total of 214 patients required CRRT during the period, 20 (9.34 %) patients were excluded, (age under 18 years old, incomplete data and 10 chronic kidney disease on dialysis). The mean age was 61.5 years (±15.47), 57.73 % men. The most frequent cause of AKI was sepsis in 30.9 % of cases. A total of 774 CRRT days were conducted with a median of 3 days per patient (range 1-19). Mean Charlson comorbidity index was 5.22 (±2.85), APACHE II score 29.65 (±6.66), total non-renal SOFA had a median of 11 (range 6-18) at the time of starting therapy. The hospital mortality was 68.4 % and early mortality was 19.07 %. In multivariate analysis: age (p = 0.004), SOFA (p = 0.038), days door-support (p = 0.03) and the presence of hypotension (p = 0.004) were independent risk factors for hospital mortality with an area under the curve of 0.76. For early death lactic acid levels (p = 0.007), glucosa (p = 0.01) and age (P = 0.02) were independent risk factors with an area under the curve of 0.73. Conclusions: Patients with AKI on CRRT have high mortality. Age, multiple organ dysfunction, hypotension and time door-support were independent mortality risk factors. Low levels of glucose and high lactate at onset of CRRT are independent risk factors of early death. 

Pediatric formulas for the anesthesiologist

The role of bystanders, first responders, and emergency medical service providers in timely defibrillation and related outcomes after out-of-hospital cardiac arrest: Results from a statewide registry

Kids save lives-Training school children in cardiopulmonary resuscitation worldwide is now endorsed by the World Health Organization (WHO)

Acute Skeletal Muscle Wasting in Critical Illness

Enhancing rehabilitation of mechanically ventilated patients in the intensive care unit: A quality improvement project

pdf 2. NICE quality standard (QS2) Stroke in Adults

The Sentinel Stroke National Audit Programme (SSNAP)

Interventions to Improve the Physical Function of ICU Survivors

Core Standards for Intensive Care Units

ICU Early Mobilization: From Recommendation to Implementation at Three Medical Centres

Revista Brasileira de Terapia Intensiva

Invasive pulmonary aspergillosis is a frequent complication of critically ill H1 N1 patients: a retrospective study

Isolation of Aspergillus in three 2009 H1 N1 influenza patients. Influenza Other Respir Viruses

Grant acknowledgement Supported by FUCAP and CIBERES

Effect of daily chlorhexidine bathing on hospitalacquired infection

Chlorhexidine bathing and health care-associated infections: a randomized clinical trial

Outcome of elderly patients undergoing intracranial meningioma resection -a systematic review and meta-analysis

References 1. An expanded definition of the adult respiratory distress syndrome

Acute Respiratory Distress Syndrome: The Berlin Definition JAMA

H1 N1 influenza A virus-associated acute lung injury: response to combination oseltamivir and prolonged corticosteroid treatment

Epidemiology, Co-Infections, and Outcomes of Viral Pneumonia in Adults: An Observational Cohort Study. Medicine (Baltimore)

Survival from severe pandemic H1 N1 in urban and rural Turkey: a case series

Prognostic importance of neutrophillymphocyte ratio in critically ill patients: short-and long-term outcomes. The American journal of emergency medicine

Reversal of neutrophil-to-lymphocyte count ratio in early versus late death from septic shock

The association between the neutrophil-to-lymphocyte ratio and mortality in critical illness: an observational cohort study

The mortality risk of over hydration in haemodialysis patients

Ultrasound assessment for extravascular lung water in patients undergoing hemodialysis. Time course for resolution

Sufficient sleep quality easily measured: a multicenter centre study in dutch ICUS

33rd International Symposium on Intensive Care and Emergency Medicine

Using Nursing Activities Score to Assess Nursing Workload on a Medium Care Unit

Nurse staffing levels and the incidence of mortality and morbidity in the adult intensive care unit: A literature review

Nursing Activities Score in the intensive care unit: Analysis of related factors. Intensive and Critical Care Nursing

Organisation and Management of Intensive Care : a prospective study in 12 European countries

Impact of performance obstacles on intensive care nurses' workload, perceived quality and safety of care and quality of working life

Burnout syndrome in critical care nursing staff

Work engagement: An emerging concept in occupational health psychology

Diagnostic Accuracy of clinical swallow assessment for oropharyngeal aspiration: a systematic review

The Cruelest Lies Are Often Told in Silence

Critical Review: Is the endoscopic swallowing assessment

18 (abstract A870). Lactate (Sepsis vs Trauma)

Jaundice in the Intensive Care Unit

Cooling techniques for targeted temperature management post-cardiac arrest

Thermoregulatory catheter-associated inferior vena cava thrombus

IVTM Intravascular Temperature Management Catheter Specifications

Prediction of postoperative pulmonary complications in a population-based surgical cohort

Lung ultrasound: routine practice for the next generation of internists

Relevance of Lung Ultrasound in the Diagnosis of Acute Respiratory Failure*

Assessment of Hemostasis after Plasma Exchange Using Rotational Thrombelastometry (ROTEM)

Prevention of pain on injection of propofol: systematic review and meta-analysis

Prevention of pain on injection with propofol: a quantitative systematic review

Mechanical ventilation guided by esophageal pressure in acute lung injury

Ultrasound assessment of antibiotic-induced pulmonary reaeration in ventilator-associated pneumonia

ARDS Definition Task Force

Endotracheal tubes cuff pressure control: does the CO2 matter? Minerva Anestesiol

Incidence and outcome of in-hospital cardiac arrest in the United Kingdom National Cardiac Arrest Audit

Delayed awakening after cardiac arrest: prevalence and risk factors in the Parisian registry

Diaphragm ultrasound as a new index of discontinuation from mechanical ventilation critical ultrasound journal

Ventilator-induced diaphragm dysfunction: time for (contr)action!

The course of diaphragm atrophy in ventilated patients assessed with ultrasound: a longitudinal cohort study Critical Care

Effect of Postextubation High-Flow Nasal Cannula vs Conventional Oxygen Therapyon

Early non-invasive ventilation treatment for severe influenza pneumonia

Use of a pathway quality improvement care bundle to reduce mortality after emergency laparotomy

The First Patient Report of the National Emergency Laparotomy Audit

Acute kidney injury enhances outcome prediction ability of sequential organ failure assessment score in critically ill patients

Médecine Intensive et Réanimation

Development and validation of a questionnaire for quantitative assessment of perceived discomforts in critically ill patients

The emotional and cognitive impact of unexpected simulated patient death

The efficiency of instructional conditions: an approach to combine mental effort and performance measures

Long-term cognitive impairment after critical illness

Satisfacción laboral de los profesionales sanitarios de un Hospital Universitario: análisis general y categorías laborales

The nurse satisfaction, service quality and nurse retention chain: Implications for management of recruitment and retention

Grant Acknowledgement AM: Moulton Foundation

Hospital/KCL

The Very Elderly Admitted to ICU: A Quality Finish? Crit Care Med

A global clinical measure of fitness and frailty in elderly people

Cognitive, functional, and quality-of-life outcomes of patients aged 80 and older who survived at least 1 year after planned or unplanned surgery or medical intensive care treatment

Outcome of elderly patients with circulatory failure. Int Care Med

Understanding and reducing disability in older adults following critical illness

Mechanical ventilation in a cohort of elderly patients admitted to an intensive care unit

Preferences of Current and Potential Patients and Family Members Regarding Implementation of Electronic Comunication Portalls in Intensive Care Units

Caring for the family of the critically ill patient

The carina as a radiological landmark for central venous catheter tip position

Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study

Duration of cardiopulmonary resuscitation and illness category impact survival and neurologic outcomes for in-hospital pediatric cardiac arrests

Duration of Prehospital Resuscitation Efforts After Out-of-Hospital Cardiac Arrest

Neurologic prognosis after cardiac arrest

Magnetic resonance imaging markers of Parkinson´s disease nigrostriatal signature

Disruption of posteromedial large-scale neural communication predicts recovery from coma

James McDonnell Foundation, the Belgian American Education Foundation, University Milano Bicocca. Fig. 67 (abstract A954). Study flowchart Intensive Care Medicine Experimental

Neurological prognostication after cardiac arrest

Strategies for Improving Survival After In-Hospital Cardiac Arrest in the United States: 2013 Consensus Recommendations: A Consensus Statement From the

Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity analysis

Early extracorporeal membrane oxygenator-assisted primary percutaneous coronary intervention improved 30-day clinical outcomes in patients with ST-segment elevation myocardial infarction complicated with profound cardiogenic shock

The current use of Impella 2.5 in acute myocardial infarction complicated by cardiogenic shock: results from the USpella Registry

Mechanical circulatory support in cardiogenic shock

Cardiac Transplantation Research Database Group, Long-term Outcomes of Cardiac Transplantation for Peripartum Cardiomyopathy: A Multiinstitutional Analysis

Cardiopulmonary Resuscitation with Assisted Extracorporeal Life Support versus Conventional Cardiopulmonary Resuscitation in Adults with In-hospital Cardiac Arrest: An Observational Study and Propensity Analysis

Patient safety, satisfaction, and quality of hospital care: cross sectional surveys of nurses and patients in 12 countries in Europe and the United States

Nurses' widespread job dissatisfaction, burnout, and frustration with health benefits signal problems for patient care

Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial

Feasibility and observed safety of interactive video games for physical rehabilitation in the intensive care unit: a case series

Is Admission to the Intensive Care Unit Associated With Chronic Opioid Use? A 4-Year Follow-Up of Intensive Care Unit Survivors

Survival to ICU discharge in ventilated patients Intensive Care Medicine Experimental

Modified early warning score with rapid lactate level in critically ill medical patients: the ViEWS-L score

Impact of matrix-assisted laser desorption ionization time-of-flight mass spectrometry on the clinical management of patients with Gram-negative bacteremia: a prospective observational study. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America

Impact of rapid organism identification via matrix-assisted laser desorption/ionization time-of-flight combined with antimicrobial stewardship team intervention in adult patients with bacteremia and candidemia. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America

Impact of rapid identification of Acinetobacter Baumannii via matrix-assisted laser desorption ionization time-of-flight mass spectrometry combined with antimicrobial stewardship in patients with pneumonia and/or bacteremia. Diagnostic microbiology and infectious disease

An international cross-sectional survey of antimicrobial stewardship programmes in hospitals

Continuous infusion of beta-lactam antibiotics in severe sepsis: a multicenter double-blind, randomized controlled trial

HPLC determination of plasma free and total tazobactam and piperacillin

Effectiveness of polymyxin Bimmobilized fiber column in sepsis: a systematic review Crit Care

Predicting functional impairment in brain tumor surgery: the Big Five and the Milan Complexity Scale *P. Ferroli

The "medial-oblique" approach to ultrasoundguided central venous cannulation-maximize the view, minimize the risk

Medial-oblique" Probe Position for Ultrasound-guided Internal Jugular Vein Cannulation: A Crossover Study

Improving survival from sudden cardiac arrest: the "chain of survival" concept

Practice parameter: prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology

Association of Bystander Interventions with neurologically intact survival among patients with Bystander-Witnessed Out of Hospital Cardiac Arrest in Japan

Performance of the revised Atlanta and determinantbased classifications for severity in acute pancreatitis

Rational Fluid Therapy for Sepsis and Septic Shock

What Do Recent Studies Tell Us? Review Article

Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock

Fluid type and the use of renal replacement therapy in sepsis: a systematic review and network meta-analysis. Int Care Med

The Third International Consensus Definitions for Sepsis and Septic Shock

Fernández-Ortega Regional University Hospital in Málaga

Rapid Diagnosis of Infection in the Critically Ill, a Multicenter Study of Molecular Detection in Bloodstream Infections, Pneumonia, and Sterile Site Infections

Grant acknowledgement Supported, in part

Cytokines and Signaling Molecules Predict Clinical Outcomes in Sepsis

The application of esophageal pressure measurement in patients with respiratory failure

Mechanical ventilation guided by esophageal pressure in acute lung injury

ECMO criteria for influenza A (H1N1)-associated ARDS: role of transpulmonary pressure

Mechanical ventilation guided by esophageal pressure in acute lung injury

Fig. 92 (abstract A1036)

Grant acknowledgment National Cheng-Kung University Hospital Grant A1038 Assessment effort and work of breathing by airway occlusion pressure versus esophageal pressure

Hospital Universitario Reina Sofia, Intensive Care Unit

Airway occlusion pressure

Effects of the prone position on respiratory mechanics and gas exchange during acute lung injury

Effects of prone position on alveolar recruitment and oxygenation in acute lung injury

Prone position reduces lung stress and strain in severe acute respiratory distress syndrome

Lateral positioning of ventilated intensive care patients: A study of oxygenation, respiratory mechanics, hemodynamics, and adverse events

The Effect of Lateral Position on Oxygenation in ARDS Patients : A Pilot Study

The open lung concept of mechanical ventilation: the role of recruitment and stabilization

Selecting the 'right' positive endexpiratory pressure level

Esophageal and transpulmonary pressures in acute respiratory failure

Spontaneous breathing during lung-protective ventilation in an experimental acute lung injury model: high transpulmonary pressure associated with strong spontaneous breathing effort may worsen lung injury

Relationship between activation of the sympathetic nervous system and renal blood flow autoregulation in cirrhosis

Hepatorenal Acute Kidney Injury and the Importance of Raising Mean Arterial Pressure

A1051 Fig. 107 (abstract A1049)

Rotational thrombolelastometry produces potentially clinical useful results within 10 min in bleeding emergency department patients: the DEUCE study

Thrombelastography and rotational thromboelastometry early amplitudes in 182 trauma patients with clinical suspicion of severe injury

A prospective study of anaemia status, haemoglobin concentration and mortality in an elderly cohort

A1061 Experience with a hospital-wide implementation of a massive transfusion protocol: before and after

A1061 References 1. Experience with a massive transfusion protocol in the management of massive haemorrhage

Rodríguez Villamizar Hospital Universitario Puerta de Hierro Majadahonda

Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College of Critical Care Medicine

End-of-life care practices in patients dead as a result of a devastating brain injury and organ donation in Spain

Prediction of potential for organ donation after cardiac death in patients in neurocritical state: a prospective observational study

Causes of family refusal for organ donation

Development of the Croatian model of organ donation and transplantation

Evaluation of organ procurement in an area under the influence of a training program

Grant acknowledgment None. A1070 Non-heart beating donor program: Seville's experience L. Martin-Villen 1,2

The variables analyzed were: total number of queries activation, of PD, of eligible donors (ED) and of real donors (RD). RD attendance times were registered and we defined out-of-hospital time (from Cardiac arrest to hospital arrival), inof-hospital time (from hospital arrival to cannulation onset), cannulation time (beginning of cannulation to perfusion onset) and perfusion (from perfusion onset to the first organ removal). We registered number and type of valid organs and tissues, number of family members or judicial negative, number of non-real-donors (NRD) and its causes

Regarding attendance times, the median time was 71 (ICR 60-75) minutes for out-of-hospital, 29 (ICR 22-33) minutes for in-of-hospital, 30 (24-35) minutes for cannulation and 135 (107-169)minutes for perfusion

Controlled donation after circulatory determination of death in Spain

Rodriguez Villamizar Puerta de Hierro Hospital, Intensive Care Unit, Majadahonda, Spain Correspondence: J. Veganzones Ramos -Puerta de Hierro Hospital

The use of lung donors older than 55 years: a review of the United Network of Organ Sharing database

Marginal donor lungs: a reassessment

Liberalization of donor criteria may expand the donor pool without adverse consequence in lung transplantation

The Human Organ Transplant Act. Legislative Acts and Guidelines, Ministry of Health

Amendment of the Human Organ Transplant Act

Lebrón-Gallardo Regional University Hospital in Málaga

Grant acknowledgment This work was funded by the 2015 Innovation Awards from the Department of Medicine

Recombinant human soluble thrombomodulin in sepsis-induced disseminated intravascular coagulation: a multicenter propensity score analysis

A1082 Successeful strategy to reduce ventilator

Getting Started Kit: Prevent Ventilator Associated Pneumonia. Cambridge MA: Institute for Healthcare Improvement

UMR_S 1158 and Hôpital Pitié-Salpêtrière, Respiratory Division and Medical ICU

Trends in Tracheostomy for Mechanically Ventilated Patients in the United States

Outcome of patients with cirrhosis requiring mechanical ventilation in ICU

The impact of organ dysfunction in cirrhosis: survival at a cost?

Does intermediate care improve patient outcomes or reduce costs

Abbreviations: TBS, tracheobronchial secretions

AKI, acute kidney injury

HD, hemodialysis A1091 Patient perceptions of physiotherapy in ICU: a qualitative study M

Dimensionamento da equipe de enfermagem da UTI-adulto de um hospital ensino -Rev. Eletr. Enf

Critérios para admissão de pacientes na unidade de terapia Intensiva e Mortalidade

Patient Acuity Rating: Quantifying Clinical Judgment Regarding Inpatient Stability

Identification of deteriorating patients on general wards; measurement of vital parameters and potential effectiveness of the Modified Early Warning Score

The epidemiology of chronic critical illness in the United States

Impact of chronic critical illness on the psychological outcomes of family members

Tracheostomy procedures in the intensive care unit: an international survey

Percutaneous tracheostomy: A 6 yr prospective evaluation of the single tapered dilator technique

Can outcomes of intensive care unit patients undergoing tracheostomy be predicted? Respir Care

Safety and complications of percutaneous tracheostomy in a cohort of 800 mixed ICU patients

Prognostic consequences of borderline dysnatremia: Pay attention to minimal serum sodium change. Crit Care

Are the dysnatremias a permanent threat to the critically ill patient?

A1100 Explore the dengue-related risk factors and death factors in

dengue hemorrhagic fever epidemic in Taiwan

Diabeted with Hypertension as Riak Factor for Adult Dengue Hemorrhagic Fever in a Predominantly Dengue Serotype 2 Epidemic: A Case Study

Characteristics of dengue epidemics in Taiwan

Modelling Risk of Cardio-Respiratory Instability as a Heterogeneous Process

Grant acknolwedgment NIH NINR R01NR013912

During two years, we included all patients with more than 20 days of stay in a medicalsurgical ICU. Previous informed consent , we collected demographics data, baseline functional status (Barthel scale), mortality intraUCI, at hospital and one-year of hospital discharge

Chronic critical illness

A1103 Comparison between 2-dimentional speckle tracking & color-tissue doppler imaging for the assessment of left ventricular global longitudinal systolic strain and strain rate in outcome prediction of sepsis

Grant acknolwedgment Italian Ministry of Health, Italy (Convenzione n.46/GR-2011-02348099); and Fondazione Sestini

Exclusion criteria: do not resuscitate order. Protocol: data were collected anonymously according to the Utstein style. Follow-up: 6 months long using registry office and telephonic interview. Data: age, sex, cerebral performance category (CPC 1 good 2 moderate disability, 3 severe disability, 4 unconscious), site of cardiac arrest, presumed etiology, initial rhythm (shockable or unshockable), witnessed event, monitored, CPR started within 1 minute. Primary end points: return of spontaneous circulation (ROSC), survival to hospital discharge and CPC 1-2. Secondary end points: 6 months survival and CPC 1-2. Statistics: numerical data are expressed as mean ± standard deviation or median (interquartile range), as percentage if ordinal data. Chi-square test for ordinal data and T Student's test for numerical data were performed. P significant if < 0.05. Results: 440 cardiac arrests, CPR was carried out in 389 cases (88 %)

Secondary end points: 12.3 % alive at 6 months; 95.5 % of them with CPC 1-2. Conclusions Our experience reflects some aspects common with other European countries: less monitored events as well as more frequent cardiac arrests in unmonitored wards 1 . RRT allowed a reduction of cardiac arrests thus reducing their incidence without modifying mortality

Incidence and outcome of in-hospital cardiac arrest in the United Kindom National Cardiac Arrest Audit

Epidemiology and outcomes of in-hospital cardiopulmonary resuscitation in the United States

Effectiveness of rapid response teams on rates of inhospital cardiopulmonary arrest and mortality: a systematic review and meta-analysis

National Intensive Care Surveillance, Quality Secretariat Building, Castle Street Hospital for Women

Correspondence: A. Beane -Network for Improving Critical Care Systems and Training (NICST), Colombo, Sri Lanka Intensive Care Medicine Experimental

Fig. 125 (abstract A1115)

Fig. 126 (abstract A1115)

Authors thank Tem International for providing a ROTEM analyzer for the study. References 1. The CONSORT Extension for Cluster Trials

MCF FIBTEM & Plasma Fibrinogen References 1. Extracorporeal Membrane Oxygenation for ARDS in Adults

Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with inhospital cardiac arrest: an observational study and propensity analysis

Critical care in resource-restricted settings

on behalf of the Task Force for Mass Critical Care. Evacuation of the ICU. Care of the Critically Ill and Injured During Pandemics and Disasters: CHEST Consensus Statement

ICU fire evacuation preparedness in London: a cross-sectional study

Fire on an intensive care unit caused by an oxygen cylinder

National questionaire survey on what influences doctors; decisions about admission to intensive care

The Faculty of Intensive Care Medicine/The Intensive Care Society

Acute care toolkit 2. High-quality acute care. Royal College of Physicians

United Kingdom; 2 Golden Jubilee National Hospital, Department of Anaesthesia

there is a need for general (non-cardiac) Intensive Care Units (ICUs) to facilitate more elective and emergency surgery for these patients. Inhaled nitric oxide, a selective pulmonary vasodilator, may be required for this purpose. Objectives: We wished to determine the availability of inhaled nitric oxide in general Scottish ICUs

We excluded tertiary paediatric, cardiothoracic and neuro-critical care units. An online survey was distributed followed, if necessary, by a telephone survey. Caldicott Guardianship approval was not required. Results: Four (20 %) general ICUs had nitric oxide immediately available

Adult congenital heart disease (grown-up congenital heart disease)

Audit of Critical Care in Scotland

Scottish Adult Congenital Cardiac Service

A1128 References 1. Triage of Patients Consulted for ICU Admission During Times of ICU-Bed Shortage

A1129 then performed a cross-sectional survey by visiting each facility, and determining characteristics for each facility

Critical care in low-income countries

Functional compromise reflected by sarcopenia, frailty, and nutritional depletion predicts adverse postoperative outcome after colorectal cancer surgery

Sarcopenia is a predictor of outcomes in very elderly patients undergoing emergency surgery

Oudemans-van Straaten HM, Beishuizen A. Low skeletal muscle area is a risk factor for mortality in mechanically ventilated critically ill patients

Acetate-versus lactate-based balanced colloids used as priming solutions for cardiopulmonary bypass: an experimental pilot study H. Cauwenberghs 1 , A. De Backer 1 , H. Neels 2 , I. Deblier 3

Correspondence: H. Cauwenberghs -ZNA Middelheim General Hospital

Methods: Following IRB approval, 20 male non-diabetics gave consent and were randomly assigned to receive either succinylatedgelatin 30 g (Geloplasma®) in 150 mEq Na+, 5 mEq K+, 100 mEq Cl-, 3 mEq Mg++, 30 mEq lactate (SID 58) or succinylated gelatin 30 g (Isogelo®) 151 mEq Na+, 4 mEq K+, 2 mEq Ca, 103 mEq Cl-, 2 mEq Mg++, 30 mEq acetate 05 and Bonferroni corrected. Results: Demographics were comparable. Acid-base variables changed similarly throughout without significant differences between groups (SIG shown in Figure 133). By contrast, glucose levels rose very significantly in the lactate group and persisted post CPB (Figure 134). Oncotic pressure, diuresis, osmolarity and oxygen uptake did not differ between groups. Discussion. Concerning acid-base variables and secondary endpoints

The perioperative period

Stewart's Textbook of AcidBase

PL Effects of intravenous solutions on acid-base equilibrium: from crystalloids to colloids and blood components

Seric pro-adrenomedullin levels in low cardiac output syndrome (LCOS) after cardiac surgery

We measured AM at 4 time-points (T1-T4): before surgery; at admission; 6 h and 18 h after surgery. Continuous data were showed as average (SD) and categorical ones in percents. Comparisons were performed with Kruskall-Wallis and ANOVA tests. The ROC approach was used to assess the predictor capacity of AM. All analyses were performed with STATA12. The Ethical Committee approved the study. Results: 68 patients were included. The average of age was 65 ± 13.9 years, and were women 62.3 %. The median (IQR) for Euroscore was 5 (2-7). Comorbidities were hypertension (59 %), diabetes mellitus (26 %) and atrial fibrillation (18.8 %). On-pump surgery was performed in the 88 % and the coronary bypass was the most frequent (31 %). The incidence of Low Cardiac Output Syndrome was 23 %. AM levels (mmols/l) were: 0.82 ± 0.46 (before surgery)

Postoperative Pro-Adrenomedullin Levels Predict Mortality in Thoracic Surgery Patients

Consensus on circulatory shock and hemodynamic monitoring. Task force of the European Society of Intensive Care Medicine

Canadian Cardiovascular Society position statement on the management of thoracic aortic disease

guidelines for the diagnosis and management of patients with Thoracic Aortic Disease

Early and midterm outcomes following surgery for acute type A aortic dissection

Importance of blood pressure control after repair of acute type a aortic dissection

135 (abstract A1134). Pro-ADM levels T1-T4

Bleeding complications associated with cardiopulmonary bypass

Prospective validation of the International Society of Thrombosis and Haemostasis scoring system for disseminated intravascular coagulation

Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation

Hospital Regional, Intensive Care, Malaga, Spain; 3 Hospital Serrania, Ronda, Spain; 4 Hospital Regional, Intensive Care, Málaga, Spain; 5 Hospital Infanta Margarita, Intensive Care, Cabra, Spain; 6 Hospital Virgen del Rocio, Intensive Care

We analyze differences in the postoperative incidence of AF in both groups. Results: Cohort of 7276 patients, Mean age 63.91 ± 12.45 years. 85.9 % was elective surgery. EuroSCORE 5.86 ± 3.14 points. ICU mortality was 7.6 %.Prior to surgery, 51.5 % and 25.8 % taking statins present a history of AF. 21.9 % postoperative AF episode presented. The patients treated before surgery with statins had AF 19.8 % vs 22.5 % (P =0.006)

The sample was divided among the patients who died and those who do not. Demographic variables, prognostic scales,type of surgery,early complications, ICU stay and mortality were compared.The variables that reached statistical significance in the univariate analysis were analyzed in multivariate logistic regression. Data expressed as mean and standard deviation, percentage, mean difference, odds ratio and corresponding confidence intervals. Statistical significance level of P < 0.05. Results: A total of 1656 patients were analyzed, ≥ 75 years 527(31.8 %) of which 32 died(6.1 %)

Predictors of postoperative complications in octogenarians undergoing cardiac surgery

A1140 Ventricular assist devices, transfusion and health-related quality of life

4 %,median RBC transfused 21

Ventricular assist devices and increased blood product utilization for cardiac transplantation. Stone ML et all

Bleeding complications and blood product utilization with left ventricular assist device implantation. Schaffer JM et all

after completion of cardiac surgery. The patients were divided into two groups: aged < 80 and age ≥80 years. Association analysis of demographic, clinical, therapeutic factors and complications during ICU stay. Univariate analysis using Chi square (Fisher if applicable) and T Student. Data expressed as percentages, means, estándar desviation (SD), mean differences (dm), odds ratio (OR), and confidence intervals 95 % (CI95%) Results: A total of 1656 significant differences in cardiac arrest

Cardiac surgery in 260 octogenarians: a case series

Outcomes and cost of cardiac surgery in octogenarians is related to type of operation: a multiinstitutional analysis

Aortic valve replacement with and without coronary artery bypass graft surgery in octogenarians: is it safe and feasible?

Short-and long-term outcomes in octogenarians after coronary artery bypass surgery

Diagnosing and treating the failing right heart

Matching dicrotic notch and mean pulmonary artery pressures: implications for effective arterial elastance. The American journal of physiology

University Hospital of Pisa, Department of Anaesthesia and Critical Care Medicine, Cardiothoracic and Vascular Anaesthesia

Venous-to-arterial CO2 differences and the quest for bedside point-of-care monitoring to assess the microcirculation during shock

Model of pCO2 gap during hypothermic cardiopulmonary bypass

Central venous O(2) saturation and venous-to-arterial CO(2) difference as complementary tools for goal-directed therapy during highrisk surgery

patients: closed-vs. opensystem

The closed-system PIVC was composed of catheter with inteof PIVCs, the incidence of bacterial colonization and PIVCs-related complications (phlebitis, extravasation, catheter occlusion and hematoma) were recorded. The protocol of the study was approved by the IRB of Tokushima University Hospital. Results: Ninety-one closed-system PIVCs and 89 open-system PIVCs were evaluated. The median indwell time did not differ between the closed-and open-system PIVCs (median

were identified from the database of the microbiological laboratory of the hospital and were included in this study as cases. Demographic data, severity of illness, risk factors for colistin-resistance (described in previous studies), clinical management and hospital outcome of all cases were recorded. MDR -Resistant to at least one agent from 3 different classes. PDR -Resistant to all classes. Results: 13 of 207 KP (6.2 %) and 8 of 445 EC (1.8 %) isolates were colistin resistant. 14 of 207 (6.76 %) of KP and 8 of 445 EC (1.8 %) isolates were colistin resistant

Colistin-resistant isolates of Klebsiella pneumoniae emerging in intensive care unit patients: first report of a multiclonal cluster

Risk factors associated with the isolation of colistin-resistant gram-negative bacteria: a matched case-control study

Combination antibiotic treatment versus monotherapy for multidrug-resistant, extensively drug-resistant,and pandrug-resistant Acinetobacter infections:a systematic review

In vitro synergistic activity of tigecycline and colistin against XDR-Acinetobacter baumannii

Woodlands Multispeciality Hospital

28 References 1. Combatting resistance in intensive care: the multimodal approach of the Spanish ICU "Zero Resistance" program. Garnacho Montero et als

during a KPC outbreak in the 10-bed ICU of a tertiary university hospital in Rio de Janeiro, Brazil. All patients admitted to the ICU were included in the study and classified as case (KPC yielded from any biological material, either considered as colonization or infection) or control (all other patients who did not have KPC isolation). Both groups were compared according to demographic data, comorbidities, sepsis diagnosis, type and time of life support, SOFA and SAPS III scores at ICU admission, length of stay (LOS) at ICU and hospital, and hospital costs, ICU and hospital mortality. Results: 63 patients were admitted during the studied period. 24 patients had KPC samples isolated from different biological material

Conclusions: During a KPC outbreak in the ICU of an academic tertiary hospital in Rio de Janeiro, the isolation of KPC associated with colonization or infection was associated with greater ICU and hospital LOS, more requirements of life-organ support, higher ICU and hospital mortality rates

Impact of carbapenem resistance on the outcome of patients' hospital-acquired bacteraemia caused by Klebsiella pneumoniae

First report of KPC-2-producing Klebsiella pneumoniae strains in Brazil

Attributable mortality rate for carbapenem-resistant Klebsiella pneumoniae bacteremia

Zero Resistance" (RZ) program was swabs (nasal, pharyngeal, axillary and rectal) were routinely performed to all patients admitted, besides diagnostic cultures when needed. Furthermore, we analysed other pathological variables and comorbidities. The difference between groups of MRB was made by Chi-square test for qualitative variables and the Kruskal-Walls test for the continuous ones. Statistical significance was set at P < .05. Results: 1651 admitted. In 136 patients were identified one or more MRB (148 in total). 60 patients (40,5 %) were ESBLs carriers, 47 (31,7 %) MRSA, 23 (15,5 %) P aeruginosa, 13 (8,8 %) Acinetobacter spp and 6 (4 %) others MRB carriers. In 36 cases (27,2 %) the presence of a MRB caused infection. Nasal swabs detected 31 % of MRB carriers (64 % of all MRSA), pharyngeal swabs 32 % (44,7 % of MRSA), axillary swabs 13 % (19 % of MRSA, 23 % of Acinetobacter), and rectal swabs 36 % (82 % of ESBLs, 46,2 % of Acinetobacter). In 4 cases (13 %) just the axillary swab was positive, and in 35 cases (36 %) the rectal was the only swab able to detect a MRB. Diagnostic cultures (blood, urine, bronchoaspirate, surgical wound and others) detected MRB in less than 30 %. The checklist did not detected neither colonization nor infection by MRB in 49 (36 %) patients (45 % MRSA, 61,5 % Acinetobacter, 30 % of ESBLs). All patients with P aeruginosa had RF, but one. There was no statistical significance between groups of MRB and other comorbidities. Conclusions: The surface cultures realized at admission detected 70 % of MRB not detected by diagnostic cultures

):e47297. A1158 Enteral paramomycin to eradicate colistin and carbepemenase resistant microorganisms in rectal colonization to prevent icu multiresistant nosocomial infections

University Hospital of Gran Canaria Dr. Negrín, Pharmacy Department

Hospital of Getafe, Intensive Care Unit

Microbiology Unit, Section Biology Pathology and Health Products

) received catecholamines and 75 (16 %) were immunocompromised. ICU mortality was 16.4 % and did not differ between ESBL carriers and non-carriers. The rate of ESBL colonization at admission and ESBL acquisition were 13.2 % and 4.3 %, respectively. Escherichia coli was the most frequently observed bacteria. The results of the univariate analysis for ESBL acquisition are presented in Table 93. In multivariate analysis, IGS-II and ICU length of stay were strongly associated with ESBL acquisition (Table 94). Discussion and conclusion: The observed rate of ESBL carriage on admission was comparable to other rates in French ICUs (15 %). Despite the unfavourable twin-bed architecture of our ICU, the incidence of ESBL acquisition was 4.3 % which was actually lower than transmission rates previously published in other ICUs. ESBL acquisition was strongly associated with ICU length of stay and severity score at admission. This study is fully consistent with previous ones challenging the geographic isolation in a non-epidemic setting and suggests that environmental contamination may not play a substantial role in the transmission of ESBL-PE

WIPO was diagnosed with LUS if, on at least one upper or lower part of both sides, the LUS profile moved from A (normal) to B (interstitial oedema) or from B to "B+", where B+ consisted in at least a doubling of the B lines number. The reference diagnosis of WIPO was established on other criteria by experts blind for LUS. Results: WIPO occurred in 15 (24 %) SBT. Among cases with WIPO, the LUS profile did not change during SBT in one case, changed for LUS signs of pulmonary oedema in 13 cases (true positives) and changed but without typical LUS signs of WIPO in one case

University Hospital of Lausanne, Intensive Care and Burn Unit

Saint Michael's Hospital and Keenan Research Centre, Interdepartmental Division of Critical Care

A1163 Introduction: Many different tools are found to predict weaning success

US) assessed excursions of the right hemi-diaphragm could be a useful measurement for prediction success in weaning from mechanical ventilation (MV

US was performed after patient met weaning criteria (according to local protocol) and it was decided to discontinue MV. Patients with neuromuscular disorders and diaphragmatic paralysis were excluded. Measurements were performed once on Pressure Support Ventilation (PS ≤ 10cmH 2 O, PEEP ≤ 5cmH 2 O). The right hemi-diaphragms of patients were evaluated by M-mode ultrasonography (Esaote MyLabGamma AC2541 1-8 MHz convex probe). The average diaphragm excursions value (DE avg ) was estimated from 3 sequential measurements. The rapid shallow breathing index (SBI), dynamic compliance (C dyn ), minute ventilation (M V ) and spontaneous tidal volume (V T spont) were obtained from the ventilator (Servo i , Maquet)

Table 99) with best DE avg cut-off value 10.5 mm (sensitivity 89 % , specificity 99 %). Conclusions: Our findings suggest that right hemi-diaphragm excursions assessed with M-mode ultrasonography is more accurate predictor of weaning success than other common weaning criteria

Weaning from mechanical ventilation

Diaphragm dysfunction assessed by ultrasonography: influence of weaning from mechanical ventilation

A1164 Cycling-off guided by real-time waveforms analysis (IntelliSync+): pilot study on next-generation PSV

Anesthesia and Intensive Care

Methods: In 6 patients under PSV, IntelliSync + was compared to to default setting, ETS opti1 decreased cycling delay and unassisted efforts at PS basal, but these favorable effects were not maintained at PS + 50. Further optimization (ETS opti2) decreased cycling and trigger delay but did not affect unassisted efforts. When IntelliSync + was activated, cycling delay was shorter and values of trigger delay and unassisted efforts were at least as low as with optimized settings of ETS. Table 100 summarizes the results obtained in the 6 conditions tested. Conclusions: Bedside optimization of ETS guided by waveforms on the ventilator screen improved PVI. Increase of pressure support level worsened PVI and mandated re-optimization of ETS

A1165 Characteristics and factors associated with prolonged weaning. A sub-analysis of the wind study G

UPRES EA3830 IRIB

Saint Michael's Hospital and Keenan Research Centre, Interdepartmental Division of Critical Care

Objectives: To determine a predictor of weaning success with a faster reaction time than respiratory rate & pulse rate. Methods: 15 patients (9 male, 6 female) on MV >21 days were included in our study diagnosed with sepsis (n = 2), pneumonia (n = 2), pancreonecrosis (n = 6), obesity hypoventilation syndrome (n = 2), intestinal obstruction (n = 3). Oxygen consumption (VO2) monitoring in different stages of MV support reduction was recorded using E-COVX

Evaluation of mortality over time in patients receiving mechanical ventilation

The use of Mechanical ventilation in intensive care unit in Russia: National epidemiological survey RUVENT-2012

A prospective, blinded evaluation of indexes proposed to predict weaning from mechanical ventilation

A randomized, controlled trial of the role of weaning predictors in clinical decision making

CPF) using ventilator built-in flow-meter to predict extubation success: a single centre study F. Gobert 1,2 , H

Inclusion criteria were: age > 18 years, intubation > 24 h, no withdrawal decision of life supporting care, eligible for scheduled weaning trial and then scheduled extubation, mechanical ventilation from Evita XL ventilator (Dräger, Germany) and patient's agreement to participate. Once daily checked criteria for weanibility were present, patients were switched to a standardised pressure support ventilation (inspiratory pressure = 7cmH2O, PEEP = 4 cmH2O, FIO2 = 0.40) for 1 h (if no chronic respiratory failure-CRF), 2 h (if CRF) or 12 h (if neuromuscular CRF). The procedure of CPF measurement was explained to the patient, who was encouraged to cough as strong as possible just before extubation. CPF measurements were done by freezing ventilator screen and scrolling the cursor to the maximal value of CPF during expiration and Tidal Volume (TV) in preceding inspiration. Three measurements were averaged. Early extubation success rate was defined as the proportion of patients who were alive and not reintubated 48 h after scheduled extubation. Median values were compared by using non parametric tests. Diagnostic performance of CPF and TV was assessed by using area under curve (AUC) of the ROC method. After having defined cut-off values for CPF and TV, we described the performance of a test combining CPF and TV values to predict the early extubation outcome. Results: During the study period, 673 patients were admitted to our ICU of who 319 were intubated and 92 patients included (Fig 146). Between the 81 patients who succeeded and the 11 patients who failed extubation, median CPF was −67.7 L/min and −57.3 L/min, respectively (p = 0.03, Fig 147A), median TV 0.646 L and 0.448 L, respectively (p = 0,078, Fig 147B), and AUC averaged 0.61 and 0.64, respectively (Fig 148A). Bi-dimensional analysis showed a synergistic effect of CPF and TV to predict early extubation success (Fig 148B). The combination of thresholds (CPF < −60 L/min and TV > 0.600 L)

Grant acknolwedgment The study has been supported by the Sigrid Juselius Foundation, Päivikki and Sakari Sohlberg Foundation

The Third International Consensus Definitions for Sepsis and Septic Shock

Acute renal failure in critically ill patients: a multinational, multicenter study

Relation between mean arterial pressure and renal function in the early phase of shock: a prospective, explorative cohort study

Grant acknolwedgment This study is supported by the National Key Technology R&D Program of China

Conclusions: In patients with chronic renal impairment who undergo cardiovascular surgery requiring CPB, a lower level of preoperative sHp is independently associated with higher risk of pAKI. A1177 Urinary angiotensinogen as a possible predictor of acute kidney injury in severe sepsis S

We aimed to compare the incidence of AKI according to RIFLE (Risk, Injury, Failure, Loss of kidney and End-stage kidney disease

United Kingdom Intensive Care Medicine Experimental

):A1180 Introduction: Cytokine elimination during continuous hemofiltration (CHF) depends largely on the character of the filter membrane

A1181 References (1) (Kidney Disease Outcomes Quality Initiative. KDIGO Clinical Practice Guidelines for Acute Kidney Injury

Antithrombin supplementation for anticoagulation during continuous hemofiltration in critically ill patients with septic shock: a case-control study

Acquired deficit of antithrombin and role of supplementation in septic patients during continuous veno-venous hemofiltration

The influence of venovenous renal replacement therapy on measurements by the transpulmonary thermodilution technique

Effect of the venous catheter site on transpulmonary thermodilution measurement variables

The influence of haemodialysis on haemodynamic measurements using transpulmonary thermodilution in patients with septic shock: an observational study

When Drugs Disappear from the Patient: eliminaiton of intravenous medication by hemodiafiltration

Cardiac arrest in intensive care unit: Case report and future recommendations

Omni® (B. Braun, Melsungen, Germany), a new third generregional citrate anticoagulation. We collected patients' characteristics, filter life time, circuit pressures, interruption of therapy duration and reasons (alarm types), achieved and targeted renal dose, metabolic parameters (serum creatinine and potassium levels and arterial base excess). In addition, we adminof total therapy time) in CVVHD-citrate mode. Mean achieved renal dose was 26.3 ml/kg/hr corresponding to 96 % of the targeted dose in CVVH-heparin mode and 29.8 ml/kg/hr corresponding to 98 % of the targeted dose in CVVHD-citrate mode. In both RRT modes, excellent metabolic control and adequate fluid balance were achieved. Overall, the interface, design and ease of use were evaluated by users as excellent. Conclusions: CRRT in both CVVH and CVVHD modes could be provided using Omni® in a safe and efficient way in ten critically ill patients. Users provided positive feedback regarding therapy setup, management and user interface. A1184 Intermittent haemofiltration outside itu led by the intensive care team. Experience at a tertiary cardiothoracic centre S

Following modification of the Aquarius haemofilter (Nikkiso), we designed and implemented a protocol for RCA with stand alone citrate administration pre filter (ACD-A (Acid Citrate Dextrose Formula-A) containing 113 mmol/L of citrate) and post dilution CVVHF using calcium containing replacement fluid (Accusol 35 containing 1.75 mmol/L Ca) and, when needed, supplementary calcium depending on systemic iCa. The protocol can deliver 25 or 35 mLs/kg/hr of CRRT. We compare the efficacy of this new protocol, which we have initially implemented in patients with a relative contraindication to heparin, to a historical cohort of patients who received CRRT with prostacyclin and or pre-dilution CVVHF but would not have been contraindicated for RCA. We also present relevant biochemical data. Methods: A prospective audit of the first 30 adult critically ill patients receiving RCA with post dilution CVVHF. Crude comparison was made with a historical group of consecutive critically ill patients who received CRRT without heparin prior to the introduction of the RCA protocol. Patients were excluded from the RCA protocol and the comparison if they had, severe acute liver injury. Data is presented as median (range) with non parametric analysis and filter survival as a Kaplan Meier for the event´filter clotting´and censored for medical cessation or technical failure. Results: There were 113 filters used in 30 patients who received RCA and 73 in 24 in the comparison group respectively. One patient (14 filters) from the RCA group was excluded from the filter survival analysis due to a triglyceride level of 18.9 mmol/L, causing repeated filter failure

Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study

Conclusions: There is no relationship between COPD patients, BMI, age and extubation failure. This new formula combine 2 parameters, DTF*RSBI, as a good parameter for extubation.Methods: We used the administrative claims data of all Diagnosis Procedure Combination (DPC) hospitals in Japan from April 1, 2011 to March 31, 2012, and retrospectively reviewed the number and outcome at discharge of patients who were on ECMO support. Results: We identified 3,336 patients who received ECMO support during the 1-year study period. The average age was 58.7, and only 6.7 % of the patients were under 20 years old. The most common diagnosis was acute coronary syndrome, followed by cardiac arrest and pulmonary embolism. The overall survival rate at discharge was 39.6 % [95%CI 38.0-42.3]. Among the 1,449 acute care DPC hospitals, ECMO support was provided in 394 hospitals (27.2 %), and therefore the annual ECMO patient volume per hospital was 8.5, which is much lower than international standards for ECMO centers. Adjusted odds ratio for discharge alive stratified by annual ECMO volume per hospital were 1.10 [95%CI 0.78-1.54] and 1.23 [95%CI 0.90-1.28] for medium (131 hospitals treating 4 to 9 patients) and high volume centers (122 hospitals treating 10 or more patients), respectively, compared with low volume centers (141 hospitals treating 3 or less patients). Conclusions: ECMO support was administered to many adult cardiac patients, and provided in a substantial proportion of acute care hospitals in Japan. No significant patient volume effect for survival discharge was found.

Funding from the American Nurses Association Impact Grant is gratefully acknowledged. history and ICU related data were analyzed. Basal fracture risk before CI was calculated using the FRAX tool (https://www.shef.ac.uk/FRAX). In January 2016, referent family doctors were contacted by phone to check out new BF occurred during the 2 years after ICU discharge. Data are expressed as median (min-max) or percentages. Unpaired data were compared using Mann-Whitney test (p < 0.05 = significant). Results: From the 1446 patients admitted in 2013, 884 had an ICU LOS ≤ 7d, 3 were < 18y, 278 died in ICU or died after an ICU LOS ≤ 7d and 31 were lost to follow-up. We analyzed 178 patients who were alive in January 2016 and 72 patients who died outside ICU during the FUP after an ICU LOS >7d. Regarding alive patients (64 % males), admission was mainly related to cardiovascular, respiratory and neurological failure, or trauma. Age was 64 (18-91)y, simplified acute physiologic score (SAPS II) was 44.5 (6-85), ICU LOS was 15 (8-106)d. According to the FRAX tool, the 10-y probability of major osteoporotic BF (major FRAX risk) was 5.4 (1-32)%. Nine patients (4 men) developed BF in 7.5 (1-28) months after CI, equivalent to a 5 % risk of new BF 2y after CI. A context of fall at home was noted in every case. Age, ICU LOS and SAPS II of these patients were not statistically different from non-fractured patients. Their major FRAX risk was 16 (3-29)%, significantly higher than non-fractured patients (p = 0.0029). Finally, among the 72 dead patients, only one 62y man experienced hip fracture at the 17 th month after CI. Conclusions: Present incidence of new BF in the 2y following severe CI with a prolonged ICU stay is similar to previously published data [3]. Patients who experienced new BF after CI had a higher FRAX risk than the non-fractured patients. Influence of CI or ICU stay on BF risk is thus questioned. However, to be relevant, our results need to be compared to a control population: this work is ongoing. Conclusions: We detected a rise in Pro-Adrenomedullin levels after cardiac surgery. The results suggest that AM could be useful for LCOS prediction. More data are necessary to confirm the role in the prediction of relevant outcomes. Conclusions: The ICU stay and early complications evaluated not differ between the two groups , except for acute renal failure and higher mortality, despite the use of shorter times in cardiac surgical in octogenarians. There is a progressive decrease in coronary artery bypass surgery in recent years in this group probably in favor of percutaneous techniques.High-risk patients who require intensive perioperative management,should be identified to reduce the incidence of postoperative complications.Conclusions: Previous renal replacement treatment and the colonized patients' long stay at ICU increase the transformation of the VRE colonization into VRE infection. Strategies to reduce the duration of ICU stay of VRE-colonized patients are the main objects to controlling VRE infection rate.

Identification of SIRS/sepsis signs by ward nurses reduces 30-days mortality in patients with sepsis M. Torsvik 1 , L.T. Gustad 2,3 , I.L. Bangstad 2 , L.J. Vinje 2 , J.K. Damås 4,5,6 , E. Solligård 6, 7, 8 , A. Mehl 2, 6,9 1 Nord University, Faculty of Health Science, Levanger, Norway; 2 Nord-Trøndelag Hospital Trust, Levanger Hospital, Internal Medicine, Levanger, number of patients had ulceration at the site of the suction port but did not suffer any complication as a result.

Drainage of pleural effusion with small bore tube in mechanically ventilated patients S.J. Lee 1 , Y.S. Cha 2 , W.-Y. Lee 1 correlate with decannulation failure but future studies are necessary in this field. Objectives: To weight all sharps containers in a Cardiac Intensive Care over a one-week period. To then review the financial implications of the cost of this waste. To review ways to redistribute this waste.Methods: All closed sharps containers were weighed. Research the cost implications of different waste types from the Waste Management team within the Trust. Look into ways of reducing this waste. Results: A total of 62 items where weighed, totaling 123.82 kg, costing £65.26. This can be broken down in six 55 L Chest Drain boxes, weighing 53.98 kg, costing £28.45. Four 22 L sharps bin, weighing 23.04 kg, costing £12.14. 13 7 L sharps bin, weighing 11.62 kg, costing £6.12, and thirty-nine 5 L sharps bins, weighing 35.18 kg, costing £18.54. This would take eight weeks to produce a tonne. Producing around 6.5 tonnes a year costing £3,425.50. Compared to other types of waste: Sharps Containers: £527 per tonne Clinical Waste: £303 per tonne Domestic Waste: £104 per tonne Recycling: £30 per tonne Clinically objects that aren't sharps are placed in sharps bins, for example arterial blood gas syringes, this item could be put into a clinical waste bin. While this suggests education is needed there are other methods to reduce Sharps Bin wastage. These include: having needle only sharps bins, and solidifying chest drain bottles post removal.Conclusions: This Cardiac Intensive Care Unit produces a volume of high-cost waste. A large proportion of this waste can be redistributed to other types of waste. This could make the unit more efficient, and reduce it's environmental burden. This audit suggests that it should be looked at other types of waste and other departments in the same manner on other units. In 2014 the bloodstream infection density related to hemodialysis catheter in the Hospital Moyses Deutsch was 13.7, which represents patient 16 in its entirety. Due to this high rate, it was necessary to review all related process through institutionalized and supervised practice, minimizing the risks of hemodialysis procedure and maintenance of the catheter, in order to directly reduce the length of hospital stay, morbidity and hospital costs. Objectives: Objective of this study was to evaluate the application of strategies according to IHI to reduce infection of the bloodstream related to CVC hemodialysis. Methods: The study was conducted in a 20-bed, medical-surgical ICU. Criteria for infection catheter related bloodtream infection are those from the CDC. Strategy was to implement the Permanent education of employees, highlighting the importance of prevention of infections;Training of new employees as the hemodialysis routine and safe and aseptic techniques; Optimize other measures that can reduce the risks, such as early removal of invasive devices.Raise awareness of nursing staff about the importance of their role in the prevention of infection, such as maintenance of the catheter with use of aseptic techniques; Disseminate monthly for teams infection rates; Make benchmarking with other services; The goals were the ICU team adhesion of 46 % achieved in six month after bundle implementation and 100 % after one year of follow up. From june 2012 on, the ICU A1108 Effect of therapeutic hypothermia on mitogen activated protein kinase pathway in the brain tissue of a swine cardiac arrest model Y. C Objective: To investigate the change in mitogen-activated protein kinase pathways in the brain tissue after therapeutic hypothermia in swine cardiac arrest model. Design. Prospective animal study Setting. University animal laboratory Subjects. Male domestic pigs (n = 24) Interventions: After the return of spontaneous circulation by cardiopulmonary resuscitation following 6 min of no flow time induced by ventricular fibrillation, pigs were randomly assigned to one of four groups (sham, normothermia, 24 hr of therapeutic hypothermia, 48 hr of therapeutic hypothermia). Therapeutic hypothermia (core temperature 32-34°C) was maintained and the pigs were then rewarmed for 8 hr. At 60 hr after the return of spontaneous circulation, the pigs were sacrificed and brain tissues were harvested. Measurement and main results: We measured the tissue levels of p38, JNK, and ERK pathway expressions in swine brain hippocampus of the four groups. The phosphorylated p38 to p38 ratio and phosphorylated JNK to JNK ratios were significantly increased in all of the intervention groups, relative to the sham group. but the phosphorylated ERK to ERK ratio was increased only in the therapeutic hypothermia groups (p-value = 0.026 in the 24 hr of therapeutic hypothermia group and p-value = 0.002 in the 48 hr of therapeutic hypothermia group, both compared to the sham group). Conclusions: Normothermia activated the p38 and JNK pathway. And did not activate the ERK pathway in ischemia-reperfusion injury after cardiac arrest. Therapeutic hypothermia, however, did not attenuate the activation of the p38 and JNK pathways, but activated the ERK pathway, which seemed to be dose dependent with the duration of therapeutic hypothermia.

Effect of permissive hypercapnia on outcome of cardiac arrest in a porcine model of cardiopulmonary resuscitation G. Babini 1 , G. Ristagno Figure 122 ). Pigs in the hypercapnic group showed a trend towards longer survival. ETCO 2 and PCO 2 were significanlty higher in the hypercaninc group compared to the normocpanic one (Table 88 ). pH and PO 2 trended to be lower in the hypercapnic group during the 4 hrs of observation. Hypercapnia was associated with significantly higher mean arterial pressure during the post-resuscitation (PR) period ( Lesser neuronal degeneration was seen in the frontal cortex in the hypercapnic group compared to the normocapnic one (Figure 122) . Neurological recovery was equivalent in the two groups ( Figure 122 ). Conclusions: Permissive hypercapnia after resuscitation was associated with better mean arterial pressure and lesser neuronal degeneration in pigs. Grant support. Laerdal Foundation for Acute Care, Norway Introduction: Pain is the main indication for utilisation of the physician staffed prehospital emergency service in Germany. 1 Data from Switzerland 2 showed that oligoanalgesia (inappropriate treatment of pain with NRS > 3) is common in trauma patients. Objectives:(1) Determination of the frequency of oligoanalgesia in trauma patients at our prehospital emergency service location in Objectives: Assessment of characteristics and outcomes of patients who suffer cardiopulmonary arrest resuscitated in a tertiary hospital, inside and outside Intensive Care Unit, according to Utstein style.Methods: A prospective cohort study was performed according to Utstein style. Every arrest occurred in the hospital "Virgen de las Nieves" (Granada, Spain) for a period of 3 years (July/09-June/12) were included. All arrest occurred in all areas of the hospital were included, except those in operating rooms and anesthesia recovery room (not attended by the resuscitation team) and those commenced in the prehospital setting. We also excluded patients in whom no resuscitation attempt was made or those suspended either by existence of a living will, by orders DNR or considered futile. The variables were grouped according to the location (inside or outside the ICU). Chi 2 test was performed when the dependent variable was qualitative and a t-Student test when it was quantitative.Results: During this period a total of 297 patients suffered at least one episode of arrest and they were resuscitated. Most frequent sex was male (61.3 %) with a median age 69 years (65.2 ± 14.2 years; interquartile 57-76 years). The cardiac origin was the most common aetiology (40.4 %). The ICU was the area most frequent location (43.8 %). When comparing the characteristics of ICU arrests with the rest of the hospital, significant differences were observed. It was most likely to have a shockable initial rhythm (χ2: 10.6; p = 0.004), younger age (62,1 ± 14,9 vs 67,5 ± 13,2 years; t = 3,3; p = 0,001), shorter interval to defibrillation (2,1 ± 2,3 vs 4,7 ± 2,4 min; t = 3,6; p = 0,001), shorter period until start of resuscitation (1,3 ± 2,3 vs 4,9 ± 4,3 min; t = 8,9; p < 0,001) and shorter total duration (15 ± 15,2 vs 24,7 ± 19,5 min; t = 4,4; p < 0,001). However, no differences were found in coronary aetiology, sex, recovery of spontaneous circulation and hospital survival (22.1 % vs 23.8 % in ICU).Conclusions: Despite higher frequency in initial shockable rhythms and lower intervals until defibrillation and resuscitation in the ICU, no differences were found in initial recovery or hospital survival.

Delayed onset of cardio-pulmonary resuscitation (CPR) does not induce hyperfibrinolysis in a piglet model of ventricular fibrillation -a pilot study in Göttingen Minipigs N. Introduction: Pulmonary arterial hypertension (PAH) is a disease with gradually increased pulmonary vascular resistance and pressure, often leads to right ventricular (RV) failure and death. Excessive proliferation of pulmonary arterial smooth muscle cells (PASMCs) is regarded as the major cause of the remodeling of pulmonary artery, whereas the underlying mechanism is largely unclear. Caffeic acid phenethyl ester (CAPE) is the main component of propolis, which is known as a versatile compound of antimitogenic, anticarcinogenic and anti-inflammatory potentials. Objectives: To investigate the effects of CAPE on the improvement of the hemodynamic function in PAH animal model and to explore the underlying mechanisms in in vitro PASMCs. Methods: Animal model of PAH symptom was induced in 200-250 grams Sprague-Dawley rats by subcutaneous injection of monocrotaline (MCT, 60 mg/kg). 2 weeks later, the MCT-induced PAH rats received intraperitoneal administration of CAPE with various dosages of 5 or 10 mg/kg once per day, for further 2 weeks. Hemodynamic functions, including RV systolic pressure (RVSP) and Fulton index, were measured before sacrifice. The lung tissues were harvested for examining the vascular remodeling of pulmonary artery. To investigate the molecular mechanisms, in vitro cultured human PASMCs challenged with either 3 % oxygen level or recombinant human PDGF (40 ng/mL), followed by the treatment of CAPE in 5 or 10 mM. The change of expression level and phosphorylation of the cellular signaling molecules, including ERK, AKT, NF-kB, or Hif-1a, were analyzed by semi-quantitative PCR and western blotting, respectively. Results: In MCT-induced PAH rats, CAPE significantly improved the hemodynamic values of RVSP, Fulton index, and attenuated the severity of pulmonary vascular remodeling. Furthermore, the administration of CAPE critically reduced the expression levels of Hif-1a, NF-kB and PDGF molecules in the lung of MCT-induced PAH rats. In vitro assay showed that an increased expression level of hif-1α and pdgf genes in hPASMCs was observed under hypoxia or PDGF stimulation, which was significantly suppressed following CAPE treatment. For chemical inhibition, we indicated that cellular signaling molecules ERK, AKT and NF-kB were involved in the up-regulation of hif-1a and pdgf genes, which were responsible for the proliferation of hPASMCs exposed to hypoxia or PDGF stimulation. In addition, CAPE also significantly promoted the number of apoptotic cells and the number of cell arrested in G0 phase of hPASMCs by TUNEL assay and SA-b-galactosidase staining, respectively. Conclusions: We showed evidence that the natural compound CAPE could provide therapeutic benefits on the reversal of experimental PAH rats. Importantly, the results further indicated that the hif-1a-mediated pdgf expression is a positive feedback mechanism underlying the pathogenesis of PAH, which was regulated by the AKT/ERK/NF-kB signaling.

Right ventricular arterial coupling after cardiac surgery: a preliminary report P. Introduction: Right ventriculo-arterial coupling (rVAC), defined as the ratio of end-systolic elastance (rEes) to pulmonary arterial elastance (rEA) is considered a sensitive method to assess right heart performance [1] . Objectives: In this study we aim to identify the feasibility of measuring rVAC in hemodynamic deranged patients undergoing complex/ emergency cardiac surgery using cardiopulmonary bypass (CPB) as an experimental model of further hemodynamic impairment. Objectives: This study aims to determine the frequency of the VRE colonization and the transformation into infection and the risk factors, which lead to infection. Methods: The patients who were hospitalized for at least 24 hours in tertiary mixed type ICU between 2012 and 2015 and had VRE colonization and VRE infection during or following their hospitalization were included in the study and their medical records were examined retrospectively. VRE rectal swab sample was taken from each patient at his arrival and once a week afterwards. When negativity was detected in the rectal swab sample, which had been taken total 3 times successively from those with positive VRE; that patient was considered VRE negative. Their demographic data, APA-CHE II scores, invasive procedures, treatments (corticosteroid, antibiotics, etc.), nutrition types, laboratory results and ICU outcome were recorded. Results: VRE colonization was detected in 110 of 1730 patients (6.4 %) admitted to ICU. VRE infection developed in 12 of 110 VRE-colonized patients (10.9 %). Among these infected patients; it was (n = 5) 41.7 % primary bloodstream infection, (n = 6) %50 urinary tract infection, (n = 1) % 8.3 pneumonia. In VRE colonized patients (64.3 %) and infected patients (91 %), the most frequent factor was E. faecium. In 67 % of the VRE-colonized patients, VRE became negative in their stay at ICU. Previous renal replacement treatment was significantly higher in statistical terms in the VREinfected group (66.7 %) when compared to VRE-colonized group (26.1 %) (p < 0.05). In the VRE-infected group, colonization with VRE lasted longer than 1 week in 10 patients (83.3 %) were determined. Demographic data, APACHE II scores, treatments, nutrition types, previous antibiotic usage and types, invasive procedures, laboratory results and ICU outcome were similar between the VRE-colonized and infected patients.Objective: To assess the value of enteral paramomycin to decontaminate patients with rectal colistin and/or carbepemenase resistant microorganisms colonization to prevent the development of ICU nosocomial infections Methods: All consecutive patients admitted to the ICU from October 2011 to September 2015, expected to require tracheal intubation for longer than 48 hours, were given SDD with a 4-day course of intravenous cefotaxime, plus enteral colistin, tobramycin, nystatin in an oropharyngeal paste and in a digestive solution. Oropharyngeal and rectal swabs were obtained on admission and once weekly. Rectal swabs colonized by colistin and/or carbepemenase resistant microorganisms were treated with enteral paramomycin 1 gram every 6 hours a day, in order to eradicate them and prevent nosocomial infections. Categorical variables were summarized as frequencies and percentages and the continuous ones as medians and interquartile ranges (IQR) or means and standard desviations. Statistical significance was set at p ≤ 0.05. Results: We applied paromomycin treatment to 58 colonized patients with rectal colistin resistant microorganisms. All of them had colonization by Extended Spectrum Beta-lactamases (ESBLs). Also, all of them but two were Klebsiella pneumonia. Out of these two, one patient was colonized by Enterobacter spp and other one by Escherichia coli. Demographic data and type of admission are shown in Fig. 142 . Forty out of 58 (68,9 %) of the studied patients the rectal swab became negative. Five out of the 58 patients were colonized by carbapenemases producing microorganisms and one of these died with persistent multirresistant rectal colonization. Only 16 out of the 40 patients that negativized the colonization received concurrent susceptible IV antibiotics. Only 1 of the paromonycin treated patients developed a mediastinitis infection due to one of the treated microorganisms. Finally, 21 patients died in the ICU. Conclusion: Our data show that enteral paramomycin is effective in treating rectal colistin and/or carbepemenase resistant microorganisms colonization allowing clinicians preventing the development of ICU nosocomial infections. Introduction: Diaphragmatic thickness increases as lung volume increases towards TLC. It has been shown that in healthy subjects, diaphragmatic thickness, increases as lung volume increases, above 0.5 of the vital capacity (VC). In mechanically ventilated patients, different levels of PEEP are used to improve oxygenation. There is no information about the diaphragmatic thickness when in ICU patients, lung volume increases with PEEP towards TLC. Methods: In patients with Acute Respiratory failure (ARF) and lower lobe atelectasis detected by Lung ECHO, two levels of PEEP (8 and 15 ± cmH2O) are used to increase lung volume and to improve oxygenation. End Expiratory Lung volume (EELV), and diaphragmatic thickness was measured at baseline (ZEEP) and at the two levels of PEEP. EELV was measured with a Nitrogen indirect dilution method and diaphragmatic thickness at the zone of apposition with echography using a 12 MHz linear probe. Statistical analysis was performed by one way ANOVA and normal distribution by Colmogorof-Smyrnof test. Results: 22 patients (17 M and 7 F) with a mean age of 58 ± 18 were studied. Diaphragmatic thickness at baseline was 0,21 cm and EELV at 1276 ml (43 %) of the predicted (2942 ml). At the intermediate and high level of PEEP diaphragmatic thickness did not change significantly (0.22 and 0.23 cm, respectively, p = 0.38) and EELV increased at 55 % (1621 ml) and 68 % (1995 ml) of the predicted. The increase in lung volume induced by PEEP was at 32 % and 40 % of the predicted VC (5049 ml). Mean PaO2/FiO2 ratio did not change significantly Conclusions: Mechanically ventilated patients for ARF have a severe reduction in their EELV or FRC. The use of PEEP reestablishes partially the EELV, but not to his normal levels(predicted FRC). Despite high levels of PEEP, diaphragmatic thickness remained constant because the increase in EELV never attained the 50 % the vital capacity.Introduction: The majority of patients entering the weaning process from mechanical ventilation (MV) in the Intensive Care Unit (ICU) will have a short and simple weaning (SW) successfully terminated within 24 hours, while other may take up to one week (difficult weaning) or longer. Studies have shown that using a sedation or a weaning protocol could reduce the length of mechanical ventilation and the weaning duration. Objectives: To describe factors associated with SW and particularly assess if sedation and weaning protocol are associated with the proportion of patients having a SW. Methods: We used the data from the WIND (Weaning accordIng New Definition) study, a prospective multicenter observational study performed in France (29 ICUs), Spain (6 ICUS) and Switzerland (1 ICU) from April to August 2013. Ventilation and weaning modalities were daily assessed until discharge in all intubated patients admitted to the participating ICUs. We defined 1) weaning attempt (WA) as a spontaneous breathing trial (SBT) or an extubation attempt (with or without SBT), 2) successful weaning as an extubation without death or invasive mechanical ventilation within 7 days. We considered patients as having a SW if weaning was successfully terminated within 24 h following their first WA. Having a protocol for sedation or for weaning (or both) was asked to each center. Quantitative and qualitative variables are presented as mean (standard deviation), median [interquartile range] or number (percentage) as appropriate. Comparisons of proportions were made using Chi2 or exact Fisher tests and continuous variables were compared using Student t-test or Wilcoxon rank sum test as appropriate. We performed a multivariable analysis of factors associated with SW by means of a logistic regression, forcing both sedation and weaning protocols in the final model. All statistical tests were two-sided and P values of 0.05 or less were considered significant. Results: Among the 2729 patients included, 2051 patients entered the weaning process and we only kept in the present analysis the 1812 patients who did not have any decision of withholding or withdrawing mechanical ventilation. Among these 1812 patients, 1413 (78 %) had a SW and 399 (22 %) had a weaning duration longer than 24 hours. Main clinical characteristics are shown in Table 95 . Conclusions: In this study of 1812 patients with a daily assessment of the weaning process, hospitalization in an ICU using a sedation protocol or a weaning protocol (as declared by the center) was not associated with a higher proportion of patients having a simple and short weaning. Admission for planned surgery, younger age, lower SOFA score at admission and shorter duration of ventilation before any weaning attempt were associated with a higher proportion of simple and short weaning.

This study benefited of a grant of the non-profit Association Départementale des Insuffisants Respiratoires (ADIR) of the Haute Normandie, France. Introduction: During weaning from prolonged ventilation overload of diaphragm as main breathing muscle should be avoided. Clinical criteria are used for determining the end of the spontaneous breathing trial (SBT) in the context of a discontinuously concept for weaning. In addition the patients subjective feeling of breathing exhaustion plays an important role. In incommunicable patients lacks this possibility for feedback.Continuous monitoring of diaphragm electrical activity could give information of respiratory muscle effort during SBT. Objectives: In tracheotomized patients undergoing prolonged weaning the relationship between the protocol-based definition of the end of a SBT and the course of the electrical activity of the diaphragm (EAdi) should be examined. Methods: Prospective observation study conducted in a 30 beds intensive care unit in an early rehabilitation clinic. 29 patients that were not communicable because of stroke (17), cerebral hypoxaemia (5), traumatic brain injury (7) have been included. Using an EAdi-catheter usually applied in NAVA (neurally adjusted ventilatory assist)-Ventilation, peak of diaphragm electrical activity (Eadi peak) was continuously recorded 30 minutes before disconnection from ventilator up to 30 minutes after reconnection. The weaning protocol contained two possibilities for terminating of the SBT: reaching clinical signs of ventilation exhausting or reaching a previously fixed time limit. Results: Median duration of mechanical ventilation at study start was 22 days and 37 days at successful weaning (28/29 patients, 1 died). 152 SBT have been recorded, 91 terminated because of exhaustion, 61 by time limit. Median duration over all was 244 minutes (exhaustion: 232/ time limit: 315). With multiple regression analysis, the relationship between the duration of the SBT and the EAdi peak was examined. Looking at all SBT, which were terminated due to exhaustion, shows that the duration of the SBT has a highly significant impact on EAdi ( p < 0.0001). The mean increase of Eadi peak was 10.899 μV (absolute) and 1.446 (relatively). In SBT terminated because the time limit has been reached, there was no significant correlation between the time and course of Eadi peak. Conclusions: Continuous recording of the electrical diaphragmatic activity during weaning of prolonged ventilation in incommunicable patients can be used as supplementary parameter in monitoring the respiratory function. Introduction: Patient-ventilator asynchronies are associated with poor outcome. It was suggested that bedside analysis of ventilator waveforms may help detecting different types of asynchrony and setting properly the ventilator [1] . Objectives: To test accuracy of a "waveform" method, based on specific signs on airway pressure (Paw) and flow curves, in detecting spontaneous respiratory activity and asynchronies in patients under Pressure Support Ventilation (PSV).Methods: 16 recordings (12 min each) of esophageal pressure (Pes), Paw and flow were obtained in obstructive (75 %) and restrictive (25 %) patients under PSV with clinical evidence of poor patient-ventilator interaction. Tracings of 4426 breaths were visually analyzed for detection of spontaneous respiratory activity both with Pes (reference method) and without Pes (waveform method) by different operators. Breaths were defined as assisted, unassisted or autotriggered, and assisted breaths as delayed triggered, early cycled or delayed cycled. The waveforms method was applied in a selection of tracings (20 min, 544 breaths) by 4 different operators for assessment of inter-rater agreement. Results: The reference method detected 6 autotriggered (0.1 %), 976 unassisted (22.1 %) and 3444 assisted (77.8 %) breaths; among assisted breaths, 897 delayed triggered (26.0 %), 1231 delayed cycled (35.7 %) and 439 early cycled (12.7 %). Table 102 shows sensitivities and specificities (95 % CI) of the waveform method in evaluating patient-ventilator interaction. The waveform method detected the start of patient's inspiration and expiration with a bias of −23 and −32 ms and a precision (±1.98 SD) of 184 and 202 ms respectively. Absolute agreement among operators was almost perfect for unassisted breaths, strong for delayed triggered, delayed cycled and early cycled breaths, and weak for autotriggered breaths. Conclusions: The waveforms method is a reliable, accurate and reproducible method to assess patient-ventilator interaction and could help optimal setting of the ventilator. Automation of this method may allow continuous monitoring of ventilated patients and/or improved breath triggering and cycling.Methods: We studied a group de mechanically patients during the weaning time, at Pressure Support Ventilation (PSV) with different levels of assistance (High 15-23 cmH2O, medium 10-14 cmH2O, low 5-9 cmH2O). Esophageal, gastric, airway pressure, and airway flow were registered, samplig 278Hz. We determined the phase difference (Φ) relationships between the neuronal times obtained from derivative flow versus esophageal or gastric signal respect to machine cycle, by calculating the phase delay, dividing by the cycle time of ventilator*360°. Times (T) definitions: T0 = onset inspiratory effort, T1/2 = effort maximum. Data were analyzed by descriptive statistical methods and are expressed as mean ± SD, medians, interquartile range (IRQ, 25-75 % quartile), and coefficient of variation (CV). The comparisons were performed by Mann-Whitney Test. The relationships between measurement methods was examined using single linear regression and Bland-Altman analysis. Results: 10 patients were studied. For all data angle phase Φ median (IRQ): T0: 1,29 (−5,06 to ,05), T1/2: −1,61 (−7,07 to 4,98). The mean comparison of T0 and T1/2 between Pes and DF did not showed statistical differences for any level of support, and correlation R 2 > 0,99. The CV for all data at the T0 of Pes and DF: 11 % and 1 %, respectively; and for the T1/2 of Pes and DF: 50 % and 14 %, respectively; without differences between levels of assistance. Table 103 below show results from Bland-Altman analysis. Figure 144 show representative tracing of DF with well-defined inflection points (arrows) at T0 and T1/2, as the onset inspiratory flow and transition from inspiratory to expiratory flow. Conclusions: The derivative of flow signal is useful to measure with accuracy neuronal and cycling times, it´s more homogeneous and precise than obtained for esophaeal or gastric pressure for all levels of assistance. The derivative of flow signal is a non-invasive signal which can be calculated easily and useful by conventional ventilator. Introduction: Sepsis has been defined as organ dysfunction as a result of the inappropriate host response to infection. [1] Renal function is often injured at the early stage of sepsis. [2] Autoregulation, which plays an important role in maintaining an adequate renal blood flow against changes in blood pressure, could be impaired during sepsis, [3] thus resulting in AKI if blood pressure fluctuates greatly. Objectives: To investigate if there is any relationship between blood pressure variability (BPV) and AKI in septic patients. Methods: Clinical data of patients admitted to our 50 bed medical ICU between 05/2014 and 03/2015 were reviewed. Continuous records of blood pressure were analysed. Blood pressure variability was calculated as the coefficient of variation (CV) of mean arterial pressure in the first 24 h of admission. AKI was defined by the KDIGO definition according to creatinine change and urine output criteria. [4] Results: 275 adult patients with sepsis (age: 59.9 ± 17.3 years old; APACHE II score: 23.6 ± 7.1; male: 63.3 %) who stayed at ICU for more than three days were identified. AKI was presented in 70 (25.5 %) of them (stage 1: n = 33; stage 2: n = 15; stage 3: n = 22). The BPV was 12.0 ± 3.7 % for the patients with AKI versus 9.8 ± 3.2 % for the others (P < 0.001). ICU mortality was 41.4 % for the AKI group compared to difference was not statistically significant regarding creatinine level on admission (p = 0.317). Moreover, there was a positive correlation between uAnCR ratio on admission with AKIN staging and creatinine level of the all studied patient in the follow up days. The cutoff value of uAnCR on admission to predict later occurrence of AKI during ICU stay was 52.24 ng/mg: at this level, (88.57 % sensitivity and 53.30 % specificity). Conclusions: Urinary angiotensinogen is a new promising biomarker in early prediction of AKI in patients with severe sepsis.

Acute kidney injury in patients with severe sepsis or septic shock: a comparison between the "risk, injury, failure, loss of kidney function, end-stage kidney disease" (RIFLE), acute kidney injury network (AKIN) and kidney disease improving global outcomes AN69ST and polymethyl methacrylate (PMMA) membranes have strong adsorption capacity. Cytokines play important roles as the main mediators affecting critically ill patients. However, differences in the cytokine elimination by specific membranes during CHF have not yet been fully investigated. Objective: The objective of this study was to determine the elimination of cytokines by AN69ST and PMMA membrane filters during CHF in a pig sepsis model. Methods: Piglets (n = 7) weighing 20-30 kg were anesthetized and administered 30 μg/kg endotoxin. The Baxter sepXiris (AN69ST membrane) and the Toray Hemofeel 1.8 W (PMMA membrane) were used as hemofilters. Samples were taken at 1, 2, 4, and 6 hours after endotoxin administration, and the inlet plasma, outlet plasma, and filtrate concentrations of TNF-α, IL-1β, IL-6, and IL-8 were measured. Clearance values were calculated for each cytokine. Results: Endotoxin administration induced increases in the inlet plasma concentrations of all cytokines measured. The AN69ST membrane filter showed higher adsorption and clearance of IL-8 than the PMMA membrane filter at 6 hours after endotoxin administration (AN69ST: 23.66 ± 17.01 mL/min; PMMA: −6.62 ± 48.89 mL/min; P < 0.05). However, the PMMA membrane filter showed higher adsorption and clearance of IL-1β than the AN69ST membrane filter. IL-6 did not appear in the filtrate of the PMMA membrane filter, while IL-8 was not eliminated in the filtrate of the AN69ST membrane filter. In addition, the filtrate concentration of TNF-α increased after its plasma concentration decreased with the PMMA membrane filter. Conclusions: Shiga et al. 1 previously reported the efficacy of cytokine absorption by AN69ST membrane filters during continuous hemodiafiltration, and Matsuda et al. 2 reported the efficacy of cytokine absorption by PMMA membranes. However, the cytokine absorption efficacy by these two membrane filters had not been directly compared. The results shown here confirm that there are differences in cytokine adsorption by the AN69ST and PMMA membrane filters. Introduction: Continuous renal replacement therapy (CRRT) in intensive care is a cornerstone in the supportive treatment arsenal. Its influence on thermodilution cardiac output measurements, and the possible influence of central venous dialysis catheter(CVDC) position, has been studied but the results are of uncertain clinical impact (1) (2) (3) . There have been case reports describing the possibility of direct aspiration into a CVDC of drugs given in adherent central venous catheter(CVC) (4, 5) . Objectives: The aim of this study was to investigate if different positions of central lines influence infused noradrenaline during continuous renal replacement therapy (CRRT) in an experimental animal model. Methods: Ten anesthetized piglets received a CVC in the right jugular vein and two CVDCs (one via the same jugular vein as the CVDC and the other through a femoral vein). After randomization the CRRT was started in either one of the CVDCs and a nitroprusside infusion was started in an auricular vein. The dose was titrated until the mean arterial pressure (MAP) was 50 mmHg and then kept constant during the rest of the experiment. After reaching the intended blood pressure an infusion of noradrenaline was started and titrated with the goal of increasing the blood pressure to a MAP of 75 mmHg during 30 minutes. After a washout period the CRRT circuit was changed to the other CVDC and the experiment was repeated. Results: The median dose of noradrenaline with the CRRT in the jugular vein was 0.34 (IQR 0.25) and in the femoral vein 0.14 (IQR 0.13) μg/kg/min (p = 0.021). Conclusions: During CRRT, the noradrenaline dose needed to reach a target blood pressure in hypotensive piglets was twice as high with the CVC and CVDC close together, compared with CVC and CVDC on opposite sides of the diaphragm. This suggests that there is a possible clearance of noradrenaline and that the clearance is affected by catheter positioning

A first evaluation of omni, a new device for renal replacement therapy P. Schlaepfer 1,2 , J.-D. Durovray 1,2 , V. Plouhinec 1 , C. Chiappa 1 , R. Bellomo 3 , A.G. Schneider 1 1 Introduction: Due to the lack of conventional dialysis facilities in our centre, intermittent renal replacement therapies (IRRT) are led and performed by the ITU team. This team comprises a group of specialist outreach nurses with the support of intensivists. IRRT are performed nocturnally by ITU nurses in level 2 areas according to our Hospital policy. Objectives: To describe the use and results of IRRT in level 2 areas in patients that have left ITU with established AKI. These therapies are directed and performed by specialist intensive care nurses with the support of the ITU medical team. Methods: Retrospective observational study that included those patients admitted to level 3 areas at Harefield Hospital during 2015 that were transferred to level 2 areas still requiring IRRT. Demographic variables were collected, along with the indication and duration of IRRT and results. Results: 1829 patients were admitted to Harefield Hospital level 3 areas during 2015, of which 229 patients required continuous renal replacement therapies (CRRT). This population included patients admitted after cardiac and thoracic surgery, heart or lung transplantation, mechanical circulatory devices, out of hospital cardiac arrests (OOHCA) and medical admissions from the cardiology or cardio-thoracic surgical wards. Demographic variables were collected, along with the indication and duration of CRRT. 31 of those patients still required intermittent renal replacement therapies at their discharge to a level 2 area. 19 of them (61.3 %) were male and the group was a median age of 49.34 years. 4 of them (12 %) were hypertensive and 6 (19 %) were diabetic. As shown in Figure 149 , the most frequent reason for admission to intensive care was cardiac surgery (25.8 %, 8 patients), followed by lung transplantation, heart transplantation and medical admissions from the transplantation ward. The reasons for admission to intensive care in the general CRRT group are also shown in Figure 149 . The most frequent indication for initiation of CRRT was metabolic acidosis (54.8 %, 17 patients), followed by a combination of uraemia and fluid overload (16.1 %, 5 patients), uraemia (12.9 %, 4 patients) and fluid overload (3.2 %, 1 patients) as shown in Figure 150 . The median time of RRT was 93 days days whilst the median time of filtration in the general RRT group was 13 days. The in-hospital mortality (after discharge from ITU) was 32.3 % and was 45.4 % in the general CRRT group. No complications were associated with the use of intermittent renal replacement therapies in level 2 areas. Conclusions: The group of patients that required intermittent renal replacement therapies beyond their discharge from ITU had longer ITU and hospital lengths of stay. These therapies were performed safely in level 2 areas by the ITU team, allowing these patients to leave level 3 areas to continue their care.