key: cord-0033858-ojvijk0o
authors: nan
title: Poster Sessions 499–563
date: 2001
journal: Intensive Care Med
DOI: 10.1007/pl00003809
sha: 8b9d43a2d06f3dca9d267af70aa98f0de45f9e64
doc_id: 33858
cord_uid: ojvijk0o

nan

INTRODUCTION. Pediatric patients with inhalation injury often develop respiratory failure and require mechanical ventilation. Human and animal studies suggest that Conventional Mechanical Ventilation (CMV) may promote over-distention-induced pulmonary lesions, which could contribute to further lung injury. Protective ventilator strategies are now being used to prevent over-distention and stretching of normal lung units. High Frequency Percussive Ventilation (HFPV) is a form of protective ventilator strategy that has been shown to be effective in the treatment of adults and children with inhalation injury (1,2). The purpose of this study was to investigate if a protective ventilator strategy such as HFPV would result in improved pulmonary function in pediatric patients with inhalation injury when compared to CMV.

METHODS. Twenty-six severely burned children with inhalation injury who were previously ventilated by HFPV(n = 13) or CMV(n = 13) were evaluated for improvements in lung function at 3 months post discharge. Study variables included; Forced Vital Capacity (FVC), Forced Expiratory Volume in One second (FEV1), Maximal Voluntary Ventilation (MVV) and Diffusion studies (DLCO) . Pulmonary function studies are presented as mean % of predicted and comparison between groups was performed by an un-paired students t-test.

A total of 26 thermally injured patients with inhalation injury were studies. Demographic data showed no significant difference between the two groups in regards to age, % TBSA burn, or previous length of ventilation. Patients who were ventilated with HFPV had a significant reduction in the peak inspiratory pressure as compared to CMV. Pulmonary function showed no significant changes in the FVC or FEV1 between groups. However the MVV and the diffusion capacity were significantly increased in patients ventilated with the HFPV.

CONCLUSION. Results indicate that the use of a protective ventilator strategy such as HFPV is effective in significantly improving maximal voluntary ventilation and the diffusion capacity. The results may be relevant when choosing a ventilator strategy for pediatric patients with inhalation injury. 

Although there are a large number of croup scores used clinically and for research, none have ever been evaluated against objective measures of respiratory function. Clinical criteria included in scores include: cyanosis (or saturation < 92 %), respiratory rate, level of consciousness, breath sounds, presence of stridor, presence of pulsus paradoxus, retractions (1 and 2) METHODS. Ethical approval was obtained and parents consented to and were present during the studies. Studies were performed on 20 patients (aged 3.6 to 23.6 months, median 11.7) with clinically severe croup and 5 normal controls (aged 5 to 31 months, median 13) undergoing EEG as part of investigation of seizures. Children were sedated with chloral hydrate (50±75 mg/kg). We recorded tidal intrathoracic pressure changes (DPes cm H 2 O) with a waterfilled oesophageal catheter-manometer and air-flow using a face-mask and Fleisch #1 pneumotachograph. Polygraph traces were digitized. We made the following measures of respiratory mechanics: Frequency (f min-1), Tidal volume (VT ml/kg), Minute ventilation (VE ml/kg/ min), Peak inspiratory flow rates (VINSP ml/s), Pressure Time Product (PTP cm H 2 O.s.min-1) and Pressure Time integral (PTINT cm H 2 O.s.min-1). RESULTS. No croup patient was clinically cyanosed and on the 14 patients where saturation data was available, median saturation was 95 % (range 92±100 %). All croup patients had inspiratory and expiratory stridor, chest wall retractions and a croupy cough. Eighteen (90 %) of croup patients had palpable pulsus paradoxus,vs no controls. INTRODUCTION. Severe Traumatic Brain Injury (sTBI) remains a major cause of death in childhood.It is considered good practice, by most neurosurgical units, to monitor intracranial pressure (ICP) in these patients. Goal directed therapy based on maintaining cerebral perfusion pressure (CPP) in order to minimize secondary brain injury, is the main justification for this practice. Doubt remains about the benefits and effects on outcome.

METHODS. Demographic data, hourly ICP and CPP measurements (if monitoring was performed) and outcome were in all patients with a primary diagnosis of sTBI admitted to the Paediatric Intensive Care Unit (PICU) at Southampton University Hospital between 1 st April 1999 and 1 st August 2000. Patients who had ICP monitoring were divided into 3 groups based on their outcome at follow up. Outcome groups were based on those used in similar studies. Inter-group ICP and CPP data were analyzed.

RESULTS. 41 patients were admitted to PICU with a primary diagnosis of sTBI, representing just over 6 % of all admissions. 31 of these had ICP monitoring on the PICU. Their median age was 9.5 years (27 male, 14 female). The majority of these injuries were sustained as a result of road traffic accidents. The crude mortality rate within these patients was 17 %. Results are shown in Table 1 . The mean ICP was significantly higher in the disabled survivors than in those who made a full recovery (p < 0.01). The CPP, however, was not statistically different (p > 0.2) between these two groups. CONCLUSION. As would be expected, the patients had higher mean ICPs in the worse outcome groups. However, goal directed therapy to maintain CPP, although apparently successful, did not appear to prevent permanent neurological impairment in a significant number of patients. This data supports the view that measurement of ICP with goal directed measures to maintain CPP, is a simplistic approach.

Larousse E, Asehnoune K, Benhamou D. Dept. AnesthØsie-RØanimation, Hopital du Kremlin-Bictre, Bictre, France

Pediatric caudal anesthesia is an effective method with a low complication rate. Little is known about its cardiovascular consequences. Transesophageal Doppler, a non-invasive method provided the opportunity for a reappraisal of caudal's haemodynamic effects.

METHODS. After parental inform consent, Ten children aged two months to five years, scheduled for lower abdominal surgery were studied. General anesthesia was induced using sevoflurane. An orotracheal tube was then inserted followed by insertion of a transesophageal Doppler probe. Caudal anesthesia was performed using 1 ml/kg of 0.25 % bupivacaine with 1/200000 epinephrine. Hemodynamic parameters were collected before (first set) and after caudal anesthesia(second set). The two sets of haemodynamic data were compared using a paired Student ttest. A p value less than 0.05 was considered significant.

RESULTS. Heart rate, systolic, diastolic and mean arterial pressures were not modified by caudal anesthesia. Descending aortic blood flow increased dramatically from 1.14 to 1.92 l/min. (p = 0.0002). In the same time, aortic ejection volume rose from 8.51 to 14.53 ml (p = 0.0002).

Simultaneously aortic vascular resistances decreased from 6279 dyn.sec.m-5 to 3901 (p = 0.005).

CONCLUSION. As previously described, caudal anesthesia did not affect heart rate and mean arterial pressure. Surprisingly, a significant increase of descending aortic blood flow was observed while systemic vascular resistances decreased. These phenomenon were probably induced by a caudal sympathetic blockade.

Pavoni V 1 , Paparella L 1 , Gianesello L 1 , Volta CA 2 , Marangoni E 2 , Alvisi V 2 , Ragazzi R 2 , Gritti G 1 . 1 Service of Anesthesia and Intensive Care, Ospedale Civile, Padova, 2 Section of Anesthesia and Intensive Care, S.Anna University Hospital, Ferrara, Italy

Brochodilating agents are commonly used in patients with chronic obstructive pulmonary disease (COPD) undergoing mechanical ventilation for acute respiratory failure (ARF). The distribution of the drug in the airways is influenced by several factors, related to the patients, to the ventilator and its circuit, to the drug used and the route of administration. Since the use of an external PEEP (PEEPe) could reduce lung inhomogeneity (1), we reasoned that the administration of salbutamol in presence of an applied PEEPe should improve the efficacy of this therapy. METHODS. Five flow limited COPD with ARF were studied before and after administration of four puffs of salbutamol. The presence of FL was detected by the NEP test. During the drug administration patients were randomly ventilated in assisted-control ventilation (a) on ZEEP and (b) with an applied PEEPe equal to 70 % of the value of intrinsic PEEP (PEEPi). Then six hours later the same patient was given of the drug in the opposite way. Data of respiratory mechanics were determined on ZEEP by the constant flow rapid occlusion method, which allowed to calculate maximal, minimal and additional resistance (Rmax,rs; Rmin,rs; Drs) and PEEPi RESULTS. The administration of bronchodilating agents with PEEPe was associated to a greater reduction of Rmin,rs [-20 % (a) VS ±36 % (b)] and PEEPi [-6 % (a) VS 16 % (b)] when compared to the administration on ZEEP (table 1) . INTRODUCTION. The relative inaccessibility of airways and alveoli for measurements and the anatomical complexity of the bronchial tree make assessment of lung disease difficult. Mathematical modelling of gas/fluid dynamics of the lung appear promising for improved understanding of pathomechanisms of lung disease and for training in mechanical ventilation.

METHODS. Based on the fundamental constitutive equations of fluid dynamics, a computer simulation model of the lung was created. This allows to model the human tracheobronchial tree in detail, including localised or diffuse, symmetric and asymmetric, homogenous and focal airway abnormalities, as well as the presence of secretions. Different ventilator settings and a variety of gas mixtures can be chosen, and modelling of turbulent or laminar flow is derived in real-time from actual flow conditions within each airway segment.

Using human morphometric data for simulation of normal lungs with CMV ventilator settings, normal ventilation patterns, pressure, and resistance are observed, further confirming the applicability of the modcl. Introducing an orotracheal tube into the model leads to an increase overall airway resistance, as does obstruction at the level of the bronchial tree, with typical prolongation of expiration. With airway obstruction, especially when combined with higher ventilation rates, air trapping is observed. In patients with asymmetric airway obstruction, shifting of air from one lung to the other (ªPendelluftº), a phenomenon known from respiratory physiology, but usually not directly observable in the ICU patient, can be observed. Furthermore, the effects of helium/oxygen ventilation, a new ventilation strategy, can be directly and quantitatively assessed, and its differential effect in upper versus lower airway obstruction can be studied.

CONCLUSION. Fluid dynamics computer simulation of the ventilated patient is feasible and allows quantitative assessment of various types of respiratory disease. This will be especially useful in the assessment of treatment effects of new ventilatory strategies (like heliox ventilation) and in the training of intensivists, because the deleterious effects of suboptimal ventilation strategies can be directly observed without endangering the patient. METHODS. twenty patients were studied, 8 with acute respiratory distress syndrome (ARDS), 8 with moderate acute lung injury (ALI), and 4 with acute-on-chronic respiratory failure (ACRF), all on controlled mechanical ventilation with constant flow. Airway flow was measured with a heated pneumotachograph interposed between the endotracheal tube and the Y piece of the ventilator. The flow signal was integrated to calculate the volume. The time constant was measured in the expiratory phase of the volume curve, adjusted to a single-exponential equation plus a constant [V (t) = V0 e _(t/RC) + b]. We calculated the respiratory rate considering an I:E ratio of 1:2 and an expiratory time of 4-fold the time constant.

RESULTS. the value of b was very small (-0.034 l), being > 0.1 l in only three patients, and correlated with the tau value (r = 0.85 

Patients with acute severe asthma present with acute respiratory acidosis due to Ventilation/Perfusion mismatch. Clinical and metabolic predictors of respiratory failure except in obvious extreme cases are sub optimal. Standard Base Excess obtained from arterial blood gas varies in proportion to metabolic disturbance and if low indicates concomitant metabolic acidosis. We hypothesized that in acute severe asthma, respiratory muscle fatigue will cause lactic acidosis and a change in base excess. This change in base excess in turn may be a predictor of respiratory failure.

Retrospective data extracted from medical records of 30 consecutive patients with acute severe asthma requiring intensive care unit admission in urban teaching hospital. 

We mentioned that patients with low electrolyte status had a higher risk for development of resporatory failure than patients with well balanced electrolytes by same APACHE-II scores. We incubated cultured lung epithelial cells with different concentrations of K + . Cell cultures with higher concentrations did express more Ion channels than those with lower concentrations.

We treated 100 Patients with serum K + lower than 3.5 mmol/l with KCl-Infusion with an predictive balance within 36 hours and 100 patients with an predictive balance within shortest time. We analysed the evidence of respiratory failure in both groups by monitoring BGA, Serum-lactate and SaO 2 .

RESULTS. The group with slow substitution did suffer from ARF significantly less than other group. There were no cardiac problems reported because of low K Levels.

In the rapid balance group the mortality was significantly higher. 25 patients died on MOF in 18 cases there was an maximum increase and activation of ion channels in lunge parenchym.

CONCLUSION. Ion channels are activated by changes in serum K Levels. Consecutively all channels have part on cellular water uptake. In clinical practice there was no evidence found for rapid balance of K Serum-Levels. Moderate k Substitution may lower the incidence of ARF. METHODS. Since observational studies have been now acknowledged as similiar valid as randomised studies (3), we performed a large prospective observational study with the same inclusion and exclusion criteria and outcome parameters used in the recent Canadian multicenter study (2). The only difference between both studies was the sucralfate dose. While the Canadian study administered sucralfate in a dose of 4 x 1 g /day, our patients received a 3 x 3 g /day regime. Preliminary endpoint was the occurrence of clinically important bleeding. The predicted bleeding incidence was calculated based on the results of the Canadian study and was compared to the actual bleeding incidence.

After approval of our ethics committee 620 mechanically ventilated patients were included in the study. Coagulation disorders were present in 95 (15.3 %) patients. Apache II-Score was 16.47 7.49 and duration of mechanical ventilation was 6.8 10.43 days. The predicted incidence of clinically important bleeding in our study was 24 (95 % confidence interval 15±36), while the actual incidence was 1 (95 % confidence interval 0±6, p < 0.01).

The results of the Canadian study-group are not valid for ICU¢s with a different dosage regime of stress ulcer prophylaxis. Since there are no hints, that the efficacy of ICU therapy in Canada and Germany is principally different, the reason for the difference of bleeding incidence may most probably be explained by the divergent sucralfate dosage . INTRODUCTION. The purpose of this study was to thoroughly elucidate the effects of body positioning on the partitioned inspiratory work done on the respiratory system (RS), chest wall (CW), and lung (L) in chronic obstructive pulmonary disease (COPD).

METHODS. 5 COPD and 5 control (with no respiratory pathology), anesthetized, paralyzed, and mechanically ventilated patients were positioned supine (S), semirecumbent (SR), and prone (P) in random order. Tracheal (RS) and esophageal (CW) pressures were determined while tidal volume (VT) was varied from 0.2 to 1.2 L by 0.2 L increments at constant flow (0.9 L/ s). RS and CW intrinsic positive end expiratory pressures (PEEPi), peak pressures, resistive pressure drops, viscoelastic pressure dissipations, and static (plateau)pressures were determined with the end expiratory/inspiratory occlusion technique (1). RS and CW pressure volume curves were constructed for each body position, and the dynamic and static components of the RS, CW, and L work per breath were determined (at VT = 0.6 and 1.2 L) as previously described (1). Statistical analysis was performed with two way analysis of variance. The Sheffe test was used for post hoc comparisons. Significance level was set at P < 0.0024 (Bonferroni correction).

A) COPD group: At 0.6 L, the total RS (WtotRS) increased from S to P due to simultaneous increases in the RS elastic work (WelRS), and static PEEPi L and CW work (WPEEPiL and WPEEPiCW respectively); from S to SR, an increase in the resistive CW work was counterbalanced by a simultaneous decrease in WPEEPiCW; from SR to P, WtotRS increased due to simultaneous increases in WelRS, WPEEPiL, and WPEEPiCW. At 1.2 L, from S to P, a decrease in dynamic L work was counterbalanced by simultaneous increases in WPEEPiL and WPEEPiCW; from S to SR, only WPEEPiCW decreased; from SR to P, a decrease in the viscoelastic L work was counterbalanced by simultaneous increases in WPEEPiL and WPEEPiCW. B) Control group: At 0.6 as well as 1.2 L, neither WtotRS nor any of its L and CW dynamic and static components were affected by the changes in body position.

In COPD, at a VT of 0.6 L, SR position seems to be an effective alternative position to S; P positioning seems disadvantageous, because it augments air trapping; at VT = 1.2 L, although WtotRS is not affected by body position change, P positioning is still associated with increased air trapping. In control patients, the effects of S, SR, and P positioning on breathing work are similar at both 0.6 L and 1.2 L. 

Central ventilatory drive has studied in several different neuromascular diseases. Myastenia Gravis has been implicated with increased neuromascular drive. The aim of this study was to identify the respiratory centre response in generalized Myasthenia Gravis.

METHODS. 15 patients with moderate generalized Myasthenia Gravis, 9 males and 6 females aged 18±69, were included in the study. 15 healthy persons with similar age and sex were used as control group. Spirometry, ventilatory muscle strength as reflected from maximum inspiratory and expiratory respiratory pressure (MIP, MEP),respiratory rate (RR), tidal volume (TV), mean inspiratory flow (Vt/Ti), Ti/Ttot, mouth occlusion pressure (P0.1), were recorded in all the patients. Statistical analysis was performed by using student t-test.

In patients with Myasthenia, forced vital capacity was decreased in 4 subjects. Mean value of forced vital capacity ( FVC) was 85 14 % of predicted. Muscle stength was markely reduced, maximum inspiratory pressure (MIP) and maximum expiratory pressure (MEP) was 52.5 and 48.22 % of predicted value respectively. There was no significant difference for parametres of control of breathing (VT, RR, Vt/Ti, Ti/Ttot, P0.1) between myasthenic patients and the control group. Data is summarized in the following METHODS. Two hundred and ten patients with several etiologies in weaning process were evaluated ( all of then with FiO 2 < 0.4 and PEEP < 8 cmH 2 O). The age ranged between 17 and 94 years. APACHE II ranged between 2 and 40 (mean 17.7). Success in weaning was obtained in 182 patients ( those who were not reintubated in the following 48 hours after the definitive disconnection from the mechanical ventilator ). The ventilators used were Evita 1 and 2 ( Drager ± Germany ). The four predictive parameters were measured before the definitive disconnection from the mechanical ventilation. Clinical and neurological stability were indispensable for the admission of the patients in the study. METHODS. Twenty one patients who needed mechanical ventilation for more than 48 hours in a general intensive care unit were prospectively evaluated in a randomized cross-over study. All patients were submitted to the three methods (EPAP, PSV and T-piece) during 30 minutes. Each method was followed by a rest time (at least 30 minutes), They were monitored by Vent-Track (Novametrix, USA). Parameters, measured at 1 and 30 minutes, were: work of breathing (WOB), minute ventilation (MV), arterial oxygenation (SatO 2 ), end-tidal CO 2 (ETCO 2 ), heart and respiratory rate (HR and RR), mean arterial pressure (MAP). Comparisons were done by a one way analysis of variance and t test. The level of significance was p < 0.05. RESULTS. The preliminary results obtained with the three methods were: The differences in the median values among the treatments groups are greater then would be expected by chance in WOB (p value = 0.001*), MV (p value = 0.002**) and SatO 2 (p value = 0.004***). (*EPAP different from PSV, PSV different from T-piece and EPAP different from T-piece; **EPAP different from PSV and PSV different from T-piece; *** EPAP different from T-piece and PSV different from T-piece). Comparison between ETCO 2 , HR, RR and MAP obtained at 1 and 30 min were not different. CONCLUSION. Those preliminary results showed that EPAP offered greater WOB than PSV or T-piece, smaller MV than PSV but similar to T-piece and similar SatO 2 to PSV but greater than T-piece. In general, EPAP offered no advantage over the other methods in patients weaning from mechanical ventilation. 

We evaluated 68 patients with COPD in weaning process (all of them with FiO 2 < 0.4 and PEEP < 8 cm H 2 O ). The ages ranged between 43 and 94 years. APACHE II ranged between 10 and 25 (mean:15.14) Success in weaning was considered when the return to mechanical ventilation wasn't necessary in the following 48 hours after definitive disconnection from the mechanical ventilator. The ventilators used was the Evita1 and 2 (Drager ± Geramany). Clinical and neurological stability were indispensable for admission the patients in this study. Index of Nemer = Quasi-estatic Lung compliance x SaO 2 /f/Vt ratio > 25 Positive prognosis in weaning; < 25 Poor prognosis in weaning Rapid Shallow Breathing Index = f/Vt ratio > 105 Poor prognosis in weaning; < 105 Positive prognosis in weaning.

Of the 68 patients evaluated, 57 were successful in weaning while 11 were unsuccessful, 5 of each died. INTRODUCTION. The aim of this study it was to determine the value of the respiratory rate to tidal volume ratio (f/TV) as a weaning parameter in mechanically ventilated COPD patients.

We studied 32 COPD patients on the day that their primary physician decided a Tpiece weaning trial. Minute ventilation (VE), respiratory rate (f), rapid shallow breathing index (f/Vt), heart rate (HR), and arterial blood pressure (BP), were measured just before and at 5, 30, and 60 minutes after the patients were disconnected from the ventilator. Arterial blood gases were measured before disconnection and at 30 and 60 minutes of the trial. Patient characteristics included age, days on mechanical ventilation, and endotracheal tube size, while laboratory data of the trial day included temperature, complete blood count, serum albumin, magnesium, phosphorus, and electrolytes.

. Group A consisted of 21 patients (mean age 70 9.2 years, ventilator days: 5.2 2.6) who were able to remain in spontaneous breathing and were successfully extubated. Group B (mean age: 71.9 4.7 years, ventilator days: 5 2.5) consisted of 3 patients who were extubated and reintubated during the next 48 hours and 8 patients who returned to ventilatory support because of clinical deterioration during the T-piece trial. Statistical analysis (x2-test) between the two groups for all measured parameters at the 4 time-points of the trial showed no differences. Mean f/Vt was high for both groups, without a rising trend on subsequent measurements during the trial for both groups. To evaluate the predictive accuracy of f/Vt ratio we used six threshold values (100, 105, 110, 115, 120, and 125) . The highest sensitivity and specificity was achieved with a threshold value of 125 at 60 minutes of the T-piece trial (81.3 and 45.5, respectively).

CONCLUSION. Rapid shallow breathing index is not an accurate predictor of weaning outcome in COPD patients. Clinical judgement seems to predict successful extubation better than any of the threshold values of the f/Vt ratio that were tested. Unless more sophisticated weaning parameters are used, measuring only the f/Vt ratio in COPD patients during a T-piece trial does not seem to help in predicting weaning outcome.

A number of regional weaning centres have been set up in Europe and North America for assessment and treatment of patients who remain ventilator dependent after critical illness 1±3 . These centres provide successful weaning for 55±70 % of such patients 3 , in a less intensive and cheaper environment, and free scarce intensive care beds. This study examines the need for such a weaning centre in the Northern region of England.

A telephone survey of all the 18 intensive care units within the Northern region of England was carried out at one-week intervals to identify patients who had been ventilated for more than two weeks. A series of questions about each patient was asked in order to exclude any patient who had required the services of an intensive care unit, other than for ventilatory support, during the preceding week, or who was receiving terminal care.

During the study period of six months, 71 patients who had only received respiratory support in the preceding week were considered appropriate for a weaning facility. 26 patients (36.5 %) were female and 45 (63.5 %) were male. Their median age was 66 years (range 22 to 89 years). All but one had a tracheostomy created. During the study period, 52 (73 %) of thes patients were successfully weaned, 12 (17 %) died and 7 remained ventilated in intensive care beds at the end of the study period. If transferred to a weaning facility, eight patients (11 %) would have required readmission to an intensive care unit for further organ support. These ventilator-dependent patients occupied intensive care beds for a median duration of eight days (range 1 to 184 days) after being identified as suitable for a weaning unit. These 71 patients occupied 1040 intensive care bed-days at hospitals in the North of England, an average of 5.7 beds per day being occupied by patients suitable for a weaning unit (5.1 % of available intensive care beds in the region). One patient identified as being suitable for a weaning unit at the beginning of the study was still ventilator-dependent at the end of the study 184 days later.

CONCLUSION. Intensive care beds are an expensive and scarce resource. This study has identified a group of patients who currently occupy intensive care beds while needing only ventilatory support, and who could be cared for in a lower-intensity regional weaning unit. INTRODUCTION. We set out to determine the incidence, risk factors and the impact on the outcome of acute myopathy in a group of patients admitted to the ICU by acute exacerbation of chronic respiratory failure that received high doses of corticosteroids.

Twenty-six patients with respiratory failure that received corticosteroids intravenously were studied prospectively. All these patients required mechanical ventilation but only seven were treated with muscle relaxants. At the admission,severity of illness was evaluated by APACHE II score. Daily, laboratory tests included: glucose, sodium, potasium, phosphorus, creatinine, and creatine phospho-kinase. Serum magnesium levels were measured twice a week. Other factors analysed were: demographic data, previous corticosteroid therapy, type of medication employed and total doses, the presence of sepsis, and bacteremia. An electrophysiological study (EPS) was performed in all cases at the onset of weaning. Acute myopathy was diagnosed if small, brief and polyphaisc motor unit action potentials characteristic of a myopathic process were present. Muscular biopsy was done when EPS was consistent with myopathy. Univariate and multivariate analysis were carried out with SPSS 9.0.

RESULTS. Nine patients (34.6 %) developed acute myopathy that was confirmed by histology in four cases (permission was denied in the rest). In the multivariate analysis, risk factors for myopathy were APACHE II score at admission (OR 2.05; 95 % CI 1.13±3.78; p < 0.001) and the development of sepsis (OR 32.05; 95 % CI 2.76±170.8; p < 0.001). Myopathy prolonged significantly mechanical ventilation and increased the length of ICU and hospital stays. Three patients with myopathy (33.3 %) and three without myopathy (17.6 %) died and in the multivariate analysis, myopathy was not an independent predictor of in-hospital mortality.

CONCLUSION. Severity of illness at the admission and the development of sepsis are risk factors associated with myopathy in critically ill patients after the administration of high doses of corticosteroids. Myopathy increases length of mechanical ventilation and in-hospital stay although mortality was unaffected. (1), but not well evaluated in the specific setting of chronic obstructive pulmonary disease (COPD) patients.

METHODS. This prospective clinical study was conducted between January 1996 and January 2001 and included 112 consecutive COPD patients requiring weaning from mechanical ventilation. During weaning procedure patients were ventilated with pressure support mode (Siemens Servo 900C) at 8 cmH 2 O. f/Vt (respiratory rate/tidal volume) ratio was calculated at one and at 60 minutes. Success of extubation was defined as the absence of reintubation or need for noninvasive ventilation within the first 48 hours. Sensitivity, specificity, positive predictive and negative predictive values were calculated and then analysed with Receiver Operating Characteristic (ROC) curves. The area under the curve (AUC) at one and at 60 minutes was calculated.

The area under ROC curves were respectively 0.53 and 0.60 at one and at 60 minutes.

CONCLUSION. The f/Vt ratio, in this group of COPD ventilated patients, was not valid in predicting weaning outcome. (1). We present a computer system that reliably acquires, stores and diplays basic information necessary for clinical effectiveness and to report activity and performance of an ICU.

The system is made of five modules: a patient database (for administrative, medical and nursing data), an information module (to facilitate transmission of useful informations), a radiological module (for easy consultation of X-ray), a laboratory module (for consultation of lab results) and an archive module (for consultation of former records).

The database was created with Filemaker pro. It allows gathering and analyse of datas to report the activity of the unit (nurse workload score, medical procedures. . .) and patient caracteristics (origin and destination, physiologic scores, diagnosis, mortality. . .). Every items has been chosen with care. The information system has been created with HTML language with allows easy sharing of documents through the Intranet of the hospital. This module contains useful informations for medical and nurse staff which are updated regularly: policies, procedures, clinical guidelines, care plans, teaching program with slides when available, monthly electronic newspaper, virtual library with selection of useful articles, report of cases of interest. . .

The coordinated system has been elaborated during the last four years. The database is easily modified according to the new exigencies of the hospital administration or medical society. Guidelines are regularly updated. This computer based tool was immediately well accepted and is a cohesion tool. We present this work to show that with the limited ressources of a non universitary hospital a motivated team may create a costumized tool which meet the special needs of an ICU. CONCLUSION. Based on our results, it is possible to establish polysomnography in the ICU and to achieve a valid diagnosis of sleep disorders. Polysomnography is not yet an instrument for routine diagnostic, but can become a valuable sophisticated method of examination. Especially educated staff is needed for its use. It could be shown that sleep disorders in intensive care medicine are a relevant problem. METHODS. The period of the study was from December 1994 to March 2001 in the Hospital Espanhol in Rio de Janeiro, a 64 bed-hospital oriented to high level elective surgery, basically oncologic. We addopted the model of complete outsourcing, including all the process in the ICU. Then, it was included purchase or rent of medical equipment, all the salaries and taxes, acquisition and sell of medical matherial and drugs, to make the bill and all other activities envolved in the ICU process.

We have started 5 years ago, begining with 4 beds, followed by 8 and then 12 beds. The results have been very good. This complete model, with all process, including pharmacy and nurses, seem to us the best, although they have more complexity and need a high cost management, compared to other outsourcing models. The set up costs are also higher. During this 7 years, we have payed to the hospital 17 % of our bill and we have had a liquid profit of 6 % CONCLUSION. Outsourcing is a good option to hospitals which don't know or don't want to manage the risks of an ICU. The complete model is the best, considering major flexibitity against high costs, giving better profits and payment of better salaries, although the risks of bad managementand bankrupt are bigger. The liquid profit for the ICU managerial group are around 6 % and the hospital receive 17 % of the bill without any risk or cost. . We report a 22 year old man who was admitted with pruritus and jaundice in December 2000 for the first time. The patient was known to suffer from chronic hepatitis C for years, which had been treated unsuccessfully with interferon A and ribavirin. Two weeks before the first admission he had taken anabolic steroids (silabolin and nandrolone) by i. m. injection for body building. The clinical examination revealed an impressive icterus and numerous scratch marks. Total serum bilirubine was 10.8 mg/dl at this time, the direct bilirubin 9.71 mg/dl. There was a slight increase of transaminases and alkaline phosphatase, blood cell count and prothrombin time were normal. Liver biopsy showed a portocentral cholestasis with mild portal cell infiltration. Treatment with cetirizine, a histamine (H1) blocker, and naltrexon, an opioid antagonist, was not effective. Administration of phenobarbital for enzyme induction failed to reduce hyperbilirubinemia and to ameliorate clinical symptomes. Since itching and jaundice worsened and the bilirubin level rose to 30 mg/dl, the patient was re-admitted to hospital in January 2001. METHODS. The molecular adsorbent recirculating system (MARS) was connected with a hemofiltration device (Baxter BM 14 + BM 11). MARS contains a synthetic hydrophilic/hydrophobic domain-presenting semipermeable membrane (pore size smaller than albumin size, 100nm thick). The opposite side of this membrane is rinsed with ligandin-like proteins (albumin) as molecular adsorbents that are regenerated online using a chromatography-like recycling system [2] . Anticoagulation was performed by continuous infusion of heparin. Three MARS sessions were run: 9 hours at day 1, 18 hours at day 2 and 24 hours at day 5. Laboratory values were controled before, 6 hours after start and at the end of albumin dialysis. RESULTS. The procedure was well tolerated by the patient. There were no haemodynamic changes and no bleeding events. The patient¢s condition significantly improved from the first session. INTRODUCTION. Endotoxemia with subsequent cellular and organ failure currently still is a serious and important complication in intensive care treatment. Monophosphoryl lipid A (MPL) is a non-toxic hydrolysed derivative of lipid A, which has been suggested to have positive immunmodulatoric influences with induction of endotoxin tolerance as a prophylactic and therapeutical intervention. Aim of the study was to objectivy the effect of pretreatment with MPL to cellular metabolism after induction of endotoxaemia using in vivo microdialysis.

After approval by the local ethics committee 10 female pigs (25 + 8 kg b. w.) were challenged with endotoxin infusion (S. fried. H909). 5 animals received low dose treatment with MPL for 5 days before stert of the experiment, while a sham group (n = 5) served as control.

Haemodynamic parameters (MAP, HR, CO, SVO 2 and blood gas analysis inclusive Lactate (ABL, Radiometer/Copenhagen) were determined in intervals of 30 min during an observation period of 5 hours. Additionally microdialysis catheters (CMA 60, Microdialysis/Sweden) were inserted into the intramuscular and intrahepatic tissue to determine the interstitial concentrations of Lactate and Glycerol.

RESULTS. No significant differences regarding the haemodynamic and blood lactate parameters were observed between both groups. After preconditing with MPL interstitial lactate concentration of the muscle and hepatic tissue showed significantly lower values after 5 hours of endotoxinemia (3.6 1.1 vs. CONCLUSION. Pretreatment with the lipid A derivative MPL before Endotoxemia as prophylactic therapeutical intervention has been described before. Using microdialysis we were able to objectivy the metabolic changes in different tissues and the favorable effect of this vaccination, although no significant differences were observed regarding the clinical parameters. Microdialysis may serve as an additional guide for objectivation of potential success of therapies in critical ill patients in intensive care units. INTRODUCTION. Measurement of certain metabolites (e. g. lactate, py-ruvate, adenosine, hypoxanthine) have been suggested to be useful for estimation of the severity of illness in intensive care patients. Using the well established method of Microdialysis it is now possible to have an access to measure cellular metabolism directly in the tissue of the critical ill patient.

We present the case of a 72 year old patient at ICU suffering from left ventricular failure and acute renal failure after cardiovascular surgery with the complication of acute myocardial infarction. A microdialysis probe (CMA 60, A. Semrau/Sprockhövel/Germany) was inserted in the abdominal subcutaneous space and perfused with Ringer`s solution (2 ml/min) using a particular perfusion pump (CMA 107). Lactate, pyruvate, glycerol and glucose were determined out of the microdialysis samples hourly over a period of 8 days (CMA 600;CMA Microdialysis/Solna/Sweden). Blood lactate was measured in intervals of 3±4 hours (ABL/Radiometer Copenhagen)

RESULTS. Concommittant to acute left ventricular decompensa-tion interstitial glycerol was elevated 3 fold from 410 mmol to 1290 mmol. Interstitial lactate/pyruvate-ratio increased from 16 to 32 during this period. After cardiac recompensation by therapy with catecholamines and implantation of a intraaortal counter pulsation both parameter decrased to normal values. Under continuous veno-venous hemofiltration (CVVH) interstitial lactate (5.5 mmol/l) was approximately 2.5 fold higher than blood lactate during whole observation time.

CONCLUSION. Interstitial glycerol as a parameter for lipolysis due to adrenergic stimulation and cell membrane degradation, as well as lactate/pyruvate ratio seem to be useful parameter for ªmetabolic stressº in seriously ill patients. Blood lactate does not seem to be a valuable parameter during CVVH. Microdialysis might be a promising additional tool in the concept of multimodal monitoring for bedside-observation of patients in the ICU. 

Microdialysis is a minimal invasive method for continuous monitoring of interstitial metabolic changes. The aim of the study was to perform microdialysis of sceletal muscles in order to assess early changes in interstitial metabolism and to compare selected biochemical parameters in skeletal muscles and blood in trauma patients.

After institutional approval 12 critically ill patients with trauma requiring ventilatory support were studied immediately after admission. Microdialysis probe CMA 60 with cutoff 20 000 Da (CMA Microdialysis Solna Sweden) was inserted under ultrasound control into the left m. quadriceps femoris. After 30 minutes recovery period microdialysis procedure with Ringer lactate solution using flow rate 35 ucl/h was started. Microdialysis samples in 60 minutes intervals were collected during 72 hours.Venous blood samples were collected every 4 hours. Lactate, glucose and urea concentrations in microdialysates and venous blood were evaluated. INTRODUCTION. Regional parameters of organ dysfunction have been claimed to be better than global haemodynamic parameters. Whether there is a difference between the use of regional or global parameters during resuscitation in critically ill patients with or without hypotension is not known.

Haemodynamic optimisation was targeted using standardised resuscitation with the mean arterial pressure (MAP), heart rate (HR), wedge pressure (PCWP), cardiac output (CO), and systemic vascular resistance as parameters. After stabilisation was achieved rapid infusion bolus was initiated to evaluate whether occult haemodynamic abnormalities could be detected and corrected. During resuscitation global haemodynamic parameters were measured simultaneously with regional tonometric (PmucosalCO 2 and pHi) and indocyanine green (ICG)dilutional parameters (CBI, clearance of ICG from blood and PDR, plasma disappearance rate of ICG)and were compared between patients with and without hypotension at admission. Data presented as mean SD.

Resuscitation increased MAP (72.5 1.5 vs 80 2, p = 0.001), CVP (11 0.8 vs. 13.5 0.7, p = 0.04), and oxygen delivery index (597 30 vs. 685 49, p = 0.04) significantly. After stabilisation significant differences persisted between the patients with and without hypotension with respect to intramucosal PmCO 2 (7.3 0.7 vs. 6.1, p = 0.03), pHi (7.20 0.03 vs. 7.31 0.02, p = 0.005), CBI (407 54 vs. 795 65, p < 0.0001), and PDR 11.8 1.5 vs 20.2 1.5, p < 0.001). Patients who subsequently died had a significant lower pHi (7.24 0.03 vs. 7.31 0.02, p = 0.03), higher PmCO 2 (7.1 0.5 vs. 6.0 0.2, p = 0.02), and lower CBI (411 56 vs. 587 60, p = 0.02), PDR (p = 0.01), and RVEDVI (p = 0.05), of which the pHi, PmCO 2 , and CBI were the most important predictors of outcome. Rapid bolus infusion after stabilisation improved gastric pHi (7.24 0.02 vs 7.27 0.03, p = 0.03) and CBI (402 25 vs. 575 28, p = 0.006) significantly in patients who were hypotensive at admission.

CONCLUSION. Resuscitation at admission of critically ill patients with shock does not require monitoring of regional parameters. However, after stabilisation regional parameters were the best predictors of outcome. Further resuscitation efforts in this phase should be aimed at influencing these regional parameters. INTRODUCTION. The typical electrocardiographic pattern associated with central nervous system diseases is characterized by deeply inverted or tall T waves, prolongation of the QT interval, and prominent U waves (1,2,3). The aim of this study is to examine the kind of ECG changes in patients with intracerebral hemorrhage in ICU. METHODS. The study group was consisted of 34 patients (14 male and 20 female) of average age 65 years. The youngest was 42 and the oldest 77. There were 20 patients with haemathoma regio capsularis (11 dex. and 9 sin.), 9 with haemathoma intracerebralis reg.temporoparietalis (6 dex. and 3 sin.) and 5 with haemorrhagio subarchnoidalis. ECG changes were evaluated during the first 72 hours from receiving in the ICU and shortly before discharge. RESULTS. The most common ECG abnormalities associated with central lesions that we found were: elevated, peaked, or notched T waves in 20 patients (58,8 %) , prolongation of the Q-T interval in 19 patients (55,8 %), ST segment depression in 11 patients (32,2 %), increased P-wave amplitude in 6 patients (17,6 %), flattening or inversion of the T wave, and U waves in 5 patients (14,6 %),increased QRS voltage in 5 patients (14,6 %) and ST segment elevation in 4 patients (11,7 %). We didn't register increased Q waves. INTRODUCTION. Controversy persist regarding the optimal management of patients with Hunt and Hess (H-H) grades IV and V after aneurysmal subarachnoid hemorrhage (SAH). Although more than 90 % of these patients die if untreated, many physicians remain reluctant to aggressively treat them. However, a more aggressive management, including the early aneurysm obliteration, could decrease this high mortality (1). We show our experience with the use of early endovascular aneurysm coil embolization in this patients.

Prospective study over a 5-year period, from 1996 to 2000. All patients admitted in poor clinical grade, defined as H-H IV or V, after aneurysmal rupture were evaluated. The aim of our study was to evaluate the morbidity and mortality in the early ( < 72 h) embolizated patients. Outcome was assessed, at discharge from ICU and nowadays (by telephone interview ), using the Glasgow Outcome Scale (GOS).

One hundred and ninety one patients with SAH were admitted in this period. 61 (32 %), were categorized as poor clinical grade. 29/61 (48 %) were excluded according to different reasons: 1) arteriography was not carried out in 14 patients (6 of them met brain death criteria at admission and 8 were excluded by age, cancer and severe chronic cardiac or respiratory illness) and only one survived; 2) 15 patients could not be embolizated by technical impossibility in 12 (10 of them by aneurysm morphology or difficult access; one by severe vasospasm and another one by minimum cerebral blood flow) and negative arteriography in other 3 patients. The remaining 32/61 patients (52 %) underwent early embolization. Of them: 1) 5 died during their ICU stay (2 by multiorganic failure; 1 by vasoespasm; 1 by brain oedema and 1 by aneurysmal rupture during the endovascular treatment). 2) 27 left ICU after a mean of 212 157 hours of mechanical ventilation and a mean ICU stay of 21 8 days. At discharge from ICU, their GOS was 4±5 (4 patients), GOS 3 (14) and GOS 2 (9; one of them died during hospital stay). However, follow-up review, after hospital discharge, showed a favorable outcome (GOS 4±5) in 18/26 patients (69 %), whereas 6 patients had a GOS 3 and only 2 patients had died (none of these caused by the SAH). Overall mortality after follow-up was 25 % (8/32). To the contrary, 11/15 (73 %) not embolizated patients had an unfavorable outcome, being noteworthy that 9 died (60 %).

CONCLUSION. In our experience an agressive management, including an early ( < 72 h) coil embolization, modifies favorably the morbidity and mortality of poor grade patients with aneurysmal SAH. We suggest that an active management, including early embolization, should be considered independently of the neurological state at admission. 

We present two cases of young women, with subacute cephalea resistant to conventional analgesia, that were admitted to our Intensive Care Unit (ICU), due to an acute mental deterioration. Both had risk factors for DST: intake of oral contraceptives and habit of smoking (first case), pregnancy and thrombocytosis (second case). The diagnosis of DST ( superior sagital sinus and transverse sinuses) was made with magnetic resonance (MR) angiography and confirmed afterward with angiography. In both cases, the computed tomography (CT) scan showed diffuse brain swelling. The first patient also had a cerebellar hematoma. Intracraneal pressure (ICP) monioring was started, demostrating maintained high pressures (range 35±45 mmHg) in the 2 patients. This finding contrasted with just a moderate brain swelling in the CT scan, and with a favourable neurologic exploration. The patients received anticoagulation and local thrombolysis with urokinase (UK).

In both cases, the transfemoral infusion of UK into the thrombosed sinus, achieved a total sinus patency. The patients received UK doses ranging from 4.5 to 5 mill IU, with a mean infusion time of 110 hours. No major bleeding complications were associated with the treatment (including absence of growth of preinfusion cerebellar hematoma. Symptoms of intracraneal hypertension ceased. On discharge from th ICU, the patients had no neurologic secuelae.In both cases, the transfemoral infusion of UK into the thrombosed sinus, achieved a total sinus patency. The patients received UK doses ranging from 4.5 to 5 mill IU, with a mean infusion time of 110 hours. No major bleeding complications were associated with the treatment (including absence of growth of preinfusion cerebellar hematoma. Symptoms of intracraneal hypertension ceased. On discharge from th ICU, the patients had no neurologic secuelae.font.

CONCLUSION. Local fibrinolisis with UK associated with anticoagulation, is an effective therapy to achieve a radiological and clinical resolution of the DST. The maintained hyperPIC in these two cases of DST, does not correlate with severe brain injury. In both cases, there was a favourable neurologic exploration and radiologic imaging. INTRODUCTION. Diffuse axonal injury (DAI) has been recognized as a common cause of unfavourable outcome in patients with severe head injury (SHI). Cerebral magnetic resonance (MR) has been shown sensitive for detection and characterization of these lesions. Gentry (1) had previously defined three major anatomic areas of DAI: Lobar DAI lesions (type 1), DAI of the corpus callosum (type 2) and brain stem DAI (type 3).

We performed a prospective study between Jan 1, 1999 and Aug 31, 2000 with seventeen patients with SHI and discrepancy between the normal CT and the neurological satatus defined as: no intracranial hypertension and abnormal awake when withdrawing sedatives. MR was performed at medium of 10 days after injury to detect DAI-type lesions. Patients with DAI were divided into the three groups previously defined. We compared the MR findings with the Glasgow Outcome Scale (GOS) at 6 months.

In fifteen patients (88 %) MR demonstrated DAI-type lesions. The GCS was lower in patients with DAI-type 2 and 3(4,75 and 5 respectively) than in patients with DAI-type 1(Glasgow 7). The correlation between DAI groups and the GOS scale is shown in the following INTRODUCTION. Previous surveys on the monitoring and treatment policies of severe traumatic brain injury (STBI) in different countries have reported a substantial heterogeneity between centers. The purpose of this survey was to study the currently practiced policies of STBI management in Israel and to compare between the pediatric and adult ICUs. METHODS. A three-page questionnaire was mailed to the directors of all the units and was completed after consensus agreement in each unit. STBI was defined as a GCS < 8 after resuscitation.

During the year 2000, 21 units treated a total of 770 STBI patients of whom 150 (20 %) were children < 16 yrs. These patients were treated exclusively in pediatric ICUs. An ICP monitoring device was employed in 125 (83 %) children compared to 434/620 (70 %) adult patients. Half of the units used an intraparenchymal device, 8 (40 %) used subarachnoid bolts, 4 (20 %) used a epidural or subdural device, and 5 (25 %) used an intraventricular catheter. Five units used more than one device. An ICP > 20 mmHg was reported by 85 % of adults ICUs as the threshold for elevated ICP, while 71 % of the pediatric ICUs considered 15 mmHg as the ICP threshold. The target cerebral perfusion pressure (CPP) was < 70 mmHg in 62 % of the adult ICUs while 85 % of the pediatric ICUs aimed for a CPP of < 50 mmHg. Fifty-two percent of the units routinely used mild hyperventilation of PaCO 2 30±35 mmHg and 33 % of the units aimed for a PaCO 2 of 35±40 mmHg. In the remaining 15 % units a PaCO 2 of 25±30 mmHg was targeted. Mannitol was routinely used by all the units in all cases of documented elevated ICP. Hypothermia ( < 34C)was employed in all cases of elevated ICP in 3/7 (43 %)of the pediatric ICUs and in 3/14 (21 %)of adult ICUs. Three units (14 %) administered corticosteroids, but only in very rare circumstances ( < 10 % of cases). Barbiturates were used only in refractory intracranial hypertension (ICP > 30 mmHg). This treatment was used in 75 % of units. Muscle relaxation was employed rarely ( < 25 % of units) and exclusively for controlling ICP. The administration of antiepileptic drugs in every patient with STBI was reported in 85 % of the pediatric ICUs and in 42 % of the adult units. Three out of 7 (43 %)ediatric ICUs reported using hypertonic solutions to decrease ICP compared with none of the adult units. The option of decompressive craniectomy following STBI was performed as a last rescue option in one or two occasions in 33 % of the units during the last 10 years.

CONCLUSION. This survey demonstrates more homogeneity in the treatment of STBI than previous surveys in other countries. In Israel both pediatric and adult units employed therapies that are supported by published data. This probably results from the fact that most STBI patients in Israel are treated in tertiary care university-affiliated hospitals. In addition, pediatric ICUs are quicker to implement new modalities of treatment such as the use of hypertonic saline and hypothermia in STBI. INTRODUCTION. Critically ill patients suffering from traumatic brain injury are often difficult to monitor and treat effectively. Though far more reliable than clinical assessment or intracerebral pressure, computer tomography is always associated with considerable stress for the severely injured patient. Thus, a serum marker which is highly specific for brain trauma would be very useful to the intensivist monitoring and treating these patients. Our goal was to determine whether S-100 B could be such a marker.

METHODS. This prospective study has been underway since 1999. Presently, it includes 70 patients. Depending upon their pattern of injury, patients are assigned to one of 3 groups: Group 1: isolated traumatic brain injury ± Group 2: traumatic brain injury in combination with polytrauma ± Group 3 (controls): polytrauma without traumatic brain injury. All patients are examined by computer tomography on admittance. S-100 B is determined by venous blood drawn within the first hours after trauma and daily for a maximum of 4 weeks. S-100 B values and courses are compared to clinical and neurological findings, laboratory controls and computer tomography.

RESULTS. All patients, including those suffering from polytrauma without traumatic brain injury, show a significant increase in S-100 B during the first hours after trauma. However, survivors differ from in their further course of S-100 B. In all survivors, S-100 B decreases after the initial peak within the first 48 hours after trauma and remains normal. In all non-survivors, however, S-100 B remains elevated and/or shows a second increase at least 48 hours before death. This pattern is the most clearly visible in patients with isolated traumatic brain injury.

CONCLUSION. In our opinion, S-100 B is a very useful marker for the intensivist managing traumatic brain injury. S-100 B measurement can be repeated regularly without any stress to the patient, supplying valuable information regarding the evaluation of ongoing therapy in traumatic brain injury on the one hand and outcome on the other. 

We have reviewed our pelvic injuried patients, trying tofind the best initial therapeutic management using the Tile classification.

METHODS. Trauma patients admitted in our trauma unit between january 1995 and december 2000. Retrospective. Variables studied included age, gender, GCS, ISS, blood products requirements, ICU stay, Tile classification, associated injuries, mortality, hemodinamic state, admission hematocrit and therapeutic management. 2 of them underwent emergent anterior external fixation and an arteriography was achieved in an A3 patient (three total Tile A3 patients), with a sacral artery injury. Tile B or C group: 67 patients. 13,4 % underwent expectant management because of admission haemodynamic stability. 25,3 % initial external fixation. 6 of them (Tile C group)nedeed aditional arteriography or vascular surgery. Tile C group included 51 patients, only 44 with admission instability. 41 % nedeed an arteriography. 4 of them requiered an external fixation next to the arteriography. Laparotomy was the initial therapeutic management in a 29,8 % of Tile B or C group patients. A quarter of them required an emergently external fixation becouse of the hemodinamic instability caused by laparotomy, with a mortality rate of 75 %. All the arteriographies were positive, with evidence of ongoing bleeding.

CONCLUSION. Almost half of Tile C group unstable pelvic injuries requiered an arteriography. Associated abdominopelvic lesions without pelvic stabilization had a high mortality rate. Arteriography is an underused therapeutic approach for pelvic injuries. 

A new, noninvasive technique for the determination of central haemodynamic parameters like myocardial contractility, peripheral resistance and volume status, is the measurement of blood flow in the descending aorta using an intraoesophageal echo-doppler probe. A good correlation with cardiac output has already been demonstrated 1,2. We currently evaluate the practicability and value of this system in the setting of early inhospital trauma management. 

It is important but difficult to detect tracheal injury exactly in the patients with subcutaneous cervical emphysema. We can easily suspect respiratory tract injury from subcutaneous emphysema or deep cervical (prevertebral) emphysema in plane neck x-ray examination. Both the tracheal injury and the pulmonary injury with a tear of the parietal pleura near apex could make subcutaneous and deep cervical emphysema. However, it is difficult to detect the site of injury, trachea or peripheral lung. Bronchoscopy often miss the existence of the upper tracheal injury, especially in intubated patients. The aim of this study is to clarify the usefulness of the helical CT and 3 dimensional spiral CT (3D-CT, ªaerographyº or ª3D-tracheographyº) for diagnosis of the existence or neglect of the tracheal injury.

METHODS. Patients with blunt neck or chest trauma and having cervical emphysema (subcutaneous emphysema, deep cervical emphysema) are examined by ordinal neck and chest CT images. If we can point out any irregular low density area around the trachea or any irregularity of tracheal ring, we performed thin slice helical CT and 3D-tracheography in the next day.

RESULTS. Six cases were examined. Irregular low density area around trachea was pointed out in two cases and irregularity of tracheal ring in one of them . For these 2 cases, we performed 3D-tracheography and we were able to detect the existence and the site of tracheal injury as the defect of continuity of tracheal lumen (air). Although other 4 cases showed prominent subcutaneous or deep cervical emphysema at emergency room, they did not show any other sign of the tracheal injury, did not need any treatment for tracheal injury, and showed prompt improvement of subcutaneous and deep emphysema.

CONCLUSION. Our strategy using ordinal CT and 3D-CT (3D-tracheography) is thought to be useful for the detect the injured site of tracheal injury in the patients with blunt neck and chest trauma having subcutaneous and deep cervical emphysema. 

Windsurfing is a popular water sport, which may be prone to accidents when certain basic rules or procedures are violated. The bibliographic data on the issue are rather limited (1). Local conditions in the Aegean sea (hundreds of islands, kilometers of isolated coasts, millions of tourists, extended summer period and rapidly changing weather conditions) exacerbate the problem and quite often test the efficacy of emergency medical service.

A study of severe windsurfing accidents in Aegean Sea was performed during the period 1998±1999. As severe windsurfing accident was defined any accident that happened during windsurfing and required the transportation of the patient to a tertiary hospital. Any transportation to tertiary hospitals from the greek islands and mainland is reported, and in vast majority of cases arranged, through the Greek National Centre of Emergency Care. A questionnaire with regard to the conditions of the accident was filled in an effort to elucidate mechanisms and conditions leading to accidents.

Our study revealed 22 cases of severe accidents due to windsurfing with a wide range of injuries including head injuries, spinal cord injuries including tetraplegia and severe extremities fractures. The hypothermic water environment represented an additional aggravating factor. For 19 of these cases air-transportation was required. Prolonged hospitalization, severe disabilities, even deaths, were the consequences of these accidents. This study examined the characteristics of these patients and the possible risk factors and conditions associated with the accidents. Accidents were not associated with previous experience and proficiency of the surfer or type of surf, but were associated with poor physical condition, difficult enviromental conditions, unknown venue, sailing alone, life jacket and wetsuit absence and foot fixation to footstraps.

CONCLUSION. Windsurfing accidents represent a constant challenge for the emergency medical system. Appropriate infrastructure at all beaches popular to surfers, such as advanced life support on the spot, easy access to special transportation facilities via boats or helicopters, and timely admission to appropriate hospital for better management of their injuries could limit the health cost of this recreational activity. INTRODUCTION. Blunt thoracic trauma that leads to intracardiac defect is uncommon, and usually connected with a high-speed traffic accident. We report a case of a ventricular septal defect combined with a lesion of tricuspidal valve and a rupture of pericardium in a polytrauma patient after falling from height. METHODS. 54 y. o. healthy man fell down from the height of approximately 15 m. He had a short time loss of consciousness with his Glasgow Coma Scale being 15 thereafter. Fractures of tibia, bilateral pubic rami, sacrum and ribs with paradoxical segment on the left side were diagnosed. Besides rib fractures his chest radiograph was normal on hospital admission. C-clamp was placed on pelvis and external fixator on tibia. Anaesthesia and recovery period were uneventful and the patient was sent to the general ward. Next morning the patient had an episode of rapid desaturation down to SpO 2 70 %. Auscultation revealed a newly appeared harsh systolic murmur and his neck veins were distended. On transoesophageal echocardiography (TEE) he showed grave insufficiency of tricuspid valve with pulmonary hypertension and dilatation of the right heart. A left to right shunt was detected on the interventricular septum. As his clinical condition was deteriorating rapidly a decision of suturation of the ventricular septal defect was made. At the operation the inspection of the heart revealed no signs of myocardial contusion, but a pericardial rupture was found. After right ventriculotomy a midmuscular ventricular septal defect measuring 1,5 cm in diameter with contused / necrotic surrounding was found. Additional findings included an avulsion of the anterior tricuspid papillary muscle, valve leaflets themselves were not injured. Despite of reversal of Heparin effect and normalisation of the coagulation profile with 4 units of fresh frozen plasma (FFP) and 4 units of platelets, increased bleeding from mediastinal drains was detected, and on immediate resternotomy diffuse bleeding was seen. Total amount of postoperative mediastinal shed blood was 2000 ml. He needed inotropic support with dopamine and norepinephrine for the first 3 postoperative days.

Postoperative TEE showed intact function of tricuspid valve and no residual ventricular septal defect. The pelvic hematomas were not enlarged in comparison with preoperative finding. The main postoperative problems were associated with his lung function ± both diffuse contusion and aspiration pneumonia in the right lower lobe. After several attempts he was finally extubated on the 12 th postoperative day and in his further course he had no cardiopulmonary problems. His ICU stay was 13 days. CONCLUSION. In the first stages of making the diagnosis of intracardiac injury after blunt thoracic trauma a high degree of suspicion is needed, and a close cooperation between intensivists, cardiac and trauma surgeons should follow, to achieve maximally aggressive treatment of the injuries. Prompt clinical diagnosis aided by echocardiography, and aggressive surgical intervention for repairing the lesions resulted in complete recovery of our patient. 

To evaluate if the initial transfusion of the different blood products is associated with the appearance of mechanical ventilation acquired nosocomial pneumonia (MVANP) in severe head injury (SHI).

A prospective case-control study matched 1:1 in a Neurotraumatological Intensive Care Unit (NIVA) during a 3 years period (1998±2000).Only SHI patients(Glasgow Coma Scale_GCS-< 8)with at least 48 hour stay in the NIVA were included in the study.Case description: MVANP patient according to the CDC criteria. Control description:Patient admitted during the study period who didn`t develop MVANP. The matched variables used were: a) age( 5 years), b) APACHE II( 4), c) Injury Severity Score(ISS) ( 4), d)cerebral lesions according to the Traumatic Coma Data Bank (TCDB) and e) length of mechanical ventilation period at least equal to the period previous to the MVANP diagnose in cases. The following variables were included:Presence or not of anemia, lowest haemoglobin value in the first 48 hours, blood transfusions, being analysed global and separatedly taking into account the different blood products transfused within the first 48 hours after admission,and performing and univariant study with a p < 0.05 being considered significant. INTRODUCTION. Remifentanil hydrochloride (R) is a potent mu-opioid agonist with fast onset (ca. 1 min) of action. R is metabolised rapidly by non-specific tissue and blood esterases with a biological half-life of < 10 min (1). R acid, the major metabolite has ca. 1/4600 of R's potency. R can therefore be titrated to optimal analgesia with predictable, rapid dissipation of effects even after prolonged infusions. R supplemented with propofol (P) or midazolam (Mid) was compared with fentanyl (F) or morphine (M) (also supplemented with P or Mid) for up to 5 days in mechanically ventilated neurotrauma patients requiring daily scheduled assessments of neurological function.

Open treatment with R, F or M was randomised 2:1:1. R infusion was started at 9 mcg/kg/h (and titrated to 18 mcg/kg/h) with or without supplemental P (Days 1±3) or Mid (Days 4±5) to pre-defined optimal sedation (Sedation Agitation Scale 1 ± 3) and analgesia (None or Mild pain). F or M were administered with P or Mid according to routine practice. The primary end-point comprised the interval from start of down-titration or stopping of study drug(s) to completion of daily neurological assessment. Safety was assessed throughout. 

Hyponatremia with natriuresis is common in patients with intracranial diseases including head injury, tumours, intracranial infection, subarachnoid haemorrhage and stroke. Two major hypotheses have been proposed to explain hyponatremia in acute brain disease: the syndrome of inappropriate secretion of antidiuretic hormone (SIADH), and the cerebral salt wasting syndrome (CSWS). SIADH is characterized by high/normal blood volume, and CSWS by a low blood volume.

We investigated the incidence of these syndromes in neurosurgical patients admitted in the Intensive Care Unit, who fulfilled inclusion criteria: 1) serum sodium < 130 mEq/L, 2) serum osmolality < 280 mOsm/kg, 3) urine sodium > 20 mEq/L, 4) urine osmolality > 200 mOsm/kg, and 5) absence of renal, adrenal, thyroid, liver and cardiac failure. Control group included normonatremic neurosurgical patients. Blood volume was measured with the Cr51-labeled autologous erythrocyte technique.

RESULTS. Hyponatremia occurred 2±19 days (median 5 days) after ICU admission. Results are shown in Table ( INTRODUCTION. Candidal infection is associated with high morbidity and mortality. Some Authors have studied several antigens and candidal metabolites in the blood of patients with suspected or confirmed candidemia or hematogenously disseminated candidal infection. Our aims is to correlate the cytokines bronchial washing and plasma levels of patients with confirmed candidemia and bronchopneumonia with severe bacterial sepsis in critical ill patients and to register changes after antifungal treatment.

METHODS. Patients involved in the study are those admitted (more than 48 hours) in ICU. We have considered 18 patients subdivided into two groups: (A) 9 patients admitted with severe sepsis with bacterial bronchopneumonia and (B) 9 patients admitted with sepsis and confirmed candidemia with bronchopneumonia.We have registered inflammatory cytokines levels in bronchial washing and plasma ( IL 6, TNF alfa, IL 1, IL12) at admission and then 1, 2, 3 and 7 days after antifungal treatment.We have measured cytokines by an ELISA assay. RESULTS. Results are shown in table 1 and 2. Also the plasma levels registered in the groups confirm the different response in cytokines levels and are correlated with bronchial levels in the two groups of patients INTRODUCTION. Blood innate immunity is inhibited after head trauma and brain surgery with increased incidence of infections (1). Neuroendocrine mediators (hypothalamic-pituitary-adrenal (HPA) axis and sympathetic nervous system (SNS)) released in blood as cytokines may participate in this inhibition. This hypothesis has been tested in seizure (SZ) surgery and malignant glioblastoma (MGB) surgery.

METHODS. With Ethics approval, patients (SZ n = 5, MGB n = 3) were compared with healthy volunteers (n = 4). Protocol: preoperative (preop), 2 hrs after surgery (postop), day 1 (D1), day 2 (D2). Innate immunity study: HLA-DR expression on monocyte (MO) (flow cytometry, Mean Fluorescence Intensity (MFI)); plasma IL10 and IL12 p40 normalised by producing cells (ELISA); neuroendocrine mediators: ACTH, cortisol, epinephrine (Epi), norepinephrine (Norepi), dopamine (Dopa), prolactin (PL), arginin vasopressin (AVP) (Radio Immuno Assay). The impact of these mediators has been tested in 6-hr LPS stimulated whole blood culture (pl) and with plasma replaced by culture medium (Med, RPMI 1640 + FCS 10 %). LPS effect on MO HLA-DR, IL10 and IL12 p40 in each condition was expressed as Stim = LPS 6 hrs ± 6 hrs without LPS. RESULTS. Preop MO HLA-DR did not differ from healthy and was depressed in postop (p < 0.01), with an altered response to LPS (table 1) . Postop inhibition was attenuated in culture medium, suggesting an impact of plasma factors. This inhibition was not related to plasma IL10 and IL12 p40 modifications. After LPS-stimulation in plasma as in medium, SZ had an anti-inflammatory profile (increase in IL10 / low IL12 p40 at D1) and MGB did not respond for both IL10 and IL12 p40. CONCLUSION. Blood innate immunity downregulation after brain surgery was profound, and sustained with hypo-stimulation capabilities. Such a pattern did not correlate with plasma IL10 and IL12 p40 levels. Cytokines release after stimulation is oriented towards anti-inflammation in SZ group, not in MGB group. Epi and cortisol impact seems negligible. A potential role of AVP and PL needs further investigations. High risk of infection may be related to the observed profound downregulation of innate immunity, in relation with an anti-inflammatory profile in SZ or global hyporesponsiveness in MGB. REFERENCE. Asadullah, Crit Care Med, 1995. That's why the study of proteinase-inhibitor balance (PIB) in sepsis and septic shock is worth while.

We checked 30 healthy and 20 puerperas with postpartum sepsis and septic shock. We evaluated the following parameters of PIB: concentration of Plasminogen (%), activity of Chimotripsine Proteinases (E/ml), activity of Antichimotripsine (IE/ml), activity of Acid-stable Inhibitor (mIE/ml), activity of alfa-2-Macroglobulin (IE/ml).

We registered conciderable increase of the Plasminogen concentration (74 % worth, p < 0.05) and activity of Acid-stable Inhibitor (110 % worth, p < 0.04) in normal puerperas in comparison with donors. 20 puerperas with postpartum sepsis and septic shock received complex infusion-trabsfusion therapy and plasmapheresis. The plasmapheresis was carried out before and after sanation of the septic locusin the plasma-exchange regime, in volume 50 % and 70 % of Circulating Plasma Volume (CPV) accordingly. Retrospectively we subdivided the puerperas with sepsis and septic shock in two groups: 1 group ± 12 women, who received plasmapheresis on the 2-nd day of the disease ± all of them recovered; 2 group ± 8 puerperas the procedure was started on the 5-th day of sepsis course and with the Polyorgan's Insufficiency with the involvement of 2 and more systems, these women died on the 10±12-th day of the disease. In the 1-st group we registered gradual normalization of the PIB parameters: the mean Plasminogen concentration in the beginning was 73.6 + 13.6 %, and at the 5-th day after Plamapheresis increased by 61.8 % from the starting level (p = 0.007); the activity of Chimotripsine Proteinases decreased on 40.6 % (p = 0.08); activity of Chimotripsine had the tendency to decrease. In the 2-nd group we did not observe any changes in PIB: the starting mean Plasminogen concentration did not exceed 50 % and afterwards remained monotonously low; the activity of Chimotripsine Proteinases and Antichimotripsine did not reliably change; the activity of Acid-stable Inhibitor had a tendency to decrease immediately after plasmapheresis, but on the 5-th day after plasmapheresis it again gradually increased up to the starting level; the activity of alfa-2-Macroglobulin remained persistently low during all the period of the disease and did not exceed 3.5 IE/ml.

CONCLUSION. Correction of the PIB is possible in the first 2 days after obstetric sepsis manifestations. After Polyorgan Insufficiency formation the plasmapheresis does is not efficient and, to our opinion, is reliably connected with the increase of lethality (up to 86±100 %).

REFERENCE. Correction of the PIB is possible in the first 2 days after obstetric sepsis manifestations.

Uhlig T 1 , Brunkhorst F 1 , Bahlmann L 2 , Schmucker P 2 , Reinhart K 1 . 1 Klinik für Anaesthesiologie und Intensivtherapie, Friedrich-Schiller-Universitaet, Jena, 2 Klinik für Anesthesiologie, Medizinische Universitaet, Luebeck, Germany

INTRODUCTION. An impaired tissue oxygenation is thought to cause complications and prolonged length of ICU-stay (1). Dtetection of tissue hypoxemia is possible by global parameters (mixed-venous oxygen saturation, lactate-concentration) but also by gastric-mucosal tonometry, which is able to detect hypoperfusion in the hepato-splanchnic region. Since splanchnic hypoperfusion is thought to be an important piece of the puzzle in the developement of impaired tissue oxygenation gastric mucosal tonometry is able to predict states of organ-dysfunction. Similar results were found for the expression of procalcitonin which is thought to be a marker for states of systemic inflammation. There is evidence for a correlation between hypoperfusion and the onset of systemic inflammation. From a clinical point of view it is difficult to detect states of tissue dysoxia due to less sensitivity and specificity of markers available. In particular indicators of global haemodynamics have to be supplemented by regional and biochemical ones. Therefore the following study asked for a correlation between tonometric measurements to the expression of procalictonin regarding aspects of outcome in cardiac ICU-patients. METHODS. After IRB approval and written informed consent 98 patients admitted to the ICU after aortocoronary bypasse surgery (ASA III-IV) were studied. Beside standard monitoring each patient received a fiberoptic pulmonary catheter for continuous measurment of cardiac output and mixed-venous oxygen-saturation. As well a nasogastric catheter was introduced for semicontinuous recirculating gas-tonometry of the gastric mucosa. Documentation of haemodynamic and tonometric parameters, arterial and mixed-venous blood gas analysis, diuresis, temperature, use of volume and catecholamines followed every 60 min from admission to the ICU until transfer to the ward. Biochemical markers like CRP, Procalicitonin (PCT), Leucocytes were taken every 24 hours. For the diagnosis of splanchnic hypoperfusion the arterio-intestinal CO 2 -difference (aiDCO 2 ) was used. If a doubling of the aiDCO 2 occurs in the frist four hours of ICU-stay patients were defined as hypoperfused. Based on sociodemographic (age, sex) indicators and well-known scoring systems (ApacheII, SAPS) the sample was matched into hypoperfused (group 1) and non-hypoperfused (group 2) patients. These groups were analyzed considering haemodynamic and inflammatory aspects as well as on aspects of outcome. RESULTS. The following statistical significant results (p < 0.05) were found: Patients of group 1 had a longer length of ICU-stay. There was an increase in PCT-concentration in group 1 but nor in group 2 at the 2 nd day of ICU stay. Patients with highest PCT concentrations in group 1 showed a higher incidence of complications. No differences between groups were found in global indicators of haemodynamics and inflammation. CONCLUSION. A combination of PCT-measurments and gastric-tonometry occurs to be a suitable clinical tool for the early onset of tissue dysoxia. The pathophysiological basis of the indicators correlation has to be studied further.

(1) Int Care Med 1995; 20: 99±104

INTRODUCTION. Increased blood lactate (L) concentrations in septic shock may be due to cellular hypoxia, but can also be affected by other factors including altered cellular metabolism reduced, lactate clearance and increased glycolysis. We hypothesized that the Lactate/Pyruvate ratio (L/P), which reflects tissue hypoxia, can be increased at the onset of SS.

METHODS. We measured L, L/P and arterial blood gases in 9 patients with septic shock (SS) and 5 patients with cardiogenic shock (CS). Blood samples were obtained within 4 h after starting adrenergic agents (dopamine and/or norepinephrine), and subsequently at 4 h intervals during the first 24 h, and then 24 h later. Arterial blood samples were immediately deproteinized with perchloric acid, and L and P were determined enzymatically using a spectrophotometer. Data are presented as median (min-max).

RESULTS. Three patients died during the first 48 h of shock. 104 measurements were obtained, including 70 in 9 SS patients and 34 in 5 CS patients. Hyperlactatemia was found in 46 (66 %) SS measurements and 31 (91 %) CS measurements (p < 0.05). L/P was simultaneously elevated L/P in 5 (7 %) SS measurements and 10 (29 %) CS measurements (p < 0.05). CONCLUSION. Hyperlactatemia is commonly associated with a normal L/P ratio not only in septic, but also in cardiogenic shock.

Int Care Med 26

PLATELET COUNT ARE CORRELATED WITH OUTCOME IN PAEDIATRIC ICU PATIENTS

Weerd de W 2 , Nijsten MW 1 , Albers MJ 2 . 1 Surgical Intensive Care Unit, 2 Pediatric Intensive Care Unit

All platelet count, leukocyte count and CRP determinations performed directly before and up to 10 days after ICU-admission were analyzed. Mortality was determined at 30 days. /L respectively (p < 0.001). Remarkably, leukocyte counts were similar in both groups at day 3: 12.6 7.5 and 13.3 12.8´109/L (NS)

Both a low platelet count during the first days after ICU-admission, and independently, a subsequent failure of the platelet count to rise are associated with increased mortality. The initial leukocyte count is not associated with mortality, with only limited differences later on. Presumably persisting systemic inflammation ± as reflected by CRP ± is associated with persisting

Blunted rise in platelet count in critically ill patients is associated with worse outcome

Blood lactate concentrations are commonly increased in patients recovering from cardiopulmonary bypass (CPB) after cardiac surgery. Although this hyperlactatemia may be due to tissue hypoxia, many patients with hyperlactatemia have adequate haemodynamic parameters and an otherwise uncomplicated course

hours after the end of the surgical procedure: 5 mL were used gor a whole blood sample and 10 mL were separated with dextran to obtain a leukocyte (WBC) rich supernatant and red blood cell (RBC) pellet. WBC and RBC were resuspended in sterile 0.9 % NaCl, and glucose was added to achieve a 10 mg/dL concentration. The three samples (whole blood, WBC, and RBC) were incubated at 37 C in a water bath for 120 min, then 0.1 mg of endotoxin (LPS) was added to each sample, and the incubation continued for another 60 min. Lactate concentrations were measured at baseline and every 30 min using a blood gas analyzer (ABL 700, Radiometer, Copenhagen, DK). The lactate production of each sample was calculated in fmol/1000 cells/min using linear regression. The relative contribution of WBC to whole blood lactate production (%WBC) was also calculated. Data were compared using a Student's T test for paired data

RESULTS. * p < 0.05 LPS vs baseline, $ p < 0.05 post CPB vs pre anesthesia

In the context of new promising and other already available efficient treatment of sepsis (1), one would find of particular interest to have a biological marker, early produced and linked to the severity of the septic state, at his disposal. PCT may match the criteria. We therefore assessed the reliability of procalcitonin (PCT), a recent bacterial infection marker

Severity of sepsis was assessed by ACCP SCCM criteria (4), and PCT levels were measured once daily until day five. The number of males and females was respectively 35 and 7, mean age 56 20 years. The APACHE II Score was 18.5 5.1 at time of admission in the ICU. Twenty one were trauma patients, 9 postcardiac surgery patients, 3 neurosurgical patients and 9 medical patients. RESULTS. Infections which occurred on day 6.9 5.7 were mainly bronchopneumonia. Sepsis occurred in 23 patients, severe sepsis in 13 patients and septic shock, defined by both hypotension requiring vasoactive agent and lactate level > 2.5 mmoles/l, in 6 patients. On day one of sepsis, PCT levels were between 0.11 and 82.6 ng/ml, (mean = 5.3, median = 0.825) in sepsis, 0.23 and 9.97 ng/ml (mean = 2.7, median = 1.08 ng/ml) in severe sepsis, 1.91 and 57

This leads to the point that, with a threshold of 1 or 5 ng/ml, the sensitivity to confirm the severe sepsis state or septic shock is respectively of 79 % and 47 %, with a specificity of 48 % and 87 %. CONCLUSION. Disappointingly, procalcitonin does not appear to be a reliable marker of severe sepsis

INTRODUCTION. Multiple organ dysfunction syndrome (MODS) is believed to result from microcirculatory failure in surgical intensive care patients. We hypothesised that degree of MODS is mirrored by simple tests of microcirculatory function. Therefore, we compared reactive hyperaemia response in the forearm using transcutaneous PO 2 /PCO 2 electrodes and laser doppler velocimetry, microvascular permeability (strain-gauge plethysmography) in legs and gastric tonometrically derived variables in patients with moderate and severe MODS. METHODS. Twenty-two patients with MODS (1, moderate MODS group: less and equal 8 points, n = 13; severe MODS group: greater and equal 9 points, n = 9) were studied. All patients were monitored including a pulmonary artery catheter and a gastric tonometer during the study. Reactive hyperaemia in the forearm skin after an arterial occlusion of 5 min was investigated using a transcutaneous PO 2 /PCO 2 -electrode heated to 37 C and a laser doppler flowmeter (Periflux 4001,Perimed, Järfella, Sweden). Fluid filtration capacity and isovolumetric venous pressure were assessed using a electromechanical sensor with automated calibration for straingauge plethysmography (Filtrass 2001, Domed GmbH, Germany) . In addition, arterial lactate concentrations, arterial-, mixed venous blood gas analysis and systemic haemodynamics were measured and systemic oxygen transport variables calculated . For statistical analysis paired Student's t-test and in cases of non-normal distribution the Wilcoxon signed-ranked test was performed. P-values < 0.05 were considered significant. Data are given as means SD. RESULTS. There were no differences in age, systemic oxygen delivery, consumption and oxygen extraction ratio between groups. Mortality in patients with moderate MODS was 15.4 %, in patients with severe MODS 55.6 % (p = 0.049). Patients with high MODS demonstrated significantly higher arterial lactate concentrations (4.5 3 mmol/l) when compared with moderate MODS (1.7 0.9 mmol/l; p = 0.04).We observed no significant differences in gastric pHi, gastric regional to arterial PCO 2 difference, capillary filtration coefficient and isovolumetric venous pressure measured in legs and the magnitude of skin reactive hyperaemia response in the forearm between patients with moderate and severe MODS. . Until today there are no standardized criteria available for diagnosis of CLS. Aim of this study was to evaluate dynamic changes of intrathoracic blood volume index (ITBVI) as a diagnostic determinant for CLS.

In a prospective clinical study 6 patients with septic shock (SAPS II-Score: 48 10), multiple organ failure and CLS were compared to 6 control patients. CLS was judged clinically by generalized oedema, positive fluid balance and weight gain. Dynamic changes of following parameters were measured before (T1 = 0 min) and after (T2 = 90 min) administration of 300 ml of albumin 20 %: ITBVI calculated by the mean transit time approach (thermodye dilution technique) using a computer system (COLD-Z021, Pulsion Medical Systems, Munich, Germany), central venous pressure (CVP), and plasma volume by photometrically determination of indocyanin green concentration (PVICG), Data are presented as mean and SD. Statistical intergroup comparison was performed using unpaired t-test. (*) p < 0.05 was considered significant.

Changes of ITBVI at T2 in CLS patients were significant smaller compared to controls (-53 126 ml/m2 vs. 103 148 ml/m2; p < 0.05). CVP averaged 18 2 mmHg in CLS patients and in controls 9 2 mm Hg at T1. Change of CVP at T2 was not different in CLS patients compared to controls (0 2 mmHg vs. 2 2 mmHg; p = 0.13). At T1 PVICG was measured in CLS patients with 43.1 7.5 ml/kg/total body weight (BW) and in controls with 57.6 10.2 ml/kg/BW (p < 0.05). Change of PVICG at T2 was not different in CLS patients compared to controls (4.3 8.7 ml/kg/BW vs. 2.5 4.2 ml/kg/BW; p = 0.65).CONCLUSION. These results suggest that measurement of dynamic changes in Intrathoracic blood volume to administration of hyperoncotic solution may serve as a useful approach to diagnosis of CLS in critically ill patients. 

To date there are two methods for calculation of the cardiac chemoreflex sensitivity (CCRS, a marker of cardiorespiratory interplay): We have recently shown that the arterial CCRS (aCCRS) is the more blunted the more MODS is pronounced (1). On the other hand the venous CCRS (vCCRS) is used for risk stratification in chronic heart failure (CHF) and sudden cardiac death (2). In the current study we aimed to apply both methods to MODS patients and we asked which method is most suitable for calculation of autonomic function in MODS.METHODS. 47 consecutive patients with MODS (APACHE II Score [AP II] > or = 20) were enrolled in this study. The aCCRS was calculated as regression slope of heart interval (HI) and PaO 2 . HI and PaO 2 were assessed at baseline, after 1/3 increase of FiO 2 and after returning to baseline (1). This regression slope was also used to calculate regression slope of HI and venous PO 2 and subsequently, called vCCRSreg. The vCCRS according to (2) was calculated as the ratio of HI alteration to that in venous PO 2 after 5 minutes of increase in FiO 2 . We also calculated this ratio for arterial PO 2 (aCRSratio). METHODS. An algorithm for haemodynamic treatment in SS, which established NA as the first drug (followed by dobutamine or adrenaline as required), was followed prospectively in all SS patients from December 1999 to August 2000. We evaluated Apache II and SOFA scores, and maximum values for C-reactive protein (CRP) and lactate. Patients were classified in three groups according NA requirement: Mild shock, NA < 0.1 mg/kg/min; Moderate shock, NA from 0.1 to 0.3 mg/kg/min; and, Severe shock, NA > 0.3 mg/kg/min.

. 56 SS patients were enrolled. The three groups were comparable in age and sex, but severe SS patients had higher APACHE and SOFA scores, lactate levels and mortality (Table) .No differences were observed between the mild and moderate groups in any variable. 

Little is known about cytokine levels in patients with acute myocardial infarction(MI) complicated by cardiogenic shock (CS) [1, 2] . In some of these patients a severe systemic inflammatory response syndrome (SIRS) is observed, which is associated with a particularly poor outcome. Serum levels of tumor necrosis factor alpha (TNF-alpha), interleukin 1 receptor antagonist (IL-1Ra) and interleukin 6 (IL-6) were measured in this setting in a prospective study. The following three groups of patients were compared: 1) acute uncomplicated MI 2) acute MI complicated by cardiogenic shock 3) acute MI with cardiogenic shock and SIRS. METHODS. Twenty-eight patients admitted to our hospital with the diagnosis of acute MI were included in the study (24 men; mean age 59 10 yrs). 16 patients had an uncomplicated acute MI, 6 MI complicated by cardiogenic shock and 6 MI with CS and SIRS. All patients were subjected to acute angioplasty with angiographic success. Cytokines were measured twohourly until creatine kinase (CK) reached its peak level (starting prior to PTCA