key: cord-0813044-37n5oivr authors: Bertini, Pietro; Guarracino, Fabio; Falcone, Marco; Nardelli, Pasquale; Landoni, Giovanni; Nocci, Matteo; Paternoster, Gianluca title: ECMO in COVID-19 patients: a systematic review and meta-analysis date: 2021-11-12 journal: J Cardiothorac Vasc Anesth DOI: 10.1053/j.jvca.2021.11.006 sha: 2f41c37f32a19fd4fe8d8593ec16873488802a9e doc_id: 813044 cord_uid: 37n5oivr Objective To analyze the survival rates of patients with COVID-19 supported with ECMO and compare survival rates of COVID-19 patients supported with ECMO to Influenza patients supported with ECMO. Design We carried out a systematic review and meta-analysis to assess the impact of ECMO as supportive therapy of COVID-19. Setting We performed a search through Cochrane, EMBASE and MEDLINE/PubMed from inception to February 19, 2021 for studies reporting hospitalized COVID-19 patients managed with ECMO. Participants A total of 134 studies was selected, including 6 eligible for the comparative meta-analysis of COVID-19 vs influenza. Interventions We pooled risk ratio and random effects model. Measurements and Main Results The primary endpoint was the overall mortality of patients with COVID-19 receiving ECMO. Of the total number of 58,472 patients with COVID-19 reported, ECMO was used in 4,044 patients. The analysis suggested an overall in-hospital mortality of 39% (95% confidence interval [0.34 to 0.43]). In the comparative analysis, COVID-19 patients on ECMO had an higher risk ratio (RR) for mortality when compared to influenza patients on ECMO: 72/164 (44%) vs 71/186 (38%) RR 1.34; 95% CI [1.05 – 1.71]; p = 0.03. Conclusions ECMO could be beneficial in COVID-19 patients according to our meta-analysis. Reported mortality rate was 39%. This systematic analysis can provide clinical advice in the current era and ongoing pandemic. Setting: We performed a search through Cochrane, EMBASE and MEDLINE/PubMed from inception to February 19, 2021 for studies reporting hospitalized COVID-19 patients managed with ECMO. Participants: A total of 134 studies was selected, including 6 eligible for the comparative meta-analysis of COVID-19 vs influenza. Interventions: We pooled risk ratio and random effects model. Veno-venous (V-V) ECMO is an invasive technique that oxygenates the blood and removes CO 2 while the failing lung is put at rest and is given time to recover. Management of ECMO patients is generally performed in tertiary care referral centers as it requires expertise in treatment of refractory respiratory failure and severe ARDS 1,2 . Although reports on the use of ECMO from previous epidemics exist 1,3-10 , dedicated guidelines were produced during the COVID-19 pandemic to help triage patients in the face of reduced resources 11 . Initial ECMO guidelines for COVID-19-related ARDS were based on pre-COVID-19 trials 1,4 and ECMO was started in patients<71 years old, with severe initial presentation, and a short duration of mechanical ventilation (MV) before ECMO (i.e.,<7 or<11 days) 12, 13 . Data on ECMO efficacy in COVID-19 related ARDS is limited and comes mainly from case reports or experiences of single centers. This systematic review and meta-analysis aimed to summarize evidence from all available studies to assess the mortality of COVID-19 patients treated with V-V ECMO. This research was carried out following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 14 (Supplemental Figure 1) . Studies reporting comparison between ECMO therapy in COVID-19 and influenza-related ARDS were also searched and the findings meta-analyzed. The study protocol is available at https://www.crd.york.ac.uk/PROSPERO/ under registration number CRD42021229145. Two authors separately reviewed all potentially eligible manuscripts (title and abstract level first, full text thereafter). Disagreements were reviewed by a third reviewer, who had a deciding vote. The quality of the included studies was assessed using the Newcastle-Ottawa scale (NOS) 15 and its modification for case series and reports, according to Murad et al. 16 . The NOS was designed to judge a study by three perspectives: the study's selection, the comparability, and the determination of the outcomes. Favorable judgment was made by awarding a star. Nine stars indicate the highest strength, and six or more stars design elevated quality. The modified NOS scale by Murad and colleagues was used for case series/reports 16 , a score of 8 marks is considered maximum as items related to comparability and adjustment (which are not relevant to non-comparative studies) and retained items that focused on selection, representativeness of cases, and ascertainment of outcomes and exposure were removed. The methodological quality of eligible studies was independently assessed by two reviewers with disagreements resolved through discussion with a third reviewer. A meta-analysis of single proportions summarizing data using the inverse variance method (95% confidence interval (CI)) 17 was conducted to examine the rate of the primary and secondary outcomes in V-V ECMO treatment for COVID-19. A meta-analysis of continuous variables was performed for some of the secondary outcomes: Mean Ratio (MR) with 95% CI was calculated, and a pooled estimate, meta-MR, was computed weighting MRs according to the variance and the number of participants in the study 18 . In comparative meta-analysis our primary measure of association was the risk ratio (RR) of mortality between groups. Secondary variables included the mean difference (MD) of ECMO duration, the duration of MV before ECMO placement, peak serum creatinine concentration, and the RR of RRT. Given the diversity of studies and populations, we did not assume a common effect size and expected considerable heterogeneity. Therefore, a priori use of the random-effects model 19 was decided using the meta, metafor, and dmetar packages for R and R-Studio Version 1.3 for macOS. This model reduces the probability of type II errors. For each study, the effects estimates are presented as squares, while proportions with their 95% CI are presented as horizontal lines. The chi-square test and the I 2 were used to assess heterogeneity between the studies. Heterogeneity was classified as low (25%), moderate (50%) or high (75%) 20 . Estimation of mean and standard deviation in studies reporting only median and interquartile was accomplished using the methodology described by Hozo et al. 21 . Results were summarized using effect estimates and their associated 95% CI. Publication bias was examined by visual inspection of the funnel plot and tested by Egger's test. 22 A p-value below 0.05 was considered suspicious of publication bias. Three prospective investigations, 82 retrospective observational analyses and 49 case reports/series of COVID-19 patients on ECMO for a total 134 studies (references to the 134 studies in Supplemental Appendix 1), were selected for the systematic review and metaanalysis of single proportions or single means, and six 23-28 of them were eligible for the comparative meta-analysis of COVID-19 ECMO patients vs influenza ECMO patients. According to the NOS 15 , the quality scores of the included studies ranged from 6 to 9 for retrospective studies, indicating elevated quality as listed in Supplemental Table 1 . Supplemental Table2 shows the score modified for case series and reports 16 . A total of 4044/58472 (6.9%) COVID-19 patients received V-V ECMO ( Table 2 ). In the 77 studies reporting sex, males were 2,606/3455 (71%) ( Table 2) (Table 2) . Overall study quality was acceptable (Supplemental Table 1 , 2). The comparative meta-analysis restricted to 6 studies (350 overall patients) showed that COVID-19 patients on ECMO had an higher RR of mortality when compared to influenza patients on ECMO: 72/164 (44%) vs 71/186 (38%) RR 1.34; 95% CI (1.05 -1.71); p = 0.03; I 2 = 0%; P of heterogeneity = 0.65 (Figure 2A) . We also found a longer mechanical ventilation duration before ECMO initiation in COVID-19 compared to influenza without difference in RRT, ECMO duration, and peak serum creatinine concentration ( Figures 2B, 3, 4 ). In this extensive systematic review, we first investigated mortality in a meta-analysis of single proportions to better understand survival rates of patients with COVID-19 supported with ECMO. We identified 134 studies, of them 102 reported an overall mortality rate of 39% (1,508/3,793 patients). This mortality rate is consistent with experience from EOLIA (37%) and CESAR (37%) randomized controlled trials, performed in the pre-COVID-19 era 29 and are much lower than the initial data of COVID-19 patients requiring ECMO reported from China in early 2020 30-33 . According to this latter evidence it seems that mortality is improving over time and this is consistent to the ELSO Registry suggesting that criteria for placing COVID-19 patients on ECMO became more stringent over time favoring survivability. However, in our comparison meta-analysis, we found six studies comparing ECMO in COVID-19 to ECMO in influenza. According to the 350 patients analyzed, mortality in COVID-19 patients on ECMO was higher than in influenza patients on ECMO (RR 1.34; 95% CI [1.05 -1.71]; p = 0.03), as previously reported 34 . Still, this data is expected to vary across time as newer COVID-19 to influenza comparisons will be reported. Interestingly, 71% of patients who underwent ECMO were male. This had already been noticed and refers not to the general population of infected people, but to the severe clinical presentations as the ones requiring V-V ECMO 35 . The causes behind this gender-based unbalance remain unclear. Social, psychological, and genetic factors could all be contributing to this sex skew. Men, who are recognized in research and practice to be more impacted by cardiovascular illnesses, diabetes, chronic pulmonary disease, hypertension, and cancer, have a high incidence of disease in most situations. 36 . All of these factors have been connected to an high COVID-19 fatality rate 37 . Even if several reports included in this meta-analysis are still incomplete as patients were still on ECMO at the moment of the publication, according to our analysis, the mean ECMO duration in COVID-19 patients was 15 days (longer than the 9 days reported in the CESAR trial 38 ). This is probably to be ascribed to a different pathophysiological representation in COVID-19 compared to pneumonia and ARDS of other etiologies and involves angiogenesis, pulmonary vasculitis and thrombosis 39 . Also, the higher ECMO duration compared to influenza might indicate a more considerable pathogenicity, leading to respiratory complications and to higher mortality 40 . A mechanical ventilation duration before ECMO initiation of 4 days is longer than pre-2020 investigations, and this might underline a lack of uniformity in intubation protocols and late calls for ECMO referral centers (overwhelmed during the pandemic surge). It also may be displaying the tendency to wait more before placing a patient on ECMO tolerating lower PaO 2 /FiO 2 ratios compared to the standard ARDS care 41 . Our comparison meta-analysis further confirmed this data: mechanical ventilation pre-ECMO duration in COVID-19 was increased by 3 days (95% CI, 2.64 to 3.59; p < 0.001) vs mechanical ventilation pre-ECMO duration in influenza. Our systematic review has limitations. First of all, by pooling observational studies, this review cannot overcome the limitations of its primary studies included which were relatively small numbers, and, still, none were based on a randomized allocation. Meta-analysis of observational studies is notoriously challenging, due to heterogeneity (in subjects, outcome definitions, study design, etc), incomplete data, and bias. Also, secondary outcome measurements were missing in many studies since the focus was often mortality. However, the present systematic review and meta-analysis is relevant and may guide current practice helping clinicians to consider patients for ECMO therapy according to the current ELSO guidelines 11 , even if only by emphasizing the limitations of the available clinical evidence. It is worth noting that the present analysis reports the highest number of COVID-19 patients treated with ECMO to date. Third, duplicate publication bias might have occurred as it is challenging to detect double ECMO runs reports, especially in large cohorts extracted from international databases. Finally, the reduced mortality ECMO duration compared to the ELSO registry could be the representation of publication bias as authors tend to publish favorable outcomes with shorter runs in less sick patients and thus overestimates survivability. To this day, this systematic review included the highest number of COVID-19 patients with ECMO outcomes. Our results suggest that ECMO could be advantageous for COVID-19 patients with ARDS. The mortality rate for patient with ARDS due to COVID who received ECMO support was 39% (95% CI 34-43%). In our view, this systematic analysis of the literature can be of benefit and provide clinical advice in the current era and ongoing pandemic. Financial support: None (the study was supported by departmental funds only). Competing interest: none declared Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome Venovenous ECMO treatment, outcomes, and complications in adults according to large case series: A systematic review Extracorporeal Life Support for Adults With Respiratory Failure and Related Indications: A Review CESAR: conventional ventilatory support vs extracorporeal membrane oxygenation for severe adult respiratory failure Extracorporeal Membrane Oxygenation for 2009 Influenza A(H1N1) Acute Respiratory Distress Syndrome Extracorporeal membrane oxygenation for severe Middle East respiratory syndrome coronavirus Extracorporeal membrane oxygenation in severe acute respiratory failure. A randomized prospective study Extracorporeal membrane oxygenation (ECMO) in patients with H1N1 influenza infection: a systematic review and metaanalysis including 8 studies and 266 patients receiving ECMO Extracorporeal Membrane Oxygenation: Beyond Cardiac Surgery and Intensive Care Unit: Unconventional Uses and Future Perspectives A meta-analysis of complications and mortality of extracorporeal membrane oxygenation Extracorporeal Membrane Oxygenation for COVID-19: Updated 2021 Guidelines from the Extracorporeal Life Support Organization Advanced Pulmonary and Cardiac Support of COVID-19 Patients: Emerging Recommendations From ASAIO-a Living Working Document Saying no until the moment is right: initiating ECMO in the EOLIA era The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration Methodological quality and synthesis of case series and case reports Meta-analysis of prevalence Ratio of means for analyzing continuous outcomes in meta-analysis performed as well as mean difference methods Meta-analysis in clinical trials Cochrane Handbook for Systematic Reviews of Interventions version 6 Estimating the mean and variance from the median, range, and the size of a sample Bias in meta-analysis detected by a simple, graphical test SARS-CoV-2 versus influenza associated acute respiratory distress syndrome requiring veno-venous extracorporeal membrane oxygenation support Comparing SARS-CoV-2 and influenza A(H1N1)pdm09-infected patients requiring ECMO -A single-centre, retrospective observational cohort experience A retrospective comparison between influenza and COVID-19-associated ARDS in a Croatian tertiary care center Outcome of acute respiratory distress syndrome requiring extracorporeal membrane oxygenation in Covid-19 or influenza -a singlecenter registry study Ventilator-associated pneumonia in patients with SARS-CoV-2-associated acute respiratory distress syndrome requiring ECMO: a retrospective cohort study Comparative outcomes between COVID-19 and influenza patients placed on veno-venous extracorporeal membrane oxygenation for severe ARDS ECMO for severe ARDS: systematic review and individual patient data meta-analysis Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Comparison between Patients Hospitalized with Influenza and COVID-19 at a Tertiary Care Center Novel Coronavirus Disease (COVID-19) in Italian Patients: Gender Differences in Presentation and Severity Sex Differences in Mortality From COVID-19 Pandemic: Are Men Vulnerable and Women Protected? JACC Case Rep Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19 Comparison of the characteristics, morbidity, and mortality of COVID-19 and seasonal influenza: a nationwide, population-based retrospective cohort study COVID-19 pneumonia: different respiratory treatments for different phenotypes?