Use of Extracorporeal Membrane Oxygenation for Adults in Cardiac Arrest (E-CPR): A Meta-Analysis of Observational Studies

Publisher Website
Google Scholar
Abstract
Published data on the use of extracorporeal membrane oxygenation (ECMO) as a supportive measure during or immediately after cardiopulmonary resuscitation (CPR) in adults (older than 18 years) shows mixed results. To assess the clinical outcomes of the use of ECMO in this modality and to look for predictors of mortality, we performed a meta-analysis (MA) of individual patients collected from observational studies. An electronic PubMed search restricted to English-language publications between 1990 and 2007, using a consensus restrictive criterion, retrieved 141 titles. After full text evaluation, 11 clinical series and nine case reports were considered appropriate and included in our MA. Data on 135 individually identified adult patients (male:female = 1.6:1) were collected. Median age for the group was 56 years (range 18-83), and the median ECMO run was 54 hours (range 0-3881). Overall survival to hospital discharge was 40% (54 of 135). The most common diagnosis leading to cardiac arrest was acute myocardial infarction (46 of 135 patients). Compared with the youngest group (17-41 years), odds ratio (OR) for mortality was higher for age group 41-56 years (OR 2.9 95%; CL, 1.6-8.2) and those older than 67 years (OR 3.4%; 95% CL, 1.2-9.7). Duration of ECMO support measured in days was also a predictor of mortality, with significant better outcome for those supported between 0.875 and 2.3 days (OR 0.2; 95% CL, 0.07-0.6). There was a negative trend in survival when manual CPR lasted >30 minutes without prompt ECMO initiation (OR 1.9; 95% CL, 0.9-4.2). This work confirms the expectations for a better survival when E-CPR is used in younger patients, for shorter periods of time and after expeditious implementation during or immediately after manual CPR. Neurologic sequelae and other major complications, although suspected to be high, are poorly described in the reviewed literature.