Abstract
Objective:
Children requiring multiple consecutive extracorporeal membrane oxygenation (ECMO) runs likely have ongoing cardiac pathology (eg, residual lesions, myocardial dysfunction) and are exposed to increased complications and end-organ failure. Often, repeat back-to-back ECMO is suggested to be futile due to poor reported survival.
Methods:
Using Extracorporeal Life Support Organization (ELSO) data (2011-2019), we evaluated children (n = 669) who received multiple cardiac ECMO runs (≥2) within 30 days interval. Factors associated with hospital mortality were evaluated using multivariable regression analysis.
Results:
Median ECMO runs was 2 (range: 2-5) including 294 (44%) patients who received extracorporeal cardiopulmonary resuscitation (ECPR). There were 250 (37%) hospital survivors. Survivors were more likely older, Caucasian, and less likely to have hypoplastic left heart syndrome, require >2 runs, receive longer support duration, require inotropes or have acidosis while on ECMO, or develop renal and neurological complications. On multivariable analysis, factors associated with death included neonates (odds ratio [OR] = 3.6, 95% CI = 1.8-7.5, P = .0002), African Americans (OR = 2.7, 95% CI = 1.4-4.9, P = .0307), longer ECMO duration (OR = 1.1, 95% CI = 1.05-1.11, P < .0001, per 10 hours), central cannulation at initial run (OR = 1.7, 95% CI = 1.1-2.8, P = .0285), renal failure (OR = 3.0, 95% CI = 1.9-4.6, P < .0001), and neurological complications (OR = 3.8, 95% CI = 2.2-6.8, P < .0001).
Conclusions:
In selected children with cardiac pathology, multiple back-to-back ECMO and/or ECPR runs are associated with 37% hospital survival. Although registry data limit the ability to clearly determine selection criteria for repeat ECMO, our findings suggest that in properly selected patients, repeat ECMO support is not futile. Ongoing assessment of support adequacy, end-organ function, and cardiopulmonary recovery is necessary as longer support and emerging complications are associated with poor survival.
Keywords
Get full access to this article
View all access options for this article.