OBJECTIVE: 

Patient selection for pediatric extracorporeal membrane oxygenation (ECMO) support has broadened over the years to include children with pre-existing neurologic morbidities. We aimed to determine the prevalence and nature of pre-ECMO neurologic disorders or disability and investigate the association between pre-ECMO neurologic disorders or disability and mortality and unfavorable neurologic outcome.

DESIGN: 

Multicenter retrospective observational cohort study.

SETTING: 

Eight hospitals reporting to the Pediatric ECMO Outcomes Registry between October 2011 and June 2019.

PATIENTS: 

Children younger than 18 years supported with venoarterial or venovenous ECMO.

INTERVENTIONS: 

The primary exposure was presence of pre-ECMO neurologic disorders or moderate-to-severe disability, defined as Pediatric Cerebral Performance Category (PCPC) or Pediatric Overall Performance Category (POPC) 3–5. The primary outcome was unfavorable outcome at hospital discharge, defined as in-hospital mortality or survival with moderate-to-severe disability (discharge PCPC 3–5 with deterioration from baseline).

MEASUREMENTS and MAIN RESULTS: 

Of 598 children included in the final cohort, 68 of 598 (11%) had a pre-ECMO neurologic disorder, 70 of 595 (12%) had a baseline PCPC 3–5, and 189 of 592 (32%) had a baseline POPC 3–5. The primary outcome of in-hospital mortality (n = 267) or survival with PCPC 3–5 with deterioration from baseline (n = 39) was observed in 306 of 598 (51%). Overall, one or more pre-ECMO neurologic disorders or disability were present in 226 of 598 children (38%) but, after adjustment for age, sex, diagnostic category, pre-ECMO cardiac arrest, and ECMO mode, were not independently associated with increased odds of unfavorable outcome (unadjusted odds ratio [OR], 1.34; 95% CI, 1.07–1.69; multivariable adjusted OR, 1.30; 95% CI, 0.92–1.82).

CONCLUSIONS: 

In this exploratory study using a multicenter pediatric ECMO registry, more than one third of children requiring ECMO support had pre-ECMO neurologic disorders or disability. However, pre-existing morbidities were not independently associated with mortality or unfavorable neurologic outcomes at hospital discharge after adjustment for diagnostic category and other covariates.