Objective: To determine the neurodevelopmental outcome of neonates who underwent extracorporeal membrane oxygenation (ECMO group) and similarly critically ill newborns with a lower Oxygenation Index who underwent conventional treatment (comparison group), and to determine whether factors such as the underlying diagnosis and the distance transported from outlying areas affect outcome.
Design: Multicentre prospective longitudinal comparative outcome study.
Setting: An ECMO centre providing services to all of western Canada and four tertiary care neonatal follow-up clinics.
Subjects: All neonates who received treatment between February 1989 and January 1992 at the Western Canadian Regional ECMO Center and who were alive at 2 years of age; 38 (95%) of the 40 surviving ECMO-treated subjects and 26 (87%) of the 30 surviving comparison subjects were available for follow-up.
Interventions: ECMO or conventional therapy for respiratory failure.
Outcome measures: Neurodevelopmental disability (one or more of cerebral palsy, visual or hearing loss, seizures, severe cognitive disability), and mental and performance developmental indexes of the Bayley Scales of Infant Development.
Results: Six (16%) of the ECMO-treated children had neurodevelopmental disabilities at 2 years of age, as compared with 1 (4%) of the comparison subjects; the difference was not statistically significant. The mean mental developmental index (91.8 [standard deviation (SD) 19.5] v. 100.5 [SD 25.4]) and the mean performance developmental index (87.2 [SD 20.0] v. 96.4 [SD 20.9]) did not differ significantly between the ECMO group and the comparison group respectively. Among the ECMO-treated subjects those whose underlying diagnosis was sepsis had the lowest Bayley indexes, significantly lower than those whose underlying diagnosis was meconium aspiration syndrome. The distance transported did not affect outcome.
Conclusions: Neurodevelopmental disability and delay occurred in both groups. The underlying diagnosis appears to affect outcome, whereas distance transported does not. These findings support early transfer for ECMO of critically ill neonates with respiratory failure who do not respond to conventional treatment. Larger multicentre studies involving long-term follow-up are needed to confirm these findings.