Abstract 1436

Purpose and Methods: The value of neonatal EEG recordings pre- or on ECMO (criteria of Tharp and Laboyrie, Crit Care Med 1983;11:926) for determining early school age performance in testable survivors (N=66, without cerebral palsy) was assessed by Weschler Preschool and Primary Scale of Intelligence-Revised (WPPSI-R), Wide Range Assessment of Memory and Language (WRAML), and Wide Range Achievement Test (WRAT) instruments. Those with normal or mildly abnormal EEG's (N=9) were compared to those with moderate or markedly abnormal EEG's (N=57). Respective gestational ages, birth weights and duration of ECMO were similar in both groups, (39±3 vs 40±2 weeks, 3.49±0.50 vs 3.18±0.50 kg, 141±79 vs 139±61 hours). Infants with moderate or markedly abnormal EEG's went onto ECMO later than infants with normal or mildly abnormal EEG's, (40±35 vs 71±55 hours, P=0.001).

Results: (ANOVA) (Table)

Table 1 No caption available
Full size table

Conclusions: Although serial neonatal EEG's may predict early mortality (23%) and major neurological morbidity (30% of survivors,Graziani, et al, J Pediatr 1994;125:969), the neonatal EEG pre- or on ECMO does not reflect long-term academic achievement in testable survivors at school age. Therefore, poor academic achievement at school age probably cannot be predicted in testable infants and children surviving with neonatal ECMO therapy, even when neonatal EEG's are moderate-markedly abnormal.

Funded by the National Institutes of Health: NS-27463 and NS-21453