Clinical predictors of outcome following mild and moderate neonatal encephalopathy in term newborns in Kathmandu, Nepal

We describe a clinical grading system for the assessment of neonatal encephalopathy developed for a large prospective study in Kathmandu. Inter-observer variability testing of our system on 27 infants showed high agreement (kappa value 0.87). Validity for the prediction of major neuro-developmental impairment at 1 y of age was tested using a cohort of 57 survivors of encephalopathy, all of whom were assessed using a combination of the Denver Developmental Screening Test and Bailey 2 at 1 y. We compared this with a modification of a scoring system previously validated in Cape Town. Both schemes converted a pretest probability of 31% (the prevalence of major impairment at 1 y of age in this cohort) to a post-test probability of 55%. This showed only marginal improvement over the traditional risk marker of neurological abnormality at discharge (post-test probability 51%). At 6 wk of age acquired microcephaly increased the probability of major impairment to 79%. Conclusions: It seems to make little difference both in practical or predictive terms whether one describes the neurological condition of the neonate using a descriptive or scoring system. The important thing is to perform repeated systematic neurological examinations on a daily basis during the neonatal period. Many clinicians will justifiably continue to use the discharge examination as the deciding factor for the need for continued neurodevelopmental surveillance.