[Long-term follow-up of neonatal hyperbilirubinemia when phototherapy was withheld because of low bilirubin levels in the spinal fluid--growth, intelligence and soft neurological signs in school children].

The significance of cerebrospinal fluid (CSF) bilirubin levels on the outcome of neonatal jaundice has been investigated by correlations between kernicterus or brain damage and CSF bilirubin levels (Torii 1974, 1981, Meisels 1984), This study is designed to clarify the significance of CSF bilirubin levels in the long-term prognosis of neonatal hyperbilirubinemia. Sixteen full-term newborn babies with hyperbilirubinemia (serum bilirubin level 20.0-30.0 mg/dl) who did not receive phototherapy because of low bilirubin in their spinal fluid (usually under 0.5 mg/dl in normal neonates) were followed in this study of growth, intelligence and neurobehavior. The subjects' neurobehavior was compared with that of a matched control group. The physical growth of the children with a history of hyperbilirubinemia was within normal limits except for one girl whose weight was less than-2 SD from the average. Verbal IQ's were lower than performance IQ's in 10 out of 14 children with a history of hyperbilirubinemia. However, no mental retardation was noted. No significant difference was found between the two groups in mean positive scores on the Garfield test. In the Prechtl test, mean score of the finger-touching test was significantly higher in the children with a history of hyperbilirubinemia than in the control group. The soft sign coefficient and the occurrence of attention deficit hyperactivity disorder were not significantly different in the two groups. In summary, a low bilirubin level (less than 0.5 mg/dl) in the spinal fluid seems to be a good prognostic sign in infants with neonatal bilirubinemia.

Determination of haem derivatives in the cerebrospinal fluid--a semi-quantitative method.

A spectrophotometric method for semi-quantitative determination of oxyhaemoglobin, methaemoglobin and bilirubin in the cerebrospinal fluid is described and evaluated. The method involves correction for CSF protein. It is based on absorbance registrations on three wavelengths; 400, 420 and 470 nm. Reference values for a healthy control group and a hyperbilirubinaemic group are presented. Evaluation of the method shows that it is well suited to semi-quantitative determination of haem derivatives in the CSF.