Bilirubin-albumin binding is important in assessing the risk of bilirubin encephalopathy in jaundiced newborns. Although exchange transfusion is recommended at specific bilirubin/albumin (B/A) ratios for babies of various birthweights and illness (Pearlman et. al. Pediatrics. 62:460, 1978), the actual unbound bilirubin (UB) concentrations at these ratios has not been determined. We therefore measured UB in 134 newborns with birthweights between 500 and 4270 g and gestational ages from 24 to 41 weeks to determine UB and its variability at reference B/A ratios. UB was measured at 288 total bilirubin concentrations and sample dilutions of 1:44 using a modified peroxidase test which allowed corrections for rate limiting dissociation of bilirubin from albumin which occurred in 88% of the UB measurements. The babies were grouped by birthweight and gestation and the equilibrium binding constants for each group determined (Klotz and Hunston. Arch Biochem Biophys 193:314, 1979). From these constants, the reference UB's and 95% confidence intervals were calculated at the reference A/B ratios. The data for gestational age are given in the table for High and Standard risk newborns as defined by Pearlman (see above reference).

Table 1
Full size table

These results show the significantly variability in UB at any given B/A ratio and suggest that actual binding measurements would be more reliable than B/A ratios in predicting risk for bilirubin toxicity.