Studies in hemolysis in glucose-6-phosphate dehydrogenase-deficient African American neonates

Background: The role of hemolysis in the mechanism and prediction of hyperbilirubinemia was contrasted between glucose-6-phosphate dehydrogenase (G-6-PD)-deficient and -normal African American neonates. Methods: Corrected end tidal carbon monoxide (ETCOc) values from the subset of male neonates born to non-smoking African American mothers, drawn from a previously published study, were analyzed. The relationship between ETCOc and bilirubin values, the latter represented as percentiles on the hour of life specific bilirubin nomogram, was determined. Hyperbilirubinemia was defined as any bilirubin value ≥95th percentile for hour of life. Results: 18.6% of 59 G-6-PD-deficient neonates developed hyperbilirubinemia, compared with 7.5% of 362 controls (relative risk 2.50, 95% confidence interval 1.31 to 4.76). As reported, ETCOc values (median, interquartile range) were significantly higher among G-6-PD-deficient neonates than controls (2.4 [2.0-2.9] vs. 2.1 [1.7-2.5] ppm, p < 0.001. However, higher ETCOc values were limited to those G-6-PD-deficient neonates with lower bilirubin percentiles: among those whose bilirubin value did not exceed the 95th percentile ETCOc was 2.30 [2.00-2.85] vs. 2.00 [1.70-2.40] ppm in controls, p = 0.001. In contrast, among the hyperbilirubinemic neonates ETCOc values were similar between G-6-PD-deficient neonates and controls: 2.7 [2.03-3.33] vs. 2.6 [2.33-3.45] ppm, p = 0.9. In the G-6-PD-deficient neonates ETCOc ≥75th percentile contributed no additional predictive value for hyperbilirubinemia (likelihood ratio 1.8). Conclusions: G-6-PD-deficient African American neonates have increased hemolysis and increased rate of hyperbilirubinemia, but the hemolysis is neither a predominant factor in the pathogenesis of hyperbilirubinemia nor is it predictive of hyperbilirubinemia, over and above the already increased risk conferred by G-6-PD deficiency.