We read the meta-analysis by Long et al. in which the association between neonatal hyperbilirubinemia and UDPglucuronosyltransferase (UGT) 1A1 gene polymorphisms was analyzed, with great interest. 1 We were particularly surprised by the authors' conclusion that UGT1A1 TATA promoter polymorphisms were not associated with an increased risk of neonatal hyperbilirubinemia in Asian subjects and that results from Caucasian populations were conflicting. Our surprise emanates from the non-inclusion in the meta-analysis of a study in which the interaction between glucose-6-phosphate dehydrogenase (G-6-PD) deficiency and UGT1A1 (TA)6/(TA)7 heterozygosity or (TA)7/(TA)7 homozygosity led to a dramatic and significant increase in the incidence of neonatal hyperbilirubinemia. 2 In that study, hyperbilirubinemia was defined as serum total bilirubin (STB) Ն15.0 mg/dL (257 mmol/L) during the first week of life. DNA from term neonates born to Sephardic Jewish mothers was analyzed for the UGT1A1 TATA promoter polymorphism and for the G-6-PD Mediterranean mutation. The variant (TA)7 promoter allele frequency was similar among G-6-PDdeficient (n = 131) and normal neonates (n = 240). Overall, 30 G-6-PD-deficient neonates (22.9%) developed hyperbilirubinemia versus 22 normal neonates (9.2%; P = 0.0005). Among those normal for G-6-PD, the UGT1A1 polymorphism had no significant effect on the incidence of hyperbilirubinemia (normal homozygotes, 9.9%; heterozygotes, 6.7%; variant homozygotes, 14.7%, NS). Furthermore, of those with the normal homozygous