Abstract
BACKGROUND: Biliary atresia (BA) is likely caused by a common phenotypic response to various triggers; one proposed trigger, cytomegalovirus (CMV), may lead to worse outcomes. The aim of this study was to determine the severity of disease and pretransplant outcomes of infants with BA, who have evidence of CMV (CMV+) at diagnosis compared with CMV-negative (CMV-) infants. METHODS: The study used data and biospecimens from the Childhood Liver Disease Research Network PROBE study of cholestatic infants. Plasma obtained at the time of hepatic portoenterostomy (HPE) of 249 infants with BA was tested for CMV by DNA-PCR and CMV-IgM. Comparisons between CMV+ and CMV- infants were made using Wilcoxon rank sum, Student t test, chi-square, or Fisher exact test. Native liver survival (NLS) outcomes were analyzed using Kaplan-Meier and Cox regression adjusting for age at HPE; pretransplant patient survival outcomes were analyzed using a competing risk model and adjusting for age at HPE. RESULTS: CMV+ infants (n = 29, 12%) underwent HPE later (67.8±13.6 d vs. 55.1±18.5 d, p = 0.0005) and had higher baseline alkaline phosphatase and aminotransferases. There was no difference between groups in jaundice clearance or NLS. The subdistribution HR of pretransplant death for CMV+ infants adjusted for age at HPE was 3.8 (p = 0.034). CONCLUSIONS: CMV infection at the time of HPE in infants with BA is not associated with worse NLS despite the association with worse liver injury, older age at HPE, and increased risk of pretransplant death adjusted for age at HPE. Continued evaluation of the consequences of CMV infection and the effects of antiviral treatment should be explored.