Biliary atresia is the leading cause of liver transplantation in children. Despite the initial description of biliary atresia over a century ago, very little is known about its onset. In the March issue of Hepatology, Matthews et al.1 raised a possible effect of environment through the assessment of decreased DNA methylation in the disrupted development of bile ducts. We take the opportunity of a 30-year cohort of all biliary atresia cases in French Polynesia to check the hypothesis of the influence of environment on biliary atresia.
Between 1979 and 2009, among the 152,945 live births in French Polynesia, 40 subjects were diagnosed with biliary atresia by a unique clinical reference center. We thus calculated a global incidence of 26.1 in 100,000 total live births, which is the highest incidence worldwide.2 Patients were of Polynesian ancestry. We ensured that the geographic distribution of births was not statistically different between patients and the total population. Birth months of the cohort and of the total population were collected over the 30 years. There was no premature birth. Radar plotting of the data clearly evidenced an unexpected shift of patient births toward July to November, which corresponds to the dry season in French Polynesia, compared to total births, which were evenly scattered (Fig. 1). Comparison of birth distribution between the dry and wet season in patients with biliary atresia versus the total population indicates a highly significant difference (P = 0.007; chi-square test).
Our observation reveals, for the first time, a significant seasonality of biliary atresia in a geographic isolate with monocentric recruitment. Previous studies with a multicentric cohort failed to demonstrate seasonality,2 although it remained debated.3, 4 Based on a whole population-based study that combines both simplified two-season climatic conditions and a population cluster, our findings cast new light to this old controversial debate. Our seasonal dynamics fits nicely with a model resulting from an infection, likely influenced by genetic background, for which the mouse models5 of biliary atresia may be regarded as paradigms. Biliary atresia may thus be considered as a dynamic disease, at least in French Polynesia. These data indicate also that the cause of biliary atresia could differ in various parts of the world.