BACKGROUND: Case reports and epidemiological studies have suggested a relationship between congenital anomalies and childhood cancer, but some potential associations remain inconsistent. In this study, we investigated the association between congenital anomalies and neuroblastoma. PROCEDURE: We used data of a case-control study on neuroblastoma conducted from 1992 to 1994, including 538 children aged 0-19 years with newly diagnosed, histologically confirmed neuroblastoma and 504 controls identified by telephone random-digit dialing and matched to cases on date of birth. Information on congenital anomalies and potential confounding factors was collected through maternal telephone interviews using a structured questionnaire. We estimated odds ratios (OR) and 95% confidence intervals (CI), adjusted for reference age at diagnosis, mother's educational level, mother's race, and household income at birth. RESULTS: An association between the maternal report of any congenital anomalies and neuroblastoma (OR = 2.58; CI = 1.57-4.25) was observed. Neuroblastoma risk increased with increasing number of anomalies per child (OR = 3.90, CI = 1.27-11.9 for two anomalies or more), and when we restricted analyses to major anomalies (OR = 7.53, CI = 2.23-25.5). Genitourinary anomalies (OR = 5.84, CI = 1.67-20.4) and cardiac anomalies (OR = 4.27, CI = 1.22-15.0) had an elevated, but imprecise neuroblastoma risk. CONCLUSIONS: Our findings support the hypothesis of an association between neuroblastoma and congenital, especially urogenital and cardiac, anomalies.