This work analyzed the incidence of risk factors in 138 cases presenting two or more of five components defining VATERL, with no other recognized unrelated anomalies: vertebral anomalies, anal atresia, esophageal atresia with or without tracheoesophageal fistula, renal anomalies, and preaxial defects of the upper limbs, including polydactyly of the thumb. The 138 infants were ascertained among 1,811,461 births examined in the 1967-1994 period by the Latin-American Collaborative Study of Congenital Malformations: ECLAMC. One healthy and one malformed control newborn infant were matched to each VATERL case. The birth prevalence rates (per 100,000 births) for VATERL were significantly lower in Venezuela (3.1) than in the other eight countries (8.8) (P < 0.001). Venezuela also had lower rates for all five VATERL defects, even after excluding the 138 VATERL cases. VATERL cases were preferentially males (male proportion 0.6261) (P < 0.02), and, when compared with healthy controls, they had a higher perinatal mortality rate (63.7%) (P < 0.005), a higher frequency of fetal losses in previous pregnancies (12.6%) (P < 0.05), and lower mean birthweights (2,361.79 +/- 809.63 g) (P < 0.005). VATERL cases showed a higher rate than matched malformed controls for prenatal exposures to drugs and physical agents (P < 0.02 and P < 0.05, respectively), although no specific pharmacological or physical group was involved. The lower birth prevalence rates found in Venezuela, for VATERL as well as for each of the five congenital anomalies involved in this association, seem to be biologically meaningful. Since we could not identify a potential risk factor, nor a common cause of underascertainment unique to the Venezuelan subsample and common to all six hospitals, no hypothesis can be advanced here for this phenomenon. Nevertheless, this unequal geographic distribution strongly suggests a common etiopathogenicity for the five congenital anomalies involved in the VATERL association.