In numerous investigations, maternal smoking increases the risk of sudden infant death syndrome (SIDS). In the present study we investigated whether prenatal risk factors for SIDS modify the effect of maternal smoking on SIDS mortality. We analysed data from a population-based cohort study (222 cases, 260,604 infants at risk) within the Westphalian Perinatal Inquiry in Germany between 1990 and 1994. In the stratified analysis, smoking was classified into non-smoking, moderate (1-10 cigarettes/d) and heavy smoking (> 10 cigarettes/d). Multiplicative interactions between smoking and other prenatal risk factors were assessed in a logistic regression model. The relative risk (RR) for maternal smoking was 2.4 (95% confidence interval 1.7-5.4) for moderate and 7.2 (5.3, 9.7) for heavy smokers. Previous established risk factors for SIDS, such as preterm birth, low birthweight, and number of prenatal visits did not increase the risk of SIDS among non-smokers, but became important risk factors among smokers. In preterm infants (< 37 weeks) of heavy smokers, the RR was 19.6 (10.4, 36.8) compared to term infants of non-smokers. Low birthweight infants (< 2500 g) of heavy smokers had a RR of 16.3 (8.4, 31.2) compared to normal weighted infants of non-smokers. Adjustment for occupational status did not change the crude estimates. The RR of < 6 prenatal visits in the heavy smoking subgroup was 14.8 (7.2, 29.6) compared to > 9 prenatal visits in the non-smoking strata. Heavy smoking potentiates other prenatal risk factors for SIDS suggesting an increased susceptibility towards the adverse effects of tobacco smoke in utero. In infants born to non-smoking mothers, prenatal risk factors are absent and postnatal factors may be of major importance.