BACKGROUND: Oral oestrogen-replacement therapy (ERT) activates blood coagulation and increases the risk of venous thromboembolism (VTE) in postmenopausal women. Transdermal ERT has little effect on haemostasis, but data assessing its effect on thrombotic process are scarce. We aimed to examine the effect of the route of oestrogen administration on VTE risk. METHODS: We did a multicentre hospital-based case-control study of postmenopausal women in France. During 1999-2002, we recruited 155 consecutive cases with a first documented episode of idiopathic VTE (92 with pulmonary embolisms and 63 with deep venous thrombosis), and 381 controls matched for centre, age, and time of recruitment. FINDINGS: Overall, 32 (21%) cases and 27 (7%) controls were current users of oral ERT, whereas 30 (19%) cases and 93 (24%) controls were current users of transdermal ERT. After adjustment for potential confounding variables, the odds ratio for VTE in current users of oral and transdermal ERT compared with non-users was 3.5 (95% CI 1.8-6.8) and 0.9 (0.5-1.6), respectively. Estimated risk for VTE in current users of oral ERT compared with transdermal ERT users was 4.0 (1.9-8.3). INTERPRETATION: Oral but not transdermal ERT is associated with risk of VTE in postmenopausal women. These data suggest that transdermal ERT might be safer than oral ERT with respect to thrombotic risk.