The objective of the study was to evaluate the venous impact of a progestogen-only contraception on women at high risk of venous thromboembolism (VTE). In this retrospective cohort study, 204 consecutive women at high risk of VTE were recruited between January 1992 and June 1997 and were prospectively followed. Women using chlormadinone acetate (CMA) at antigonadotropic doses (n=102) were matched by age and date of referral and history of venous thrombosis with women who had no hormonal contraception (n=102). During follow-up (mean of 33 months), nine episodes of VTE were observed: three in women receiving CMA and six in nontreated women. Using the Cox model to adjust for confounding variables such as age, thrombophilia and body mass index, the relative risk of VTE associated with the use of CMA was not significant [relative risk: 0.8 (0.2-3.9)]. These reassuring results need to be confirmed in other prospective studies.