Oral contraception and risk of a cerebral thromboembolic attack: results of a case-control study.

OBJECTIVE To assess the risk of cerebral thromboembolism in women using low dose oral contraceptives. DESIGN A retrospective case-control study. SETTING All Danish medical, neurological, neurosurgical, and gynaecological departments. SUBJECTS All 794 women in Denmark aged 15-44 who had suffered a cerebral thromboembolic attack during 1985-9 and 1588 age matched randomly selected controls. RESULTS Of 692/1584 case/control questionnaires sent out, 590/1396 (85.3%/88.1%) were returned. Among the cases, 15 refused to participate, 69 had a revised or unreliable diagnosis, 40 had had thromboembolic disease previously, 13 were pregnant, and 152 had a disease predisposing to a cerebral thromboembolic attack. Of the 323 cases without a known predisposition, 320 reported use or non-use of oral contraception. Among the 1396 controls, eight refused to participate, were mentally retarded, or lived abroad; 18 returned an uncompleted questionnaire; 17 had had thromboembolic disease previously; 31 were pregnant; and 130 had a disease predisposing to a cerebral thromboembolic attack. Thus 1198 non-predisposed controls were available, among whom 1197 reported use or non-use of oral contraception. Among the 320 cases, 116 (36.3%) were oral contraceptive users at the time of the cerebral thromboembolic attack. By comparison there were 191 users (16.0%) among the 1197 controls, giving a crude odds ratio of 3.0. After multivariate analysis, including confounder control for age, smoking, years of schooling, and trend in use of different types of oral contraceptives during 1985-90, pills containing 50 micrograms oestrogen were associated with an odds ratio for cerebral thromboembolic attack of 2.9 (95% confidence interval 1.6 to 5.4), those containing 30-40 micrograms oestrogen an odds ratio of 1.8 (1.1 to 2.9), those containing progestogen only an odds ratio of 0.9 (0.4 to 2.4). The odds ratio did not change with increasing age or with duration of oral contraceptive use. A 50% increased risk of a cerebral thromboembolic attacks among cigarette smokers (after confounder control) was independent of oral contraception status and age. CONCLUSION Low dose oral contraceptives are associated with an increased risk of cerebral thromboembolic attack. Combined or sequential pills containing 30-40 micrograms oestrogen are associated with a one third reduced risk compared with preparations containing 50 micrograms oestrogen. Progestogen only pills did not increase the risk of a cerebral thromboembolic attack.