Risk factors for fatal venous thromboembolism in young women: a case-control study.

ABSTRACT

PIP: The relationship between oral contraceptive (OC) use and fatal thromboembolism due to the recent availability of low dose OCs is reexamined. This case-control study of thromboembolism involved 60 women 16-39 years who died from the underlying case of pulmonary or venous thromboembolism between 1986-88 in England and Wales. 115 living controls were matched by age and marital status with the records of those who had died. From the medical record, information was obtained on medical history, drug history, menstrual status, obstetric history, contraceptive methods, parity, height, weight, blood pressure, and smoking habits. Conditional logistic regression was used to estimate odds ratios (OR) with the EGRET statistical package at 95% confidence intervals (CI) on 60 matched case-control sets. The results of the descriptive analysis indicate that 46 cases were venous thrombosis and 14 pulmonary. 90% of the cases were confirmed by autopsy at dissection of the veins or lungs. 44% of the cases were25 years, 27% were 25-34, and 30% were 35-39 years. 40% of the 115 controls were 25 years, 33% were 25-34, and 21% were 35-39 years. 8% of cases had a previous history of deep vein thrombosis and 7% had superficial vein thrombosis. There was no thrombosis history among controls. With a history of any venous thrombotic episode, there is an OR fatal venous thromboembolism estimated at 4.0 (95% CI, 1.4-11.5). 12% of cases and 3% of controls had had an accident or undergone an operation 3 months prior to death. The estimated OR associated with accidents or surgery within 3 months was 11.1 (95% CI, 1.3-92.5). 67% of cases and 25% of controls reported at least 1 major diagnosis. 5 of the cases but no controls were mentally retarded, and 4 had a history of epilepsy. These cases were the only ones statistically significant. 29% of the 38 cases and 40% of controls were current smokers. 1.6 (95% CI, 0.7-3.4) was the estimated OR of fatal venous thrombosis associated with current use of OCs. After controlling for the presence of other conditions, the OR rose to 2.1 (95% CI, 0.8-5.2). 15 of 26 cases were using OCs with 35 mg of estrogen, 1 with 50 mg, 10 with multiphasic pills, and 3 unknown; of the 45 controls, 24, 2, 17, and 2, respectively, used these preparations. Of the 7 cases and 3 controls having had an operation of accident 3 months prior, 4 cases but no controls used OCs. There is confirmation of increased risk due to accidents or surgery. It appears that fatal risk from OC use has been reduced; however, these data do not speak to the risk of nonfatal venous thromboembolism.