PURPOSE: Estrogen replacement therapy is believed by many physicians to cause thrombophlebitis and to be contraindicated in women at risk for this disease. However, clinical data supporting this assumption are scant, and further investigation is required. PATIENTS AND METHODS: We tested the estrogen-thrombophlebitis association in a case-control study. Charts of all consecutive women aged 45 years or older with a primary or secondary discharge diagnosis of thrombophlebitis, venous thrombosis, or pulmonary embolism were reviewed; 121 cases and 236 controls matched for age, year of admission, admitting service, and socioeconomic status were obtained. Hormone use and nonuse were validated in a subset of randomly selected women. RESULTS: Cases and controls, whose average age was 65 years, did not differ significantly on matching variables or on current use of exogenous estrogen (5.1% of cases versus 6.3% of controls). Other analyses that variously excluded women with a past history of thrombosis, women less than 50 years of age, women with thrombosis occurring after admission, and women whose estrogen use was indeterminate also did not support an increased risk of thrombotic disease. Adjustment for the presence of independent thrombotic risk factors did not alter the odds ratio for estrogen use. CONCLUSION: This case-control study of older women, unselected for other thrombotic risk factors, does not support the commonly held assumption that replacement estrogen increases the risk of venous thrombosis.
PURPOSE: Estrogen replacement therapy is believed by many physicians to cause thrombophlebitis and to be contraindicated in women at risk for this disease. However, clinical data supporting this assumption are scant, and further investigation is required. PATIENTS AND METHODS: We tested the estrogen-thrombophlebitis association in a case-control study. Charts of all consecutive women aged 45 years or older with a primary or secondary discharge diagnosis of thrombophlebitis, venous thrombosis, or pulmonary embolism were reviewed; 121 cases and 236 controls matched for age, year of admission, admitting service, and socioeconomic status were obtained. Hormone use and nonuse were validated in a subset of randomly selected women. RESULTS: Cases and controls, whose average age was 65 years, did not differ significantly on matching variables or on current use of exogenous estrogen (5.1% of cases versus 6.3% of controls). Other analyses that variously excluded women with a past history of thrombosis, women less than 50 years of age, women with thrombosis occurring after admission, and women whose estrogen use was indeterminate also did not support an increased risk of thrombotic disease. Adjustment for the presence of independent thrombotic risk factors did not alter the odds ratio for estrogen use. CONCLUSION: This case-control study of older women, unselected for other thrombotic risk factors, does not support the commonly held assumption that replacement estrogen increases the risk of venous thrombosis.