Hormone Therapy and Risk of Myocardial Infarction : A National Register Study

There was a discrepancy between data in the Women's Health Initiative (WHI) showing increased risk of coronary artery disease and stroke associated with continuous combined hormone therapy (HT), and prior observational studies suggesting that HT would have a primary preventive effect on cardiovascular disease. A previous study suggested that WHI data and observational results would be more in agreement if time since initiation of HT was controlled. Some investigators believed that the WHI data may not be applicable for healthy younger peri-menopausal women. This population-based observational study evaluated the influence of age, various regimens, routes of administration, duration of use, progestagen types, and estrogen dose on the risk of myocardial infarction (MI) associated with HT. National registry information on 698,098 healthy Danish women, aged 51-69, were followed from 1995 to 2001. Exposure to HT was recorded on a daily basis from a central prescription registry. Relative risk (RR) estimates were analyzed by Poisson regression analysis. A total of 4957 MI events were identified within the 6 year study period. Compared to women who never used HT, current use of hormones did not increase the overall risk of MI (RR, 1.03; 95% confidence interval [CI], 0.95-1.11). Among women in the youngest age group (51-54 years), however, there was an increased risk of MI (adjusted RR, 1.24; 95% CI, 1.02-1.51). The adjusted RR was 0.92 with a 95% CI of 0.80-1.06 in the older age groups. Increased risk of MI was also associated among women in the youngest age group with longer duration of HT (>4 years); this association was not found in the older age groups. The highest risk of MI among all age groups was found with the continuous combined regimen (adjusted RR, 1.35; 95% CI, 1.18-1.53). No significant risk was associated with cyclic combined therapy, unopposed estrogen, or tibolone. The risk associated with the dermal route for unopposed estrogen was significantly lower than for oral delivery (P ≤.04). No overall indication of increased risk of MI was found with increasing doses of estrogen or progestagen type. These data did not show any overall association between HT and risk of MI among older women who currently used HT; increased risk was found in younger women. Longer duration of HT also increased the risk among younger women, but not older women.