Recent findings from landmark clinical trials, such as the Women's Health Initiative (WHI)1,2, have significantly altered current thinking about the role of postmenopausal hormone therapy (HT) in cardiovascular disease (CVD) prevention. HT was a therapy widely accepted by women and physicians, often used for CVD prevention purposes, based on observational epidemiologic data and plausible biological mechanisms.
Prior observational studies consistently supported a 35% to 50% lower risk of coronary heart disease (CHD) with postmenopausal HT in primary prevention,3 and short-term clinical studies demonstrated favorable effects on serum lipid profiles (10% to 15% increases in high-density lipoprotein (HDL) cholesterol and comparable reductions in low-density lipoprotein (LDL) cholesterol),4 endothelial function, vascular tone, and oxidative status.5
However, until recently, little randomized clinical trial data were available to guide clinical decision-making about HT, or to assess the net benefit/risk balance of HT. Several previous studies had suggested risks such as breast cancer, venous thromboembolic events, and potential adverse effects on intermediate markers (coagulation markers, triglycerides, and inflammatory markers including C-reactive protein).5
In July 2002, the WHI,1 the first large scale primary prevention trial of HT which enrolled 16,808 healthy postmenopausal women, reported a 29% increased risk of CHD (95% confidence interval, 2% to 63%) after a mean of 5.2 years of treatment with oral conjugated equine estrogen (0.625mg daily) plus medroxyprogesterone acetate (2.5mg daily). There was a 41% increased risk of stroke in the estrogen-progestin group (95% confidence interval (CI), 7% to 85%), and more than two-fold increased risk of venous thromboembolic events. The 26% increase in risk of invasive breast cancer was of borderline statistical significance (nominal 95% CI, 0 to 59%). The benefits of HT were confined to a 37% reduction in colorectal cancer, which was nominally statistically significant, and a significant 34% reduction in hip fractures.The summary global index therefore showed a 15% increase in risk of an adverse event in the estrogen-progestin group compared with the placebo group (see Table 1).
While the absolute risks and benefits of HT in WHI were small in magnitude, over the 5.2 years of the trial, about one more in 100 women taking combined estrogen-progestin HT experienced an adverse global index event than those taking placebo (see Table 2).The net risk:benefit ratio was therefore unfavorable, particularly in the context of primary prevention of chronic diseases.
In March 2004, the estrogen-only arm of WHI was terminated.2 In this trial, 10,739 postmenopausal women aged 50 to 79 years with prior hysterectomy were randomized to oral conjugated equine estrogen (0.625mg daily) or placebo and followed for 6.8 years (mean). As shown in Table 1, estrogen alone did not appear to affect the risk of CHD, but increased the risk of stroke by 39% (95% CI, 10% to 77%). Risk of hip fracture was decreased by 39% (95% CI, 41% to 91%) with estrogen, compared with placebo. Interestingly, no increase in breast cancer risk was observed. There were also no significant differences in risks of pulmonary embolism, colon cancer, or in the global risk of an adverse event.
Several recently published clinical trials in secondary prevention have also demonstrated that in women with pre-existing coronary artery disease, combined estrogen-progestin HT was associated with an early increase in the risk of CHD events (Heart Estrogen/progestin Replacement Study, (HERS) trial),6 and no change in angiographic measures of atherosclerosis (Estrogen Replacement Angiographic trial, (ERA)).7
In the Women's Estrogen for Stroke (WEST) trial,8 compared with placebo, estradiol treatment did not affect the risk of death or nonfatal stroke. Death due to stroke (predominantly ischemic) was more common in the estradiol group (relative risk (RR) 2.9%, 95% CI 0.9% to 9.0%), although not statistically significant.
The knowledge gained from these clinical trials has transformed clinical practice and considerably weakened the role of HT in disease prevention. While HT has documented benefits (reduction in menopausal symptoms, osteoporotic fracture risk and possibly colon cancer), these are outweighed by significant risks (with combined HT, increased risk of CHD, stroke, breast cancer, venous thrombosis, with estrogen only HT, increased risk of stroke). Recent data also suggest that estrogen alone or combined with progestin may increase the risk of cognitive decline and dementia.
Therefore, given the increased risk of clinical CVD events with HT in trials of primary and secondary prevention, HT should not be initiated or continued for the purpose of preventing CVD or its complications. When using HT to treat menopausal symptoms, current recommendations emphasize the importance of limiting the duration of treatment as much as possible, given the possible risks with longer treatment, and using the lowest effective dose.
Whether any role for HT exists in CVD prevention, which may be influenced by the optimal timing of initiation of treatment, and dose, formulation, and duration of hormones used, remains to be determined. Both the WHI estrogen plus progestin and estrogen-alone trials suggest that women who initiate HT closer to the time of menopause have lower CHD risk with HT than women starting later in life; this hypothesis warrants further study.
The results of WHI and other clinical trials of HT, and their enormous impact on our current understanding of the appropriate clinical role of HT, also reinforce the importance of clinical trials to test the net benefits and risks of therapeutic interventions, and of applying well-proven methods of risk reduction in CVD prevention.