Hormone Replacement Therapy and Cardiovascular Disease

Robert A. Wild, MD, MPH

The Magnitude of the Problem

The incidence of cardiovascular disease (CVD) in women increases substantially during the decades following menopause. In addition, the proportion of deaths attributable to CVD increases with increasing age. Statistics show that CVD is the leading specific cause of death for postmenopausal women, accounting for more deaths than all other causes combined.^[83,84] One of every 2 women who reaches age 50 will eventually die of heart disease or stroke; while only 1 woman in 25 will die of breast cancer.^[85] Moreover, since 1984, there has been a declining death rate from CVD in men, but mortality from CVD in women has increased.^[85]

Cardioprotective Effects

Because estrogen loss and aging are associated with a dramatic increase in CVD in women, copious research has focused on elucidating the potential cardioprotective effects of estrogen. The degree to which estrogen deficiency increases the risk of CVD in women, however, remains a subject of debate.^[86]

Many studies, including The Framingham Heart Study, have found the incidence of CVD in postmenopausal women higher than that of premenopausal women of the same age range.^[87-89] However, although the largest increase in coronary mortality in women coincides with menopause, vital statistics data do not support that menopause, apart from chronological aging, increases the risk of CVD.^[90] What seems likely is that aging and estrogen decline both contribute to the increased risk of CVD.

The increasing levels of cholesterol observed in postmenopausal women may account, in part, for an increased risk of CVD. The levels of low-density lipoprotein cholesterol (LDL-C), very-low-density lipoprotein, and other lipoprotein lipid particles increase in these women. In addition, the oxidation of LDL-C is enhanced. High-density lipoprotein cholesterol (HDL-C) levels also seem to decrease over time. The effects of aging vs estrogen deficiency on the HDL-C decline are difficult to separate.

Although approximately one quarter of estrogen's CVD protective effect is thought to be related to lipid metabolism, and although lipid metabolism and vessel wall physiology are interrelated, estrogen's direct effects on each layer of the vessel and the blood component-vessel wall interface are considered equally or more important in terms of CVD risk. After menopause, blood flow to all vessels decreases. Prostacyclin production decreases, levels of endothelin and nitric oxide increase, and levels of angiotensin-converting enzyme decrease.

Primary Prevention of Cardiovascular Disease

More than 30 observational studies (the majority but not all) have reported that ERT or cyclic HRT is associated with a reduced risk of coronary heart disease (CHD) in postmenopausal women.^{[74, 91-104]} The clinical implication of these studies is that HRT may be effective for primary prevention of CHD in menopausal women.

Long-term HRT is associated with reduced prevalence or incidence of CVD of 25% to 50% compared with no treatment.^[105-107] A variety of mechanisms through which estrogen exerts its beneficial effects on the cardiovascular system have been elucidated. These include: effects on lipoprotein metabolism and such non-lipid factors as fibrinogen, blood pressure, insulin sensitivity, effects on body fat distribution, and the protective effect of estrogen on blood vessels.^[108-113] ERT decreases LDL-C levels by about 10% and increases HDL-C levels by a similar amount.^[114] In addition, women who used estrogen had significantly less coronary artery stenosis than those who did not use estrogen.^[115]

The Postmenopausal Estrogen/Progestin Intervention Study (PEPI) found that women ages 45 to 65 years, randomized to either ERT and/or various progestin-plus-estrogen regimens, experienced similar decreases in total and LDL-C levels and increases in HDL-C, compared with women taking a placebo.^[116] The improvement in the cholesterol profile was most striking in patients treated with ERT, while patients taking estrogen plus progestins showed somewhat variable improvements. In addition, this clinical trial demonstrated that women taking estrogen, alone or in combination with progestin, had lower fibrinogen levels.

Elevated levels of plasminogen-activator inhibitor-1 (PAI-1) and tissue-type plasminogen activator, independent predictors of CHD, likely contribute to a postmenopausal woman's risk of CVD. Both molecules reflect fibrinolytic activity. A recent study^[111] found that estrogen alone or in combination with progestin reduced PAI-1 levels by approximately 50% and was associated with enhanced systemic fibrinolysis.

The data regarding the potential effect of HRT in stroke prevention in older women are less consistent. A recent study of older, postmenopausal women, with a high rate of past or current HRT use, found no evidence that HRT was associated with a lower incidence of stroke.^[117] Other studies have shown that estrogen replacement is associated with a reduced stroke mortality as well as stroke incidence in older women.^[96]

Secondary Prevention of CVD

A number of researchers have examined whether estrogen confers a protective effect on women with established coronary artery disease (CAD). Many observational studies have demonstrated that ERT/HRT use is associated with reduced risk of death and future events in women with CAD (50% to 90%).^[118-122] However, the first randomized, prospective controlled trial to examine the effects of ERT on secondary CVD outcomes -- the Heart and Estrogen/Progestin Replacement Study (HERS) -- contradicted these results.^[123,124]

The HERS trial showed no benefit of 4 years of continuous estrogen and synthetic progestin treatment in women with established coronary disease (RR = 0.99, 95% CI = 0.80-1.22). The 2700 study participants, with an average age of 67.7 years, were taking various medications to treat their CVD and had significant comorbidities.

In the HERS trial, in fact, HRT was associated with more secondary cardiac events compared with placebo in the first year of treatment. By the fourth and fifth years, there was a significant reduction in the relative risk of an additional cardiac event. As a result, although evidence supports the use of HRT in postmenopausal women for the primary prevention of CVD, further studies are needed to establish its role in secondary prevention.^[125,126]

Clinicians, however, should be wary of relying solely on the results of the HERS trial. This trial evaluated an older cohort of women with significant CVD. Only 46% of patients received statins and diet modification in addition to HRT. The majority of those taking statins did not reach target goals for lipid reduction. HRT may not be able to improve on results obtained with this type of care during the first 2 years of treatment.

In addition, the HRT regimen used in HERS trial consisted of a fixed combination of estrogen and progestin. A comparison group taking estrogen alone was not included. Because progestins are known to reduce some benefits of estrogen on the cardiovascular system, one cannot exclude the possibility that estrogen alone might be beneficial.

Another recently completed, unpublished prospective randomized trial examining the effects of HRT in women with existing CAD also tends to discount its role in secondary prevention. The Estrogen Replacement and Atherosclerosis trial is a randomized, placebo-controlled, double-blind trial evaluating the effects of ERT with or without continuous low-dose progestin vs placebo on the progression of atherosclerosis.^[127,128] The study included 309 postmenopausal women with at least 1 coronary stenosis greater than 30%. According to data presented at the American College of Cardiology Scientific Session 2000, after 3 years, there were no differences between the 3 study groups in terms of the primary outcome, minimal lumen diameter.

Potential Cardiovascular-Related Risks of HRT

Results of other studies have shown similar patterns. That is, HRT is associated with an increase in the frequency of cardiac events during early months of treatment, followed by a reduction in incidence over the long term. The Cholesterol and Recurrent Events (CARE) trial examined cardiac events in patients with existing CHD and high-LDL cholesterol.^[129] After 1 year, subjects in the pravastatin group had about a 7.5% increase in second events compared with a 5% increase in subjects treated with a placebo. After about 2.5 years of treatment, the incidence of second events in HRT users was observed.

The ongoing Women's Health Initiative (WHI), a primary prevention trial of more than 25,000 healthy women, has released similar data. Preliminary results show that during the first 2 years of treatment, there was a small increase in the number of heart attacks, strokes, and blood clots in women taking estrogen or estrogen with a progestin compared with women taking a placebo. The results, however, were not significant enough to merit cessation of the trial.

Recent observational studies have suggested that HRT causes a 2- to 4-fold increase in the risk of deep venous thrombosis and pulmonary embolism.^[130-134] This finding was supported by results of the HERS trial in which an increased risk for venous thromboembolism (VTE) among women taking HRT was noted.^[120,135] A recently published report on this data shows that women assigned to hormone therapy had a 3-fold increase in risk for thromboembolic events compared with those assigned to placebo.^[136] The relative risk was similar for women with and without other risk factors for VTE. Women with known cardiac disease are at higher risk for VTE. All the current HRT options appear to be associated with an increase in risk. Fortunately, the actuarial risk is small.

It is known that elevated levels of inflammation factors predict an increased risk of vascular disease. In an effort to elucidate a mechanism for the increased risk of cardiac events during early months of therapy, researchers examined the effects of HRT on inflammation in 365 women.^[137] Patients taking 4 different hormone regimens had large sustained increases in the concentration of C-reactive protein and a decrease in soluble E-selectin compared with patients treated with placebo (P = .0001). The final concentrations of C-reactive protein were 85% higher compared with placebo while E-selectin (which can be considered an anti-inflammatory marker) was 18% lower than the baseline level. This increase in levels of C-reactive protein might be related to the adverse early effects of estrogen therapy.

In summary, it seems prudent to avoid a fixed estrogen-progestin regimen in older women with severe CVD to attempt to improve the patient's condition in the short term. Women who are already on HRT, however, should continue with the regimen. Furthermore, clinicians should be aware of the increased risk for VTE when considering the relative benefits and risks of HRT. Women with a history of VTE, cancer, lower-extremity fracture, or immobilization should avoid postmenopausal HRT.

Do Progestins Counteract the Benefits?

But how does one "oppose" estrogen's proliferative effects on the uterus with progestin without significantly attenuating the beneficial cardiovascular effects? Progestin is known to attenuate estrogen's effect on HDL-C, blood flow, and vasomotion.^[138] In addition, progestin interferes with improved stress reactivity of estrogen and may affect carbohydrate metabolism. Some experts have suggested that the lack cardiovascular protection in the HERS trial^[122,139] was due to the use of a fixed combination of estrogen and progestin in older women with compromised cardiovascular function.

Various progestin preparations have been studied in an effort to diminish its attenuating effects. Results of the PEPI trial^[140] suggest that oral micronized progesterone is more beneficial in terms of the HDL-C profile. Also, vaginal formulations enhance uterine delivery while minimizing systemic effects.

Diabetes and HRT

Postmenopausal women with diabetes are at increased risk for CVD. Several studies suggest that HRT decreases the CVD risk in postmenopausal women with diabetes.^[141-143] For example, Kaplan and colleagues^[142] conducted a case-control study of postmenopausal estrogen use and risk of incident myocardial infarction (MI) in diabetic women. The RR of MI for current estrogen users was 0.51 (95% CI: 0.22-1.15) compared with never-users. And among current estrogen users, risk of MI tended to decline with each additional year of estrogen use.

The precise mechanism by which HRT may improve control of diabetes requires further investigation. Moderate doses of estrogen are known to increase insulin sensitivity and, therefore, should be beneficial for patients with type-2 diabetes. Some studies indicate that estrogen may decrease the production of glucose in the liver and improve pancreatic function.^[144] Again, however, data are still emerging and larger prospective trials are required to clarify the effect of estrogen on CVD.

Clinical Implications

It remains unclear whether the benefits of HRT seen in observational studies are largely the result of a process of self-selection by which healthier individuals are prescribed HRT or by other selection biases. Such biases may not only exaggerate the benefit but also underestimate the magnitude of adverse events.

Only randomization will help eliminate known and unrecognized sources of bias. Although observational studies are quite suggestive, only large, well-designed randomized trials can reliably test whether estrogen and/or estrogen plus progestin reduces the risk of CVD. Two such trials -- the Women's Health Initiative in the United States and the Women's International Study of long Duration Oestrogen after

Menopause (WISDOM) trial in the UK -- are underway and should provide results in the next 5 years.