Prevention of Cardiovascular Disease: The Role of Hormone Replacement Therapy

During this symposium, conducted during the annual meeting of the American Society for Reproductive Medicine in Orlando, Florida, a distinguished panel presented the most recent information about cardiovascular disease and its associated modifiable and nonmodifiable markers, in regard to the role of hormone replacement therapy.

This program was supported by an educational grant from Ortho-McNeil Pharmaceutical, Inc.

Introduction

The loss of endogenous estrogen at menopause coincides with a sharp increase in the risk of cardiovascular disease (CVD) and cardiovascular deaths in women (Kramarow E et al. Health and Aging Chartbook. Hyattsville, MD. Nat’l Center for Health Statistics. 1999). Despite this, and notable within today’s environment of increased public awareness about the health risks, is the discrepancy between women’s perceptions of their cardiovascular disease risk, and the actual reality. According to a poll conducted by the American Heart Association in 1997, 61% of women thought cancer was the major cause of death (American Heart Association. CVD: Women’s Perceived Risk. 1997). In fact, CVD is the leading cause of death in women, killing one of every two women—more than all forms of cancer combined. These faulty perceptions have broad implications, from lowered vigilance to risk management, to failure to recognize symptoms of an acute coronary event, to a misapprehension of the benefits and risks of hormone replacement therapy.

Results from numerous observational studies over the past three decades have shown that postmenopausal estrogen (ERT) is associated with a number of cardiovascular benefits (Grodstein F et al. Ann Intern Med 2000;133:933-941). According to newer information from randomized trials, the concomitant use of some progestins with estrogen as hormone replacement therapy (HRT) can attenuate the positive effects of estrogen, such as lipid effects, to varying degrees, depending on the type and dose (Williams JK et al. J Am Coll Cardiol 1994;24(7):1757-61; ; JAMA 1995;273:199-208). 

The weight and meaning of accumulating evidence regarding HRT and CVD and its clinical implications served as the impetus for this symposium, with a distinguished faculty chaired by John H. Mattox, MD, Professor of Clinical Obstetrics, Gynecology, and Community Medicine at the University of Arizona. Participating as faculty were Robert D. Langer, MD, MPH, Professor of Family and Preventive Medicine at the University of California, San Diego; and Lee P. Shulman, MD, Professor and Deputy Chair of the Department of Obstetrics and Gynecology at the University of Illinois at Chicago. 
 
 
 

Estrogen as a Cardioprotective Agent

The protective effects of estrogen on the cardiovascular system extend beyond the lipid paradigm, said Dr. Lee Shulman, to “a direct, intrinsic cardiac effect—a vascular effect and an atherogenic effect.” Direct effects on the vasculature include rapid, increased vasodilation and association with increased levels of nitric oxide (White RE et al .Cardiovasc Res 2002;53(3):650-661). Over the longer-term, inhibition of vascular injury and atherosclerosis via changes in gene expression are effects mediated through steroid hormone receptors, ERa, and ERb (Karas RH et al . Proc Natl Acad Sci USA 1999;96(26):1513-6); Iafrati MD et al. Nat Med 1997;3(5):545-8). 

Positive influences on serum lipoproteins account for about 30% of estrogen’s CVD protection in the form of a decrease in LDL-cholesterol, and an increase in HDL-cholesterol which is estimated to confer a 3-5% decrease in morbidity and mortality from cardiovascular disease (Levitt JI. Arch Intern Med 2002;162(1):108; Discussion 109-10). Estrogen also promotes a decrease in damaging lipoprotein(a) (Godsland IF. Fertil Steril 2001;75(5):898-915). 

Beyond its vascular and lipid effects, estrogen improves carbohydrate metabolism (and may have a positive impact in women with diabetes) (Darko DA et al. Diabetes Res Clin Pract 2001;54(3): 157-64). It decreases platelet aggregation and exerts an antioxidant effect (Bar J et al. Haemost 2000;84(4):695-700; Tranquilli AL et al. Gynecol Endocrinol 1995;9(2):137-41). 

The effect of estrogen on the progression of subclinical atherosclerosis in healthy postmenopausal women without preexisting CVD was investigated in a randomized, controlled study, The Estrogen in the Prevention of Atherosclerosis Trial (EPAT). This 2-year study demonstrated that 17b-estradiol (1 mg day) significantly reduced the progression of subclinical atherosclerosis, versus placebo. The effects were assessed by measuring changes in intima-media thicknesses of the common carotid artery using ultrasound (Annals Intern Med 2001, 135: 939-953). 

In a recently published trial, the association between HRT and longitudinal changes in blood pressure were evaluated in a normotensive postmenopausal population (N=226) (Scuteri A et al. Ann Intern Med 2001;21(135): 229-38.Patients participating in the Baltimore Longitudinal Study of Aging were followed for 5.7+/-5.3 years. Lifestyle variables, blood pressure, and cardiovascular risk factors were measured at baseline and approximately every two years thereafter in women taking HRT (n=77) and women using neither hormone (n=129). The investigators found that, over time, average systolic blood pressure increased less in HRT users than in non-users, independent of other cardiovascular risk factors (Figure 1). Diastolic blood pressure did not change statistically in either group. 

In one retrospective observational study of 2,268 women who were followed for 10 years after undergoing coronary angiography, estrogen use was associated with increased survival rates compared to survival rates of never-users of estrogen (Sullivan JM et al. Arch Intern Med 1990;150:2557-2562). Likewise with a retrospective analysis of women who had survived a first myocardial infarction and took ERT: these individuals had a reduced relative risk for both reinfarction and death, versus non-users (Newton KM et al. J Epidemiol 1997;145:269-277). 

More is Learned About Progestins

Looking back at results from over 40 observational studies of hormone replacement therapy conducted over the past 30 years, Dr. Robert D. Langer pointed to a trend of reduced CVD protection as treatment regimens evolved from estrogen only to combination regimens of estrogen plus a progestin. 

Initial groundbreaking research characterized the differing cardiovascular effects of HRT regimens and different progestins used an oophorectomized cynomolgus monkey model (Clarkson, Adams, Wagner. Wake Forest University). In these studies, animals were given different forms of HRT, after which researchers evaluated specific cardiac vascular parameters. They found that the estrogen-related benefits of improved vascular reactivity of the coronary artery and reduction in coronary plaque formation were attenuated by the administration of medroxyprogesterone acetate (MPA), but not by progesterone (Williams et al. J Am Clin Cardiol 1994; 24:1757-1761). Furthermore, it appeared that physiological levels of 17b-estradiol (E2) plus progesterone protected the animals against chemically induced vasospasm (Miyagawa K et al. Nat Med 1997;3:324-327). E2 plus the progestin MPA did not confer such protection. In vitro studies that followed identified other differences among progestins related to aortic endothelial cell-mediated LDL oxidation (Zhu X-D et al. Atherosclerosis 2000;148:31-41). Levonorgestrel and MPA, for example, were found to oppose the cell-protective antioxidant effects of estrogen significantly more than norgestimate (NGM) or progesterone.

Dr. Langer pointed out that the Postmenopausal Estrogen/Progestin Interventions (PEPI) trial is, to date “the only completed long-term clinical trial of commonly used HRT regimens in healthy postmenopausal women.” The objective of PEPI was to compare changes in lipid levels over a 3-year interval among healthy postmenopausal women receiving HRT vs placebo. Since HDL-C is the strongest lipid predictor of coronary risk in women, it was selected as the primary outcome.(The Writing Group for the PEPI trial. JAMA 1995;273:199-208). Patients were randomized to receive either placebo, conjugated equine estrogen 0.625 mg (CEE) alone, CEE plus cyclic micronized progesterone 200 mg (MP), CEE plus continuous MPA 2.5 mg, or CEE plus cyclic MPA 10 mg. Its findings were that HDL-C improved in all active treatment groups versus placebo, but the unopposed CEE and CEE plus MP arms had significantly greater increases in HDL-C than either the CEE plus continuous MPA or the CEE plus cyclic MPA (P<.04). Overall, the PEPI trial demonstrated that MPA attenuated the estrogen benefit, while micronized progesterone did not. (All active treatment arms showed similar decreases from baseline in LDL-C levels and increases from baseline in triglyceride levels.)

Recent trials of shorter duration have tested newer HRT regimens offering potentially fewer compromises to the benefits of estrogen. For example, norgestimate offers a promising alternative to other progestins. It has been shown to be nearly identical to progesterone in binding affinity for both the progesterone and androgen receptors (Phillips A et al. Contraception. 1990; 41:399-410). One randomized trial compared the regimen of E2 1 mg plus NGM 90 µg, in a 3-days-off 3-days-on, regimen (available as Ortho Prefest) with E2 1 mg alone. Patients in both arms of this study had beneficial increases from baseline in HDL-C and decreases in LDL-C(Lobo RA et al. Am J Obstet Gynecol 2000;182:41-49). Dr. Langer commented that “…the HDL benefit with norgestimate was strikingly similar to results for progesterone in other studies, in its lack of attenuation of the beneficial estradiol effect on lipids.” Furthermore, he said, the significant increase from baseline in triglyceride levels associated with unopposed E2 was minimized in the E2 1 mg plus NGM 90 µg arm to an insignificant increase from baseline. In a subset analysis of lipid changes, more than 50% of women in the E2 1 mg plus NGM 90 µg arm with unfavorable baseline lipid levels (based on the National Cholesterol Education Program [NCEP] Guidelines for cholesterol management/ATP III) shifted into the desired range during the study for each lipid marker that was assessed (Figure 2).

Another progestin, norethindrone acetate (NETA) has recently become available as part of combination regimens. In a one- year trial E2 2 mg plus NETA 1 mg was compared with E2 1 mg/NGM 90 µg . In women who took NETA, HDL-C levels were significantly decreased from baseline. HDL increased in women on NGM. The importance of this difference is underscored by the fact that the E2 dose in the NETA regimen was twice that generally used in the US. (Ylikorkala O, et al. Clin Ther 2000;22:622-636).

Other studies have tested regimens of CEE and MPA besides those evaluated in the PEPI trial, e.g., 0.45 mg CEE + 1.5 mg MPA. Modest increases in HDL-C from baseline were seen, but these were lower than with unopposed CEE or CEE plus micronized progesterone. No CEE/MPA regimen diminishes the increase in triglycerides associated with CEE (Data on file. Wyeth-Ayerst Company. Lobo RA et al. Fertility Steril 2001;76:13-24). 

A summary of findings of lipid impacts with HRT regimens currently under study and available as therapy reveals that only E2 1 mg plus NGM 90 µg has been shown to exert positive effects on all three lipid parameters: (Lobo RA et al. Am J Obstet Gynecol 2000;182:41-49). 

Dr. Langer concluded by noting, as did Dr. Shulman, that studies with HRT have in the past overwhelmingly focused on single progestin. Dr. Langer predicted, “We should expect to see differences in the hard outcomes—true cardiovascular outcomes—if we use some of the less attenuating or potentially more beneficial progestogens.” 
  

Patient Management: A Review of the Evidence

Of the three major prospective studies of ERT/HRT in the prevention of CVD, two have evaluated secondary prevention in women with established cardiovascular disease. Only one has been designed to investigate the potential for HRT to reduce primary CVD risk. 

The large HERS study, enrolling 2,763 women with established coronary heart disease, had as its primary endpoint nonfatal MI or death from coronary heart disease. Its initial results were disappointing, with an effectively null impact from the HRT intervention arm of CEE 0.625 mcg + MPA 2.5 mg (Hulley S et al. JAMA 1998;280:605-613). But a caveat is in order in regard to the initial HERS data, Dr. Mattox said. The results were based on data from the first year of the study. However, a protective effect from the HRT arm has been emerging in subsequent years 2, 3, 4 and 5; and additional data are to be published as HERS 2. It is thought that these data will show a downward trend in numbers of second CVD events in women remaining on the HRT arm of the trial for the longer-term (Hulley S et al. JAMA 1998;280:605-613). 

The ERA (Estrogen Replacement in Atherosclerosis) study examined the effects of ERT/HRT on the progression of coronary disease, versus placebo. Subjects were postmenopausal women with at least 1 coronary stenosis of >30%, randomized to CEE 0.625 mcg alone; CEE 0.625 mcg + MPA 2.5 mg; or placebo (Herrington DM et al. N Engl J Med 2000;343:622-629). The hope of the ERA investigators was to see a change in vessel diameter that would reflect a modification of atherogenesis. But after an average follow-up of 3.2 years, no differences were seen in vessel diameter incidence of CV events, and there were no difference in results from intent-to-treat or on-treatment analyses. The relationship of this outcome to clinical morbidity was unclear. Whether MPA attenuated the beneficial effect of estrogen is a question warranting further study.

A third important prospective trial is the very sizable (over 27,000 participants) Women’s Health Initiative (WHI), with over 27,000 participants. WHI differs from HERS and ERA in that it is a primary prevention trial, enrolling women at average risk for cardiovascular disease. Treatment arms were: CEE 0.625 in women who had undergone a hysterectomy, CEE 0.625 + MPA in women with an intact uterus, or placebo. In a 2-year interim safety report, an increase in events in active treatment groups was observed. It is important to note that the events were combined events (CHD, stroke, and TE), and that as a preliminary report, it focused on safety. It did not indicate final findings. The study’s safety monitoring board endorsed continuation of the study as the numbers of adverse outcomes were less than originally predicted from the study design.

When looking to these landmark trials for clinical guidance, Dr Mattox urged the audience to consider the important fact that both HERS and ERA were secondary prevention studies, and that subjects were older than women typically considered for initiation of HRT (HERS mean age, 67; ERA mean age, 66). Another consideration, raised by other presenters, is the implication of use of a single HRT regimen, when it is known now that there are differences in compounds that have different effects on the vascular system.

Based on newer data and trends that are currently emerging, Dr. Mattox said that the effects of HRT on CVD will likely be seen to differ in several ways:

• By compound (both estrogen and progestin)
• By extent of CVD at initiation of HRT
• By type of event (CHD, stroke, TE)
• By time since the start of HRT (considering initiation of use, early phase, and long-term use)

The American Heart Association (AHA) recommendations on HRT and CVD state that HRT should not be initiated for women with existing heart disease (secondary prevention). For primary prevention, AHA states that there are currently insufficient data to assert that HRT should be initiated for the sole purpose of primary prevention of CVD. Finally, the AHA recommends that HRT initiation and continuation should be based on established non-coronary benefits and risks; possible coronary benefits and risks; and patient preference (Mosca L et al. Circulation 2001;104:499-503).

The North American Menopause Society (NAMS), in an official response to the AHA recommendations, stressed the distinction between HRT use for primary prevention (in healthy women) and secondary prevention in women with pre-existing cardiovascular disease, and asserted that the data do not, in fact, contradict the use of HRT for primary prevention (The North American Menopause Society. [Press Release]. July 27, 2001).

The American College of Obstetricians and Gynecologists (ACOG) is currently reviewing all relevant information on the issue of HRT and CVD (American College of Obstetricians and Gynecologists [Press Release]. July 26, 2001). Its recommendations to come, Dr. Mattox stated, will be based on “much of the data presented this evening.” 

Dr. Mattox proposed that patient management guidelines take into consideration all of the known benefits of HRT in addition to the current Adult Treatment Panel III (ATP III) guidelines of the National Cholesterol Education Program (NCEP) for management of high LDL levels with a statin, and consider the effects of HRT on lipid profiles. 

All contents Copyright © 1999 - 2002 Medical Association Communications