Without question, hormone replacement therapy has become the most controversial issue in menopausal gynecology. The following overview of this issue was written by Anthony A. Luciano, M.D. (a founding member of the Womens' Surgery Group).
Womens' Health Initiative (WHI) Perspective
After our usual Saturday morning match, my tennis partner, Rafael, asked if I still prescribed estrogen and progestin therapy (EPT) for menopausal women, given the recent WHI findings of increased cardiovascular disease, stroke, and breast cancer associated with its use. "Of course", I said. In fact, most of my symptomatic postmenopausal patients continue on estrogen or EPT to control their menopausal symptoms. In addition, my wife Mary continues to take EPT. Before he asked, I emphatically declared that I very much loved my wife of almost 30 years. But I certainly understood his concern regarding EPT. In fact, shortly after the WHI findings became widely publicized, my wife, like several of her friends and many of my patients, discontinued EPT. But after 8-10 weeks of not taking hormones, just like several other symptomatic postmenopausal women, Mary resumed therapy because of difficulty tolerating the vasomotor symptoms and the associated disturbances of sleep patterns, moods, etc. Several other postmenopausal women, however, were able to cope with the transient recurrence of menopausal symptoms, and I was happy to encourage them to permanently discontinue ET/EPT, since the original reason for starting the therapy (vasomotor symptoms) was no longer an issue and the secondary, long-term benefits of EPT had been seriously challenged by WHI data.
Although the WHI has significantly affected our clinical practice regarding the long-term therapy with EPT, it has not precluded its use in symptomatic postmenopausal women, whose distressing symptoms and quality of life are dramaticaly improved by EPT.
Overview and implications of the WHI data
The WHI study was designed in 1991-1992 to determine the role of long-term therapy with either unopposed estrogen (ET) or combined estrogen and progestin (EPT) in postmenopausal women’s health. For the evaluation of long-term ET, the study enrolled 10,739 hysterectomized women who were randomized to either conjugated equine estrogen (CEE), marketed as Premarin, 0.625 mg daily or placebo. For the evaluation of long-term EPT, it enrolled 16,608 asymptomatic women with intact uterus, aged 50 to 79 years, who were randomized to either CEE 0.625 mg plus medroxyprogesterone acetate (MPA) 2.5 mg, marketed as Prempro, or placebo.
The primary focus of the trial was to quantify the reduction in the incidence of cardiovascular disease events versus the increased incidence of breast cancer associated with the use of ET/EPT. Secondary additional benefits of EPT included the reduction of osteoporosis related fractures, colon cancer and overall mortality, versus the increased incidence of thromboembolic events, endometrial cancer and other adverse reactions. From these data they would calculate total benefits and risks and reach a final conclusion regarding the overall safety of long term ET/ EPT use.
It is important to emphasize that the WHI study did not evaluate the benefits of ET/EPT on vasomotor symptoms, night sweats, and vaginal dryness, which are the most common reasons for their use. The goal of the study was to determine whether or not ET/EPT should be considered for long term use by postmenopausal women to prevent cardiovascular disease, stroke, fractures, colon cancer, and overall mortality. Although the study was designed for 8-year duration, assessment of the data was carried out semiannually by an independent Data Safety Monitoring Board (DSMB) to determine trends.
Results of the EPT (Estrogen + Progestrone) Arm of WHI Study
During the 10th interim analysis of the data from the EPT arm of the study (May 31, 2002), the DSMB found that the adverse effects in cardiovascular disease originally noted during the fifth interim analysis in 1999 persisted. Moreover, the increased incidence of breast cancer had crossed the designated boundary and the global index. This supported a finding of overall harm for ETP. The DSMB concluded that the evidence of harm for breast cancer, coronary heart disease, stroke and thromboembolic events outweighed the evidence of benefit for fractures and colorectal cancer. Because the overall risks of ETP outweighed its benefits, on July 9, 2002, after an average of 5.2 years of observation, the study arm evaluating the combined EPT versus placebo was halted.
It is important to notice that the arm of the WHI study designed to examine the use of unopposed estrogen, CEE 0.625 mg daily, was not terminated.
When the EPT arm of the study was halted, participants using the study drug, Prempro, had an increased incidence of 26% of breast cancer, 41% of stroke, 29% ofcoronary heart disease events, and 110% of thromboembolic events. On the benefits side, they experienced a 37% reduction in colorectal cancer and 34% reduction in hip fractures. But these reductions were not sufficient to offset the increased risk. It is important to point out that there was no difference in mortality rates between the EPT and the placebo groups.
Because relative risks are confusing and often misunderstood, the results were also reported in absolute values, expressed in the number of additional events for treated women-years. Thus among 10,000 women taking EPT for a year, there will be 8 more cases of invasive breast cancer, 8 more strokes, 8 more pulmonary embolic events, and 7 more myocardial infarctions, but 6 fewer colorectal cancers and 5 fewer hip fractures. Although the absolute risk for an individual woman remains quite small (<0.1% per year), when counting all events over the 5.2 years of the trial, the number of women experiencing a global index event was about 100 more per 10,000 women taking EPT than in those taking placebo.
Extrapolating to the population at large to a longer treatment duration, the use of EPT could account for tens of thousands additional adverse events. Because of this, the Writing Group for the WHI concluded that EPT should not be prescribed for the purpose of preventing cardiovascular disease, stroke, colon cancer or fractures.
These results do not preclude prescribing EPT for the purpose of alleviating vasomotor symptoms, night sweats, vaginal dryness and other disturbing symptoms that affect a large number of menopausal women. Indeed, the ability of ET/EPT to relieve vasomotor symptoms remains unquestioned and unequalled among all currently available medical or alternative therapies.
The WHI did not assess vasomotor symptoms relief; in fact, women with severe hot flashes were excluded from the study. Consequently, the major benefits and the most common reasons for prescribing EPT were not considered in the risk- benefit ratio and in the assessment of the global index. Had this been done, and had the study included symptomatic women with significant hot flashes, it is likely that the risk-benefit ratio might have yielded different results and perhaps different conclusion. Nevertheless, the data from the WHI study does provide valuable information, which allows us to discuss benefits and risks of ETP more precisely and more objectively, especially for asymptomatic postmenopausal women.
Impact of WHI on clinical practice
Prior to the WHI results becoming public, the most common reasons for prescribing ET/EPT were vasomotor symptoms, night sweats, vaginal dryness and mood swings associated with menopause. Secondary reasons for encouraging postmenopausal women to use EPT included the protective effects on cardiovascular disease, osteoporosis and perhaps Alzheimer’s disease. Of course, the risks of breast cancer and thromboembolic events were also considered, but we believed that “they paled in comparison to the purported benefits”. The data from the EPT arm of the WHI study has dramatically changed the secondary reasons for prescribing EPT. In fact what used to be considered secondary benefits of EPT (cardiovascular disease and stroke) must now be considered additional risks of EPT, which should further discourage women from its long-term use.
Prior to the WHI data, the cardiovascular protective effect of ET/EPT had been substantiated by several epidemiological studies, as well as experimental and clinical studies, which consistently reported the beneficial effects of ET/EPT on lipid profile and directly on vessel walls in decreasing arteriosclerosis and inducing vasodilatation. Were all these studies erroneous or is there some reason to explain the differences in the findings before and after the WHI study? It has been stated that observational studies have pitfall selection bias and compliance bias. It has been repeatedly stated that women who take hormones are typically healthier, better educated and wealthier and have fewer coronary risk factors. Consequently, the observed favorable outcome may reflect more their healthier status than their hormone therapy. Although these confounding factors might have played a role in exaggerating the benefits, it is unlikely that they would have masked the increased risk currently observed with the WHI and other recently published prospectively randomized studies ( HERS 4 and WEST 5 studies). Moreover, the other benefits (osteoporosis and colon cancer protection) and risks (thromboembolic events and breast cancer) previously reported in observational studies have been appropriately confirmed by the prospectively randomized studies. Only the cardiovascular disease risks have not been confirmed, instead they have been completely negated. Why?
Although there is no clear answer, it may be explained by the significant improvement in cardiovascular health care that women have experienced during the past 2 decades, which might have obscured the potential cardiovascular benefits of EPT. Indeed, 20 and 30 years ago, cardiovascular disease was thought to be primarily a disease of men. Very few women were evaluated or treated for hypertension and/or hyperlipidemia, two major cardiovascular disease risk factors. In these women, ET/EPT might have been therapeutic, indirectly by improving lipid profile and directly by decreasing plaque formation and by inducing vasodilatation through the regulation of nitric oxide, prostaglandin synthetase and membrane ionic permeability.
During the past 15 years, women’s heath care has changed dramatically, so that women with hypertension or hyperlipidemia, especially those who participate in clinical trials, are commonly being treated with potent lipid lowering agents and antihypertensive medications. The cardiovascular protective effects of these medicines would obscure any potential vascular benefits of ET/EPT. Consequently only the adverse vascular effects of ET/EPT would be noted, since it’s potential beneficial effects have already been usurped by the stronger cardiotropic statins and antihypertensive drugs.
In addition, had the patients at risk for fracture been treated with biphosphonates, I doubt very much that the protective effects of ET/EPT on fractures would have been noted. This hypothesis is supported by the recent publication of the Estrogen in the Prevention of Atherosclerosis Trial (EPAT). In this prospective study, 222 postmenopausal women with LDL cholesterol levels in excess of 130 were randomized to receive either 17-beta estradiol or placebo and were angiographically monitored for progression of coronary atherosclerosis. The results revealed that the women randomized to 17-beta estradiol experienced significantly less progression of atherosclerosis than placebo group. However, in those study patients who were also taking lipid-lowering therapy (statins), 17-beta estradiol therapy did not confer additional benefit on coronary atherosclerosis. It appears then that for cardiovascular protection, ET/EPT served its purpose in women’s health when there was a need for it, prior to the widespread use of cardioprotective agents.
Currently, with the vastly improved women’s health care, the role of ET/EPT in cardiovascular protection is neither needed nor safe. Indeed it is risky. The primary reasons for prescribing EPT remain the same, but the secondary gains are no longer there, because for most women the risks are greater. The WHI data are very useful in allowing us to quantify these risks.
The fact that the estrogen-only arm of the WHI study continues unabated, supports the hypothesis that the progestin in the EPT preparation may be mostly responsible for the increased risk of breast cancer as well as atherosclerosis, coronary heart disease, and stroke. Moreover, some may argue that MPA may be particularly important for the cardiovascular risks and that a different progestin, such as micronized progesterone or norethindrone acetate, may not confer similar risks. Indeed the PEPI study found that the addition of MPA to CEE in postmenopausal women negated some of the beneficial effects of CEE on the lipid profile, whereas the addition of micronized progesterone did not.
Moreover, animal studies have found that MPA but neither NETA nor micronized progesterone negated the beneficial effects of estrogen on coronary plaque formation in animal studies. It appears that different progestins may have variable effects on atherosclerosis, and that among the various clinically available progestins, MPA may be particularly deleterious to cardiovascular health. However, the results of these animal studies may not apply to humans but may serve as hypothesis that need to be tested in prospectively randomized human clinical trials. Whether the WHI EPT findings can be generalized to other hormone preparations is unclear. For the present, the burden of proof for greater safety and efficacy lies with the sponsors of the other products.
Recommendations for clinical practice
In response to the WHI data, the American College of Obstetricians and Gynecologists (ACOG) issued guidelines regarding the administration of ET/EPT and the management of menopausal women symptomatic and or at risk for osteoporosis, breast cancer and cardiovascular disease. I believe these guidelines are appropriate and very useful given the current knowledge of ET/EPT in women’s health. The ACOG recommendations and my comments are listed below.
Individualize treatment. Women will tolerate and respond differently to different preparations and doses. I tend to use natural estrogens and progesterone as much as possible because I can easily measure serum levels which help me assess patients’ response and compliance. I especially prefer the transdermal preparation, because it is more physiologic, yields constant serum levels throughout the 24-hour day and avoids first pass through the liver, which is associated with non physiologic alteration of hepatic enzymes and proteins, especially binding globulins for steroids, thyroid and sex hormones as well as coagulation factors.
I use the lowest dose that will achieve benefits and avoid side effects. This is a very important recommendation. It is true of all therapies that we use, but it is especially important for ET/EPT, which can result in life threatening side effects. In a prospective, observational study conducted in healthy postmenopausal women for the primary prevention of cardiovascular disease, it was reported that the relative risk of stroke increased progressively with each incremental increase in the dose of unopposed CEE. The relative risks (confidence intervals) of stroke with CEE at the daily dose of 0.3 mg, 0.625 mg, and 1.25 mg were 0.54 (0.28-1.06), 1.35 (1.08-1.68), and 1.63 (1.18-2.26) respectively. Consequently, one should start with the lowest dose (0.3 mg for CCE, 0.5 mg for oral estradiol , 0.035mg for transdermal estradiol). This dose may be progressively increased according to the patient’s response but seldom should exceed double the starting dose. Similar guidelines should be followed for progestins.
Use ET for women without a uterus and EPT with a uterus. The importance of using EPT in women with a uterus is well known to all gynecologists. However, the optimal preparation and administration of HRT remains elusive. It is clear by now that not all progestins have the same physiologic effects on the body, especially on cardiovascular health. The type, sequence, dose and duration of progestin administration should be carefully reevaluated.
Originally, when it became evident that unopposed estrogen caused endometrial cancer in women with a uterus, progestins (usually in the form of MPA) was administered for 7, 10, or 12 days each month, mimicking the cyclic production of progesterone by the premenopausal ovary. Although effective in protecting against endometrial cancer, the cyclic administration of EPT induced undesirable uterine bleeding which caused many women to stop EPT. To avoid cyclic bleeding and to improve compliance, the continuous and combined EPT preparations were introduced with prolonged exposure to progestin. Since it appears from the WHI study that progestins may be primarily responsible for the increased risks of CVD, stroke and breast cancer, it may be prudent to minimize the duration of progestin treatment in postmenopausal women. Several studies have demonstrated that the cyclic administration of progestin (two weeks every 2-3 months) may be as protective against endometrial cancer as the monthly administration. Moreover, if the minimally effective dose of both daily estrogen and cyclic progestin (every 2-3 months) are used, most women will not experience significant bleeding.
Long-term use of continuous combined, fixed dose HRT in asymptomatic women should be stopped. Asymptomatic women do not need (nor will they benefit) from ET/EPT and there is no need for them to use it. The potential benefit of EPT on bone loss and fractures can be duplicated with other medications specifically designed for this purpose. The biphosphonates (alondranate and risedronate) are safe, effective, convenient (once per week dose) and not associated with life threatening side effects. Raloxifene (a selective estrogen receptor modulator) is also a good alternative for the prevention and treatment of osteoporosis in patients not at risk for thromboembolic events. Unlike ET/EPT, raloxifene is not associated with an increased risk of either breast cancer or cardiovascular events. In fact, data from the MORE study suggest that, besides protecting against bone loss and vertebral fractures, raloxifene may be protective against the risk of breast cancer by as much as 75%. Raloxifene was found not to increase risk or events of cardiovascular disease.
Reassess the benefit-risk ratio with each patient annually. Annual visits should include a review of the reasons for taking ET/EPT and the current knowledge regarding the risk benefit ratio of this therapy. It is quite possible that when all the data from the WHI studies are available and reassessed, the benefit-risk ratio may change and consequently our recommendations may be different. Currently I assess the benefit-risk ratio by explaining the data as it is now, and I remind the patient that the most important (if not the only) reason for taking ET/EPT is the control of menopausal symptoms. To find out whether the menopausal symptoms have resolved or have become more tolerable, I invite the patient to discontinue the therapy for a period of 4-8 weeks. If the symptoms recur and are intolerable, or if in the patient’s mind the benefit risk ratio is positive, we resume therapy. If it is negative, we stop and consider alternatives. Our role should be to help the patient arrive at her own conclusion regarding her individual benefit-risk ratio, based on the scientific data we provide and the symptoms that she experiences.
Until the recent publication of the WHI study, women initiated and continued ET/EPT for a number of reasons, including vasomotor symptoms and related quality of life issues (sleeping, mood, vaginal dryness), osteoporosis prevention and treatment, and cardiovascular protection. With the new knowledge from the WHI study, we now know that EPT confers no cardiovascular disease protection but, instead, carries additional risk.
Although ET/EPT has been shown to be protective against osteoporosis and related fractures, the associated risk of cardiovascular and thromboembolic events and breast cancer preclude its use for prevention of osteoporosis in favor of safer alternatives, such as raloxifene and biphosphonates. Raloxifene may be especially attractive to older postmenopausal women not at risk for thromboembolism or vasomotor symptoms, because of its associated protective effects on breast cancer risk and potential benefits on cardiovascular disease.
Currently, the only indications for EPT in menopausal women are vasomotor symptoms and associated quality of life issues. When used for these reasons, women should be advised to take the lowest effective dose of these medications for as short a time as possible. I favor estradiol either orally or transdermally, starting with daily doses of 0.5 or .05 mg respectively. If symptoms are not adequately controlled and serum estradiol levels are less than 50 pcg/ml, I increase the daily doses to .75mg and .075 mg, up to a maximum of 1 mg and .100 mg respectively. For the progestin preparation, I recommend micronized progesterone orally, 200 mg at bedtime for 2 weeks every 1-3 months. If patients cannot tolerate micronized progesterone, I recommend norethindrone acetate in combination with estradiol either orally (Activella) or transdermally (Combipatch).
The duration of therapy must be individualized and reassessed yearly, according to the current knowledge of the risks and benefits and the patients’ symptoms. This may require periodic interruption of the therapy to assess the patient's symptoms. Our responsibility is to provide patients with current scientific knowledge of the risks and benefits of EPT. Each postmenopausal woman should use this knowledge (in combination with her menopausal symptoms and response to HRT) to determine whether or not to continue or initiate therapy. This approach, which requires ongoing dialogue between physician and patient, empowers each postmenopausal woman to actively participate with and take responsibility for her own health care.
1. Writing group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women’s Health Initiative randomized control trial. JAMA 2002;288:321-3
2. Barret-Connor E, and Bush TL. Estrogen and coronary heart disease in women. J Am Med Assoc 1991;265:1861-67.
3. Farhar MY, Lavigne MC, Ramwell PW. The vascular protective effects of estrogen. The vascular protective effects of estrogen. FASEB J 1996;10:615-24.
4. Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in menopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998;280:605-13.
5. Viscoli CM, Brass LM, Kernan WN, Sarrel PM, Suissa S, Horwitz RI. A clinical trial of estrogen-replacement therapy after ischemic stroke. N Engl J Med 2001;345:1243-1249.
6. Hodis HN, Mack WJ, Lobo RA, Shoupe D, Sevanian A, Mahrer PR, Selzer RH, Liu CR, Liu CH, Azen SP. Estrogen in the prevention of atherosclerosis. Ann Intern Med 2001;135:939-953.
7. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Artherioscler Thromb Vasc Biol 1997;17:217-221.
8. Riis BJ et al Norethisterone acetate in combination with estrogen:Effects on the skeleton and other organs. Am J Obstet Gynecol 2002;187:1101-16.
9. Alexanderson P, Haarbo J, Sandholdt I, Shalmi M, Lawaetz H, Christiansen C. Norethindrone acetate enhances the antiatherogenic effect of 17beta-estradiol: a secondary prevention study of aortic atherosclerosis in ovariectomized cholesterol-fed rabbits. Arterioscler Thromb Vasc Biol 1998;18:902-7.
10. The Writing group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. JAMA 1995;273(3):199-208
11. Adams MR, Kaplan JR, Manuch SB, et al. Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomized monkeys: lack of an effect of added progesterone. Arteriosclerosis 1990;10:1051-1057.
12. Grodstein F, Manson JE, Golditz GA, WillettWC, Speizer FEStampfer MJ. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med 2000;133:933-41
13. Cauley JA, Norton L, Lippman ME et al. Continued breast cancer risk reduction in postmenopausal women treated with raloxifene: 4-year results from MORE trial. Multiple outcomes of raloxifene evaluation. Breast Cancer Res Treat. 2001Jan;65(2):125-34.
14. Barrett-Connor E, Grady D, Sashegyi A, et al. Raloxifene and cardiovascular events in osteoporotic postmenopausal women: four-year results from the MORE randomized trial. JAMA 2002Feb20;287(7):847-57