Return to American Society for Reproductive Medicine                                    Print This
HRT: Translating New Methods and Formulations into Patient Satisfaction

Discontinuation of HRT: The Reasons and Some Solutions

“The field of hormone replacement therapy (HRT) has been challenged now as never before,” began speaker Phillip Sarrel, MD. Dr. Sarrel is Professor of Obstetrics and Gynecology and Psychiatry at the Yale University School of Medicine, and his career spans more than 25 years during which time there have been several important alterations in the perceived role of HRT. In the United States and elsewhere, much effort has been directed to the study of hormone replacement therapy. Still, according to Dr. Sarrel and other experts, the education of women using hormone replacement therapy may be lacking, as physicians and other healthcare professionals strive to interpret and deliver menopause healthcare in an environment of rapidly accumulating new data.

Among women in the United States who experience hot flushes, vaginal dryness, and other symptoms of ovarian hormonal insufficiency, only a minority has ever used HRT. One-third has never consulted a physician regarding HRT, and one-third have consulted, but never started HRT. Among those who have started taking HRT, only 8-12% has been compliant for more than two years. According to a pharmaceutical company analysis of Premarin, the average usage time, from the writing of a prescription, was only 141 days.

Why do more than 85% of women who start HRT discontinue treatment? According to a survey conducted by Dr. Sarrel and his colleagues and presented at the North American Menopause Society (NAMS. 1997, Boston, MA), women who are within two years of their last menstrual cycle discontinue HRT due to a fear of breast cancer. Women taking HRT who are more than 2 years from their last menstrual flow most often discontinue due to adverse effects. First among adverse effects that lead to discontinuation are unscheduled bleeding. Others are persistence or aggravation of menopausal symptoms, bloating and weight gain, irritability and depression, and an inadequate sexual effect (Sarrel et al. NAMS. 1997, Boston, MA).

Several factors relating to the HRT regimen may explain the adverse effects (Table 1).

Fluctuation of serum estrogen, specifically the acute estrogen depletion of the early menopause, produces the most severe symptoms, including hot flushes, midcycle “unscheduled” bleeding. HRT may induce other arterial reactions that include migraine, angina, and myocardial infarction (Sarrel P. Human Reproduction Update 1999;5: 205-209). A too-high dose of HRT can cause breast tenderness, which is a symptom very frightening to women. A too potent progestin can serve to overwhelm the effects of estrogen and cause symptoms of its own, Dr. Sarrel said, as can an imbalance of the estrogen and progestin components of HRT. This was demonstrated in the PEPI trial, which showed clearly that medroxyprogesterone acetate (MPA) canceled out the beneficial effects of estrogen on HDL, while micronized progestin had a neutral effect (JAMA 1995;273:199-208). A discussion of relative progestogen potencies was provided by Sitruk-Ware (Menopause 2002;9(1):6-15). In addition, Rosano et al. provided an example of the detrimental effects of a too potent progestin in women with cardiovascular disease (Rosano GMC, et al. J Am Coll Cardiol 2000;36:2134-59).

The risk/benefit ratio of HRT is greatly affected by the progestin, as many studies to date have shown. If estrogen were to partner with an ideal progestin, that progestin would be protective to the endometrium, have no adverse effects on arterial structure and function, have no adverse psychiatric or sexual effects, and would not compromise any benefits of estrogen, Dr. Sarrel said.

Another factor causing symptoms that may not be corrected by some HRT regimens is depletion of androgens. The impact of depleted androgens in meno-pausal women was demonstrated by Sherwin et al. (Psychosom Med 1987; 49:397-409), in a study of surgically sterilized women who experienced a loss of sexual desire and function. In this study, estradiol (E2) was given with testosterone, and evaluated for effects on sexual desire, arousal, sexual fantasies, coitus and orgasm. Compared with control and estrogen alone, the combination of estrogen + testosterone was successful in inducing a significant improvement in all four parameters. However, the dose of testosterone used in Sherwin et al. was pharmacologic. Dr. Sarrel also conducted a study showing that the combination of esterified estrogens plus methyltestosterone improved both sexual desire and sexual sensation (Sarrel PM et al. J Reprod Med 1998;43:847-858) (Figure 1).

Paradoxically, non-compliance itself is a significant source of adverse effects in women taking HRT, and in this instance, patient education may play a corrective role. Dr. Sarrel emphasized the importance of a structured and thoughtful approach to menopausal healthcare. Physicians rendering care to women should take the time in the clinic to discuss the benefits and risks of HRT and answer patient questions, in addition to conducting breast and pelvic exam with a Pap test (Sarrel PM et al. NAMS. 1997, Boston). Based on data from Dr. Sarrel’s extensive research in the area, the way in which menopause healthcare is delivered can affect adherence. Adherence, in its turn, may be a key to realizing the benefits of any treatment, including HRT, and in controlling its risks (Horwitz R et al. Arch Intern Med 1993;153(16): 1863-8. Schiff I, et al. Menopause 1998;5:69-76).

HRT and Cardiovascular Health

Approximately twenty years worth of observational studies into the cardiovascular effects of the hormones in HRT suggested that high dose unopposed estrogen reduced the incidence of coronary events, and that combinations of estrogen and progestin were associated with smaller reductions. The first major clinical trial to evaluate the differences in progestins was the Postmenopausal Estrogens/Progestins Interventions (PEPI) trial (JAMA 1995;273:199-208), said speaker Robert D. Langer, MD, MPH, Professor of Family and Preventive Medicine at the University of California San Diego. “PEPI was a three-year study designed in the late 1980s to look in detail at the differences in major cardiovascular risk factors across forms of HRT.” Its primary endpoint was change in HDL cholesterol, thought at that time to be one of the major mechanisms involved in the cardiovascular protection conferred by HRT. PEPI investigators also evaluated other domains of cardiovascular risk, including blood pressure, insulin resistance, and clotting factors, as well as endometrial safety and effects on bone density.

PEPI tested the form of estrogen most commonly used in HRT — conjugated equine estrogens (CEE) — at a dose of .625 milligrams daily. This was combined with two different progestogens: medroxyprogesterone acetate (MPA) cyclical (10 mg 12 days per cycle) and continuous (2.5 mg daily), and cyclical micronized progesterone (200 mg 12 days per cycle). PEPI showed that micronized progesterone did not significantly attenuate the CEE mediated rise in HDL, while both continuous and cyclical MPA did.

Insulin resistance is a strong risk factor for coronary disease in women, Dr. Langer said — “about two to three times stronger than in men.” Both MPA combinations had an adverse effect on 2-hour post-challenge glucose that was not found with CEE plus micronized progesterone or estrogen alone.

Using the PEPI trial data to predict changes in 10-year coronary risk based on standard assumptions for changes in major cardiovascular risk factors, Dr. Langer said it was expected that the benefit from CEE plus either cyclical or continuous MPA would be “about a 5% decline in coronary risk.” For micronized progesterone, the decline was projected at 19%, and for CEE alone, the expectation was for a 23% reduction.

In July of 2002, Women’s Health Initiative (WHI) investigators stopped treatment in the trial testing CEE plus MPA 2.5 mg in 17,000 women because of emerging harm from breast cancer. Importantly, coronary, stroke and venous thromboembolic events were not decreased, but instead increased, in the active treatment group. Dr. Langer made the point that what happened in the WHI was in many ways similar to what happened in HERS (Heart and Estrogen/ Progestin Replacement Study). That is, most of the harm was done in the first year of treatment (Table 2). Notably, these increases in risk were the same for women with a known cardiovascular disease as for those with no known history, and did not preferentially target older women. In fact, Dr. Langer said, the women who experienced the greatest hazard were the youngest women — from ages 50 to 59.

What do these data and those from prior studies such as HERS tell us? According to Dr. Langer and other experts, they suggest that the early cardiovascular events are probably not mediated by classical cardiovascular risk factors. “My guess is that we’re going to find one or a few factors that code for an extraordinarily high risk for early cardiovascular events in susceptible women, and the remaining population is not at excess risk. Candidates include some factors related to first-pass hepatic metabolism of estrogen. If they are implicated, transdermal estrogen could eliminate the risk.” He added that the overall attributable risk — the actual number of cases or coronary events — was relatively small; treatment added less than one-tenth of one percent to CV risk. The WHI estrogen-only trial continues. If that study finds overall benefit, Dr. Langer said, the effects would be due to long-term changes in risk factors, and the difference between the findings for the two parts of the study would be attributable to the progestin.

Langer noted that studies in non-human primates have demonstrated differences in the vascular effects of MPA and progesterone. In those studies, estrogen protected against coronary vasospasm; this effect was blocked by MPA but not by progesterone. In long-term studies, estrogen reduced the development of atheroma; that benefit was blocked by MPA but not by progesterone. Like the PEPI results, these studies suggest that there are clinically relevant differences in progestogens. But, to date, human trials of combination HRT evaluating disease rates have only tested combinations with continuous MPA. Other options, like micronized progesterone or local vaginal or intra-uterine administration of progestogenic agents may detract less from the putative benefits of postmenopausal estrogen.

Progestogens Effects on Bone, Breast, and Quality of Life

At the present moment, there are more questions than answers about
HRT, said speaker Lorraine Fitzpatrick, MD, Professor of Internal Medicine and Director of the Women’s Health Fellowship at the Mayo Clinic. Fortunately, there are many HRT options (or HT, as the NIH suggests calling hormone replacement therapy), even though long-term data are lacking. But some things are known. For example, in PEPI, all active treatments were equally effective in protecting against vasomotor symptoms. Breast discomfort was slightly more severe with MPA than with micronized progesterone. Weight gain in PEPI was less in any active treatment groups compared to controls, but compared with subjects randomized to MPA, subjects receiving micronized progesterone had more weight loss.

Another study evaluated quality of life issues in women who were switched from MPA to oral micronized progesterone. There, women experienced improvements in a number of menopausal parameters: somatic vasomotor symptoms, anxiety and symptoms consistent with depression.

Studies looking at the connection between breast cancer and progestogens are “all over the map,” Dr. Fitzpatrick said. But an important question to answer is whether progestogens increase proliferation and stimulate growth factors and cytokines, because of the possibility that such stimulation increases mutational rates.

Breast density may be affected by progestogens, according to some studies, and the connection between increased breast density and risk of breast cancer has been established. “When breast tissue biopsies were analyzed from women who were taking an estrogen alone versus and estrogen and progestogen, said Dr. Fitzpatrick, “…there was less epithelial and more fatty tissue in patients who were taking estrogen alone.” As breast density increases, so does risk for breast cancer — by almost 5.2-fold if 75% of the breast tissue is epithelial rather than fatty tissue. PEPI data did not support the increase of breast density with longer term use of HRT, but, as Dr. Fitzpatrick said, a myriad of studies have indeed shown that HRT is associated with an increased relative risk for breast cancer.

The effect of HRT versus calcium, vitamin D and placebo on osteoporosis has been evaluated in a number of studies. In a review of 59 randomized trials (45 prevention and 12 treatment), of at least one year in duration, Dr. Fitzpatrick found three trials suggestive of a protection with HRT from vertebral fractures, and four that were suggestive of protection against non-vertebral fractures. The WHI showed a “reduction in fractures at all sites.” By contrast, there is consistent evidence that HRT increases bone density at all sites with both estrogen alone and the combination of estrogen and any progestogen in a dose- dependent manner.

Estradiol Percutaneous Gel

Among the newer options in postmenopausal hormone therapy (not yet available in the United States) is an estradiol percutaneous gel. This product, which was first made available in France in early 1975 and remains its most popular form of HRT, has been available in Canada for the past four years. Speaker Fay Weisberg, MD Assistant Professor at the University of Toronto, explained that its benefit versus systemic HRT is that it avoids the first pass liver metabolism, diffusing directly through the epidermis to the bloodstream. In addition, there is less nausea with the transdermal form than with oral therapy, it avoids the GI tract, and, as Dr. Weisberg added, “it is as effective as oral therapy for relief of symptoms.” There are, however, some disadvantages with the gel that include skin irritation, skin markings, and some problems with adherence.

The gel is typically applied to the skin over a large surface area on a daily basis, including both arms, the abdomen or the inner thigh. It contains 0.06% 17-beta estradiol, and once applied, Dr. Weisberg said, it is absorbed within two minutes. Once absorbed, the gel is unaffected by the application of moisturizers or sunscreen. Physiological plasma concentrations of estradiol are reached quickly, with a surface-area and dose dependent relationship to time of symptom relief as well as efficacy, studies have shown. “It relieves conditions of urethral atrophy and improves sleep patterns as well,” said Dr. Weisberg.

A growing body of data suggest that the estradiol gel works as well as oral estrogen. In a study by Christiansen (Maturitas 1987;9:207), estradiol gel plus calcium was compared to estradiol gel plus placebo, placebo estradiol gel plus calcium, and placebo estradiol gel plus placebo calcium. Endpoints were symptoms of menopause using the Kupperman Index, and including measures of hot flushes, paresthesia, insomnia, nervousness, melancholia, vertigo, fatigue, arthralgias/myalgias, headaches, and palpitations. The findings: “In month three, and continuing to month 24, there was a significant improvement in Kupperman Index parameters in patients who received the gel as well as the gel plus calcium,” said Dr. Weisberg. No significant differences in safety were found after two years between the gel and the placebo/calcium or the gel/calcium.

Other studies with estradiol gel have demonstrated its equivalence with oral CEE in effects on bone density, a possibly significant cardiovascular finding of no increases over time in renin substrate, and no change in triglycerides compared to increases with oral conjugated estrogen.

Preferential medical indications for transdermal estradiol over oral estrogen include smokers, patients with liver or gall bladder disease, persons with insulin sensitivity, persons who are lactose intolerant, and patients who dislike the pill or cannot take oral therapy.

Return to American Society for Reproductive Medicine                                    Print This

All contents Copyright © 1999 - 2003 Medical Association Communications