This article is more than 5 years old. Information may no longer be current.
Early initiation of menopausal hormone therapy did not affect atherosclerosis progression
Early initiation of menopausal hormone therapy improved some markers of CVD, but had no effect on the progression of atherosclerosis 4 years, according to new data from the KEEPS study.
Researchers conducted the randomized, double blind, placebo-controlled trial to determine whether hormone therapy was associated with cardioprotective effects in women who recently entered menopause.
The study population included women aged 42 to 58 years who were between 6 and 36 months from their last menses, had no history of CVD, and had plasma follicle-stimulating hormone levels ≥35 IU/L and/or estradiol levels <147 pmol/L.
Participants were randomly assigned one of three therapies: oral conjugated equine estrogens (Premarin, Pfizer) 0.45 mg/day (n=230); transdermal 17 beta-estradiol (Climara, Bayer HealthCare) 50 mcg/day (n=222); or placebo (n=275). Those assigned oral conjugated equine estrogens and transdermal 17 beta-estradiol also received oral progesterone (Prometrium, Abbott Laboratories) 200 mg twice monthly.
The primary outcome was change in carotid artery intima-media thickness. Secondary outcomes included changes in markers of CVD risk. Participants were followed for 4 years, at which point carotid artery intima-media thickness was available for 79.8% of participants, of whom 63.8% of all participants were still receiving study medications.
Similar atherosclerosis progression
The researchers reported a similar increase in carotid artery intima-media thickness during follow-up in all three hormone therapy groups, at a mean rate of 0.0076 mm/year.
The difference in rates of change between the oral conjugated equine estrogen group and the placebo group was 0.0008 mm/year (95% CI, –0.0012 to 0.0029); the difference in rates of change between the estradiol group and the placebo group was 0.0005 mm/year (95% CI, –0.0016 to 0.0026).
Change in coronary artery calcium score was also similar between the hormone therapy groups (risk difference for oral conjugated equine estrogen group vs. placebo group, –3.6 percentage points (95% CI, –11.4 to 4.1); risk difference for estradiol group vs. placebo group, –2.1 percentage points (95% CI, –10 to 5.7).
Some CV risk factors affected
Compared with women assigned placebo, those assigned oral conjugated equine estrogen showed increased HDL (mean difference, 2.75 mg/dL; 95% CI, 1.27-4.22), decreased LDL (mean difference, –5.1 mg/dL; 95% CI, –8.73 to –1.47), increased C-reactive protein (mean difference, 10.48 nmol/L; 95% CI, 4.95-15.91) and increased sex hormone-binding globulin (mean difference, 38.71 nmol/L; 95% CI, 34.98-42.44) during follow-up.
Compared with women assigned placebo, those assigned estradiol showed decreased total cholesterol (mean difference, –5.52 mg/dL; 95% CI, –9.76 to –1.28), decreased non-HDL (mean difference, –4.23 mg/dL; 95% CI, –7.86 to –0.61) and decreased Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) score (mean difference, –0.23; 95% CI, –0.43 to –0.04) during follow-up.
The researchers reported no between-group differences in BP or interleukin-6 levels.
Vaginal bleeding was more common among women assigned estrogen compared with placebo (P<.001). Other adverse events, including serious adverse events, did not differ based on treatment.
“The North American Menopause Society and other groups support the use of [menopausal hormone therapy] for relief of menopausal symptoms in women at low risk for known complications of this therapy. Results of KEEPS are consistent with this recommendation and provide novel information that, in recently menopausal women at low cardiovascular risk, 4 years of [menopausal hormone therapy] neither increases nor decreases atherosclerosis progression as measured by [carotid intima-media thickness] or [coronary artery calcium] score. To the extent that these imaging methods predict CVD events, our findings suggest that [menopausal hormone therapy] neither is a risk nor is protective in the population studied,” S. Mitchell Harman, MD, PhD, of the Kronos Longevity Research Institute and the Phoenix Veterans Affairs Health Care System, and colleagues concluded.
Disclosure: The study was funded by the NIH and the Aurora Foundation. Some study medications were donated by Abbott Pharmaceuticals and Bayer HealthCare. See the full study for a list of the researchers’ relevant financial disclosures.
Perspective
Back to Top
The KEEPS trial is a well-designed and long-anticipated study that provides us with new information from several perspectives. KEEPS was ostensibly designed to confirm the timing hypothesis — the concept that CV benefits would be realized if hormone therapy (HT) were given to healthy, recently postmenopausal women without atherosclerosis. The timing hypothesis was an attempt to reconcile the Women’s Health Initiative findings that indicated no CV benefit and possible harm when HT is given to older women who may already have atherosclerosis with years of observational studies reporting coronary benefits in younger women starting HT close to menopause. Unfortunately, for proponents of the timing hypothesis, KEEPS does not provide supporting evidence.
Healthy women between 6 and 36 months from their last menses and aged 42 to 58 years were randomly assigned to oral conjugated estrogen 0.45 mg/day (lower than the 0.625 mg dose used in the Women’s Health Initiative) or transdermal estradiol 50 mcg/day, both with progesterone 200 mg 12 days per month, or placebo. The primary endpoint for KEEPS was carotid artery intima-media thickness, and the secondary endpoint was coronary artery calcium (CAC). Neither endpoint confirmed the timing hypothesis.
Other interesting and important information from KEEPS relates to the different estrogen delivery routes used in the study. Findings confirmed that oral and transdermal estrogen produce different effects on lipid parameters, insulin resistance, and the levels of estradiol, estrone and sex hormone-binding globulin. Oral estrogen raised HDL, triglycerides, estrone and sex hormone-binding globulin while lowering LDL. Transdermal estrogen decreased insulin resistance and raised estradiol levels. The clinical significance of these differences is not entirely clear, especially with regard to hormone levels, and it is hoped that future KEEPS publications will comment on the potential significance.
Observational studies have suggested lower risk for thrombosis with transdermal estrogen, but there was no signal of greater safety with transdermal estrogen in KEEPS, where there was one thrombotic event in the transdermal HT arm and none in the oral HT arm. KEEPS was not large enough to demonstrate a risk in the incidence of thrombosis.
Of potential concern is the report of more endometrial cancers in the HT groups (not statistically significant, but noteworthy). This finding differs from findings in the Women’s Health Initiative where the risk of endometrial cancer with use of HT (CE 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily) showed a nonsignificant decrease in risk in the HT group compared with the placebo group that persisted in the follow-up period. (HR=0.81; 95% CI 0.48-1.46; Anderson. JAMA. 2003; Heiss. JAMA. 2008]
Another interesting finding is that after 4 years, there was no significant difference in vasomotor symptoms between the transdermal HT group and the placebo group. Of note, the oral HT treatment appeared to suppress vasomotor symptoms more effectively than the transdermal HT throughout the study. Equivalence issues across various estrogen products have always been challenging.
It is interesting to look at the questions raised by KEEPS. There was no demonstrated carotid artery intima-media thickness benefit or risk in healthy, recently menopausal women with little to no atherosclerosis at baseline. We know from the Women’s Health Initiative that there was CV harm in older women who presumably had atherosclerosis at the time of initiating HT. That begs the question: Is there a subpopulation of postmenopausal women who might derive CV benefit, and, if so, how might they be identified?
Another question that remains is whether 200 mg of micronized progesterone for 12 days per month provides adequate endometrial protection. Looking at markers of endometrial stimulation — vaginal bleeding, endometrial hyperplasia, and endometrial cancer— there is evidence that the women assigned to the two hormone arms were experiencing more endometrial stimulation. Combining endometrial cancer and endometrial hyperplasia, there were four, two and one cases in the oral HT, transdermal HT and placebo groups, respectively. The report of bleeding occurring outside the expected time was three to four times more frequent in the two HT groups compared to placebo. Unpredictable bleeding is a quality of life issue for many women. We know from our experience in the 1990s that unexpected bleeding was a primary reason women discontinued HT.
Overall, the KEEPS findings support the position of the North American Menopause Society that endorses the use HT for menopausal symptoms in recently postmenopausal who have no contraindication [NAMS HT Position Statement 2012]. These are welcome findings for clinicians and for recently menopausal women who would like to take HT for their symptoms.
We look forward to seeing more publications from the KEEPS trial as well as the anticipated results from the ELITE trial to further enhance our understanding of the complex role HT plays in the lives of postmenopausal women.