ABSTRACT & COMMENTARY
News from the KEEPS Study: HRT Does Not Decrease Progression of Atherosclerosis Over 4 Years of Treatment
By Jeffrey T. Jensen, MD, MPH
Synopsis: Although treatment with oral conjugated estrogens or transdermal estrogen improved some surrogate markers of cardiovascular disease risk, no reduction in the progression of atherosclerosis was seen over 4 years of treatment.
Source: Harman SM, et al. Arterial imaging outcomes and cardiovascular risk factors in recently menopausal women: A randomized trial. Ann Intern Med 2014;161:249-260.
The KEEPS (Kronos Early Estrogen Prevention) study was a randomized, controlled trial of postmenopausal hormone therapy designed to assess the effects of treatment in women who were within 36 months of their last menses, the typical time interval for initiation of therapy. Healthy (no history of cardiovascular disease [CVD]) recently postmenopausal women were randomized to receive oral conjugated estrogens o-CEE 0.45 mg/d (Premarin®, Pfizer Pharmaceuticals), transdermal estradiol 50 mcg/d (t-E2, Climara®, Bayer HealthCare), or placebo. Estrogen-treated subjects also received oral progesterone capsules, 200 mg/d (Prometrium®, Abbott), on days 1-12 of each month. The treatment allocation was double-blinded; all estrogen-treated subjects received both a patch and daily pill that were active or placebo according to assignment, plus the monthly progesterone capsules. To conceal allocation, placebo-treated subjects also received a placebo patch, pill, and capsule. The primary outcome was the annual change from baseline in carotid artery intimamedia thickness (CIMT) measured by ultrasonography. Secondary outcomes included changes from baseline in other markers of cardiovascular risk including coronary artery calcium (CAC) score (measured by chest CT scan at baseline and end of study) and several biochemical endpoints; HDL-C, LDL-C, triglycerides, interleukin-6, C-reactive protein, sex hormone-binding globulin, glucose, and insulin. A total of 727 women were randomized: o-CEE (230, 31.6%), t-E2 (222, 30.5%), placebo (275, 37.8%). Participants had a mean age of 52.7 years and were an average of 1.4 years after menopause. At 48 months, CIMT was available for 580 women (79.8%); of these 464 (63.8%) were still using the assigned study medications. The mean duration of treatment was 37.4 months for o-CEE, 34.6 months for t-E2, and 37.6 months for placebo.
A similar increase in mean CIMT (approximately 0.007 mm/year) was seen in all three groups during the 4 years of observation. The change in CAC scores (an increase of 17.4% o-CEE, 18.9% t-E2, and 21.0% placebo group) were also not statistically significant. Treatment with o-CEE resulted in a decrease in LDL-C and an increase in HDL-C, C-reactive protein, and SHBG. The insulin resistance score decreased with t-E2. About half the subjects (o-CEE 49.1%, t-E2 47.3%, and placebo 47.6%) experienced at least one adverse event, but most of these were mild. The study was not powered to assess serious adverse events.
The KEEPS study was designed to determine whether early initiation of hormonal replacement therapy (HRT) could prevent the development of CVD in postmenopausal women. One of the major criticisms of the Women’s Health Initiative (WHI) study is that the population studied was asymptomatic and approximately 10 years postmenopausal — a decade older than the age at which women commonly start HRT. The difference from the positive benefits seen in observational studies suggests that a critical window for initiation of treatment may exist. In fact, follow-up reanalysis of the WHI combined CEE/medroxyprogesterone acetate treatment group documented a nonsignificant trend toward protection from CVD in women < 10 years postmenopausal in contrast with the elevated risk observed in women starting therapy more than 20 years after menopause.1
Since the investigators in the KEEPS study did not have the resources or time to enroll a cohort as large as WHI, the enrollment and outcomes were more modest. Only surrogate measures of CVD were evaluated. Treatment was divided between an oral and transdermal approach. But all women received an oral treatment with micronized progesterone. The PEPI study demonstrated that the favorable effects of oral CEE on lipids were attenuated with medroxyprogesterone acetate, but not affected with micronized progesterone.2 However, we still don’t understand the role of postmenopausal progestin replacement therapy outside of endometrial protection.
Our current understanding of HRT is that thrombosis is the principle risk.3 Thrombosis is related to estrogen-induced changes in hepatic globulins. Current research supports the use of non-oral routes of administration of estrogen and the use of estradiol rather than ethinyl estradiol to avoid the first pass effect of oral therapy on the liver.4
The KEEPS study evaluated the factor through which thrombosis mediates most heart attacks, the development of atherosclerotic plaques. Although no difference was seen in this study, there are several limitations that might explain the results. The first is that, in contrast to WHI, the investigators were too careful to exclude women at risk for CVD from the study. By enrolling a healthy low-risk population, the ability to demonstrate a difference may have been limited. All women in the study were required to have a baseline CIMT of < 50 Agatston units. These women may be at lower risk for progression. Another explanation is that the time for follow-up was too short. However, there is no indication from the data that any trend toward protection with either treatment was emerging during the later years of the study. All of the treatment groups showed a slow trend toward progressive arterial narrowing.
While the negative results of KEEPS do not settle the question about the critical window of treatment hypothesis, they do demonstrate that combined HRT does not increase atherosclerotic progression. Given this, I remain convinced that transdermal (or vaginal) estradiol treatment makes sense for women using estrogen replacement, as the thrombosis mechanism remains an important factor. Data from the ESTHER study showed no increase in the risk of DVT in users of transdermal estradiol compared to non-users, but an increase in risk in women using oral products.5 It would take another large study the size of WHI to evaluate whether an alternative HRT regimen actually protects against CVD. I am not optimistic about seeing this funded.
The take-home clinical message is that reduction in CVD is not a primary indication for HRT or a secondary health benefit. However, the results from KEEPS are consistent with other studies that suggest otherwise healthy women initiating HRT at or shortly after menopause do not increase their risk of adverse cardiovascular outcomes.
- Harman SM, et al. Timing and duration of menopausal hormone treatment may affect cardiovascular outcomes. Am J Med 2011;124:199-205.
- Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. The Writing Group for the PEPI Trial. JAMA 1995;273:199-208.
- Harman SM. Estrogen replacement in menopausal women: Recent and current prospective studies, the WHI and the KEEPS. Gend Med 2006;3:254-269.
- Sandset PM, et al. Mechanisms of thrombosis related to hormone therapy. Thromb Res 2009;123(Suppl 2):S70-73.
- Canonico M, et al. Hormone therapy and venous thromboembolism among postmenopausal women: Impact of the route of estrogen administration and progestogens: The ESTHER study. Circulation 2007;115:840-845.