Age Moderates the Short-Term Effects of Transdermal 17β-Estradiol on Endothelium-Dependent Vascular Function in Postmenopausal Women
Objective— We evaluated age and coronary heart disease (CHD) as potential moderators of the effects of 17β-estradiol on vascular endothelial function in postmenopausal women.
Methods and Results— In a double-blind crossover design, 100 postmenopausal women aged 50 to 80 years were randomized to each of 3 transdermal patches, releasing 17β-estradiol (0.05 mg/d), 17β-estradiol (0.05 mg/d)+ norethindrone acetate (NETA, 0.14 mg/d), and placebo. Flow-mediated dilation (FMD) and response to 400 μg sublingual glyceryl trinitrate (GTN-D) were assessed approximately 18 hours after patch placement. Age, but not CHD, moderated the FMD response to treatment (P=0.01). For women in their fifties, the estradiol patch was associated with improved FMD (7.69±4.79%) compared with placebo (4.81±5.97%, P<0.05), but the estradiol+norethindrone patch response (5.81±4.85%) was not significantly different from placebo. Women in their sixties and seventies showed no alterations in FMD response to either active patch. GTN-D response declined with advancing age (P<0.01), with women in their seventies exhibiting blunted GTN-D response compared with younger women.
Conclusions— The cardiovascular benefits of natural estrogen supplementation on vascular endothelial function may be dependent on postmenopausal age, with improved vascular function evident only in the early postmenopausal years. Short-term FMD response to estradiol might help stratify individual differences in risks versus benefits of HRT.
The widespread use of estrogen supplementation by hormone replacement therapy (HRT) was founded on epidemiological evidence of its cardiovascular and other health benefits for postmenopausal women.1–3 However, clinical trial data now indicate that HRT does not lower cardiovascular risk and may increase risk in some women.4–7 Accounting for these contradictory findings remains an important endeavor that should result in a better understanding of the risks and benefits of HRT. One potentially important difference between the epidemiological studies and clinical trials is the age of onset of HRT use. Although participants in the observational studies typically had been taking HRT since menopause, clinical trials participants started HRT on average 10 years or longer after menopause.8 This 10-year difference correlates roughly with a tripling in the prevalence of coronary heart disease when comparing women in the perimenopausal years with those ages 60 to 79 years, the age range that predominated in clinical trials.9 This consideration also raises the possibility that the presence of underlying coronary disease may have been more common among clinical trials participants by the time they were started on HRT than women participating in observational studies, and HRT benefits may be diminished in the presence of coronary disease.10 There is also evidence to suggest that the cardioprotective benefits of supplementation with natural estrogens, such as 17β-estradiol, may be superior to the conjugated equine estrogens used in clinical trials.11
See page 1669
Augmentation of the vascular endothelial vasodilator response is one important mechanism by which estrogen supplementation may benefit cardiovascular health.12 Noninvasive assessment of endothelial function by ultrasound measurement of brachial artery flow-mediated dilation (FMD) has provided a useful clinical research tool in this context.13 In the Cardiovascular Health Study, favorable effects of estrogen on FMD response were limited, and evident only in a subgroup of postmenopausal women without clinical evidence of atherosclerotic disease.14 These findings were in contrast to evidence from previous studies that demonstrated augmented FMD response to estrogen.15,16 Postmenopausal age again may help explain these discrepant findings: In the initial studies showing that estrogen augmented FMD, women were younger (typically in their fifties); in contrast, the Cardiovascular Health Study sample was comprised of women over 65 years.
In the present study we evaluated age and preexisting CHD in postmenopausal women as potential moderators of improved FMD associated with natural estrogen (17β-estradiol) supplementation, both unopposed and in combination with progesterone (norethindrone acetate). Transdermal administration was chosen because at the time the study was designed, one hypothesis for the failure of HRT in then existing, completed randomized clinical trials related to the oral route of administration. It was also unclear at that time whether unopposed or combination HRT was more or less effective or harmful. Because a specific and less tested route of administration along with a different form of HRT was being tested (17β-estradiol rather than conjugated equine estrogens), we felt it prudent to study both unopposed and combination regimens in our study.
The study sample consisted of 100 postmenopausal women, aged 50 to 80 years. Both healthy women and women with established coronary heart disease (CHD) (defined by cardiac catheterization documenting ≥50% occlusion of at least one major coronary vessel) were actively recruited. For healthy women, eligibility criteria included no prior cardiovascular disease diagnosis and no prior cardiovascular events determined by health history and physical examination. Postmenopausal status was defined by amenorrhea ≥12 months, and was confirmed by a reproductive hormone panel. Exclusion criteria included: use of HRT or selective estrogen receptor modulators (SERMS) within 30 days of enrollment; congestive heart failure NYHA Class > II; pacemaker dependency; uncontrolled hypertension (defined by a resting blood pressure ≥180/105 mm Hg); persistent atrial fibrillation or tachyarrhythmia, myocardial infarction (MI) or percutaneous transluminal coronary angioplasty (PTCA) within 30 days of enrollment; coronary artery bypass grafting (CABG) within 3 months of enrollment; uncorrected valvular disease; hypertrophic or restrictive cardiomyopathy; uncorrected thyroid disease; persistent tachyarrhythmia; current nitrate use; and body mass index (BMI) ≥40 kg/m2. The protocol was approved by the Institutional Review Board at Duke University Medical Center, and written informed consent was obtained from all study participants before assessments.
A randomized, double-blind, crossover design was used in which participants were exposed to each of 3 transdermal patches, releasing 17β-estradiol (0.05 mg/d), 17β-estradiol (0.05 mg/d)+norethindrone acetate (NETA, 0.14 mg/d), and placebo. Each patch was placed on the upper quadrant of the buttock, with study assessments scheduled approximately 18 hours after placement when plasma concentrations were close to peak. Treatment order was fully randomized using all 6 possible orders of exposure to the 3 patch types, and all study personnel were blinded to the order of treatment. A washout period of 5 to 7 days was required between treatments. Studies were performed in a quiet room with temperature regulated at 22 to 24°C.
Assessment of FMD
FMD of the brachial artery was assessed a minimum of 4 hours after a light fat-free caffeine-free breakfast that was provided. Participants were instructed to avoid use of vasoactive medications (list provided to participants) from midnight until after the FMD assessments were completed the following morning. Additionally, all participants were called by study staff on the evening before testing and were reminded about the instructions. Compliance with these instructions was confirmed before conducting the FMD study.
Longitudinal B-mode ultrasound images of the brachial artery, 4 to 6 cm proximal to the antecubital crease, were obtained using an Acuson Aspen ultrasound platform with an 11-MHz linear array transducer. All images were obtained by the same sonographer (blinded to treatment), who had extensive experience with the FMD technique and held the transducer manually. Images were obtained under the following conditions: (1) after 10 minutes of supine relaxation; (2) during reactive hyperemia, induced after inflation for 5-minute to supra-systolic pressure (≈200 mm Hg) of a pneumatic occlusion cuff placed around the forearm, and (3) after administration of 400 μg sublingual glyceryl trinitrate (GTN) spray. End-diastolic images were stored to a magnetic-optical disk, and arterial diameters were measured as the distance between the proximal and distal arterial wall intima-media interfaces using PC-based software (Brachial Analyzer Version 4.0, Medical Imaging Applications LLC).
Peak hyperemic flow was assessed by Doppler velocity measurement during the first 10 seconds after deflation of the occlusion cuff, and hyperemic flow response was defined as the percent change in flow relative to resting baseline. Peak FMD response was assessed from 10 to 120 seconds after deflation of the cuff, with peak arterial diameter quantified using polynomial curve fitting, and FMD was defined as the maximum percent change in arterial diameter relative to resting baseline. In an unpublished evaluation of 20 healthy men and women who underwent our FMD assessment protocol on consecutive days, repeat FMD values showed a correlation of r=0.81, P<0.001, a mean difference of 0.64±2.67%, and a coefficient of variation of 26.73%. Glyceryl trinitrate dilation (GTN-D) was defined as peak arterial diameter 3 to 5 minutes after GTN administration, and expressed as percent change from resting baseline. Assessments of FMD and GTN-D were performed by a single reader who was blinded to participants’ identity and treatment condition.
Female Reproductive Hormones
On completion of the FMD assessment at each visit, a blood sample was drawn from an antecubital vein and collected in a serum separator tube, left to clot for 30 minutes, centrifuged at room temperature for 15 minutes at 3000 rpm, transferred to a serum transfer tube, and refrigerated before being assayed the same day for serum 17-β-estradiol and follicle-stimulating hormone (FSH) by immunochemiluminometric assay (Labcorp Corporation).
Characteristics of participants with established CHD were compared with healthy participants using Student t tests and Chi-square tests. Repeated measures analysis of covariance (ANCOVA) tests were used to evaluate the effects of healthy versus CHD status and of age on FMD and GTN response. Primary analyses evaluated age as a continuous variable, and where age was a significant factor, its effects were illustrated by categorizing age by decade. Covariates included baseline arterial diameter, ethnicity, cardiovascular medications, and smoking status. A probability value of P<0.05 was used to denote statistical significance. All analyses were performed using the SAS System, Release 8.2.
CHD Versus Healthy Participants
CHD patients (n=49) were comparable to healthy participants (n=51) in most characteristics, including age, BMI, minority composition, blood pressure, and percent smokers (Table 1). Participants with CHD were younger at menopause, whether it was natural (47±7 versus 51±6 years, P<0.05) or surgical (40±10 versus 49±8 years, P<0.01), and had therefore experienced a greater number of postmenopausal years (22±11 versus 16±8 years, P<0.01) than healthy participants. A greater percentage of CHD patients were taking ACE-Inhibitors (51% versus 6%, P<0.001), β-blockers (80% versus 10%, P<0.001), statins (76% versus 4%, P<0.001), and aspirin (67% versus 24%, P<0.001), and CHD patients had lower HDL (53±15 versus 62±19 mg/dL, P<0.05) and LDL (119±37 versus 140±38 mg/dL, P<0.01) cholesterol. A greater percentage of CHD patients were diabetic (18% versus 4%, P<0.05), had previous diagnosis of hypertension (63% versus 27%, P<0.001), and were hypercholesterolemic (71% versus 27%, P<0.001).
The study sample age range spanned 3 decades, with participants in their fifties (n=19), sixties (n=45), and seventies (n=36); age by decade was used to illustrate the effects of age, which was considered a continuous variable in the planned primary analyses. As shown in Table 2, older participants were shorter in height (P<0.01) and weighed less (P<0.05) (though they were comparable in BMI), and were less likely to be minorities (P<0.05). Age was strongly related to the number of years since menopause (P<0.0001).
Endothelial FMD Response
CHD Versus Healthy Participants
CHD diagnosis was not associated with altered FMD response compared with healthy women (F(1,90)=0.51, P=0.48), and CHD status did not demonstrate an interaction effect with patch (F(2,180)=1.38, P=0.26; FMD responses (%±SD): CHD placebo=4.73±4.09; CHD estradiol=4.62±4.96; CHD estradiol+NETA=4.51±3.49; Healthy placebo=4.51±4.46; Health estradiol=4.54±4.55; Healthy estradiol+NETA=4.08±4.44). Where age was categorized by decade, the 3-way interaction between age, CHD status, and patch was also nonsignificant (P=0.19).
A repeated measures ANCOVA evaluating determinants of FMD that included age as a continuous variable revealed a significant effect for patch (F(2,180)=3.29, P<0.05), age (F(1,90)=10.31, P<0.002), and a patch by age interaction (F(2,180)=4.61, P=0.011). The effects of age on FMD (mean±SD) during exposure to placebo, estradiol, and estradiol plus NETA patches are shown in Table 3, which compares women in their fifties, sixties, and seventies as a means of illustrating the effects of age. A repeated measures ANCOVA evaluating age categorized by decade confirmed the age by patch interaction (F(4,180)=2.62, P=0.03). Post-hoc evaluation of patch by decade revealed that FMD was modified by patch condition in women aged 50 to 59 years, but not in 60- to 69- or 70- to 79-year-old women (Table 3). For women in their fifties, the estradiol patch was associated with a significantly greater FMD compared with placebo (P<0.05), whereas the combination estradiol plus NETA patch did not result in a significant change in FMD. The association of advancing age with the FMD response to the estradiol patch is illustrated in the Figure, which shows that estradiol improved FMD response (P<0.005) for women in their fifties, but not for women in their sixties and seventies.
Previous HRT use was not a significant determinant of FMD response to the active patches, and when included in our statistical models this covariate did not alter our observations for CHD status and age. Hyperemic peak flow response to forearm occlusion (percent change from baseline) was unrelated to CHD diagnosis (P=0.68; means±SD: CHD placebo=73±43; CHD estradiol=71±43; CHD estradiol+NETA=75±44; Healthy placebo=81±45; Healthy estradiol=79±42; Healthy estradiol+NETA=79±50), but diminished with advancing age (P=0.046; Table 4). Transdermal patch condition was unrelated to hyperemic flow (P=0.54), and there was no interaction between age and patch (P=0.62), suggesting that the observed effects for FMD were not secondary to parallel changes in hyperemic flow.
Nitroglycerin (GTN-D) Response
Nitroglycerin induced vasodilation (GTN-D; mean±SD) during exposure to placebo, estradiol, and estradiol plus NETA patches is shown by decade in Table 5. There was no effect of transdermal patch on GTN-D, but age was independently associated with altered GTN-D (P<0.01). Women in their seventies demonstrated significantly lower GTN-D than women in their fifties and sixties (P<0.05) under all patch conditions. CHD diagnosis was not associated with altered GTN-D response compared with healthy women (P=0.86).
Estrogen Manipulation Check
Plasma 17-β-estradiol levels were 88±47 pg/mL and 80±41 pg/mL for the estradiol-alone and estradiol + NETA treatment periods, respectively, and 24±24 pg/mL for the placebo period. Both active patches were associated with significantly higher 17-β-estradiol levels than placebo (P<0.001). Neither CHD/Healthy status nor age was related to plasma 17-β-estradiol responses to the active patches.
The present study shows that, in postmenopausal women, age is an important moderator of the short-term effects of 17β-estradiol on vascular endothelial function. Women aged 50 to 59 years demonstrated a marked augmentation of FMD response after approximately 18 hours of exposure to transdermal estradiol when compared with transdermal placebo. However, women aged 60 to 79 showed no evidence of improved FMD in the presence of supplemental estradiol. These effects were independent of whether participants had a prior diagnosis of CHD, which was not a determinant of the FMD response to estradiol. Our GTN-induced vasodilation observations did not exhibit age-related sensitivity to estradiol, suggesting that the vascular effects of estradiol may be relatively specific to modification of the endothelial vasodilator mechanism.
We found that when estradiol supplementation was accompanied by progesterone (norethindrone acetate, 0.14 mg/d), the improvement in FMD seen in women in their fifties was blunted and was not significantly different from placebo. Previous observations for progesterone have been mixed, with some studies reporting that progesterone abolishes the augmentation of FMD induced by estradiol,17 whereas others indicate that the estradiol response is attenuated minimally.18 It is of note that in our study we used norethindrone instead of the more widely used progestin, medroxyprogesterone. Dose of progesterone also may be important, with some evidence of lower doses having lesser effects on FMD improvements.18 Interestingly, a review of the potentially deleterious impact of progesterone on the vascular benefits of estrogen concluded that presence of coronary atherosclerosis and years since menopause are likely to be more significant factors moderating the HRT response.19
In the Cardiovascular Health Study, postmenopausal FMD response was found generally to diminish with advancing age.14 Because the Cardiovascular Health Study included only women over 65 years of age, its finding that HRT was not associated with a greater FMD response is consistent with our observations that estradiol failed to augment FMD in women 60 years or older. In contrast, studies that had earlier demonstrated augmented FMD response associated with estrogen supplementation typically included postmenopausal women in their fifties.14,15 In a recent study of women with premature ovarian failure, ranging in age from 23 to 40 years, a 6-month HRT intervention resulted in a doubling of the preintervention FMD response, restoring it to a level comparable to a healthy control group.17 Our observations extend these findings, by showing that in a randomized, double-blind, crossover design comprised of women ranging in age from 50 through 80 years, age was an important moderator of the FMD response to transdermal 17β-estradiol (0.05 mg/d) supplementation, with augmentation of FMD occurring only in women in their fifties.
The mechanisms accounting for how postmenopausal age moderates the FMD response to estradiol are not well understood and merit further research. Possible mechanisms accounting for our observed augmentation of FMD after only brief exposure (18 hours) to estradiol include increased bioavailability of nitric oxide through enhanced synthesis and reduced breakdown.20 In our study sample, the presence of documented CHD was not associated with altered FMD response to estradiol, but postmenopausal age may nonetheless have served as a proxy for atherosclerotic vascular disease (as suggested by age-related blunting of FMD independent of CHD status), despite it having gone undiagnosed clinically in our “healthy” study participants.9 Several lines of evidence suggest that natural estrogens, such as 17β-estradiol, are more likely to convey vascular benefits than the conjugated equine estrogens.11 Moreover, it has been suggested that 17β-estradiol delivered by transdermal patch, as used in the present study, may be the optimal approach to achieving postmenopausal cardiovascular protection.11 Indeed, it remains a possibility that our observations for FMD response to acute estrogen supplementation may be relatively specific to17β-estradiol when administered by transdermal patch.
The Framingham data revealed trends toward a cardioprotective effect of HRT for women aged 50 to 59 years, but increased likelihood of adverse effects among older women.1,21 Furthermore, the WHI estrogen-alone trial noted that younger women treated with conjugated equine estrogen appeared at reduced risk of CHD.22 Evidence from animal studies further underscores the relevance of postmenopausal age. Studies on cynomolgus monkeys have shown that cardiovascular benefits from HRT were observed only when therapy was initiated at the time of ovariectomy, whereas benefits were absent when HRT was delayed for 2 years.23 The discrepant findings between epidemiological and clinical trials HRT evidence may be attributable in part to the possibility that age at initiation of hormone therapy is a moderator of the vascular effects of exogenous estrogen. Women in the epidemiological studies initiated hormone therapy at an average age of 51 years, whereas the HRT clinical trials tended to enroll women who on average were in their sixties.10 Dubey and colleagues,11 in their comprehensive review, underscore the importance of timing of estrogen supplementation if cardiovascular benefits are to be achieved. Although our study included women who had used HRT in the past, our observations relating to age are nonetheless also consistent with the theory that vascular benefits of estradiol supplementation may be dependent on the duration of the estrogen-deficient state.24
The prognostic significance of vascular endothelial function has been documented in several recent studies. Brachial artery FMD has been shown to be an independent predictor of cardiovascular events, including death, and stroke in patients with CAD,25,26 and after vascular surgery in patients with peripheral vascular disease.27 Brachial FMD also demonstrates sensitivity to interventions resulting in improved cardiovascular prognosis.28 This evidence raises the possibility that short-term intervention effects on FMD, such as its augmentation in the presence of 17β-estradiol, may provide a useful marker of longer-term clinical outcomes. In other words, FMD response could be a useful biomarker, or surrogate end point, in the stratification of individual differences in risks versus benefits of HRT. The Kronos Early Estrogen Prevention Study (KEEPS) is an ongoing study in which women who are close to menopause (no more than 36 months after the cessation of menses) will be randomized to HRT or placebo.29 The ELITE Study is another recently initiated NIH-funded clinical trial, in which postmenopausal women will be randomized according to number of years since menopause, to test the hypothesis that 17β-estradiol will reduce the progression of early atherosclerosis if initiated soon after menopause when the vascular endothelium is relatively healthy. Although neither of these trials is designed to evaluate the effect of HRT on clinical end points, observations from these studies will provide further insight into the physiological effects of HRT in the early postmenopausal years.
In conclusion, age appears to be an important moderator of improved vascular endothelial function associated with short-term 17β-estradiol transdermal patch exposure. These observations add to a body of emerging evidence suggesting that the cardiovascular benefits of natural estrogen supplementation may occur only when initiated in the early postmenopausal years.
Sources of Funding
This study was supported by grant NR05281 from the National Institute of Nursing Research, National Institutes of Health, Bethesda, MD, and grant MO1-RR-30, National Center for Research Resources, Clinical Research Centers Program, National Institutes of Health.
Original received July 7, 2006; final version accepted May 7, 2007.
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vittinghoff E. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. J Am Med Assoc. 1998; 280: 605–613.
Manson JE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong ND, Crouse JR, Stein E, Cushman M; Women’s Health Initiative Investigators. Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med. 2003; 349: 523–534.
Hodis HN, Mack WJ, Lobo RA. Randomized controlled trial evidence that estrogen replacement therapy reduces the progression of subclinical atherosclerosis in healthy postmenopausal women without preexisting cardiovascular disease. Circulation. 2003; 108: e5.
Rosamond W, Flegal K, Friday G, Furie K, Go A, Greenlund K, Haase N, Ho M, Howard V, Kissela B, Kittner S, Lloyd-Jones D, McDermott M, Meigs J, Moy C, Nichol G, O’Donnell CJ, Roger V, Rumsfeld J, Sorlie P, Steinberger J, Thom T, Wasserthiel-Smoller S, Hong Y; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics - 2007 Update: A Report from the Am Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2007; 115: e69–e171.
Dubey RK, Imthurn B, Barton M, Jackson EK. Vascular consequences of menopause and hormone therapy: importance of timing of treatment and type of estrogen. Cardiovasc Res. 2005; 66: 295–306.
Koh KK. Effects of estrogen on the vascular wall: vasomotor function and inflammation. Cardiovasc Res. 2002; 55: 714–726.
Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D, Charbonneau F, Creager MA, Deanfield J, Drexler H, Gerhard-Herman M, Herrington D, Vallance P, Vita J, Vogel R; International Brachial Artery Reactivity Task Force. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery: a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol. 2002; 39: 257–265.
Herrington DM, Espeland MA, Crouse JR III, Robertson J, Riley WA, McBurnie MA, Burke GL. Estrogen replacement and brachial artery flow-mediated vasodilation in older women. Arterioscler Thromb Vasc Biol. 2001; 21: 1955–1961.
Koh KK, Blum A, Hathaway L, Mincemoyer R, Csako G, Waclawiw MA, Panza JA, Cannon RO 3rd. Vascular effects of estrogen and vitamin E therapies in postmenopausal women. Circulation. 1999; 100: 1851–1857.
Kalantaridou SN, Naka KK, Papanikolaou E, Kazakos N, Kravariti M, Calis KA, Paraskevaidis EA, Sideris DA, Tsatsoulis A, Chrousos GP, Michalis LK.. Impaired endothelial function in young women with premature ovarian failure: normalization with hormone therapy. J Clin Endocrinol Metab. 2004; 89: 3907–3913.
Wakatsuki A, Okatani Y, Ikenoue N, Fukaya T. Effect of medroxyprogesterone acetate on endothelium-dependent vasodilation in postmenopausal women receiving estrogen. Circulation. 2001; 104: 1773–1778.
Gerhard M, Walsh BW, Tawakol A, Haley EA, Creager SJ, Seely EW, Ganz P, Creager MA. Estradiol therapy combined with progesterone and endothelium-dependent vasodilation in postmenopausal women. Circulation. 1998; 98: 1158–1163.
Koh KK, Sakuma I. Should progestins be blamed for the failure of hormone replacement therapy to reduce cardiovascular events in randomized controlled trials?. Arterioscler Thromb Vasc Biol. 2004; 24: 1171–1179.
Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res. 2002; 53: 605–619.
Heitzer T, Schlinzig T, Krohn K, Meinertz T, Munzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation. 2001; 104: 2673–2678.