Effect of postmenopausal hormone replacement on atherosclerosis in femoral arteries

Transcription

1 Maturitas 41 (2002) Effect of postmenopausal hormone replacement on atherosclerosis in femoral arteries Peter Angerer *, Stefan Störk, Wolfgang Kothny, Clemens von Schacky Klinikum der Uni ersität München, Institut und Poliklinik für Arbeits- und Umweltmedizin, Medizinische Klinik-Innenstadt, Ziemssenstraße 1, München, Germany Received 26 January 2001; received in revised form 9 July 2001; accepted 16 July 2001 Abstract Objecti es: On the basis of epidemiological and experimental data, it has been supposed that hormone replacement therapy (HRT) inhibits atherosclerosis in postmenopausal women. This randomized controlled trial examined whether 1 mg 17 -estradiol daily, combined cyclically with mg gestodene in every month (HRT 1), or in every third month (HRT 2) slows the increase of intima-media thickness in femoral arteries compared with no HRT. Methods: Healthy postmenopausal women (n=321) with an increased risk for future vascular disease as indicated by 1 mm of intima-media thickness in the carotid arteries were equally randomized to one of the three groups for 48 weeks. Ultrasound scans of femoral arteries were recorded at study start and study end, together with a thorough clinical examination and laboratory work-up. Results: Complete scans were obtained in 260 of the 264 subjects who participated until study end. Mean maximum intima-media thickness of four femoral artery segments (common and superficial, both sides) was mm (mean S.D.) at study start. It increased by , , and mm in the HRT 1, HRT 2 and no HRT groups, respectively (HRT 1 versus no HRT, HRT 2 versus no HRT; both P 0.2). Compared with no HRT, HRT significantly lowered follicle stimulating hormone, low-density lipoprotein cholesterol, and fibrinogen. Conclusions: In this 1-year trial, irrespective of the progestogen dose used, HRT with 1 mg 17 -estradiol did not inhibit progression of femoral artery atheroslerosis in postmenopausal women with subclinical vascular disease Elsevier Science Ireland Ltd. All rights reserved. Keywords: 17 -Estradiol; Hormone replacement therapy; Atherosclerosis; Femoral arteries; Intima-media thickness; Randomized controlled trial Abbre iations: ABI, ankle/brachial (systolic blood pressure) index; CAD, coronary artery disease; FSH, follicle stimulating hormone; HDL, high-density lipoprotein cholesterol; HRT, hormone replacement therapy; IMT, intima-media thickness; LDL, low-density lipoprotein cholesterol; PAD, peripheral arterial disease. * Corresponding author. Tel.: ; Fax: address: peter.angerer@arbeits.med.uni-muenchen.de (P. Angerer) /02/$ - see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S (01)

2 52 P. Angerer et al. / Maturitas 41 (2002) Introduction Although several investigations have addressed the question of whether hormone replacement therapy (HRT) for postmenopausal women slows progression of atherosclerosis, the answer still remains unclear. Some epidemiological studies suggested a reduced risk for angiographically significant coronary artery disease [1], and for atherosclerosis in the carotid arteries [2 4], whereas others observed no beneficial effect of HRT [5]. Randomized controlled trials demonstrated that oestrogen alone or combined with different doses of progestogen does not inhibit progression of atherosclerosis in coronary or carotid arteries of postmenopausal women [6,7]. Data on HRT and atherosclerosis in peripheral arteries of postmenopausal women are scarce. Recently, in the Rotterdam Study, a lower risk for haemodynamically relevant peripheral arterial disease in previous long-term users of HRT was observed [8]. However, in the randomized Heart and Oestrogen/Progestogen replacement study (HERS), peripheral arterial disease (PAD) events were not reduced by HRT [9]. PAD is frequent in postmenopausal women, increasing with age [8,10]. It causes pain [8,11] and it indicates a substantially increased cardiovascular mortality [12]. Involvement of proximal (e.g. femoral) arteries was associated with a higher mortality than involvement of distal (e.g. tibial) arteries [13]. In this randomized, controlled trial, we investigated the effect of 17 -estradiol combined with standard dose and with low dose progestogen (gestodene) on progression of atherosclerosis during 48 weeks as monitored by serial ultrasound measurements. In the present paper, we report on the progression of atherosclerosis in the femoral arteries. The hypothesis was that the increase of intima-media thickness would be smaller in both HRT groups compared with the control group. 2. Methods 2.1. Subjects and design The Postmenopausal Hormone REplacement against Atherosclerosis (PHOREA) trial was randomized, controlled, observer blind, and conducted according to the International Conference for Harmonisation guidelines for Good Clinical Practice (ICH-GCPP). PHOREA investigated the effect of two different HRT regimens compared with no HRT on intermediate endpoints of atherosclerotic disease and on risk factors for atherogenesis. The primary outcome measure of the part of the trial that examined progression of atherosclerosis was the averaged change of intima-media thickness (IMT) of six segments of the carotid arteries. The change of IMT in the femoral arteries was the secondary outcome measure. The study design and the results of the primary outcome measure have been reported in detail elsewhere [7]. Eligibility criteria included female gender, postmenopausal status as defined by last regular bleeding or surgical menopause 1 year ago; or, in the case of hysterectomy, follicle stimulating hormone (FSH) levels 40 IU/l, age between 40 and 70 years, and IMT 1 mm in at least one of the predefined segments of the carotid arteries as detected by high-resolution ultrasound. Excluded were women with malignant disease, those for whom HRT was necessary or was intolerable, women with myocardial infarction within the past 6 months or with coronary artery disease (CAD) that required treatment for angina, history of ischaemic cerebrovascular disease, blood pressure greater than 200/110 mmhg after 5 min rest, with any other contraindication against HRT or with any other serious illness. All subjects gave informed consent. The study was approved by the ethics committee of the faculty of medicine of the University of Munich. Subjects were randomly allocated to three groups. The HRT 1 group received tablets containing 1 mg 17 -estradiol each day, with addition of mg gestodene on days of each 4-week cycle (standard dose progestogen). The HRT 2 group received oestrogen following the same regimen, but gestodene was added in each third cycle only (low dose progestogen). The no HRT group received no oestrogen and no progestogen at all. The duration of treatment was 48 weeks, corresponding to 12 cycles. An indepen-

3 P. Angerer et al. / Maturitas 41 (2002) dent clinical research organization (Biometrisches Zentrum für Therapiestudien, Munich, Germany) monitored the adherence to the study protocol, verified all source data, and provided the verified database. It also ensured blinding of the ultrasound reader. Subjects were seen in the university hospital outpatient centre at study start, and in weeks 12, 22, and 48. At study start and study end, arterial IMT was documented by ultrasound examination. At each visit, history was taken and a physical examination was performed, including blood pressure measurements in recumbent position at rest and extensive laboratory work-up. General health advice and advice for treatment of risk factors followed the guidelines of the American Heart Association [14]. The study medication was dispensed at each visit, and all blisters and all remaining tablets were collected during subsequent visits. Subjects documented daily study drug intake and vaginal bleeding in a diary. Compliance with medication was assessed by this diary, by FSH measurement, and by pill count. The intention-to-treat population was defined as all randomized subjects who completed both ultrasound examinations. The valid case population was defined as all subjects who did not violate any eligibility criteria at randomization or during the study, took at least 91.7% (11 cycles) of the study medication and did not use any additional HRT (including topical application) during the trial Ultrasound outcome measures Atherosclerosis was measured as thickness of the intimal and medial layers of the artery as visualized by high-resolution 7.5 MHz ultrasound (Apogee CX Color; ATL, Bothel, WA, USA) [15]. An RMI 413 tissue-mimicking phantom was used to monitor instrument performance. Subjects rested recumbent for the examination. For the measurement of the femoral IMT, the probe was held in an anterior posterior position, and two segments were visualized at both sides: the distal 10 mm of the common femoral artery, and the proximal 10 mm of the superficial femoral artery. The optimal longitudinal view of the maximum IMT in sufficient quality, i.e. clear bloodintima and media-adventitia boundaries over the full length of the segment, was recorded digitally. The examination was repeated at study end according to the same protocol, using the same angles of interrogation. Only the far wall was considered for calculation of outcome measures since ultrasound measurement of the near wall underestimates the true histological thickness and proper visualization is often difficult [5,15]. The measurement of the carotid IMT has been described previously [7]. In brief, for the common and internal carotid artery and the carotid bifurcation, left and right side, the maximum IMT of the far wall was determined at study start and at study end, using the same angles of interrogation. To ensure blindness of sonographers with respect to treatment, their contact with the participant was limited to the ultrasound examination. On all recordings, a five-digit random number replaced the name of the subject and the date of examination. The keyed list containing names, dates and five-digit numbers was stored at the clinical research organization until the end of the trial. Sonographers (W.K. and P.A.) had completed a training program with more than 1000 scans of carotid and femoral arteries prior to the trial. All readings were carried out by W.K. to avoid interobserver variation. In addition, to assess reproducibility, scans were performed in 30 subjects in identical manner as already described within 1 week after the baseline or follow-up scan and evaluated blindly. Reproducibility for the femoral arteries, expressed as the mean difference between two measurements and its standard deviation [16], was: mean maximum IMT, mm; single maximum IMT, mm. IMT was measured twice from the digitized image, at the site of its maximum thickness within each segment, and both results were averaged. The software NIH-Image (National Institutes of Health, Bethesda, USA) was used, on a Power Macintosh 8100/80 with a high-resolution screen. The mean of the maximum far wall IMT values for both femoral artery segments on both sides

4 54 P. Angerer et al. / Maturitas 41 (2002) was calculated as mean maximum IMT per subject. The highest IMT of all femoral segments was the single maximum IMT per subject. In addition, the mean of the maximum far wall IMT values of both sides was calculated for the common and the superficial femoral artery per subject separately Statistical analysis Sample size estimation was based on similar assumptions as for the study on carotid atherosclerosis progression [7]: with reference to what had been demonstrated in another primary prevention trial, the annual atherosclerotic progression rate of IMT was estimated to be mm in untreated women [17]. HRT was estimated to result in a 50% slowing of progression. To achieve a power of 90% and a P value of 0.05, 80 subjects per group were required, resulting in 320 subjects altogether after adding the estimated drop-out rate of 33% [18]. Normality and homogeneity of variances of the outcome measures were assessed by Lilliefors test and Levene s test, respectively. Differences between HRT 1 and no HRT, and between HRT 2 and no HRT, were examined by independent sample t-test or chi-squared test, according to the nature of the data. For the analysis of the ultrasound outcome measures, the P value was adjusted for multiple comparisons according to the Bonferroni Holm method. For other outcome measures, no adjustment was made since the analysis was exploratory. A P value 0.05 was defined as significant. All calculations were performed on a Power Macintosh 7600/120 using SPSS (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Baseline data Of 321 subjects randomized, 264 remained in the study and had a second ultrasound examination after 48 weeks (intention-to-treat population). Their baseline characteristics are presented in Table 1. In 260 of 264 subjects, both scans of the femoral artery IMT were completed and readable. In the no HRT group, 72 subjects were considered as valid cases, 69 in the HRT 1 group, and 73 in the HRT 2 group (i.e. 214 of 264) according to the criteria described in Section 2 (valid case population). Of these 214, 210 had complete femoral artery ultrasound scans. Fifty-seven subjects ended participation prematurely. Compared with subjects who completed the trial in the respective groups, those who stopped in the no HRT group had a higher number of hysterectomies and longer duration of previous HRT, and those who stopped in the HRT 2 group had a lower number of hysterectomies and shorter duration of previous HRT. At baseline, in the femoral arteries mean maximum IMT was mm and single maximum IMT was mm. In the carotid arteries, the respective values were and mm. Thus, participants who discontinued did not differ from those who completed the trial until study end with respect to IMT data. Among the 264 subjects who participated until study end, no difference between treatment groups was observed at study start, except for fewer hysterectomies in the no HRT and HRT 2 group compared with HRT 1 (Table 1) Femoral artery IMT At baseline, 166 subjects (64%) had a femoral IMT of 1 mm in at least one of the four segments examined. In the intention-to-treat analysis, overall mean maximum IMT increased by mm (mean of all groups). All differences of change of the femoral IMT as presented in Table 2 were not statistically significant (all P values 0.2). Results of the valid case analysis of the femoral artery outcome measures were almost identical to the results of the intention-to-treat analysis (data not shown) Other characteristics No subject reported symptoms suggestive of PAD (e.g. claudication), neither at study start nor during the study.

5 P. Angerer et al. / Maturitas 41 (2002) The following changes were significantly different between HRT groups and no HRT (Table 3): FSH decreased by approximately onethird in the HRT groups, whereas it increased slightly in the no HRT group. Low-density lipoprotein cholesterol (LDL) and fibrinogen decreased in both HRT groups and increased slightly in the no HRT group. High-density lipoprotein cholesterol (HDL) decreased slightly in all groups with a significantly smaller change in the HRT 2 group. There was a small increase in weight in the no HRT group, whereas weight decreased by a small amount in both HRT groups. More subjects in the no HRT group took antihypertensive medication during follow-up compared with HRT 1. Fewer subjects in the HRT 1 group compared with the no HRT group were current smokers at study end. Changes of other factors with a potential effect on IMT (plasma glucose, blood pressure, physical activity, and lipid lowering medication) were similarly distributed among treatment groups during followup (Table 3). Table 1 Baseline data, intention-to-treat population Variable, unit No HRT HRT 1 HRT 2 N Age, years Body mass index, kg/m Blood pressure systolic, mmhg Blood pressure diastolic, mmhg Coronary artery disease, n 5 (5.4%) 1 (1.2%) 1 (1.2%) Diabetes mellitus, n 5 (5.4%) 4 (4.7%) 3 (3.5%) Current smokers, n 20 (21.5%) 12 (14.0%) 12 (14.1%) Current smokers: cigarettes smoked per day Physical activity, h per week Subjects on antihypertensive drugs, n 27 (29%) 19 (22.1%) 29 (34.1%) Subjects on lipid lowering drugs, n 6 (6.5%) 5 (5.8%) 6 (7.1%) LDL cholesterol (P), mmol/l (mg/dl) ( ) ( ) ( ) HDL cholesterol (P), mmol/l (mg/dl) ( ) ( ) ( ) Triglycerides (P), mmol/l (mg/dl) ( ) ( ) ( ) Fibrinogen (P), g/l (mg/dl) (271 60) (287 47) (287 56) Glucose (P), mmol/l (mg/dl) ( ) ( ) ( ) Follicle stimulating hormone (S), IU/l Years since menopause/hysterectomy Ever user of HRT, n 61 (66.3%) 59 (68.6%) 65 (76.5%) Ever user of HRT, months of use Hysterectomy, n 21 (22.6%)* 41 (47.7%)* 28 (32.9%)* Mean maximum IMT in CA, mm Single maximum IMT in CA, mm Mean maximum IMT in FA, mm Single maximum IMT in FA, mm Data with plus minus values are means S.D. CA, Carotid arteries; FA, femoral arteries; P, plasma; S, serum. * Comparison of no HRT versus HRT 1, P 0.001; comparison of HRT 1 versus HRT 2, P=0.050.

6 56 P. Angerer et al. / Maturitas 41 (2002) Table 2 Femoral artery IMT: absolute change from study start to study end Variable, unit No HRT HRT 1 HRT 2 N Mean maximum IMT, mm Single maximum IMT, mm Mean maximum IMT, common femoral artery, mm Mean maximum IMT, superficial femoral artery, mm Study end minus study start (positive values indicate increase), intention-to-treat analysis. HRT 1 versus no HRT and HRT 2 versus no HRT, all comparisons P 0.2. Data with plus minus values are means S.D Exploratory subgroup analysis In order to detect any influence of uneven distribution of baseline characteristics on change of IMT in femoral arteries, exploratory analysis was performed in the following subgroups: previous hysterectomy (yes/no), history of current smoking (yes/no), antihypertensive medication at any time during the study (yes/no), and lipid lowering medication at any time during the study (yes/no). In any subgroup, change of femoral artery IMT was not different between HRT groups and the no HRT group Relationship between progression of carotid and femoral atherosclerosis Change of the mean maximum IMT of the six carotid segments and of the two femoral segments correlated significantly, in all subjects (r=0.36, P 0.001): in HRT 1, r=0.30, P=0.007; in HRT 2, r=0.44, P 0.001; in no HRT, r=0.39, P Discussion Combined sequential HRT with 17 -estradiol and gestodene for 48 weeks did not slow progression of subclinical atherosclerosis in femoral arteries. In this randomized controlled trial, all subjects were at increased risk for vascular disease events as defined by the inclusion criterion of 1 mm IMT in at least one segment of the carotid arteries [19,20]. In addition, 64% of subjects had 1 mm IMT of the femoral arteries and most had vascular risk factors. LDL and fibrinogen decreased by HRT within the expected range. No difference in progression was observed between the standard dose and the low dose progestogen regimen. Subjects in this study had atherosclerosis but less than 3% had experienced any clinical event (e.g. myocardial infarction or coronary intervention). From the clinical point of view, this represents a primary prevention situation. However, these subjects carry a high risk for future vascular disease events [19,20]. On the basis of epidemiological observations, it was suggested that healthy women at increased cardiovascular risk might benefit most from HRT [21]. For women with manifest CAD, two large randomized intervention trials recently demonstrated that conjugated equine estrogens combined with medroxyprogesterone acetate do not prevent cardiovascular events [22] and do not slow progression of angiographically observed coronary atherosclerosis [6]. The latter finding was independent from the addition of medroxyprogesterone acetate [6], which has been accused of antagonizing the beneficial effect of estrogens [21 23]. In response to these new findings, the perspective has shifted to HRT as a means of primary prevention [24]. However, there are no trials with clinical endpoints to corroborate this concept. We and other workers have recently shown that HRT does not slow progression of carotid atherosclerosis in clinically healthy women [7,25]. Pooled data from randomized trials that focused mostly on bone-related outcome measures in primarily healthy women showed a non-significantly higher odds ratio for cardiovascular events in women taking HRT compared

7 P. Angerer et al. / Maturitas 41 (2002) with those taking placebo [26]. Peripheral artery atherosclerosis is an established independent intermediate endpoint for cardiovascular morbidity and mortality [13,27,28]. The lack of a favourable effect in the present trial contributes to the recent findings that do not support the use of HRT in primary prevention of cardiovascular disease. In addition to its role as intermediate end point for vascular disease mortality, atherosclerosis of the lower extremities is a disease entity of its own. The highly significant correlation between progression of carotid and femoral IMT indicates that both can be understood as sequelae of a common generalized disease. However, only 13% (r 2 ) of the variation of femoral artery IMT progression could be explained by carotid artery IMT progression. To the best of our knowledge, the present study is the only randomized trial to date that examined the effect of HRT on femoral artery atherosclerosis. It is in line with the results of HERS where no reduction of incident peripheral arterial events was found [9]. Our results did not confirm observations from the epidemiological Rotterdam Study that HRT given for at least 1 year was associated with a decreased likelihood for PAD as defined by a low ankle/ brachial systolic blood pressure index (ABI) [8]. Potential bias in epidemiological studies would exaggerate the benefit of HRT [29]. Alternatively, since ABI compares Table 3 Other characteristics: absolute change from study start to study end Variable, unit No HRT HRT 1 HRT 2 P value No HRT vs No HRT vs HRT 1 HRT 2 N Weight, kg Blood pressure systolic, mmhg Blood pressure diastolic, mmhg Current smokers at last 21 (22.6%) 9 (10.6%) 10 (11.8%) follow-up, n Physical activity, h per week Subjects on antihypertensive 28 (30.1%) 14 (16.3%) 20 (23.5%) Subjects on lipid lowering drugs, 14 (15.1%) 11 (12.8%) 11 (12.9% drugs, n a n a LDL cholesterol (P) (mmol/l) (mg/dl) ( ) ( ) ( ) HDL cholesterol (P) (mmol/l) (mg/dl) ( ) ( ) ( ) Triglycerides (P) mmol/l (mg/dl) ( ) ( ) ( ) Fibrinogen (P) g/l (mg/dl) (7 41) ( 21 48) ( 20 43) Glucose (P), mmol/l (mg/dl) ( ) ( ) ( ) Follicle stimulating hormone (S), IU/l Mean maximum IMT in carotid arteries, mm Mean of all follow-up visits minus study start unless otherwise indicated (positive values indicate increase). Data with plus minus values are means S.D. a During follow up.

8 58 P. Angerer et al. / Maturitas 41 (2002) systolic blood pressure in the tibial artery and in the brachial artery, it may be influenced by several mechanisms as there is proximal atherosclerotic lumen narrowing, perfusion through collateral arteries, and stiffening of the artery that must be compressed by the blood pressure cuff [11]. HRT may influence some of these processes by vasodilation, by improved endothelial function or by reduced arterial stiffness [30 32]. Stiffness seems to be especially altered in smokers [33], who account for a large proportion of women with PAD [8]. The lower prevalence of pathological ABI was observed in women who predominantly had used unopposed oestrogen [8]. In our study, an antagonizing action of progestogen is unlikely to be responsible for oestrogen s lack of effect since no difference was demonstrated in the group that took only one-third of the standard dose progestogen. The favourable influence on LDL and fibrinogen was similar in both HRT regimens. However, the increase of IMT in femoral arteries was somewhat smaller in the HRT groups than in the no HRT group. Sample size had been estimated for the expected change of IMT in the carotid arteries, which was the primary outcome measure [7]. Consequently, if we speculate that atherosclerosis in femoral arteries progresses more slowly than in carotid arteries, a larger sample size or a longer duration of treatment would have been necessary. If we hypothesize that the observed difference was not explained by chance (as we assume on the basis of the P value), a sample size of 500 subjects per group would have been necessary to obtain a power of 90%. Thus, our study may have missed an unexpectedly slow effect of HRT on femoral atheroslerosis. The dose of 17 -estradiol recommended for postmenopausal HRT is 1 or 2 mg [34]. Lipoprotein changes are similar with large or small doses of oral oestrogen [34]. In order to minimize side effects we used 1 mg 17 -estradiol. FSH decreased to levels close or below the value of 40 IU/l. When compared with standard HRT regimens composed of mg conjugated equine oestrogen plus cyclic medroxyprogesterone acetate, as studied in the Postmenopausal Oestrogen/ Progestin Interventions (PEPI) Trial [23], the HRT regimen used in the present study resulted in a slightly smaller decrease of LDL, a small decrease or no change of HDL, but a more desirable change in fibrinogen. The biological effects of the HRT used in the present trial are thus comparable with standard therapy. Other limitations of our study have been discussed in detail previously [7] and are therefore only briefly mentioned: subjects and the physicians taking care of them in the outpatient clinic were not blinded, but the ultrasound scans were performed and read by blinded personnel, thus reducing the likelihood of bias to a minimum. Visible thickening of the arterial wall and reduction of lumen by atherosclerosis is one, although important, pathomechanism of atherosclerotic disease. However, we did not study endothelial function or other phenomena that may also be important for the pathology of the vessel wall and may be favourably influenced by oestrogen [30]. In conclusion, for healthy postmenopausal women at increased risk for vascular disease events, this randomized trial demonstrated that HRT with 1 mg 17 -estradiol did not slow progression of atherosclerosis of femoral arteries. Conversely, HRT decreased LDL and fibrinogen as expected. This negative finding is in line with other recent randomized trials. Thus, evidence is mounting that oestrogen replacement does not substantially slow progression of atherosclerosis. Acknowledgements This trial was supported by a grant from Münchener Universitätsgesellschaft, München, made possible by Schering AG, Berlin. The study medication was also provided by Schering AG. References [1] Sullivan JM, Vander Zwaag R, Lemp GF, et al. Postmenopausal estrogen use and coronary atherosclerosis. Ann Intern Med 1988;108: [2] Espeland MA, Applegate W, Furberg CD, Lefkowitz D, Rice L, Hunninghake D. Estrogen replacement therapy and progression of intimal-medial thickness in the carotid arteries of postmenopausal women. ACAPS Investigators. Asymptomatic Carotid Atherosclerosis Progression Study. Am J Epidemiol 1995;142:

9 P. Angerer et al. / Maturitas 41 (2002) [3] Punnonen R, Jokela H, Heinonen PK, Aine R, Dastidar P. Hormone replacement therapy and atherosclerosis. J Reprod Med 1995;40: [4] Westendorp IC, Veld BA, Bots ML, et al. Hormone replacement therapy and intima-media thickness of the common carotid artery: the Rotterdam study. Stroke 1999;30: [5] Nabulsi AA, Folsom AR, Szklo M, White A, Higgins M, Heiss G. No association of menopause and hormone replacement therapy with carotid artery intima-media thickness. Atherosclerosis Risk in Communities (ARIC) Study Investigators. Circulation 1996;94: [6] Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med 2000;343: [7] Angerer P, Störk S, Kothny W, Schmitt P, von Schacky C. Effect of oral postmenopausal hormone replacement on progression of atherosclerosis. A randomized, controlled trial. Arterioscl Thromb Vasc Biol 2001;21: [8] Westendorp IC, in t Veld BA, Grobbee DE, et al. Hormone replacement therapy and peripheral arterial disease: the Rotterdam study. Arch Intern Med 2000;160: [9] Hsia J, Simon JA, Lin F, et al. Peripheral arterial disease in randomized trial of estrogen with progestin in women with coronary heart disease: the heart and Estrogen/ Progestin replacement study. Circulation 2000;102: [10] Vogt MT, Cauley JA, Kuller LH, Hulley SB. Prevalence and correlates of lower extremity arterial disease in elderly women. Am J Epidemiol 1993;137: [11] Weitz JI, Byrne J, Clagett GP, et al. Diagnosis and treatment of chronic arterial insufficiency of the lower extremities: a critical review. Circulation 1996;94: [12] Vogt MT, Cauley JA, Newman AB, Kuller LH, Hulley SB. Decreased ankle/arm blood pressure index and mortality in elderly women. J Am Med Assoc 1993;270: [13] Vogt MT, Wolfson SK, Kuller LH. Segmental arterial disease in the lower extremities: correlates of disease and relationship to mortality. J Clin Epidemiol 1993;46: [14] Grundy SM, Balady GJ, Criqui MH, et al. Guide to primary prevention of cardiovascular diseases. A statement for healthcare professionals from the Task Force on Risk Reduction. American Heart Association Science Advisory and Coordinating Committee. Circulation 1997;95: [15] Wong M, Edelstein J, Wollman J, Bond MG. Ultrasonicpathological comparison of the human arterial wall. Verification of intima-media thickness. Arterioscler Thromb 1993;13: [16] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1: [17] Salonen R, Nyyssonen K, Porkkala E, et al. Kuopio Atherosclerosis Prevention Study (KAPS): a populationbased primary preventive trial of the effect of LDL lowering on atherosclerotic progression in carotid and femoral arteries. Circulation 1995;92: [18] Dawson-Saunders B, Trapp RG. Estimating and comparing means. In: Basic and Clinical Biostatistics. London: Appelton and Lange, 1994:120. [19] Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study. Circulation 1997;96: [20] O Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson SK Jr. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med 1999;340: [21] Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117: [22] Hulley S, Grady D, Bush T, et al. 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: [23] The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. J Am Med Assoc 1995;273: [24] Nabel EG. Coronary heart disease in women-an ounce of prevention. N Engl J Med 2000;343: [25] de Kleijn MJ, Bots ML, Bak AA, et al. Hormone replacement therapy in perimenopausal women and 2-year change of carotid intima-media thickness. Maturitas 1999;32: [26] Hemminki E, McPherson K. Impact of postmenopausal hormone therapy on cardiovascular events and cancer: pooled data from clinical trials. Br Med J 1997;315: [27] Balbarini A, Buttitta F, Limbruno U, et al. Usefulness of carotid intima-media thickness measurement and peripheral B-mode ultrasound scan in the clinical screening of patients with coronary artery disease. Angiology 2000;51: [28] Kristenson M, Lassvik C, Bergdahl B, et al. Ultrasound determined carotid and femoral atherosclerosis in Lithuanian and Swedish men: the LiVicordia study. Atherosclerosis 2000;151: [29] Barrett Connor E, Grady D. Hormone replacement therapy, heart disease, and other considerations. Annu Rev Public Health 1998;19: [30] Mendelsohn ME, Karas RH. The protective effects of

10 60 P. Angerer et al. / Maturitas 41 (2002) estrogen on the cardiovascular system. N Engl J Med 1999;340: [31] Gerhard M, Walsh BW, Tawakol A, et al. Estradiol therapy combined with progesterone and endothelium-dependent vasodilation in postmenopausal women. Circulation 1998;98: [32] Waddell TK, Rajkumar C, Cameron JD, Jennings GL, Dart AM, Kingwell BA. Withdrawal of hormonal therapy for 4 weeks decreases arterial compliance in postmenopausal women. J Hypertens 1999;17: [33] Angerer P, Kothny W, Stork S, von Schacky C. Hormone replacement therapy and distensibility of carotid arteries in postmenopausal women: a randomized, controlled trial. J Am Coll Cardiol 2000;36: [34] Barrett Connor E. Hormone replacement therapy. Br Med J 1998;317: