FERTILITY AND STERILITY VOL. 70, NO. 2, AUGUST 1998 Copyright 1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Prospective, randomized, controlled study of the effect of hormone replacement therapy on peripheral blood flow velocity in postmenopausal women Tze K. Lau, M.R.C.O.G.,* Din Wan, M.B., B.S.,* So F. Yim, M.R.C.O.G.,* John E. Sanderson, M.D., and Christopher J. Haines, M.D.,* The Chinese University of Hong Kong, The Prince of Wales Hospital, Shatin, Hong Kong Objective: To investigate the effect of different types and methods of delivery of hormone replacement therapy (HRT) on peripheral vascular flow velocity in postmenopausal women. Design: A prospective, randomized, operator-blinded, controlled study. Setting: A hormone replacement clinic in a university teaching hospital. Patient(s): Sixty-eight women who had undergone surgical menopause. Intervention(s): No treatment, oral estrogen, continuous combined estrogen and progestogen, or percutaneous estrogen. Main Outcome Measure(s): The pulsatility indices of the brachial, dorsalis pedis, popliteal, and radial arteries were measured under standardized conditions before the commencement of HRT and after 2 and 6 months of treatment. Serum E 2 levels were measured at each visit. Result(s): There was an inverse correlation between the serum E 2 levels and the pulsatility indices. There was a significant reduction in the pulsatility index in at least one of the four arteries after 2 months of HRT in all the treatment groups but not in the control group. The effect of HRT on the pulsatility index persisted until the completion of the study in all the treatment groups. Conclusion(s): These results confirm that the administration of HRT is associated with a reduction of the pulsatility index, and hence an increase in blood flow in the peripheral arteries; this change in the pulsatility index is related directly to serum E 2 levels. The percutaneous route of administration of estrogen was at least as effective as oral treatment in improving peripheral vascular flow velocity. The beneficial effect of estrogen was not affected by the addition of a progestogen. (Fertil Steril 1998;70:284 8. 1998 by American Society for Reproductive Medicine.) Key Words: Hormone replacement therapy, peripheral artery, pulsatility index Received November 12, 1997; revised and accepted March 6, 1998. Supported by the Hong Kong Obstetrical and Gynaecological Trust Fund, Hong Kong. Reprint requests: Tze K. Lau, M.R.C.O.G., Department of Obstetrics and Gynaecology, The Prince of Wales Hospital, Shatin, New Territories, Hong Kong (FAX: 852-2636-0008; E-mail: tzekinlau@cuhk.edu.hk). * Department of Obstetrics and Gynaecology. Department of Medicine. 0015-0282/98/$19.00 PII S0015-0282(98)00148-4 Menopause is associated with an increase in the incidence of cardiovascular disease (1), but this can be prevented in part by the administration of hormone replacement therapy (HRT) (2 4). Approximately one third of the cardioprotective effect of estrogen is thought to be due to beneficial changes in circulating lipids and lipoproteins (5). Another important mechanism is a direct effect of estrogen on blood vessels, where it has been shown to produce vasodilatation and improve blood flow. Several methods have been used to measure changes in blood flow after the administration of HRT, one of which is the measurement of the pulsatility index with the use of Doppler ultrasound (US). With this method, significant reductions in the pulsatility index have been observed in the internal carotid arteries (6, 7) and an increase in peak flow velocity has been found in the aorta (8). However, there is only limited information on the effect of HRT on the peripheral vessels, which are of smaller caliber and should reflect the behavior of resistance vessels better than the larger vessels previously studied (9). Almost all studies of the effect of estrogen on peripheral flow velocity have been conducted using oral estrogen. The use of percutaneous estrogen gels and transdermal patches are also popular methods of delivering estrogen for the purpose of HRT, but there is limited information regarding the effect of parenteral administration on flow velocity (9). Different 284
methods of delivery induce variable changes in the lipid profile (10), and this presumably is due to the hepatic firstpass effect. It is not known whether the route of administration of estrogen affects blood flow. Postmenopausal women who still have a uterus should receive a progestogen in addition to estrogen to prevent endometrial hyperplasia (11). The results of earlier studies suggested that the addition of a progestogen negated the beneficial effect of estrogen on lipids and lipoproteins, but several more recent reports have concluded that the use of the newer progestogens produces little effect on lipids (12), and continuous combined HRT appears to confer no deleterious effect on lipids and lipoproteins (13). Although progestogens have been investigated extensively with respect to the lipid profile, there is no information on whether progestogens influence the effect of estrogen on blood flow. In a previous small observational study, we found that the administration of oral estrogen with or without a progestogen was associated with a significant reduction in the pulsatility index in the peripheral arteries (14). However, this was a small pilot study in which there was no control group and no blinding. The aim of the present study was to substantiate the results of the original study. MATERIALS AND METHODS Study Design We conducted a prospective, randomized, operatorblinded, controlled study of the effect of HRT on the peripheral arteries in postmenopausal women. The study protocol was approved by the clinical research ethics committee of the local institution. Each subject was randomly allocated to one of four groups for a period of 6 months. Doppler US studies were performed at the time of recruitment and after 2 and 6 months of treatment. These periods were chosen as indicators of the acute and more prolonged effects of treatment. At the time of each flow study, blood was taken for the measurement of E 2. All subjects were referred back to the Hormone Replacement Clinic after the study for continuing care. Each group consisted of 17 patients. Based on the results of our previous study (14), the sample size calculation suggested that 8 patients were needed in each group to detect a mean ( SD) reduction in the pulsatility index with treatment of 13.9 8.6 compared with a mean change of zero in the control group, with 80% power and an alpha error of 0.05. An extra 9 patients in each group were recruited to overcome possible error in the assumptions used in the sample size calculation and to allow for dropouts. Wilcoxon s signed-rank test was used to examine the difference in the pulsatility index before and after treatment. The linear trend across different levels of serum E 2 (see later for details of categorization of E 2 level) was tested by the one-way analysis of variance (ANOVA). The Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL) was used for the analysis of all data. P.05 was considered statistically significant. Patients Patients were recruited from the Hormone Replacement Therapy Clinic of The Prince of Wales Hospital in Hong Kong. All patients newly referred to the clinic who had undergone surgical menopause were recruited. This included those who had undergone a total abdominal hysterectomy and bilateral salpingo-oophorectomy for benign gynecologic conditions before natural menopause. We selected women who had undergone a hysterectomy to avoid the problem of bleeding with the use of unopposed estrogen. Exclusion criteria included [1] the use of any form of HRT in the preceding 3 months, [2] a history of hypertension or cardiovascular disease, [3] diabetes mellitus, [4] a contraindication to HRT, and [5] smoking. Written informed consent was provided by all patients who agreed to participate in the study. Treatment Plan The patients were assigned to one of four groups according to computer-generated random numbers. The treatment protocols were as follows: 1. No treatment (controls). 2. Oral E 2, 2 mg/d (Estrofem; Novo Nordisk, Bagsvaerd, Denmark). 3. Oral E 2, 2 mg/d, and norethisterone acetate, 1 mg/d (Kliogest; Novo Nordisk). 4. Percutaneous E 2, 2.5 g/d (Oestrogel; Besins-Iscovesco, Paris, France). Doppler Studies The peripheral arteries were studied using color duplex pulsed Doppler US (Aloka S-680) with a steered 7.5-MHz peripheral vascular US probe (Aloka UST-5518-7.5). All studies were performed by the same investigator in a quiet room with constant light and a room temperature between 21 C and 22 C. Patients fasted for at least 2 hours before the study and rested in a supine position for 15 minutes before any flow velocity measurements were taken. The follow-up studies were performed at a similar time of day as the initial studies, and the investigator who performed the flow studies was blinded to the treatment that the patients received. Before the study was initiated, the US equipment settings (including preprocessing, postprocessing, and sampling volume of the Doppler gate) were tested for each of the peripheral arteries to determine the optimal configuration for the examinations. The same settings then were used on each occasion for all patients. The pulsatility indices of four peripheral arteries on the right side of the body were examined using methods previously described (14). The pulsatility index is defined as (S-D)/A, where S is the peak systolic frequency shift, D is FERTILITY & STERILITY 285
TABLE 1 Characteristics at recruitment of 68 patients who had undergone surgical menopause. Characteristic No treatment (n 14) Oral estrogen (n 15) Combined estrogen and progestogen (n 13) Percutaneous estrogen (n 16) Age 44.8 4.1 43.8 3.6 45.4 5.4 45.1 4.6 Duration of surgical menopause (y) 0.49 0.15 0.47 0.22 0.41 0.11 0.54 0.21 Pulsatility index at recruitment Brachial artery 9.3 6.2 9.0 7.8 8.4 4.0 7.9 2.7 Dorsalis pedis artery 19.0 13.8 25.1 30.4 21.3 15.8 17.7 11.5 Popliteal artery 14.4 7.4 12.9 9.5 17.3 16.6 14.0 9.3 Radial artery 6.9 4.6 7.9 5.5 9.7 8.3 11.9 8.5 the end-diastolic frequency shift, and A is the temporal mean frequency shift over one cardiac cycle. The reproducibility of the Doppler measurements was tested before the commencement of the study. The investigator measured the pulsatility index of the brachial artery on a volunteer 10 times, 10 minutes apart, under the conditions described earlier. The mean ( SD) of the pulsatility index was 3.24 0.15 (range, 2.99 3.48). Serum E 2 Assessment The serum E 2 level was measured in all patients at each visit. A total of 259 samples were available and included both control and treated patients. Because of the limitation of the assay used, samples with E 2 levels of 90 pmol/l were reported as 90 pmol/l instead of as a numeric result. Therefore, E 2 concentrations were categorized into groups (e.g., 101 pmol/l, 101 200 pmol/l, 201 200 pmol/l) for subsequent analysis. RESULTS A total of 68 patients were recruited, with 17 in each study group. There were 4 dropouts, 1 from the control group, 2 from the continuous combined treatment group, and 1 from the percutaneous estrogen group. Each of these patients had completed only the first blood flow study. Five patients, 1 from the control group, 2 from the oral estrogen group, and 2 from the combined treatment group, were excluded from data analysis because their serum hormonal profiles at recruitment were not compatible with the menopausal state. One further patient from the control group was excluded because there was a substantial increase in the E 2 concentration (to 1,500 pmol/l) at the first follow-up visit, most likely due to the use of an exogenous supplement. There was no statistically significant difference between the four groups in age, duration of menopause, or the baseline pulsatility index of all four arteries studied at the time of recruitment (Table 1). Tables 2 and 3 show the difference in the pulsatility indices after 2 months and 6 months of treatment, respectively. There was an initial reduction in the pulsatility index in one or more arteries in all the treatment groups at the first follow-up study, but not in the control group. There was little change in the pulsatility index between 2 and 6 months of treatment. There were significant reductions in the pulsatility index in one or more arteries at TABLE 2 Paired difference in the pulsatility index after 2 months of hormone replacement therapy. Artery Brachial Dorsalis pedis Popliteal Radial Control (n 14) 3.1 6.7 3.3 17.5 1.8 9.3 1.2 3.0 Oral estrogen (n 15) 4.8 7.8* 14.7 30.9* 6.0 9.5* 2.9 5.5 Continuous combined estrogen and progestogen (n 13) 2.2 3.2* 4.6 10.2 5.2 21.1 1.7 5.8 Percutaneous estrogen (n 16) 1.7 2.7* 2.1 13.2 1.9 14.7 6.2 7.4* * P.05 (versus pretreatment pulsatility index). 286 Lau et al. HRT and peripheral arterial flow Vol. 70, No. 2, August 1998
TABLE 3 Paired difference in the pulsatility index after 6 months of hormone replacement therapy. Artery Brachial Dorsalis pedis Popliteal Radial Control (n 14) 3.4 7.7 5.3 19.1 2.9 8.8 1.8 5.2 Oral estrogen (n 15) 1.9 6.7 14.5 34.8 4.1 8.8* 1.1 9.3 Continuous combined estrogen and progestogen (n 13) 2.2 2.0* 9.7 23.1 7.2 17.3 3.3 8.0 Percutaneous estrogen (n 16) 1.6 3.7 6.7 12.5* 5.2 10.6* 6.3 9.9* * P.05 (versus pretreatment pulsatility index). 6 months compared with pretreatment measurements in all the treatment groups, but not in the control group. The reduction in the pulsatility index at 6 months was demonstrated in three of four arteries in the group that received percutaneous estrogen, but in only one of four arteries in the other treatment groups. There was a significant linear trend toward a lower pulsatility index with an increasing level of serum E 2 in three of the four arteries studied, namely the brachial artery (P.013), the dorsalis pedis artery (P.005), and the popliteal artery (P.001). An example using the dorsalis pedis artery is shown in Figure 1. The reduction in the pulsatility index appeared to be most FIGURE 1 Relation between the pulsatility index of the dorsalis pedis artery and different serum E 2 concentrations (total sample, n 259). There was a trend toward a lower pulsatility index with an increasing E 2 level (P.005, ANOVA with linear trend). Solid circles represent means, with 95% confidence interval represented by the bars. profound after 2 months of treatment, because there was a significant reduction in the pulsatility index in three of the four arteries after 2 months of treatment compared with only one of the four arteries after 6 months of treatment. In contrast, the reduction in the pulsatility index appeared to be most profound after 6 months of treatment among those who received percutaneous estrogen. DISCUSSION This is the first randomized study on the effect of HRT on peripheral vascular flow. Previous observational studies have demonstrated a beneficial effect on the Doppler parameters of the central arteries after HRT in postmenopausal women, with either a reduction in the pulsatility index or an increase in peak flow velocity (6 8). Gilligan et al. (9) reported an improvement in endothelial function in the peripheral blood vessels after the acute intra-arterial administration of E 2. However, in most of these studies, there was a lack of controls and measurement bias could not be excluded because the treatment the patients received was known to the operator. The pulsatility index is an indirect measurement of the vascular resistance and therefore of blood flow (15). To minimize errors in Doppler examinations, all measurements in this study were performed under standard conditions and environments by one operator, in whom the intra-observer error of measurement had been found to be only 4.6%. We found that there was a significant reduction in the pulsatility index after 2 months of treatment in at least one vessel in all the patients who received HRT, but not in the control patients. Similar results were obtained after 6 months of treatment. However, the reduction in the pulsatility index after 2 months was most convincing with oral treatment, as significant changes were demonstrated in three of four vessels, compared with only one of four vessels with continuous combined or percutaneous treatment. After 6 months, however, there were significant reductions in the pulsatility index in three of four vessels with percutaneous estrogen, but in FERTILITY & STERILITY 287
only one vessel with the other treatment regimens. It thus appeared that oral HRT had a more profound short-term effect on the peripheral arteries. The difficulties involved in conducting these studies did not allow for a definite conclusion regarding whether percutaneous treatment confers the greatest long-term benefit, although this appears to be the case. The level of E 2 correlated negatively with the pulsatility index, and this suggests that the concentration of circulating E 2 is a major determinant of the reduction in the pulsatility index regardless of the route of administration of treatment. To our knowledge, this is the first study to investigate the effect of the different methods of estrogen administration on peripheral vascular flow velocity. Estrogen receptors have been reported to be present in the vessel wall (16), and E 2 can initiate the release of nitric oxide, a potent vasodilator (17). We demonstrated that the percutaneous method of treatment is at least as effective as oral therapy. Our results also support the hypothesis that estrogen replacement acts directly on the vessel wall and causes vasodilatation. The effect of the addition of progestogen also was studied. Reductions in the pulsatility index were found in at least one vessel at both 2 and 6 months in patients who received continuous combined HRT. At 6 months, the magnitude of the reduction was similar to that observed with the use of unopposed estrogen, suggesting that the addition of the progestogen does not significantly affect the blood flow increase in response to oral estrogen treatment. We conclude that HRT is associated with a reduction in the pulsatility index of the peripheral arteries, which may represent a direct vasodilator effect of estrogen on the peripheral vasculature. This change appears to take place regardless of the route of administration of estrogen and does not appear to be influenced by the addition of a progestogen. References 1. Sarrel PM. Ovarian hormones and the circulation. Maturitas 1990;12: 287 8. 2. Grady D, Rubin SM, Petitti DB, Fox CS, Black D, Ettinger B, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117:1016 37. 3. Stampfer MJ, Colditz GA, Willett WC, Manson JE, Rosner B, Speizer FE, et al. Postmenopausal estrogen therapy and cardiovascular disease. N Engl J Med 1991;325:756 62. 4. Bush TL, Barrett-Connor E, Cowan LD, Criqui MH, Wallace RB, Suchindran CM, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-up Study. Circulation 1987;75:1102 9. 5. Orimo A, Inoue S, Ikegami A, Hosoi T, Akishita M, Ouchi Y, et al. Vascular smooth muscle cells as target for estrogen. Biochem Biophys Res Commun 1993;195:730 6. 6. Gangar KF, Vyas S, Whitehead M, Crook D, Meire H, Campbell S. Pulsatility index in internal carotid artery in relation to transdermal oestradiol and time since menopause. Lancet 1991;338:839 42. 7. Penotti M, Nencioni T, Gabrielli L, Farina M, Castiglioni E, Polvani F. Blood flow variations in internal carotid and middle cerebral arteries induced by postmenopausal hormone replacement therapy. Am J Obstet Gynecol 1993;169:1226 32. 8. Pines A, Fisman EZ, Levo Y, Averbuch M, Lidor A, Drory Y, et al. The effects of hormone replacement therapy in normal postmenopausal women: measurements of Doppler-derived parameters of aortic flow. Am J Obstet Gynecol 1991;164:806 12. 9. Gilligan DM, Badar DM, Panza JA, Quyyumi AA, Cannon III RO. Effects of estrogen replacement therapy on peripheral vasomotor function in postmenopausal women. Am J Cardiol 1995;75:264 8. 10. Crook D, Cust MP, Gangar KF, Worthington M, Hillard TC, Stevenson JC, et al. Comparison of transdermal and oral estrogen/progestin hormone replacement therapy: effects on serum lipids and lipoproteins. Am J Obstet Gynecol 1992;166:950 5. 11. Paterson M, Wade-Evans T, Sturdee DW, Thom MH, Studd JWW. Endometrial disease after treatment with oestrogens and progestogens in the climacteric. Br Med J 1980;280:822 4. 12. Lobo AL, Pickar JH, Wild RA, Walsh B, Hirvonen E. Metabolic impact of adding medroxyprogesterone acetate to conjugated estrogen therapy in postmenopausal women. Obstet Gynecol 1994;84:987 95. 13. Writing Group for PEPI Trial: Effects of estrogens or estrogen/progestin regimens on heart disease risk factors in postmenopausal women: the postmenopausal estrogen/progestin trial. JAMA 1995;273:199 208. 14. Lau TK, Chung TKH, Haines CJ. A preliminary study on the effect of hormone replacement therapy on peripheral flow velocity in postmenopausal women. Maturitas 1997;26:53 6. 15. Woodcock JP, Gosling RG, Fitzgerald DE. A new noninvasive technique for assessment of superficial femoral arterial obstruction. Br J Surg 1972;59:226 31. 16. Lin AL, Shain SA. Estrogen-mediated cytoplasmic and nuclear distribution of rat cardiovascular estrogen receptors. Arteriosclerosis 1985; 5:668 77. 17. Gislard V, Millar V, Van Houte P. Effects of 17 -oestradiol on endothelium-dependent responses in the rabbit. J Pharmacol Exp Ther 1988;244:19 22. 288 Lau et al. HRT and peripheral arterial flow Vol. 70, No. 2, August 1998