Gender-Specific Differences in the Effects of Testosterone and Estrogen on the Development of Atherosclerosis in Rabbits
Abstract The aim of the present study was to investigate whether there are gender-specific differences in the effects of testosterone and estrogen on the process of atherogenesis. Thirty-two castrated male and 32 ovariectomized female rabbits were separated into 4 study groups of 8 males and 8 females each and received postoperatively a 0.5% cholesterol diet for 12 weeks. During this period either no hormones, estradiol (1 mg/kg body wt/week), testosterone (25 mg/kg body wt/week IMM), or estrogen combined with testosterone in above dosages were administered. Computerized morphometric analysis of the intimal thickening in the proximal aortic arch showed a significant inhibitory effect of estrogen in female and of testosterone in male animals (P<.05). In the group with combined treatment, the plaque size in both sexes was smaller than in the animals of the control group (P<.05). These differences were independent of changes in plasma lipid parameters. The incorporation of 5′-bromo-2′-deoxyuridine, associated with cell proliferation, into cells of the neointima was not significantly affected by the different hormone application regimens in males. In females, the incorporation rate was significantly lowered in the estrogen treated group compared with the control group (P<.05). Due to the observed differences in the sex specific atheroprotective effects of testosterone and estrogen, these data suggest that complex hormone interactions, which are independent of changes in plasma lipids, may play an important role in the process of atherogenesis.
- Received September 25, 1996.
- Accepted April 4, 1997.
Coronary heart disease (CHD) is the main cause of death in Western societies. There is a difference in incidence between men and women related to the woman’s menopausal status. While CHD is rare prior to menopause, the incidence in women 10 years after menopause is nearly identical to that of men at the same age.1 Changes in the woman’s serum estrogen level are assumed to be responsible for these differences.2 Clinical studies have shown a protective effect of estrogen replacement therapy on the development of CHD in postmenopausal women.3 This has been confirmed in experimental studies, where different effects of estrogen on the process of atherogenesis could be documented, including its influence on the serum lipid profile, direct actions on the metabolism of the vessel wall involving the deposition of cholesterol, an influence on vascular motility, and calcium antagonistic functions.4 5 6 The exact mechanisms of these effects are still unclear. Knowledge about the role of testosterone in the development of atherosclerosis is even more limited. Experimental and clinical data are conflicting, showing variously that testosterone had either no,7 8 negative9 10 or positive effects11 12 13 on the process of atherosclerosis, or on clinical symptoms14 15 16 in males.
Recently, Adams et al17 found an atherogenic effect of testosterone in female cynomolgus monkeys and hypothesized that testosterone might be responsible for the increased incidence of atherosclerosis in men. The aim of the present study was to investigate whether there are gender specific differences in the effects of testosterone alone, estrogen alone, or testosterone in combination with estrogen on plaque development and the percentage of cells undergoing DNA synthesis in the neointima of the aortic arch in an experimental animal model.
A total of 64 New Zealand White rabbits (32 male and 32 female) were chosen for the present study. The animals were approximately 15 weeks of age. The males had an average body weight (BW) of 2.61±0.27 kg, and the females 2.74±0.18 kg. They were housed singly and exposed to 12-hour light cycles. For the surgical procedures, castration in males or bilateral ovariectomy in females, the animals were anesthetized with ketamine HCl (30 mg/kg BW) and xylazine (5 mg/kg BW) intramuscularly (IM). Postoperatively, they were assigned to 4 study groups of 8 male and 8 female rabbits each. All were fed an atherogenic 0.5% cholesterol diet (Altromin Inc) for a 12-week period. At the end of the experiment, the animals were killed and the aortic arches were prepared for further histological examination. All procedures were reviewed and approved by the Institutional Animal Care Subcommittee.
Estradiol valerate (Progynon Depot 10, Schering AG) and/or testosterone enanthate (Testoviron Depot, Schering AG) were administered IM once a week at dosages of 1 mg and 25 mg/kg BW, respectively. Group 1, the control group, received no hormone treatment. Group 2 received estrogen, group 3 testosterone and group 4 a combination of estrogen and testosterone in the above dosages.
Plasma Hormone Levels
Blood samples for determination of plasma hormone levels were drawn on day 0 and on day 84, ie, one week after the last hormone injection. Plasma levels of 17 β-estradiol and testosterone were determined using commercially available radioimmunoassays (RIA standard kits, Biermann Inc).
Plasma and Lipoprotein Cholesterol
At the beginning of the experiment, immediately prior to castration or ovariectomy, and at the end of the experiment on day 84, blood samples were collected in EDTA tubes and processed immediately. After centrifugation at 3000 rpm at 4°C for 10 minutes, the plasma total cholesterol concentration (TC) was measured by a standard enzymatic method (Boehringer Inc).
The area under the curve (AUC) was calculated by integrating the curve specified by the total plasma cholesterol concentrations from the 4 blood collections. It represents the long-term plasma cholesterol concentration, to which the arteries were exposed over the whole study period.
At the beginning and at the end of the experiment preparative ultracentrifugation of the plasma was performed using a Beckman ultracentrifuge equipped with a Ti 50.3-rotor (Beckman Instruments). The procedure to isolate the lipoproteins was performed as described in the “Manual of Laboratory Operations” of the NIH.18 After ultracentrifugation, cholesterol (UZ-C) and high density lipoprotein cholesterol (HDL-C) were measured enzymatically. Low density lipoprotein cholesterol (LDL-C) was calculated using the following equation: LDL-C=UZ-C - HDL-C. Very low density lipoprotein cholesterol (VLDL-C) was also calculated, using the equation: VLDL-C=Total plasma cholesterol (TC) - UZ-C.
The number of cells in the plaque undergoing DNA synthesis was determined by application of 5′-bromo-2′-deoxyuridine (BrdU), a thymidine analogue. Briefly, 100 mg BrdU and 75 mg 2′-deoxycytidine/kg BW (both from Sigma Co) were applied as subcutaneous neck depots 18 hours before the animals were killed.19 In addition, the animals received 30 mg BrdU and 25 mg deoxycytidine/kg BW IM 18 and 12 hours before the end of the experiment.
Fixation and Histological Procedures
The animals were killed after the final blood collection by intravenous injection of a barbiturate (T 61, Hoechst Veterinär Co). The ascending aortic arch was dissected, divided into 3 segments and immersion-fixed in a 0.1 mol/L cacodylate-buffered 2% paraformaldehyde solution for at least 24 hours. After paraffin embedding, the samples were prepared for further histological examination. First, the segments of each aortic arch were sectioned serially (slice thickness 4 μm) until the maximum plaque extent could be determined. Then the serial sections of the sample showing the greatest plaque extent were used for all further morphometric and histological examinations.
The border between the neointimal plaque and the media, the lamina elastica interna, was identified by Elastica van Gieson staining. The plaque size was measured morphometrically (expressed in mm2) in three serial cross sections of the aortic arch using a digital image analyzer with a software package (Bilaney Consultants Co). The mean cross-sectional area of these 3 serial cross sections was calculated.
For all immunohistochemical staining procedures, sections of the aortic vessel incubated with or without primary antibody were used as positive and negative controls, respectively. Alpha-actin is known to be a highly specific marker for smooth muscle cells (SMCs). For the identification of neointimal cells as SMCs, immunohistochemical staining in cross sections was performed as described earlier19 using a monoclonal antibody specific for alpha-actin (Renner Inc).
The percentage of BrdU labeled cells in the plaque was determined as follows: During a period of 18 hours before the end of the experiment BrdU was incorporated into the DNA of proliferating cells. These cells were identified using a monoclonal antibody specific for BrdU (Bio Cell Consulting). Immunohistochemical detection of BrdU labeled cells was performed in cross sections using biotin-avidin and hemalaune staining. For calculating the percentage of BrdU labeled cells, four diametrically arranged segments of one sample were examined using an ocular grid. In these segments the total intimal cell number as well as the number of BrdU-labeled cell nuclei were counted. To ensure that BrdU was incorporated into replicating cells, the labeling index of the small intestinal mucosa was also determined, where BrdU was incorporated into about 30% of the cell nuclei.
Statistical evaluations were carried out with the help of the statistical software JMP 3.1.6 (SAS).
Original data are expressed as mean± SD. For statistical analysis, an arcsin transformation of BrdU data and a logarithmic transformation for data of the intimal area, the plasma lipoprotein, the total cholesterol values were performed to normalize distributions and to standardize variances. Two-way analysis of variance (ANOVA) was performed to evaluate the influences of the different hormone regimens on body weight increase, lipid parameters (AUC, VLDL-C, LDL-C, HDL-C), plaque size and BrdU incorporation, including the factors gender, study group, and gender-study group interaction. In cases where no interaction effect was found (P>.05), the respective interaction factor was omitted. Subsequently, modified t tests were performed to identify the sources of the differences, and to compare the different hormone treatment groups to the control group. When an interaction effect could be identified, comparisons to the control group were performed for each gender separately.
In addition, gender differences in plaque size and in the incorporation of BrdU between the hormone treated groups and the respective control groups were compared.
Analysis of covariance (ANCOVA) was used to examine possible confounding effects from the different lipid parameters or from body weight on the extent of neointimal plaque development and the percentage of BrdU labeled cells.
Differences were considered to be statistically significant when p was less than 0.05. P-values were adjusted by the method of Bonferroni and Holm for each ANOVA and for each set of subsequent t tests as well.
One male rabbit of the group treated with estrogen died of unknown causes 66 days after the beginning of the experiment and the data were therefore excluded from evaluation.
Body Weight and Plasma Lipid Parameters
Analysis of body weight increase during the 12-week period showed a significant difference among the 4 study groups (P<.0001). No gender or interaction effect was seen. Only rabbits treated with estrogen showed a significantly greater increase in body weight compared with the control group (Table 1⇓).
TC levels in the groups were similar at the beginning of the study. At the end of the experiment TC levels were significantly different among the study groups (P=.002). No gender or interaction effect was seen. Only animals treated with testosterone showed a significantly lower level of TC compared with the control group (P=.0004).
When long-term plasma cholesterol concentrations summarized for the 12-week study period with the AUC were analyzed, a significant difference between the study groups was found (P<.0001). No gender or interaction effect was observed. AUC values were significantly lower in the estrogen treated group (group 2, P=.004), as well as in the group with combined treatment (group 4, P<.0001), compared with the control group.
At the end of the experiment, VLDL-C (P=.004), LDL-C (P=.0001) and HDL-C (P<.0001) levels differed significantly among the study groups. In LDL-C (P=.012) and HDL-C (P=.016) significant differences due to gender were seen. In HDL-C a significant interaction effect was also found (P=.002), hence the genders were analyzed separately.
In all animals, VLDL-C was lower in the hormone treated groups compared with the control group (group 2, P=.008; group 3, P=.001; group 4, P=.002). LDL-C levels were higher in the estrogen treated animals compared with controls (P=.001). In females, HDL-C levels were lower in group 3 (P<.0001) and group 4 (P=.0003) compared with the control group (Table 1⇑).
Plasma Hormone Levels
Hormone treatment increased the respective estrogen and/or testosterone plasma levels considerably and comparably in the corresponding male and female groups (Table 1⇑).
The structure and composition of the plaques were generally similar in all study groups, based on histomorphological criteria. Serial cross sections of all groups stained with alpha-actin specific for SMCs showed homogenous actin filaments. Most were localized in the luminal part of the intima. The plaques proliferated concentrically with foam cells predominant centrally and SMCs luminally (Fig 1⇓). The cells which showed BrdU incorporation were identified as SMCs by comparing their locations shown by the staining procedures in the serial cross sections.
There was a significant group (P<.0001), gender (P=.03), and interaction (P=.0003) effect on the development of neointimal plaques (P=.0003). Therefore, males and females were analyzed separately. In males, plaque size was less in both the testosterone group (1.2±1.02 mm2, P=.002) and the combination group (0.48±0.41 mm2, P<.0001) than in the control group (3.55±1.30 mm2). In females, plaque size was less in the estrogen group (1.29±1.11 mm2, P=.001) and the combination group (1.11±0.99 mm2, P=.0001) compared with the control group (4.28±1.76 mm2).
In addition, gender differences in plaque size between the hormone treated groups and the respective control groups were compared. The only significant difference between the corresponding groups was found in the testosterone treated males (3.55±1.30 mm2 versus 1.2±1.02 mm2) and females (4.28±1.76 mm2 versus 6.1±1.38 mm2), compared with their respective control groups (P=.004), (Fig 2⇓).
ANCOVA found no confounding effect from body weight or the different lipid parameters on plaque development, except for the AUC, where a confounding effect was seen (P=.01). The significant differences in plaque development between the study groups seen after adjustment for the AUC, were the same as those shown to be significant by the ANOVA applied without adjustment for AUC (Table 2⇓).
BrdU Labeled Cells in the Plaque
Analysis of BrdU incorporation showed a significant group (P=.002) and interaction (P=.0002) effect, but no gender effect (P=.70). Due to the interaction effect, males and females were analyzed separately. In males, there were no significant differences between the different hormone treated groups and the control group. In females, the percentage of BrdU labeled cells in the estrogen treated group was significantly reduced compared with the control group (2.97± 1.81% versus 8.70±3.90%), whereas no significant differences between the control group and the other hormone groups were observed.
In addition, gender differences in BrdU incorporation between the hormone treated groups and the respective control groups were compared. The only significant difference between the corresponding groups was found in the estrogen-treated animals. The change in BrdU incorporation in females (from 8.70±3.90% to 2.97±1.81%) was greater than that in males (from 7.00±1.70% to 7.70±3.60%), compared with their respective control groups (P=.003) (Fig 3⇓).
ANCOVA did not reveal any confounding effects from body weight or the different lipid parameters on the percentage of BrdU labeled cells in the plaques.
The P-values from the two-way ANOVA for the factors group, gender, and gender-group interaction, as well as the adjustment for AUC for plaque extent and the percentage of BrdU labeled cells in the neointima, are summarized in Table 2⇑.
In this experimental model a gender specific effect of estrogen and testosterone on the development of atherosclerosis could be demonstrated.
Atheroprotective effects of estrogen have been observed in a variety of animal models including chickens, rabbits and primates.4 7 20 21 22 The cholesterol-fed rabbit has become a widely used model since it was first described by Anitschkow23 in 1914. One advantage of this model is that atherosclerotic lesions can be induced rapidly and reliably within 8 to 12 weeks.20 24 25 Recent studies have demonstrated that the effects of estrogen and progesterone on atherogenesis seen in cholesterol-fed rabbits, were comparable to those seen in monkeys.4 26 Data from epidemiological studies are in agreement with the results found in animal studies.27
The hyperlipidemia produced in this model is characterized by a considerable elevation of VLDL-C in contrast to humans, where a moderate elevation of LDL-C is observed.28 29 VLDL-C is generally considered to be the more atherogenic lipoprotein in cholesterol-fed rabbits.30 Although the distribution pattern of atherosclerotic plaques in rabbits is different than that in humans,31 all other aspects of the formation of atherosclerotic lesions, such as the cell types involved, the initiation, and the maturation process, are very similar to the atherosclerosis seen in type IIa human hypercholesterolemia.32
In our experiments significant differences were observed in plaque development between the control group and the hormone treated groups, independent of changes in lipid parameters. These results are consistent with previous studies indicating that there are mechanisms of sex hormones which influence the development of the atherosclerotic plaque independent of lipid changes.4 33 34 Considerable experimental data have increased our knowledge concerning the possible beneficial actions of estrogen in the process of atherogenesis. These include influences in lipid and carbohydrate metabolism, coagulation parameters, antioxidative activity, its ability to restore endothelial dysfunction and induce vascular relaxation, calcium antagonistic potency, and antiproliferative effects on SMCs. Furthermore, it may influence the composition of atheromatous plaques.2 35 36 37 38 About 25% of the atheroprotective effects of estrogen are thought to be due to its influences on lipid metabolism.39
In agreement with the results of other studies, our data showed that estrogen reduced the development of atherosclerotic plaques in female rabbits.26 30 34 40 41 In male animals, no protective effect of estrogen, when given alone, was found.34 This supports the clinical data showing that long-term application of estrogen in men suffering from CHD was of no benefit.42 43 However, plaque extent was significantly reduced in male rabbits when estrogen and testosterone were given simultaneously. Probably the gender specific effect of testosterone predominated.
In contrast to the numerous data from studies concerning the role of estrogen, there are relatively little data concerning the effects of testosterone on serum lipid parameters and the process of atherogenesis. The results of studies investigating the effect of testosterone are inconclusive and discrepant.9 10 11 12 13 In several studies testosterone levels in men suffering from CHD or surviving myocardial infarction were analyzed. In the majority of these studies the testosterone levels were found to be lower, but in several studies the levels were normal.44 Testosterone therapy was shown to have a beneficial effect on angina pectoris in male patients.14 15 16
In experimental studies with cholesterol-fed male rabbits, Ludden et al7 as well as Larsen et al8 did not find significant changes in the cholesterol content of the aortic vessel wall after treatment with testosterone compared with untreated control groups. Recently, a relaxing effect of testosterone on isolated artery segments of both male and female rabbits fed a cholesterol free diet was demonstrated. In that experiment, the testosterone effect was both receptor and lipid independent.45 Those results support the hypothesis, that certain mechanisms mediate the effect of testosterone directly at the artery vessel wall.
Adams et al17 found that the administration of testosterone increased the plaque content in the coronary arteries of female ovariectomized cynomolgus monkeys. Different from our study protocol, the animals were fed a cholesterol diet over a period of eight months before they received hormone treatment. The authors proposed that testosterone might also be atherogenic in males and therefore responsible for gender differences in the development of atherosclerosis. However, we found that testosterone was atheroprotective in male rabbits. An atherogenic effect in female animals, shown in the study of Adams et al,17 did not reach statistical significance in our experiment.
The proliferation of SMCs is an important step in the development of atherosclerotic plaques.46 Recently, a reduced proliferation of SMCs in female but not in male rabbits treated with estrogen was described.34 We did not find a significant effect of estrogen alone on either proliferation rate or plaque development in males. Testosterone or combined hormone treatment did not alter the percentage of BrdU labeled cells in the neointima, although a reduction of plaque development could be demonstrated.
In females the percentage of cells in the plaque undergoing DNA synthesis and the plaque development were both significantly reduced by application of estrogen alone, whereas testosterone treatment showed no significant effect on either. In females treated with the combined hormone regimen the rate of incorporation of BrdU into cells was not altered, although the plaque size was reduced significantly. Probably testosterone exerted a negative effect on the proliferation rate, but did not reduce the atheroprotective effect of estrogen on plaque size.
These data conflict with results of Rosenfeld and Ross,47 who demonstrated an inverse relationship between the percentage of total cells labeled by thymidine and the size of atherosclerotic lesions. Our data suggest, that the influence of sex steroids on plaque development might be mediated by different mechanisms, which need further investigation.
In conclusion, a protective effect of estrogen in female and of testosterone in male rabbits was seen. When both hormones were applied simultaneously, the gender specific hormone predominated in its protective action on plaque size. It was able to influence the percentage of BrdU labeled cells in female, but not in male animals.
Due to these gender-specific differences in the action of testosterone and estrogen our data suggest that there are complex interactions of sex hormones involved in the process of atherogenesis. Whether and/or to what degree these mechanisms of sex hormone interaction involve specific sex hormone receptors, hormone independent receptors, or receptor independent actions, needs to be elucidated by further experiments.
Selected Abbreviations and Acronyms
|AUC||=||area under the curve|
|CHD||=||coronary heart disease|
|b.d.l.||=||below detectable limit|
|HDL-C||=||high density lipoprotein cholesterol|
|LDL-C||=||low density lipoprotein cholesterol|
|SMCs||=||smooth muscle cells|
|TC||=||total plasma cholesterol|
|VLDL-C||=||very low density lipoprotein cholesterol|
This work was in part supported by a grant of the Dr Karl Kuhn Foundation, University of Tübingen. We thank P.F. Kahle, MD, for his critical review of the manuscript. The authors gratefully acknowledge the technical assistance of H. Bacher, M. Heilig, M. Holz, G. Kaletta, and C. Lenz.
McGill HC Jr, Stern MP. Sex and atherosclerosis. Atheroscler Rev. 1979;4:157-242.
Adams MR, Kaplan JR, Manuck SB, Koritnik DR, Parks JS, Wolfe MS, Clarkson TB. Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomized monkeys: lack of an effect of added progesterone. Arteriosclerosis. 1990;10:1051-1057.
Losordo DW, Kearney M, Kim EA, Jekanowski J, Isner JM. Variable expression of the estrogen receptor in normal and atherosclerotic coronary arteries of premenopausal women. Circulation. 1994;89:1501-1510.
Collins P, Rosano GMC, Sarrel PM, Ulrich L, Adamopoulos S, Beale CM, McNeill JG, Poole-Wilson PA. 17 -Estradiol attenuates acetylcholine-induced coronary arterial constriction in women but not in men with coronary heart disease. Circulation. 1995;92:24-30.
Ludden JB, Bruger M, Wright IS. Experimental atherosclerosis, IV: effect of testosterone propionate and estradiol dipropionate on experimental atherosclerosis in rabbits. Arch Pathol. 1942;33:58-62.
Goldberg RB, Rabin D, Alexander AN, Doelle GC, Getz GS. Suppression of plasma testosterone leads to an increase in serum total and high density lipoprotein cholesterol and apoproteins A-I and B. J. Clin Endocrinol Metab. 1985;60:203-207.
Khaw KT, Barrett-Connor E. Endogenous sex hormones, high density lipoprotein cholesterol, and other lipoprotein fractions in men. Arterioscler Thromb. 1991;11:489-494.
Heller RF, Wheeler MJ, Micallef J, Miller NE, Lewis B. Relationship of high density lipoprotein cholesterol with total and free testosterone and sex hormone binding globulin. Acta Endocrinol. 1983;104:253-256.
Adams MR, Williams JK, Kaplan JR. Effects of androgens on coronary artery atherosclerosis and atherosclerosis-related impairment of vascular responsiveness. Arterioscler Thromb Vasc Biol. 1995;15:592-570.
Manual of Laboratory Operations, Lipid Research Clinics Program, Vol 1. Lipid and Lipoprotein Analysis. Bethesda, Md: National Heart and Lung Institute, NIH, National Institutes of Health; 1974.
Hanke H, Strohschneider T, Oberhoff M, Betz E, Karsch KR. Time course of smooth muscle cell proliferation in the intima and media of arteries following experimental angioplasty. Circ Res. 1990;67:651-659.
Pick R, Stamler J, Rodbard S, Katz LN. The inhibition of coronary atherosclerosis by estrogens in cholesterol-fed chicks. Circulation. 1952;6:276-280.
Anitschkow N. Über die Atherosklerose der Aorta beim Kaninchen und über deren Entstehungsbedingungen. Beitr. path. Anatomie u. allgem. Pathologe. 1914;59:306-348.
Haarbo J, Leth-Espensen P, Stender S, Christiansen C. Estrogen monotherapy and combined estrogen-progesterone replacement therapy attenuate aortic accumulation of cholesterol in ovariectomized cholesterol-fed rabbits. J Clin Invest. 1991;87:1274-1279.
Wissler RW, Vesselinovitch D. Differences between human and animal atherosclerosis. In: Schettler G, Weizel A, eds. Atherosclerosis III. Berlin/Heidelberg/New York: Springer; 1974:319-325.
Rosenfeld ME, Tsukada T, Gown AM, Ross R. Fatty streak initiation in Watanabe heritable hyperlipemic and comparably hypercholesterolemic fat-fed rabbits. Arteriosclerosis. 1987;7:9-23.
Sulistiyani, Adelman SJ, Chandrasekaran A, Jayo J, St. Clair RW. Effect of 17 -dihydroequilin sulfate, a conjugated equine estrogen, and ethynylestradiol on atherosclerosis in cholesterol-fed rabbits. Arterioscler Thromb Vasc Biol. 1995;15:837-846.
Buja LM, Kita T, Goldstein JL, Watanabe Y, Brown MS. Cellular pathology of progressive atherosclerosis in the WHHL rabbit: an animal model of familial hypercholesterolemia. Arteriosclerosis. 1983;3:87-101.
Wagner JD, Clarkson TB, St. Clair RW, Schwenke DC, Shively CA, Adams MR. Estrogen and progesterone replacement therapy reduces low density lipoprotein accumulation in the coronary arteries of surgically postmenopausal cynomolgus monkeys. J Clin Invest. 1991;88:1995-2002.
Hanke H, Hanke S, Finking G, Muhic-Lohrer A, Mück AO, Schmahl FW, Haasis R, Hombach V. Different effects of estrogen and progesterone on experimental atherosclerosis in female versus male rabbits: quantification of cellular proliferation by bromodeoxyuridine. Circulation. 1996;94:175-181.
Wild RA. Estrogen: Effects on the cardiovascular tree. Obstet Gynecol. 1996;87:27S-35S.
Bush TL, Barrett-Connor E. Noncontraceptive estrogen use and cardiovascular disease. Epidemiol Rev. 1985;7:80-104.
Gerhard M, Ganz P. How do we explain the clinical benefits of estrogen? From bedside to bench. Circulation. 1995;92:5-8.
Clarkson TB. Estrogens, progestins, and coronary heart disease in cynomolgus monkeys. Fertil Steril. 1994;62(suppl 2):147S-151S.
Hough JL, Zilversmit DB. Effect of 17 beta estradiol on aortic cholesterol content and metabolism in cholesterol-fed rabbits. Arteriosclerosis. 1986;6:57-63.
Yue P, Chatterjee K, Beale C, Poole-Wilson PA, Collins P. Testosterone relaxes rabbit coronary arteries and aorta. Circulation. 1995;91:1154-1160.
Rosenfeld ME, Ross R. Macrophage and smooth muscle cell proliferation in atherosclerotic lesions in WHHL and comparably hypercholesterolemic fat-fed rabbits. Arteriosclerosis. 1990;10:680-687.