The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
Volume 328:1069-1075 April 15, 1993 Number 15
NextNext

Association of Hormone-Replacement Therapy with Various Cardiovascular Risk Factors in Postmenopausal Women
Azmi A. Nabulsi, Aaron R. Folsom, Alice White, Wolfgang Patsch, Gerardo Heiss, Kenneth K. Wu, Moyses Szklo, for The Atherosclerosis Risk in Communities Study Investigators

 

This Article
-Abstract

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background Most epidemiologic studies of cardiovascular disease in postmenopausal women suggest that estrogen-replacement therapy has a protective effect. The effects of the use of estrogen combined with progestin are less well studied.

Methods To examine the associations of hormone-replacement therapy with concentrations of plasma lipids and hemostatic factors, fasting serum concentrations of glucose and insulin, and blood pressure, we studied 4958 postmenopausal women participating in a population-based investigation. Using cross-sectional data, we classified the women into four groups according to their use of hormone-replacement therapy: current users of estrogen alone, current users of estrogen with progestin, nonusers who had formerly used these hormones, and nonusers who had never used them.

Results Current users had higher mean levels of high-density lipoprotein cholesterol, its subfractions high-density lipoprotein2 and high-density lipoprotein3, and apolipoprotein A-I than nonusers, and lower mean levels of low-density lipoprotein cholesterol, apolipoprotein B, lipoprotein(a), fibrinogen, antithrombin III, and fasting serum glucose and insulin. However, current users of estrogen alone had higher triglyceride, factor VII, and protein C levels than either nonusers or current users of estrogen with progestin. After making certain assumptions, we estimated that the findings, if causal, would translate into a reduction of 42 percent in the risk of coronary heart disease in users of hormones as compared with nonusers. Women using estrogen with progestin would have an even greater estimated benefit.

Conclusions A randomized trial is needed to eliminate possible selection biases in our observational study that are related to the prescription of replacement hormones. Nevertheless, hormone-replacement therapy appears to be associated with a favorable physiologic profile, which probably mediates its protective effects on cardiovascular disease. The use of estrogen combined with progestin appears to be associated with a better profile than the use of estrogen alone.

In 1985, about 3 million postmenopausal women in the United States were using hormone-replacement therapy to alleviate symptoms associated with menopause1. More than 20 studies have been conducted to assess the effects of this therapy on coronary heart disease2. The majority of those studies, particularly the prospective studies, revealed a protective effect of the use of estrogen-replacement therapy2.

Hormone-replacement therapy may modify the risk of coronary heart disease by several mechanisms; possibilities include alterations in plasma concentrations of lipoproteins,3 hemostatic factors,3,4 glucose, and insulin4 and of blood pressure5. Epidemiologic studies of the physiologic effects of hormone-replacement therapy have largely been limited to the postmenopausal use of estrogen alone. Data on the combined use of estrogen and progestin, currently recommended as hormone-replacement therapy for women who have not had a hysterectomy, are more limited6. In this analysis, we assessed the associations of therapy with exogenous estrogen alone or estrogen combined with progestin with physiologic variables related to cardiovascular risk in a large sample of postmenopausal women.

Methods

This cross-sectional analysis used data from the Atherosclerosis Risk in Communities study7. The study cohort comprised four population samples of men and women 45 to 64 years old from the following regions: Forsyth County, North Carolina; Jackson, Mississippi; selected suburbs of Minneapolis; and Washington County, Maryland. The total cohort included 15,800 subjects, approximately one quarter from each region. All subjects in Jackson and 14 percent of those in Forsyth County were black; almost all the others were white. The response rate -- i.e., the proportion of eligible subjects who completed the base-line examination -- was 46 percent in Jackson and approximately 65 percent in each of the three other communities.

At base-line examinations conducted from 1986 to 1989, the subjects underwent measurement of blood pressure while seated, anthropometric measurements, venipuncture (after fasting for 12 hours), B-mode ultrasonography of the carotid arteries, and a set of interviews that recorded their medical history, level of physical activity, reproductive history, medication use, and other variables. Plasma total triglyceride8 and cholesterol9 were measured by enzymatic methods, with dextran-magnesium precipitation10 for the measurement of high-density lipoprotein (HDL) cholesterol and its HDL3 subfraction; the level of the HDL2 subfraction was calculated by subtraction. Low-density lipoprotein (LDL) cholesterol was calculated with the Friedewald equation11. Apolipoprotein A-I12 and apolipoprotein B13 were measured by radioimmunoassay. Lipoprotein(a) was measured by enzyme-linked immunosorbent assay14. The following reliability coefficients (between-subject variance/total variance) were obtained through repeated analysis testing of a sample of the subjects over several weeks: triglycerides, 0.85; LDL cholesterol, 0.91; HDL cholesterol, 0.94; HDL3 cholesterol, 0.70; HDL2 cholesterol, 0.77; apolipoprotein A-I, 0.60; and lipoprotein(a), 0.95. Fibrinogen, factor VII, and factor VIII were measured by coagulation tests, von Willebrand factor and protein C by enzyme-linked immunosorbent assays, and antithrombin III by thrombin inactivation15. The following reliability coefficients were obtained: fibrinogen, 0.72; factor VII, 0.78; factor VIII, 0.86; von Willebrand factor, 0.68; protein C, 0.56; and antithrombin III, 0.42.

The blood pressure was measured three times with a random-zero sphygmomanometer while subjects were seated after they had rested for five minutes. The value for systolic blood pressure used in the analysis was the average of the second and third measurements. Women were considered to have diabetes if they identified themselves as diabetic, were taking hypoglycemic medications, or had elevated serum glucose concentrations (fasting level >= 140 mg per deciliter [7.8 mmol per liter]; nonfasting level >= 200 mg per deciliter [11.1 mmol per liter]). Fasting serum glucose was measured with a hexokinase-glucose-6-phosphate dehydrogenase method, and fasting serum insulin with a commercial radioimmunoassay (125Insulin Kit, Cambridge Medical Diagnostics, Billerica, Mass.); values for glucose and insulin are reported only for fasting nondiabetic subjects. The body-mass index (the weight in kilograms divided by the square of the height in meters) was computed from the weight measured to the nearest pound (0.45 kg) and the height measured to the nearest centimeter; an index of 27.3 or more indicated overweight16. The circumference of the waist at the level of the umbilicus and the maximal circumference of the hips were measured to the nearest centimeter; the values were rounded down and expressed as the waist-to-hip ratio. An index of physical activity in sports (sports index), ranging from 1 (a low level of activity) to 5 (a high level), was derived with the Baecke questionnaire17. The subjects' smoking and drinking status, educational level, number of years since menopause, and use of antihypertensive medication were assessed through interviews. Prevalent cardiovascular disease was indicated by a history of angina or intermittent claudication elicited with the use of the Rose questionnaire18; a history of myocardial infarction or stroke diagnosed by a physician and reported by the subject; a pathologic Q wave on electrocardiography; or a history of cardiovascular surgery, angioplasty, or carotid endarterectomy reported by the subject.

The use of hormone-replacement therapy was ascertained in the reproductive history elicited by a trained interviewer. To enhance accuracy, the subjects were requested to bring containers of all medications used during the two weeks before the examination.

This report is based on data on postmenopausal women, both black and white, who were free of cardiovascular disease at their base-line visit. They were classified as postmenopausal if they had not menstruated during the two years before the examination. The postmenopausal women were classified according to the type of menopause: surgical (women with bilateral oophorectomy), natural (this classification also included women with hysterectomy and at least one intact ovary who were 55 years old or older), or of uncertain cause (nonmenstruating women whose ovarian status was unknown, such as those with a hysterectomy and at least one intact ovary who were younger than 55 years of age). Of the 8705 women 45 to 64 years old, we excluded 2442 who were premenopausal or perimenopausal, 5 who had primary amenorrhea, 187 who had incomplete data, 26 who were using estrogen creams only, and 52 who were using hormones other than estrogen alone or combined with progestin; we did not exclude a small number of women who were using transdermal preparations of estradiol. We also excluded 17 women who were Asians or American Indians, 153 women who were using hypolipidemic medications, and 865 women with cardiovascular disease. Thus, we included 4958 postmenopausal women in our analysis.

The 4958 women were assigned to one of four groups: current users of estrogen alone, current users of estrogen with progestin, nonusers who had formerly used hormones, and nonusers who had never used them. The type of hormone-replacement therapy prescribed for the former users could not be determined accurately. We had two main hypotheses of interest: postmenopausal women currently receiving hormone-replacement therapy differed significantly from nonusers (both groups together [former and never]) in their levels of blood lipids, lipoproteins, hemostatic factors, glucose, insulin, and blood pressure; and current users of estrogen alone differed significantly from current users of estrogen with progestin in these variables. The study data were insufficient to assess associations with the individual preparations of estrogens and progestins used, dosage, or duration of use.

The characteristics of the women were first expressed in terms of their prevalence among the four groups. Unadjusted mean levels of lipids, lipoproteins, hemostatic factors, glucose, insulin, and blood pressure (both systolic and diastolic) were then computed for the groups. The study hypotheses were tested by generating three statistical contrasts in analysis-of-covariance models using the general linear-modeling programs of the SAS19 computer package -- i.e., a comparison of current hormone users with nonusers, a comparison of current users of estrogen alone with current users of estrogen combined with progestin, and a comparison of current users of estrogen alone with all other subjects (for levels of triglycerides, factor VII, and protein C only). The covariates included the subjects' age, body-mass index, and sports index (all treated as continuous variables) and race, smoking, drinking, diabetes, level of education, use of antihypertensive medications, and study region, modeled with the use of dummy variables.

Results

Approximately 63 percent of the women studied had never used oral replacement hormones, and 16 percent had formerly used them (Table 1). Only 21 percent of the women used them currently; of these, about 83 percent were using estrogen alone (primarily conjugated estrogen [Premarin]) and 17 percent were using estrogen with progestin (primarily conjugated estrogen [Premarin] with medroxyprogesterone acetate [Provera]). White women were more likely to be using estrogen with progestin than were black women.

View this table:
[in this window]
[in a new window]
  		Table 1. The Four Study Groups According to Use of Replacement Hormones and Race.

 
As shown in Table 2, the characteristics and lifestyles of the four study groups varied. Race-specific analyses similar to the analysis summarized in Table 2 did not reveal substantial racial differences in the patterns of these features.

View this table:
[in this window]
[in a new window]
  		Table 2. Distribution of the Subjects' Characteristics According to the Use of Replacement Hormones.

 
Unadjusted mean values for physiologic variables are shown in Table 3, but since the patterns of these values were similar to those of adjusted values, we have confined our discussion to the latter (Table 4). Current users of estrogen alone had significantly higher mean levels of triglycerides than the other three groups combined (P<0.001); the mean in this group was about 10 mg per deciliter (0.11 mmol per liter) higher than the mean in the group currently using estrogen with progestin and about 18 mg per deciliter (0.20 mmol per liter) and 21 mg per deciliter (0.23 mmol per liter) higher than the means in the group that formerly used hormones and the group that never used them, respectively. As compared with nonusers, current users had significantly higher levels of HDL, HDL3, and HDL2 cholesterol and apolipoprotein A-I (P<0.001), with differences of approximately 9 mg per deciliter (0.23 mmol per liter), 5 mg per deciliter (0.11 mmol per liter), 5 mg per deciliter (0.13 mmol per liter), and 18 mg per deciliter (180 mg per liter), respectively; when these variables were compared in the two groups of current users (users of estrogen alone and users of estrogen with progestin), the values in both groups were similar. As compared with nonusers, current users had significantly lower levels of LDL cholesterol, apolipoprotein B, and lipoprotein(a) (P<0.05), with differences of about 16 mg per deciliter (0.40 mmol per liter), 4 mg per deciliter (38 mg per liter), and 15 µg per milliliter, respectively; when these variables were compared in the two groups of current users, the values in both groups were similar. The ratio of LDL cholesterol to apolipoprotein B was lower in current users than in nonusers (1.47 vs. 1.80, P>0.05).

View this table:
[in this window]
[in a new window]
  		Table 3. Unadjusted Values (Means ±SD) for Physiologic Variables, According to Use of Replacement Hormones.

 
View this table:
[in this window]
[in a new window]
  		Table 4. Adjusted Mean Values for Physiologic Variables, According to Use of Replacement Hormones.

 
As also shown in Table 4, current hormone users had significantly lower adjusted mean levels of fibrinogen than nonusers, with a difference of about 0.16 g per liter (P<0.001); the two groups of current users had similar mean fibrinogen levels. Current users had lower adjusted mean levels of antithrombin III than nonusers, with a difference of about 4 percent (P = 0.002); the two groups of current users had similar levels. Current users of estrogen alone had significantly higher mean levels of factor VII and protein C than the three other groups combined (P<0.001); the differences between this group and the others were about 11 percent for factor VII and 0.19 µg per milliliter for protein C. There were no significant differences among the groups in the levels of factor VIII or von Willebrand factor.

Current users had significantly lower adjusted mean fasting concentrations of serum glucose and insulin than nonusers (Table 4), with differences of about 2 mg per deciliter (0.13 mmol per liter) and 1.3 micro U per milliliter (9.7 pmol per liter), respectively (P<0.001); the two groups of current users had similar concentrations. There were no significant differences between current hormone users and nonusers or between the two groups of current users in the adjusted mean systolic or diastolic blood pressure.

The analyses summarized in Table 3 and Table 4 were repeated after women with reported cardiovascular disease were included; the findings were similar to those in the subgroup free of cardiovascular disease.

Discussion

Postmenopausal women receiving hormone-replacement therapy typically have half the risk of cardiovascular disease of nonusers2,20. The aim of our analysis was to provide additional information about physiologic variables that may be responsible for this benefit. The strengths of the Atherosclerosis Risk in Communities study are its large, population-based, biracial sample of women and its standardized measurements. Some limitations of its cross-sectional data are an inability to study associations between variables and hormones according to the type, dose, duration of use, and mode of administration, especially in former users, and possible selection bias related to nonresponse to the survey. An analysis of nonrespondents (unpublished data) suggested that they were more likely to smoke and have a lower socioeconomic status than were persons who did respond, but it is not apparent whether this finding might have influenced associations between hormonereplacement therapy and physiologic variables.

The most important limitation of this study, however, is that an observational study may be biased by unknown selection factors influencing the prescription and use of hormone-replacement therapy. The associations described may reflect differences between hormone users and nonusers, rather than the effects of the hormones themselves. Although we controlled for differences between hormone users and nonusers (Table 2), the control may not have been fully effective. To verify the statistical models, we also analyzed data on a subgroup of women who were white, were more than 55 years old, had continued their education beyond high school, did not smoke, were not obese, and did not have diabetes; the associations were virtually the same. Nevertheless, a randomized long-term trial of various combinations of hormones would allow more definitive conclusions.

Lipids and Lipoproteins

Users of estrogen alone had higher levels of HDL and HDL2 cholesterol and apolipoprotein A-I than nonusers, as previously reported21,22. Users of estrogen alone also had higher levels of HDL3 cholesterol; previous reports about this subfraction have been inconsistent22,23. It is believed that estrogen suppresses hepatic lipase activity, elevating levels of HDL2 and HDL cholesterol21,23. Users of estrogen alone had lower levels of LDL cholesterol, as previously reported21,22. Estrogen appears to lower the level of LDL cholesterol by increasing its rate of clearance from plasma,21,22 but the levels of apolipoprotein B, the principal apoprotein in LDL, were also lower.

Previous clinical trials in postmenopausal women have generally reported that the addition of a progestin opposes many beneficial effects of estrogen by lowering the level of HDL cholesterol, mainly the level of HDL2,20,21,24 without changing that of HDL323. Progestins appear to increase hepatic lipase activity, thus increasing the catabolism of HDL2 and lowering the levels of both HDL2 and HDL cholesterol24. Progestins do not appear to change the levels of LDL cholesterol significantly21,24. Our analysis revealed that users of estrogen with progestin and users of estrogen alone had similar levels of HDL, HDL2, and HDL3 cholesterol, apolipoprotein A-I, LDL cholesterol, apolipoprotein B, and lipoprotein(a). This similarity may be due to the fact that the majority of the users of estrogen with progestin were taking medroxyprogesterone acetate, a progestin with low levels of androgenic activity, which has slight effects23,25 or no effect26 on lipoprotein levels when used alone25 or in combination with estrogen23. It also has less influence on hepatic lipase activity than most other progestins, as suggested by a clinical trial in premenopausal women 40 to 50 years old27. Another survey28 of postmenopausal women recently found no difference in the levels of HDL and LDL cholesterol in users of conjugated estrogen alone and those in users of conjugated estrogen with medroxyprogesterone acetate.

The effects of estrogen combined with progestin on lipoprotein(a), a recently described cardiovascular risk factor, have not been widely reported. Hormone use was associated with a reduced level of lipoprotein(a), a finding consistent with a previous trial of conjugated estrogen combined with medroxyprogesterone acetate29. The possibility that hormone-replacement therapy may lower the level of lipoprotein(a) is interesting, because lipoprotein(a) levels appear to be genetically determined for the most part and resistant to most environmental influences or lifestyle factors29,30.

In addition to this study, other cross-sectional studies and clinical trials have reported that the use of estrogen alone increases plasma triglyceride levels in postmenopausal women21,22,28. Estrogen appears to increase the hepatic synthesis of very-low-density lipoprotein (VLDL) triglycerides, particularly large VLDL21,22. Large VLDL is directly catabolized by the liver rather than delipidated to small VLDL and LDL, and therefore its elevation may be less harmful than that of other triglycerides22. In contrast, we found no elevation of triglyceride levels in the users of estrogen with progestin. This confirms the findings of some previous studies,28,31 but not all,26 that the combined use of conjugated estrogen and medroxyprogesterone acetate does not adversely affect triglyceride levels. Progestin apparently lowers triglyceride levels by increasing the clearance or decreasing the synthesis (or by causing both actions) of VLDL and triglycerides24,32. We conclude from our data that lipid levels appear not to be adversely affected by the addition of a low dose of progestin to estrogen-replacement therapy. In fact, the lipid profile may be improved.

Hemostatic Factors

Plasma concentrations of fibrinogen and factor VII, two coagulation factors, have been directly associated with the incidence of cardiovascular disease33. Deficiencies of antithrombin III and protein C predispose persons with these features to venous thrombosis,34,35 but high levels may reflect a response to thrombogenesis36. In this study, current users of hormones had lower levels of fibrinogen and antithrombin III than nonusers. Most investigators have found no difference in levels of fibrinogen37,38,39 or antithrombin III37,38 between users of estrogen alone and users of estrogen with progestin. The association between factor VII levels and the use of estrogen alone in this study has also been observed in clinical trials40. However, in our study, factor VII levels were not influenced by the use of estrogen combined with progestin, as previously reported37. Most likely, the increase in triglyceride levels during the use of estrogen alone was responsible for the parallel change in factor VII levels41. It has been suggested that triglycerides activate phospholipase C-sensitive factor VII complexes41. In the light of recent evidence linking elevated triglyceride levels to atherogenesis42 and elevated factor VII levels to cardiovascular disease,33,43 our findings regarding the differential associations of the use of estrogen alone and the use of estrogen with progestin may be important clinically.

The association between the use of hormones and protein C levels mirrored that between hormones and factor VII levels: the levels in users of estrogen alone were higher than the levels in the other groups. The use of oral contraceptives in fertile women also appears to increase factor VII and protein C levels36,44. Levels of factor VIII and von Willebrand factor were not altered by hormone-replacement therapy in our subjects; levels of these factors have been reported to be increased in postmenopausal women taking 10 µg of ethinyl estradiol, but not in those taking 2 mg of estradiol valerate45.

Glucose and Insulin

In agreement with our analysis, a recent cross-sectional study46 reported that fasting serum concentrations of glucose and insulin were lower in users of conjugated estrogen alone and users of conjugated estrogen with medroxyprogesterone than in nonusers. Estrogen and progestin may influence glucose and insulin levels by altering body composition and pancreatic {beta}-cell function46.

Blood Pressure

We observed no difference in blood pressure (both systolic and diastolic) between users of either estrogen alone or estrogen with progestin and nonusers. Most studies have found either no change or even a reduction in blood pressure with estrogen use5. One study (the Rancho Bernardo Study)28 reported that systolic and diastolic blood pressure was lowered by the use of estrogen combined with progestin but not by the use of estrogen alone.

Hormone-Replacement Therapy and the Risk of Cardiovascular Disease

The epidemiologic evidence of a protective effect of estrogen-replacement therapy against cardiovascular disease is compelling20. Our study confirms previous explanatory studies of the effects of estrogen and extends studies of estrogen with progestin, suggesting physiologic effects that may mediate the protection. We did not attempt to address the overall risks and benefits of hormone replacement in postmenopausal women since this topic has been discussed by others47. Likewise, because our study was not a randomized trial, we cannot rule out selection bias related to hormone replacement or certain other noncausal explanations of the findings. Nevertheless, it is of interest to estimate the potential effect of these physiologic findings, if causal, on the risk of coronary heart disease. In a clinical trial in men, a reduction of 1 mg per deciliter (0.026 mmol per liter) in the LDL cholesterol level decreased the risk of coronary heart disease by 1 percent48. In observational studies of men, an increase of 1 mg per deciliter (0.026 mmol per liter) in the HDL cholesterol level decreased the risk by 2 percent49. In observational studies that included women, a reduction of 0.01 g per liter in the fibrinogen level decreased the risk by about 0.5 percent33,50. If these associations are independent, additive, and causal, our observation in hormone users of a reduction of 16 mg per deciliter (0.40 mmol per liter) in the LDL cholesterol level, an increase of 9 mg per deciliter (0.23 mmol per liter) in the HDL cholesterol level, and a reduction of 0.16 g per liter in the fibrinogen level would represent a sizable reduction of 42 percent in the risk of coronary heart disease in users as compared with nonusers. The reduction in fasting levels of glucose and insulin would further reduce the risk in hormone users. Furthermore, the reduction of 9 percent in the factor VII levels in users of estrogen with progestin would reduce their risk by about 18 percent, as compared with users of estrogen alone, assuming that a 1 percent decrease in the factor VII level would reduce the risk of death due to coronary heart disease by 2 percent33. The reduction of 10 mg per deciliter (0.11 mmol per liter) in the triglyceride level would probably reduce the risk associated with the use of estrogen with progestin even more.

These results provide further arguments for a long-term, randomized, controlled trial to produce conclusive evidence of the risks and benefits of the use of various hormone preparations in postmenopausal women.

Supported under contracts (N01-HC-55015, N01-HC-55016, N01-HC-55018, N01-HC-55019, N01-HC-55020, N01-HC-55021, and N01-HC-55022) with the National Heart, Lung, and Blood Institute.

We are indebted to Dr. John Eckfeldt, Dr. Robert Rock, Dr. Spencer Brown, Dr. Woody Chambless, Dr. Fredric Romm, Dr. Paul McGovern, Dorothy Buckingham, Leone Reed, Audrey Papp, and Andrea Finch (Laboratory Committee, Atherosclerosis Risk in Communities study); to Victoria Nabulsi, Laura Kemmis, Lynn Kitzerow, Sally Ingersoll, Larry Crum, Ding Ye Zhao, Ann Howard, and Gretchen Marcucci for assistance in the preparation of the manuscript; and to field-center technicians Elsie Bacon, Karen Barr, Carol Christman, Lisa Field, Amy Haire, Bryna Lester, Stella Loehr, Sharada Lyer, Barbara Mariotti, Catherine McCormick, Gail Murton, Joan Nelling, Virginia Overman, Delilah Posey, Cathy Rachui, Sue Ware, Shirley Willis, and Virginia Wyum for sample preparation.


Source Information

From the Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis (A.A.N., A.R.F.); the ESP Division, Burroughs-Wellcome Company, Research Triangle Park, N.{beta}(A.W.); the Atherosclerosis Clinical Laboratory, Methodist Hospital, Houston (W.P.); the School of Public Health, Department of Epidemiology, University of North Carolina, Chapel Hill (G.H.); the Division of Hematology-Oncology, University of Texas Medical School, Houston (K.K.W.); and the Johns Hopkins School of Hygiene and Public Health, Baltimore (M.S.).

Address reprint requests to Dr. Folsom at the Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S. Second St., Suite 300, Minneapolis, MN 55454.

References

  1. Kennedy DL, Baum C, Forbes MB. Noncontraceptive estrogens and progestins: use patterns over time. Obstet Gynecol 1985;65:441-6. 
  2. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47-63. [CrossRef][Medline]
  3. Notelovitz M. Estrogen replacement therapy: indications, contraindications, and agent selection. Am J Obstet Gynecol 1989;161:1832-1841. [Medline]
  4. Barrett-Connor E. Putative complications of estrogen replacement therapy: hypertension, diabetes, thrombophlebitis, and gallstones. In: Korenman SG, ed. The menopause: biological and clinical consequences of ovarian failure; evolution and management. Norwell, Mass.: Serono Symposia, 1990:199-209.
  5. Lobo RA. Estrogen replacement therapy and hypertension. Postgrad Med 1987;14:48-54.
  6. Bush TL, Barrett-Connor E, Cowan LD, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-Up Study. Circulation 1987;75:1102-1109. [Free Full Text]
  7. The ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. Am J Epidemiol 1989;129:687-702. [Free Full Text]
  8. Nagele U, Hagele EO, Sauer G, et al. Reagent for the enzymatic determination of serum total triglycerides with improved lipolytic efficiency. J Clin Chem Clin Biochem 1984;22:165-174. [Medline]
  9. Siedel J, Hagele EO, Ziegenhorn J, Wahlefeld AW. Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin Chem 1983;29:1075-1080. [Free Full Text]
  10. Warnick GR, Benderson JM, Albers JJ. Quantitation of high-density-lipoprotein subclasses after separation by dextran sulfate and MG2+ precipitation. Clin Chem 1982;28:1574-1574.abstract 
  11. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502. [Abstract]
  12. Brown SA, Rhodes CE, Dunn K, Gotto AM Jr, Patsch W. Effect of blood collection and processing on radioimmunoassay results for apolipoprotein A-I in plasma. Clin Chem 1988;34:920-924. [Free Full Text]
  13. Brown SA, Epps DF, Dunn JK, et al. Effect of blood collection and processing on radioimmunoassay results for apolipoprotein B in plasma. Clin Chem 1990;36:1662-1666. [Free Full Text]
  14. ARIC: Atherosclerosis Risk in Communities study: manual 8: lipid and lipoprotein determinations. Chapel Hill, N.C.: National Heart, Lung, and Blood Institute, 1987.
  15. Papp AC, Hatzakis H, Bracey A, Wu KK. ARIC hemostasis study. I. Development of a blood collection and processing system suitable for multicenter hemostatic studies. Thromb Haemost 1989;61:15-19. [Medline]
  16. National Center for Health Statistics, Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight: United States, 1976-80. Vital and health statistics. Series 11. No. 238. Washington, D.C.: Government Printing Office, 1987. (DHHS publication no. (PHS) 87-1688.)
  17. Baecke JAH, Burema J, Frijters JER. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982;36:936-942. [Free Full Text]
  18. Rose GA, Blackburn H, Gillum RF, Prineas RJ, eds. Cardiovascular survey methods. 2nd ed. World Health Organization monograph series no. 56. Geneva: World Health Organization, 1982.
  19. SAS user's guide: statistics, version 5 rev. ed. Cary, N.C.: SAS Institute, 1988.
  20. Lobo RA. Estrogen and cardiovascular disease. Ann N Y Acad Sci 1990;592:286-94, 334. [Medline]
  21. Sacks FM, Walsh BW. The effects of reproductive hormones on serum lipoproteins: unresolved issues in biology and clinical practice. Ann N Y Acad Sci 1990;592:272-85, 334. [Medline]
  22. Walsh BW, Schiff I, Rosner B, Greenberg L, Ravnikar V, Sacks FM. Effects of postmenopausal estrogen replacement on the concentrations and metabolism of plasma lipoproteins. N Engl J Med 1991;325:1196-1204. [Abstract]
  23. Tikkanen MJ, Nikkila EA, Kuusi T, Sipinen SU. High density lipoprotein-2 and hepatic lipase: reciprocal changes produced by estrogen and norgestrel. J Clin Endocrinol Metab 1982;54:1113-1117. [Free Full Text]
  24. Tikkanen MJ, Kuusi T, Nikkila EA, Sipinen S. Post-menopausal hormone replacement therapy: effects of progestogens on serum lipids and lipoproteins: a review. Maturitas 1986;8:7-17. [CrossRef][Medline]
  25. Silfverstolpe G, Gustafson A, Samsioe G, Svanborg A. Lipid metabolic studies in oophorectomised women: effects on serum lipids and lipoproteins of three synthetic progestogens. Maturitas 1982;4:103-111. [CrossRef][Medline]
  26. Weinstein L. Efficacy of a continuous estrogen-progestin regimen in the menopausal patient. Obstet Gynecol 1987;69:929-932. [Medline]
  27. Tikkanen MJ, Nikkila EA, Kuusi T, Sipinen S. Different effects of two progestins on plasma high density lipoprotein (HDL2) and postheparin plasma hepatic lipase activity. Atherosclerosis 1981;40:365-369. [CrossRef][Medline]
  28. Barrett-Connor EL, Wingard DL, Criqui MH. Postmenopausal estrogen use and heart disease risk factors in the 1980s: Rancho Bernardo, Calif., revisited. JAMA 1989;261:2095-2100. [Free Full Text]
  29. Soma M, Fumagalli R, Paoletti R, et al. Plasma Lp(a) concentration after oestrogen and progestagen in postmenopausal women. Lancet 1991;337:612-612. [Medline]
  30. Hegele RA. Lipoprotein (a): an emerging risk factor for atherosclerosis. Can J Cardiol 1989;5:263-265. [Medline]
  31. Farish E, Fletcher CD, Hart DM, Teo HT, Alazzawi F, Howie C. The effects of conjugated equine oestrogens with and without a cyclical progestogen on lipoproteins, and HDL subfractions in postmenopausal women. Acta Endocrinol (Copenh) 1986;113:123-127. [Free Full Text]
  32. Nikkila EA, Tikkanen MJ, Kuusi T. Effects of progestins on plasma lipoproteins and heparin-releasable lipases. In: Bardin CW, Milgrom E, Mauvais-Jarvis P, eds. Progesterone and progestins. New York: Raven Press, 1983:411-20.
  33. Meade TW, Mellows S, Brozovic M, et al. Haemostatic function and ischaemic heart disease: principal results of the Northwick Park Heart Study. Lancet 1986;2:533-537. [Medline]
  34. Esmon CT. Protein-C: biochemistry, physiology, and clinical implications. Blood 1983;62:1155-1158. [Free Full Text]
  35. Thaler E, Lechner K. Antithrombin III deficiency and thromboembolism. Clin Haematol 1981;10:369-390. [Medline]
  36. Meade TW, Stirling Y, Wilkes H, Mannucci PM. Effects of oral contraceptives and obesity on protein C antigen. Thromb Haemost 1985;53:198-199. [Medline]
  37. Gordon EM, Williams SR, Frenchek B, Mazur CA, Speroff L. Dose-dependent effects of postmenopausal estrogen and progestin on antithrombin III and factor XII. J Lab Clin Med 1988;111:52-56. [Medline]
  38. Notelovitz M, Kitchens CS, Ware MD. Coagulation and fibrinolysis in estrogen-treated surgically menopausal women. Obstet Gynecol 1984;63:621-625. [Medline]
  39. Notelovitz M, Kitchens C, Ware M, Hirschberg K, Coone L. Combination estrogen and progestogen replacement therapy does not adversely affect coagulation. Obstet Gynecol 1983;62:596-600. [Medline]
  40. Meade TW. Clotting factors and ischaemic heart disease: the epidemiologic evidence. In: Meade TW, ed. Anticoagulants and myocardial infarction: a reappraisal. New York: John Wiley, 1984:91-111.
  41. Skartlien AH, Lyberg-Beckmann S, Holme I, Hjermann I, Prydz H. Effect of alteration in triglyceride levels on factor VII-phospholipid complexes in plasma. Arteriosclerosis 1989;9:798-801. [Free Full Text]
  42. Lapidus L, Bengtsson C, Lindquist O, Sigurdsson JA, Rybo E. Triglycerides -- main lipid risk factor for cardiovascular disease in women? Acta Med Scand 1985;217:481-489. [Medline]
  43. Balleisen L, Schulte H, Assmann G, Epping P-H, van de Loo J. Coagulation factors and the progress of coronary heart disease. Lancet 1987;2:461-461. 
  44. Malm J, Laurell M, Dahlback B. Changes in the plasma levels of vitamin K-dependent proteins C and S and of C4b-binding protein during pregnancy and oral contraception. Br J Haematol 1988;68:437-443. [Medline]
  45. Lindberg UB, Crona N, Stigendal L, Teger-Nilsson AC, Silfverstolpe G. A comparison between effects of estradiol valerate and low dose ethinyl estradiol on haemostasis parameters. Thromb Haemost 1989;61:65-69. [Medline]
  46. Barrett-Connor E, Laakso M. Ischemic heart disease risk in postmenopausal women: effects of estrogen use on glucose and insulin levels. Arteriosclerosis 1990;10:531-534. [Free Full Text]
  47. Barrett-Connor EL. The risks and benefits of long-term estrogen replacement therapy. Public Health Rep 1989;104:Suppl:62-65. [Medline]
  48. Lipid Research Clinics Program. The Lipid Research Clinics Coronary Primary Prevention Trial results. I. Reduction in incidence of coronary heart disease. JAMA 1984;251:351-364. [Free Full Text]
  49. Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease: four prospective American studies. Circulation 1989;79:8-15. [Free Full Text]
  50. Kannel WB, Wolf PA, Castelli WP, D'Agostino RB. Fibrinogen and risk of cardiovascular disease: the Framingham Study. JAMA 1987;258:1183-1186. [Free Full Text]

 

This Article
-Abstract

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Hormone Replacement and Cardiovascular Risk Factors
Zumoff B., Crook D., Stevenson J. C., Daubresse J.-C., Bluming A. Z., Nabulsi A. A., Folsom A. R., White A., Martin K. A., Freeman M. W.
Extract | Full Text  
N Engl J Med 1993; 329:1041-1043, Sep 30, 1993. Correspondence

This article has been cited by other articles:


HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved.