| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From Social and Scientific Systems, Inc., and the National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Maryland.
Address reprint requests to: Maureen I. Harris, PhD National Diabetes Data Group National Institute of Diabetes and Digestive and Kidney Diseases National Institutes of Health 6707 Democracy Boulevard, Room 695, MSC-5460 Bethesda, MD 20892-5460 E-mail: mh63q{at}nih.gov
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Cross-sectional data from the third National Health and Nutrition Examination Survey (1988–1994) included levels of hemoglobin A1c in women with diagnosed diabetes and levels of hemoglobin A1c, fasting and 2-hour glucose, and fasting insulin and C-peptide in women without diagnosed diabetes. We compared mean values for these measures among never, current, and past users of HRT with adjustment for confounders. Types of hormones were not studied.
Results: Hormone replacement therapy was used by 8.6% of diabetic women and 16.7% of women without diagnosed diabetes; 19.3% and 18.5%, respectively, had used HRT in the past. Current use approximately doubled among diabetic women between 1988–1991 and 1991–1994. Current users had lower hemoglobin A1c and fasting plasma glucose levels but higher 2-hour glucose levels compared with never and past users. After adjustment for confounding factors, hemoglobin A1c levels were 0.1% lower, fasting glucose levels were 3 mg/dL lower, and 2-hour glucose levels were 15 mg/dL higher in current users. Fasting serum insulin and C-peptide levels were not associated with HRT use. Duration of HRT use among current users and time since cessation among former users were not associated with measures of glucose metabolism.
Conclusion: The prevalence of HRT in the United States among diabetic women is approximately half that of women without diabetes diagnoses, although it appears to be increasing. Postmenopausal hormones appear to have no adverse effect on basal glucose metabolism but are associated with slightly elevated postchallenge glucose levels.
Few studies have focused on the effect of hormone replacement therapy (HRT) on glucose metabolism, despite the high prevalence of diabetes in postmenopausal women. A beneficial effect of HRT on glucose metabolism might explain the association between HRT and cardiovascular disease.1 A lower risk of type 2 diabetes was observed among current users of HRT compared with never-users in two cohort studies.2,3 Results of three community-based studies suggested that HRT is associated with lower fasting glucose and fasting insulin levels.1,4,5 A study of combined estrogen and progesterone showed decreased fasting glucose levels and insulin levels compared with placebo.6 Other studies had discrepant results. Some showed little or no impairment, and others showed deterioration in glucose tolerance.7–13 Those studies included selected groups that were not necessarily representative of US women currently taking HRT.
To study whether HRT was associated with elevated levels of glycoslyated hemoglobin, fasting and 2-hour glucose, or fasting insulin or C-peptide, we analyzed data from a large, population-based survey of a representative sample of US women at least 45 years of age.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Of 5167 women at least 45 years of age who participated in the interview, 4518 had examinations. Women who were pregnant (n = 1) and those who had menstruated in the past 12 months (n = 521) were excluded. Women 55 years of age or older were considered postmenopausal (n = 3403). Among women 45–54 years of age (n = 593), we excluded those with unknown menstrual status (n = 23), unknown hysterectomy status (n = 5), and those who had hysterectomies with conservation of at least one ovary (n = 220). One hundred forty women could not be classified as never, current, or past users of HRT. We identified women with histories of diabetes or diabetes therapy and excluded women with gestational diabetes and 12 women with missing data. Thus, we included 569 diabetic women and 3027 nondiabetic women.
Hemoglobin A1c was measured in all women by using high-performance liquid chromatographic assay, as was done in the Diabetes Control and Complications Trial.14 The upper limit of normal for hemoglobin A1c was 6.1%, defined as the mean plus two SDs (5.27% + 0.86%) for women with fasting plasma glucose levels less than 110 mg/dL and 2-hour postchallenge glucose levels less than 140 mg/dL. Only hemoglobin A1c was measured in diabetic women. In nondiabetic women, fasting plasma glucose, fasting serum insulin, and C-peptide were measured after overnight fasts of at least 8 hours (n = 1216). Nondiabetic women 45–74 years of age had 75-g oral glucose challenges and blood samples drawn 2 hours later (n = 796). Plasma glucose was measured by using a hexokinase enzymatic reference method (Cobas Mira Chemistry System; Roche Diagnostic Systems, Inc., Montclair, NJ). Serum insulin and C-peptide were measured by using radioimmunoassay (Pharmacia Insulin RIA kit; Pharmacia Diagnostics, Inc., Fairfield, NJ, and Bio-Rad Lyphocheck immunoassay; ECS Division of Bio-Rad Laboratories, Anaheim, CA, respectively) in the Department of Child Health, University of Missouri. The average interassay coefficient of variation was 8.4% for insulin and 11.2% for C-peptide. Intraassay coefficients were less than 10% for both measures.
Undiagnosed diabetes and impaired fasting glucose were defined by fasting glucose levels of at least 126 mg/dL and 110–125 mg/dL, respectively.15 Women who were currently using pills or the intradermal patch were considered current users. Women using HRT in the form of creams, injections or suppositories (0.9% of diabetic women and 2.5% of women without diabetes) were not considered current users. Body mass index (BMI) was calculated from measured height and weight (kg/m2), and waist-to-hip ratio was calculated from measured waist and hip circumferences. An index of physical activity based on performance of nine activities (walking, jogging or running, bicycling, swimming, aerobics, dancing, calisthenics, gardening or yard work, and weight lifting) was calculated by summing the products of frequencies over the last month and intensity ratings.
Statistical analyses were done by using SAS software (Statistical Analysis System, Inc., Cary, NC) with appropriate weights to adjust for the stratified sampling scheme. We used SUDAAN16 to calculate standard errors and tests of statistical significance. Analysis of covariance was used to adjust means for the effects of covariates. Values for hemoglobin A1c, glucose, insulin, and C-peptide were log-transformed because of skewed distributions. In the multivariable models, we tested age, race/ethnicity, parental history of diabetes, BMI, waist-to-hip ratio, physical activity, cigarette smoking, alcohol intake, level of education, and medications known to affect glucose levels for associations with measures of glucose metabolism. Those covariates were retained in the final models if they had a P value < .10 to include marginally statistically significant confounders.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Among diabetic women, mean hemoglobin A1c value was statistically insignificantly lower among current users than never or past users (Table 1
). Current users were younger than never or past users (P < .05), were more often non-Hispanic white persons and were more educated than never users (P < .05), and had longer durations of HRT use than past users (P < .05). The small number of current users (n = 29) precluded further investigation of the association between HRT and hemoglobin A1c.
|
Mean hemoglobin A1c was lower in current users of HRT than in never or past users (Table 2). Mean levels of fasting glucose, insulin, and C-peptide were statistically insignificantly lower in current users. Mean 2-hour glucose levels were statistically insignificantly higher in current users than never and past users. Prevalence of impaired fasting glucose among women without diagnosed diabetes was lower in current users of HRT (5.3%) than in never (14%; P = .001) or past users (13.9%; P = .04). Prevalence of undiagnosed diabetes was also lower among current users (2.6%) than never (5.1%; P = .10) or past users (5.5%; P = .29). The small number of cases among current HRT users precluded further study of the relation of HRT to impaired fasting glucose and undiagnosed diabetes.
|
shows demographic, lifestyle, medical, and anthropometric factors in women without diabetes according to HRT use. Compared with never or past users, current users were younger, were more often non-Hispanic white persons and less often non-Hispanic black persons, more frequently reported parental histories of diabetes, were leaner, had lower waist-to-hip ratios, were more physically active, were more likely to consume alcohol, and were more educated. They also were less likely to have had a natural menopause. Compared with past users, current users were older at first HRT and had used it longer. Current users took HRT almost continuously since beginning therapy, whereas past users took HRT for 4–5 years after menopause.
|
shows age-adjusted and multivariable-adjusted geometric means for hemoglobin A1c, fasting and 2-hour glucose, and fasting insulin and C-peptide levels. Body mass index and waist-to-hip ratio were associated with all measures of glucose metabolism. Race/ethnicity, parental history of diabetes, smoking, and alcohol intake were included in the model for hemoglobin A1c; race/ethnicity and parental history of diabetes were included in the model for fasting glucose; race/ethnicity was included in the model for 2-hour glucose level; BMI squared, alcohol intake, and education were included in the model for fasting insulin level; and race/ethnicity, BMI squared, smoking, and alcohol intake were included in the model for C-peptide level. Multivariable adjustment attenuated the differences in hemoglobin A1c among the three HRT categories, but the mean for current users compared with never and past users remained 0.1% lower. Fasting glucose levels were approximately 3 mg/dL lower in current HRT users than in never and past users in the adjusted models. Multivariable adjustment increased the difference in 2-hour glucose levels (approximately 15 mg/dL) between current users and never and past users. Fasting insulin and C-peptide levels did not differ statistically significantly among the three groups after multivariable adjustment.
|
The associations of HRT with all measures were similar when the multivariable models (Table 4
) were rerun excluding women who were taking any medication that might affect glucose metabolism (diuretics, angiotensin-converting enzyme inhibitors, calcium-channel blockers, ß-blockers, 
-blockers, adrenal corticosteroids, or nasal corticosteroids). Adjusted differences between current and never users for hemoglobin A1c and for fasting glucose were similar when examined by parental history of diabetes, age, race/ethnicity, BMI, menopausal type, or hysterectomy status (data not shown).
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Our results for diabetic women were consistent with those of small randomized trials that showed better glycemic controls in diabetic women using HRT17,18 ( Friday KE. Estrogen replacement therapy improves glycemic control and high density lipoprotein cholesterol concentrations in postmenopausal type II diabetic women [abstract]. Diabetes 1998;47:A357) and a cohort study ( Ferrara A, Karter A, Glei D, Ackerson L, Darbinian J, Selby J. Hormone replacement therapy is associated with better glucose control in a multi-ethnic population with type 2 diabetes: The Northern California Kaiser Permanente Diabetes Registry [abstract]. Diabetes 1998;47:A152). Our results for fasting glucose in nondiabetic women were similar to those of other cross-sectional studies that observed lower levels (2–3 mg/dL) in HRT users compared with nonusers.1,4,5 Those studies found lower fasting insulin levels. In our data, after adjustment for potential confounders, fasting insulin and C-peptide levels were not associated with HRT. Current users of HRT had higher 2-hour glucose values than did never or past users (10 mg/dL with age adjustment and 15 mg/dL after full adjustment). Higher 2-hour glucose values also were found in women who took HRT in the Postmenopausal Estrogen/Progestin Intervention Study5 but not in women in the Rancho Bernardo study.1
We could not evaluate type of HRT preparation. No differences in the relation between current HRT use and hemoglobin A1c value and fasting glucose level were found according to whether women had had hysterectomies (a proxy for using combination therapy or estrogen alone). Two other studies did not observe differences in glucose metabolism by type of HRT.1,5
Our study is limited by its cross-sectional design. The lower values for hemoglobin A1c, fasting glucose, insulin, and C-peptide in current users and their healthier profiles could have resulted from selection and other biases. Women who use HRT are healthier before use and have lower insulin levels.19 Potential users might have been screened to exclude women with adverse risk profiles. Gabal and colleagues3 noted that until recently conjugated estrogen was contraindicated in diabetic women, and they suggested that women at risk of diabetes also might have been discouraged from using it. We found the prevalence of HRT among diabetic women to be half that of nondiabetic women. Lower fasting glucose levels in current users might also be explained by cessation of therapy by women who developed hyperglycemia before the survey. In our data, women who had stopped using HRT more recently did not have elevated glucose values compared with those who had stopped in the more distant past, suggesting that discontinuation was not prompted by development of irreversible abnormalities in glucose metabolism. Data on very recent cessation were sparse, limiting our ability to assess it as a possible bias.
|
Footnotes |
|---|
Received February 2, 2000. Received in revised form May 31, 2000. Accepted June 22, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Manson JE, Rimm EB, Colditz GA, Willett WC, Nathan DM, Arky RA, et al. A prospective study of postmenopausal estrogen therapy and subsequent incidence of non-insulin-dependent diabetes mellitus. Ann Epidemiol 1992;2:665–73.[Medline]
3. Gabal LL, Goodman-Gruen D, Barrett-Connor E. The effect of postmenopausal estrogen therapy on the risk of non-insulin-dependent diabetes mellitus. Am J Public Health 1997;87:443–5.
4. Nabulsi AA, Folsom AR, White A, Patsch W, Heiss G, Wu KK, et al. Association of hormone-replacement therapy with various cardiovascular risk factors in postmenopausal women. The Atherosclerosis Risk in Communities Study Investigators. N Engl J Med 1993;328:1069–75.
5. Espeland MA, Hogan PE, Fineberg WE, Howard G, Schrott H, Waclawiw MA, et al. Effect of postmenopausal hormone therapy on glucose and insulin concentrations. PEPI Investigators. Postmenopausal Estrogen/Progestin Interventions. Diabetes Care 1998;21:1589–95.[Abstract]
6. The Writing Group for the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women. JAMA 1995;273:199–208.[Abstract]
7. Spellacy WN, Buhi WC, Birk SA. The effect of estrogens on carbohydrate metabolism: Glucose, insulin, and growth hormone studies on 171 women ingesting Premarin, mestranol, and ethinyl estradiol for six months. Am J Obstet Gynecol 1972;114: 378–92.[Medline]
8. De Cleyn K, Buytaert P, Coppens M. Carbohydrate metabolism during hormonal substitution therapy. Maturitas 1989;11:235–42.[Medline]
9. O’Sullivan AJ, Ho KKY. A comparison of the effects of oral and transdermal estrogen replacement on insulin sensitivity in postmenopausal women. J Clin Endocrinol Metab 1995;80:1783–8.[Abstract]
10. Cagnacci A, Tuveri F, Cirillo R, Setteneri AM, Melis GB, Volpe A. The effect of transdermal 17ß-estradiol on glucose metabolism of postmenopausal women is evident during the oral but not the intravenous glucose administration. Maturitas 1997;28: 163–7.[Medline]
11. Colacurci N, Zarcone R, Mollo A, Russo G, Passaro M, Seta L, et al. Effects of hormone replacement therapy on glucose metabolism. Panminerva Med 1998;40:18–21.[Medline]
12. Godsland IF, Gangar K, Walton C, Cust MP, Whitehead MI, Wynn V, et al. Insulin resistance, secretion and elimination in postmenopausal women receiving oral or transdermal hormone replacement therapy. Metabolism 1993;42:846–53.[Medline]
13. Stevenson JC, Crook D, Godsland IF, Lees B, Whitehead MI. Oral versus transdermal hormone replacement therapy. Int J Fertil Menopausal Stud 1993;38:30–5.
14. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.
15. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183–97.[Medline]
16. Shah BV, Barnwell BG, Bieler GS. SUDAAN User’s Manual, Release 6.40. Research Triangle Park, NC: Research Triangle Institute, 1995.
17. Brussaard HE, Gevers Leuven JA, Frolich M, Kluft C, Krans HM. Short-term oestrogen replacement therapy improves insulin resistance, lipids and fibrinolysis in postmenopausal women with NIDDM. Diabetologia 1997;40:843–9.[Medline]
18. Andersson B, Mattsson LA, Hahn L, M
in P, Lapidus L, Holm G, et al. Estrogen replacement therapy decreases hyperandrogenicity and improves glucose homeostasis and plasma lipids in postmenopausal women with noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1997;82:638–43.
19. Matthews KA, Kuller LH, Wing RR, Meilahn EN, Plantinga P. Prior to use of estrogen replacement therapy, are users healthier than nonusers? Am J Epidemiol 1996;143:971–8.
This article has been cited by other articles:
|
|
 |
|
  R. E. van Genugten, K. M. Utzschneider, J. Tong, F. Gerchman, S. Zraika, J. Udayasankar, E. J. Boyko, W. Y. Fujimoto, S. E. Kahn, and and the American Diabetes Association GENNID Study Effects of Sex and Hormone Replacement Therapy Use on the Prevalence of Isolated Impaired Fasting Glucose and Isolated Impaired Glucose Tolerance in Subjects With a Family History of Type 2 Diabetes Diabetes, December 1, 2006; 55(12): 3529 - 3535. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Selvin, J. Coresh, S. H. Golden, L. L. Boland, F. L. Brancati, and M. W. Steffes Glycemic Control, Atherosclerosis, and Risk Factors for Cardiovascular Disease in Individuals With Diabetes: The Atherosclerosis Risk in Communities study Diabetes Care, August 1, 2005; 28(8): 1965 - 1973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Body Mass Index and Insulin Resistance Obstet. Gynecol., October 1, 2004; 104(4_suppl): 5S - 10S. [Full Text] [PDF]
|
|
|
|
 |
|
 B. V. Howard, J. Hsia, P. Ouyang, L. Van Voorhees, J. Lindsay, A. Silverman, E. L. Alderman, M. Tripputi, and D. D. Waters Postmenopausal Hormone Therapy Is Associated With Atherosclerosis Progression in Women With Abnormal Glucose Tolerance Circulation, July 13, 2004; 110(2): 201 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Rossi, G. Origliani, and M. G. Modena Transdermal 17-{beta}-Estradiol and Risk of Developing Type 2 Diabetes in a Population of Healthy, Nonobese Postmenopausal Women Diabetes Care, March 1, 2004; 27(3): 645 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Newton, A. Z. LaCroix, S. R. Heckbert, L. Abraham, D. McCulloch, and W. Barlow Estrogen Therapy and Risk of Cardiovascular Events Among Women With Type 2 Diabetes Diabetes Care, October 1, 2003; 26(10): 2810 - 2816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Barrett-Connor Diabetes and Heart Disease Diabetes Care, October 1, 2003; 26(10): 2947 - 2958. [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, S. Nomura, Y. Ikoma, E. Yamamoto, T. Ito, T. Mitsui, K. Tamakoshi, and S. Mizutani Effects of Postmenopausal Hormone Replacement Therapy on HbA1c Levels Diabetes Care, April 1, 2003; 26(4): 1088 - 1092. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Lokkegaard, A T Pedersen, B L Heitmann, Z Jovanovic, N Keiding, Y A Hundrup, E B Obel, and B Ottesen Relation between hormone replacement therapy and ischaemic heart disease in women: prospective observational study BMJ, February 22, 2003; 326(7386): 426 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Kanaya, D. Herrington, E. Vittinghoff, F. Lin, D. Grady, V. Bittner, J. A. Cauley, and E. Barrett-Connor Glycemic Effects of Postmenopausal Hormone Therapy: The Heart and Estrogen/progestin Replacement Study: A Randomized, Double-Blind, Placebo-Controlled Trial Ann Intern Med, January 7, 2003; 138(1): 1 - 9. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Crespo, E. Smit, A. Snelling, C. T. Sempos, and R. E. Andersen Hormone Replacement Therapy and Its Relationship to Lipid and Glucose Metabolism in Diabetic and Nondiabetic Postmenopausal Women: Results from the Third National Health and Nutrition Examination Survey (NHANES III) Diabetes Care, October 1, 2002; 25(10): 1675 - 1680. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
