| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Obstetrics and Gynecology (R.S.L., C.L.G.) and Health Evaluation Sciences (A.R.K.), Penn State College of Medicine, Hershey, Pennsylvania 17033; and Division of Endocrinology, Metabolism, and Molecular Medicine (A.D.), Feinberg School of Medicine, Northwestern University, Chicago, Illinois 60611
Address all correspondence and requests for reprints to: Richard S. Legro, M.D., Department of Obstetrics-Gynecology, 500 University Drive, Pennsylvania State University College of Medicine, M. S. Hershey Medical Center, Hershey, Pennsylvania 17033. E-mail: rsl1{at}psu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
IGT is a major risk factor for type 2 DM as well as an independent risk factor for cardiovascular disease and sudden death, particularly in women (8). Several landmark clinical trials have shown lifestyle changes or pharmacologic interventions can delay the development of type 2 DM in individuals with IGT (9, 10). Accordingly, there has been renewed interest in identifying individuals at risk for IGT. Multiple professional groups including the American Diabetes Association (11), The American College of Obstetricians and Gynecologists (12), and a joint American Society of Reproductive Medicine/European Society of Human Reproduction and Endocrinology PCOS Consensus Workshop Group (13) have published guidelines recommending screening women with PCOS for glucose intolerance and diabetes with an oral glucose tolerance test (OGTT). Nevertheless, there have been limited studies of the natural history of glucose intolerance in this population to determine whether conversion rates to DM are similar to those in other populations. For example, there is a substantial spontaneous reversion rate to normal glucose tolerance, especially given the known variability in oral glucose testing (14, 15).
Two uncontrolled studies (4, 16) have suggested that the risk of conversion to DM is very high in women with PCOS. However, the attributable risk due to a diagnosis of PCOS, in addition to other risk factors such as aging or obesity, is unknown. These considerations are important when developing recommendations for the interval in rescreening women with PCOS with oral glucose tolerance testing. We performed this study to determine the changes in glucose tolerance over time in women with PCOS, but we also studied a group of reproductively normal women, without IGT or type 2 DM, to assess the independent impact of PCOS.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We studied 94 women (71 women with PCOS and 23 control women) at the M. S. Hershey Medical Center of the Pennsylvania State University. The ethnicity of the subjects was 69 Caucasian and two African-American women with PCOS and 21 Caucasian and two African-American control women. The Investigational Review Board at the Hershey Medical Center approved these studies, and all subjects gave written informed consent to both visits. At the baseline visit, all women were in good health, euthyroid, and, for at least 1 month before each study, not taking any medication (except for oral contraceptive agents, which were stopped for 3 months before study) known to affect sex hormone or carbohydrate metabolism.
The diagnosis of PCOS was made by the presence of chronic anovulation (six or fewer menses per year) in association with elevated circulating androgen levels (1). Nonclassical 21-hydroxylase deficiency, hyperprolactinemia, and androgen-secreting tumors were excluded by appropriate tests before the diagnosis of PCOS was made (2). No woman with known diabetes was enrolled in the study at baseline. Control women with 27- to 35-d menstrual cycles and no hirsutism (Ferriman Gallwey score < 8) were studied as a reference population. To control for conditions altering insulin action, control women did not engage in regular aerobic exercise, nor did they have a history of hypertension or diabetes or a family history of DM in a first-degree relative, and all had normal glucose tolerance (NGT) at baseline (17, 18, 19). Control women had NGT based on a 75-g, 2-h OGTT applying World Health Organization (WHO) criteria (20).
This study was intended as an initial study to provide data for estimating sample size for a larger study, and we sought a sample size of approximately 75 women with PCOS and approximately 25 control women to examine trends in both populations. Women who had participated in prior studies were randomly recruited for this study until our projected sample size was reached. No specific recommendations were made for management of their PCOS between visits, and care was given by their referring physicians. At the follow-up visit, similar conditions applied to both control women and women with PCOS, except that at the time of the follow-up visit, we did not exclude women on current medications. Therefore, 10 women were receiving oral contraceptives or depot contraceptive agents (PCOS: n = 8, controls: n = 2). Four women with PCOS were being treated with metformin, two of whom had been diagnosed with DM at the baseline visit and were excluded from all analyses given metformin’s insulin-sensitizing and glucose-lowering effects (21).
Study protocol
All studies were performed 0800–1000 h after a 3-d, 300-g carbohydrate diet and an overnight fast or 10–12 h. Studies were performed in women with PCOS without regard to the last episode of vaginal bleeding and in controls from d 1 to 10 of the follicular phase of the menstrual cycle. Height and weight were obtained on all subjects, and body mass index (BMI) was calculated. BMI, blood pressure, waist circumference, and hip circumference were obtained at both visits as reported (22). An iv catheter was inserted, and the vein was kept open with an infusion of 0.9% normal saline at 30 cc/h. All subjects received a 75-g oral glucose load, and blood samples were obtained at 0 and 120 min (3). Additional blood samples were obtained at 30, 60, and 90 min in 64 women with PCOS and eight control women at the baseline visit and in 60 women with PCOS and 22 control women at the follow-up visit.
Testosterone (T), non-SHBG, bound testosterone (uT), and dehydroepiandrosterone sulfate (DHEAS) levels were determined in the fasting (0 h) blood sample (23). Plasma glucose levels were determined in all samples by the glucose oxidase technique. Hemoglobin A1c (follow-up visit only) levels were determined by HPLC in the fasting (0 h) sample (normal range 4.0–6.0%). Insulin levels were determined in all samples with a double-antibody radioimmunoassay method using reagents obtained from Linco Research, Inc. (St. Charles, MO). In seven control subjects, insulin levels at the baseline visit were determined using reagents from Diagnostic Products Corp. (Los Angeles, CA) (23). A validation study between the two assays showed they were highly correlated. For the 0 h insulin level, the correlation coefficient between the two assays was R = 0.96 (n = 76); for the 2-h insulin level, the correlation value was 0.98 (n = 86) (our unpublished data). All assays had method coefficients of variation of 10% or less (23, 24).
Fasting glucose levels were classified by the revised 1997 American Diabetes Association (ADA) criteria. Normal fasting glucose (NFG) was less than 110 mg/dl (<6.06 mmol/liter), impaired fasting glucose (IFG) was 110–126 mg/dl (6.06–6.94 mmol/liter), and type 2 diabetes (fasting DM) was 126 mg/dl or more (>6.94 mmol/liter) (25). Conversion was viewed as worsening of category over time, and reversion was viewed as an improvement. For ADA fasting glucose categories, conversion is a change from NFG to IFG or type 2 DM or a change from IFG to type 2 DM, and reversion is an improvement from type 2 DM to IFG or NFG or from IFG to NFG. Glucose tolerance was assessed by WHO criteria (20). NGT is defined as a 2-h glucose postchallenge value less than 140 mg/dl (<7.71 mmol/liter), IGT is defined as a 2-h postchallenge value from 140 to 199 mg/dl (7.71–10.96 mmol/liter), and type 2 DM is defined as a 2-h postchallenge value of 200 mg/dl or more (>11.02 mmol/liter). For WHO glucose tolerance categories, conversion is a change from NGT to IGT or type 2 DM or a worsening from IGT to type 2 DM, and reversion is a change from type 2 DM to IGT or NGT or from IGT to NGT. An insulin sensitivity index (ISI0,120) was determined based on the fasting (0 h) and 2-h OGTT glucose and insulin levels (26). This index is highly correlated with the rate of whole-body glucose disposal determined by the euglycemic hyperinsulinemic clamp technique (26).
Data analysis
Because all control women had NGT at the baseline visit using the WHO criteria, valid inferential comparisons could be made only between the NGT controls and women with PCOS who had NGT at the baseline visit. Descriptive statistics, however, are reported for the non-NGT at baseline women with PCOS for informational purposes and to assess the reversion rate within women with PCOS. Continuous data are reported as the mean (±SD). Repeated-measures analysis of covariance models, adjusting for baseline age and BMI, were fit to the data to assess the effect of time within and between NGT women with PCOS and NGT control women for all continuous outcomes. P values and 95% confidence intervals (CIs) from the repeated-measures models were adjusted for multiple testing using Bonferroni’s method for each outcome. Categorical data were analyzed using either a Pearson’s 
2 test or Fisher’s exact test with odds ratios and associated 95% CIs reported where appropriate. Furthermore, the analyses were conducted twice by including and excluding women receiving contraceptive steroids with no significant differences observed between the results. Therefore, we included the subjects receiving contraceptive agents in the reported analyses.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Control women were older and women with PCOS were heavier. Women with PCOS were more hyperandrogenemic and had higher mean levels of glucose and insulin during the OGTT. Women with PCOS had evidence for insulin resistance by fasting measures such as insulin levels or glucose to insulin ratios or the ISI.
|
reports changes in these parameters at the follow-up visit for the women with NGT at the baseline visit. The mean follow-up was between 2 and 3 yr for both groups (baseline NGT PCOS: 2.8 ± 1.8 yr; baseline NGT controls: 2.9 ± 2.1 yr) (Table 2). There were few changes over time within each group (either PCOS or control). There was no change in ISI0,120 for the baseline NGT control group or the baseline NGT PCOS group. There also was little difference in changes over time in these parameters between the two groups.
|
, whereas the distribution of calculated insulin sensitivities, ISI0,120, in PCOS and control women is found in Fig. 2. There were no significant differences in individual measures at specific time points during the OGTT or in integrated measures of glucose and insulin (Table 2
) within or between groups. We did not measure hemoglobin A1c levels at the baseline visit. At the follow-up visit the mean hemoglobin A1c level was 6.0 ± 0.8% in women with PCOS who had baseline NGT, compared with 5.3 ± 0.7% in the control women who had NGT (P = 0.003).
|
|
The change in characteristics from baseline to follow-up was examined in women with baseline NGT using the WHO criteria. The characteristic changes were assessed within the WHO change in glucose tolerance group (PCOS conversion, PCOS unchanged, control conversion, and control unchanged) (data not shown). Women with PCOS who converted had a significant interval decrease in ISI0,120. The mean change over time in ISI0,120 was –14.89 [95% CI (–23.1, –6.7), P < 0.001]. No significant association was found with any fasting or 2-h measure of insulin sensitivity.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Furthermore, the glycohemoglobin levels at the upper range of normal at the follow-up visit suggest that ambient glucose levels throughout the day are primarily within the normal range in women with PCOS, although these levels are significantly higher than those in the control women. Insulin sensitivity calculated from the OGTT measures did not change in our study groups. Although a crude measure of ß-cell function, the lack of change in area under the curve (AUC) insulin secretion at the follow-up visit, compared with the baseline visit, in the face of stable glucose levels in women with PCOS suggest maintenance of adequate ß-cell function for the degree of peripheral insulin resistance (27).
Our results differ from previous studies of changes in glucose tolerance in women with PCOS (4, 16). Ehrmann et al. (4) reported a significant increase in the mean 2-h glucose value over a 3-yr average period of follow-up to 161 ± 9 from 139 ± 6 mg/dl in a follow-up study of 25 PCOS women. Our population did not show a significant change in 2-h glucose levels over a similar period of follow-up. Recently Norman et al. (16) reported a trend toward worsening glucose tolerance in a study of 67 women with PCOS after an average follow-up of 6.2 yr. The conversion rate for normoglycemic women at baseline was 9% (five of 54) to IGT and 8% (four of 54) to type 2 diabetes. For women with IGT at baseline, 54% (seven of 13) had type 2 diabetes at follow-up. Our study showed a worse prognosis for women with PCOS and NGT at baseline and a better prognosis for women with PCOS and IGT at baseline, who had a higher reversion rate to NGT and a lower conversion rate to type 2 DM than the study of Norman et al. (16). The latter study, however, had a substantially longer follow-up period of 6 yr, compared with our follow-up interval of approximately 3 yr.
In other populations, IGT is a risk factor for the development of type 2 diabetes with an average conversion rate of 1–5% per year (28, 29). In the present study, women with PCOS and IGT had a similar risk with a net conversion of 6% to type 2 diabetes (three of 50) over approximately 3 yr, or 2% per year. There are certainly other groups of women at higher risk for developing type 2 diabetes, such as Latina women with a history of gestational DM, whose cumulative conversion rates to diabetes may be as high as 50% over 5 yr or 10% per year (30). The effect of PCOS is more pronounced in the 40% conversion to IGT, given NGT at baseline.
It is important to note that this study confirms that at both baseline and follow-up, fasting glucose levels in PCOS tend to be normal and furthermore are not associated with WHO categories of glucose tolerance as we have previously reported (3). Limited studies of glycohemoglobin levels in women with PCOS have shown they tend to be normal, even in those women with IGT (31), and our results are consistent with this study. Elevated fasting glucose and glycohemoglobin levels are better predictors of both macro- and microvascular disease (32, 33). However, large population studies have shown that IGT with normal fasting glucose levels identifies populations with an adverse cardiovascular risk profile and a large attributable risk of sudden death (8, 34). Furthermore, IGT is better predictor of these outcomes than IFG (8, 34). A fasting glucose level is therefore inadequate to identify all women with PCOS at increased risk for developing type 2 diabetes or vascular complications.
Our sample size limits the power to detect predictive factors for worsening glucose tolerance. Increasing age is a risk factor, and in many studies so is decreasing insulin sensitivity. However as a group there was no change in insulin sensitivity in women with PCOS over time. The ISI0,120 may not be the best marker of insulin sensitivity but is better than homeostatic measures alone. Another derived index from all values obtained during the OGTT, which is more closely correlated with insulin sensitivity as obtained from clamp studies, has been advocated by Matsuda and DeFronzo (35). However, when we analyzed our data in the applicable subset (data not shown), we found our results were the same with no change in insulin sensitivity over time. In larger population studies, increasing age, BMI, an unfavorable body fat distribution, and parameters of insulin sensitivity and secretion during the OGTT have been predictive for conversion to diabetes (28, 30). Our population is more obese, compared with the general population, and do not include substantial numbers of lean women with PCOS. In our study we also lacked more sensitive measures of body fat distribution such as a computed tomography scan, and undetected changes in visceral fat may have affected the results.
There were many limitations to our study. Our control group was small and varied in age and weight, compared with our PCOS group. There was an inherent selection bias because we were unable to repeat OGTTs in all women due to both logistic and financial constraints. Therefore, it is possible that the women who returned for repeat testing were more likely to be abnormal. A prospective case-control study with a predetermined interval of follow-up would be a stronger design.
Medications were also a confounder. We elected to include women on oral contraceptives in the analysis because we found no difference in OGTT responses at follow-up whether these women on oral contraceptives were included or excluded. This finding is consistent with many previous small studies that have shown minimal (n = 23) (36) or no effect (n = 20) (37) of oral contraceptive agents on glucose tolerance or diabetes risk in women with PCOS. However, there are also similarly small studies that have reported short-term deterioration of glucose tolerance in hyperandrogenic women (n = 16) (38) and women with PCOS (39) treated with oral contraceptive agents for weeks or months. Thus, there are insufficient data to draw any conclusions about the long-term effects of oral contraceptives on glucose metabolism in women with PCOS. We also excluded women currently using metformin because multiple studies and a recent metaanalysis have shown a significant improvement in glycemic parameters in women with PCOS treated with this medication (21). This beneficial effect may have led to an underestimation of the conversion risk to worsening glucose tolerance, if women were studied in a medication-free state. However it is difficult, if not beyond the emerging standard of care in the larger clinical arena, to find women with hyperandrogenism, chronic anovulation, and insulin resistance free of all confounding medications.
Our study had a low power based on sample size to detect differences in glucose tolerance within and between groups. This study was intended as an initial study to provide data for estimating sample size for a larger prospective study. Based on our study findings, and choosing the end point of conversion to IGT from NGT, we would need a sample size of 117 PCOS cases and 117 controls to have at least 85% power to detect a difference between the PCOS conversion proportion of 0.40 and the control conversion proportion of 0.22 (odds ratio of 2.4 of a women with PCOS converting, compared with control) from NGT to IGT using a two-sided 2 test having a 5% significance level. Furthermore, a substantially larger sample size would be required to detect differences in the conversion of IGT to type 2 DM based on the low rate of conversion in the present study. Our study supports that women with PCOS should be periodically rescreened for diabetes similar to other IGT populations but probably less frequently than on an annual basis.
|
Footnotes |
|---|
First Published Online March 29, 2005
Abbreviations: AUC, Area under the curve; BMI, body mass index; CI, confidence interval; DHEAS, dehydroepiandrosterone sulfate; DM, diabetes mellitus; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; ISI, insulin sensitivity index; NFG, normal fasting glucose; NGT, normal glucose tolerance; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome; T, testosterone; uT, unbound T.
Received September 16, 2004.
Accepted March 22, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Mukherjee, N. Shaikh, S. Khavale, G. Shinde, P. Meherji, N. Shah, and A. Maitra Genetic variation in exon 17 of INSR is associated with insulin resistance and hyperandrogenemia among lean Indian women with polycystic ovary syndrome Eur. J. Endocrinol., May 1, 2009; 160(5): 855 - 862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. I. Cheang, J. M. Huszar, A. M. Best, S. Sharma, P. A. Essah, and J. E. Nestler Long-term effect of metformin on metabolic parameters in the polycystic ovary syndrome Diabetes and Vascular Disease Research, April 1, 2009; 6(2): 110 - 119. [Abstract] [PDF]
|
|
|
|
 |
|
 A.J. Goverde, A.J.B. van Koert, M.J. Eijkemans, E.A.H. Knauff, H.E. Westerveld, B.C.J.M. Fauser, and F.J. Broekmans Indicators for metabolic disturbances in anovulatory women with polycystic ovary syndrome diagnosed according to the Rotterdam consensus criteria Hum. Reprod., March 1, 2009; 24(3): 710 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Palomba, A. Falbo, F. Zullo, and F. Orio Jr. Evidence-Based and Potential Benefits of Metformin in the Polycystic Ovary Syndrome: A Comprehensive Review Endocr. Rev., February 1, 2009; 30(1): 1 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Ciaraldi, V. Aroda, S. Mudaliar, R. J. Chang, and R. R. Henry Polycystic Ovary Syndrome Is Associated with Tissue-Specific Differences in Insulin Resistance J. Clin. Endocrinol. Metab., January 1, 2009; 94(1): 157 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Svendsen, L. Nilas, K. Norgaard, J.-E. B. Jensen, and S. Madsbad Obesity, body composition and metabolic disturbances in polycystic ovary syndrome Hum. Reprod., September 1, 2008; 23(9): 2113 - 2121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Nestler Metformin for the Treatment of the Polycystic Ovary Syndrome N. Engl. J. Med., January 3, 2008; 358(1): 47 - 54. [Full Text] [PDF]
|
|
|
|
 |
|
 H. I Katcher, R. S Legro, A. R Kunselman, P. J Gillies, L. M Demers, D. M Bagshaw, and P. M Kris-Etherton The effects of a whole grain enriched hypocaloric diet on cardiovascular disease risk factors in men and women with metabolic syndrome Am. J. Clinical Nutrition, January 1, 2008; 87(1): 79 - 90. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. E. S. Salley, E. P. Wickham, K. I. Cheang, P. A. Essah, N. W. Karjane, and J. E. Nestler POSITION STATEMENT: Glucose Intolerance in Polycystic Ovary Syndrome A Position Statement of the Androgen Excess Society J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4546 - 4556. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hahn, M. Backhaus, M. Broecker-Preuss, S. Tan, T. Dietz, R. Kimmig, M. Schmidt, K. Mann, and O. E Janssen Retinol-binding protein 4 levels are elevated in polycystic ovary syndrome women with obesity and impaired glucose metabolism Eur. J. Endocrinol., August 1, 2007; 157(2): 201 - 207. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez, A. Jaramillo, G. Enriquez, E. Miro, A. Lopez-Bermejo, D. Dunger, and F. de Zegher Polycystic ovaries after precocious pubarche: relation to prenatal growth Hum. Reprod., February 1, 2007; 22(2): 395 - 400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Lo, S. L. Feigenbaum, G. J. Escobar, J. Yang, Y. M. Crites, and A. Ferrara Increased Prevalence of Gestational Diabetes Mellitus Among Women With Diagnosed Polycystic Ovary Syndrome: A population-based study Diabetes Care, August 1, 2006; 29(8): 1915 - 1917. [Full Text] [PDF]
|
|
|
|
|
|
E. Tasali, E. Van Cauter, and D. A. Ehrmann Relationships between Sleep Disordered Breathing and Glucose Metabolism in Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 36 - 42. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
