Comment on "Prevalence and Predictors of Risk for Type 2 Diabetes Mellitus and Impaired Glucose Tolerance in Polycystic Ovary Syndrome"1

Mario Ciampelli, Anna Maria Fulghesu and Antonio Lanzone

Department of Obstetrics and Gynecology Catholic University of Sacred Heart 00168 Rome, Italy

We have read with interest the report by Legro and coworkers (1) about the incidence of abnormal glucose tolerance among a large group of patients with polycystic ovary syndrome (PCOS). The authors found a prevalence of 31.1% for impaired glucose tolerance (IGT) and 7.5% for noninsulin dependent diabetes mellitus (NIDDM), thus suggesting that about 4 out of 10 PCOS patients could show abnormalities of glucose tolerance. In nonobese PCOS women, 10.3% IGT and 1.5% diabetes were found. These data strongly support the idea of PCOS as an endocrinometabolic syndrome, characterized by a very high prevalence of glucose metabolism disturbances. In another recent report, Ehrman and others (2) found in a group of 122 PCOS women, a prevalence of 35% for IGT and 10% for NIDDM, thus suggesting a substantially higher prevalence of glucose abnormalities in PCOS women when compared with age- and weight-matched populations of women with normal ovaries. Both reports hypothesize that the main pathogenetic factors to explain such metabolic complications could consist of obesity (particularly upper-body obesity), genetic factors (Ehrman et al. (2) showed that 83% of the diabetic women with PCOS had a first-degree relative with NIDDM), and insulin resistance, which is a characteristic of obese as well as lean PCOS patients (1, 2). Recently, in a similar study protocol (3), we analyzed the endocrinometabolic features of 110 PCOS women divided into four groups according to their body mass index (BMI) and their insulin response to a glucose load. In our population, which consisted of all native Italians, we did not find subjects affected by NIDDM, while we found IGT only in obese hyperinsulinemic PCOS women, with a prevalence of 15.5% (6.4% if all 110 evaluated subjects were considered).

Our data are clearly in contrast with the other above cited reports (1, 2), and we wonder how such a discrepancy could be explained. We think that a main factor could be the anthropometric characteristics of the investigated patients; indeed, the mean BMI of our obese PCOS individuals (approximately 30–31 kg/m2) was superimposable to that by Legro et al., even if in this last case all PCOS patients (and not only the obese ones) were considered. Furthermore Ehrman and coworkers (2) found mean BMI values of 36.9 kg/m2 for the IGT group and 41.0 kg/m2 for the NIDDM group. For our population it is very difficult for a patient to reach a BMI of 35 kg/m2, whereas a value of 40 kg/m2 is very unusual.

It is thus conceivable that the American obese PCOS individuals are much compromised from a metabolic point of view when compared with Italian people, and that factors other than PCOS status could be involved in the glucose abnormalities of American people, perhaps environmental factors or dietary habits.

Concerning the lean PCOS subjects, we re-evaluated data by Legro and coworkers by excluding the overweight patients; we found a prevalence of 5.2% for IGT and 1.7% for NIDDM. In this case, what could be the risk factors, other than PCOS status? The authors suggested a key role for insulin resistance, which is defined as a constant in PCOS. However, in this case also, there is a discrepancy between the American and the European studies (4), in which a normal insulin-stimulated glucose uptake was found in lean PCOS women. On the other hand several data seem to indicate an imbalance between insulin resistance and hyperinsulinemia in PCOS; indeed, an improvement of insulin sensitivity with no significant effects on early insulin response to IVGTT was found in a group of obese PCOS women (5). Furthermore we demonstrated that, in PCOS subjects, naltrexone (an oral opioid antagonist) administration was able to reduce the insulin response to a glucose load without influencing the peripheral insulin resistance (6). The above cited data seem to confirm indirectly that reduced insulin sensitivity and hyperinsulinemia in response to a glucose load are two distinct, but not necessarily related, features of the insulin disorders of the syndrome, thus arguing against the presence of hyperinsulinemia as a simple compensatory ß-cell response to insulin resistance, as hypothesized by the authors in explaining the pathogenetic mechanisms leading to IGT and NIDDM in lean PCOS subjects.

Finally, Legro et al. (1) found IGT in 14% of control women, whereas they reported that the prevalence rates of glucose disorders in a major population-based study of U.S. women of similar age is 8.8%. How do the authors explain the slightly higher prevalence found in their data? Are there other additional hypothetical risk factors in the analyzed populations?

After all, we think that the message rising from the data by Legro and coworkers, and confirmed by Ehrman et al. (2), is that the presence of glucose abnormalities in nearly half the PCOS women is quite alarming, probably due to risk factors other than PCOS status or related features and not necessarily applicable to European populations. We would welcome an authors’ opinion about our comments.

Footnotes

1 Address correspondence to: Antonio Lanzone, Department of Obstetrics and Gynecology, Catholic University of the Sacred Heart, L. Go A. Gemelli 8, Rome, Italy 00168. Back

Received March 16, 1999.

References

  1. Legro RS, Kunselman AR, Dodson WC, Dunaif A. 1999 Prevalence and predictors of risk for type 2 Diabetes Mellitus and impaired glucose tolerance in polycystic ovary syndrome: a prospective, controlled study in 254 affected women. J Clin Endocrinol Metab. 84:165–169.[Abstract/Free Full Text]
  2. Ehrman DA, Barnes RB, Rosenfield RL, Cavaghan MK, Imperial J. 1999 Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care. 22:141–146.[Abstract]
  3. Ciampelli M, Fulghesu AM, Cucinelli F, et al. 1999 Impact of insulin and body mass index on metabolic and endocrine variables in polycystic ovary syndrome. Metabolism. 48:167–172.[Medline]
  4. Ciampelli M, Lanzone A. 1998 Insulin and polycystic ovary syndrome: a new look at an old subject. Gynecol Endocrinol. 12:277–292.[Medline]
  5. Holte J, Bergh T, Berne C, Wide L, Lithell H. 1995 Restored insulin sensitivity but persistently increased early insulin secretion after weight loss in obese women with polycystic ovary syndrome. J Clin Endocrinol Metab. 80:2586–2593.[Abstract]
  6. Fulghesu AM, Ciampelli M, Guido M, et al. 1998 Role of the opioid tone in the pathophysiology of hyperinsulinemia and insulin resistance in polycystic ovarian disease. Metabolism. 47:158–162.[Medline]