This article is more than 5 years old. Information may no longer be current.
Metabolic abnormalities persist for lean women with PCOS
Click Here to Manage Email Alerts
The prevalence of obesity in a cohort of Italian women with polycystic ovary syndrome was much lower than that of similar women living in the United States; however, metabolic risks for these women persist despite having a lower BMI, according to findings published in Clinical Endocrinology.
“PCOS presents with several phenotypes, and in order to determine the metabolic risks it is necessary to study a less obese population,” Roger A. Lobo, MD, a professor of obstetrics and gynecology/reproductive endocrinology at Columbia University, New York, told Endocrine Today. “As shown in our study, even when women are less obese and perhaps have a better diet (Mediterranean), they still have certain metabolic risk factors. This is also true even if they are ovulatory. The less obese phenotype C (ovulatory) women had abnormalities in glucose and lipids, suggesting an important role for androgens as well. Only women with phenotype D, if they are not obese, have risk factors similar to controls.”
Lobo and colleagues analyzed data from 1,215 women with PCOS stratified by the four Rotterdam phenotypes, along with 108 matched ovulatory, non-hyperandrogenic women, evaluated at the department of health sciences at the University of Palermo in Italy between July 2008 and June 2018 (mean age, 24 years). Phenotype A is defined as chronic anovulation, hyperandrogenism and polycystic ovaries on ultrasound. Phenotype B is defined as chronic anovulation and hyperandrogenism, but without polycystic ovaries on ultrasound. Phenotype C represents women who are ovulatory with regular cycles but with hyperandrogenism and polycystic ovaries; phenotype D represents women who have irregular cycles and polycystic ovaries on ultrasound, but with no evidence of hyperandrogenism. Researchers measured BMI, fasting glucose, total cholesterol, HDL and LDL cholesterol and triglycerides. Women also underwent an oral glucose tolerance test.
Within the cohort, 57.7% had phenotype A, 9.2% had phenotype B, 30% had phenotype C and 3.1% had phenotype D.
Compared with controls, women with PCOS overall were more likely to have obesity (31% vs. 6%), metabolic syndrome (6.6% vs. 2.1%), type 2 diabetes (2.1% vs. 1.4%) and prediabetes (13.1% vs. 3.5%; P < .01 for all). Women with PCOS were also more likely to have low HDL cholesterol (45% vs. 8%) and higher LDL cholesterol (17.9% vs. 2.5%; P < .01 for all) vs. controls.
Researchers found that women with phenotype B had a higher BMI and a greater prevalence of obesity vs. women with other PCOS phenotypes and controls (P < .01), followed by phenotype A and phenotype C. BMI and obesity prevalence was similar when comparing women with phenotype D and controls, according to researchers.
Compared with controls and women with phenotype D, women with phenotypes A, B and C had higher levels of serum insulin and insulin resistance as measured by homeostatic model assessment of insulin resistance (HOMA-IR), according to researchers. There were no between-group differences in mean fasting blood glucose.
Across phenotypes, women with phenotype B also had the highest prevalence of metabolic syndrome (17.6%), followed by women with phenotype A (6.5%), phenotype C (3.6%) and women with phenotype D (2.7%), who had a prevalence similar to controls (2.5%; P < .01 for all).
“In spite of the relatively lower prevalence of obesity and metabolic syndrome in our PCOS population, some lipid alterations were very common,” the researchers wrote. “In particular, low HDL [cholesterol] and high LDL [cholesterol] values were observed in 44% and in 18% of our young, relatively lean Mediterranean women with PCOS from Sicily. These lipid alterations are similar to those found in U.S. studies. Only [triglycerides] were very different ... with only 3.5% of our patients having this lipid alteration, compared to 31.4% in the U.S. studies. This difference may be explained, at least in part, by differences in diet in the two populations.”
Multivariate analysis showed that BMI predicted abnormalities in fasting glucose and triglycerides (P < .01); however, there was no association with total testosterone, according to the researchers.
“Screening for metabolic disorders is important in all women with PCOS, but particularly if they are overweight or obese,” Lobo said.
Lobo added that long-term, prospective follow-up of the various PCOS phenotypes is needed to determine what risks women have going forward. – by Regina Schaffer
For more information:
Roger A. Lobo , MD, can be reached at Columbia University, Department of Obstetrics and Gynecology, 622 W. 168th Street, New York, NY 10032; email: ral35@columbia.edu.
Disclosures: The authors report no relevant financial disclosures.
Click Here to Manage Email Alerts