Severity of hyperandrogenism in PCOS can differ across ethnicities

Women from Iceland, Boston had differing testosterone, LH and DHEAS levels.

Issue: December 2006

Women of different ethnic backgrounds from Iceland and Boston, had significant differences in some of the reproductive features of polycystic ovary syndrome. Researchers did not see differences with regard to glucose and insulin in these populations.

“Polycystic ovary syndrome is diagnosed based on a constellation of signs and symptoms that have varied across time and between investigators,” the researchers wrote in the Journal of Clinical Endocrinology and Metabolism. “Ethnicity also influences the variability of associated symptoms, both on a genetic and environmental basis. Cross-population studies are therefore necessary to ascertain key differences in phenotypes that may point to genetic variability or genetic or environmental modifiers of the PCOS phenotype in different ethnic groups.”

Corrine K. Welt, MD, an assistant professor of medicine at Massachusetts General Hospital, and colleagues conducted the study.

PCOS was defined based on National Institutes of Health criteria to “ensure a severely affected phenotype that would presumably capture the metabolic and reproductive abnormalities known to accompany this syndrome.”

Study populations

The study included 105 women in Iceland and 262 women in Boston. An Icelandic control group of 32 women without PCOS was also included, and historic controls with regular menses and documented ovulation were used for the Boston group. The researchers defined PCOS as chronic oligomenorrhea and clinical and/or biochemical hyperandrogenism. All patients underwent a transvaginal ultrasound and had blood samples drawn for measurements of lipid, glucose, insulin, gonadotropin and sex steroid levels.

Of the 262 women in Boston, 172 reported their race as non-Hispanic white. This group was compared with the Icelandic population, all of whom reported their race as white. A secondary analysis among the different ethnic groups in Boston was also performed.

Icelandic women were taller and weighed more than Boston women, but BMI did not differ between the groups. Iceland patients had greater hip circumference and a smaller waist-hip ratio. Although diastolic blood pressure did not differ, Icelandic women had higher systolic BP; also, a higher percentage of Icelandic women had elevated systolic BP (above 130 mm Hg; 34.4% vs. 22.4%; P<.05). Both the Ferriman Gallwey scores and the percentage of individuals with an elevated Ferriman Gallwey score, acne and acanthosis were lower in Icelandic women with PCOS.

Reproductive differences

Icelandic women had lower luteinizing hormone levels (23.1 IU/L vs. 27.6 IU/L; P<.05) as well as LH-follicle-stimulating hormone ratios (2.3 vs. 2.7; P<.05) than women from Boston.

“In addition, androstenedione levels and the androstenedione to testosterone ratio were higher, and testosterone, free testosterone and [dehydroepiandrosterone] concentrations were lower in Icelandic compared with Boston participants with PCOS,” the researchers wrote. Among women with a BMI above 30, those in Iceland had higher androstenedione levels than did those in Boston (4.2 ng/mL vs. 3.5 ng/mL; P<.001).

Researchers did not find many significant differences with regard to the metabolic features of PCOS between populations. Glucose, insulin, cholesterol and triglycerides were similar in Boston and Icelandic women, although HDL cholesterol levels were lower in Icelandic women. There were no differences in the prevalence of impaired fasting glucose, type 2 diabetes or metabolic syndrome.

Ultrasound data showed that Icelandic women had smaller maximum ovarian volumes and a smaller number of follicles observed in a single plane compared with Boston women. There were no differences in the proportion of ovaries found to meet the criteria for polycystic ovary morphology (>90% in both groups).

Study significance

“The study demonstrates that there can be differences in the severity and manifestations of hyperandrogenism in women with PCOS in different ethnic groups,” Welt told Endocrine Today. “We have also demonstrated differences in HDL and systolic BP in different ethnic groups suggesting potential differences in cardiovascular risk.”

Welt said that the differences in testosterone between the groups suggest that there could be differences in the enzyme 17 beta-hydroxysteroid dehydrogenase. “The differences in LH and DHEAS could also be related to subtle genetic variations between the populations.”

With regard to how these findings might affect diagnosis of PCOS, Welt said that androstenedione could be an important factor in documenting hyperandrogenism, especially among ethnic groups with less hirsutism. Also, patients with irregular menses and documented hyperandrogenism probably do not require ultrasound confirmation, as polycystic ovary morphology is found in virtually all of this subset of women.

“This study points to the need for further research into ethnic differences in PCOS,” Welt said. “These studies would facilitate comparison of data across ethnic groups and allow pooling of patient data for large-scale genetic studies.” –by Dave Levitan

For more information:

Welt CK, Arason G, Gudmundsson JA, et al. Defining constant versus variable phenotypic features of women with polycystic ovary syndrome using different ethnic groups and populations. J Clin Endocrinol Metab. 2006;91:4361-4368.