Birth Control Pills May Curb PCOS-Related Diabetes Risk

Certain oral contraceptives were associated with a lower risk for type 2 diabetes and prediabetes in women with polycystic ovary syndrome (PCOS), according to a retrospective, population-based study.

An analysis of 64,051 women with PCOS found an approximately twofold higher risk for type 2 diabetes (adjusted HR 2.04, 95% CI 1.89-2.20, P<0.001) and dysglycemia (aHR 1.87, 95% CI 1.78-1.97, P<0.001) compared with the general population, reported Wiebke Arlt, MD, of the University of Birmingham in England, and colleagues.

This elevated risk appeared to be similar across overweight, underweight, and normal weight women, the group wrote online in Diabetes Care.

However, combined oral contraceptive pills (COCPs) appeared to offset this risk during the median 3.5 years follow-up, as women prescribed them saw a 16% to 17% reduced risk for type 2 diabetes. The risk reduction was similar for contraceptives with an antiandrogenic progestin component (aHR 0.84, 95% CI 0.73-0.97, P=0.020) and without (aHR 0.83, 95% CI 0.72-0.94, P=0.005).

Arlt and colleagues also said that in a secondary case-control analysis of 2,407 women with PCOS, those prescribed oral contraceptives had a 28% reduced odds for developing dysglycemia (aOR 0.72, 95% CI 0.59-0.87). This reduced risk was observed among all BMI subgroups.

“These data suggest that, rather than obesity in isolation, PCOS-specific factors, including androgen excess, underpin the increased metabolic risk,” the researchers wrote. “We found that those women with PCOS and hirsutism, a clinical feature of androgen excess, had a further increased risk of dysglycemia.”

Previous research has linked androgen excess with insulin resistance and type 2 diabetes risk, they also noted.

“We knew from previous, smaller studies, that women with PCOS have an increased risk of type 2 diabetes,” Arlt added in a statement. “However, what is important about our research is that we have been able to provide new evidence from a very large population-based study to show for the very first time that we have a potential treatment option — combined oral contraception — to prevent this very serious health risk.”

COCPs, which are widely prescribed to women with PCOS for menstrual cycle regulation, can exert antiandrogen effects through two distinct mechanisms, the study authors explained. The estrogen component in COCPs increases the production of sex hormone binding globulin in the liver, thereby reducing the concentration of free testosterone capable of binding and activating androgen receptors. Additionally, they noted some progestins used in COCPs can convey additional antiandrogenic action through androgen receptor blockade.

Arlt’s group analyzed medical records from a large U.K. primary care database, The Health Improvement Network. The 64,051 women with PCOS were matched with 123,545 control subjects. The mean age of the cohort was 30.5, and the mean BMI was 25.6.

The secondary nested pharmacoepidemiological case-control study included 2,407 women with PCOS. Approximately 43% of these women were prescribed oral contraceptives. This smaller study investigated the relationship between COCPs and dysglycemia, which the researchers defined as an HbA1c of 6.0% or higher. Type 2 diabetes was defined as an HbA1c of 6.5% or higher.

All models were adjusted for age, BMI, socioeconomic status, ethnicity, smoking status, as well as a record of hypertension, hypothyroidism, and use of prescription lipid-lowering medications.

The researchers noted some limitations to their study. It is possible that the lower dysglycemia risk in women with PCOS on oral contraceptives might be the result of a prescription-by-indication bias, explaining that women with cardiovascular risk factors such as obesity, dyslipidemia, and hypertension may have been less likely to be prescribed contraceptives.

In addition, the research was based on diagnostic codes in the medical records. However, some women may have had undiagnosed PCOS, seeing how the prevalence of PCOS in the study was lower than other published community prevalence data.

Finally, the retrospective analyses could not establish causality. A large-scale randomized clinical trial would be necessary for that, the study authors said.