Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SAT 532-553-Hyperandrogenic Disorders
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-544
Karla A Temple*, Rachel Leproult, Harry Whitmore, Babak Mokhlesi, Eve Van Cauter and David A Ehrmann
University of Chicago, Chicago, IL
Obstructive sleep apnea (OSA) is a frequent co-morbidity of PCOS and is an independent risk factor for insulin resistance.  Adiponectin, an adipocytokine, acts to decrease hepatic glucose output thereby attenuating hyperinsulinemia and insulin resistance.  Studies that have examined the impact of OSA on adiponectin levels have been primarily conducted in men.  The purpose of this study was to determine the impact of OSA on mean adiponectin levels in women with and without PCOS and to examine the relationships between adiponectin levels, leptin levels, and insulin resistance in both groups of women.

Thirty-two women (PCOS n=18, Control n=14) had a 2-h 75-g OGTT to exclude diabetes, an overnight polysomnogram (PSG), and blood sampling every 20 min over a 24-h period to assess adiponectin and leptin concentrations.  The HOMA-IR was calculated from fasting glucose and insulin values.  OSA severity was quantified by the Apnea-Hypopnea Index (AHI).  All analyses were controlled for BMI, and results are shown as mean ± SE.

Both PCOS and controls were obese (BMI: 40.7±7.0 vs. 44.9±9.7 kg/m2, respectively, p=0.167).  PCOS women were younger than non-PCOS women (26.6±5.9 vs. 32.9±6.5 yr, p=0.008).  Twenty-eight percent of PCOS women versus 36% of control women had OSA (AHI >5 events/hour), and 33% of PCOS women versus 14% of control women had prediabetes (IFG and/or IGT).  Mean 24-h adiponectin levels were lower in PCOS than non-PCOS women (PCOS OSAneg: 5.5; PCOS OSApos: 4.3; Control OSAneg: 8.4; Control OSApos: 5.6 ug/mL; p=0.005) and were inversely related to severity of OSA (p=0.019).  In multiple regression analyses controlling for BMI and AHI, the expected inverse relationship between adiponectin and leptin was present in control women (p=0.003), but not in PCOS women (p=0.893).  Similarly, the expected inverse relationship between adiponectin levels and HOMA-IR was present in control women (p=0.017) but not in PCOS women (p=0.491).  

CONCLUSIONS: Women with PCOS have lower adiponectin levels than non-PCOS women and this difference is exacerbated by the presence and severity of OSA.  Contrary to control women, adiponectin levels in PCOS women are not related to leptin levels and do not correlate with fasting insulin resistance.  Low adiponectin levels in PCOS women, particularly those with OSA, may play a role in their increased risk of cardiometabolic dysfunction.

Nothing to Disclose: KAT, RL, HW, BM, EV, DAE

*Please take note of The Endocrine Society's News Embargo Policy at

Sources of Research Support: This work was supported by NIH grants P50-HD057796, R01 HL075079, P60-DK020595, UL1-TR000430, the Blum-Kovler Family Foundation and the ResMed Foundation.