Comparison of changes in fasting glucose among adolescents with Polycystic Ovary Syndrome to obese controls; a retrospective cohort study

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SAT 596-621-Pediatric Endocrinology /Steroids and Puberty
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-611
Asma Javed*, Patricia Simmons, Aida Nadim Lteif, Seema Kumar and Alice Yoonju Chang
Mayo Clinic, Rochester, MN

Polycystic Ovary Syndrome (PCOS) is a condition of androgen excess associated with impairment in insulin sensitivity and can be diagnosed as early as adolescence. Adolescents with PCOS have a higher prevalence of glucose intolerance when compared to healthy adolescent controls in cross sectional studies. Although higher risk for developing impaired fasting glucose (IFG) is established in obese adolescents, whether PCOS is associated with higher risk for IFG compared to obese adolescents is not known. We sought to evaluate whether there is a greater increase in fasting glucose over time in adolescents with PCOS compared to overweight/obese adolescent girls (OA) without PCOS. METHODS:

We conducted a retrospective cohort study of 310 adolescents with PCOS compared to 250 obese adolescents without PCOS. All adolescents were followed at an academic medical center, Mayo Clinic Rochester, MN, from 1996 to 2012. Participants included for this analysis were 248 adolescent girls, ages 13 to 18 years, (98=PCOS, 150 OA) who had two or more measurements of fasting glucose separated by at least a 6 month interval. We excluded adolescents with IFG or diabetes at diagnosis of PCOS or obesity. Almost exclusively, PCOS was diagnosed by a single pediatric gynecology specialist using Rotterdam criteria.  Adolescent girls were diagnosed with overweight/obesity by CDC criteria based on age specific BMI percentiles. Multivariate linear regression was used to assess predictors for changes in fasting glucose. RESULTS:

At diagnosis, adolescents with PCOS had significantly lower BMI (kg/m2) (PCOS: 33.2 +/- 0.76 vs OA: 37.4 +/- 0.6 p <0.001) and older age (PCOS: 15.65 years +/- 0.15 vs OA: 14.87 +/- 0.13 p value <0.001) . Adolescents with PCOS had shorter length of follow up (PCOS: 3.66 years +/- 0.33 vs OA: 4.34 +/- 0.27 p=0.02). The baseline fasting glucose (mg/dl) was not significantly different between groups (PCOS: 88 +/- 0.67 vs OA:  88.3 +/- 0.54 p = 0.74).

Mean change in fasting glucose (mg/dl) from time of diagnosis to last follow-up was 2.9 ± 1.2 mg/dl for PCOS and 4.6 ± 1 for OA (p=0.29). In the multivariate linear regression model, significant predictors for change in fasting glucose were age (p<0.01), BMI (p <0.001) and fasting glucose (p <0.001) at time of diagnosis. Within the PCOS group, BMI was a significant predictor for development of IFG (p = 0.03). There was no significant difference in the incidence of IFG (PCOS 16.3% v. OA 20%, p=0.46) over a median of 4.1 years of follow-up. Median time to diagnosis of IFG (years) was 1.25 (IQ range 0.65-6.06) versus 4.75 for controls (IQ  range 1.11-5.85 p = 0.27). CONCLUSIONS:

Adolescent girls with PCOS do not have a significantly different change in fasting glucose compared to obese adolescents despite having a lower BMI. Among adolescents with PCOS, greater BMI increases the risk for developing IFG over time.

Nothing to Disclose: AJ, PS, ANL, SK, AYC

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