Fasting Glucose Changes in Adolescents with Polycystic Ovary Syndrome Compared with Obese Controls: A Retrospective Cohort Study

Abstract

Objective To compare changes in fasting glucose among adolescents with polycystic ovary syndrome (PCOS) to obese adolescents without PCOS. Methods Retrospective cohort study of 310 adolescents with PCOS and 250 obese adolescents, (ages 13–18 years) seen at Mayo Clinic, Rochester MN, from 1996–2012. Included for analysis were 98 adolescents with PCOS and 150 obese adolescents who had 2 or more fasting glucose measurements separated by at least 6 months. Adolescents with impaired fasting glucose or diabetes were excluded. Multivariate models were used to assess predictors of change in fasting glucose. Results At diagnosis, adolescents with PCOS had lower body mass index (BMI) (kg/m2) and older age than obese adolescents (P<.001). Adolescents with PCOS had shorter follow up (years) (P=.02). Baseline fasting glucose (mg/dl) was not different between groups. Mean change in fasting glucose (mg/dl/year) was 2.4± 9.4 mg/dl/year for PCOS and 2.2±6.2 mg/dl/year for obese adolescents (P=.83). Significant predictors for change in fasting glucose were BMI and fasting glucose at diagnosis (P<.01). Within the PCOS cohort, BMI was a significant predictor for development of IFG (P=.003). Prevalence of hypertension (HTN) increased in the PCOS cohort from baseline to follow up (P=.02). PCOS and BMI were significantly associated with development of HTN in the entire cohort. Conclusions Adolescent girls with PCOS do not show a significant change in fasting glucose or an increased risk for the development of IFG compared to obese adolescents. BMI, not PCOS status, was the strongest predictor for changes in fasting glucose and development of IFG over time.

Natural history of polycystic ovary syndrome: A systematic review of cardiometabolic outcomes from longitudinal cohort studies

“…One study (Table 3b) showed that conversion rates from normal glucose tolerance (NGT) to IGT were not significantly different between women with PCOS and controls 32 . In another study, women with PCOS had a significantly higher prevalence of IGT than those without PCOS at baseline, but the prevalence was similar at follow up in both groups 44 . Of the studies that compared the change in glycaemic status based on oral glucose tolerance test (OGTT) within the PCOS group, two studies 32,47 indicated a significant increase in the 2‐h glucose levels from baseline, but one study 40 showed no significant change in the 2‐h glucose levels.…”

“…Characteristics of studies included in this review are outlined in Table 1. Seven of the studies were conducted in the United States, 27,36,39–41,44,56 six in Italy, 29–31,38,48,49 three studies in Iran, 26,45,50 three in Sweden 37,51,52 and each of the remaining 12 studies were conducted in Australia, 53 the Netherlands, 28 Denmark, 54 Turkey, 32 Bosnia, 55 Thailand, 46 Venezuela, 43 Spain and Germany, 35 Croatia, 34 Poland, 42 Korea 33 and Hong Kong 47 …”

“…Seventeen studies reported change in BMI, 28–32,34,35,37,39,40,42,47,49,50,52–54 10 studies reported changes in lipid profiles, 26,29,34,40,42,44,47–50,52,54,56 10 studies described HTN, 26,40,42,44,47,48,50,52,54,56 T2DM risk was reported by 11 studies 27,32,33,36,38,45–47,50,55,56 and 15 studies reported IR and/or IGT 28,29,31,32,34,36,40,42–44,47,48,50,52,54 . The majority of investigators described more than two outcomes, which explains the overlap in a number of studies.…”

“…Only three studies were conducted among young girls and adolescents who were aged between 1 year 53 and 15 years 35,44 from the start of follow up, three other studies 37,51,52 reported outcomes in women aged 49 years at baseline and the remaining studies were conducted in participants whose age ranged from 21.8 to 38 years. The mean follow‐up period ranged between 2 and 32 years.…”

“…Three studies 40,44,48 demonstrated that changes in fasting glucose levels are similar between women with and those without PCOS. Within the PCOS group, two studies 28,34 showed no significant difference in fasting glucose, while five other studies 29,32,36,43,47 demonstrated a significant increase.…”