Prevalence and Predictors of Prediabetes in Adolescents and Young Adults with Turner Syndrome: A Cross-Sectional Study from Eastern India.

BACKGROUND

Individuals with Turner syndrome (TS) have a high risk for prediabetes/type 2 Diabetes Mellitus (T2DM). There is scarce data regarding risk factors for prediabetes in TS, specially from South Asia.

METHODS

We conducted a cross-sectional study on girls with TS aged 12-30 years who had achieved pubertal stage B3 and above-spontaneously or with oestrogen. Anthropometric measurements and biochemical tests were conducted, and medical records were reviewed for details about pubertal onset and progression, growth hormone (GH) and oestrogen therapy.

RESULTS

Out of 129 patients with TS in our database, 99 met the criteria for inclusion, mean age 18.33+/-3.78 years and mean BMI 20.57+/- 3.71 kg/m. Prevalence of prediabetes was 23.23%. Plasma-glucose measured after 75 g-oral-anhydrous-glucose-load (OGTT-PPG) identified five additional prediabetes cases, who had normal fasting plasma glucose (FPG) or HbA1c%. Compared to those without prediabetes, TS with prediabetes (n = 23) had higher mean body weight, BMI, waist circumference (WC) [42.02+/- 5.83 vs 36.22+/-8.07, 22.77+/-2.78 vs 19.91+/- 3.72, 85.26+/- 3.52 vs 81.08+/- 4.59, pall < 0.03 ], higher median WC-to-height ratio (WHtR) and WC-to-hip ratio (WHR)((0.64 [0.6-0.69] vs 0.59[0.56- 0.66], 0.9[0.84-1.12] vs 0.85[0.75-1.01], p < 0.02), and higher LDL-cholesterol, triglycerides, and greater prevalence of hepatosteatosis (47.1% vs 21.1%, < 0.01). Among GH recipients (n = 36), those with prediabetes had delayed initiation and shorter duration of GH therapy. There were no differences in cardiometabolic parameters or the prevalence of diabetes between different karyotypic variants of TS. BMI, WC and WHR had significant positive correlation with FBG, OGTT-PPG and HbA1c% (p < 0.004). Delay in oestrogen initiation had a significant correlation with OGTT-PPG (Spearman's-rho = 0.69, < 0.004). BMI, WHR and pubertal status were independent predictors for prediabetes (OR: 1.27 [1.03-1.57]), 1.18 [1.04-1.34]) and 0.09[0.02-0.38], respectively, pall < 0.02), but karyotype was not. BMI had the highest sensitivity [cut-off: 21.04 kg/m (sensitivity: 82.6%, specificity: 62.2%) and WHR had the highest specificity [cut-off: 0.89 (sensitivity: 73.9%, specificity 78.4%)] for predicting prediabetes.

CONCLUSION

Indian girls with TS have a high risk for prediabetes, irrespective of underlying karyotype and should be screened with oral glucose challenge to identify prediabetes. Timely intervention against central obesity and early initiation of GH and oestrogen should be ensured in TS. Late presenting girls should be closely monitored for dysglycaemia before and during treatment with GH and/or oestrogen.