Body fat patterning in lean Asian Indians with diabetes: Case-control study

To perform body fat patterning in Asian-Indian individuals with T2D.

A total of 53 patients with recent-onset diabetes and 106 non-diabetic controls were included from screened 261 individuals. Data was divided into 2 groups; overweight/obese [(BMI ≥23 kg/m²); 45 diabetic, 84 non-diabetic] and lean [(BMI <23 kg/m²); 8 diabetic, 22 non-diabetic]. Anthropometry (weight, height, BMI, waist, hip circumference, waist-hip ratio) and lipids, adiponectin and hsCRP were measured. Body composition (BC) was assessed by bioimpedance analysis (BIA) and Dual Energy X-ray absorptiometry (DEXA). We analyzed the association of visceral adipose tissue (VAT) with anthropometric measures to identify predictors of diabetes.

Total body fat percentage was comparable between patients with T2D and non-diabetic controls in both, obese [35.0 ± 9.1% vs 36.8 ± 8.4%, p = 0.29 (BIA), 40.1 ± 6.7 vs 46.6 ± 4.1%, p = 0.052 (DEXA) and lean [25.1 ± 5.6% vs 26.0 ± 6.7%, p = 0.74 (BIA), 35.3 ± 4.8% vs 34.1 ± 6.3%, p = 0.72 (DEXA) study group. Individuals of T2D (obese or lean) had significantly higher visceral fat rating (BIA), VAT area, volume, mass and VAT corrected for total body fat percentage (DEXA). Obese T2D had lower muscle mass (57.0 ± 6.4% vs 60.0 ± 5.5%, p = 0.03) than obese controls. Intra-abdominal visceral fat (IAVF) [(VFR, VAT (mass/area/volume) and VAT mass corrected for body fat)] had the best sensitivity (71%) for incident diabetes.

Higher Intra-abdominal visceral fat and not total body fat is associated with incident diabetes independent of BMI. IAVF estimation by either BIA or DEXA should be performed to predict diabetes especially in lean individuals.