Type 2 diabetes in non-obese Indian subjects is associated with reduced leptin levels: study from Mumbai, Western India

• Published in: Molecular and cellular biochemistry 2005-07, Vol.275 (1-2), p.143-151
• Main Authors: Marita, A Rosalind, Sarkar, Joy A, Rane, Sujata
• Format: Article
• Language: English
• Subjects: Adult; Aged; Blood Glucose - metabolism; Body Composition - physiology; Body mass index; C-Peptide - blood; Cohort Studies; Cross-Sectional Studies; Diabetes; Diabetes Mellitus, Type 2 - blood; Female; Gender; Glucose Tolerance Test; Humans; India; Insulin; Insulin - blood; Leptin - blood; Male; Middle Aged; Population Surveillance; Proinsulin - blood; Regression Analysis; Sex Factors; Urban Population; Women
• ISSN: 0300-8177; 1573-4919
• DOI: 10.1007/s11010-005-1204-7

Asian Indian subjects have a high tendency to develop Type 2 diabetes even though obesity is relatively uncommon. We evaluated the serum leptin levels in a group of non-obese Type 2 diabetic patients from Mumbai, Western India. Cross sectional study. A total of 104 subjects consisting of 28 with Type 2 diabetes, 16 with impaired glucose tolerance and 60 age and sex-matched control subjects were given 75 g oral glucose tolerance test. Fasting serum leptin (IRMA), insulin and C-peptide were measured along with fasting and 2 h plasma glucose. The relation between these variables was studied by univariate and multiple regression analysis. Type 2 diabetes was associated with marked (50-60%) reduction in serum leptin levels, in both men and women. Women, but not men, with impaired glucose tolerance exhibited 60% lower leptin. Serum leptin levels were positively correlated to body mass index (BMI; r = 0.501, p = 0.001) and calculated body fat percent (r = 0.525, p = 0.001) in all the study subjects with a better correlation in the normal subjects (r = 0.562 for BMI and 0.735 for body fat). On the other hand, serum leptin showed significant correlation to serum insulin (r = 0.362, p = 0.008) only in subjects with diabetes or IGT. In the multiple regression model, BMI was the only independent predictor of leptin, in all the subjects. However, in subjects with diabetes or impaired glucose tolerance, waist circumference (p = 0.003), gender (p = 0.007) and body fat (p = 0.009) were significant predictors of leptin, besides BMI. Gender-specific multiple regression revealed serum insulin as an independent predictor of leptin in men (p = 0.026). Therefore, lower serum leptin levels in diabetes is partly due to increased waist circumference, decreased BMI and male sex. These observations are consistent with the view that leptin levels in this cohort of non-obese Indians from Mumbai exhibit gender-specific relationship partly attributed to changes in serum insulin and waist circumference in men and to changes in BMI, in women.