Combining Fasting Plasma Glucose with Gamma-glutamyl Transferase Improves the Sensitivity to Predict Incident Diabetes in Asian Indian Men with Impaired Glucose Tolerance.

Objective: To study the associations of baseline gamma-glutamyltransferase (GGT) and alanine transaminase (ALT) with incident diabetes among Asian Indian men with impaired glucose tolerance (IGT).

Methods: In a 2 year prospective, randomised, controlled primary prevention study of diabetes, among 537 IGT men aged 35-55 years, 123 incident diabetes (DM) cases occurred. Anthropometric {body mass index (BMI), waist circumference (WC)}, and laboratory measurements (fasting, 30 min and 2 hr plasma glucose (2 hr PG), HbA1c and plasma insulin, lipid profile, ALT, GGT) were estimated at baseline (Clinical Trial Identification No: NCT00819455). Predictive associations of baseline GGT and ALT values during the study were assessed using appropriate statistical methods.

Results: Baseline GGT but not ALT was significantly higher in incident diabetes cases. Mean (95% CI) GGT decreased in subjects who reverted to normal glucose tolerance (NGT), whereas it increased in subjects who deteriorated to diabetes (NGT:-3.5 (-6.4 to -0.6); IGT:0.3 (-3.0 to 2.4); DM:8.3 (3.6 to 13.0) UL(-1); P < 0.0001). The risk of DM significantly increased with increasing baseline GGT after adjusting for confounders such as BMI, alcohol drinking, 2 hr PG and insulin resistance (2.02[1.35-3.02]; P = 0.001). Receiver operating characteristic curve showed that the model comprising of baseline fasting plasma glucose (FPG) and GGT (area-under-curve(AUC)[95% CI]: 0.668 [0.613-0.722]; P < 0.0001) was equally sensitive in identifying subjects with risk of diabetes as compared to 2 hr PG (AUC [95% CI]: 0.670 [0.614-0.725]; P < 0.0001) and HbA1c (AUC [95% CI]: 0.677 [0.619-0.734]; P < 0.0001) alone.

Conclusions: GGT was an independent predictor of incident diabetes. Combination of GGT and FPG offers a simple and sensitive tool to identify subjects at high risk of developing diabetes.