OBJECTIVE: We sought to examine the association between plasma concentrations of liver enzymes gamma-glutamyltransferase (GGT) and alanine transaminase (ALT) and incident diabetes, prospectively. RESEARCH DESIGN AND METHODS: We conducted a case-cohort analysis of data from participants mainly aged 35-65 years in the European Prospective Investigation into Cancer and Nutrition-Potsdam Study. The analytic sample included 787 participants with incident diabetes and 2,224 participants without diabetes. RESULTS: Concentrations of GGT and ALT were significantly associated with incident diabetes after extensive adjustment. Compared with participants in the lowest quintile of GGT, the adjusted hazard ratios for increasing quintiles were 1.13 (95% CI 0.66-1.93), 1.67 (1.01-2.77), 2.77 (1.71-4.49), and 2.67 (1.63-4.37), respectively (P for linear trend <0.001). Compared with participants in the lowest quintile of ALT, the adjusted hazard ratios for incident diabetes were 0.93 (0.56-1.53) for quintile 2, 1.28 (0.83-1.96) for quintile 3, 1.35 (0.88-2.07) for quintile 4, and 1.93 (1.27-2.92) for quintile 5 (P for linear trend = 0.002). The magnitude of the associations were higher among men than women for GGT (P = 0.004) but did not differ significantly between men and women for ALT (P = 0.307). CONCLUSIONS: Concentrations of GGT and ALT were significant predictors of incident diabetes in this study, even at concentrations still considered to be within the normal range.
OBJECTIVE: We sought to examine the association between plasma concentrations of liver enzymes gamma-glutamyltransferase (GGT) and alanine transaminase (ALT) and incident diabetes, prospectively. RESEARCH DESIGN AND METHODS: We conducted a case-cohort analysis of data from participants mainly aged 35-65 years in the European Prospective Investigation into Cancer and Nutrition-Potsdam Study. The analytic sample included 787 participants with incident diabetes and 2,224 participants without diabetes. RESULTS: Concentrations of GGT and ALT were significantly associated with incident diabetes after extensive adjustment. Compared with participants in the lowest quintile of GGT, the adjusted hazard ratios for increasing quintiles were 1.13 (95% CI 0.66-1.93), 1.67 (1.01-2.77), 2.77 (1.71-4.49), and 2.67 (1.63-4.37), respectively (P for linear trend <0.001). Compared with participants in the lowest quintile of ALT, the adjusted hazard ratios for incident diabetes were 0.93 (0.56-1.53) for quintile 2, 1.28 (0.83-1.96) for quintile 3, 1.35 (0.88-2.07) for quintile 4, and 1.93 (1.27-2.92) for quintile 5 (P for linear trend = 0.002). The magnitude of the associations were higher among men than women for GGT (P = 0.004) but did not differ significantly between men and women for ALT (P = 0.307). CONCLUSIONS: Concentrations of GGT and ALT were significant predictors of incident diabetes in this study, even at concentrations still considered to be within the normal range.
Authors: Cris A Slentz; Lori A Bateman; Leslie H Willis; A Tamlyn Shields; Charles J Tanner; Lucy W Piner; Victoria H Hawk; Michael J Muehlbauer; Greg P Samsa; Rendon C Nelson; Kim M Huffman; Connie W Bales; Joseph A Houmard; William E Kraus Journal: Am J Physiol Endocrinol Metab Date: 2011-08-16 Impact factor: 4.310