Objective: Shortening of telomere length has been reported in several conditions including Type 2 diabetes and atherosclerosis. The aims of this study were (1) to assess whether telomere shortening occurs at the stage of pre-diabetes, i.e., impaired glucose tolerance (IGT) and (2) whether telomere shortening was greater in Type 2 diabetic subjects with atherosclerotic plaques. Methods: Subjects with impaired glucose tolerance (IGT) (n = 30), non-diabetic control subjects (n = 30), Type 2 diabetic patients without (n = 30) and with atherosclerotic plaques (n = 30) were selected from the Chennai Urban Rural Epidemiology Study (CURES), an ongoing epidemiological population-based study. Southern-blot analysis was used to determine mean terminal restriction fragment (TRF) length, a measure of average telomere size, in leukocyte DNA. Levels of thiobarbituric acid reactive substances (TBARS), protein carbonyl content (PCO) and high sensitive C-reactive protein (hs-CRP) were measured by standard methodologies. Carotid intima-media thickness (IMT) was assessed by high resolution B-mode ultrasonography. Results: The mean (+/- S.E.) TRF lengths were significantly lower in IGT subjects (6.97 +/- 0.3 kb; p = 0.002) and lower still in Type 2 diabetic subjects without plaques (6.21 +/- 0.2; p = 0.0001) and lowest in Type 2 diabetic subjects with atherosclerotic plaques (5.39 +/- 0.2; p = 0.0001) when compared to control subjects (8.7 +/- 0.5). In IGT subjects, TRF length was positively correlated to HDL cholesterol and negatively correlated to glycated hemoglobin (HbA(1c)), TBARS, PCO, HOMA-IR and IMT. In multiple linear regression analysis, presence of diabetes, HDL cholesterol and increased TBARS levels appear as significant determinants of telomere shortening. Conclusion: Telomere shortening is seen even at the stage of IGT. Among subjects with Type 2 diabetes, those with atherosclerotic plaques had greater shortening of telomere length compared to those without plaques. (c) 2006 Elsevier Ireland Ltd. All rights reserved.