Salivary telomere length and the risks of prediabetes and diabetes among middle-aged and older adults: findings from the Health and Retirement Study

Aim To assess the association of telomere length (TL) with prediabetes/diabetes and to explore the potential factors affecting TL among individuals with prediabetes/diabetes by weight status. Methods This study included 3,379 eligible adults (aged 45-85 years, males: 42%) from the US Health and Retirement Study in 2008. TL was assayed using quantitative PCR of saliva (T/Sratio). Linear and nonlinear associations between TL and prediabetes/diabetes were assessed using the logistic regression and restricted cubic spline model, respectively, adjusting for TL-plate numbers, age, sex, race, body mass index, lifestyles, diabetes medications, and cardiometabolic parameters (blood pressure, C-reactive protein, and total cholesterol). Multiple linear regression was used for testing any factors associated with TL. Results Among 3,379 participants, 868 (25.7%) had prediabetes with a mean TL of 1.34 +/- 0.37 (T/S ratio) and 858 (25.4%) had diabetes with a mean TL of 1.36 +/- 0.43 (T/S ratio). Neither linear nor nonlinear association of TL with prediabetes/diabetes was significant by weight status. Age was negatively associated with TL in both normal-weight (beta = - 0.002, p = 0.025) and overweight/obese (beta = - 0.002, p = 0.006) prediabetes, but non-significant in normal-weight and overweight/obese diabetes. BMI and cardiometabolic parameters were not associated with TL in prediabetes/diabetes by weight status. Conclusions Salivary TL was not associated with prediabetes/diabetes among the US middle-aged and older adults. Further longitudinal studies are required to establish the link between TL and diabetes development and to identify potential factors affecting TL shortening, particularly in normal-weight diabetic patients.