J Shen¹, K Reesink¹, SJPM Eussen^1,2, A Koster², C van der Kallen^1,3, BE de Galan^1,3,4 , CG Schalkwijk¹, MMJ van Greevenbroek¹

¹CARIM, Cardiovascular Research Institute Maastricht , Maastricht University, Maastricht, the Netherlands. ²CAPHRI, Care and Public Health Research Institute, Maastricht University, Maastricht, the Netherlands. ³Department of Internal Medicine, Maastricht University Medical Center+ (MUMC+), Maastricht, the Netherlands. ⁴Department of internal medicine, Radboud University Medical Center, Nijmegen, the Netherlands.

E-mail: jie.shen@maastrichtuniversity.nl

Background: Arterial stiffness (AS) has been associated with insulin resistance (IR), yet few to no studies have related AS to IR in specific tissues. Also, knowledge on the association of AS with incident diabetes is limited and not completely elucidated. Therefore, we evaluated the relationship of AS with IR in liver, muscle and adipose tissue as well as with incident diabetes.

Methods: In 7277 population-based participants (59.8±8.7 years, 50.5% men, 23.0% type 2 diabetes, oversampled), we performed an oral glucose tolerance test and derived measures of hepatic insulin resistance index (HIRI), muscle insulin sensitivity index (MISI), adipocyte insulin resistance index (ATIRI) for tissue-specific IR respectively, and the Matsuda index for whole-body IR. AS was measured as carotid-femoral pulse wave velocity (cf-PWV). Cross-sectional associations between cf-PWV and measures of IR were assessed with multiple linear regression. The longitudinal association between baseline cf-PWV and incident diabetes, through 10-year follow-up questionnaire data, was evaluated by Cox regression. All main variables were standardized, after ln-transformation if skewed. The analyses were adjusted for potential confounders including demographics, physical activity, diet, body mass index, heart rate, and mean arterial pressure.

Results: In the adjusted models, cf-PWV was cross-sectionally associated with worse whole-body IR, indicated by lower Matsuda index [standardized β = -0.07; 95% CI: -0.11,-0.02] , but not with tissue-specific IR (HIRI [0.04; -0.01,0.09], MISI [-0.03; -0.08,0.03] or ATIRI [0.01; -0.04,0.06]). During a median follow-up of 9.1 [7.1, 10.1] years, 179 participants developed diabetes. After adjustment, cf-PWV was not significantly associated with new-onset diabetes [standardized Hazard Ratio=1.03, 95%CI: 0.86,1.24].

Conclusion: In this middle-aged population, aortic stiffness, as reflected by cf-PWV, was related to whole-body IR, but not significantly to either tissue-specific IR or to the development of type 2 diabetes during a median 9.1 years of follow-up.