Impaired autonomic function is a complication of type 2 diabetes mellitus (DM2), but may also be involved in its development. For this reason, this study looked at the association of autonomic function with the incidence of DM2 in a homogeneous Caucasian population.

This Hoorn study was a prospective population-based study of individuals aged 50–75 years. For the 631 participants, the standard deviation of all normal-to-normal intervals (SDNN) and eight other parameters of autonomic function were calculated at baseline. Fasting and 2-h glucose were measured during follow-up by oral glucose tolerance test (OGTT). DM2 at baseline and follow-up was ascertained by questionnaire and OGTT. After excluding participants with DM2 at baseline, the association of parameters of autonomic function with incident diabetes was examined using logistic-regression analysis while adjusting for possible confounders.

After excluding those with known (n=67) or newly diagnosed (n=126) DM2 at baseline and those missing follow-up data (n=140), 298 participants were eligible for the study (182 with normal glucose tolerance, 19 with impaired fasting glucose and 97 with impaired glucose tolerance). During a median follow-up of 9.2 (range 4.5–11.1) years, 94 incident cases of DM2 were observed. After adjusting for confounding variables, the DM2 odds ratio was 1.12 (95% CI: 0.77, 1.64) per SDNN increase. Results for other parameters of autonomic function were similar.

The present study found no evidence of an association between autonomic function and DM2 incidence in a population at high risk of diabetes. This implies that previously observed associations between autonomic function and glucose metabolism in cross-sectional settings may reflect reverse causation.