Abstract
OBJECTIVE: To evaluate whether the control status of type 2 diabetes mellitus (DM) influences the progression of Parkinson's disease (PD). METHODS: We conducted a prospective cohort study from March 2009 to August 2020. Patients at baseline were categorized into DM and non-DM groups, and those with DM were further classified into the well and poorly controlled DM groups based on the 7.0% of glycated hemoglobin (HbA1C) levels. Multivariate Cox proportional hazards regression models were used to explore the predictors for PD-related outcomes by hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: Of the 379 patients enrolled, 49 (12.9%) had DM, and 22 of DM (44.9%) were poorly controlled. The adjusted HRs were 2.060 (95% CI 1.165-3.641) for United Rating Scale (UPDRS) III score increased ≥14 in the poorly controlled-DM group, and 1.066 (95% CI 0.572-1.986) in the well-controlled DM group, relative to the non-DM group (p trend = 0.025), after adjusting for sex, age, age of onset, body mass index, and UPDRS III and Montreal Cognitive Assessment (MoCA) scores at baseline. The adjusted HRs were 2.079 (95% CI 1.212-3.566) for reaching Hoehn & Yahr stage ≥3 in the poorly controlled DM group, and 0.879 (95% CI 0.413-1.871) in the well-controlled DM group, compared with the non-DM group (p trend = 0.021). Time to death or time to MoCA 3-point decrease were not significantly different among the three groups. INTERPRETATION: Poorly controlled DM is an independent risk factor contributing to motor progression in PD. Our study highlights the importance of adequate control of diabetes in PD.