Abstract:
Objective To classify subtypes among middle-aged and elderly type 2 diabetes mellitus (T2DM) patients aged between 35‒74 years in Shanghai suburbs, to compare their characteristics and analyze incidence risk for cerebrovascular disease among these subtypes, so as to promote personalized and precise treatment of T2DM.
Methods A total of 7 792 patients with T2DM who completed a baseline survey from 2016 and 2019 were selected as the research subjects, based on the data from a natural population cohort and biobank in Shanghai suburbs. Patients were stratified by gender and clustered into subtypes using k-means method based on baseline parameters including the age at T2DM diagnosis, body mass index (BMI), fasting blood glucose, and triglyceride to high-density lipoprotein cholesterol ratio (TG/HDL-C). Patients were followed up until March 31, 2023. Multivariate Cox regression models were used to analyze the association between subtypes and incidence risk for cerebrovascular disease, and those with cerebrovascular disease within 1 year of follow-up survey were excluded from sensitivity analysis.
Results Among the 7 792 patients with T2DM, 3 615 were males and 4 177 were females. Stratified by gender, 4 subgroups were identified through k-means clustering analysis, namely poor blood glucose control subgroup, severe insulin-resistant subgroup, younger onset subgroup, and older onset subgroup. The median follow-up time was 4.30 years, during which 1 960 cerebrovascular disease events were observed (844 in males, 1 116 in females). After adjusting for smoking, alcohol consumption, weekly exercise, family history of diabetes mellitus, and duration of diabetes mellitus, among male patients, the incidence risk for cerebrovascular disease was lower in the younger onset subgroup (HR=0.59, 95%CI: 0.48‒0.73, P<0.001), poor blood glucose control subgroup (HR=0.81, 95%CI: 0.65‒1.00, P=0.046), and severe insulin-resistant subgroup (HR=0.61, 95%CI: 0.50‒0.75, P<0.001), compared to the older onset subgroup. While among female patients, the incidence risk for cerebrovascular disease was also lower in the younger onset subgroup (HR=0.68, 95%CI: 0.57‒0.80, P<0.001), poor blood glucose control subgroup (HR=0.73, 95%CI: 0.60‒0.89, P=0.002), and severe insulin-resistant subgroup (HR=0.72, 95%CI: 0.61‒0.85, P<0.001), compared to the older onset subgroup. Results of the sensitivity analysis were consistent with the main findings.
Conclusion Among middle-aged and elderly T2DM patients in suburban Shanghai, both male and female patients have the highest incidence risk for cerebrovascular disease in the older onset subgroup. Subtyping of T2DM patients can help to identify the high-risk populations of cerebrovascular disease.