Associations of fasting plasma glucose with all-cause mortality and cardiovascular events in older Chinese diabetes patients: A population-based cohort study

Abstract

Aims/introduction: Our aim was to investigate the optimal fasting glucose (FPG) range in Chinese older adults with type 2 diabetes, and to clarify whether the optimal range varies according to the control of risk factors.

Materials and methods: The baseline survey for the cohort study began in 2018, with follow up ending in 2022. Our study enrolled 59,030 older diabetes patients with no history of cardiovascular disease (CVD). Participants were divided into nine groups based on their baseline glycemic status. The association between FPG and the risk of adverse outcomes was mainly estimated by multivariate Cox proportional risk models and restricted spline analysis.

Results: During the 4-year follow-up period, a total of 5,637 deaths and 4,904 CVD events occurred. The associations of FPG with mortality and CVD events showed J-shaped curves. Among all-cause deaths, the hazard ratios for FPG ≤4.50 and >11.50 mmol/L were 1.50 (95% confidence interval [CI] 1.31-1.71) and 1.84 (95% CI 1.67-2.02). Among CVD, the hazard ratios for FPG ≤4.50 and >11.50 mmol/L were 1.31 (95% CI 1.13-1.53) and 1.71 (95% CI 1.54-1.89), respectively. The optimal FPG ranges of all-cause mortality and CVD were 5.50-7.50 and 4.50-7.50 mmol/L, respectively. For participants with at least two risk factors, the optimal FPG levels were higher than those with fewer risk factors.

Conclusions: In older Chinese diabetes patients, the FPG ranges related to the minimum death and CVD event rates were 5.50-7.50 and 4.50-7.50 mmol/L, respectively. Patients with more cardiovascular risk factors had higher optimal blood glucose ranges than those with fewer risk factors.

Keywords: Cardiovascular disease; Diabetes patient; Fasting plasma glucose.

Figure 1

Hazard ratios (HRs) and 95% confidence intervals (CIs) for all‐cause mortality and cardiovascular events according to fasting glucose levels at baseline. The solid line represents HR, the shaded area represents 95% CI and the reference value represents the lowest risk. There was a J‐shaped relationship between fasting glucose and the risk of all‐cause death (a) and cardiovascular events (b), with the lowest risk points being 7.27 and 6.94 mmol/L, respectively.

Figure 2

Hazard ratios (HRs) and 95% confidence intervals (CIs) for all‐cause mortality and cardiovascular events according to fasting glucose levels at baseline stratified by uncontrolled risk factors (≥2, <2). The solid line represents HR, the shaded area represents 95% CI and the reference value represents the lowest risk. (a–d) The associated lowest risk points were 7.14, 7.40, 7.03 and 7.43 mmol/L, respectively.