Abstract
OBJECTIVES: Evidence on the association between fasting blood glucose and mortality in non-diabetic patients who had a stroke is limited. We aimed to investigate the association of baseline fasting plasma glucose (FPG) with 1 year all-cause mortality in non-diabetic patients with acute cerebral infarction (ACI). DESIGN: A multicentre prospective cohort study. SETTING: Four grade A tertiary hospitals in the Xi'an district of China. PARTICIPANTS: A total of 1496 non-diabetic patients within 7 days of ACI were included. MAIN OUTCOME MEASURES: The outcome was 1 year all-cause mortality. Baseline FPG was analysed as a continuous variable and was divided into four quartiles (group Q1-group Q4). We used multivariable Cox regression analyses, curve fitting and Kaplan-Meier (K-M) analyses to explore the association of baseline FPG with 1 year all-cause mortality in non-diabetic patients with ACI. RESULTS: After controlling for confounders, multivariable Cox regression analyses indicated a 17% increase in 1 year all-cause mortality for every 1 mmol/L of baseline FPG increase (HR=1.17, 95% CI 1.02 to 1.35, p=0.030). Patients from the Q4 group had 2.08 times increased hazard of 1 year all-cause mortality compared with the Q1 group (HR=2.08, 95% CI 1.13 to 3.82, p=0.019), while the survival rate of patients in group Q4 was decreased compared with that in other groups (p<0.001). The curve fitting revealed a positive but non-linear association of baseline FPG with 1-year all-cause mortality in non-diabetic patients with ACI. CONCLUSION: In non-diabetic patients with ACI, elevated baseline FPG is an independent risk factor for 1-year all-cause mortality, and the two are positively and non-linearly associated. These results suggest that high FPG should be seen as a concern in non-diabetic patients with ACI.