Supplementary Material for: Pre-stroke long-term fasting blood glucose level and glucose variability with post-stroke all-cause mortality in non-diabetic ischemic stroke patients from Northwestern China

Introduction: The influence of pre-stroke long-term fasting blood glucose (FBG) and its variability on post-stroke all-cause mortality in ischemic stroke (IS) patients without diabetes remains unclear. This study aimed to investigate their independent and joint effects on all-cause mortality in this specific population. Methods: We included 193,705 non-diabetic IS patients from the Northwest China Real-world and Population-based Tianshan Cohort 2019 to 2023. Cox proportional hazards regression models assessed the associations of impaired fasting glucose (IFG; 6.1-6.9 mmol/L) and high FBG variability with post-stroke all-cause mortality risk. Additive and multiplicative interactions between the two exposures were evaluated using the delta method and likelihood ratio tests, respectively. Results: 14,481 patients (7.5%) died during a median follow-up of 0.87 years. IFG showed no significant association with mortality risk. Higher FBG variability (LogVIM Q5 vs Q1) significantly raised mortality risk (HR 1.56, 95% CI: 1.48-1.64; Ptrend < 0.001). A significant interaction between IFG and high FBG variability was observed on both additive (RERI 0.30 95% CI 0.13-0.49, P = 0.006) and multiplicative scales (HR 1.34, 95% CI 1.13-1.58; P = 0.001). Conclusions: Long-term pre-stroke FBG variability is an independent predictor of post-stroke all-cause mortality among non-diabetic IS patients, while IFG is not. However, IFG amplifies the effect of FBG variability. Their combination shows synergistic risk elevation.