Data Management And Sharing System of China National Clinical Research Center for Neurological Diseases

Journal：Cardiovascular Diabetology

Publication type: Article

IF(IF): 9.3

PMID: 37633905 Link To Pubmed Full text Download

Stroke is one of the major causes of death and disability in the world[1], and imposes a heavy economic burden globally[2]. Ischemic stroke accounts for approximately 80% of all strokes and is associated with many risk factors, such as hyperlipidemia[3], diabetes[4], hypertension[5] and insulin resistance[6]. Insulin resistance, which occurs commonly in type 2 diabetes (T2D) and also nearly 50% stroke patients without diabetes[7], is considered as a new independent predictor of ischemic stroke[8], and also related with higher rates of disability in stroke patients[9]. One randomized controlled trial (RCT) showed that pioglitazone, which improves insulin sensitivity, could decrease the risk of stroke or myocardial infarction in nondiabetic patients with a recent transient ischemic attack (TIA) or ischemic stroke[10]. In addition, some studies showed that insulin resistance is also related to some independent risk factors of poor stroke outcomes such as hypertension[11], hypertriglyceridemia and high-density lipoprotein (HDL)[12]. The interaction of insulin resistance and these factors presents difficulties when evaluating the influence of insulin resistance on stroke outcomes. In one mathematical analysis, intervention of insulin resistance can prevent approximately 42% cardiovascular disease, and additional therapy such as hypotensive and antidiabetic drugs can further decrease the risk[13]. Currently, hyperinsulinemic-euglycemic clamp is the golden standard test measuring insulin resistance. However, it is costly, invasive and unavailable in the real-world practice[14]. Therefore, some noninvasive methods were investigated to measure the insulin resistance, such as homeostasis model assessment (HOMA)[15], triglyceride glucose index (TyG index)[16] and estimated glucose disposal rate (eGDR)[17]. Estimated glucose disposal rate was originally developed as a validated score to measure the insulin resistance in type 1 diabetes (T1D) based on the waist circumference (WC), hypertension and glycated hemoglobin A (HbA1c)[8]. Recently, some studies found that a level of eGDR is associated with a decreased risk of stroke, all-cause mortality and cardiovascular disease in patients with T1D[18] and T2D[8]. However, the relationship between eGDR and stroke outcomes in ischemic stroke patients is unknown. In this study, we aimed to investigate the associations between insulin resistance, evaluated by eGDR formula, and the stroke recurrence, disability, all-cause mortality and cardiovascular event in patients with acute first-ever ischemic stroke.

The study cohort is described using standard descriptive statistics in terms of averages with standard deviation for continuous variables and counts with percentages for categorical variables. The incidence rates of cardiovascular events, stroke recurrence and functional outcome were presented with 95% exact Poisson confidence intervals (CI). The association between eGDR (mg/kg/min) was categorised as previously used cut-off levels: < 4, 4–5.99, 6–7.99, and ≥ 8 mg/kg/min. Cardiovascular events and stroke recurrence were illustrated using Kaplan–Meier curves. Primary and secondary outcomes were analytically assessed using three Cox regression models; model 1: unadjusted; model 2: adjusted for age and sex; model 3: adjusted for age, sex, and other related variables. Given that HbA1c, WT and presence of hypertension are part of eGDR calculation, adjustments for these variables were not made when analysing the association between eGDR and outcomes. A similar methodology has been used in the past for the analysis of eGDR. In addition, to identify a subgroup being at even higher risk of cardiovascular events and mortality within our study population, all models were stratified according to sex, age, diabetes and thrombolysis.