Non-insulin-based insulin resistance indices predict early neurological deterioration in elderly and middle-aged acute ischemic stroke patients in Northeast China.

Insulin resistance (IR) has a strong association with acute ischemic stroke (AIS) occurrence and poor prognosis of afflicted patients. However, the relation between early neurological deterioration (END) risk and IR in elderly and middle-aged patients remains to be thoroughly studied. Here, we investigated the relationship between four indicators of IR and the risk of END in middle-aged patients patients with AIS. The study retrospectively analyzed 1696 elderly and middle-aged patients having AIS between January 2019 and June 2023. Within 7 days of admission, the patients were then stratified relying upon alternations in the National Institutes of Health Stroke Scale. Subsequently, we employed logistic regression analyses for assessing each index correlation with END on the basis of the tertiles of TyG index (TyGI), triglyceride to high-density lipoprotein ratio (TG/HDL), TyG-BMI, alongside IR metabolic score (METS-IR). These four indicators were significantly heightened in the END group (n = 680) in comparison to the non-END group (n = 1016). When grouping using tertiles, the four aforementioned indicators emerged as independent risk factors for END occurrence, whether or not adjusted for confounding factors. The results revealed a progressive elevation in END occurrence risk with the rise in the tertile of each indicator. Finally, we utilized receiver operating characteristic (ROC) curves for assessing the indicators' predictive power. TyG-BMI, TyGI, TG/HDL, and METS-IRs' area under the curve (AUC) were, respectively, 0.736 (95% CI: 0.712-0.761; P < 0.001), 0. 694 (95% CI: 0.668-0.721; P < 0.001), 0.684 (95% CI: 0.658-0.711; P < 0.001), and 0.722 (95% CI: 0.697-0.747; P < 0.001). IR is associated with END risk in middle-aged AIS patients. TyG-BMI, TyGI, TG/HDL, and METS-IR are independent risk factors of END in elderly and middle-aged AIS patients. Simultaneously, these four IR indicators have significant predictive power for END.