A multimarker panel for predicting short-term outcome after intravenous thrombolysis in acute ischemic stroke: a retrospective study
TL;DR
The combined assessment significantly improves predictive accuracy compared to individual markers alone, supporting its potential clinical utility for early risk stratification and individualized treatment decision-making and require prospective multicenter validation before routine clinical implementation can be recommended.
Abstract
Objective To evaluate the predictive efficacy of a multimarker panel combining the National Institutes of Health Stroke Scale (NIHSS) score immediately after intravenous thrombolysis (IVT) with the Systemic Immune-Inflammation Index (SII), baseline blood glucose (Glu), and plasma uric acid (UA) for short-term neurological prognosis in patients with acute ischemic stroke (AIS). Methods This single-center retrospective cohort study enrolled 132 consecutive AIS patients who received IVT at Chengdu Second People’s Hospital between January 2021 and December 2021. Clinical demographics, laboratory parameters, and NIHSS scores at admission (NIHSS1) and immediately after thrombolysis (NIHSS2) were collected. The modified Rankin Scale (mRS) at 90 days was used to classify patients into good prognosis (mRS 0-2) and poor prognosis (mRS 3–6) groups. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for poor prognosis. Receiver operating characteristic (ROC) curve analysis, calibration assessment, and decision curve analysis were conducted to evaluate and compare the predictive performance of individual and combined markers. Subgroup and sensitivity analyses were performed to validate model robustness. Results Among the 132 enrolled patients, 55 (41.67%) had poor prognosis at 90 days. Multivariate logistic regression analysis identified NIHSS2 score (OR = 1.317, 95%CI: 1.092–1.588, P = 0.004), baseline Glu (OR = 1.389, 95%CI: 1.083–1.781, P = 0.010), UA (OR = 1.008, 95%CI: 1.001–1.014, P = 0.028), and SII (OR = 1.001, 95%CI: 1.000–1.002, P = 0.039) as independent risk factors for poor prognosis. The combined model incorporating these four indicators demonstrated superior predictive performance with an area under the curve (AUC) of 0.928 (95%CI: 0.870–0.966, P < 0.001), sensitivity of 98.18%, specificity of 72.73%, and Youden index of 0.709. The combined model significantly outperformed individual predictors: UA (Z = 5.276, P < 0.001), SII (Z = 4.031, P < 0.001), baseline Glu (Z = 3.595, P < 0.001), and NIHSS2 alone (Z = 2.124, P = 0.034). Calibration assessment demonstrated good fit (Hosmer-Lemeshow χ2 = 6.847, P = 0.553), and decision curve analysis confirmed net clinical benefit across threshold probabilities of 10–80%. Subgroup analyses demonstrated consistent excellent performance across age, sex, stroke severity, and thrombolytic agent subgroups (AUC range: 0.91–0.95). Leave-one-out sensitivity analysis confirmed that each biomarker contributed unique prognostic information to the combined model. Conclusion The multimarker panel combining post-thrombolysis NIHSS score, SII, baseline Glu, and UA provides a simple, readily available, and highly effective approach for predicting short-term neurological prognosis in AIS patients after IVT. The combined assessment significantly improves predictive accuracy compared to individual markers alone, supporting its potential clinical utility for early risk stratification and individualized treatment decision-making. However, these findings require prospective multicenter validation before routine clinical implementation can be recommended.