Predictive value of preoperative and postoperative controlling nutritional status scores and their changes in older patients with large vessel occlusion stroke after endovascular thrombectomy.

Background: Over the past decade, endovascular thrombectomy (EVT) for acute large vessel occlusion (LVO) has advanced significantly. However, older patients who present with large-vessel occlusion and are treated with EVT tend to have poorer outcomes compared to younger individuals. This study aims to investigate whether preoperative and postoperative controlling nutritional status (CONUT) scores can predict clinical outcomes and their dynamic changes in older patients with large vessel occlusion stroke following EVT. Methods: A total of 441 patients with LVO who underwent EVT were enrolled from two stroke units. These patients were divided into a prospective training cohort and a prospective validation cohort. Based on the 90-day modified Rankin Scale (mRS) score, patients were categorized into either a favorable outcome group or an unfavorable outcome group. Univariate analysis and logistic regression were used to screen candidate predictors. A nomogram was developed to predict the risk of unfavorable outcomes. The predictive performance of the model was evaluated using receiver operating characteristic (ROC) curve analysis. Model calibration was assessed with the Hosmer-Lemeshow (H-L) test and calibration curves. Decision curve analysis (DCA) was applied to evaluate the clinical utility of the model. Results: Multivariable logistic regression analysis revealed that admission NIHSS (OR = 1.096; 95% CI: 1.031-1.166; Conclusion: The CONUT score-based nomogram model demonstrates comparable predictive performance for predicting adverse outcomes at 90 days in older patients with LVO undergoing EVT, whether applied preoperatively or postoperatively. This tool provides clinicians with an accurate and effective method for early prediction and timely treatment planning in this specific patient population.