Abstract
BACKGROUND: Acute ischemic stroke with large vessel occlusion (AIS-LVO) poses a grave threat to the health of the elderly, exhibiting a high degree of disability and mortality. Post-stroke ventilator-associated pneumonia (VAP) significantly impairs neurological recovery and worsens clinical outcomes. This study aimed to construct and validate a prognostic nomogram to forecast VAP risk in elderly patients who underwent endovascular therapy (EVT) with AIS-LVO. METHODS: We retrospectively analyzed a total of 536 patients with AIS-LVO who endured EVT under mechanical ventilation at the Dongguan Hospital of Guangzhou University of Chinese Medicine from August 2018 to March 2025. After applying inclusion/exclusion criteria, 240 elderly patients were randomly split into two groups: training (n = 168) and validation (n = 72), maintaining a 7:3 ratio. Using the least absolute shrinkage and selection operator regression (LASSO) for feature selection followed by multivariable logistic regression, we identified independent predictors for nomogram construction. Model performance was assessed through the area under receiver operating characteristic (ROC), calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC). RESULTS: Six independent predictors were identified: gender (OR 0.34, 95% CI 0.13∼0.85), nasogastric intubation (OR 7.56, 95% CI 1.77∼32.25), postoperative platelet-to-lymphocyte ratio(PLR) (OR 1.01, 95% CI 1.01∼1.02), postoperative neutrophil-to-lymphocyte ratio (NLR) (OR 1.22, 95% CI 1.02∼1.45), admission white blood cell(WBC) (OR 1.25, 95% CI 1.04∼1.49)and prognostic nutritional index (PNI) (OR 0.85, 95% 0.79∼0.92). The nomogram demonstrated excellent discrimination (AUROC 0.880, 95% CI 0.826∼0.933) and good calibration. DCA and CIC confirmed clinical utility across a wide probability threshold range. CONCLUSION: We developed and validated an effective nomogram incorporating six clinically accessible parameters to forecast VAP risk in elderly stroke patients post-EVT. This tool has the potential to expedite early high-risk patient identification and conduct preventive measures to enhance patient clinical outcomes.