Development and preliminary validation of the score: a novel clinical-biomarker model for predicting stroke-associated pneumonia in acute ischemic stroke patients

Background Stroke-associated pneumonia (SAP) is a common and serious complication following acute ischemic stroke (AIS), contributing to increased morbidity and mortality. Existing clinical scores often lack accuracy in capturing the biological mechanisms underlying SAP. This study aims to develop and validate a novel clinical-biomarker scoring system to predict the occurrence of SAP in patients with AIS and to evaluate its predictive performance. Results SAP was diagnosed in 26 of 55 patients (47%). Dysphagia ( p = 0.004) and congestive heart failure ( p = 0.013) were significant clinical predictors. Glial fibrillary acidic protein (GFAP), triggering receptor expressed on myeloid cells-1 (TREM-1), and interleukin-10 (IL-10) were identified as relevant biomarkers. The five-variable score (GFAP, TREM-1, IL-10, smoking, and dysphagia) achieved an area under the curve (AUC) of 0.725 (95% CI: 0.592–0.859, p = 0.001). A simplified clinical implication score (GFAP, smoking, and dysphagia) demonstrated slightly better predictive accuracy (AUC 0.760, 95% CI: 0.627–0.893, p = 0.001). Conclusions Integrating biomarkers with clinical factors enhances SAP prediction in AIS patients. The simplified three-variable model demonstrated good predictive ability and practicality for bedside use, supporting its potential as a useful tool for early risk stratification of SAP in AIS patients.