血管内治療後の大血管閉塞脳梗塞患者における肺炎リスク予測のための予後予測ノモグラムの開発と検証

Post-stroke hypostatic pneumonia (HP) significantly impairs neurological recovery and worsens prognosis in patients with acute ischemic stroke with large vessel occlusion (AIS-LVO). This study aimed to develop and validate a prognostic nomogram for predicting hypostatic pneumonia risk following endovascular therapy (EVT) in AIS-LVO patients.

We retrospectively analyzed 650 consecutive AIS-LVO patients who underwent endovascular therapy with mechanical ventilation at Dongguan Hospital of Guangzhou University of Chinese Medicine from September 2018 to March 2025. After applying inclusion/exclusion criteria, 412 patients were randomly split into two groups: training (n = 288) and validation (n = 124), maintaining a 7:3 ratio. Using least absolute shrinkage and selection operator (LASSO) regression for feature selection followed by multivariable logistic regression, we identified independent predictors for nomogram construction. Model performance was assessed through the receiver operating characteristic curve (ROC), calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC).

Four independent predictors were identified: admission Glasgow Coma Scale (GCS) score (OR 0.77, 95% CI 0.68-0.86), postoperative 48 h fever (OR 2.77, 95% CI 1.52-5.02), postoperative 48 h neutrophil-to-lymphocyte ratio (NLR) (OR 1.15, 95% CI 1.08-1.22), and ASPECTS (OR 0.74, 95% CI 0.63-0.87). The model had an area under the curve (AUC) of 0.829 (95% CI: 0.781-0.877) in the training cohort and 0.817 (95% CI 0.732-0.903) in the validation cohort, which means it was good at making predictions. Calibration curves revealed good alignment between predicted and observed probabilities in the training cohort. The validation cohort retained satisfactory calibration, with only modest overestimation of risk. DCA and CIC consistently indicated the nomogram's applicability in diverse clinical settings.

We developed and validated an effective nomogram incorporating four clinically accessible parameters to predict the risk of hypostatic pneumonia after EVT. This tool may facilitate early high-risk patient identification and guide preventive therapy to improve clinical outcomes.