Development and validation of a bedside nomogram for predicting severe hospital-acquired pneumonia in traumatic brain injury patients based on early semi-quantitative chest CT assessment

Introduction Traumatic brain injury (TBI) often leads to severe hospital-acquired pneumonia (sHAP), which impairs recovery but remains difficult to predict early. We aimed to develop a bedside nomogram based on early chest CT features for individualized sHAP risk assessment in TBI patients. Methods This retrospective study included TBI patients screened for sHAP. Early semi-quantitative chest CT assessment (ground-glass opacity [GGO] and consolidation scores) was combined with clinical and laboratory parameters. LASSO regression and subset selection were used for model development, followed by 10-fold cross-validation and independent validation. Results An eight-variable model was established, including age, GCS score, mechanical ventilation, white blood cell count, lymphocyte count, hemoglobin, GGO score, and consolidation score. Multivariable analysis identified consolidation score (OR 1.20, 95% CI 1.03–1.39) and GGO score (OR 1.10, 95% CI 1.01–1.20) as independent risk factors. The nomogram showed robust discrimination with AUC of 0.880 (95% CI 0.851–0.908) in training and 0.871 (95% CI 0.824–0.918) in validation cohorts, with good calibration and clinical net benefit. Conclusions This validated bedside nomogram enables early, individualized sHAP risk stratification in TBI management, supporting targeted clinical decision-making.