Abstract
Pneumonia is a common and serious complication among patients with traumatic brain injury (TBI), significantly impacting their prognosis. The Braden Scale (BS), a tool for assessing the risk of pressure ulcers, has been shown to effectively predict outcomes in various conditions. This study aims to evaluate the predictive value of the Braden Scale for TBI-related pneumonia and to develop a pneumonia prediction model for TBI patients in the intensive care unit (ICU). Clinical data of TBI patients who had a BS documented on ICU admission were retrospectively extracted from the Medical Information Mart for Intensive Care database. Restricted cubic spline (RCS) was employed to evaluate the nonlinear relationship between BS and pneumonia. Univariable and multivariable logistic analyses identified risk factors for TBI-related pneumonia, facilitating the construction of a nomogram model. The nomogram underwent internal validation through receiver operator characteristic (ROC) curve, calibration curve, and decision curve analysis. This study involved 2,175 TBI patients, among whom 516 (23.7%) with pneumonia. The total BS score and each subscale score of pneumonia patients were significantly lower than those of non-pneumonia patients (all P < 0.001). RCS analysis revealed a nonlinear L-shaped relationship between BS and the risk of pneumonia. Multivariable analysis identified eight risk factors for TBI-related pneumonia, including BS < 15, male, calcium, heart failure, chronic pulmonary disease, sepsis, respiratory, and temperature. By incorporating these variables, the predictive nomogram achieved an area under the ROC curve (AUC) of 0.803 (95% CI, 0.782-0.824) and had a well-fitted calibration curve and good clinical practicability. Furthermore, the predictive value of the nomogram surpasses that of common severity score, including the Glasgow Coma Scale (AUC = 0.530), Simplified Acute Physiology Score II (AUC = 0.588), Sequential Organ Failure Assessment (AUC = 0.610), and Acute Physiology Score III (AUC = 0.607). The admission BS is a simple and reliable indicator for predicting TBI-related pneumonia. Incorporating BS into predictive model may enhance the assessment of pneumonia risk in TBI patients, thereby facilitating the identification of high-risk populations and promoting timely intervention.