Cross-Institutional Validation of the UChicago PBI Imaging Score in Civilian Penetrating Brain Injury

<title>Abstract</title> <p>Background: Civilian penetrating brain injury (PBI) remains one of the most lethal forms of traumatic brain injury, with early prognostication posing a significant clinical challenge. Existing prognostic tools are limited, particularly in their incorporation of injury-specific radiographic features for acute risk stratification. The UChicago PBI Imaging Score was developed as a head computerized tomography (HCT)-based prognostic score for in-hospital mortality after civilian PBI. We evaluated its performance in an independent cohort from the original institution and an external validation cohort. Methods We conducted a two-cohort validation study using data from two Level 1 trauma centers: the University of Chicago Medical Center (UCMC) and the University of Oklahoma Health Sciences Center (OUHSC). The UCMC cohort served as an independent validation cohort from the original institution, and the OUHSC cohort served as an external validation cohort. The score’s discriminatory performance for in-hospital mortality was assessed using receiver operating characteristic curve analysis, with area under the curve (AUC) values calculated for each cohort and pooled validation sample. Results A total of 108 patients were included (36 from OUHSC and 72 from UCMC). Median age was 27.7 years (IQR, 21.6–39.4), and overall, in-hospital mortality was 54.6%. Non-survivors had lower presenting Glasgow Coma Scale scores than survivors [median 3.0 (IQR, 3.0–4.0)]. The UChicago PBI Imaging Score demonstrated strong discrimination for in-hospital mortality, with AUCs of 0.86 (95% CI, 0.73–0.99) in the OUHSC external validation cohort, 0.83 (95% CI, 0.73–0.93) in the UCMC independent validation cohort, and 0.84 (95% CI, 0.77–0.91) in the pooled validation sample. Conclusion The UChicago PBI Imaging Score demonstrated strong discriminatory performance for in-hospital mortality in both cohorts. Used as an adjunct to clinical assessment, the score may improve early risk stratification, provide a standardized imaging framework, and support PBI research.</p>