Zone of Injury Determined by Free Air on Computed Tomography Scans Predicts Open OTA 42A-C Tibia Fracture Complications.

Abstract

To define a CT-derived zone-of-injury metric, incorporating normalized soft-tissue air extent and BMI and secondly to determine if this metric was associated with adverse outcomes following an open OTA 42A-C tibia fractures. Design: Retrospective cohort study. Level I trauma center. A retrospective review of patients in a tibia fracture registry (2012-2024) meeting inclusion criteria (age ≥18 years old, open OTA 42A-C fractures, preoperative full length tibia CT imaging, ≥6-month follow-up) was performed.Outcome Measures and Comparisons: The CT-based ZOI was measured as the longitudinal extent of soft-tissue air (mm) normalized to tibial length (mm) (ZOIsoft/Tibial Length). The primary outcome was composite complications including fracture related infection, amputation, or nonunion. A logistic regression model using ZOIsoft/Tibial Length and BMI generated predicted probabilities for composite complications. Model discrimination was assessed via area under receiver operating characteristic (AUROC) analysis and compared to Gustilo-Anderson classification using the DeLong test. An optimal probability threshold was derived statistically (Youden Index) for dichotomizing patients into high- and low-risk cohorts. Fifty-five patients (58 fractures) met inclusion criteria (mean age 40.0 ± 15.1 years; 81.0% male; mean follow-up 16.9 ± 9.5 months). Soft-tissue ZOI and BMI were the significant predictors of composite complications (p = 0.006, 0.061). The CT-based ZOI model (log(p/1-p) = 0.601 + (3.343 × soft-tissue ZOI/Tibial Length) + (-0.106 × BMI) demonstrated superior discrimination (AUROC = 0.752) compared to Gustilo-Anderson (AUROC = 0.581, p = 0.042). Patients above the derived threshold (0.403) had significantly worse outcomes: composite complication rate 64.0% vs. 18.2% (p < 0.001) and nonunion (52.0% vs. 9.1%, p < 0.001). Amputation (20.0 vs 3.0%, p = 0.075) and fracture-related infection (32.0% vs. 15.2%, p = 0.203) were not significant. A novel CT-based ZOI metric integrating soft-tissue injury extent as measured by soft-tissue air and BMI independently predicted overall complications risk. This newly described CT-based ZOIsoft metric provided superior prognostic accuracy compared to Gustilo-Anderson classification and may enhance early risk stratification in open tibia fractures. Prognostic Level III.