Relationship between wrist morphology and scaphoid fracture: a CT-based case control study

Aims: To determine whether wrist morphology measured on computed tomography (CT)—in particular, ulnar variance (UV), radial inclination (RI), radial height (RH), scaphoid volar tilt (SVT), and lunate volar tilt (LVT)—is independently associated with scaphoid fracture after a fall on the outstretched hand. Methods: We performed a retrospective, CT-based case–control study of 78 patients (39 CT-confirmed scaphoid fractures; 39 controls). Standardized multiplanar reconstructions were used to obtain true coronal and fossa-centered sagittal planes. Two radiologists (≥10 years’ experience) made masked, duplicate measurements; inter- and intra-observer reliability was assessed using ICCs, and good-to-excellent agreement was confirmed. Group comparisons employed t-tests or Mann–Whitney U as appropriate; independent associations were evaluated with multivariable logistic regression. Discrimination for UV was examined with ROC analysis. Results: UV was significantly more negative in the fracture group (−1.9±1.6 mm) than in controls (−1.1 ± 1.6 mm; p=0.047). In the multivariable model including RI, UV, RH, SVT, and LVT, UV was the only independent predictor (B=-0.315; OR=0.73; 95% CI 0.54–0.98; p=0.035); other parameters were not independently associated (all p>0.05). Model fit was modest (Nagelkerke R²=0.092). Using UV alone, the AUC was 0.613 (95% CI 0.479–0.732). The Youden-optimal cutoff UV