Occult fractures of the tibial plateau: 3D augmented X-ray pilot study.

Abstract

Purpose: Tibial plateau fractures are commonly missed on conventional radiographs (CR), particularly when subtle or nondisplaced. This study evaluates the diagnostic performance of Multitom Rax, a novel cone-beam computed tomography (CBCT)-based 3D augmented X-ray system that provides multiplanar tomographic reconstructions alongside CR, for the detection of tibial plateau (TP) fractures in the emergency setting.

Methods: Over a four-month period, 23 emergency department patients with suspected knee fractures underwent 3D Augmented X-ray (Multitom RAX, Siemens Healthineers). Four blinded radiologists with varying experience levels sequentially interpreted conventional radiographs and 3D tomographic images. Reader confidence in presence or absence of fracture was rated on a five-point Likert scale. Reference standard was established by MRI, standard CT, surgery or consensus of three expert musculoskeletal radiologists. Diagnostic performance was evaluated via area under the receiver operating characteristic curve (AUC) and compared between modalities. Differences in AUC between modalities were assessed using confidence interval-based comparison at a 95% significance level (α = 0.05) with the null hypothesis of no difference in AUC.

Results: Five occult TP fractures were identified among 23 patients (21.7%, 95% CI: 7.5-43.7%). Only two were detected using CR by the most experienced reader, whereas all five were correctly identified on 3D reconstructions without false positives by more experienced radiologists. The per reader-averaged AUC across patients increased from 0.748 ± 0.087 with CR to 0.933 ± 0.043 with 3D. The difference in AUC (mean 0.185, 95% CI: [0.019, 0.369]) was significant at the 95% level (p = 0.023).

Conclusion: 3D Augmented X-ray demonstrated superior sensitivity and diagnostic confidence in detecting subtle TP fractures compared to CR. These preliminary findings support the potential value of 3D CBCT-based imaging in the evaluation of acute knee trauma and warrant validation in larger studies.