Purpose: To evaluate the diagnostic accuracy of high-spatial-resolution ultrasonography (US) in the diagnosis of scaphoid fractures.
Materials and methods: In 72 hours after acute wrist trauma, 15 consecutive patients were examined for possible scaphoid fractures clinically and with conventional radiographs, including scaphoid views. Thereafter, high-spatial-resolution US was performed by two experienced radiologists blinded to the results of the previously performed investigations. High-spatial-resolution US of the scaphoid bones was performed from the palmar, lateral, and dorsal directions in the longitudinal and transverse planes. US findings indicative of a scaphoid fracture were cortical discontinuity and/or periosteal elevation. Finally, magnetic resonance (MR) images (short inversion time inversion-recovery, T1- and T2*-weighted) (ie, the standard) of the affected wrist were obtained and evaluated for a possible scaphoid fracture by two radiologists in consensus.
Results: Nine of 15 patients had scaphoid fractures. Seven (78%) of nine patients had positive findings at high-spatial-resolution US and five (56%) had such findings at conventional radiography (ie, four occult scaphoid fractures), with an accuracy of 87% and 73%, respectively. Two (50%) of four radiographically occult scaphoid fractures were depicted with high-spatial-resolution US. Sonographic findings of scaphoid fractures were either cortical discontinuity (n = 4), periosteal elevation (n = 2), or a combination of these two findings (n = 1).
Conclusion: High-spatial-resolution US is a reliable diagnostic tool for the evaluation of occult scaphoid fractures and should be considered an adequate alternative diagnostic tool prior to computed tomography or MR imaging.