Experimental evaluation of scaphoid X-series, carpal box radiographs, planar tomography, computed tomography, and magnetic resonance imaging in the diagnosis of scaphoid fracture.

Aim: In this study, we evaluated scaphoid X-series, Carpal Box radiographs (longitudinal and transverse), planar tomography, computed tomography (CT), and magnetic resonance imaging (MRI) in the diagnosis of scaphoid fracture. The aim of this study was to evaluate the planar technique in the diagnosis of scaphoid fracture. The use of planar tomography, CT, and MRI was to see whether these methods are useful in the diagnosis of scaphoid fracture when other diagnostics modalities remains negative.

Methods: Twenty-eight embalmed human cadaver specimens were used, in 23 of which fractures of the scaphoid were produced mechanically. Scaphoid X-series, Carpal Box posterior-anterior radiographs in ulnar deviation (X-CB), Carpal Box posterior-anterior views with the hand in 15-degree supination and ulnar deviation (X-CB 15-degree) were acted in all specimens, CT in eight wrists, planar tomography in seven wrists, and MRI in five wrists. The anatomic analysis of the specimens was used as the gold standard for comparison. Scaphoid X-series, Carpal Box radiographs, and planar tomography were judged independently and in a blind fashion by six observers, and CT and MRI were also judged independently and in a blind fashion by three radiologists. The observers were asked if they could recognize a scaphoid fracture. The agreement among the six observers for the scaphoid X-series and X-CB was measured.

Results: In the 23 fractured wrists, scaphoid X-series, X-CB, X-CB 15-degree, was true positive in 12, 14, and 15 wrists, respectively, whereas these methods were true negative in cadaver wrists 1, 3, and 5. CT was true positive in five of five fractured wrists and true negative in three of three negative wrists. Planar tomography was true positive in one of four fractured wrists and true negative in two of three nonfractured wrists. MRI was obtained in five wrists (one without a fracture), of which the fracture was recognized in only two. The highest agreement between observers was found in the X-CB 15-degree.

Conclusion: From the planar investigated methods, the 15-degree Carpal Box posterior-anterior, longitudinal and transverse views were most accurate in recognizing scaphoid fracture with also the highest agreement between the observers.