Reliability of semi-pronated ulnar deviation PA view (billiard view) of the wrist in evaluating the scaphoid.

Various methods assessing the scaphoid have been reported because of its unique position. In our hospital, pre- and postoperative evaluation of the scaphoid alignment has been carried out in the billiard view, a combination of 45° pronated oblique with ulnar deviation posteroanterior (PA) view, in addition to 3DCT scan and PA and lateral view. This study compared the intra- and inter-observer reliabilities of the lateral intrascaphoid angle and scaphoid length (SL) and scaphoid height (SH) on the billiard view. A total of 60 patients who underwent surgery for scaphoid nonunion were identified and the preoperative and the final follow-up postoperative plain radiographs were used for measurement. Three observers assessed each image. Intra- and inter-observer reliability was determined using intra-class correlation (ICC) coefficients. Intra-observer reliability was all excellent ranging between 0.855 and 0.992. Inter-observer reliability ranged between 0.292 and 0.983. SL and SH demonstrated excellent agreement, while ISA demonstrated poor to moderate agreement. The best method for assessing the scaphoid in simple radiograph remains debatable, but our current data suggest that measuring SL and SH on the billiard view is reproducible and can be used for evaluating restoration of scaphoid alignment.