gms | German Medical Science

Objectives/Interrogation: Screws protruding the proximal pole of the scaphoid during retrograde scaphoid fixation with cannulated screws may lead to unfavorable outcomes. Conventional intraoperative radiographic views have limitations to detect this complication.

We propose a modification of the Skyline view of the distal radius (Joseph et al 2011, Riddick et al. 2012) to assess the proximal pole of the scaphoid tangentially.

The aim of this study was to evaluate the ability of a new radiographic view), tangential to the proximal pole of the scaphoid, the Scaphoid Skyline View (SSV), to detect intrarticular screws during retrograde fixation of the scaphoid.

Methods: 9 fresh frozen adult cadaveric wrists without degenerative radiographic signs were used. A retrograde screw was inserted with a volar percutaneous retrograde technique.

Radiographic views studied were: posteroanterior (PA), PA with ulnar deviation (PAUD), lateral (L), oblique pronated (OP), PA with extension and ulnar deviation (PAEUD) and the SSV.

SSV is obtained with aligning the forearm at 30º respect to the Xray beam and maximum wrist flexion.

Baseline X rays were recorded with screw flush to the proximal pole, visualized directly through a limited dorsal capsulotomy. The screw tip was advanced sequentially 0.5 mm measured with a digital caliper through the dorsal approach. After every screw advance all views were repeated until screw prominence a was noticed in all images.

The mean articular protrusion (mm) required to detect intrarticular screws was compared between different radiographic views using oneway ANOVA.

Accuracy of the different radiographic view was studied with ROC curve analysis. Statistically significant differences were considered with p<0.05.

Results: The mean (SD) protruding length (in mm) required to detect articular screws with different radiographic views was:

PA 1.9 (1), PAUD 1.8 (1.1), OP 1.4 (0.9), L 1.1 (0,7), PAEUD 1.2 (0.7), SSV 0.8 (0.3). Statistically significant differences were found between groups (oneway ANOVA, p=0.014, F=3,756). SSV had the highest accuracy with an area under curve of 0.92 (95% Confidence interval: 0.831-1).

Conclusions:

• SSV was able to detect intrarticular screws with less than 1mm, being the most accurate view.
• Other radiographic views showed similar ability to detect protruding screws as described in literature [1].
• This radiographic view could avoid complications when used during retrograde fixation of the scaphoid with cannulated screws.