Do Your Routine Radiographs to Diagnose Cam Femoroacetabular Impingement Visualize the Region of the Femoral Head-Neck Junction You Intended?

ABSTRACT

PURPOSE: To use computer models and image analysis to identify the position on the head-neck junction visualized in 10 radiographic views used to quantify cam morphology. METHODS: We generated 97 surface models of the proximal femur from computed tomography scans of 59 control femurs and 38 femurs with cam morphology-a flattening or convexity at the femoral head-neck junction. Each model was transformed to a position that represents the anteroposterior, Meyer lateral, 45° Dunn, modified false-profile, Espié frog-leg, modified 45° Dunn, frog-leg lateral, cross-table, 90° Dunn, and false-profile views. The position on the head-neck junction visualized from each view was identified on the surfaces. This position was then quantified by a clock face generated on the plane of the head-neck junction, in which the 12-o'clock position indicated the superior head-neck junction and the 3-o'clock position indicated the anterior head-neck junction. The mean visualized clock-face position was calculated for all subjects. Analysis was repeated to account for variability in femoral version. A general linear model with repeated measures was used to compare each radiographic view and anteversion angle. RESULTS: Each radiographic view provided visualization of the mean clock-face position as follows anteroposterior view, 1201; Meyer lateral view, 108; 45° Dunn view, 140; modified false-profile view, 201; Espié frog-leg view, 214; modified 45° Dunn view, 235; frog-leg lateral view, 245; cross-table view, 300; 90° Dunn view, 313; and false-profile view, 344. Each view visualized a different position on the clock face (all P < .001). Increasing simulated femoral anteversion by 10° changed the visualized position of the head-neck junction to a more clockwise position (range, 007 to 029; all P < .001), whereas decreasing anteversion by 10° visualized a more counterclockwise position (range, -023 to -008; all P < .001). CONCLUSIONS: Ten common radiographic views used to identify cam morphology visualized different clock-face positions of the head-neck junction. Our data will help clinicians to understand the position of the head-neck junction visualized for each radiographic view and make educated decisions in the selection of radiographs acquired in the clinic. CLINICAL RELEVANCE Our findings will aid clinicians in choosing a set of radiographs to capture cam morphology in the assessment of patients with hip pain.