In contrast to the acetabular cup version (CV), stem version (SV) measurement on the crosstable lateral radiograph has not been validated. We, therefore, investigated: (1) the difference of SV on the crosstable lateral radiograph from that measured using computed tomography data and (2) the reliability of SV measurement on the crosstable lateral radiograph.
We examined the correlations of the crosstable-CV and crosstable-SV with true-CV (operative cup anteversion) and true-SV (neck version relative to the posterior condylar line) obtained using computed tomography data from 103 patients. We also produced the digital reconstructed crosstable lateral radiograph with adjusted hip flexion and rotation and defined SV on this image as the true-crosstable SV.
Significant positive correlations were noted between the crosstable-CV and true-CV, with a correlation coefficient of 0.79. In contrast, the correlation between the crosstable-SV and true-SV was very low (r = 0.30), with significantly higher deviation from the true-SV values when compared with the crosstable-CV (standard deviation, 11.6° vs 6.5°, P < .001). The true-crosstable SV was correlated with true-SV (r = 0.81); however, the average difference was 23.5° ± 8.9°.
Crosstable-SV differed from true-SV in the measurement plane and did not correlate well with the true-SV, suggesting the unreliability of its measurements. Therefore, the crosstable-SV is not recommended for clinical use.