ABSTRACT
BACKGROUND: Computed tomography (CT) scans of the shoulder are often not well aligned to the axis of the scapula and glenoid. The purpose of this paper was to determine the effect of sagittal rotation of the glenoid on axial measurements of anterior-posterior (AP) glenoid width and glenoid version attained by standard CT scan. In addition, we sought to define the angle of rotation required to correct the CT scan to optimal positioning. METHODS: A total of 30 CT scans of the shoulder were reformatted using OsiriX software multiplanar reconstruction. The uncorrected (UNCORR) and corrected (CORR) CT scans were compared for measurements of both (1) axial AP glenoid width and (2) glenoid version at 5 standardized axial cuts. RESULTS: The mean difference in glenoid version was 2.6% (2° ± 0.1°; P = .0222) and the mean difference in AP glenoid width was 5.2% (1.2 ± 0.42 mm; P = .0026) in comparing the CORR and UNCORR scans. The mean angle of correction required to align the sagittal plane was 20.1° of rotation (range, 9°-39°; standard error of mean, 1.2°). CONCLUSION: These findings demonstrate that UNCORR CT scans of the glenohumeral joint do not correct for the sagittal rotation of the glenoid, and this affects the characteristics of the axial images. Failure to align the sagittal image to the 12-o'clock to 6-o'clock axis results in measurement error in both glenoid version and AP glenoid width. Use of UNCORR CT images may have notable implications for decision-making and surgical treatment.
