2015 ISAKOS Biennial Congress ePoster #1006
Comparison of Axial Length and Version Measurements of Standard 2-Dimensional and Anatomically Aligned Computerized Tomography Scans
Daniel Gross, MD, Boston, MA UNITED STATES
Petar Golijanin, Research Coordinator, Boston, MA UNITED STATES
Guillaume D. Dumont, MD, Columbia, SC UNITED STATES
Stephen A. Parada, MD, Augusta, GA UNITED STATES
Anthony A. Romeo, MD, Scarsdale, NY UNITED STATES
Matthew T. Provencher, MD, Vail, CO UNITED STATES
Massachusetts General Hospital, Boston, MA, USA
FDA Status Cleared
Summary: Standard computed tomography (CT) scans of the shoulder are formatted in the plane of the body as opposed to the scapula and glenoid, resulting in imaging studies that distort the glenohumeral anatomy potentially resulting in errors when measuring the glenoid, both in terms of anterior-posterior (AP) glenoid length and glenoid version.
Standard computed tomography (CT) scans of the shoulder are formatted in the plane of the body as opposed to the scapula and glenoid, resulting in imaging studies that distort the glenohumeral anatomy potentially resulting in errors when measuring the glenoid, both in terms of anterior-posterior (AP) glenoid length and glenoid version. The hypothesis of this project is that the orientation of 2-dimensional (2D) CT scans performed in standard CT protocols is not optimized, resulting in substantial differences in glenoid AP length and version measurements.
Study design: Case Series
Thirty patients were selected from a database of normative CT shoulder imaging studies. Inclusion criteria to qualify a glenohumeral CT scan for this study was minimal to absent glenoid bone loss, absence of osteoarthritis, and no evidence of degenerative joint disease or joint space narrowing. The CT scans as originally obtained were defined as standard (STD). STD CT scans were reformatted using Osirix ™ multi-planar reconstruction (MPR) so that the scan was oriented to the plane of the scapula in the coronal and sagittal planes, and the axial cuts on the glenoid were oriented perpendicular to the 12 to 6 o’clock axis. These reformatted scans were defined as corrected (CRT). The amount of correction to orient the glenoid out of the plane of the body to optimal axial and coronal alignment was measured. The axial AP length and axial version measurements were then taken at five normalized cuts across a best-fit circle of the inferior glenoid, in both the STD and CRT images.
STD CT scans in the plane of the body needed a mean correction for coronal inclination of 5.53 degrees (range, -10.74 to 4.52, SD 5.51). Axial version required a mean correction of 35.53 degrees (range, 17 to 55, SD 9.17). Over five cuts, there was a mean difference in glenoid version of 2.39 degrees (2.6%), (range, 2.20 to 2.55, SD 0.12), and a mean difference in axial AP length of 1.23 mm (5.2%) (range, 0.83 to 1.77 mm) depending on the exact axial cut. This corresponds to anywhere from 3-9% error in measurement of the AP length of the glenoid.
These findings demonstrate that STD CT scans of the glenohumeral joint do not correct for the axial plane and coronal inclination of the glenoid by a large degree. This results in substantial error in both version and length (AP) measurements of the glenoid. In order to obtain an ideal glenohumeral joint representation, one should correct the axial and coronal images to be aligned with the plane of the glenoid and not in the plane of the body, as this may have notable implications for decision making and surgical treatment.