Comparison of DSA and LSA angle measurements in postoperative radiographs and computed tomography of patients with reverse total shoulder arthroplasty.

The distalization shoulder angle (DSA) and lateralization shoulder angle (LSA) are radiographic measurements used to assess the position of the implant and its impact on shoulder function in reverse shoulder arthroplasty (RSA). The present study aims to determine (a) whether LSA and DSA measurements obtained from radiographs and CT scans of the same patients yield comparable results and (b) the inter-observer reliability of radiographic and CT measurements. LSA and DSA measurements of 88 shoulders in 80 patients were performed on radiographs and multiplanar reconstructions of CT scans. A paired t-test was used to compare angle measurements both between imaging modalities (radiograph versus CT) and between observers. Inter-method correlation was evaluated using Spearman's Rho and inter-observer agreement was assessed using the intraclass correlation coefficient (ICC). There was a very good inter-method correlation for LSA (Spearman's Rho = 0.876), with the total mean differences between radiographic and CT measurements being 2° (SD ± 3.6°, range -8.5 to 5°, p < 0.001). The ICC for LSA was 0.691 for radiographs and 0.721 for CT scans. There was also a very good inter-method correlation for DSA (Spearman's Rho = 0.855), with the total mean difference between radiographic and CT measurements being 3.5° (SD ± 4.8°, range -9.5 to 12°, p < 0.001). The ICC for DSA was 0.858 for radiographs and 0.768 for CT scans. CT-derived measurements of the DSA and LSA demonstrate high reproducibility, showing strong agreement with radiographic measurements. Strong inter-observer reliability can be expected for both angles in both radiographic and CT assessments.