BackgroundImplant position is critical in the outcomes of reverse and anatomic total shoulder arthroplasties (RTSA and ATSA). Studies to date focused on glenoid orientation but the literature on scapular position is lacking. This study aims to assess scapular coronal inclination (SCI) on erect shoulder X-rays compared with that on shoulder planning software. In addition, we compare SCI on erect shoulder X-rays before and after RTSA and ATSA. MethodsThis is a single-center retrospective review of erect X-rays of patients who had ATSA and RTSA performed by a single surgeon (J.T.) between January 2013 and May 2021. SCI was measured using the Scapular Coronal Axis (SCA) and the Scapular Spine Line–Medial Border (SSL-MB) relative to the horizontal axis. To achieve our primary aim, the SCA was measured on preoperative X-rays and compared to those in shoulder planning software. The difference in SSL-MB angles preoperatively and postoperatively was our secondary outcome. X-rays were screened to ensure acceptable rotation in the sagittal and axial planes, using coracoclavicular distance and coracoacromial angle, respectively, to enable reliable comparison. A repeat measurement was performed 1 month apart to determine intraobserver reliability. Three investigators performed measurements separately to determine interobserver reliability. ResultsThere were 177 total shoulder arthroplasties (TSAs) identified during the study period but only 24 ATSA and 29 RTSA fulfilled the selection criteria. The SCA had an inferior tilt of mean −16.3° (95% confidence interval [CI] −18.38 to −14.13) compared to horizontal (0°) in planning software. Using the SSL-MB, the mean difference for ATSA was 0.5° (95% CI −1.59° to 2.56°) and the mean difference for RTSA was −0.2° (95% CI −2.73° to 2.37°). The SSL-MB differences were not statistically significant. All interobserver and intraobserver Intraclass Correlation Coefficients measurements were ≥ 0.85 except 0.73 for intraobserver coracoclavicular distance. ConclusionIn conclusion, the native SCA was tilted inferiorly at a mean of −16.3° and not perfectly horizontal like portrayed in shoulder planning software. Surgeons need to be aware of this to accurately place the glenoid component in TSA. There was no change in SCI after TSA. Correlation with clinical outcomes is needed.