Humeral resting abduction influences LSA and DSA: a radiographic and CT-derived 3D model analysis of rTSA using the Medacta shoulder system

The purpose of this study is to examine the effect of humeral abduction, as expressed by the humeral abduction resting angle (HARA) and scapular HARA (SHARA) on the lateralization shoulder angle (LSA) and distalization shoulder angle (DSA) using a) three-dimensional computed tomography-derived reverse total shoulder arthroplasty (rTSA) preoperative models, and b) consecutive postoperative radiographs obtained at multiple follow-up time points. Three-dimensional computed tomography-derived models generated using preoperative planning software and postoperative radiographs of patients who underwent rTSA using the Medacta Shoulder System, with at least one year of follow-up, were analyzed. In the 3D models, HARA was adjusted to 10&#xb0;, 20&#xb0;, and 30&#xb0;, and LSA and DSA were measured at each position and compared using one-way analysis of variance. For each patient, consecutive radiographs were obtained on postoperative day 1 and at 3, 6, and 12 months postoperatively. HARA and SHARA were measured on each radiograph. Then, pairs of consecutive radiographs were grouped according to their difference in HARA: <10&#xb0;, 10&#xb0;-20&#xb0;, and >20&#xb0;. Independent-samples t-tests were used for pairwise comparisons. Intrarater reliability and inter-rater reliability were evaluated using the intraclass correlation coefficient (ICC). Pearson's and Spearman's coefficients were applied to investigate possible correlations. In total, 45 preoperative 3D models and 135 postoperative radiographs were analyzed. In the 3D model analysis, mean DSA decreased from 48&#xb0; at 10&#xb0; of HARA to 44&#xb0; at 20&#xb0;, and 39&#xb0; at 30&#xb0; (P < .001), whereas mean LSA increased from 84&#xb0; at 10&#xb0; of HARA to 86&#xb0; at 20&#xb0;, and 89&#xb0; at 30&#xb0; (P < .001). When examining radiographs, variability in HARA was present in 95% of the cases. The mean HARA difference between pairs of consecutive radiographs was 10&#xb0; &#xb1; 10&#xb0;. All comparisons performed in the respective pairs of radiographs showed the same pattern: in every pair, the radiograph with the higher HARA and SHARA demonstrated significantly higher LSA and significantly lower DSA (P < .001). The intrarater ICC was 0.98 for DSA and 0.99 for LSA for 3D measurements, whereas the inter-rater ICC was 0.77 for DSA, 0.79 for LSA, 0.86 for HARA, and 0.81 for SHARA for radiographic measurements. There was a statistically significant, moderate positive correlation between HARA, SHARA, and LSA and a statistically significant moderate to strong negative correlation between HARA, SHARA, and DSA in both 3D models and radiographs (P < .001). In conclusion, this study shows that shoulder resting abduction significantly affects both LSA and DSA after rTSA. Higher humeral abduction results in higher LSA and lower DSA in both 3D models and radiographs. Humeral abduction should therefore be taken into account when planning lateralization and distalization in rTSA, and it should be controlled in future studies evaluating correlations with clinical outcomes.