**4. Discussion**

The most important finding of this study was that there was only a negligible correlation between radiographic measurements and clinical and functional outcomes following primary rTSA using a design with a humeral inclination of 135◦. Even though statistically significant correlations between postoperative outcomes scores and radiographic measurements were found, these observations are of limited predictive value for outcomes following rTSA. The data gathered from this multi-center study indicate that the importance of radiographic measurements and their correlation with outcomes following rTSA may be limited.

In the postoperative setting, lateralization in rTSA can be expressed by different radiographic variables, including LHO, LSA, and COR. In their retrospective study, Jeon et al. demonstrated that an increased postoperative LHO was found to be a significant risk factor for poor restoration of postoperative AFE, when using an implant designed to be medialized [21]. In contrast, the data from this study showed that in a cohort using a lateralized implant, no significant relationship between preoperative and postoperative LHO and AFE could be demonstrated. However, in this study, post-LHO was found to significantly influence final ASES score, which may be of limited predictive value, as this finding did not allow for drawing a definite conclusion.

Increasing the lateralization of the COR in rTSA using a medialized implant design has been shown to result in greater active ROM [9]. As only few studies have focused on measuring COR in lateralized implants [18,29], the authors from this study could not find a significant relationship between COR, clinical outcomes, and final active ROM when using a lateralized rTSA design. Similar to a previous study by Boutsiadis et al. [20], who reported that patients achieved the highest degree in postoperative AFE and AABD with a DSA between 40◦ and 65◦ and the highest degree in AER when having a LSA of 75◦ to 95◦, the findings of this study demonstrated a significant correlation between LSA and AFE as well as DSA and AFE. Additionally, the highest degree in AER was noted in patients having LSA values between 75◦ and 95◦; however, a direct correlation of LSA and DSA with final AER and AABD could not be confirmed [20]. This may be explained by existing differences in implant designs being used, as all included patients uniformly underwent rTSA using a lateralized design with a humeral inclination of 135◦. In contrast, Boutsiadis et al. included patients with two different implant designs (145◦ and 155◦ humeral inclination) [20].

Additionally, a positive correlation between LSA and DSA could be shown, which is consistent with the current literature [20]. In a lateralized rTSA design, a lower LSA, which corresponds to a more medialized implant, is associated with a larger DSA, indicating a greater distance between the acromion and humerus. To this, the findings from this study sugges<sup>t</sup> that a LSA greater than 100◦ correlates with a DSA of less than 40◦, thus reducing final AFE. Considering the current literature, lateralization of rTSA has been shown to increase postoperative AFE and AER by restoring the anatomic COR, while optimizing recruitment of the muscle fibers [6,30,31]. However, LSA was noted to be between 75◦ and 95◦ for optimal implant lateralization, with excessive lateralization resulting in less active ROM.

Increasing distalization, in order to improve tension on the deltoid muscle, has been shown to play an important role in rTSA [4]. In a computer-based model using different humeral offset and stem designs, Lädermann et al. demonstrated a strong positive linear relationship between AHD and AFE and AABD [6]. Furthermore, the authors showed that AHD decreased by 6 mm when switching from a 155◦ inlay design to a 135◦ onlay design. Even though it was shown that a higher AHD, expressed as arm lengthening, was related to a higher degree in AFE [6], the exact amount of arm lengthening remains inconclusive [6,18,32,33].

First introduced by Moor et al. [26], the CSA has been reported to be a reproducible radiographic index. As a larger CSA has been found to be associated with degenerative rotator cuff tears, there is still limited knowledge regarding its influence on rTSA [34]. Even though Roberson et al. [34] reported improved AFE in patients with a lower CSA, no significant relationship between CSA and clinical outcomes scores or final active ROM was found in this study.

Taking these findings into account, there remains a debate regarding the ideal placement of the components during rTSA to most sufficiently restore active ROM while minimizing complications such as component loosening and scapular notching. Additionally, as this study further verified, there is still a considerable lack of data in assessing radiographic prosthesis positioning in correlation with clinical outcomes. This may lead shoulder surgeons to overestimate current data and the importance of radiographic measurements and their correlation with outcomes following rTSA.

There are several limitations to the study. First, the study cohort was not randomized. Second, although outcomes were collected prospectively, data were reviewed retrospectively, which could create selection bias. Third, all radiographic measurements are highly dependent on patient orientation during radiographic imaging, as angles and distances are influenced by the position of the scapula as well as rotation of the humerus. However, this reflects daily clinical practice, as radiographic imaging, even if standardized, can show significant variances. Fourth, the multi-center design of this study including five surgeons from different sites leads to differences in implant positioning and intraoperative and postoperative outcomes. However, for the purpose of this study, the authors intended to demonstrate that even with high experience and expertise in this field, the observations and findings from this study and its subsequent comparison to the current literature should be interpreted with careful attention.
