Background. There has been increased focus on understanding the risk factors associated with scapular notching in reverse shoulder arthroplasty (RSA). The purpose of this study is to evaluate the effect of scapular morphology and surgical technique on the occurrence of scapular notching using the notching index as a comprehensive predictive tool. Methods. Ninety-one patients treated with a primary RSA were followed for a minimum of 24 months. Using a previously published notching index formula ((PSNA × 0.13) + (PGRD)), a notching index value for all patients was calculated. Radiographic assessment of patients were grouped by Nerot grade of scapular notching, group mean differences for prosthetic scapular neck angle (PSNA), peg glenoid rim distance (PGRD), preoperative scapular neck angle (SNA), notching index and clinical outcomes were compared. Results. Seventy-five (82%) of the ninety one shoulders in the study developed scapular notching. There was no significant difference in average notching index for group 1, 31.8 ± 4.4, and group 2, 33.1 ± 7.2. No significant difference was demonstrated between the groups for SNA (102.7° vs. 105.4°, p = 0.3), PSNA (125.8° vs. 124.5°, p = 0.82), PGRD (15.4 mm vs. 16.8 mm, p = 0.47) or in clinical outcomes between groups. Discussion. Our results demonstrated an overall low notch index that lacked specificity in predicting notching for this cohort. These results suggest that perhaps PSNA and prosthetic design are more significant contributors to notching with certain scapular morphology

Introduction. Reverse Shoulder Arthroplasty (RSA) is recognized to be an effective solution for rotator cuff deficient arthritic shoulders, but there are still concerns about impingement and range of motion (ROM). Several RSA biomechanical studies have shown that humeral lateralization can increase ROM in planar motions (e.g. abduction). However, there is still a debate whether humeral lateralization should be achieved with a larger sphere diameter or by lateralizing the center of rotation (COR). The latter has shown to decrease the deltoid moment arm and increase shear forces, where the former may pose challenges in implanting the device in small patients. The aim of this study was to evaluate how humeral lateralization achieved by varying COR lateral offset and glenosphere diameter in a reverse implant can affect impingement during activities of daily living (ADLs). Methods. Nine shoulder CT scans were obtained from healthy subjects. A reverse SMR implant (LimaCorporate, IT) was virtually implanted on the glenoid and humerus (neck-shaft angle 150°) as per surgical technique using Mimics software (Materialise NV). Implant positioning was assessed and approved by a senior surgeon. The 3D models were imported into a validated shoulder computational model (Newcastle Shoulder Model) to study the effects of humeral lateralization. The main design parameters considered were glenosphere diameter (concentric Ø36mm, Ø40mm, Ø44mm) and COR offset (standard, +2mm, +5mm), for a total of 9 combinations for each subject; −10°, 0° and 10° humeral components versions were analyzed. The model calculated the percentage of impingement (intra-articular, contact of cup with scapula neck and glenoid border; extra-articular, contact of humerus with acromion and coracoid) during 5 ADLs (hand to opposite shoulder, hand to back of head, hand to mouth, drink from mug and place object to head height). Results. On average, the Ø40mm and Ø44mm glenosphere resulted in significantly less impingement across ADLs compared to Ø36mm (−31% and −35% respectively). Humeral version and lateralization had no significant effect on impingement for the Ø44mm glenosphere. However, lateralization of +5 mm substantially reduced impingement on the Ø36mm glenosphere but the effect was significant only for the neutral 0° version (−42%) and 10° (−50%) anteversion. Discussion and Conclusions. The results of this study suggest that, for the SMR Reverse prosthesis, humeral lateralization through the increase of glenosphere diameter was the most efficient way to reduce impingement during ADLs compared to the lateralization of the COR. Humeral version can also affect the impact of lateralization on impingement during ADLs; in this study, the impingement for the Ø36mm glenosphere with 10° retroversion was not decreased through lateralization; this may be related to the combined effect of version and scapular morphology. Considering that using larger glenosphere diameter without offsetting the COR theoretically does not reduce overall deltoid lever arm nor increase the shear forces on the glenoid component, this should be the preferable option whenever possible. However, concerns over soft tissue over-tensioning may necessitate the use of a smaller diameter glenosphere in some patients