Final ROM after reverse total shoulder arthroplasty (RSA) is multifactorial and affected by the implant’s position, soft-tissue tension, muscle function, appropriate rehabilitation, and even preoperative factors . Increasing impingement-free ROM is important for improving the patient’s functional outcome and avoiding scapular notching. Scapular notching is common after RSA; it may occur in as many as 96% of these procedures . Notching has been shown to result from friction in internal rotation (IR), extension, and external rotation . One notching study showed impingement in IR in 18% of patients . The study reported that small changes in the glenosphere’s position (3.4-mm inferiorization or 6.2-mm lateralization) could have avoided most of the impingement.
The ability to obtain full IR after RSA is often overlooked. Internal rotation is important for various self-care and hygiene activities with previous studies showing that 102° of IR is necessary to complete functional tasks associated with common shoulder scores , but it is often not predictably addressed by RSA . A recent meta-analysis showed that 20% of patients had difficulty with toileting after RSA . Another study showed that patients lost the ability to wash their backs (30%), tuck in their shirts (29%), and manage toileting (12%) after bilateral RSA . Finally, IR has been shown to correlate to patient outcomes, with improved IR in patients with more-favorable global clinical outcome scores, Constant-Murley scores, and pain scores . Rotational motion limitations, with a resultant decrease in function, have been shown to yield unsatisfactory outcomes . Studies have shown that the implant’s configuration, including glenosphere inferiorization or lateralization and humeral neck-shaft angle, has an effect on ROM, but IR is often not reported [13, 29]. A systematic review showed that IR was not reported in any form in 35% of RSA articles and when it was reported, various methods were used . Impingement limits IR, and the absence of important muscle tendon units, such as the absence or nonrepair of the subscapularis or release of the latissimus dorsi, pectoralis major, or teres major, may also limit IR. One study showed improved IR with subscapularis repair during RSA , but the benefit depends on the integrity of the repair . The reported effect on the patient’s clinical outcome is variable [12, 34], leading many surgeons to choose not to repair the subscapularis. Even when repair is performed, a functional subscapularis cannot overcome bony impingement; thus, efforts such as subscapularis repair to address the internal rotator muscle groups would be ineffective if passive IR were limited by preventable impingement. Templating software programs have been developed for planning RSA procedures and have the ability to assist surgeons in achieving the desired placement of their implants . Using this software we can identify the location of bony impingement allowing surgeons to adjust the preoperative plan to avoid unwanted contact and improve IR.
Therefore, we performed a CT-modeling study, in which we asked: What reverse total shoulder arthroplasty implant position improves the range of impingement free internal rotation without compromising other motions (external rotation and extension)?
Patients and Methods
Twenty-five CT scans were uploaded into three-dimensional (3-D) surgical planning software (Imascap, Brest, France) to measure impingement-free ROM in various implant configurations (Fig. 1). This 3-D planning software has been shown to reliably measure the glenoid’s morphology and assist with appropriate component positioning in shoulder arthroplasty . Virtual implantation of an RSA humeral component with an onlay design (Aequalis Ascend Flex, Wright Medical, Bloomington, MN, USA) was performed using an anatomic neck cut. The baseline configuration consisted of a 29-mm baseplate placed at the inferior edge of the glenoid, a 36-mm standard glenosphere, 3 mm of baseplate lateralization from the native glenoid surface, a humeral component with a neck-shaft angle of 145°, and anatomic humeral version (Fig. 2). Lateralization was accomplished with the use of a bone graft as described by Boileau et al. . This configuration was taken through a simulated ROM until bony impingement was encountered. Additional test configurations (Table 1) with variable anterior and inferior offset (7.5 mm and 10 mm), lateralization (10 mm), glenosphere size (39 mm), neck-shaft angle (135°), and humeral version (+ 10° of anteversion) in isolation and in combination were similarly taken through simulated ROM and the results were compared with the baseline configuration. We also evaluated other ROM parameters including extension and external rotation to determine if the gains in IR obtained were associated with other types of motion losses.
Table 1. -
|Baseline: 29-mm baseplate with 3-mm lateralization, 36-mm centered glenosphere, anatomic humeral version, and neck-shaft angle of 145°
|Additional 7 mm of baseplate lateralization (total of 10 mm)
|Increased humeral anteversion by 10°
|Varus neck-shaft angle (135°)
|Increased glenosphere size to 39 mm
|Decreased baseplate size to 25 mm; used 36-mm sphere with 2-mm eccentricity placed anteriorly
|Decreased baseplate size to 25 mm; used 36-mm sphere with 2-mm eccentricity placed inferiorly
|Decreased baseplate size to 25 mm; used 39-mm sphere with 3-mm eccentricity placed anteriorly
|Decreased baseplate size to 25 mm; used 39-mm sphere with 3-mm eccentricity placed inferiorly
|Combined increased lateralization, increased humeral anteversion, 25-mm baseplate, 39-mm glenosphere with 3-mm eccentricity inferiorly, and varus neck-shaft angle of 135°
Twenty-five CT scans were obtained from a deidentified teaching database maintained by the senior author (GW). We included 25 consecutive patients with primary osteoarthritis and Walch A1 glenoid morphology . Patients with A1 glenoids were chosen to prevent changes in ROM caused by substantial glenoid deformity or wear.
Our data were found to be nonparametrically distributed by the Shapiro-Wilk test so comparisons were assessed using Wilcoxon’s signed rank test for quantitative variables. The values for quantitative variables are expressed as medians and interquartile ranges. The significance was evaluated by calculating the p value. All comparisons were tested at the two-tailed 0.05 significance level.
The final simulation (Fig. 3), which combined lateralization, inferiorization, varus neck-shaft angle, increased glenosphere size, and increased humeral anteversion, resulted in a greater improvement in IR than any single parameter change did (baseline IR 85° [IQR 73° to 90°], combined changes IR 119° [IQR 113° to 121°], median difference 37° [IQR 32° to 43°]; p < 0.001) (Table 2). Individual alterations in implant size, configuration, and position resulted in an improvement in IR (p < 0.05), except for the lateralized configuration (Table 2).
Table 2. -
Comparison of IR between configurations
||Median IR (IQR) (°)
||Median difference compared with baseline (IQR) (°)
||p value compared with baseline configuration
||85 (73 to 90)
||84 (76 to 90)
||3 (-3 to 9)
|Increased humeral anteversion
||90 (81 to 99)
||9 (9 to 9)
|135° neck-shaft angle
||95 (87 to 99)
||11 (8 to 14)
||91 (82 to 95)
||5 (4 to 8)
|36-mm sphere, eccentric anterior
||97 (88 to 102)
||13 (11 to 18)
|36-mm sphere, eccentric inferior
||98 (90 to 103)
||13 (12 to 21)
|39-mm sphere, eccentric anterior
||105 (97 to 107)
||20 (15 to 25)
|39-mm sphere, eccentric inferior
||104 (97 to 107)
||19 (17 to 24)
||119 (113 to 121)
||37 (32 to 43)
IR = internal rotation; IQR = interquartile range.
Inferiorization of the glenosphere increased impingement-free IR. With a 29-mm baseplate at the inferior edge of the glenoid and a 36-mm glenosphere, the baseline inferior overhang of the glenosphere was 3.5 mm. Increasing inferiorization by decreasing the baseplate size to 25 mm and using a 36-mm glenosphere with 2 mm of inferior eccentricity resulted in a total inferior offset of 7.5 mm and led to an increase in IR compared with the baseline measurement (98° [90° to 103°], median difference from baseline 13° [12° to 21°]; p < 0.001). Additional inferiorization totaling 10 mm, which was obtained by using a 25-mm baseplate and a 39-mm glenosphere with 3 mm of inferior offset, resulted in an increase in IR, which was an improvement over the baseline configuration’s values (104° [97° to 107°], median difference 19° [17° to 24°]; p < 0.001) and was greater than those for the 7.5-mm inferior configuration (median difference 4° [4° to 7°]; p < 0.001). Anteriorization of the glenosphere overhang had a similar effect on IR, as obtained by inferiorization, and showed a stepwise improvement, but resulted in worse extension (37° [24° to 64°] versus 50° [39° to 96°], median difference 16° [12° to 22°]; p < 0.001) and external rotation (45° [36° to 63°] versus 49° [39° to 63°], median difference 5° [3° to 6°]; p < 0.001) than the inferiorly offset spheres did (Table 3).
Table 3. -
Comparison of external rotation and extension between inferior and anterior eccentric glenospheres
||Median difference (IQR)
|Median ER (IQR) (°)
||49 (39 to 63)
||45 (36 to 63)
||5 (3 to 6)
|Median extension (IQR) (°)
||50 (39 to 96)
||37 (24 to 64)
||16 (12 to 22)
ER = external rotation; IQR = interquartile range.
After glenosphere inferior offset, the greatest improvement in IR was obtained by use of a more varus 135° neck-shaft angle. Improved IR was also seen with a larger glenosphere. Finally, increased humeral version also improved IR, but the increased humeral anteversion was associated with a reciprocal loss of external rotation (15° [5° to 30°] versus 21° [12° to 33°]; median difference 7° [1° to 7°]; p = 0.004) and therefore did not improve global rotational motion. The only configuration without improvement in IR was after 7 mm of additional glenoid lateralization (10 mm total) (84° [76° to 90°], difference from baseline 3° [-3° to 9°]; p = 0.09).
Internal rotation is important for daily activities  and after RSA, impingement free-IR can help to decrease scapular notching . Limitations in IR are not always addressed by RSA  and frequently lead to functional deficits [26, 31]. We used a 3-D CT-based computer model to template RSAs and simulate ROM to find bony impingement limiting IR and then modified the template to find which configuration changes a surgeon could use to improve IR. The results showed that the greatest improvement in IR was obtained by combining lateralization, inferiorization, varus neck-shaft angle, increased glenosphere size, and increased humeral anteversion. Each of these changes also showed some IR improvement in isolation except for lateralization. If these findings are used for preoperative planning and surgical instrumentation, then constructs free of bony impingement can result and any ROM limitations that are found can be managed by soft-tissue releases intraoperatively.
There are limitations to this study. First, this was a computational model and does not consider the soft tissues. Without knowing whether capsular contractures are present and whether the subscapularis was repaired, healed appropriately, and is functional, we cannot determine the IR that would result in the clinical setting. However, just as preoperative templates help us achieve desired surgical goals, computer models can help us know what to study clinically. Passive impingement-free IR must be present to allow the internal rotator muscles to function. If bone or implant impingement restricts IR, the internal rotator muscles cannot surpass it. Finding factors that improve IR in a computer model can help us when generating clinical hypotheses and when evaluating clinical outcomes. Also, when trying to improve IR intraoperatively, knowing from a preoperative template where bony impingement will occur will help surgeons know whether further soft-tissue releases should be considered or if the maximum amount has already been achieved. We also did not report on all ROM because IR was the focus of this study; many other studies have evaluated other ROM. Additional configurations could have been evaluated, including an inlay versus onlay humeral design, glenoid version , and glenoid tilt [21, 37]. Surgeons, however, should be cautious increasing glenoid tilt with eccentric spheres because despite its common use clinically it results in a shearing force across the glenoid . Baseplate size was also not evaluated as an independent factor but has previously been shown to have no effect on IR . With the ever-increasing number of implants available, it is not feasible to examine every option and we chose these based on their commonality. Still, surgeons should be aware that these findings cannot be applied to every implant. Lastly, although we reported that the above-mentioned configurations will improve IR, in this study, we could not determine whether these configurations are possible in every patient or have other yet unknown adverse effects. We examined the effect on other ROM including external rotation and extension as well to ensure gains in IR were not associated with losses in other ROM but again the computer model is limited to evaluation of bone and implant and other potential downsides related to the soft tissues cannot be determined. Despite this limitation, these implant configurations are currently available and have been shown in the evidence to have good clinical outcomes [3, 8, 11, 22]. Further objective clinical evaluations of the implant configurations discussed in this study are warranted to determine the extent to which inferiorization or lateralization can be used safely.
What Configuration Increases IR without Loss of Global ROM?
A combination of techniques can be used to increase IR; in this study, improvements after combined lateralization, inferiorization, varus neck-shaft angle, increased glenosphere size, and increased humeral anteversion were almost as large as the sum of each individual improvement (37° versus 42°). The largest gain in IR from a single parameter change was seen after inferiorization and we advocate its use in RSA. As the improvements increased with added inferiorization, a surgeon can adjust the amount of overhang to achieve the desired IR. Inferior overhang has also been shown to decrease scapular notching [10, 17, 19, 25] and lower acromial stresses , which may lead to fewer stress fractures. Surgeons should be careful not to over lengthen the arm when increasing inferior overhang as it can worsen outcomes . Overlengthening can be combated with eccentric humeral trays in onlay designs or by use of a more varus neck shaft angle, which will also help to improve IR and decrease notching [16, 24, 36]. Ultimately, using a combination of techniques can prevent over lengthening while still allowing for improved IR.
Lateralization in our study was not shown to improve IR on its own, which is consistent with previous evidence [1, 36]. One prior study did show an increase in IR when the baseplate was lateralized from 0 mm to 5 mm but not from 5 mm to 10 mm . Our study examined a change from 3 mm to 10 mm without finding a difference. This shows that the potential benefits of lateralization on IR may be seen early but are not enhanced by further lateralization. These findings should be further evaluated with large clinical studies. Despite this, lateralization is important for improving global ROM and limiting scapular notching and should therefore be considered when performing RSA [1, 13, 24, 36].
In our study, changes in humeral version were also shown to affect IR, but increases in anteversion resulted in reciprocal decreases in external rotation and must be considered if this method is to be used to improve IR. These results highlight that there is an arc of rotation, and placing the humeral implant into anteversion shifts the rotational arc to favor IR (Fig. 4). Conversely, placing the humeral implant into retroversion shifts the rotational arc to favor external rotation.
The effect of a larger glenosphere is less clear. Increasing glenosphere size also results in some increased overhang, and we cannot differentiate whether the sphere size actually plays a role in improving IR or whether it is due to the resultant overhang. Some previous studies have shown improved IR with a larger sphere [30, 37], while others have shown no improvement in IR with a larger sphere but found improvement with an eccentric sphere , making it unclear where the benefit is derived from. Additionally, the effect of sphere size on polyethylene wear has recently been examined  with larger spheres resulting in greater volumetric polyethylene wear and smaller spheres resulting in greater linear penetration into the polyethylene. These results do not clearly identify one sphere size as superior to others but should be considered when determining which sphere to use.
The largest improvements in IR were seen after combining baseplate lateralization, inferior glenosphere overhang, increased glenosphere diameter, varus humeral neck shaft angle, and increased humeral component anteversion in our computer modeling study. Each of these modifications in isolation also improved IR with the exception of lateralization. Surgeons can use these findings when templating and performing RSA to improve impingement-free IR in their patients. When using these findings as part of a preoperative plan, surgeons can refer to the resultant IR and determine whether further soft tissue releases could be beneficial. Validation of these findings through detailed reporting of implant configuration and increased reporting of resultant IR in clinical studies is needed.
1. Athwal GS, MacDermid JC, Reddy KM, Marsh JP, Faber KJ, Drosdowech D. Does bony increased-offset reverse shoulder arthroplasty decrease scapular notching? J Shoulder Elbow Surg. 2015;24:468-473.
2. Boileau P, Moineau G, Roussanne Y, O'Shea K. Bony increased-offset reversed shoulder arthroplasty: minimizing scapular impingement while maximizing glenoid fixation. Clin Orthop Relat Res. 2011;469:2558-2567.
3. Boileau P, Morin-Salvo N, Bessière C, Chelli M, Gauci MO, Lemmex DB. Bony increased-offset-reverse shoulder arthroplasty: 5 to 10 years' follow-up. J Shoulder Elbow Surg. 2020;S1058-2746.
4. Boileau P, Watkinson DJ, Hatzidakis AM, Balg F. Grammont reverse prosthesis: design, rationale, and biomechanics. J Shoulder Elbow Surg. 2005;14:147S-161S.
5. Cazeneuve JF, Cristofari DJ. Long term functional outcome following reverse shoulder arthroplasty in the elderly. Orthop Traumatol Surg Res. 2011;97:583-589.
6. Chae SW, Kim SY, Lee H, Yon JR, Lee J, Han SH. Effect of baseplate size on primary glenoid stability and impingement-free range of motion in reverse shoulder arthroplasty. BMC Musculoskelet Disord. 2014;15:417.
7. Collin P, Matsukawa T, Denard PJ, Gain S, Lädermann A. Preoperative factors influence the recovery of range of motion following reverse shoulder arthroplasty. Int Orthop. 2017;41:2135-2142.
8. Collotte P, Erickson J, Vieira TD, Domos P, Walch G. Midterm clinical and radiologic results of reverse shoulder arthroplasty with an eccentric glenosphere. J Shoulder Elbow Surg. 2020;29:976-981.
9. Dedy NJ, Gouk CJ, Taylor FJ, Thomas M, Tan SLE. Sonographic assessment of the subscapularis after reverse shoulder arthroplasty: impact of tendon integrity on shoulder function. J Shoulder Elbow Surg. 2018;27:1051-1056.
10. de Wilde LF, Poncet D, Middernacht B, Ekelund A. Prosthetic overhang is the most effective way to prevent scapular conflict in a reverse total shoulder prosthesis. Acta Orthop. 2010;81:719-726.
11. Erickson BJ, Frank RM, Harris JD, Mall N, Romeo AA. The influence of humeral head inclination in reverse total shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2015;24:988-993.
12. Friedman RJ, Flurin PH, Wright TW, Zuckerman JD, Roche CP. Comparison of reverse total shoulder arthroplasty outcomes with and without subscapularis repair. J Shoulder Elbow Surg. 2017;26:662-668.
13. Gutiérrez S, Comiskey CA 4th, Luo ZP, Pupello DR, Frankle MA. Range of impingement-free abduction and adduction deficit after reverse shoulder arthroplasty. Hierarchy of surgical and implant-design-related factors. J Bone Joint Surg Am. 2008;90:2606-2615.
14. Gutiérrez S, Walker M, Willis M, Pupello DR, Frankle MA. Effects of tilt and glenosphere eccentricity on baseplate/bone interface forces in a computational model, validated by a mechanical model, of reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2011;20:732-739.
15. Haggart J, Newton MD, Hartner S, Ho A, Baker KC, Kurdziel MD, Wiater JM. Neer Award 2017: wear rates of 32-mm and 40-mm glenospheres in a reverse total shoulder arthroplasty wear simulation model. J Shoulder Elbow Surg. 2017;26:2029-2037.
16. Keener JD, Patterson BM, Orvets N, Aleem AW, Chamberlain AM. Optimizing reverse shoulder arthroplasty component position in the setting of advanced arthritis with posterior glenoid erosion: a computer-enhanced range of motion analysis. J Shoulder Elbow Surg. 2018;27:339-349.
17. Kolmodin J, Davidson IU, Jun BJ, Sodhi N, Subhas N, Patterson TE, Li ZM, Iannotti JP, Ricchetti ET. Scapular notching after reverse total shoulder arthroplasty: prediction using patient-specific osseous anatomy, implant location, and shoulder motion. J Bone Joint Surg Am. 2018;100:1095-1103.
18. Krämer M, Bäunker A, Wellmann M, Hurschler C, Smith T. Implant impingement during internal rotation after reverse shoulder arthroplasty The effect of implant configuration and scapula anatomy: a biomechanical study. Clin Biomech. 2016;33:111-116.
19. Lädermann A, Denard PJ, Boileau P, Farron A, Deransart P, Walch G. What is the best glenoid configuration in onlay reverse shoulder arthroplasty? Int Orthop. 2018;42:1339-1346.
20. Lädermann A, Gueorguiev B, Charbonnier C, Stimec BV, Fasel JH, Zderic I, Hagen J, Walch G. Scapular notching on kinematic simulated range of motion after reverse shoulder arthroplasty is not the result of impingement in adduction. Medicine. 2015;94:e1615.
21. Li X, Knutson Z, Choi D, Lobatto D, Lipman J, Craig EV, Warren RF, Gulotta LV. Effects of glenosphere positioning on impingement-free internal and external rotation after reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2013;22:807-813.
22. Müller AM, Born M, Jung C, Flury M, Kolling C, Schwyzer HK, Audigé L. Glenosphere size in reverse shoulder arthroplasty: is larger better for external rotation and abduction strength? J Shoulder Elbow Surg. 2018;27:44-52.
23. Namdari S, Yagnik G, Ebaugh DD, Nagda S, Ramsey ML, Williams GR Jr, Mehta S. Defining functional shoulder range of motion for activities of daily living. J Shoulder Elbow Surg. 2012;21:1177-1183.
24. Nelson R, Lowe JT, Lawler SM, Fitzgerald M, Mantell MT, Jawa A. Lateralized center of rotation and lower neck-shaft angle are associated with lower rates of scapular notching and heterotopic ossification and improved pain for reverse shoulder arthroplasty at 1 year. Orthopedics. 2018;41:230-236.
25. Nyffeler RW, Werner CM, Gerber C. Biomechanical relevance of glenoid component positioning in the reverse Delta III total shoulder prosthesis. J Shoulder Elbow Surg. 2005;14:524-528.
26. Rojas J, Joseph J, Liu B, Srikumaran U, McFarland EG. Can patients manage toileting after reverse total shoulder arthroplasty? A systematic review. Int Orthop. 2018;42:2423-2428.
27. Rojas J, Joseph J, Srikumaran U, McFarland EG. How internal rotation is measured in reverse total shoulder arthroplasty: a systematic review of the literature. JSES Int. 2019;4:182-188.
28. Roy JS, Macdermid JC, Goel D, Faber KJ, Athwal GS, Drosdowech DS. What is a successful outcome following reverse total shoulder arthroplasty? Open Orthop J. 2010;4:157-163.
29. Samitier G, Alentorn-Geli E, Torrens C, Wright TW. Reverse shoulder arthroplasty. Part 1: systematic review of clinical and functional outcomes. Int J Shoulder Surg. 2015;9:24-31.
30. Torrens C, Guirro P, Miquel J, Santana F. Influence of glenosphere size on the development of scapular notching: a prospective randomized study. J Shoulder Elbow Surg. 2016;25:1735-1741.
31. Triplet JJ, Kurowicki J, Berglund DD, Rosas S, Horn BJ, Levy JC. Loss of functional internal rotation following various combinations of bilateral shoulder arthroplasty. Surg Technol Int. 2018;33:326-331.
32. Walch G, Badet R, Boulahia A, Khoury A. Morphologic study of the glenoid in primary glenohumeral osteoarthritis. J Arthroplasty. 1999;14:756-760.
33. Walch G, Vezeridis PS, Boileau P, Deransart P, Chaoui J. Three-dimensional planning and use of patient-specific guides improve glenoid component position: an in vitro study. J Shoulder Elbow Surg. 2015;24:302-309.
34. Werner BC, Wong AC, Mahony GT, Craig EV, Dines DM, Warren RF, Gulotta LV. Clinical outcomes after reverse shoulder arthroplasty with and without subscapularis repair: the importance of considering glenosphere lateralization. J Am Acad Orthop Surg. 2018;26:e114-e119.
35. Werner BS, Ascione F, Bugelli G, Walch G. Does arm lengthening affect the functional outcome in onlay reverse shoulder arthroplasty? J Shoulder Elbow Surg. 2017;26:2152-2157.
36. Werner BS, Chaoui J, Walch G. The influence of humeral neck shaft angle and glenoid lateralization on range of motion in reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2017;26:1726-1731.
37. Werner BS, Chaoui J, Walch G. Glenosphere design affects range of movement and risk of friction-type scapular impingement in reverse shoulder arthroplasty. Bone Joint J. 2018;100:1182-1186.
38. Werner CM, Steinmann PA, Gilbart M, Gerber C. Treatment of painful pseudoparesis due to irreparable rotator cuff dysfunction with the Delta III reverse-ball-and-socket total shoulder prosthesis. J Bone Joint Surg Am. 2005;87:1476-86.
39. Wong MT, Langohr GDG, Athwal GS, Johnson JA. Implant positioning in reverse shoulder arthroplasty has an impact on acromial stresses. J Shoulder Elbow Surg. 2016;25:1889-1895.