Summary
Surgeons and manufacturers can control many things about the reverse total shoulder (RSA), including component design, size and placement.
Computer models and laboratory tests attempt to predict the particular RSA geometries that restore muscle lengths and avoid unwanted contact between bone and implant.
This post asks a narrower question: do any of these choices make the shoulder work better or reduce complications for patients receiving an RSA?
Published trials found little effect on how patients did. Five randomized trials each tested a single aspect of geometry: 6 mm of lateralization on the glenoid side, a baseplate tilted 10° downward, a humeral component with a 135° rather than 155° neck-shaft angle, and, in two trials, a humeral component in neutral version rather than 30° of retroversion. Four found no difference in function scores, pain or motion at one to two years. The fifth found that 30° of humeral component retroversion gave more external rotation with the arm at the side, while neutral version gave slightly better scores on two questionnaires; the ASES score did not differ. Some observational comparisons, such as a more lateral glenosphere, a lower neck-shaft angle, a baseplate inclined slightly upward or a baseplate in marked anteversion or retroversion, showed more motion or less notching on X-ray, but not better scores. Where lateralized designs did score better, the difference was small: about 5 to 6 points on the 100-point Subjective Shoulder Value. Two studies that did report better scores compared a surgeon's historical outcomes with later ones, after a switch to a lateralized design or to matching humeral component retroversion to each patient's own anatomy. In such comparisons, later patients may do better for other reasons, such as surgeon experience and patient selection. In one study of shoulders placed with the same retroversion, the more recent patients did better than the earlier ones.
Implant labels such as "lateralized" or "135° inlay" describe the implant, but not where the parts end up in the patient. Reaming and seating change that. When researchers measured the final position on X-rays in 630 shoulders, patients with similar geometry had widely different outcomes.
Reproducing a preoperative plan more precisely is not associated with better outcomes. Patient-specific guides made from patients' CT scan put the glenoid baseplate closer to the surgeon's preoperative plan, and robotics and augmented or virtual reality aim to do the same. However, in the one study that measured function, patients whose baseplate was placed with patient-specific guides did no better at two years than those whose baseplate was placed without them.
Moreover, the scapula itself sits at a different angle in each person: in 93 patients whose baseplate was placed with patient-specific guides at a planned 0° inclination, the inclination relative to the floor ranged from 1° to 53° downward.
Notching. The humeral component can wear a notch in the scapular neck below the glenoid. Lateralized designs and a glenosphere placed low, overhanging the inferior glenoid, reduce notching on X-ray. In a randomized trial, a 135° humeral component reduced notching from 58% to 21% without changing patient-reported outcomes. How much notching harms the patient is less clear: one series found no effect in the first few years; others found slightly lower scores and less motion.
Fractures of the acromion. Stress fractures of the acromion and scapular spine are the complication most often blamed on lateralization. The largest study covered 6,320 reverses at 15 U.S. centers. There, the strongest predictors were patient factors: inflammatory arthritis, a massive cuff tear, osteoporosis, prior surgery and female sex. Findings on glenoid-side lateralization and arm lengthening conflicted. One surgeon who began medializing the glenoid in high-risk patients saw fractures fall from 9% to 2%, but this study compared earlier patients with later ones. At another center, fractures were rarer when the acromion was farther from the glenosphere center than the greater tuberosity was (0.4% vs 5.0%). Fractures matter to patients: in one series, those who had them had less forward elevation and were less likely to return to their activities.
Instability. In one surgeon's series, a baseplate tilted more superiorly after surgery was associated with instability, and patients reporting instability had lower ASES scores.
Few studies had enough dislocations to test the effect of implant position.
How often RSAs go wrong varies with the report. Pooling 52 published series gave a complication rate of 9.4% and a revision rate of 2.6%. A Swiss center that followed 854 reverses prospectively found complications in 18% and further surgery in about 10%. The most common complications were acromial fracture and glenoid component loosening. Patients with a complication scored well below those without.
A 2026 review concedes that whether a particular geometry helps patients "is unclear," and that design changes "should ultimately be evaluated by clinical outcomes."
What we know and what we don't. In randomized trials, single changes in baseplate tilt, glenoid lateralization, neck-shaft angle and humeral component version produced no or only small differences in patient-reported results.
It has not been demonstrated that any particular geometry gives consistently better patient-reported outcomes.
The evidence in detail
A recent post covered the factors a surgeon could use in choosing between an anatomic and a reverse total shoulder.
When a reverse is selected, the surgeon would like to know which geometry leads to optimal comfort, function and satisfaction for the patient, and to fewer complications that need further care?
That is the focus of this post.
Another recent post, examined how each element of reverse geometry can be measured on postoperative radiographs: baseplate inclination, inferior glenosphere overhang, glenoid and humeral lateralization, distalization, humeral component version, neck-shaft angle and liner capture. It showed that many measures in this literature do not measure what they are named for, that angles often stand in for millimeters, and that values shown on the box are often reported in place of the geometry actually achieved in the shoulder.
This post asks the question: does the geometry change what the patient experiences?
Only clinical studies are considered: studies of patients, with outcomes reported by patients or measured on them.
Computer models, cadaver studies and finite-element analyses are not considered.
A 2026 review of the reverse in cuff-intact shoulders makes the same point. It describes the case for a 135° humeral component and 5–10 mm of glenoid lateralization, which rests on computational and biomechanical studies, then concedes that whether these translate into better motion, strength or patient-reported outcomes "is unclear," and that "design changes should ultimately be evaluated by clinical outcomes and PROs" (Wong et al., Curr Rev Musculoskelet Med 2026 [1]).
Reviews can blur this line. A 2022 systematic review of humeral and glenoid version in the reverse recommends 0° to 20° of retroversion for both components. Nine of its 11 included studies were biomechanical, one was clinical-radiographic, and one described implant design (Berton et al., J Clin Med 2022 [2]). Its recommendation is therefore not used here.
Geometry and clinical function
In studies of patients, no position or geometry of the reverse has been shown consistently to improve patient-reported function by an amount the patient would notice. Of five randomized trials, four found no difference in patient-reported function and the fifth found small, mixed differences. The observational associations are weak.
Geometry | Best clinical evidence | Effect on patient-reported function |
Glenoid lateralization, 6 mm vs nonlateralized-distalized | RCT, one surgeon, 53 of 66 enrolled analyzed, any indication including revision (Southam et al., JSES 2023 [3]) | No difference in ASES, PROMIS upper extremity or rotation-dependent daily activities, notching or acromial fracture at 2 years |
Glenosphere +4 mm vs standard | Matched cohort, 102 pairs (King et al., OTSR 2023 [4]) | No difference in pain, motion or scores |
Lateralized vs nonlateralized glenospheres, bony or metallic | Meta-analysis of 16 comparisons, 865 shoulders; low levels of evidence (Nunes et al., JSES 2021 [5]) | No difference in ASES, Constant or SST; Subjective Shoulder Value 6 points higher and pain 1 point lower with lateralization |
Glenosphere 0 vs 4 mm with a 135° humeral component | Cohort, 82 patients (Harmsen et al., JSES Int 2022 [6]) | No difference reported |
Humeral version, neutral vs 30° retroversion | Double-blinded RCT, 66 patients (Wiater et al., JBJS 2024 [7]) | No difference in rotation, elevation, strength, pain, PROMIS or ASES |
Humeral version, neutral vs 30° retroversion (second trial) | Single-blinded RCT, one center, 60 randomized and 57 analyzed; 87% massive cuff tears or cuff tear arthropathy; one lateralized system (Ventura-Parellada et al., JSES 2026 [8]) | 30° gave more external rotation at the side (50° vs 38°). Neutral gave better SST (76 vs 67 of 100, about one item of 12) and SF-12 physical scores (45 vs 39). No difference in ASES, Constant or pain at 2 years |
Humeral component neck-shaft angle, 135° vs 155° | Single-surgeon RCT, 100 randomized, 68 analyzed at 2 years or more (9 revised shoulders excluded); irreparable cuff tear or osteoarthritis; neutral glenosphere; one inlay system (Gobezie et al., JSES 2019 [9]) | No difference in ASES (74 vs 78), SANE (74 vs 76), SST (8 vs 7), pain, forward flexion (132° vs 135°) or external rotation (29° vs 30°). Powered for a 10° difference in flexion, not for patient-reported outcomes |
Humeral component neck-shaft angle, 135° vs 145°, one onlay system | Retrospective, 103 shoulders, 30 at 135°; glenoid lateralization and inferior eccentricity also differed between groups (Neyton et al., JSES 2023 [10]) | No difference in Constant (66 vs 68), Subjective Shoulder Value (77% vs 83%) or motion |
Humeral version matched to each patient's CT (10°–40°) vs fixed 20° | Retrospective, one surgeon, 80 patients, nearly all cuff tear arthropathy; fixed 20° in 2007–2012, individualized afterward (Oh et al., JSES 2020 [11]) | Individualized group better: ASES 80 vs 73, SST 7.3 vs 5.9, ER 49° vs 39°, flexion 142° vs 129°. But 21 later patients whose individualized angle was 20° or less (mean 19°) also did better than the earlier fixed-20° group, so the difference cannot be attributed to version |
Glenoid tilt, 10° inferior vs neutral | RCT, 42 analyzed (Edwards et al., JSES 2012 [12]) | No difference in clinical outcome |
Glenoid inclination as measured, superior (median 6°) vs neutral or inferior (median −4°) | Retrospective, one surgeon, 154 patients, one lateralized system (Mahendraraj et al., BJJ 2021[13]) | No difference in ASES, SANE, pain or motion at 2 years |
Glenoid baseplate version measured on the axillary radiograph: 10° or more of anteversion, neutral, 10° to 20° of retroversion, or 20° or more of retroversion | Retrospective, one institution, 115 shoulders with osteoarthritis or cuff tear arthropathy; one system whose glenosphere is two-thirds of a sphere; from our group (Sheth et al., Int Orthop 2025 [14]) | No difference in SST, ASES or change in SST, and no linear relation between version and SST; no difference in complications or revisions. Patients with 10° or more of anteversion more often reported difficulty putting on a coat (86% vs 42% with neutral version) |
Inferior glenosphere overhang, measured, by tertile (<7.1, 7.1–9.9, >9.9 mm) | Retrospective, one center, 284 shoulders, one medialized-glenoid, lateralized-humerus system (Dean et al., JSES 2022[15]) | High overhang gained 12° more active elevation than low (49° vs 37°), at the MCID threshold. No differences in ASES, Constant, SPADI or SST changes reached the MCID |
Measured global lateralization and distalization (LSA, DSA) | Two registries, 630 reverses for cuff tear arthropathy, 5 implant systems (Imiolczyk et al., JSES 2024 [16]) | No linear or curved relation of either angle to Constant score, Subjective Shoulder Value or motion at 2 years |
Measured lateralization and distalization angles, split into glenoid and humeral parts | Retrospective, 217 patients (Kirkham et al., JSES 2026 [17]) | A greater increase in the lateralization shoulder angle and a smaller distalization angle went with a slightly better Subjective Shoulder Value (a ratio of 1.01 per degree, not related to an MCID); no association with ASES |
Three designs: 155° inlay (Grammont), 135° inlay with 4 mm lateralized glenosphere, 145° onlay with lateralized baseplate | Two-center registry, 226 patients with cuff tear arthropathy; not randomized (Freislederer et al., BMC Musculoskelet Disord 2024[18]) | No difference in Constant score through 2 years. Flexion and abduction best with the 145° onlay; rotation better with both lateralized designs |
The same three designs in primary osteoarthritis | Two-center registry, 97 patients; groups differed before surgery: supraspinatus intact in 56% of the onlay group and 11–12% of the others, 98% women in the 135° group (Imiolczyk et al., BMC Musculoskelet Disord 2025 [19]) | Subjective Shoulder Value 90% with the 155° design vs 95% and 93% with the lateralized designs (P = .046); no difference in Constant score; more motion with the lateralized designs |
155° humeral component with 2.5 mm lateral offset vs 135° humeral component with 6–10 mm lateralized glenosphere | Retrospective, one surgeon, 107 consecutive patients; the surgeon switched designs in mid-series (Zitkovsky et al., JAAOS 2020[20]) | Lateralized 135° group better: ASES 87 vs 77, SANE 95 vs 85, ER 51° vs 40°; flexion did not differ. Lateralization, neck-shaft angle, implant system and era all changed together |
Design category: medial glenoid–medial humerus, medial glenoid–lateral humerus, lateral glenoid–medial humerus | Meta-analysis of 42 studies (Burden et al., BJJ 2021 [21]) | No difference in pooled Constant or ASES scores among the three |
Global lateral humeral offset and humeral version on CT, relative to each patient's preoperative anatomy | 30 patients, CT before and after surgery, isokinetic dynamometer at a mean of 2.4 years (Charles et al., J Biomech 2024 [22]) | Greater offset predicted greater strength, but offset more than 5 mm beyond the patient's own was associated with less strength; external rotation was greatest with version within 10° of native and offset at or medial to native. No patient-reported outcome |
Two other findings deserve mention. When the subscapularis was repaired and the glenosphere was lateralized, ASES improvement was smaller than with either alone (Werner et al., JAAOS 2018 [23]). A meta-analysis of nine nonrandomized studies of lateralized humeral designs (lower neck-shaft angle or onlay tray) found 7° more external rotation, but no difference in Constant or ASES scores (Dhillon et al., Cureus 2021 [24]).
Before-and-after comparisons. The two studies in the table with clearly better scores, Oh [11] and Zitkovsky [20], each compared a surgeon's earlier patients with later ones. Experience, technique, rehabilitation and patient selection all change over time, and none of these can be separated from the implant change. The Oh study tests this directly. Later patients who happened to receive about the same 20° as the earlier group also did better, which points to the era rather than the version. In the Zitkovsky study, the later group also had shorter follow-up.
Implant category is not placement. Most of these studies classify shoulders by what the manufacturer's catalog says the implant does: a "lateralized" glenosphere, a "135° inlay" humeral component. A deeply seated onlay behaves as an inlay; an incompletely seated inlay behaves as an onlay. The lateralization a patient receives depends on the glenoid reaming, the seating and the humeral cut, not only the label. Studies that measure the result on the patient's radiograph, such as LSA and DSA, are a step forward. In the largest of them, patients with identical angles had widely different outcomes.
Putting the baseplate where it was planned. A 2026 meta-analysis of 16 studies (532 shoulders) found that patient-specific guides brought the glenoid component closer to the plan for inclination and central pin position, but not for version. The mean errors were 3.0° for inclination, 1.7 mm for pin position and 3.8° for version. Only one included study measured function, and it found no difference at two years. The authors conclude that better accuracy "has not been shown to improve clinical outcomes and PROMS" (Vicary-Watts et al., JSEA 2026 [25]).
Even a baseplate placed exactly as planned on the scapula does not sit the same way in every patient. Ninety-three patients had their baseplate placed with patient-specific guides at a planned 0° inclination. Each patient's resting scapular rotation differed, so inclination relative to gravity averaged 18° downward, ranging from 1° to 53°. At 9 weeks, a more open resting angle between the arm and the scapula predicted better ASES scores. By 26 weeks, no radiographic measure, including global inclination, predicted outcome (Shekhbihi et al., JSES 2026 [26]; "Neutral" inclination is therefore a statement about the scapula, not about how the joint is loaded when the patient stands.
Geometry and complications
The clinical evidence links geometry to notching and to acromial and scapular spine fractures, but the studies point in different directions on the key variable, glenoid lateralization.
How often. A meta-analysis of 52 studies of primary reverses (5,824 shoulders, mean follow-up 3.9 years) found an overall complication rate of 9.4% and a revision rate of 2.6%. Notching was reported in 14.4% (Galvin et al., JSES 2022[27]). A single center that collected data prospectively on 854 primary reverses, 99% with one 155° onlay system, found higher rates. Complications occurred in 156 shoulders (18%), and 82 (about 10%) needed further surgery. The most common complications were acromial fracture (5.3%), glenoid loosening (4.3%), instability (2.7%) and humeral fracture or loosening (2.5%). Patients with a complication had a mean relative Constant score of 57% vs 81% and a Subjective Shoulder Value of 53% vs 80% (Kriechling et al., BJJ 2022 [28]). The authors suggest their onlay design may have added acromial stress, but the series had no comparison group. The difference between these two sources is itself a finding: complication rates depend on how carefully patients are followed. Across design categories, pooled glenoid loosening was 6% with medial glenoid–medial humerus designs vs 0% with medial glenoid–lateral humerus designs, a finding the reviewers rated as low-strength evidence (Burden et al. [21]).
Notching. In a randomized trial, 10° of inferior tilt did not reduce notching: 75% with tilt vs 86% without, all placed 3 mm inferior (Edwards et al. [12]). In the randomized neck-shaft angle trial, with a neutral glenosphere, notching was 21% with a 135° humeral component and 58% with 155°, while patient-reported outcomes did not differ (Gobezie et al. [9]). Within one onlay system, notching was 53% at 145° and 30% at 135°, though glenoid lateralization also differed (Neyton et al. [10]). In primary osteoarthritis, notching was 22% with a 155° design, 16% with a 135° lateralized design and 9% with a 145° onlay (Imiolczyk et al. [19]). A meta-analysis of lateralized glenospheres found lower odds of notching (odds ratio 0.14; Nunes et al. [5]). A +4 mm glenosphere was associated with less notching in one matched cohort (King et al. [4]) but not in another that used a 135° humeral component: 24.4% vs 19.5% (Harmsen et al. [6]). Lateralized humeral designs reduced notching in pooled nonrandomized studies (odds ratio 0.17; Dhillon et al. [24]). In the randomized lateralization trial, notching did not differ (Southam et al. [3]). In a registry comparison of three designs, notching was 50% with the Grammont design, 24% with the 135° inlay and 14% with the 145° onlay (Freislederer et al. [18]). Pooled across 42 studies, notching was 52% with medial glenoid–medial humerus designs, 18% with medial glenoid–lateral humerus designs and 12% with lateral glenoid–medial humerus designs (Burden et al. [21]). When one surgeon switched from a 155° design with 2.5 mm of offset to a 135° design with 6–10 mm, notching fell from 41% to 15% and heterotopic ossification from 72% to 35% (Zitkovsky et al. [20]). Measured glenosphere overhang was protective: notching was 18%, 11% and 3% in the low, middle and high overhang tertiles (Dean et al. [15]). Superior inclination up to about 6° on a lateralized design did not increase notching (Mahendraraj et al. [13]). In 517 shoulders with one 135° inlay system, 6 mm or more of glenoid lateralization reduced notching in women (5.3% vs 15.6%), and glenosphere overhang was protective; lateralization was the implant's nominal value (Pak et al., JSES 2026 [29]). Whether notching matters to the patient is less clear. In one series it had no effect on outcomes at short-term follow-up (Harmsen et al. [6]). In the 517-shoulder series, notched shoulders had 13° less forward flexion and ASES scores 7 points lower, below the MCID (Pak et al. [29]). In another, notching rose from 11.9% at 2 years to 19.5% at 5 years and was associated with worse scores, motion and complications (Shelley et al., J Orthop 2020 [30]).
Acromial and scapular spine fractures. These are the complications most often blamed on lateralization.
Study | Patients | Associated with more fractures | Associated with fewer fractures or no effect |
6,320 primary and revision reverses, 15 US institutions; fracture rate 3.8% | Strongest: patient factors, including inflammatory arthritis (OR 2.29), massive cuff tear (2.05), osteoporosis (2.00), prior surgery (1.82) and female sex (1.74). Geometry: total glenoid lateral offset (OR 1.06), greater increase in LSA (1.42), greater postoperative LSA (1.76); the LSA ratios appear to be across the interquartile range, not per degree | More lateral humeral offset (OR 0.74, across the interquartile range; see below). Distalization: no association | |
470 primary reverses, all with one 135° inlay humeral component; fracture rate 5.5% | Greater increase in acromiohumeral distance (OR 2.21 per cm); each mm of inferior glenosphere overhang (+19%); dominant arm (OR 2.21) | Glenoid lateralization 0–8 mm: no association (7.4%, 5.6%, 4.4%, 6.0%) | |
1,146 reverses, one institution; 21 fractures (1.8%) vs 84 matched controls | Less humeral lateralization (4.1 vs 8.4 mm), higher glenoid inclination, osteoporosis; less distalization with more medialization of the center of rotation | — | |
Fracture rate 1.1% | Less lateral greater-tuberosity offset, more arm lengthening, thinner acromion | — | |
282 primary reverses, one surgeon; 13 fractures (4.6%); before vs after the surgeon began medializing the glenoid in higher-risk patients | Glenoid lateralization over 5 mm by implant specifications (OR 14.5, 95% CI 1.75 to 119.6); 12 of 13 fractures in lateralized glenoids | Fracture rate fell from 9.2% to 2.3%, and from 10.7% to 3.1% in patients with two or more risk factors | |
526 reverses, one medialized-glenoid, lateralized-humerus onlay system; 360 after propensity matching | Acromion closer to the glenosphere center than the greater tuberosity: fractures 5.0%, subacromial notching 10.8% | Acromion as far or farther: fractures 0.4%, notching 0%. ASES, SPADI and SST did not differ | |
426 primary reverses, three systems; 26 scapular spine fractures | Onlay humeral component (11.9% vs 4.7% inlay); greater postoperative acromiohumeral distance (37.5 vs 33.7 mm) | Global lateralization (52.8 vs 53.9 mm): no association | |
860 reverses, one 155° onlay system; 45 fractures (5%) | Scapular spine (Levy III) fractures only: larger postoperative LSA (89° vs 83°), less distalization (acromiohumeral distance 33 vs 38 mm) | Acromial (Levy I and II) fractures: no radiographic predictor | |
920 reverses, one surgeon, one lateralized-glenoid 135° inlay system; 47 fractures (5.1%) vs 141 matched controls | — | No difference in global lateralization, acromiohumeral interval, delta angle, humeral seating or acromioclavicular joint morphology |
The patient factors are consistent across these studies: bone quality, inflammatory arthritis, female sex and cuff deficiency. The geometric findings are not. Glenoid lateralization was associated with fracture in two studies [31] [35] and not in two others [32] [39]. In one of the two that found it, lateralization came from implant specifications, earlier patients were compared with later ones, and there were 13 fractures [35]. Greater distalization went with more fractures in two studies [32] [37], with fewer scapular spine fractures in one [38], and with neither in two [31] [39]. Only one study measured the humeral side alone: more lateral humeral offset, from a line down the center of the humeral shaft to a parallel line through the glenosphere–humeral interface, went with fewer fractures [31]. The measures called humeral lateralization by Schenk, about 50 mm from glenoid to humerus [33], by Werthel [34] and by Haidamous [37] span the whole reconstruction and include the glenoid contribution. Moverman's odds ratios for lateral humeral offset (0.74) and for the lateralization shoulder angle (1.42 and 1.76) are labeled per millimeter or per degree, but the paper's Figure 2 does not fit that reading. Across the interquartile range of lateral humeral offset (6.7 to 11.9 mm), the odds of fracture fall by about the reported 0.74, or about 6% per millimeter. Across the plotted range of change in the lateralization shoulder angle (about 37°), the odds only double, or about 2% per degree; a ratio of 1.42 per degree would multiply the odds more than 100,000-fold over that range. Our earlier post discusses this further. One measure combines the two directions: whether the acromion sits farther from the glenosphere center than the greater tuberosity does. In one large series it was associated with fewer fractures [36]. All of these studies are observational, and the 135° inlay series was from a research group listed under the manufacturer's affiliation [32]. Fractures matter to patients. Those with scapular spine fractures had less forward flexion (120° vs 135°) and were less likely to return to their activities (71% vs 92%) [37]. In another series, patients with fractures were less satisfied despite equivalent ASES, pain and motion (Werthel et al. [34]), a reminder that a score can miss what a patient feels.
Instability. In 97 reverses from one surgeon, 13% reported some instability and 4% had a dislocation that needed reduction. A greater change toward superior baseplate inclination from before to after surgery was associated with instability (odds ratio 1.08 per degree). The paper does not agree with itself on the postoperative inclination: its text reports higher odds of instability with a larger β angle (odds ratio 1.15), while its regression table gives 0.92 and the unstable shoulders had the smaller mean angle, 81° against 88°. No subscapularis was repaired and lateralized glenospheres were excluded. Patients reporting any instability had lower ASES scores, 61 against 72 (Tashjian et al., CORR 2018 [40]). A systematic review of ten retrospective studies pooled an odds ratio of 2.48 (95% CI 1.22 to 5.07) for instability when the subscapularis was not repaired; the authors call their findings hypothesis-generating rather than causal (Ulenberg et al., JSES 2026 [41]). Whether the tendon can be repaired depends partly on its quality, which that comparison cannot account for. In the 217-patient series, a larger postoperative lateralization shoulder angle was associated with reoperation for baseplate failure or instability (odds ratio 1.11 per degree), but on only 5 events [17]. The randomized version trial had one revision for instability (Wiater et al. [7]), as did a matched series of reverses for cuff-intact osteoarthritis (Baek et al. [42]).
Where this leaves us
Demonstrated in patients: in randomized trials, glenoid tilt (0° vs 10° inferior), 6 mm of glenoid lateralization and a 135° vs 155° humeral component did not change patient-reported function; the 135° component reduced notching from 58% to 21%. Of two trials of humeral version (neutral vs 30°), one found no difference; the other found more external rotation at 30° and slightly better SST and SF-12 physical scores at neutral, with no difference in ASES. Observational studies of measured inclination, baseplate version and overhang found no difference in patient-reported scores. Meta-analyses of design categories and of lateralized glenospheres found no difference in ASES or Constant; the lateralized glenospheres gave a Subjective Shoulder Value about 6 points higher.
Reported, but confounded: better scores after a surgeon changed to a lateralized 135° design or to individualized version, and fewer acromial fractures after a surgeon began medializing the glenoid in high-risk patients. In each, later patients were compared with earlier ones.
Associated with complications in one or a few retrospective series: superior baseplate inclination and an unrepaired subscapularis with instability; an acromion closer to the glenosphere center than the greater tuberosity with acromial and scapular spine fracture. The strongest predictors of fracture remain the patient's own: bone quality, inflammatory arthritis, female sex and cuff deficiency.
Inferred, not yet demonstrated: that any particular reverse geometry, "anatomic" or otherwise, makes patients better. The observational associations with notching and fracture are real but conflicting, and they come mostly from catalog categories rather than measured position. Placing the baseplate more accurately has not yet been shown to help either.
How important is RSA geometry to the outcome realized by the patient?
Cooper's hawk
Follow on twitter: https://twitter.com/shoulderarth
Follow on facebook: https://www.facebook.com/frederick.matsen
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/
References
1. Wong LH, Lurie BM, Davies MR. Reverse total shoulder arthroplasty in the setting of an intact rotator cuff: a contemporary review of current indications, outcomes, and surgical considerations. Curr Rev Musculoskelet Med. 2026;19:70. doi
2. Berton A, Longo UG, Gulotta LV, De Salvatore S, Piergentili I, Calabrese G, Roberti F, Warren RF, Denaro V. Humeral and glenoid version in reverse total shoulder arthroplasty: a systematic review. J Clin Med. 2022;11:7416. doi
3. Southam BR, Bedeir YH, Johnson BM, Hasselfeld KA, Kloby MA, Grawe BM. Clinical and radiological outcomes in lateralized versus nonlateralized and distalized glenospheres in reverse total shoulder arthroplasty: a randomized control trial. J Shoulder Elbow Surg. 2023;32:1420-1431. doi
4. King JJ, Hones KM, Wright TW, Roche C, Zuckerman JD, Flurin PH, Schoch BS. Does isolated glenosphere lateralization affect outcomes in reverse shoulder arthroplasty? Orthop Traumatol Surg Res. 2023;109(4):103401. doi
5. Nunes B, Linhares D, Costa F, Neves N, Claro R, Silva MR. Lateralized versus nonlateralized glenospheres in reverse shoulder arthroplasty: a systematic review with meta-analysis. J Shoulder Elbow Surg. 2021;30:1700-1713. doi
6. Harmsen SM, Robaina J, Campbell D, Denard PJ, Gobezie R, Lederman ES. Does lateralizing the glenosphere center of rotation by 4 mm decrease scapular notching in reverse shoulder arthroplasty with a 135° humeral component? JSES Int. 2022;6(3):442-446. PMC
7. Wiater JM, Lee JYJ, Shields EJW, Childers K, Dery L, Koueiter D. Humeral component version has no effect on outcomes following reverse total shoulder arthroplasty: a prospective, double-blinded, randomized controlled trial. J Bone Joint Surg Am. 2024;106(13):1181-1188. doi
8. Ventura-Parellada C, Goñalons-Giol F, Alonso-Rodriguez Piedra J, Gamez-Baños F, Jovell-Fernandez E, Mora-Guix JM. Humeral component retroversion in the Comprehensive Reverse Shoulder System arthroplasty: rotations, clinical outcomes, and quality-of-life analysis in a prospective randomized study. J Shoulder Elbow Surg. 2026;35:2-10. doi
9. Gobezie R, Shishani Y, Lederman E, Denard PJ. Can a functional difference be detected in reverse arthroplasty with 135° versus 155° prosthesis for the treatment of rotator cuff arthropathy: a prospective randomized study. J Shoulder Elbow Surg. 2019;28:813-818. doi
10. Neyton L, Nigues A, McBride AP, Giovannetti de Sanctis E. Neck shaft angle in reverse shoulder arthroplasty: 135 vs. 145 degrees at minimum 2-year follow-up. J Shoulder Elbow Surg. 2023;32:1486-1493. doi
11. Oh JH, Sharma N, Rhee SM, Park JH. Do individualized humeral retroversion and subscapularis repair affect the clinical outcomes of reverse total shoulder arthroplasty? J Shoulder Elbow Surg. 2020;29:821-829. doi
12. Edwards TB, Trappey GJ, Riley C, O'Connor DP, Elkousy HA, Gartsman GM. Inferior tilt of the glenoid component does not decrease scapular notching in reverse shoulder arthroplasty: results of a prospective randomized study. J Shoulder Elbow Surg. 2012;21:641-646. doi:10.1016/j.jse.2011.08.057. PubMed
13. Mahendraraj KA, Shields MV, Grubhofer F, Golenbock SW, Jawa A. Reassessing glenoid inclination in reverse total shoulder arthroplasty with glenosphere lateralization. Bone Joint J. 2021;103-B(2):360-365. doi
14. Sheth MM, Schiffman CJ, Whitson AJ, Matsen FA III, Hsu JE. Baseplate version in reverse shoulder arthroplasty: does excessive retroversion or anteversion affect functional activities of daily living? Int Orthop. 2025;49:1119-1125. doi
15. Dean EW, Dean NE, Wright TW, Hao KA, Schoch BS, Farmer KW, Struk AM, King JJ. Clinical outcomes related to glenosphere overhang in reverse shoulder arthroplasty using a lateralized humeral design. J Shoulder Elbow Surg. 2022;31:2106-2115. doi
16. Imiolczyk JP, Imiolczyk T, Góralczyk A, Scheibel M, Freislederer F. Lateralization and distalization shoulder angles do not predict outcomes in reverse shoulder arthroplasty for cuff tear arthropathy. J Shoulder Elbow Surg. 2024;33:121-129. doi
17. Kirkham MS, Guy CR, Puzzitiello RN, Staten TM, Christy KB, Zhang C, et al. Do radiographic measurements of glenoid or humeral lateralization and distalization affect post-operative outcomes for reverse total shoulder arthroplasty? J Shoulder Elbow Surg. 2026;35:2336-2345. doi
18. Freislederer F, Moroder P, Audigé L, Schneller T, Ameziane Y, Trefzer R, Imiolczyk JP, Scheibel M. Analysis of three different reverse shoulder arthroplasty designs for cuff tear arthropathy – the combination of lateralization and distalization provides best mobility. BMC Musculoskelet Disord. 2024;25:204. doi
19. Imiolczyk JP, Audigé L, Freislederer F, Schneller T, Ameziane Y, Touet A, Scheibel M. Clinical and radiological comparison of three different reverse shoulder arthroplasty designs for patients with primary osteoarthritis. BMC Musculoskelet Disord. 2025;26:577. doi
20. Zitkovsky HS, Carducci MP, Mahendraraj KA, Grubhofer F, Jawa A. Lateralization and decreased neck-shaft angle reduces scapular notching and heterotopic ossification. J Am Acad Orthop Surg. 2020;28:e1073-e1080. doi
21. Burden EG, Batten TJ, Smith CD, Evans JP. Reverse total shoulder arthroplasty: a systematic review and meta-analysis of complications and patient outcomes dependent on prosthesis design. Bone Joint J. 2021;103-B(5):813-821. doi
22. Charles SJ-C, LeVasseur C, Rai A, Kane G, Munsch M, Hughes J, Anderst W, Lin A. Anatomic restoration of lateral humeral offset and humeral retroversion optimizes functional outcomes following reverse total shoulder arthroplasty. J Biomech. 2024;176:112372. doi
23. 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(5):e114-e119. Link
24. Dhillon G, Warren M, Assiotis A, et al. Reverse shoulder arthroplasty humeral lateralisation: a systematic review. Cureus. 2021;13(11):e19845. doi
25. Vicary-Watts RJ, Bayne JR, Dom K, Grant AL, Wilkinson MPR, Morse LP. Glenoid preparation in shoulder arthroplasty, is patient-specific instrumentation more accurate than standard instrumentation? A systematic review and meta-analysis. J Shoulder Elb Arthroplast. 2026;100116 (pre-proof). doi
26. Shekhbihi A, Raiss P, Modelhart M, Randlkofer LM, Herbst EC, Moroder P. The global glenoid component inclination: why scapulothoracic orientation should be considered when defining glenoid component inclination in reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2026 (online ahead of print). doi
27. Galvin JW, Kim R, Ment A, Durso J, Joslin PMN, Lemos JL, Novikov D, Curry EJ, Alley MC, Parada SA, Eichinger JK, Li X. Outcomes and complications of primary reverse shoulder arthroplasty with minimum of 2 years' follow-up: a systematic review and meta-analysis. J Shoulder Elbow Surg. 2022;31:e534-e544. doi
28. Kriechling P, Zaleski M, Loucas R, Loucas M, Fleischmann M, Wieser K. Complications and further surgery after reverse total shoulder arthroplasty: report of 854 primary cases. Bone Joint J. 2022;104-B(3):401-407. doi
29. Pak T, Kilic AI, Ardebol J, Gobezie R, Sears BW, Lederman E; ShARC; Werner BC, Denard PJ. Glenoid-sided lateralization decreases scapular notching with a 135° humeral component Arthrex reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2026;35:e224-e232. doi
30. Shelley RJ, DeFoor MT, Parada SA, Crosby LA. Clinical implications of scapular notching at 2 and 5-year follow-up after reverse total shoulder arthroplasty. J Orthop. 2020;21:384-389. Link
31. Moverman MA, Puzzitiello RN, Glass EA, et al., Jawa A; ASES Complications of RSA Multicenter Research Group. Implant-positioning and patient factors associated with acromial and scapular spine fractures after reverse shoulder arthroplasty. J Bone Joint Surg Am. 2024;106:1384-1394. doi
32. Pak T, Ardebol J, Menendez ME, Gobezie R, Sears BW, Lederman E; ShARC group; Werner BC, Denard PJ. Robert H. Cofield, MD, Award for Best Oral Presentation 2023: up to 8 mm of glenoid-sided lateralization does not increase the risk of acromial or scapular spine stress fracture following reverse shoulder arthroplasty with a 135° inlay humeral component. J Shoulder Elbow Surg. 2024;33:S1-S8. doi
33. Schenk P, Aichmair A, Beeler S, Ernstbrunner L, Meyer DC, Gerber C. Acromial fractures following reverse total shoulder arthroplasty: a cohort controlled analysis. Orthopedics. 2020;43(1):15-22. PubMed
34. Werthel JD, Schoch BS, van Veen SC, Elhassan BT, An KN, Cofield RH, Sperling JW. Acromial fractures in reverse shoulder arthroplasty: a clinical and radiographic analysis. J Shoulder Elb Arthroplast. 2018;2:2471549218777628. doi
35. Eisenberg MT, Hui C, Schwartz K, Qubain L, Amini MH. In reverse shoulder arthroplasty, selectively medializing the glenoid in higher-risk patients is associated with reduced risk of acromial stress fracture. J Shoulder Elbow Surg. 2026;35:2354-2361. doi
36. Kawashima I, King JJ, Wright JO, Struk AM, Banks SA, Wright TW. Shoulder geometry after reverse total shoulder arthroplasty with a medialized glenoid and a lateralized humerus predicts subacromial notching and acromial or scapular spine fractures. J Shoulder Elbow Surg. 2026;35:40-48. doi
37. Haidamous G, Lädermann A, Frankle MA, Gorman RA II, Denard PJ. The risk of postoperative scapular spine fracture following reverse shoulder arthroplasty is increased with an onlay humeral stem. J Shoulder Elbow Surg. 2020;29:2556-2563. doi
38. Kriechling P, Weber F, Karczewski D, Borbas P, Wieser K. Predictive factors of acromial fractures following reverse total shoulder arthroplasty: a subgroup analysis of 860 shoulders. JSES Int. 2023;7:812-818. doi
39. Polisetty T, Cannon D, Grewal G, Vakharia R, Levy JC. Radiographic and anatomic variations on postoperative acromion fractures after inlay and lateralized reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2023;32:76-81. doi
40. Tashjian RZ, Martin BI, Ricketts CA, Henninger HB, Granger EK, Chalmers PN. Superior baseplate inclination is associated with instability after reverse total shoulder arthroplasty. Clin Orthop Relat Res. 2018;476:1622-1629. doi
41. Ulenberg N, Herbstreit S, Dudda M, Vogel C, Mester B. Risk factors for instability and dislocation after reverse total shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2026 (in press). doi
42. Baek CH, Lim C, Kim JG, Kim BT, Kim SJ. Comparative outcomes of anatomic and reverse total shoulder arthroplasty for rotator cuff-intact glenohumeral osteoarthritis: a propensity score-matched, age-stratified exploratory analysis. J Shoulder Elb Arthroplast. 2026;10:100050. doi
References checked against the full papers or against publisher and PubMed records, November 2026.
