Abstract
Background
Two classes of primary reverse total shoulder arthroplasty (rTSA), inlay (in-rTSA), and onlay (on-rTSA) were compared to determine differences in rates of revision.
Methods
Between 1 January 2012 and 31 December 2020, all primary in-rTSA or on-rTSA procedures were compared from a large national arthroplasty registry by cumulative percentage revision (CPR). Kaplan–Meier estimates of survivorship and hazard ratios from Cox proportional hazard models adjusted for age, gender, glenosphere size, and humeral fixation determined any associations to the risk of revision.
Results
Of the 14,807 in-rTSA and 6590 on-rTSA procedures, the CPR at seven years was 4.9%. There was an increased risk of revision for in-rTSA vs on-rTSA (p = 0.039) when adjusted for age, gender, glenosphere size, and humeral fixation. Glenosphere size <38 mm adjusted for age and gender (p = 0.016) increased the revision risk. Revision for instability/dislocation occurred more often for in-rTSA vs on-rTSA (p < 0.001) in the first three months. Males had a higher rate of revision than females for in-rTSA (3months+, p = 0.001) and for on-rTSA (p < 0.001).
Discussion
Care should be taken when considering in-rTSA particularly in males, and if preoperative planning suggests a small (<38 mm) glenosphere.
Level of evidence
Level III, therapeutic study. Original article.
Keywords: primary shoulder arthroplasty, reverse shoulder arthroplasty, inlay reverse shoulder arthroplasty, onlay reverse shoulder arthroplasty
Introduction
Reverse total shoulder arthroplasty (rTSA) is now the most common shoulder joint replacement undertaken in Australia. 1 Part of the advantage of rTSA is the wider range of indications compared to anatomic shoulder arthroplasty. 2 Recent results of modern modular implants are encouraging. 3 The lateralisation of the humeral component was popularised in North America in 2004, compared to the medialised Grammont reverse design.2,4 Since then, rTSA has been classified based on the centre of rotation or the position of the humerus relative to the glenosphere. 5 Recently, rTSA categorisation has recognised the proximal humeral metaphyseal component with the associated neck angle changes. 6
The rTSA proximal humeral component metaphyseal region has two designs: inlay, where the metaphyseal component lies deep into the epiphyseal osteotomy; or onlay, where the metaphyseal component lies on top of the epiphyseal osteotomy. An inlay reverse shoulder arthroplasty (in-rTSA) gives a theoretical advantage of improved range of motion (external and internal rotation) and the potential disadvantage of scapular notching. 7 Onlay reverse shoulder arthroplasty (on-rTSA) provides potential gains in shoulder elevation but there is an increased risk of spine of scapular fracture.8–10
Clinical studies have not reported a difference in revision rates between these two humeral designs.8,11 A comparative observational study of a large number of rTSAs from a national registry may provide further insights. The aim of this study was to compare the rate of revision for in-rTSA and on-rTSA using data from a large arthroplasty registry.
Materials and methods
The Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) began data collection on 1 September 1999 for hip and knee and since April 2004 for shoulder procedures. The Registry has documented almost all shoulder arthroplasty procedures Australia-wide since November 2007. These data are externally validated against patient-level data provided by all Australian state and territory health departments. The external and internal validation process has been well documented.1,12–14 A sequential, multilevel matching process is used to identify any missing data which are subsequently obtained by follow-up with the relevant hospital. Each month, in addition to internal validation and data quality checks, all primary procedures are linked to any subsequent revision involving the same patient, joint, and side. Data are also matched biannually to the Australian National Death Index data to control for mortality.
The study population included all primary rTSA procedures undertaken for any diagnosis except fracture (to remove varying outcomes with different tuberosity fracture management) that incorporated either an inlay or onlay modular metaphyseal humeral component with any reverse shoulder arthroplasty implant type. Those undertaken with extended fixation peg glenoid components (to remove Bio-reverse shoulder arthroplasty procedures) or monoblock humeral components were excluded. Given on-rTSA became available nationally in 2011, we determined an appropriate study period was from 1 January 2012 to 31 December 2020 ensuring similar implant technologies. The analysis was adjusted for age, gender, glenosphere size, and humeral fixation. The analyses were repeated with the SMR L2 prosthesis (Lima Corporate, San Daniele del Friuli, Italy) removed from the in-rTSA cohort. The SMR L2 was withdrawn from the Australian market during 2012 due to higher than anticipated rates of revision. 1 To further investigate the effect of prosthesis class on revision, an additional analysis was performed restricting to prostheses used in more than 250 procedures for both in-rTSA and on-rTSA to reflect commonly utilised implants in our country.
The two most common primary diagnoses osteoarthritis, and rotator cuff arthropathy were analysed to determine whether diagnosis was associated with a risk of revision for in-rTSA compared to on-rTSA. This ensured >300 cases in each diagnosis category for both cohort groups. Further age and gender-adjusted analyses on the effects of glenosphere size, reasons for revision, humeral fixation, and gender were compared between in-rTSA and on-rTSA.
Statistical analysis
Kaplan–Meier estimates of survivorship were used to report the time to revision, with censoring at the time of death and closure of the dataset at the end of December 2020. The unadjusted cumulative percent revision (CPR), with 95% confidence intervals (CIs), was calculated using unadjusted point-wise Greenwood estimates. Age and gender-adjusted hazard ratios (HR) were calculated from Cox proportional hazard models to compare the rate of revision between groups. Additional adjustments for glenosphere size and humeral fixation were made on the primary comparison of in-TSA and on-rTSA and the prostheses effect sub-analyses. The assumption of proportional hazards was checked analytically for each model. If the interaction between the predictor and the log of time was statistically significant in the standard Cox model, then a time-varying model was estimated. Time points were selected based on the greatest change in hazard, weighted by a function of events. Time points were iteratively chosen until the assumption of proportionality was met, and HRs were calculated for each selected time period. For the current study, if no time period was specified, the HR was calculated over the entire follow-up period. All tests were two-tailed at 5% levels of significance. Statistical analysis was performed using SAS software version 9.4 (SAS Institute Inc., Cary, North Carolina).
Results
There were 21,397 rTSA procedures, of which 14,807 were in-rTSA and 6590 were on-rTSA. The proportion of males and females in the two groups was similar. The mean patient age for in-rTSA and on-rTSA were also similar. Patient body mass index (BMI) and American Society of Anaesthesiologists – Physical Status Classification (ASA score) were similar in both groups (Table 1). The type of humeral fixation and glenosphere sizes is included with patient characteristics in Table 1. Those diagnoses with over 300 cases of both in-rTSA and on-rTSA were osteoarthritis and rotator cuff arthropathy. There were 7497 (50.6%) in-rTSA, 3561 (54%) cases of osteoarthritis, and rotator cuff arthropathy as the primary diagnosis in 6557 (44.3%) in-rTSA and 2714 (41.2%) on-rTSA.
Table 1.
Patient and prosthesis characteristics of cohort groups in-rTSA and on-rTSA.
| Variable | Inlay | Onlay | Total |
|---|---|---|---|
| Follow-up years | |||
| Mean ± SD | 3.7 ± 2.3 | 2 ± 1.6 | 3.2 ± 2.3 |
| Median (IQR) | 3.5 (1.7, 5.4) | 1.7 (0.7, 2.9) | 2.7 (1.3, 4.7) |
| Minimum | 0 | 0 | 0 |
| Maximum | 9 | 8.9 | 9 |
| Age | |||
| Mean ± SD | 73.9 ± 7.8 | 73.4 ± 7.9 | 73.7 ± 7.9 |
| Median (IQR) | 74 (69, 79) | 74 (69, 79) | 74 (69, 79) |
| Age group | |||
| <55 | 205 (1.4%) | 106 (1.6%) | 311 (1.5%) |
| 55−64 | 1458 (9.8%) | 692 (10.5%) | 2150 (10%) |
| 65−74 | 5859 (39.6%) | 2743 (41.6%) | 8602 (40.2%) |
| ≥75 | 7285 (49.2%) | 3049 (46.3%) | 10,334 (48.3%) |
| Gender | |||
| Male | 5983 (40.4%) | 2808 (42.6%) | 8791 (41.1%) |
| Female | 8824 (59.6%) | 3782 (57.4%) | 12,606 (58.9%) |
| ASA score a | |||
| 1 | 402 (3.1%) | 192 (3%) | 594 (3%) |
| 2 | 5691 (43.6%) | 2784 (43%) | 8475 (43.4%) |
| 3 | 6608 (50.6%) | 3345 (51.7%) | 9953 (51%) |
| 4 | 344 (2.6%) | 152 (2.3%) | 496 (2.5%) |
| 5 | 3 (0%) | 1 (0%) | 4 (0%) |
| BMI b | |||
| Underweight (<18.50) | 66 (0.7%) | 42 (0.7%) | 108 (0.7%) |
| Normal (18.50−24.99) | 1678 (16.6%) | 976 (16.4%) | 2654 (16.5%) |
| Pre Obese (25.00−29.99) | 3500 (34.7%) | 2053 (34.5%) | 5553 (34.6%) |
| Obese Class 1 (30.00−34.99) | 2839 (28.1%) | 1601 (26.9%) | 4440 (27.7%) |
| Obese Class 2 (35.00−39.99) | 1302 (12.9%) | 852 (14.3%) | 2154 (13.4%) |
| Obese Class 3 (≥40.00) | 711 (7%) | 428 (7.2%) | 1139 (7.1%) |
| Humeral Fixation | |||
| Cementless | 13,809 (93.3%) | 6222 (94.4%) | 20,031 (93.6%) |
| Cemented | 998 (6.7%) | 368 (5.6%) | 1366 (6.4%) |
| Glenosphere Size c | |||
| <38 mm | 3348 (22.6%) | 3003 (48.6%) | 6351 (30.3%) |
| 38–40 mm | 7139 (48.3%) | 1804 (29.2%) | 8943 (42.6%) |
| >40 mm | 4300 (29.1%) | 1378 (22.3%) | 5678 (27.1%) |
| Total | 14,807 | 6590 | 21,397 |
SD: standard deviation; IQR: interquartile range; ASA: American Society of Anesthesiologists; BMI: body mass index (kg/m2).
Excludes 1875 procedures with unknown ASA score.
Excludes 5349 procedures with unknown BMI.
Excludes 425 procedures with unknown Glenosphere size.
When considering all procedures irrespective of primary diagnosis, there were 530 (3.8%) primary in-rTSA procedures and 158 (2.4%) primary on-rTSA procedures revised during the study period. The CPR at seven years was 4.9% (95% CI 4.4, 5.4) for in-rTSA and 4.9% (95% CI 3.3, 7.2) for on-rTSA (Table 2). There was a higher rate of revision for in-rTSA compared to on-rTSA (HR = 1.22 (95% CI 1.01, 1.47), p = 0.039 adjusting for age, gender, glenosphere size and humeral fixation) (Figure 1). Removing the SMR L2 prosthesis did not change the difference in the revision rate (HR = 1.21 (95% CI 1.00, 1.46), p = 0.050). Limiting the analysis to prostheses with >250 procedures in both in-rTSA and on-rTSA classes also did not change the difference in the revision rate (HR = 1.24 (95% CI 1.02, 1.51), p = 0.030).
Table 2.
CPR of in-rTSA and on-rTSA (all diagnoses).
| CPR | 1 year | 2 years | 3 years | 4 years |
|---|---|---|---|---|
| Inlay | 2.3 (2.1, 2.6) | 3.0 (2.7, 3.3) | 3.5 (3.2, 3.8) | 3.8 (3.5, 4.2) |
| Onlay | 1.8 (1.5, 2.2) | 2.7 (2.3, 3.2) | 3.0 (2.5, 3.5) | 3.8 (3.1, 4.6) |
| CPR | 5 years | 6 years | 7 years | 8 years |
| Inlay | 4.2 (3.8, 4.6) | 4.6 (4.1, 5.0) | 4.9 (4.4, 5.4) | 5.1 (4.6, 5.7) |
| Onlay | 4.0 (3.3, 4.9) | 4.0 (3.3, 4.9) | 4.9 (3.3, 7.2) |
CPR: cumulative percentage revision.
Figure 1.
Cumulative percentage revision (CPR) of in-rTSA and on-rTSA (all diagnoses).
Patient gender was analysed for both cohort groups. Males with in-rTSA had a higher rate of revision compared to males with on-rTSA for the first three months only adjusted for age (HR = 1.60 (95% CI 1.19, 2.15), p = 0.001) with no difference after that time. Males with in-rTSA had a higher rate of revision compared to females with in-rTSA for the entire period (0–3 months HR = 2.60 (2.00, 3.40), p < 0.001; three months + HR = 1.40 (95% CI 1.14, 1.73), p = 0.001). Males with on-rTSA also had a higher rate of revision compared to females for on-rTSA (Entire period HR = 2.01 (95% CI 1.46, 2.77), p < 0.001). When the outcome for females were compared, there was no difference between in-rTSA and on-rTSA revision rates (Figure 2).
Figure 2.
Cumulative percentage revision (CPR) of primary reverse shoulder arthroplasty by class of primary and gender.
Reverse shoulder arthroplasties with glenosphere sizes <38 mm were associated with an increased risk of revision for in-rTSA compared to on-rTSA adjusted for age and gender (HR = 1.44 (95% CI 1.07, 1.94), p = 0.016), but there was no difference in the rate of revision for glenospheres sized 38–40 mm or >40 mm. Revision for instability/dislocation was not associated with the smallest (<38 mm) glenosphere size.
Revision diagnoses with a greater than 10% frequency for both in-rTSA and on-rTSA revision included; instability/dislocation (29.2% and 27.8%), infection (25.5% and 24.7%), loosening (18.5% and 17.1%), and fracture (11.7% and 12.7%, respectively). In-rTSA had a higher rate of revision for instability/dislocation compared to on-rTSA for the first three months only adjusting for age and gender (HR = 2.24 (95% CI 1.39, 3.63), p < 0.001) with no difference after this time. In addition, the revision risk for instability/dislocation was higher for the first three months period with the removal of SMR L2 prosthesis (HR = 2.27 (95% CI 1.41, 3.68), p < 0.001) and when limiting the analysis to prostheses with >250 procedures in both in-rTSA an on-rTSA (HR = 2.19 (95% CI 1.34, 3.58), p = 0.001).
When compared, there was no difference in the rate of revision of in-rTSA and on-rTSA undertaken for osteoarthritis (HR = 1.23 (95% CI 0.95, 1.58), p = 0.117) or rotator cuff arthropathy (HR = 1.18 (95% CI 0.90, 1.56), p = 0.231) adjusted for age and gender. The type of humeral fixation was not associated with revision risk for either in-rTSA or on-rTSA.
Discussion
The in-rTSA has been the most prevalent reverse shoulder arthroplasty design in Australia, as evidenced by this study. Despite this popularity, we found an increased risk of revision of in-rTSA compared to on-rTSA as measured by CPR over the entire study period. Our study also found excluding SMR L2 (Lima Corporate, San Daniele del Friuli, Italy), or including the most commonly used prostheses (>250 procedures) did not confound the increased rate of revision. Kenon et al. compared 100 rTSA of either glenoid lateralised or medialised type. They found no difference in outcome measures (except increased forward elevation for medialised type) and complication rates, but revision rate was not included. 11 Whilst Merolla et al. found when comparing 74 shoulders of either Grammont design to onlay type that clinical scores and complications were similar, but the onlay type had lower scapular notching and higher external rotation. 8 Scapular spine fractures were reported in 4.3% of 485 onlay type reverse shoulder arthroplasty by comparison. 15 On modelling range of motion with differing onlay tray inclinations compared to a Grammont design, Lädermann et al. found increasing tray inclination improved adduction, extension, and external rotation most of all. 16 None of these studies suggested that there were differences in revision risk, perhaps because of small sample sizes and relatively short follow-up.
The revision rates of both cohorts were consistent with other registry series reporting primary rTSA.12,17,18 Despite this there may be explanatory variables that may separate the outcome between the two classes of reverse shoulder arthroplasty. Revision for instability/dislocation was more likely to occur in the first three months after primary reverse shoulder arthroplasty surgery for in-rTSA compared to on-rTSA. Instability/dislocation has been a known risk factor for primary anatomic shoulder arthroplasty 12 and revision shoulder arthroplasty. 13 Glenosphere size increased the risk of revision for in-rTSA compared to on-rTSA. The smaller glenosphere also increased the likelihood of instability/dislocation. It has been previously noted that smaller glenosphere size increased revision rates of rTSA for primary osteoarthritis but not rotator cuff arthropathy. 14 We note that assessing the cohort without the SMR L2 (Lima Corporate, San Daniele del Friuli, Italy) or with only the most commonly used prostheses (>250 procedures) did not confound the revision risk for instability/dislocation. The relationship between instability/dislocation, small glenosphere size, and revision risk appears to suggest tension length susceptibility or adduction impingement for in-rTSA and would expectedly present early rather than after years from primary surgery.
There have been other described influences on outcomes for primary reverse shoulder arthroplasty. Age and gender have been previously identified as risk factors for the revision of primary reverse shoulder arthroplasty. 19 There have also been clinical effects with improvements in patient reported outcome measures (PROMs) in males compared to females.20–22 We statistically adjusted for age and gender to exclude confounding effects, but we note being male was associated with revision risk in both in-rTSA and on-rTSA. In addition, cementless humeral components have not been shown to influence the outcome of reverse replacements patients, 9 and we noted the number of cemented humeral implants was below 15% in both cohorts. Primary diagnosis has been noted as a revision risk for 1st revision rTSA but not in this study of primary reverse shoulder arthroplasty. 13
This study has limitations and strengths. The two groups had differing sample sizes but similar length of observation, indicating short to medium-term differences only. It is a registry-based observational study (which cannot determine causation) that only considers revision as the end point rather than clinical performance such as range of motion or relief of pain as PROMs were not included. However, this study includes all practising orthopaedic surgeons in our country, and all implants enrolled in the two cohorts, indicative of real-world practise and revision. The observations represent the overall associations not the best possible for any given diagnosis or prosthetic type. 3 The decision for revision is a complex patient–surgeon process and cannot be accounted for in a registry analysis. Groups were well matched for BMI, ASA, and comparisons were adjusted for age, gender, glenosphere size, and humeral fixation but other unknown factors may be associated with revision rates.
Conclusion
In this large comparative registry series, the CPR of in-rTSA and on-rTSA was <5% at seven years after primary surgery. However, there was an increased risk of revision of in-rTSA compared to on-rTSA over the entire study period. Glenosphere size and gender were associated with revision risk. There was an increased revision rate for instability/dislocation for in-rTSA compared to on-rTSA in the first three months after primary rTSA. We advise, on the basis of this study, there should be caution undertaking in-rTSA, especially in men, particularly if preoperative planning suggests utilising a small (<38 mm) glenosphere.
Acknowledgements
The authors thank the AOANJRR staff, orthopaedic surgeons, hospitals, and patients whose data made this work possible.
Footnotes
Authors’ Note: IRB approval: (QAA14/2016) Declaration under Part VC of the Health Insurance Act 1973. Previously presented in part to the AOA ASM virtual Meeting in November 2021.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethics: The AOANJRR is approved by the Australian Federal Government as a declaration of quality assurance activity under Section 124X of the Australian Federal Health Insurance Act 1973. All investigations were conducted in accordance with the ethical principles of research (The Helsinki Declaration II). The AOANJRR is funded by the Commonwealth of Australia Department of Health and Ageing.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The AOANJRR is funded by the Commonwealth of Australia Department of Health and Ageing.
ORCID iD: David R J Gill https://orcid.org/0000-0002-0023-630X
References
- 1.Australian Orthopaedic Association National Joint Replacement Registry. Hip, Knee & Shoulder Arthroplasty: Annual Report 2019 , https://aoanjrr.sahmri.com/annual-reports-2019 (accessed 14 November 2021).
- 2.Boileau P, Watkinson D, Hatzidakis AM, et al. Neer award 2005: the Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J Shoulder Elbow Surg 2006; 15: 527–540. [DOI] [PubMed] [Google Scholar]
- 3.Kang JR, Dubiel MJ, Cofield RH, et al. Primary reverse shoulder arthroplasty using contemporary implants is associated with very low reoperation rates. J Shoulder Elbow Surg 2019; 28: S175–S180. [DOI] [PubMed] [Google Scholar]
- 4.Cuff D, Pupello D, Virani N, et al. Reverse shoulder arthroplasty for the treatment of rotator cuff deficiency. J Bone Joint Surg 2008; 90: 1244–1251. [DOI] [PubMed] [Google Scholar]
- 5.Routman HD, Flurin P-H, Wright TW, et al. Reverse shoulder arthroplasty prosthesis design classification system. Bull Hosp Jt Dis 2015; 73: S5–14. [PubMed] [Google Scholar]
- 6.Beltrame A, Di Benedetto P, Cicuto C, et al. Onlay versus inlay humeral steam in reverse shoulder arthroplasty (RSA): clinical and biomechanical study. Acta Biomed 2019; 90: 54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ernstbrunner L, Andronic O, Grubhofer F, et al. Long-term results of reverse total shoulder arthroplasty for rotator cuff dysfunction: a systematic review of longitudinal outcomes. J Shoulder Elbow Surg 2019; 28: 774–781. [DOI] [PubMed] [Google Scholar]
- 8.Merolla G, Walch G, Ascione F, et al. Grammont humeral design versus onlay curved-stem reverse shoulder arthroplasty: comparison of clinical and radiographic outcomes with minimum 2-year follow-up. J Shoulder Elbow Surg 2018; 27: 701–710. [DOI] [PubMed] [Google Scholar]
- 9.Phadnis J, Huang T, Watts A, et al. Cemented or cementless humeral fixation in reverse total shoulder arthroplasty? A systematic review. Bone Joint J 2016; 98: 65–74. [DOI] [PubMed] [Google Scholar]
- 10.Simovitch RW, Friedman RJ, Cheung EV, et al. Rate of improvement in clinical outcomes with anatomic and reverse total shoulder arthroplasty. J Bone Joint Surg 2017; 99: 1801–1811. [DOI] [PubMed] [Google Scholar]
- 11.Kennon JC, Songy C, Bartels D, et al. Primary reverse shoulder arthroplasty: how did medialized and glenoid-based lateralized style prostheses compare at 10 years? J Shoulder Elbow Surg 2020; 29: S23–S31. [DOI] [PubMed] [Google Scholar]
- 12.Gill DR, Graves SE, Rainbird S, et al. A comparison of revision rates for osteoarthritis of primary reverse total shoulder arthroplasty to primary anatomic shoulder arthroplasty with a cemented all-polyethylene glenoid: analysis from the Australian Orthopaedic Association National Joint Replacement Registry. Clin Orthop Relat Res 2021; 479: 2216–2224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gill DR, Page RS, Graves SE, et al. The rate of 2nd revision for shoulder arthroplasty as analyzed by the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Acta Orthop 2021; 92: 258–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Page R, Beazley J, Graves S, et al. Effect of glenosphere size on reverse shoulder arthroplasty revision rate: an analysis from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). J Shoulder Elbow Surg 2022. Jun;31(6): e289–e301. [DOI] [PubMed] [Google Scholar]
- 15.Ascione F, Kilian CM, Laughlin MS, et al. Increased scapular spine fractures after reverse shoulder arthroplasty with a humeral onlay short stem: an analysis of 485 consecutive cases. J Shoulder Elbow Surg 2018; 27: 2183–2190. [DOI] [PubMed] [Google Scholar]
- 16.Lädermann A, Denard PJ, Boileau P, et al. Effect of humeral stem design on humeral position and range of motion in reverse shoulder arthroplasty. Int Orthop 2015; 39: 2205–2213. [DOI] [PubMed] [Google Scholar]
- 17.Fevang B-TS, Lie SA, Havelin LI, et al. Risk factors for revision after shoulder arthroplasty: 1,825 shoulder arthroplasties from the Norwegian arthroplasty register. Acta Orthop 2009; 80: 83–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lehtimäki K, Rasmussen JV, Mokka J, et al. Risk and risk factors for revision after primary reverse shoulder arthroplasty for cuff tear arthropathy and osteoarthritis: a nordic arthroplasty register association study. J Shoulder Elbow Surg 2018; 27: 1596–1601. [DOI] [PubMed] [Google Scholar]
- 19.Craig RS, Lane JC, Carr AJ, et al. Serious adverse events and lifetime risk of reoperation after elective shoulder replacement: population based cohort study using hospital episode statistics for England. Br Med J 2019; 364:l298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Friedman RJ, Cheung EV, Flurin P-H, et al. Are age and patient gender associated with different rates and magnitudes of clinical improvement after reverse shoulder arthroplasty? Clin Orthop Relat Res 2018; 476: 1264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Leathers MP, Ialenti MN, Feeley BT, et al. Do younger patients have better results after reverse total shoulder arthroplasty? J Shoulder Elbow Surg 2018; 27: S24–S28. [DOI] [PubMed] [Google Scholar]
- 22.Wong SE, Pitcher AA, Ding DY, et al. The effect of patient gender on outcomes after reverse total shoulder arthroplasty. J Shoulder Elbow Surg 2017; 26: 1889–1896. [DOI] [PubMed] [Google Scholar]