JSES - 2026-05-12 - Journal Article
Surgical Options for the Management of Severe Glenoid Structural Deficiencies in the Setting of Revision Reverse Shoulder Arthroplasty: A Systematic Review.
Henke RT, Ries RJ, Tobin JG, Rogalski BL, Friedman RJ, Li X, Eichinger JK
Topics
Key Takeaway
In revision RSA with severe glenoid defects, custom implants achieved MCID in 100% of outcome measures versus bone graft constructs that failed MCID for external rotation (67%) and internal rotation (50%), though complication rates were 19% vs 28% respectively.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This systematic review compared bone grafting with conventional implants versus custom implants for severe glenoid bone deficiency in revision RSA, analyzing 232 shoulders from 13 studies. Bone graft cohorts achieved MCID for ASES, Constant, SST, abduction, and forward elevation, but only 67% and 50% achieved MCID for external and internal rotation, respectively; custom implants achieved MCID across all measures. Complication rates were 19% (custom) vs 28% (bone graft), with re-revision rates similar at 52% vs 48%; femoral head allograft drove 85% of re-revisions in the bone graft subgroup.
Key Limitation
The analysis is derived predominantly from Level IV case series with no direct head-to-head comparison, making it impossible to control for glenoid defect size, classification, or bone quality across cohorts.
Original Abstract
BACKGROUND
Over the last decade, due to the increase in the number of shoulder arthroplasty procedures performed, revision to reverse total shoulder arthroplasty (rTSA) has increased by 392%. This rise has been associated with a greater incidence of complications, including implant loosening and bone loss requiring alternative strategies for management of failed arthroplasty with glenoid bone deficiency. The purpose of this paper is to review the current literature and compare outcomes of different grafts and custom implants used to address severe glenoid defects for revision of failed shoulder arthroplasty.
METHODS
A systematic search of three databases (PubMed, Elsevier, CINAHL) identified articles on severe bone loss during revision to rTSA. Articles were categorized into bone graft types, including allograft, autograft, and custom implant cohorts. Complications were evaluated, and outcomes were compared to previously published Minimal Clinically Important Difference (MCID) values.
RESULTS
Thirteen articles produced 556 shoulders (57% female), mean age 68 (35 - 93 age range), all with a two-year minimum follow up. After filtering those shoulders which underwent revision rTSA, 232 shoulders remained. Bone grafting accounted for 58% of cases, while 32% received custom implants. Shoulders in which bone grafting was utilized had ASES, Constant scores, and SST that all achieved MCID thresholds. Range of motion testing showed all shoulders achieved MCID for abduction and forward elevation, while only 67% and 50% of shoulders achieved MCID for external rotation and internal rotation, respectively. In the custom implant cohort, all shoulders (100%) achieved the minimal clinically important difference (MCID) in all outcome measures. Complication rates (19% vs 28%) differed between cohorts, with implant loosening (25%), instability (21%), and fracture (17%) being the most common types. The percentage of complications necessitating re-revisions was similar between the bone grafted (48%) and custom implant (52%) cohorts. Subgroup analysis of bone graft type revealed, complications requiring additional revision were predominantly seen in shoulders reconstructed with femoral head allografts (85%), compared with only 15% involving iliac crest or proximal humeral head autografts.
CONCLUSION
Both strategies, including bone grafting with conventional implants and the use of custom implants for structural glenoid defects, achieved MCIDs and meaningful improvements in patient function and pain. Subgroup analysis suggests that autograft may be a more reliable option than allograft in the setting of severe glenoid defects during revision rTSA. However, these findings are derived primarily from Level IV studies and may be subject to multiple sources of bias; therefore, higher-quality comparative studies are needed to confirm these observations.