Abstract
Purpose
We performed a systematic review of the studies including clinical/functional outcomes and complications of bone grafting for glenoid defects in reverse total shoulder arthroplasty (RTSA).
Methods
The PubMed and Cochrane databases were searched for relevant papers.
Results
Thirteen articles were included. The mean clinical/functional subjective scores significantly improved postoperatively. The implant revision rate for primary and revision RTSA was 3.1% and 21.1% respectively. The reoperation rate was 3.5% and 24.4% respectively.
Conclusions
There was moderate evidence that bone grafting is effective for glenoid defects in primary RTSA. Further high-quality research is required about revision RTSA for moderate-to-severe glenoid defects.
Keywords: Glenoid bone grafting, Reverse shoulder arthroplasty, Glenoid bone loss, Systematic review, Glenohumeral osteoarthritis, Rotator cuff arthropathy
1. Introduction
Severe glenoid bone loss in patients undergoing reverse total shoulder arthroplasty (RTSA) is considered a significant challenge for the treating physician.1 Insufficient glenoid bone stock jeopardizes initial fixation, proper positioning of the baseplate and predisposes to increased inferior scapular notching and impingement, glenoid component loosening and instability as well as decreased kinematics and soft tissue tensioning.2
There is wide variety of glenoid defect patterns according to their etiology and this fact plays a significant role in the treatment strategy.3 Glenohumeral arthritis has been associated with posterior wear,4 rotator cuff arthropathy to superior wear,3 chronic anterior instability to anterior wear,5 and revision cases to universal bone loss.6 Various classification schemes exist which succeed to describe the glenoid bone loss, although they fail to offer treatment guidelines.4,7,8
Different surgical techniques have been proposed for the management of glenoid bone defects in patients treated with RTSA. These options include preferential reaming,9 augmented implant components,10 bone grafting,11 patient-specific instrumentation,12 and, more recently, bony-increased offset-reverse shoulder arthroplasty (BIO-RSA),13 and custom-made implants.14 Despite these options, the use of bone graft to reconstruct the glenoid holds a certain appeal as not only is the glenoid defect addressed, but future revisions are easier to manage with increased available bone stock.11 Structural autografts can be harvested from the iliac crest,1,15,16 the humeral head15,17 and nonstructural from cancellous bone.16,18 In contrast, allograft options include cancellous allograft bone chips,19 femoral neck3,18 and head derived allograft.20,21
A number of clinical trials have been published assessing the role of bone grafting for the management of glenoid bone defects in patients undergoing RTSA. However, no systematic review of the literature has been published to date. For this reason, the aims of this study were four-fold: (1) what is the revision rate of RTSA when combined with bone grafting for the management of glenoid bone defects? (2) What is the complication rate of RTSA when combined with bone grafting for the management of glenoid bone defects? (3) Are there differences in revision and complication rates between primary and revision RTSA when combined with bone grafting? (4) Does the type of bone graft play a significant role on the RTSA outcome? Our primary hypothesis is that the outcomes of RTSA combined with bone grafting would be proven satisfactory in patients with glenoid bone defects.
2. Methods
2.1. Search criteria
The US National Library of Medicine (PubMed/MEDLINE) and the Cochrane Database of Systematic Reviews were queried for publications from January 1980 to January 2019 utilizing the following keywords: “glenoid” AND “reverse” AND “shoulder” AND “arthroplasty”.
2.2. Inclusion and exclusion criteria
The inclusion criteria were: 1) studies describing human subjects of any age and gender, 2) studies that include a population of at least 10 patients who were originally treated with a RTSA combined with bone graft for the management of glenoid bone defects, 3) studies that follow patients for a minimum of 24 months after surgery, 4) studies that provide a clinical/functional and/or radiographic outcome measure (e.g. patient-reported outcome scores, postoperative complications, functional scores, range of motion, pain scale, etc).
The exclusion criteria were: 1) review articles, 2) case studies with less than 10 patients, 3) technical notes, 4) corrigenda, 5) editorial notes, 6) non full-text articles, 7) studies without any clinical/functional or radiographic outcome, 8) studies in which patients were treated with RTSA and augments or custom-made implants or patient-specific instrumentation, 8) studies with mixed patient cohort, consisting of both RTSA and anatomical TSA, 9) studies in which no subjects underwent RTSA and bone grafting and 10) non-English language publications.
2.3. Data collection
Two authors independently conducted the search. All authors compiled a list of articles not excluded after application of the inclusion and exclusion criteria. Discrepancies between the authors were resolved by discussion. During initial review of the data, the following information was collected for each study: title, author, year published, study design, number of patients, number of joints, gender, indication for surgery, type of interventional treatment performed, type of graft, type of implant used, success percentage of treatment (revision-free patients), reoperation rate, complication rate, range of motion, clinical/functional subjective scores, implant shift, and radiographic graft union-incorporation.
The level of evidence in the included studies was determined using the Oxford Centre for Evidence-Based Medicine- Levels of Evidence.22 The methodological quality of each study and the different types of detected bias were assessed independently by each reviewer with the use of modified Coleman methodology score.23
3. Results
The literature search identified 431 abstracts that were examined to determine the efficacy of RTSA with bone grafting for the management of glenoid bone loss (Fig. 1). Following application of the inclusion-exclusion criteria, 13 articles were found eligible for our systematic analysis.5,6,13,24, 25, 26, 27, 28, 29, 30, 31, 32, 33 A summary flowchart of our literature search according to PRISMA guidelines can be found in Fig. 1.
Fig. 1.
The flowchart of the selection of the studies.
Four out of the 13 studies (30.8%) were prospective5,6,25,29 and nine (69.2%) were retrospective.13,24,26, 27, 28,30, 31, 32, 33 Two studies (15.4%) had a level of evidence III,6,33 whereas all other studies had a level of evidence IV.5,13,24, 25, 26, 27, 28, 29, 30, 31, 32 The modified Coleman methodology score was 47.8, while it ranged from 3733 to 6025. All studies were characterized by selection, detection and performance bias (Table 1).
Table 1.
Type of study, level of evidence, modified Coleman methodology score, potential high risk of bias, and conflicts of interest amongst authors.
| Authors | Type of study | Level of evidence | Modified Coleman score | Possible high risk of bias | Conflict of interest |
|---|---|---|---|---|---|
| Ozgur et al.24 | retrospective | IV | 44 | Selection, detection and performance | yes |
| Lanzone et al.25 | prospective | IV | 60 | Selection, detection and performance | no |
| Lorenzetti et al.26 | retrospective | IV | 50 | Selection, detection and performance | yes |
| Garofalo et al.27 | retrospective | IV | 52 | Selection, detection and performance | yes |
|
Jones et al. 28 |
retrospective | IV | 44 | Selection, detection and performance | yes |
|
Werner et al. 5 |
prospective | IV | 52 | Selection, detection and performance | no |
| Ernstbrunner et al.29 | prospective | IV | 52 | Selection, detection and performance | yes |
| Lopiz et al.30 | retrospective | IV | 46 | Selection, detection and performance | no |
|
Boileau et al. 13 |
retrospective | IV | 50 | Selection, detection and performance | yes |
| Tashjian et al.31 | retrospective | IV | 47 | Selection, detection and performance | yes |
| Gupta et al.32 | retrospective | IV | 40 | Selection, detection and performance | yes |
| Mahylis et al.33 | retrospective | III | 37 | Selection, detection and performance | no |
| Wagner et al.6 | prospective | III | 48 | Selection, detection and performance | yes |
3.1. General characteristics
In total, 539 patients (173 males and 366 females) were included in this analysis. Amongst them, 437 patients underwent RTSA with bone graft (81.1%) and 102 patients (18.9%) received RTSA without bone graft (served as control group).6 The mean age of patients ranged from 67 years24 to 75.3 years.30 Their mean follow-up ranged from 24 months24,25 to 4.9 years.5 (Table 2).
Table 2.
Demographic characteristics and mean follow-up of the patients.
| Authors | Number of patients | Mean age (years) |
Sex | Mean follow-up |
|---|---|---|---|---|
|
Ozgur et al. 24 |
20 | 67 | 11F | 24 months |
| 9M | ||||
|
Lanzone et al. 25 |
16 | 75 | 10M | 24 months |
| 6F | ||||
| Lorenzetti et al.26 | 57 | 73 | 17M | 46 months |
| 40F | ||||
| Garofalo et al.27 | 26 | 68.5 | 8M | 32 months |
| 18F | ||||
|
Jones et al. 28 |
44 | 69.1 | 20M | 40.6 months |
| 24F | ||||
|
Werner et al. 5 |
21 | 71 | 3M,18F | 4.9 years |
| Ernstbrun-ner et al.29 | 41 | 73.5 | 13M, 28F | 2.8 years |
|
Lopiz et al. 30 |
20 | 75.3 | 2M, 18F | 38 months |
|
Boileau et al. 13 |
54 | 73 | 13M, 41F | 36 months |
|
Tashjian et al. 31 |
14 | 75 | 4M, 10F | 2.6 years |
|
Gupta et al. 32 |
54 | 70.4 | 8M, 46F | 29.6 months |
|
Mahylis et al. 33 |
30 | 67.4 | 16M, 14F | 2.9 years |
|
Wagner et al. 6 |
142 (40 study group +102 control group) | 69 | 50M, 92F | 3.1 years |
(M: males, F: females, Not rep.: not reported).
3.2. Indication for RTSA
Three papers (23.1%) dealt with RTSA as revision procedure,6,24,33 three papers (23.1%) included mixed cases of primary and revision RTSA,28,30,32 while the remaining seven papers examined only patients with primary RTSA.5,13,25, 26, 27,29,31
From the 279 patients who underwent primary RTSA for a known cause, 140 (50.2%) were suffering from rotator cuff arthropathy, 16 (5.7%) from rheumatoid arthritis, 28 (10%) from osteoarthritis, 41 (14.7%) from instability, and 28 (10%) from fracture-dislocation. In addition, 10 patients (3.6%) had undergone failed rotator cuff repair in the past, two patients (0.7%) had fracture sequelae, and one patient (0.4%) was suffering from neuropathic arthropathy. Finally, 13 cases (4.7%) were characterized as post-traumatic without any further clarification.
From the 81 patients who had previously undergone RTSA which failed, there were 30 cases (37%) of glenoid loosening, 20 cases (24.7%) of instability, 16 cases (19.8%) of infection, five cases (6.2%) of rotator cuff insufficiency, four cases (4.9%) of progressive glenoid arthrosis, four cases (4.9%) of glenoid wear, and two cases (2.5%) of failed humeral hemiarthroplasty.
3.3. Assessment of glenoid defect
In the three papers that included exclusively patients undergoing revision RTSA,6,24,33 the size of the glenoid defect was estimated intraoperatively, after the removal of the primary components. In eight of the remaining papers, the size of the glenoid defect was calculated with the use of preoperative computed tomography (CT),5,13,25, 26, 27,30, 31, 32 while four of these studies26,27,30,32 utilized three-dimensional CT reconstruction. Preoperative X-rays were used as well in six out of the eight papers that used CT.5,25,27,30, 31, 32 Finally, one study29 used only preoperative X-rays for the calculation of the glenoid defect (Table 3).
Table 3.
Type of RTSA, type of graft and grade of glenoid bone loss per study.
| Author(s) | Type of RTSA | Type of graft | Glenoid bone defect |
|---|---|---|---|
| Ozgur et al.24 | Tornier (South Bloomington, MN, USA), except for one patient who had a Tornier humeral stem and a DonJoy baseplate (Austin, TX, USA). |
Allograft (8 from femoral shaft, 11 from femoral neck/head, 5 from proximal humerus). No information about number of screws for fixation. | Not reported. |
| Lanzone et al.25 | Lima SMR reverse (Systema Multiplana Randelli, Udine, Italy) |
Humeral head hexagonal autograft. No information about number of screws for fixation. | 13 B2 and 3C glenoids (Walch classification) |
| Lorenzetti et al.26 | Reverse Shoulder Prosthesis DJO Surgical | Autograft (52 patients; 91%), femoral head allograft (5 patients; 9%). No information about number of screws for fixation. | 16 glenoids were grade E1 and 19 glenoids were grade E3 (Sirveaux classification). 9 glenoids were grade 3,(Levigne classification) 3 glenoids were type A2, 2 were type B2 and 2 were type C (Walch classification), 6 glenoids with severe bone loss were unable to be classified |
| Garofalo et al.27 | 22 AequalisTM Reversed II, 4 Ascend flex reverse shoulder arthroplasty (Tornier, Edina, MN). |
Humeral head autograft, one screw for its fixation. | Not reported. |
| Jones et al.28 | Equinoxe RTSA; Exactech, Inc., Gainesville, FL, USA | 30 patients received an autograft (29 humeral heads and 1 iliac crest) and 14 patients received an allograft from femoral head. No information about number of screws for fixation. |
Not reported. |
| Werner et al.5 | 8 Delta reverse shoulder prostheses (DePuy Orthopedics, Warsaw, IN, USA) and 13 Aequalis reversed shoulder prostheses (Tornier, Saint-Ismier, France). |
Autograft from humeral head. No information about number of screws for fixation. | Antuna classification: Central defects:- Peripheral defects: grade I: 1 grade II: 3 grade III: 5 Combined defects: grade I:- grade II: 10 grade III: 2 Glenoid bone loss averaged 45% preoperatively. |
| Ernstbrunner et al.29 | 32 (78%) Comprehensive Reverse Shoulder Prosthesis (Biomet, Warsaw, IN, USA), 6 (15%) Delta Xtend (DePuy Orthopedics, Warsaw, IN, USA), and 3 (7%) Encore Reverse Shoulder (DJO Surgical, Austin, TX, USA) |
Autograft (humeral head or one iliac crest): 39 Allograft (CanPac or femoral head): 2 No information about number of screws for fixation. |
Moderate to severe glenoid erosion was present in 39 patients (98%) (mild glenoid erosion: bone loss confined to the peripheral part of the glenoid; moderate glenoid erosion: extending from the periphery of the glenoid to its midline; severe glenoid erosion: extending beyond the midline of the glenoid) |
| Lopiz et al.30 | Delta Xtend (DePuy Orthopaedics, Warsaw, IN, USA) in 16 cases and SMR Modular Shoulder System with the Axioma TT Metal Back (Systema Multiplana Randelli, Lima-LTO, Italy) in 4 cases. |
Humeral head autograft (35%), frozen allograft of the femur or tibia (65%). No information about number of screws for fixation. |
Not reported. |
| Boileau et al.13 | Aequalis reversed prosthesis (Wright-Tornier) |
Autologous, trapezoidal humeral head bone graft. No information about number of screws for fixation. | Walch classification:8 shoulders were A2, 15 were B2, 7 were C. Favard system: 15 shoulders were E2, 21 were E3, 3 were E4. Combined vertical and horizontal glenoid bone deficiency was present in 15 patients. |
| Tashjian et al.31 | 1 Trabecular Metal Reverse Total Shoulder Arthroplasty (Zimmer, Warsaw, IN, USA) and 13 Aequalis Reversed Shoulder Arthroplasty (Tornier, Bloomington, MN, USA) |
Humeral head autograft. No information about number of screws for fixation. | Rotator cuff tear arthropathy: 7 (Favard classification E2: 3 E3: 4) Glenohumeral osteoarthritis: 5 (Walch classification A2: 1 B3: 4) Anterior glenoid fracture: 2 |
| Gupta et al.32 | Flat-back glenoid components (Delta III; DePuy International Ltd., Leeds, UK) were used until 2006, and a convex-shaped backside glenoid baseplate (Delta Xtend; DePuy, Warsaw, IN, USA) was used since 2007. |
Autograft (humeral head or iliac crest). No information about number of screws for fixation. | Not reported. |
| Mahylis et al.33 | DePuy Delta Xtend (DePuy Synthes, Warsaw, IN, USA) in 21 patients and Aequalis Reversed Shoulder (Tornier, Edina, MN, USA) in 9. |
15 patients received structural iliac crest bone autograft, 15 patients received nonstructural bone allograft. No information about number of screws for fixation. | Not reported. |
| Wagner et al.6 | Comprehensive Reverse Shoulder (Biomet, Warsaw, Indiana) in 20 shoulders; Encore Reverse Shoulder Prosthesis (DJO Surgical, Austin, Texas) in 10 shoulders; the Delta III and Delta Xtend (DePuy Orthopaedics, Warsaw, Indiana) in one and seven shoulders, respectively; Aequalis Reversed Shoulder (Tornier, Edina, Minnesota) in two shoulders. |
Autograft (iliac crest or humerus): 14 Allograft [CanPac, femur, and/or DBX (demineralized bone matrix)]: 20 Both: 6 No information about number of screws for fixation. |
Not reported. |
In total, the glenoid bone defect was graded in six out of the 13 studies (46.2%). Based on the studies which used Walch classification to assess patients with glenohumeral osteoarthritis, 12 glenoid defects (out of 58; 20.7%) were graded as type A2, 30 as type B2 (out of 58; 51.7%), four as B3 (out of 58; 6.9%), and 12 as type C (out of 58; 20.7%). Based on the studies which made use of the Favard classification to classify patients with rotator cuff arthropathy, there were 18 cases of type E2 (out of 46; 39.1%), 25 of type E3 (out of 46; 54.3%) and three of type E4 (out of 46; 6.5%).
3.4. Type of graft
Six studies out of the 13 studies (46.2%) reported that they made use of bone autograft,5,13,25,27,31,32 while allograft was purely used in one study (7.7%).24 In the remaining six studies (46.2%), both autograft and allograft were used in different cases.6,26,28, 29, 30,33 (Table 3).
In total, the vast majority of patients were treated with bone autograft (342 out of 406 patients; 84.2% of all grafts). Regarding the origin of the autograft, 205 patients (out of 342 who had autograft; 60.0%) received autograft harvested from the humeral head, while iliac crest autograft was used in 17 patients (out of 342; 4.9%) and no information regarding the origin of autograft was provided in 120 patients (out of 342; 35.1%). Furthermore, bone allograft was applied in 58 cases (out of 406 patients; 14.3%). Amongst them, there were 19 cases treated with femoral head allograft, 11 cases in which the origin of the allograft was the proximal femur, eight cases who received femoral shaft allograft, five cases with proximal humerus allograft, while in the remaining 15 cases the origin was not clarified. Finally, in six cases, both autograft and allograft were used.
3.5. Clinical/functional subjective scores
All studies included in this review showed significant improvement of the mean postoperative clinical/functional subjective scores compared with the respective mean preoperative values. Seven out of the 13 studies (53.8%) made use of the Constant score.5,13,25,27,28,30,32 Both the Simple Shoulder Test (SST)6,26,28,29,31,32 and the American Shoulder and Elbow Surgeons (ASES) score6,26,28, 29, 30, 31 were reported in six studies (46.2%), while the Visual Analogue Scale (VAS) pain score was used in four out of the 13 studies (30.8%).26,28,30,31 In addition, three out of the 13 studies (23.1%) made use of the Subjective Shoulder Value (SSV).13,25,29 Finally, one33 out of the 13 studies (7.7%) utilized the Penn Shoulder Score (PSS) and the Veterans RAND 12-item (VR-12) health survey (mental and physical component score), while another one (7.7%) measured the Shoulder Pain and Disability Index (SPADI).28 (Table 4).
Table 4.
Mean ROM and clinical/functional subjective scores before and after surgery.
| Author(s) | Preoperative mean score and range of motion | Postoperative mean score and range of motion |
|---|---|---|
| Lanzone et al.25 | CS: 31 SSV: 25 |
CS: 61 (sign.dif.) SSV: 70 (sign.dif.) |
| Lorenzetti et al.26 | ASES: 36.7 SST: 1.6 VAS pain: 6.2 VAS function: 3.1 Forward elevation: 69.7° Abduction: 64.8° External rotation: 29.4° Internal rotation: 2.7 (0:no internal rotation, 1: to the greater trochanter, 2: L5:S1, 3: L3-L4, 4: L1-L2, 5: T12, 6: T10-T11, 7: T8:T9, 8: T6:T7) |
ASES: 75.3 (sign.dif.) SST: 7 (sign.dif.) VAS pain: 1.6 (sign.dif.) VAS function: 7.4 (sign.dif.) Forward elevation: 142.1° (sign.dif.) Abduction: 132.5° (sign.dif.) External rotation: 53.7° (sign.dif.) Internal rotation: 4.6 (sign.dif.) |
| Garofalo et al.27 | Active elevation:- Abduction:- External rotation:- CS:- |
Active elevation: 135° Abduction: 122° External rotation: 30° CS: 68.2 |
| Jones et al.28 | ASES: 36 CS: 30 SST: 3.2 SPADI: 85.2 Abduction: 69° Forward flexion: 75° External rotation: 14° VAS pain: 6.3 |
ASES: 75 (sign.dif.) CS: 57.7 (sign.dif.) SST: 8.2 (sign.dif.) SPADI: 35 (sign.dif.) Abduction: 102° (sign.dif.) Forward flexion: 116° (sign.dif.) External rotation: 26° (sign.dif.) VAS pain: 1.9 (sign.dif.) Insignificant differences between the allograft and autograft groups preoperatively. Insignificant differences in postoperative outcome measures, pain scores and range of motion between the allograft and autograft groups. Insignificant differences in average improvement for all measures between the allograft and autograft groups. |
| Werner et al.5 | CS: 5.7 age- and gender-adjusted CS: 7.8% Elevation: 35° Abduction: 25° External rotation: 2.4° |
CS: 57.2 (sign.dif.) age- and gender-adjusted CS: 81.3% (sign.dif.) Elevation: 128° (sign.dif.) Abduction: 113° (sign.dif.) External rotation: 8.4° |
| Ernstbrunner et al.29 | Abduction: 59° ASES:- SST: 6.6 SSV:- |
Abduction: 149° (sign.dif.) ASES: 77 SST: 9 SSV: 90 |
| Lopiz et al.30 | CS: 30.7 ASES: 33.8 Abduction: 59° Forward flexion: 70° External rotation: 10° VAS pain: 7.4 |
CS: 51.3 (sign.dif.) ASES: 67.6 (sign.dif.) Abduction: 104° (sign.dif.) Forward flexion: 113° (sign.dif.) External rotation: 13° VAS pain: 3.3 (sign.dif.) |
| Boileau et al.13 | Active forward elevation: 85° Active external rotation: 12° Active internal rotation S1 CS: 31 SSV: 30 |
Active forward elevation: 148° (sign.dif.) Active external rotation: 24° (sign.dif.) Active internal rotation: L4 (sign.dif.) CS: 68 (sign.dif.) SSV: 83 (sign.dif.) |
| Tashjian et al.31 | Passive forward elevation: 106° Active forward elevation: 80° Active adducted external rotation: 18° Active adducted internal rotation: 1.3 (Active adducted internal rotation was coded as follows: 0 points = lateral thigh, 2 = buttock, 4 = lumbosacral junction, 6 = L3, 4 = sacrum; 8 = T12, 10 = interscapular region) VAS pain: 8.1 SST: 1.8 ASES: 22 |
Passive forward elevation: 136° Active forward elevation: 130° (sign.dif.) Active adducted external rotation: 29° Active adducted internal rotation: 3.1 VAS pain: 2.5 (sign.dif.) SST: 6.5 (sign.dif.) ASES: 66 (sign.dif.) |
|
Gupta et al. 32 |
CS: 17.9 SST: 1.6 |
CS: 78.9 (sign.dif.) SST: 7.5 (sign.dif.) |
| Mahylis et al.33 | PSS: 16 VR-12 MCS: 49 VR-12 PCS: 30 Active forward elevation: 78° Active external rotation: 18° |
PSS: 81 (sign.dif.) VR-12 MCS: 55 (sign.dif.) VR-12 PCS: 38 (sign.dif.) Active forward elevation: 150° (sign.dif.) Active external rotation: 34° (sign.dif.) |
| Wagner et al.6 | Abduction: 54° ASES: SST: |
Abduction: 112° (sign.dif.) ASES: 66.0 SST: 6.0 |
ASES: American Shoulder and Elbow Surgeons, SST: Simple Shoulder Test, SSV: Subjective Shoulder Value, VAS: Visual Analogue Scale, CS: Constant score, PSS: Penn Shoulder Score, VR-12: Veterans RAND 12-item health survey, MCS: mental component score, PCS: physical component score, SPADI: Shoulder Pain and Disability Index, sign.dif.: significant difference).
3.6. Range of motion (ROM)
Ten out of the 13 studies (76.9%) examined postoperative ROM,5,6,13,26, 27, 28, 29, 30, 31,33 while three studies (23.1%) did not comprise measurements of shoulder ROM24,25,32 All studies which compared preoperative to postoperative mean ROM showed that the latter was significantly improved at the final follow-up (Table 4).
3.7. Revision and reoperation rates
Overall, the all-cause reoperation rate was 8.9% (39 out of 437 patients), while the implant revision rate was 7.6% (33 out of 437 patients) Specifically, 22 patients (5.0%) had revision of both the humeral and glenoid components, six patients (1.4%) underwent isolated revision of the glenoid component, and five patients (1.1%) had isolated revision of the humeral component. The most common causes for revision surgery were aseptic loosening (14 out of 437 cases; 2.5%), infection (9 out of 437 cases; 2.1%) and dislocation (4 out of 437 cases; 0.9%). There was also one case (0.2%) of pending periprosthetic fracture, and one case (0.2%) in which the surgeon failed to attain satisfactory screw purchase, and primary central peg fixation was inadequate.
3.7.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA5,13,25, 26, 27,29,31 the all-cause implant revision rate of primary RTSA, when combined with glenoid bone grafting, was 3.1% (7 out of 229 patients), while the all-cause reoperation rate was 3.5% (8 out of 229 patients).
3.7.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the all-cause implant re-revision rate of revision RTSA, when combined with glenoid bone grafting, was 21.1% (19 out of 90 patients), while the all-cause re-reoperation rate was 24.4% (22 out of 90 patients).
3.8. Graft union-incorporation and implant shift
There were 402 cases with graft union (92% of RTSAs with bone graft) and 35 cases (8% of RTSAs with bone graft) with inadequate graft incorporation noted in follow-up X-rays. Among them, four cases required revision surgery. Among these four cases, there were two cases with concomitant glenoid aseptic loosening and one case with concomitant glenoid aseptic loosening and implant shift. In addition, six cases (1.4% of RTSAs with bone graft) of glenoid baseplate shift were documented. Three of these six cases required revision surgery. Among the three cases, there were two cases where there was concomitant aseptic glenoid loosening.
3.8.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 there were 12 cases (out of 229; 5.2%) with inadequate graft incorporation. Two of these 12 cases required revision surgery, while the other 10 cases were followed-up without requiring any intervention. Furthermore, there was one case (out of 229; 0.4%) of glenoid baseplate shift in a patient treated with primary RTSA which did not require reoperation.
3.8.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 there were 11 cases (out of 90; 12.2%) with inadequate graft incorporation. One of these 11 cases required revision surgery, while the other 10 cases were followed-up without requiring any intervention. Furthermore, there were four cases (out of 90; 4.4%) of patients treated with revision RTSA who were postoperatively found with glenoid baseplate shift; two of them required revision surgery.
3.9. Aseptic loosening
Aseptic loosening was reported in seven out of the 13 papers (53.8%). Twenty two cases (5%) were found which were divided in three cases of humeral component loosening and 19 cases of glenoid component loosening. Seventeen out of these 19 cases (77.3%) required reoperation, while 14 of them had revision surgery.
3.9.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 the rate of aseptic loosening of primary RTSA, when combined with glenoid bone grafting, was 3.1% (7 out of 229 patients). Five patients had aseptic loosening of the glenoid component and four of them were revised. Moreover, there were two patients with aseptic loosening of the humeral components who both required revision surgery.
3.9.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the rate of aseptic loosening of revision RTSA, when combined with glenoid bone grafting, was 7.8% (7 out of 90 patients). All seven patients had aseptic loosening of the glenoid component and six of them were revised.
3.10. Periprosthetic fractures
In seven out of the 13 papers (53.8%), intraoperative or postoperative fractures were reported. In total, there were 18 fractures (4.1%), two of which required reoperation (11.1%). Ozgur et al. reported one case of pending periprosthetic fracture which was treated with revision of the humeral component as well as one case of periprosthetic humeral fracture which was treated conservatively.24 Lorenzetti et al. found one patient with periprosthetic fracture (1.8% of their patient cohort) who was treated with open reduction and internal fixation.26 Also, there were five acromial or scapular spine fractures (8.8%) treated conservatively.26 Moreover, Ernstbrunner et al. pointed out two cases of scapular spine fractures and one case of greater tuberosity fracture (7.3%) which were treated conservatively.29
3.10.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 the rate of periprosthetic fractures of primary RTSA, when combined with glenoid bone grafting, was 4.8% (11 out of 229 patients). One patient required open reduction and internal fixation of the fracture, while no one of the patients required implant revision.
3.10.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the rate of periprosthetic fractures of revision RTSA, when combined with glenoid bone grafting, was 2.2% (2 out of 90 patients). One patient required implant revision, while the other one was treated conservatively.
3.11. Dislocations
Dislocations were noticed in seven out of the 13 studies (53.8%). Totally, 12 cases with dislocation were noted (2.7%), four of which required revision surgery (33.3%). Gupta et al. found three cases of dislocation (5.6% of their patient cohort) without further clarifying their method of treatment.32 In the study of Mahylis et al.,33 there was one patient with dislocation (3.3%) who was conservatively treated with closed reduction, whereas Wagner et al. found two patients with dislocated shoulders (5%) who underwent revision RTSA by replacing both the humeral component and the glenosphere.6
3.11.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 the rate of dislocations of primary RTSA, when combined with glenoid bone grafting, was 1.3% (3 out of 229 patients). All of these patients were treated conservatively.
3.11.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the rate of dislocations of revision RTSA, when combined with glenoid bone grafting, was 5.6% (5 out of 90 patients). Four patients required reoperation, while one patient was treated conservatively.
3.12. Infection
Infection was reported in seven out of the 13 studies of the review (53.8%). In total, 12 cases of infection were noted (2.7%), 10 of which underwent reoperation (83.3%) and nine of these 10 cases required re-implantation surgery. In addition, there was one case of superficial infection which was treated conservatively. Specifically, five patients in the study of Ozgur et al. (25% of their patient cohort) underwent removal of their graft secondary to infection.24 One patient had a revised RTSA combined with a new allograft positioned to the glenoid after the first allograft was removed as a result of low grade infection.24 Two patients were treated with 2-stage revision, while two patients underwent a resection arthroplasty for low grade infection.24 In addition, Jones et al.28 reported two cases of deep joint infection (4.5%), one of which was treated with 2-stage revision and the other one with permanent antibiotic spacer, while Boileau et al. found one case of infected glenoid loosening (1.9%) which did not undergo surgery.13
3.12.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 the rate of periprosthetic joint infection of primary RTSA, when combined with glenoid bone grafting, was 0.9% (2 out of 229 patients). No one from these two patients required revision surgery.
3.12.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the rate of periprosthetic joint infection of revision RTSA, when combined with glenoid bone grafting, was 7.8% (7 out of 90 patients). All seven patients required implant revision.
3.13. Radiographic and other complications not requiring surgery
Totally, 107 cases with complications not requiring surgery were found (24.5%). The three most common complications amongst them were scapular notching (65 cases) (60.7%), glenoid radiolucency (19 cases) (17.8%), and humeral radiolucency (11 cases) (10.3%).
3.13.1. Primary RTSA
Based on the studies that separately reported the outcomes of patients treated with primary RTSA,5,13,25, 26, 27,29,31 the rate of non-surgical complications of primary RTSA, when combined with glenoid bone grafting, was 14.8% (34 out of 229 patients). There were 23 cases of scapular notching, eight cases of heterotopic ossification, one broken humeral tray, one pulmonary embolism and one transient brachial plexus palsy.
3.13.2. Revision RTSA
Based on the studies that separately reported the outcomes of patients treated with revision RTSA,6,24,33 the rate of non-surgical complications of revision RTSA, when combined with glenoid bone grafting, was 27.8% (25 out of 90 patients). There were 12 cases of scapular notching, 11 cases of radiographic glenoid lucency, and two cases of radiographic humeral lucency.
3.14. RTSA with autograft versus allograft for the management of severe glenoid bone defects
Autografts and allografts were compared in three studies included in this analysis.28,30,33 All three studies reported no significant differences between the autograft- and the allograft-treated group in the outcomes.28,30,33 Specifically, in the study by Mahylis et al.,33 in which 15 patients received structural iliac crest bone autograft and 15 patients received nonstructural bone allograft, no significant differences between the two groups were observed in terms of bone graft integration, implant position, scapular notching, implant shift or failure, clinical scores and range of motion. In addition, Lopiz et al.30 assessed bone graft integration in autograft- and allograft-treated patients to find non-significant differences amongst groups (100% of autografts and 92% of allografts were fully incorporated). Moreover, Jones et al.28 documented no significant differences in postoperative clinical and functional outcome variables between autograft- and allograft-treated patients.
4. Discussion
The key finding of this study was that the use of primary RTSA combined with glenoid bone grafting was associated with satisfactory clinical, functional and radiographic outcomes in the short term. Specifically, all mean clinical/functional subjective scores as well as ROM were significantly improved after surgery. Primary RTSA with bone grafting resulted in high survival rates (96.9%) and low rates of complications requiring surgery (3.5%). In addition, the rates of radiographic graft non-union (5.2%) and glenoid component shift (0.4%) were also low. Based on these findings, it is suggested that primary RTSA with glenoid bone grafting might be a safe and effective technique for the management of patients with glenoid bone loss who require RTSA.
On the contrary, revision RTSA with glenoid bone grafting was associated with substantially lower implant survivorship (78.9%) and higher complications requiring reoperation (24.4%) compared to primary RTSA with glenoid bone grafting. In addition, the rates of radiographic graft non-union (12.2%), non-surgical complications (27.8%) were much higher compared to the respective rates of primary RTSA with glenoid bone grafting. Taking into consideration these findings, it might be assumed that revision RTSA with glenoid bone is associated with deteriorated outcomes compared to primary RTSA with glenoid bone grafting.
Aseptic loosening was the most common (5%) complication leading to failure after RTSA combined with glenoid bone grafting, while the rate of aseptic loosening in primary cases (3.1%) was lower compared to revision cases (7.8%). Periprosthetic joint infection was less frequently documented (2.7%) as a reason for failure as well as postoperative dislocations (4.1%). Once again, both the rates of periprosthetic joint infection and postoperative dislocations were lower in primary cases (0.9% and 1.3%, respectively) compared to revision cases (7.8% and 5.6%, respectively). Finally, regardless of the graft type that was used, fractures of the bone graft were very rarely observed in both primary and revision cases.
Bone grafting of the glenoid defect was used both in the setting of primary and revision RTSA surgery. The most common indication for primary surgery was rotator cuff arthropathy (50.2%). The three most frequent indications for conversion of a shoulder arthroplasty to RTSA combined with glenoid bone grafting were glenoid loosening (30.7%), glenohumeral instability (24.7%), and periprosthetic joint infection (19.8%).
The type and amount of glenoid bone loss was diagnosed either preoperatively (CT, X-rays) or intraoperatively. Various classification systems (Table 3) were used in the studies included in this review showing that there are still controversies regarding the optimal way to grade these bone defects. There is a need for critical appraisal of the validity of classification systems currently in use and the development of a consensus system that will permit comparison between the published results of different techniques dealing with glenoid bone loss in patients with rotator cuff arthropathy.
There was a complete lack of studies examining the impact of RTSA design,34 glenosphere lateralization,35 glenosphere diameter36 or humeral neck-shaft angle.35 As for the type of graft, either autograft or allograft was used in the studies which were included in this analysis, both resulting in high survival and graft incorporation rates. There were three studies which directly compared the outcome of patients treated with autografts and allografts. All three studies reported no significant differences between the autograft- and the allograft-treated group.28,30,33 Based on this finding, it seemed that the type of graft used in conjunction with RTSA might not affect the postoperative outcomes.
The studies assessed in this review had several limitations. First of all, there was a complete lack of studies with long-term follow-up, while most studies reported only short-term outcomes. In addition, there were no level I and II controlled trials, while all apart two studies6,33 were level IV case series. The quality of these studies was moderate based on the modified Coleman methodology score and potential selection, detection and performance bias might have influenced the results. However, the type of treatment was clearly defined and homogenous among studies, while the survival rates remained high in all studies included in this review.
Another limitation was that there were not any controlled studies to directly compare the outcomes of primary and revision RTSA with bone grafting. However, although it cannot be supported by any statistical meta-analysis due to the heterogeneity of the studies, the differences that were observed in all clinical and radiographic outcomes were towards the same direction, suggesting that primary RTSA might have higher survival rates and lower complications compared to revision RTSA.
5. Conclusions
There was moderate evidence to show that bone grafting is effective in the short term for the management of glenoid bone defects in patients undergoing primary RTSA. In contrast, further research of higher quality is required to generate more evidence-based conclusions to guide the optimal management of patients with moderate-to-severe glenoid bone defects who undergo revision RTSA.
Ethical approval
Not applicable.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
None.
References
- 1.Mizuno N., Denard P.J., Raiss P., Walch G. Reverse total shoulder arthroplasty for primary glenohumeral osteoarthritis in patients with a biconcave glenoid. J Bone Joint Surg Am. 2013;95:1297–1304. doi: 10.2106/JBJS.L.00820. [DOI] [PubMed] [Google Scholar]
- 2.Riboh J.C., Garrigues G.E. Bone grafting in shoulder arthroplasty. Orthopedics. 2012;35:966–973. doi: 10.3928/01477447-20121023-11. [DOI] [PubMed] [Google Scholar]
- 3.Seidl A.J., Williams G.R., Boileau P. Challenges in reverse shoulder arthroplasty: addressing glenoid bone loss. Orthopedics. 2016;39(1):14–23. doi: 10.3928/01477447-20160111-01. Jan-Feb. [DOI] [PubMed] [Google Scholar]
- 4.Walch G., Badet R., Boulahia A., Khoury A. Morphologic study of the glenoid in primary glenohumeral osteoarthritis. J Arthroplast. 1999;14(6):756–760. doi: 10.1016/s0883-5403(99)90232-2. Sep; [DOI] [PubMed] [Google Scholar]
- 5.Werner B.S., Böhm D., Abdelkawi A., Gohlke F. Glenoid bone grafting in reverse shoulder arthroplasty for long-standing anterior shoulder dislocation. J Shoulder Elb Surg. 2014;23(11):1655–1661. doi: 10.1016/j.jse.2014.02.017. Nov. [DOI] [PubMed] [Google Scholar]
- 6.Wagner E., Houdek M.T., Griffith T. Glenoid bone-grafting in revision to a reverse total shoulder arthroplasty. J Bone Joint Surg Am. 2015; Oct 21;97(20):1653–1660. doi: 10.2106/JBJS.N.00732. [DOI] [PubMed] [Google Scholar]
- 7.Sirveaux F., Favard L., Oudet D., Huquet D., Walch G., Mole D. Grammont inverted total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff. Results of a multicentre study of 80 shoulders. J Bone Joint Surg Br. 2004;86:388–395. doi: 10.1302/0301-620x.86b3.14024. [DOI] [PubMed] [Google Scholar]
- 8.Antuna S.A., Sperling J.W., Cofield R.H., Rowland C.M. Glenoid revision surgery after total shoulder arthroplasty. J Shoulder Elb Surg. 2001;10(3):217–224. doi: 10.1067/mse.2001.113961. May-Jun. [DOI] [PubMed] [Google Scholar]
- 9.Sears B.W., Johnston P.S., Ramsey M.L., Williams G.R. Glenoid bone loss in primary total shoulder arthroplasty: evaluation and management. J Am Acad Orthop Surg. 2012;20:604–613. doi: 10.5435/JAAOS-20-09-604. [DOI] [PubMed] [Google Scholar]
- 10.Wright T.W., Roche C.P., Wright L., Flurin P.H., Crosby L.A., Zuckerman J.D. Reverse shoulder arthroplasty augments for glenoid wear. Comparison of posterior augments to superior augments. Bull Hosp Jt Dis. 2015;73(Suppl 1):S124–S128. [PubMed] [Google Scholar]
- 11.Malhas A.M., Granville-Chapman J., Robinson P.M. Reconstruction of the glenoid using autologous bone-graft and the SMR Axioma TT metal-backed prosthesis: the first 45 sequential cases at a minimum of two years' follow-up. Bone Joint Lett J. 2018;100-B(12):1609–1617. doi: 10.1302/0301-620X.100B12.BJJ-2018-0494.R1. Dec. [DOI] [PubMed] [Google Scholar]
- 12.Villatte G., Muller A.S., Pereira B., Mulliez A., Reilly P., Emery R. Use of Patient-Specific Instrumentation (PSI) for glenoid component positioning in shoulder arthroplasty. A systematic review and meta-analysis. PLoS One. 2018;13(8) doi: 10.1371/journal.pone.0201759. Aug 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Boileau P., Morin-Salvo N., Gauci M.O. Angled BIO-RSA (bony-increased offset-reverse shoulder arthroplasty): a solution for the management of glenoid bone loss and erosion. J Shoulder Elb Surg. 2017;26(12):2133–2142. doi: 10.1016/j.jse.2017.05.024. Dec. [DOI] [PubMed] [Google Scholar]
- 14.Debeer P., Berghs B., Pouliart N., Van den Bogaert G., Verhaegen F., Nijs S. Treatment of severe glenoid deficiencies in reverse shoulder arthroplasty: the Glenius Glenoid Reconstruction System experience. J Shoulder Elbow Surg Epub ahead of print. 13 Feb 2019 doi: 10.1016/j.jse.2018.11.061. [DOI] [PubMed] [Google Scholar]
- 15.Neyton L., Boileau P., Nové-Josserand L., Edwards T.B., Walch G. Glenoid bone grafting with a reverse design prosthesis. J Shoulder Elb Surg. 2007;16(3 Suppl):S71–S78. doi: 10.1016/j.jse.2006.02.002. [DOI] [PubMed] [Google Scholar]
- 16.Melis B., Bonnevialle N., Neyton L. Glenoid loosening and failure in anatomical total shoulder arthroplasty: is revision with a reverse shoulder arthroplasty a reliable option? J Shoulder Elb Surg. 2012;21:342–349. doi: 10.1016/j.jse.2011.05.021. [DOI] [PubMed] [Google Scholar]
- 17.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(9):2558–2567. doi: 10.1007/s11999-011-1775-4. Sep; [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bateman E., Donald S.M. Reconstruction of massive uncontained glenoid defects using a combined autograft-allograft construct with reverse shoulder arthroplasty: preliminary results. J Shoulder Elb Surg. 2012;21:925–934. doi: 10.1016/j.jse.2011.07.009. [DOI] [PubMed] [Google Scholar]
- 19.Elhassan B., Ozbaydar M., Higgins L.D., Warner J.J. Glenoid reconstruction in revision shoulder arthroplasty. Clin Orthop Relat Res. 2008;466:599–607. doi: 10.1007/s11999-007-0108-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.LoDe Biase C.F., Ziveri G., De Caro F., Roberts N., Delcogliano M. Reverse shoulder arthroplasty using a “L” shaped allograft for glenoid reconstruction in a patient with massive glenoid bone loss: case report. Eur Rev Med Pharmacol Sci. 2014;18(1 Suppl):44–49. [PubMed] [Google Scholar]
- 21.Klein S.M., Dunning P., Mulieri P., Pupello D., Downes K., Frankle M.A. Effects of acquired glenoid bone defects on surgical technique and clinical outcomes in reverse shoulder arthroplasty. J Bone Joint Surg Am. 2010;92:1144–1154. doi: 10.2106/JBJS.I.00778. [DOI] [PubMed] [Google Scholar]
- 22.Burns P.B., Rohrich R.J., Chung K.C. The levels of evidence and their role in evidence-based medicine. Plast Reconstr Surg. 2011;128(1):305–310. doi: 10.1097/PRS.0b013e318219c171. Jul. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sambandam S.N., Gul A., Priyanka P. Analysis of methodological deficiencies of studies reporting surgical outcome following cemented total-joint arthroplasty of trapezio-metacarpal joint of the thumb. Int Orthop. 2007; Oct;31(5):639–645. doi: 10.1007/s00264-006-0240-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ozgur S.E., Sadeghpour R., Norris T.R. Revision shoulder arthroplasty with a reverse shoulder prosthesis: use of structural allograft for glenoid bone loss. Der Orthopäde. 2017;46(12):1055–1062. doi: 10.1007/s00132-017-3494-3. Dec. [DOI] [PubMed] [Google Scholar]
- 25.Lanzone R., Carbone S., Albino P., Cassio J.B., Métais P. Retroverted glenoid reconstruction using glenoid plate in reverse shoulder arthroplasty. Musculoskelet Surg. 2017;101(Suppl 2):121–127. doi: 10.1007/s12306-017-0481-0. Dec. [DOI] [PubMed] [Google Scholar]
- 26.Lorenzetti A., Streit J.J., Cabezas A.F. Bone graft augmentation for severe glenoid bone loss in primary reverse total shoulder arthroplasty: outcomes and evaluation of host bone contact by 2D-3D image registration. JBJS Open Access. 2017; Jul 28;2(3) doi: 10.2106/JBJS.OA.17.00015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Garofalo R., Brody F., Castagna A., Ceccarelli E., Krishnan S.G. Reverse shoulder arthroplasty with glenoid bone grafting for anterior glenoid rim fracture associated with glenohumeral dislocation and proximal humerus fracture. Orthop Traumatol Surg Res. 2016;102(8):989–994. doi: 10.1016/j.otsr.2016.09.009. Dec. [DOI] [PubMed] [Google Scholar]
- 28.Jones R.B., Wright T.W., Zuckerman J.D. Reverse total shoulder arthroplasty with structural bone grafting of large glenoid defects. J Shoulder Elb Surg. 2016;25(9):1425–1432. doi: 10.1016/j.jse.2016.01.016. Sep. [DOI] [PubMed] [Google Scholar]
- 29.Ernstbrunner L., Werthel J.D., Wagner E., Hatta T., Sperling J.W., Cofield R.H. Glenoid bone grafting in primary reverse total shoulder arthroplasty. J Shoulder Elb Surg. 2017;26(8):1441–1447. doi: 10.1016/j.jse.2017.01.011. Aug. [DOI] [PubMed] [Google Scholar]
- 30.Lopiz Y., García-Fernández C., Arriaza A., Rizo B., Marcelo H., Marco F. Midterm outcomes of bone grafting in glenoid defects treated with reverse shoulder arthroplasty. J Shoulder Elb Surg. 2017;26(9):1581–1588. doi: 10.1016/j.jse.2017.01.017. Sep. [DOI] [PubMed] [Google Scholar]
- 31.Tashjian R.Z., Granger E., Chalmers P.N. Structural glenoid grafting during primary reverse total shoulder arthroplasty using humeral head autograft. J Shoulder Elb Surg. 2018 Jan;27(1):e1–e8. doi: 10.1016/j.jse.2017.07.010. [DOI] [PubMed] [Google Scholar]
- 32.Gupta A., Thussbas C., Koch M., Seebauer L. Management of glenoid bone defects with reverse shoulder arthroplasty-surgical technique and clinical outcomes. J Shoulder Elb Surg. 2018;27(5):853–862. doi: 10.1016/j.jse.2017.10.004. May. [DOI] [PubMed] [Google Scholar]
- 33.Mahylis J.M., Puzzitiello R.N., Ho J.C., Amini M.H., Iannotti J.P., Ricchetti E.T. Comparison of radiographic and clinical outcomes of revision reverse total shoulder arthroplasty with structural versus nonstructural bone graft. J Shoulder Elb Surg. 2019;28(1):e1–e9. doi: 10.1016/j.jse.2018.06.026. Jan. [DOI] [PubMed] [Google Scholar]
- 34.Lawrence C., Williams G.R., Namdari S. Influence of glenosphere design on outcomes and complications of reverse arthroplasty: a systematic review. Clin Orthop Surg. 2016;8(3):288–297. doi: 10.4055/cios.2016.8.3.288. Sep; [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ferle M., Pastor M.F., Hagenah J., Hurschler C., Smith T. Effect of the humeral neck-shaft angle and glenosphere lateralization on stability of reverse shoulder arthroplasty: a cadaveric study. J Shoulder Elb Surg. 2019;28(5):966–973. doi: 10.1016/j.jse.2018.10.025. May. [DOI] [PubMed] [Google Scholar]
- 36.Langohr G.D., Giles J.W., Athwal G.S., Johnson J.A. The effect of glenosphere diameter in reverse shoulder arthroplasty on muscle force, joint load, and range of motion. J Shoulder Elb Surg. 2015;24(6):972–979. doi: 10.1016/j.jse.2014.10.018. Jun. [DOI] [PubMed] [Google Scholar]