Abstract
Aims
To report early (two-year) postoperative findings from a randomized controlled trial (RCT) investigating disease-specific quality of life (QOL), clinical, patient-reported, and radiological outcomes in patients undergoing a total shoulder arthroplasty (TSA) with a second-generation uncemented trabecular metal (TM) glenoid versus a cemented polyethylene glenoid (POLY) component.
Methods
Five fellowship-trained surgeons from three centres participated. Patients aged between 18 and 79 years with a primary diagnosis of glenohumeral osteoarthritis were screened for eligibility. Patients were randomized intraoperatively to either a TM or POLY glenoid component. Study intervals were: baseline, six weeks, six-, 12-, and 24 months postoperatively. The primary outcome was the Western Ontario Osteoarthritis Shoulder QOL score. Radiological images were reviewed for metal debris. Mixed effects repeated measures analysis of variance for within and between group comparisons were performed.
Results
A total of 93 patients were randomized (46 TM; 47 POLY). No significant or clinically important differences were found with patient-reported outcomes at 24-month follow-up. Regarding the glenoid components, there were no complications or revision surgeries in either group. Grade 1 metal debris was observed in three (6.5%) patients with TM glenoids at 24 months but outcomes were not negatively impacted.
Conclusion
Early results from this RCT showed no differences in disease-specific QOL, radiographs, complication rates, or shoulder function between uncemented second-generation TM and cemented POLY glenoids at 24 months postoperatively. Revision surgeries and reoperations were reported in both groups, but none attributed to glenoid implant failure. At 24 months postoperatively, Grade 1 metal debris was found in 6.5% of patients with a TM glenoid but did not negatively influence patient-reported outcomes. Longer-term follow-up is needed and is underway.
Cite this article: Bone Jt Open 2021;2(9):728–736.
Introduction
Total shoulder arthroplasty (TSA) is an effective surgical procedure to manage patients suffering from end-stage shoulder arthritis with an intact rotator cuff. Issues related to the glenoid implant, however, remain a common complication affecting both clinical and patient-based outcomes.1,2
TSA is often deemed a success or failure based on revision surgery, but what happens if the patient’s expectations are not met? Restoring a patient’s quality of life (QOL) is an important treatment goal of TSA and warrants consideration.3 QOL is a multidimensional concept that considers various aspects of life such as physical, mental, emotional, and social functioning.4 Developed in 2001, the Western Ontario Osteoarthritis of the Shoulder (WOOS) score is a rigorously validated disease-specific QOL instrument. It was designed for patient evaluation but also for research, specifically, to serve as the primary outcome in clinical trials involving glenohumeral osteoarthritis (OA).5,6
All-polyethylene glenoid implants have been used for decades in primary TSA. A porous tantalum biomaterial, trabecular metal (TM) component was developed to address problems seen with the cemented polyethylene implants, notably chronic loosening. The first-generation TM glenoid was recalled due to high failure rates,7 and a second-generation, constructed from the same materials (ultra-high-molecular-weight polyethylene (UHMWPE) and TM) was introduced but with two distinctly different design features. The UHMWPE portion was compression-moulded into a TM keel, with a cavity within each peg to expand the area of interdigitation creating a monoblock design, and an anterior and posterior peg were added increasing the number of pegs from three to five.8 The second-generation design is the subject of our study (Figure 1).
Fig. 1
The second-generation Trabecular Metal glenoid.
Published studies comparing all-polyethylene and TM glenoid components are retrospective in design; some involve the first-generation TM implant, and although most report joint-specific or general health outcomes, none report disease-specific QOL using the WOOS questionnaire.8-14 The authors designed a randomized controlled trial to compare patient-reported outcomes including the WOOS, as well as clinical and radiological outcomes in patients undergoing a TSA with a second-generation uncemented TM glenoid versus a cemented polyethylene (POLY) glenoid at five years postoperatively. The purpose of the current paper is to report early (two-year) findings from this trial with a null hypothesis that there will be no statistical difference in disease-specific QOL WOOS scores between groups at two years postoperative.
Methods
Approval to conduct this randomized controlled trial was granted by the University of British Columbia Clinical Research Ethics Board and the study was registered with clinicaltrials.gov (Identifier: NCT01539122). Five fellowship-trained surgeons (PYKC, WDR, DP, ZDZ, and FL) from three centres participated. Patients were screened for eligibility and had surgery between June 2012 and December 2016 (Table I). Final eligibility was determined intraoperatively prior to randomization.
Table I.
Study inclusion and exclusion criteria.
| Inclusion |
|---|
| Aged between 19 and 79 yrs |
| Diagnosis of primary glenohumeral osteoarthritis |
| Exclusion |
| Significant glenoid bone loss on a preoperative CT scan or intraoperatively* |
| Major joint trauma |
| Avascular necrosis |
| Rotator cuff or inflammatory arthropathy |
| Chronic dislocations |
| Massive cuff tear |
| Previous shoulder surgery (other than arthroscopic debridement or acromioplasty) |
| Active joint or systemic infection, muscle paralysis |
| Charcot arthropathy |
| Life expectancy < 2 yrs |
| Unable to read or speak English |
-
*
Final assessment made intraoperatively prior to randomization.
Surgical procedures
Patients were positioned in a beach chair position and received a general anaesthetic with or without a brachial plexus regional block. A deltopectoral approach was used, and biceps tenodesis performed. The subscapularis tendon was treated according to surgeon preference (tenotomized or peel off the lesser tuberosity). Osteophytes were removed and the humeral head osteotomized. The humeral canal was progressively reamed, and the trial-broaching stem left in situ to protect the proximal humerus during glenoid exposure. Once complete glenoid exposure was achieved, the glenoid bone stock was assessed for eligibility in the study. If adequate bone stock was identified (adequate defined as a contained defect suitable for an anatomical keeled implant), patients were randomized to either an uncemented second-generation TM glenoid (Zimmer Biomet, USA) or a cemented all-polyethylene glenoid (Zimmer Biomet). Randomization was performed using a data centre (EmPOWER Health Research, Canada) and stratified by surgeon. All patients received an uncemented Bigliani-Flatow humeral prosthesis (Zimmer Biomet). Placement of the glenoid component was carried out and no cement was used for the TM glenoid. The subscapularis was repaired and the incision closed.
Postoperatively, patients in both groups were admitted to hospital overnight and discharged the next day with routine postoperative instructions and rehabilitation protocols. Follow-up intervals were: six weeks, six-, 12-, and 24 months postoperatively. Patients were blinded to their treatment assignment.
Outcomes
The primary outcome was the Western Ontario Osteoarthritis of the Shoulder (WOOS) Index score. The WOOS is a patient-reported, disease-specific instrument measuring quality of life.6 An overall score from 0 to 100 is obtained where 0 represents a major dysfunction in shoulder-related quality of life and 100 represents the best possible score.
Secondary patient-reported outcome measures (PROMs) including the American Shoulder and Elbow Surgeons (ASES) score,15 the 12-Item Short Form Health Survey Questionnaire (SF-12),16 and the EuroQol-five dimensions (EQ-5D).17 The ASES is a shoulder-specific functional assessment tool and is scored out of 100 points, where 100 signifies maximum function and no pain. The SF-12 is patient-reported measure of global health status. Two summary scales can be derived. The Mental Component Summary (MCS) is based on questions related to vitality, social functioning, emotional, and mental health items while the Physical Component Summary (PCS) is based on physical functioning, bodily pain, physical, and general health items.18 The EQ-5D is a patient-reported, standardized measure of health-related quality of life.
Clinical assessment
Clinical examination included active range of motion (ROM), shoulder strength, and stability. A goniometer was used for ROM measurements. Complications, revisions, and reoperations were also documented. Revision surgery was defined as partial or complete arthroplasty of an original prosthesis due to component failure or infection, and reoperation was defined as surgery to address a soft-tissue problem.19
Radiological evaluation
Plain radiological images included anteroposterior (AP), axillary, and lateral views, and were evaluated by the treating surgeon for radiolucency, as well as implant subsidence, position, and migration. Radiolucent lines (RLLs) around the glenoid implant and humeral stem were categorized as present/absent and severity reported in millimetres. Metal debris was assessed by one surgeon, the principal investigator (PYKC), according to Endrizzi et al9 (Table II). Glenoid morphology was classified according to Walch on preoperative CT images.20
Table II.
Endrizzi et al9 metal debris classification system.
| Grade 0 | No evidence of metallic debris |
|---|---|
| Grade 1 | Debris at the bone-metal interface |
| Grade 2 | Intra-articular soft tissue metallic debris |
| Grade 3 | Visible, but incomplete fracture of the metal component |
| Grade 4 | Component breakage and displacement |
Sample size calculation
Sample size was based on the following parameters: the primary outcome (WOOS), a two-sided test, significance level of 0.05, and power of 0.8. The final endpoint was 24 months postoperatively. Estimates for the WOOS mean, standard deviation (SD), and a minimally clinical importance difference of 15% were based on a similar population of patients published by JOINTS Canada.21 These parameters resulted in the need for 34 patients per treatment group. An additional 18 patients per group were recruited to account for pre- and post-randomization attrition resulting in 104 patients in total.
Statistical analysis
Continuous variables were described via means and SDs, and quantiles, while categorical variables were described per frequency tables. Descriptive statistics were generated overall as well as stratified by group and study interval. Comparisons of primary and secondary outcome variables, within and between groups, were made via mixed effects repeated measures analysis of variance (ANOVA). Standardized residuals demonstrated normality per normal quantile-quantile (Q-Q) plots. All analyses were performed using SAS v9.4 (SAS Institute, USA). A p-value < 0.05 was considered statistically significant.
Results
A total of 93 patients were randomized intraoperatively to either a TM (46) or POLY (47) glenoid component. Patient flow through the study is presented in Figure 2. Patient demographic details, previous surgery, and glenoid morphology were similar between groups. (Table III)
Fig. 2
CONSORT flow diagram.
Table III.
Patient demographic details.
| Characteristic | TM (46) | POLY (47) |
|---|---|---|
| Sex, n (%) | ||
| Male | 24 (52) | 23 (49) |
| Female | 22 (48) | 24 (51) |
| Mean age, yrs (SD) | 66.5 (6.4) | 68.4 (5.5) |
| Mean BMI, kg/m2 (SD) | 28.8 (5.8) | 28.3 (5.3) |
| Mean duration of symptoms, mths (SD) | 57.8 (49.5) | 63.6 (57.7) |
| Dominant arm, n (%) | ||
| Left | 6 (13) | 5 (11) |
| Right | 38 (83) | 41 (87) |
| Ambidextrous | 1 (2) | 1 (2) |
| Not reported | 1 (2) | 0 (0) |
| Operative limb, n (%) | ||
| Right | 21 (46) | 27 (57) |
| Left | 25 (54) | 20 (43) |
| Smoking status, n (%) | ||
| No | 44 (96) | 46 (98) |
| Yes | 2 (4) | 1 (2) |
| Alcohol status, n (%) | ||
| No | 22 (48) | 18 (38) |
| Yes | 24 (52) | 28 (60) |
| Not reported | 0 | 1 |
| Previous surgery on study shoulder, n (%) | ||
| No | 45 (98) | 45 (96) |
| Yes | 1 (2) | 2 (4) |
| Previous surgery type, n | ||
| Tendon shortening | 1 | N/A |
| Subacromial decompression | N/A | 1 |
| Debridement | N/A | 1 |
| Walch classification of glenoid morphology, n (%) | ||
| A1 | 8 (17) | 9 (19) |
| A2 | 6 (13) | 5 (10.5) |
| B1 | 4 (9) | 5 (10.5) |
| B2 | 27 (59) | 28 (60) |
| Not reported | 1 | 0 |
-
Demographics of those eligible but not randomized (n = 11): mean age 65.5 yrs (SD 8.3); male/female: 7/4; mean BMI 32.3 kg/m2 (SD 6.7).
-
N/A, not applicable; SD, standard deviation.
Clinically significant improvements from baseline to 24 months postoperatively were seen in both groups across all patient-reported outcomes (Table IV). No statistically significant differences in WOOS scores were observed between groups at 24 months, however a statistical difference was detected at six months postoperatively in favour of the TM group (TM = 90.8 vs POLY = 84.9; p = 0.030, mixed effects repeated measures ANOVA). Interestingly, findings with SF-12 Mental scores were similar to the WOOS. A significant difference was found at six months postoperatively favouring the TM group (TM = 59.0 vs POLY = 55.3; p = 0.013, mixed effects repeated measures ANOVA), however no other differences were detected. With respect to the ASES, SF-12 Physical, and EQ-5D index and VAS, no statistical or clinically important differences were detected between groups during the 24-month follow-up period.
Table IV.
Patient-reported outcome measures for each group over time.
| Mean outcome (SD) | TM | POLY | Adjusted MD (95% CI) | p-value* |
|---|---|---|---|---|
| WOOS | ||||
| Preop | 32.1 (21) | 26.6 (15) | 5.5 (-1.9 to 13.0) | 0.143 |
| 6 wks | 60.3 (21) | 59.1 (17) | 1.2 (-6.7 to -9.1) | 0.763 |
| 6 mths | 90.8 (10) | 84.9 (14) | 5.9 (0.6 to 10.8) | 0.030 |
| 12 mths | 90.4 (16) | 91.3 (14) | -0.9 (-6.6 to 5.5) | 0.848 |
| 24 mths | 91.9 (13) | 93.1 (11) | -1.2 (-6.1 to 3.8) | 0.641 |
| ASES Patient Portion | ||||
| Preop | 35.8 (16) | 37.1 (14) | -1.3 (-8.7 to 6.0) | 0.716 |
| 6 wks | 61.4 (19) | 64.1 (15) | -2.7 (-9.9 to 4.3) | 0.437 |
| 6 mths | 86.2 (16) | 81.9 (14) | 4.3 (-2.2 to 10.8) | 0.187 |
| 12 mths | 88.7 (14) | 89.0 (11) | -0.3 (-5.1 to 5.3) | 0.971 |
| 24 mths | 88.8 (15) | 89.1 (13) | -0.3 (-6.0 to 5.7) | 0.961 |
| SF-12 Physical | ||||
| Preop | 35.8 (8) | 36.5 (7) | 0.7 (-4.3 to -2.3) | 0.551 |
| 6 wks | 39.9 (9) | 38.5 (7) | 1.4 (-2.0 to 4.5) | 0.461 |
| 6 mths | 48.5 (9) | 46.8 (9) | 1.7 (-2.9 to 4.6) | 0.659 |
| 12 mths | 47.0 (11) | 49.1 (8) | -2.1 (-5.7 to 2.4) | 0.415 |
| 24 mths | 47.8 (11) | 47.8 (8) | 0.0 (-3.6 to 4.4) | 0.840 |
| SF-12 Mental | ||||
| Preop | 54.1 (10) | 52.9 (10) | 1.2 (-2.9 to 5.5) | 0.549 |
| 6 wks | 56.6 (9) | 55.7 (8) | 0.9 (-2.8 to 4.2) | 0.701 |
| 6 mths | 59.0 (5) | 55.2 (8) | 3.8 (0.8 to 6.6) | 0.013 |
| 12 mths | 56.5 (6) | 55.7 (8) | 0.8 (-1.9 to 3.9) | 0.505 |
| 24 mths | 56.5 (7) | 56.2 (6) | 0.3 (-2.0 to 3.2) | 0.656 |
| EQ-5D Index Scale | ||||
| Preop | 0.62 (0.2) | 0.64 (0.2) | -0.02 (-0.10 to 0.06) | 0.615 |
| 6 wks | 0.79 (0.1) | 0.80 (0.1) | -0.01 (-0.06 to -0.03) | 0.587 |
| 6 mths | 0.90 (0.1) | 0.86 (0.1) | 0.04 (-0.07 to -0.08) | 0.100 |
| 12 mths | 0.89 (0.1) | 0.88 (0.1) | 0.02 (-0.03 to -0.07) | 0.475 |
| 24 mths | 0.88 (0.1) | 0.89 (0.1) | -0.01 (-0.06 to -0.05) | 0.778 |
| EQ-5D VAS Slider | ||||
| Preop | 72.0 (17.4) | 72.3 (15.9) | -0.3 (-7.0 to 6.5) | 0.944 |
| 6 wks | 72.4 (23.8) | 78.9 (15.0) | -6.5 (-14.6 to 1.8) | 0.127 |
| 6 mths | 83.7 (15.9) | 83.6 (13.4) | 0.1 (-6.4 to 5.8) | 0.923 |
| 12 mths | 78.5 (16.8) | 81.9 (18.1) | -3.4 (-10.0 to 4.6) | 0.458 |
| 24 mths | 80.0 (15.3) | 83.0 (9.9) | -3.0 (-7.6 to 3.2) | 0.421 |
-
*
Mixed effects repeated measures analysis of variance.
-
ASES, American Shoulder and Elbow Surgeons score; CI, confidence interval; EQ-5D, EuroQol five-dimension questionnaire; MD, mean between-groups difference; POLY, polyethylene component; SD, standard deviation; SF-12, 12-Item Short Form Health Survey; TM, trabecular metal; VAS, visual analogue scale; WOOS, Western Ontario Osteoarthritis of the Shoulder score.
Active forward elevation and external rotation improved in both groups from baseline and there were no clinically significant differences between groups at any postoperative interval (Table V).
Table V.
Active range of motion for operative and nonoperative arm by group over time.
| Timepoint | Mean TM ROM, ° (SD) | Mean POLY ROM, ° (SD) | ||
|---|---|---|---|---|
| Forward elevation | Operative | Non-op | Operative | Non-op |
| Preop | 103 (25) | 144 (20) | 103 (29) | 146 (25) |
| 6 wks | 90 (21) | 145 (26) | 98 (25) | 143 (27) |
| 6 mths | 132 (16) | 145 (22) | 129 (19) | 146 (20) |
| 12 mths | 138 (16) | 144 (26) | 138 (16) | 145 (17) |
| 24 mths | 136 (21) | 141 (28) | 142 (17) | 147 (20) |
| External rotation (at side) | ||||
| Preop | 14 (18) | 47 (22) | 19 (22) | 47 (23) |
| 6 wks | 26 (16) | 47 (21) | 23 (17) | 46 (24) |
| 6 mths | 45 (19) | 50 (22) | 47 (26) | 51 (27) |
| 12 mths | 49 (18) | 57 (32) | 46 (20) | 49 (19) |
| 24 mths | 52 (19) | 47 (22) | 50 (19) | 53 (29) |
| External rotation (abduction) | ||||
| Preop | 17 (30) | 56 (28) | 19 (30) | 52 (28) |
| 6 wks | 19 (25) | 58 (27) | 18 (23) | 52 (32) |
| 6 mths | 50 (24) | 60 (26) | 47 (24) | 55 (28) |
| 12 mths | 53 (25) | 56 (30) | 54 (21) | 62 (23) |
| 24 mths | 58 (25) | 56 (30) | 60 (21) | 62 (24) |
-
POLY, polyethylene component; ROM, range of motion; SD, standard deviation; TM, trabecular metal.
Table VI presents strength at baseline and 24 months postoperatively. The greatest difference between groups was seen with external rotation at 24 months, where 57% of TM patients versus 67% of POLY patients achieved full strength. Six complications (12.8%) occurred in the POLY group and five (10.9%) in the TM group (Table VII). No glenoid implant failures or revisions were observed in either group.
Table VI.
Strength (Grade 1 to 5) by group at baseline and 24 months postoperatively.
| Variable | TM, n (%) | POLY, n (%) | ||
|---|---|---|---|---|
| Forward elevation | Baseline | 24 mths | Baseline | 24 mths |
| 1 | 0 | 1 (2) | 0 | 0 |
| 2 | 3 (7) | 0 | 1 (2) | 0 |
| 3 | 6 (13) | 0 | 10 (21) | 0 |
| 4 | 25 (54) | 10 (22) | 21 (45) | 13 (28) |
| 5 | 9 (17) | 31 (67) | 11 (23) | 31 (66) |
| Not reported | 4 (9) | 4 (9) | 4 (9) | 3 (6) |
| Abduction | ||||
| 1 | 0 | 1 (2) | 0 | 0 |
| 2 | 4 (9) | 0 | 1 (2) | 0 |
| 3 | 8 (17) | 0 | 14 (30) | 1 (2) |
| 4 | 21 (46) | 14 (30) | 19 (40) | 12 (26) |
| 5 | 9 (20) | 27 (60) | 9 (19) | 31 (66) |
| Not reported | 4 (9) | 4 (9) | 4 (9) | 3 (6) |
| External rotation | ||||
| 1 | 0 | 1 (2) | 0 | 0 |
| 2 | 5 (10) | 0 | 2 (4) | 0 |
| 3 | 6 (13) | 1 (2) | 11 (23) | 0 |
| 4 | 23 (50) | 14 (30) | 25 (55) | 13 (28) |
| 5 | 8 (17) | 26 (57) | 4 (9) | 31 (67) |
| Not reported | 4 (8) | 4 (9) | 4 (9) | 3 (6) |
| Internal rotation | ||||
| 1 | 0 | 1 (2) | 0 | 0 |
| 2 | 2 (4) | 0 | 3 (7) | 0 |
| 3 | 7 (15) | 0 | 11 (23) | 0 |
| 4 | 20 (44) | 9 (20) | 18 (38) | 12 (26) |
| 5 | 13 (28) | 32 (70) | 11 (23) | 32 (68) |
| Not reported | 4 (9) | 4 (9) | 4 (9) | 3 (6) |
| Pain with testing? | ||||
| No | 3 (7) | 37 (80) | 7 (15) | 39 (83) |
| Yes | 39 (85) | 5 (11) | 36 (77) | 5 (11) |
| Not reported | 4 (9) | 4 (9) | 4 (9) | 3 (6) |
-
Not reported - missing or patient couldn’t perform.
-
POLY, polyethylene component; TM, trabecular metal.
Table VII.
Complications and, when relevant, modes of failure for each group.
| Group | Age, yrs | Time of AE, mths | Complication type (traumatic or atraumatic; humeral or glenoid implant) | Treatment and outcome | WOOS score at AE | WOOS score at 24 mths | Metal debris |
|---|---|---|---|---|---|---|---|
| TM | 64 | 0 Intraop |
Periprosthetic fracture; atraumatic; humeral stem |
Nonoperative; fracture healed |
24.2 | 100 | None |
| TM | 62 | 6 | Periprosthetic fracture and instability; traumatic (patient fell); posterior subluxation and humeral fracture | Reoperation; ORIF and inferior capsule shift |
71.9 | 76.0 | None |
| TM | 67 | 21 | Loosening - aseptic; atraumatic; humeral stem | Revision to cemented humeral stem (glenoid: no change) | 56.8 | 56.8 | None |
| TM | 71 | 22 | Instability; atraumatic (posterior dislocation while patient was stretching); humeral | Nonoperative; reduced;no further treatment |
91.2 | 91.2 | None |
| TM | 67 | 24 | Loosening - aseptic; atraumatic; humeral stem | Revision to cemented humeral stem (glenoid: no change) | 50.0 | 50.0 | None |
| POLY | 75 | 0 Intraop |
Periprosthetic fracture; atraumatic; glenoid | Nonoperative | 55.5 | 100 | N/A |
| POLY | 61 | 2 | Subscapularis and supraspinatous tear; traumatic (patient fell) | Revision; rotator cuff repair and downsizing of humeral head | 55.6 | 96.8 | N/A |
| POLY | 78 | 3 | Periprosthetic fracture; atraumatic (patient was doing exercises and felt a pop); greater tubercle fracture | Nonoperative | 6.8 | 91.6 | N/A |
| POLY | 67 | 3 | Subscapularis and supraspinatous tear; atraumatic | Reoperation; conversion to rTSA. |
53.1 | 97.7 | N/A |
| POLY | 67 | 5 | Periprosthetic fracture; traumatic (accident); humeral stem only | Nonoperative | 49.3 | 94.7 | N/A |
| POLY | 59 | 5 | Subscapularis failure; traumatic | Reoperation; major shoulder contracture release + subscapularis repair | 16.5 | 42.8 | N/A |
-
AE, adverse event; N/A, not applicable; ORIF, open reduction and internal fixation; POLY, polyethylene component; rTSA, reverse total shoulder arthroplasty; TM, trabecular metal; WOOS, Western Ontario Osteoarthritis Shoulder.
Grade 1 metal debris was detected in three patients (6.5%) at 24 months. There was no evidence of debris in these patients prior to the 24-month interval and their mean WOOS scores were not negatively impacted (mean 24-month WOOS score = 89.7 (standard deviation (SD) 11)).
There was no radiological evidence of glenoid component migration in either group. At 12 months postoperatively, one patient in the POLY group had superior migration of the humeral implant in keeping with a rotator cuff tear. RLL around the glenoid implant were observed in eight patients (17%) in the POLY group but the severity was minor in most patients (6 = 1 mm; 1 = 2 mm; 1 = 6 mm). In the TM group, ‘possible’ lucency around the glenoid was reported in one patient.
With respect to the humeral stem, subsidence was observed in four (8.5%) patients in the POLY group and seven (15.2%) in the TM group and RLLs were observed in three (6.4%) and five (10.9%) patients in the POLY and TM groups, respectively.
Discussion
Early postoperative findings from this randomized trial comparing an uncemented second-generation TM versus cemented POLY glenoid showed no clinically relevant differences with respect to disease-specific QOL, pain, or shoulder function at two years postoperatively. No revisions or reoperations were attributed to the glenoid in either group during the 24-month study period, and complication rates were similar between groups. Metal debris was identified in 6.5% of patients (n = 3) with a TM glenoid however the severity of debris was of the lowest grade (Grade 1) and did not negatively influence QOL or shoulder function.
While data from retrospective studies and case series are valuable and necessary, randomized trials provide more powerful data and are needed to determine efficacy of a new technique or implant. Results from our randomized trial support the findings from seven retrospective case-series involving the second-generation TM glenoid. These studies had longer follow-up periods, ranging from 38 to 80 months, but their conclusions were consistent with our trial: there were no catastrophic implant failures with the second-generation TM implants, and good patient outcomes were reported with respect to pain, satisfaction, and function.8-14 The patient-reported outcomes documented in these studies were either joint-specific or general health measures and are valuable for broad comparisons. However, none reported the disease-specific QOL WOOS, which has been referenced by some as the best instrument for assessing shoulder arthroplasty.22
Our trial is the first to compare a second-generation TM versus POLY glenoid using WOOS data and demonstrates that disease-specific QOL significantly improves with both implants after TSA, and that no differences between implants were detected at two years postoperatively. At six months postoperatively, a statistical difference in the WOOS and SF-12 Mental scores favoured the TM group. The clinical relevance of 5.9 points on the WOOS and 3.8 on the SF-12 Mental Score is unclear, particularly when all other clinical and functional outcomes were comparable between groups at all other timepoints.
It is well documented that TSA survivorship declines over time. The Australian Orthopaedic Association National Joint Arthroplasty Registry (AOANJRR) 2019 Annual Report documented TSA revision rates (all prostheses) of: 2.9% (two yrs), 7.8% (five yrs), 12.4% (ten yrs), and 14.9% (12 yrs).23 These rates are similar to an American prospective cohort, involving 2,588 TSAs performed between 1976 to 2008, that showed revision rates of 5.8%, 9.8%, and 18.6% at five, ten, and 20 years, respectively.24 Another large American-based cohort study involving 5,566 patients and over 30 glenoid designs offers evidence that implant design is associated with differential risks of revision.25 Systematic review and registry data have shown higher failure rates with metal-backed glenoids versus all-polyethylene implants, and that uncemented glenoids are five times more likely to require revision than cemented.23,26,27 The importance of exercising caution when interpreting and comparing these cohort studies and registry data cannot be overstated. Many of the datasets include first-generation non-porous metal-backed components. Further, they include a diverse range of patient characteristics, implantation techniques, and follow-up periods. A recent systematic review highlights these factors by reporting no difference in failure rates between cemented polyethylene glenoids and ‘modern’ metal-backed glenoids in the mid-term follow-up period (< 36 months), however in the longer term (> 72 months) the modern metal-backed glenoids had lower rates of failure, RRLs, and loosening.28
The issue of periprosthetic metal debris is a growing concern with TM components. Metal debris preceded implant failure with some first-generation TM and metal-backed designs, and warrants monitoring.9,19 Our study found a lower rate of debris (6.5%) and severity (Grade 1) when compared to the literature, and might be attributed to our shorter follow-up period. Chen et al10 identified debris in 24% of patients in their series at a mean 80 months with four patients categorized as Grade 1 and one patient as Grade 2. A subsequent retrospective study by the same author explored differences in metal debris between cemented and uncemented second-generation TM glenoids.11 Although there was no difference in debris rate between groups, the cemented glenoids had more severe debris (Grade 2 and 4) than uncemented (Grade 1 to 2). Our early findings are similar to those seen in Chen et al’s11 uncemented cohort. Endrizzi et al9 observed Grade 1 to 3 metal debris in 44% of patients at five years postoperatively. The incidence and severity of debris increased over time, and there was no relationship between the presence of debris and outcomes, radiolucency, glenoid size, or preoperative Walch classification. Watson et al13 reported 11% of patients with metal debris at 34 months’ follow-up, however the authors did not use the Endrizzi grading system to categorize debris severity. The correlation between higher rates of debris and longer-term follow-up is further supported by an unpublished retrospective series that documented 18 of 40 (45%) of patients with metal debris at 65 months’ follow-up (Grade 1 = 15, Grade 2 = 3).29 Metal debris was not correlated with worse outcomes in all studies mentioned above, and this is consistent with our trial. A correlation between metal debris and implant failure may exist, but requires longer-term follow-up.
The presence or progression of RLL at the implant-bone interface following TSA can denote component loosening. RLLs were not appreciated with the uncemented TM implant in our trial, which was in direct contrast with the 17% RLL seen in our cemented POLY group, and also with Chen et al,11 who reported statistically greater RLLs with uncemented versus cemented TM glenoids (64% and 29%, respectively). Endrizzi et al9 also reported RLL with cemented TM implants at a rate of 36%. These differences may be attributed to a variety of factors such as glenoid wear, implantation technique, or follow-up duration.
Revising TM components can be difficult and deserves consideration. Despite the many benefits of biological fixation, it can make prosthesis removal extremely difficult resulting in significant bone loss.13,30 Additionally, one cannot leave the metal tray in place and only replace the polyethylene implant due to the TM monoblock design.30-32
Limitations of this study include the short-term follow-up period, variability with multiple surgeons performing the surgeries, and only one surgeon performing the metal debris evaluations. Future studies should consider reliability of metal debris assessment. Longer-term follow-up is needed to understand implant survivorship and whether RLL or metal debris rates increase over time.
In conclusion, early results from this randomized controlled trial comparing an uncemented TM versus cemented POLY glenoid showed no differences with respect to disease-specific QOL, radiological outcomes, complication rates, or shoulder function between groups. Revision surgeries and reoperations were reported in both groups, but none attributed to glenoid implant failure. Grade 1 metal debris was found in 6.5% of patients with a TM glenoid but did not negatively influence QOL. Longer-term follow-up is needed and is currently underway.
References
1. Gauci MO , Cavalier M , Gonzalez JF , et al. Revision of failed shoulder arthroplasty: epidemiology, etiology, and surgical options . J Shoulder Elbow Surg . 2020 ; 29 ( 3 ): 541 – 549 . Crossref PubMed Google Scholar
2. Hsu JE , Hackett DJ , KV V , Matsen FA . What can be learned from an analysis of 215 glenoid component failures? J Shoulder Elbow Surg . 2018 ; 27 ( 3 ): 478 – 486 . Crossref PubMed Google Scholar
3. Hasan SS , Leith JM , Campbell B , Kapil R , Smith KL , Matsen FA . Characteristics of unsatisfactory shoulder arthroplasties . J Shoulder Elbow Surg . 2002 ; 11 ( 5 ): 431 – 441 . Crossref PubMed Google Scholar
4. Cella DF . Quality of life: concepts and definition . J Pain Symptom Manage . 1994 ; 9 ( 3 ): 186 – 192 . Crossref PubMed Google Scholar
5. Kirkley A , Griffin S , Dainty K . Scoring systems for the functional assessment of the shoulder . Arthroscopy . 2003 ; 19 ( 10 ): 1109 – 1120 . Crossref PubMed Google Scholar
6. Lo IK , Griffin S , Kirkley A . The development of a disease-specific quality of Life measurement tool for osteoarthritis of the shoulder: The western Ontario Osteoarthritis of the shoulder (WOOS) index . Osteoarthritis Cartilage . 2001 ; 9 ( 8 ): 771 – 778 . Crossref PubMed Google Scholar
7. Budge MD , Nolan EM , Heisey MH , Baker K , Wiater JM . Results of total shoulder arthroplasty with a monoblock porous tantalum glenoid component: a prospective minimum 2-year follow-up study . J Shoulder Elbow Surg . 2013 ; 22 ( 4 ): 535 – 541 . Crossref PubMed Google Scholar
8. Merolla G , Chin P , Sasyniuk TM , Paladini P , Porcellini G . Total shoulder arthroplasty with a second-generation tantalum trabecular metal-backed glenoid component: Clinical and radiographic outcomes at a mean follow-up of 38 months . Bone Joint J . 2016 ; 98-B ( 1 ): 75 – 80 . Crossref PubMed Google Scholar
9. Endrizzi DP , Mackenzie JA , Henry PDG . Early debris formation with a porous tantalum glenoid component: Radiographic analysis with 2-year minimum follow-up . J Bone Joint Surg Am . 2016 ; 98-A ( 12 ): 1023 – 1029 . Crossref PubMed Google Scholar
10. Chen RE , Brown AM , Greenstein AS , Mannava S , Voloshin I . Radiographic and clinical outcomes of second generation Trabecular Metal glenoid components in total shoulder arthroplasty . Shoulder & Elbow . 2020 ; 13 ( 4 ): 416 – 425 . Google Scholar
11. Chen RE , Brown AM , Greenstein AS , Miller RJ , Mannava S , Voloshin I . Cemented versus uncemented fixation of second-generation trabecular metal glenoid components: Minimum 5-year outcomes . J Shoulder Elbow Surg . 2021 ; 30 ( 4 ): e147 – e156 . Crossref PubMed Google Scholar
12. Panti JP , Tan S , Kuo W , Fung S , Walker K , Duff J . Clinical and radiologic outcomes of the second-generation Trabecular MetalTM glenoid for total shoulder replacements after 2-6 years follow-up . Arch Orthop Trauma Surg . 2016 ; 136 ( 12 ): 1637 – 1645 . Google Scholar
13. Watson ST , Gudger GK , Long CD , Tokish JM , Tolan SJ . Outcomes of Trabecular Metal-backed glenoid components in anatomic total shoulder arthroplasty . J Shoulder Elbow Surg . 2018 ; 27 ( 3 ): 493 – 498 . Crossref PubMed Google Scholar
14. Styron JF , Marinello PG , Peers S , Seitz WH . Survivorship of trabecular metal anchored glenoid total shoulder arthroplasties . Tech Hand Up Extrem Surg . 2016 ; 20 ( 3 ): 113 – 116 . Crossref PubMed Google Scholar
15. Richards RR , An KN , Bigliani LU , et al. A standardized method for the assessment of shoulder function . J Shoulder Elbow Surg . 1994 ; 3 ( 6 ): 347 – 352 . Crossref PubMed Google Scholar
16. Ware J , Kosinski M , Keller SD . A 12-item Short-Form Health survey: construction of scales and preliminary tests of reliability and validity . Med Care . 1996 ; 34 ( 3 ): 220 – 233 . Crossref PubMed Google Scholar
17. EuroQol Research Foundation . EQ-5D-3L user guide . 2018 . https://euroqol.org/publications/user-guides ( date last accessed 5 August 2021 ). Google Scholar
18. Ware JE , Kosinski M , Keller SD . SF-12: How to Score The SF-12 Physical and Mental Health Summary Scales . Boston, MA : The Health Institute, New England Medical Centre . 1995 . Google Scholar
19. Boileau P , Moineau G , Morin-Salvo N , Avidor C , Godenèche A , Lévigne C , et al. Metal-backed glenoid implant with polyethylene insert is not a viable long-term therapeutic option . J Shoulder Elbow Surg . 2015 ; 24 ( 10 ): 1534 – 1543 . Google Scholar
20. Walch G , Badet R , Boulahia A , Khoury A . Morphologic study of the glenoid in primary glenohumeral osteoarthritis . J Arthroplasty . 1999 ; 14 ( 6 ): 756 – 760 . Crossref PubMed Google Scholar
21. Litchfield RB , McKee MD , Balyk R , Mandel S , Holtby R , Hollinshead R , et al. Cemented versus uncemented fixation of humeral components in total shoulder arthroplasty for osteoarthritis of the shoulder: a prospective, randomized, double-blind clinical trial-A JOINTs Canada Project . J Shoulder Elbow Surg . 2011 ; 20 ( 4 ): 529 – 536 . Crossref PubMed Google Scholar
22. Wright RW , Baumgarten KM . Shoulder outcomes measures . J Am Acad Orthop Surg . 2010 ; 18 ( 7 ): 436 – 444 . Crossref PubMed Google Scholar
23. Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) . Hip, Knee & Shoulder Arthroplasty: Annual Report . 2019 . https://aoanjrr.sahmri.com/annual-reports-2019 ( date last accessed 5 August 2021 ). Google Scholar
24. Singh JA , Sperling JW , Cofield RH . Revision surgery following total shoulder arthroplasty: analysis of 2588 shoulders over three decades (1976 to 2008 ). J Bone Joint Surg Br . 2011 ; 93-B ( 11 ): 1513 – 1517 . Crossref PubMed Google Scholar
25. Dillon MT , Chan PH , Prentice HA , Burfeind WE , Yian EH , Singh A , et al. The association between glenoid component design and revision risk in anatomic total shoulder arthroplasty . J Shoulder Elbow Surg . 2020 ; 29 ( 10 ): 2089 – 2096 . Crossref PubMed Google Scholar
26. Papadonikolakis A , Matsen FA . Metal-backed glenoid components have a higher rate of failure and fail by different modes in comparison with all-polyethylene components: A systematic review . J Bone Joint Surg Am . 2014 ; 96-A ( 12 ): 1041 – 1047 . Crossref PubMed Google Scholar
27. Sharplin PK , Frampton CMA , Hirner M . Cemented vs. uncemented glenoid fixation in total shoulder arthroplasty for osteoarthritis: a New Zealand Joint Registry study . J Shoulder Elbow Surg . 2020 ; 29 ( 10 ): 2097 – 2103 . Crossref PubMed Google Scholar
28. Kim DM , Aldeghaither M , Alabdullatif F , Shin MJ , Kholinne E , Kim H , et al. Loosening and revision rates after total shoulder arthroplasty: a systematic review of cemented all-polyethylene glenoid and three modern designs of metal-backed glenoid . BMC Musculoskelet Disord . 2020 ; 21 ( 1 ): 114 . Crossref PubMed Google Scholar
29. Yeoh J , Chin P , Regan W , Sasyniuk TM , Sayre E , Lim B . Results of second-generation trabecular metal glenoid total shoulder arthroplasties . In: Bone Joint J Montreal, QC . 2019 . Google Scholar
30. Gauci MO , Bonnevialle N , Moineau G , Baba M , Walch G , Boileau P . Anatomical total shoulder arthroplasty in young patients with osteoarthritis: all-polyethylene versus metal-backed glenoid . Bone Joint J . 2018 ; 100-B ( 4 ): 485 – 492 . Crossref PubMed Google Scholar
31. Castagna A , Delcogliano M , de Caro F , Ziveri G , Borroni M , Gumina S , et al. Conversion of shoulder arthroplasty to reverse implants: clinical and radiological results using a modular system . Int Orthop . 2013 ; 37 ( 7 ): 1297 – 1305 . Crossref PubMed Google Scholar
32. Katz D , Kany J , Valenti P , Sauzières P , Gleyze P , El Kholti K . New design of a cementless glenoid component in unconstrained shoulder arthroplasty: a prospective medium-term analysis of 143 cases . Eur J Orthop Surg Traumatol . 2013 ; 23 ( 1 ): 27 – 34 . Crossref PubMed Google Scholar
Author contributions
P. Y. K. Chin: Conceptualization, Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
W. D. Regan: Conceptualization, Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
D. Plausinis: Conceptualization, Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
Z. Zarzour: Conceptualization, Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
F. Leung: Conceptualization, Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
T. Sasyniuk: Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
B. Lim: Investigation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing.
E. C. Sayre: Formal analysis, Writing – original draft, Writing – review & editing.
Funding statement
The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other non-profit organization with which one or more of the authors are associated.
ICMJE COI statement
P. Y. K. Chin reports consultancy fees from Zimmer Biomet. Z. Zarzour, W. D. Regan, B. Lim, D. Plausinis, F. Leung, T. Sasyniuk, and P. Y. K. Chin report an institutional grant from Zimmer Biomet to fund the research for this study.
Open access funding
Open access funding was provided by a Zimmer Biomet grant via the Vancouver Costal Health Research Institute (F12-01182).
Acknowledgements
The authors wish to acknowledge the contributions of Alyssa Andrew, Linh Cajigas, Krista Clark, Kayla Johnston, Diana Hart-McConnell, Jinelle Lim, Amy MacDonald, Dan MacLeod, and Giselle Roy for their assistance with data collection.
Ethical review statement
University of British Columbia Clinical Research Ethics Board, H12-00323.
© 2021 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
