Abstract
Background:
There is currently no national total shoulder arthroplasty (TSA) database in Canada. As a first step toward a national registry, a regional shoulder database was initiated in 2017. In this study, we describe the implementation of and initial findings from the shoulder database and patient-reported outcomes program.
Methods:
The registry is intended to capture all shoulder arthroplasty procedures in the province of Manitoba and includes surgeon-reported operative details and data on patient-reported outcome measures (PROMs). Every surgeon submits data. We included primary and revision procedures performed since 2017 in the retrospective case series study. We assessed registry coverage by the rate of return of surgeon-completed operative forms and patient-reported outcomes questionnaires, respectively. We determined the incidence of revision by the number of primary procedures linked to a revision within 1 year of surgery.
Results:
A total of 1044 TSA procedures occurred during the study period. Overall, 65.0% were anatomic (n = 679) and 35.0% were reverse TSA procedures (n = 365). Of the 1044 surgeries that took place, the capture rate was 92.0% (n = 960) for operative data. The capture rate for PROM questionnaires was 78.6% preoperatively and 65.8% postoperatively. Four primary procedures were linked with a revision within 1 year of surgery (0.4%). The most frequent diagnosis reported was degenerative arthritis (n = 558/817). High satisfaction (n = 569/636) and improved joint-specific and general quality-of-life PROMs were reported at 1 year.
Conclusion:
The provincial shoulder database demonstrates the early stages of a registry, which contains useful, granular data and is an opportunity to fill an important gap in Canadian arthroplasty data.
Abstract
Contexte:
Il n’existe actuellement pas de base de données nationale sur les arthroplasties totales de l’épaule (ATE) au Canada. Une base de données régionale sur l’épaule a été mise en place en 2017 afin de servir de première étape à un registre national. Dans la présente étude, nous décrivons la mise en œuvre de la base de données sur l’épaule et d’un programme de résultats rapportés par la patientèle ainsi que les premiers résultats relatifs à ces initiatives.
Méthodes:
Le registre a pour objectif d’enregistrer toutes les arthroplasties de l’épaule effectuées au Manitoba ainsi que les détails relatifs à l’intervention chirurgicale rapportés par le chirurgien ou la chirurgienne et les données relatives aux mesures des résultats déclarés par les patients et patientes (MRDPP). Chaque chirurgien ou chirurgienne soumet ses données. Nous avons inclus dans cette étude de série de cas rétrospective les chirurgies primaires et de révision effectuées depuis 2017. Nous avons évalué la couverture du registre au moyen du taux de retour des formulaires d’intervention chirurgicale remplis par les chirurgiens et chirurgiennes et des questionnaires sur les résultats déclarés par la patientèle, respectivement. Nous avons déterminé l’incidence des chirurgies de révision au moyen du nombre de chirurgies primaires liées à une révision dans l’année suivant l’intervention chirurgicale.
Résulats:
Au total, 1044 ATE ont été effectuées au cours de la période de l’étude. En tout, 65,0 % de celles-ci ont été réalisées au moyen d’une prothèse anatomique (n = 670) et 35,0 %, avec une prothèse inversée (n = 365). Parmi les 1044 interventions réalisées, le taux de saisie de données relatif à l’intervention chirurgicale était de 92,0 % (n = 960), et celui relatif aux questionnaires de MRDPP était de 78,6 % avant l’intervention et de 65,8 % après celle-ci. Quatre chirurgies primaires ont été liées à une révision dans l’année suivant l’intervention (0,4 %). Le diagnostic le plus souvent mentionné était l’arthrose (n = 558/817). On a noté une satisfaction élevée (n = 569/636) ainsi qu’une amélioration des MRDPP quant à l’articulation visée et de la qualité de vie en général 1 an après l’intervention chirurgicale.
Conclusion:
La base de données provinciale sur l’épaule constitue l’un des stades précoces d’un registre; elle contient des données granulaires et utiles et constitue une occasion de combler un besoin important quant aux données canadiennes sur l’arthroplastie.
Joint replacement registries have contributed to improved outcomes in arthroplasty for more than 50 years. They track the performance of surgeries by connecting intraoperative details to various metrics, including patient demographic characteristics, revision, reoperation, and complications.1 Registries have demonstrated benefit to multiple partners: patients, health care providers, and policy-makers all benefit from the cycle of continuous improvement encouraged by registry programs. The inclusion of patient-reported outcome measures (PROMs) in many registries is supporting a shift toward value-based, patient-centric care.2–5 Globally, hip and knee replacement registries are widespread. The rate of total shoulder arthroplasty (TSA) has increased substantially over the last decade; however, there are relatively few programs established to register and analyze the outcomes of shoulder replacement procedures.6–9 Additionally, there are few reports in the literature that describe the implementation of shoulder replacement registry programs.6,7,10,11 In Canada, hip and knee arthroplasty procedures have been systematically documented within the Canadian Joint Replacement Registry (CJRR) since 2001.3 There is currently no national TSA database in Canada. As a first step toward a national initiative, a regional shoulder registry was initiated in the Canadian province of Manitoba in 2017.
The Manitoba Joint Replacement Registry (MJRR) is a regional medical registry established in 2001 that stores and maintains clinical, surgical, and prosthesis information related to all hip and knee surgeries performed in Manitoba. The program expanded to include shoulder arthroplasties in April 2017. The registry data set includes surgeon-reported operative details and patient-reported data. The registry is intended to capture all shoulder arthroplasty procedures in the province of 1.5 million residents. All TSAs in Manitoba are performed out of 2 major institutions by 6 fellowship-trained shoulder surgeons who have all committed to submitting their cases to the MJRR. The MJRR is publicly funded through the provincial health care system. An advisory committee is responsible for directing the MJRR and is composed of local surgeons, provincial decision-makers, and other relevant representatives. The MJRR operations are managed by a designated team of administrative staff who are responsible for registry data collection, entry, and maintenance. A summary of registry data is provided annually in a report to each participating surgeon. Registry data are also reported to the provincial orthopedic standards and quality committee with the goal of supporting surgeons, hospitals, and regions in improving the quality of care they provide.
In this study, we aimed to describe the implementation and initial findings of the shoulder database and PROMs program.
Methods
This was a retrospective case series study in Manitoba. We included all primary and revision anatomic TSA and reverse TSA procedures between Apr. 1, 2017, and Mar. 31, 2022. We excluded hemiarthroplasty and fracture cases on initial findings owing to the known differences in postoperative outcomes.12,13
Registry PROMs database consent and eligibility
Patient-reported data are self reported on a voluntary basis and collected independently from operative data. Consent to participate in the PROMs program occurs at the time of questionnaire completion, and all patients are contacted with the opportunity to participate in the PROMs program at each time point.
Definition of revision
The MJRR maintains a list of all scheduled procedures, which is reconciled with surgeon-provided data on a monthly basis. The registry collects information on all revision cases. Revision is defined as any reoperation to insert or replace 1 or more implant components. Revision cases linked to a registered primary procedure are tracked to determine the development of new revision cases over time.
Operative data collection and elements
Operative details are collected at the time of surgery for all eligible patients. Variables include procedure side, anesthesia, diagnosis, type of replacement, prosthesis used, revision or reoperation status, surgical approach, antibiotics, and deep vein thrombosis prophylaxis (Appendix 1, available at www.canjsurg.ca/lookup/doi/10.1503/cjs.014624/tab-related-content). In addition to manually completed procedural details, device labels (stickers), which feature unique identifiers (barcodes) associated with implants, are collected and scanned into the MJRR.
Registry data are manually reviewed for completion by the registry team and included in the patient chart. Each operative form requires a surgeon signature following completion; registry forms that do not have a signature are flagged by the team for review. Each participating surgeon is required to revise and sign flagged operative reports before inclusion in the chart. Data are returned to a central site, where they are entered, stored, and maintained within a single central database on an ongoing basis. Each procedure is assigned a unique registry index number, which links each patient questionnaire back to the same patient and procedure.
Patient-reported outcomes
Patient-reported data are voluntary and collected from each consenting patient using paper questionnaires (Table 1). Patients are provided forms preoperatively in the preadmission clinic and contacted 1 year after the procedure via mail-out. We included 1-year follow-up data associated with TSA procedures between Apr. 1, 2017, and Mar. 31, 2022. Return envelopes with paid postage are provided with mail-out questionnaires, and patient response is voluntary. The distribution and collection of PROMs questionnaires and operative data are a collaborative effort between registry and hospital staff. Data on PROMs are manually entered on an ongoing basis by the registry team, and all information is stored in a central database. The selections for both generic and joint-specific PROMs tools meet recommendations outlined by the International Society of Arthroplasty Registries (ISAR) PROMs Working Group.5,14 Validated PROMs questions are routinely used in patients undergoing TSA.2,15–17 A list of all procedures performed within the study time frame served as the denominator for patient and surgeon participation. The capture rate of the registry (%) was determined by the frequency of surgeon and patient questionnaires returned within the study time frame. All blank questionnaires were considered missing. Incomplete data were recorded as entered on the form. The incidence of revision in the registry was determined by the number of primary procedures linked to a revision procedure within the first year following surgery. All patients included in the study were eligible for 1-year follow-up. Patients who died before 1 year were excluded from follow-up analysis.
Table 1.
Patient-reported variables and collection time points
| Variable | Validated | Version | Time point | |
|---|---|---|---|---|
| Preoperative | Postoperative 1 year | |||
| Consent | N | – | ✓ | ✓ |
| Date completed | N | – | ✓ | ✓ |
| Height | N | – | ✓ | |
| Weight | N | – | ✓ | |
| Medical history | N | Asked to specify conditions by selecting all that apply: heart disease; high blood pressure; lung disease; diabetes; ulcer or stomach disease; kidney disease; liver disease; anemia or other blood disease; cancer; depression; osteoarthritis or degenerative arthritis other than shoulder/elbow; back pain, rheumatoid arthritis; other medical problem, please specify | ✓ | |
| ASES score (x/100.00) | Y | Modified (1994) Visual analogue scale Activities of daily living |
✓ | ✓ |
| EQ-5D score (x/1.000) | Y | Version EQ-5D-5L (2009) Original language |
✓ | ✓ |
| Satisfaction | N | Asked to answer, What is your overall satisfaction with this shoulder replacement surgery? Very satisfied; Satisfied; Neutral (neither Satisfied nor Unsatisfied); Unsatisfied; Very unsatisfied. |
✓ | |
| Complications | N | Asked to specify, Yes/No/Don’t know for the following: Dislocation of this shoulder requiring treatment by a doctor; Infection of this shoulder, requiring oral or intravenous antibiotics; Infection of this shoulder requiring surgery; Further surgery for problems with this shoulder replacement. |
✓ | |
ASES = American Shoulder and Elbow Surgeons; EQ-5D = EuroQol 5-Dimension; EQ-5D-5L = EuroQol 5-Dimension, 5-Level; N = no; Y = yes.
We independently reviewed surgeon and patient-reported data. We stratified data by procedure type (primary v. revision, and anatomic v. reverse TSA). We used a 2-sided 2-sample t test for the comparison of continuous variables and a χ2 test or Fisher exact test for categorical variables. The patient-reported, multi-question outcome measures American Shoulder and Elbow Surgeons (ASES) and EuroQol 5-Dimension (EQ-5D) required a response to all questions in the survey to calculate the score. We used a Canadian time-trade-off derived value set to convert the multi-question EQ-5D results into a single index score.18 We assessed the difference between baseline and the 1-year follow-up for the continuous outcome measures ASES and EQ-5D with a 2-sided paired t test for patients completing both the preoperative and postoperative questionnaires. We reviewed missing patient-reported data to explore potential bias in the PROMs program. Each patient was categorized as a respondent or nonrespondent based on the return of PROMs questionnaires at baseline and at 1 year following surgery. We compared respondents and nonrespondents with a 2-sided 2-sample t test for the comparison of continuous variables and a χ2 test or Fisher exact test for categorical variables. We conducted all statistics using IBM SPSS Statistics software version 28.00, with the significance level set at a p value less than 0.05.
Results
We identified 1044 patients in the MJRR who had a TSA during the study period (Table 2); 679 were anatomic TSAs (primary n = 625; revision n = 54) and 365 were reverse TSAs (primary n = 334; revision n = 31). Out of the 1044 surgeries identified by the MJRR, 960 operative reports were completed (92.0%). Preoperatively, 821 patients returned PROMs questionnaires (78.6%). Five patients died before 1-year follow-up (1039 had 1-year follow-up). Compared with preoperative PROMs, fewer questionnaires (n = 684) were returned 1 year post-operatively (65.8%), and 574 patients (55.2%) completed both preoperative and postoperative questionnaires. Four revision procedures were linked with a registered primary procedure in the first year following surgery (n = 4/959), which is 0.4% of patients.
Table 2.
Summary of all total shoulder arthroplasty procedures captured (reported by surgeons) in the database from Apr. 1, 2017, to Mar. 31, 2022
| Variable | No. (%) of procedures* | p value† | ||
|---|---|---|---|---|
| All | Anatomic TSA | Reverse TSA | ||
| Procedure count | 1044 | 679 | 365 | 0.8 |
| Primary | 0959 (91.9) | 625 (92.0) | 334 (91.5) | |
| Revision | 85 (8.1) | 54 (8.0) | 31 (8.5) | |
| Age, yr, mean ± SD | 69.8 ± 9.6 | 68.4 ± 9.7 | 72.2 ± 8.9 | < 0.001 |
| Sex, female | 558 (53.4) | 340 (50.1) | 218 (59.7) | 0.003 |
| Procedure side, left | 470 (45.0) | 335 (49.3) | 135 (37.0) | < 0.001 |
| Diagnosis‡ | ||||
| Unreported | 227 | 117 | 110 | |
| Reported | 817 | 562 | 255 | |
| Degenerative arthritis | 558 (68.3) | 482 (85.8) | 76 (29.8) | < 0.001 |
| Inflammatory arthritis | 19 (2.3) | 9 (1.6) | 10 (3.9) | 0.04 |
| Congenital problem | 6 (0.7) | 1 (0.2) | 5 (2.0) | 0.01 |
| Cuff tear arthropathy | 187 (22.9) | 45 (8.0) | 142 (55.7) | < 0.001 |
| Malunion/non-union | 6 (0.7) | 3 (0.5) | 3 (1.2) | 0.4 |
| Post-traumatic osteoarthritis | 18 (2.2) | 11 (2.0) | 7 (2.7) | 0.5 |
| Other§ | 45 (5.5) | 26 (4.6) | 19 (7.4) | 0.1 |
SD = standard deviation; TSA = total shoulder arthroplasty.
Unless stated otherwise.
p values were obtained from 2-sample 2-sided t tests for continuous variables, the χ2 test was used for categorical variables, and the Fisher exact test was applied where expected cell counts were insufficient.
Surgeon participation was voluntary; response may not be available for all variables.
Includes osteonecrosis, infection, tumour, and any other diagnosis not listed (Appendix 1, available at www.canjsurg.ca/lookup/doi/10.1503/cjs.014624/tab-related-content).
Demographic data recovered from electronic medical records and surgeon-reported items are outlined in Table 2. A higher proportion of patients undergoing anatomic TSA were diagnosed with degenerative arthritis than patients undergoing reverse TSA (85.8% v. 29.8%; p < 0.001). A higher proportion of patients undergoing reverse TSA were diagnosed with inflammatory arthritis (3.9% v. 1.6%; p = 0.04), a congenital problem (2.0% v. 0.2%; p = 0.01), and cuff tear arthropathy (55.7% v. 8.0%; p < 0.001). Eighty-five revision procedures were captured in the registry in the first 5 years.
Patient-reported data are outlined in Table 3. A significant improvement (pre- to postoperatively) was observed in the ASES and EQ-5D scores of primary (ASES mean change 44.8, 95% confidence interval [CI] 42.4 to 47.3; p < 0.001; EQ-5D mean change 0.187, 95% CI 0.169 to 0.204; p < 0.001) and revision groups (ASES mean change 19.0, 95% CI 7.9 to 30.2; p < 0.001; EQ-5D mean change 0.131, 95% CI 0.045 to 0.218; p = 0.002).
Table 3.
Patient-reported data available for primary and revision total shoulder arthroplasty procedures from Apr. 1, 2017, to Mar. 31, 2022 (1-year follow-up data up to Oct. 31, 2023)
| Variable | Preoperatively; no. (%) of procedures* | Postoperatively; no. (%) of procedures* | ||||
|---|---|---|---|---|---|---|
|
|
|
|||||
| Primary | Revision | p value† | Primary | Revision | p value† | |
| BMI, mean ± SD | 32.6 ± 7.5 (n = 487) | 31.26 ± 6.4 (n = 33) | 0.3 | – | – | |
|
| ||||||
| ASES (x/100), mean ± SD | 32.6 ± 15.7 (n = 565) | 34.1 ± 19.8 (n = 40) | 0.6 | 77.47 ± 21.84 (n = 541) | 53.64 ± 25.18 (n = 38) | < 0.001 |
|
| ||||||
| EQ-5D (x/1.000), mean ± SD | 0.563 ± 0.233 (n = 724) | 0.548 ± 0.247 (n = 51) | 0.6 | 0.774 ± 0.195 (n = 617) | 0.696 ± 0.214 (n = 37) | 0.02 |
|
| ||||||
| Morbidity burden‡ | 768 | 53 | 0.5 | – | – | |
|
| ||||||
| 0 | 214 (27.9) | 19 (35.8) | – | – | ||
|
| ||||||
| 1 | 101 (13.2) | 5 (9.4) | – | – | ||
|
| ||||||
| 2 | 135 (17.6) | 12 (22.8) | – | – | ||
|
| ||||||
| 3 | 157 (20.4) | 9 (17.0) | – | – | ||
|
| ||||||
| ≥ 4 | 161 (21.0) | 8 (15.1) | – | – | ||
|
| ||||||
| Satisfaction | – | – | – | 636 | 40 | < 0.001 |
|
| ||||||
| Very unsatisfied | – | – | 24 (3.8) | 4 (10.0) | ||
|
| ||||||
| Unsatisfied | – | – | 10 (1.6) | 4 (10.0) | ||
|
| ||||||
| Neutral | – | – | 33 (5.2) | 7 (17.5) | ||
|
| ||||||
| Satisfied | – | – | 162 (25.5) | 9 (22.5) | ||
|
| ||||||
| Very satisfied | – | – | 407 (64.0) | 16 (40) | ||
|
| ||||||
| Postoperative complications | – | – | 623 | 39 | ||
|
| ||||||
| Dislocation | – | – | – | 18 (2.9) | 4 (10.3) | 0.1 |
|
| ||||||
| Superficial infection — antibiotics | – | – | 9 (1.4) | 3 (7.7) | 0.07 | |
|
| ||||||
| Deep infection — surgery | – | – | 6 (1.0) | 1 (2.6) | 0.4 | |
|
| ||||||
| Future surgery | – | – | 15 (2.4) | 7 (17.9) | 0.001 | |
ASES = American Shoulder and Elbow Surgeons; BMI = body mass index; EQ-5D = EuroQol 5-Dimension; SD = standard deviation.
Unless stated otherwise.
p values were obtained from 2-sample 2-sided t tests for continuous variables, the χ2 test was used for categorical variables, and the Fisher exact test was applied where expected cell counts were insufficient.
Patient reported per Table 1.
No baseline differences existed between respondents and nonrespondents at 1 year. Nonrespondents at 1 year were younger (68.0 v. 70.7; p < 0.001), reported different morbidity burden (p = 0.02), and reported lower quality of life (EQ-5D 0.477 v. 0.591; p < 0.001) preoperatively than respondents (Table 4 and Table 5).
Table 4.
A comparison of questionnaire respondents and nonrespondents 1 year postoperatively using demographic information and preoperative patient-reported items captured in the database from Apr. 1, 2017, to Mar. 31, 2022
| Variable | No. (%) of respondents n = 684 |
Mean ± SD | No. (%) of non-respondents n = 355 |
Mean ± SD | 95% CI | p value* |
|---|---|---|---|---|---|---|
| Age, yr | 684 | 70.7 ± 8.6 | 355 | 68.0 ± 11.0 | −4.05 to −1.42 | < 0.001 |
| Sex | 684 | 355 | 0.4 | |||
| Female | 360 (52.6) | 196 (55.2) | ||||
| Male | 324 (47.4) | 159 (44.8) | ||||
| BMI | 385 | 32.1 ± 7.3 | 135 | 33.6 ± 7.9 | −0.026 to 2.89 | 0.054 |
| Morbidity burden† | 574 | 244 | 0.02 | |||
| 0 | 168 (29.3) | 65 (26.6) | ||||
| 1 | 60 (10.5) | 46 (18.9) | ||||
| 2 | 107 (18.6) | 39 (16.0) | ||||
| 3 | 123 (21.4) | 42 (17.2) | ||||
| ≥ 4 | 116 (20.2) | 52 (21.3) | ||||
| Preoperative ASES | 430 | 33.4 ± 16.1 | 172 | 31.0 ± 15.8 | −5.17 to 0.501 | 0.1 |
| Preoperative EQ-5D | 580 | 0.591 ± 0.227 | 192 | 0.477 ± 0.235 | −0.151 to −0.076 | < 0.001 |
ASES = American Shoulder and Elbow Surgeons; BMI = body mass index; CI = confidence interval; EQ-5D = EuroQol 5-Dimension; SD = standard deviation.
p values were obtained from 2-sample 2-sided t tests for continuous variables, the χ2 test was used for categorical variables, and the Fisher exact test was applied where expected cell counts were insufficient.
Patient reported per Table 1.
Table 5.
A comparison of missing baseline data based on questionnaire response at the 1-year follow-up time point
| Variable | No. (%) of respondents n = 821 |
Mean ± SD | No. (%) of non-respondents n = 223 |
Mean ± SD | 95% CI | p value* |
|---|---|---|---|---|---|---|
| Age, yr | 821 | 69.7 ± 9.3 | 223 | 70.2 ± 10.7 | −0.924 to 1.91 | 0.5 |
| Sex | 821 | 223 | 0.5 | |||
| Female | 434 (52.9) | 124 (55.6) | ||||
| Male | 387 (47.1) | 99 (44.4) | ||||
| Satisfaction | 569 | 107 | 0.7 | |||
| Very unsatisfied | 23 (4.0) | 5 (4.7) | ||||
| Unsatisfied | 13 (2.3) | 1 (0.9) | ||||
| Neutral (neither satisfied nor dissatisfied) | 32 (5.6) | 8 (7.5) | ||||
| Satisfied | 141 (24.8) | 30 (28.0) | ||||
| Very satisfied | 360 (63.3) | 63 (58.9) | ||||
| 1-year ASES | 483 | 76.3 ± 22.6 | 96 | 73.9 ± 24.0 | −7.43 to 2.59 | 0.3 |
| 1-year EQ-5D | 549 | 0.773 ± 0.191 | 105 | 0.749 ± 0.224 | −0.066 to 0.017 | 0.2 |
ASES = American Shoulder and Elbow Surgeons; CI = confidence interval; EQ-5D = EuroQol 5-Dimension; SD = standard deviation.
p values were obtained from 2-sample 2-sided t tests for continuous variables, the χ2 test was used for categorical variables, and the Fisher exact test was applied where expected cell counts were insufficient.
Discussion
The establishment and maintenance of registry data comes with a plethora of administrative and logistical challenges. The exponential increase in TSA procedures over the last decade is an example of how registry programs must evolve with advancements in medical technology and practice to remain relevant and useful. The MJRR experience is not unique and speaks to the effort involved in establishing and maintaining an arthroplasty registry program. The ISAR was established in 2004 with the goal of supporting registry development and promoting global alignment of registry data.1,14 This study shares the groundwork of a successful shoulder registry program that aligns with the literature, other national registry programs, and ISAR recommendations. These early findings may contribute to international conversations regarding registry data and TSA.
The fundamental purpose of the MJRR shoulder program is to accumulate high volumes of accurate and useful arthroplasty-related data. Administrative framework, surgeon buy-in, and human resources were all vital components of the registry groundwork. The data collected by the MJRR shoulder program are the product of initial planning. All shoulder surgeons in the province were actively engaged in the registry, and 92.0% (960/1044) of TSA surgeries were captured in the database. The capture rates of PROMs were relatively high; the questionnaire return rate was 78.6% preoperatively and 65.8% 1 year postoperatively.19,20
Completeness of registry data is a constant challenge, for both small and large programs. A recent report from the American Academy of Orthopaedic Surgeons Shoulder and Elbow Registry (AAOS SER) states an approximate 20% linkage of preoperative data with follow-up ASES scores.21 The successful patient and provider engagement observed in the MJRR shoulder program is likely a result of its integration with standard-of-care activities. The MJRR shoulder database benefits from the well-established administrative workflows in place as part of the pre-existing hip and knee program. Despite early success, the compliance for MJRR operative reports on a per-variable, question-by-question basis varied within the MJRR (e.g., approximately 20% of reports were missing a diagnosis [Table 2; n = 227 diagnosis unreported]), and 55% of patients completed both pre-operative and postoperative surveys. Most arthroplasty registry programs resort to quality-control tactics, such as systematic auditing, to improve compliance.2 The national CJRR hip and knee arthroplasty registry has successfully tied registry reporting to surgical billing as an incentive for surgeon participation.3 To improve the completeness and accuracy of data captured, both billing incentives and systematic auditing would be valuable next steps for the MJRR. To improve patient engagement in the PROMs collection, a mechanism to follow up on missed responses would be valuable. Electronic PROMs have recently been explored as a method to reduce administrative burden of PROMs.4 Advancements in technology and regulation across Canadian provincial health care systems mean that electronic SMS or email-based PROMs collection may also be a viable option for the MJRR.
Rates of revision reported in literature for TSA vary owing to differences in definition, follow-up time frame, patient population, and shoulder arthroplasty procedure examined. The early findings in the MJRR indicate a low rate of early revision at 1 year (0.4%).19,21 Long-term follow-up registry data on revision successfully supports device monitoring and improved outcomes in shoulder, hip, and knee arthroplasty.19,20,22 The linkage of primary and revision cases for long-term monitoring in the MJRR will be a valuable resource as the database matures.
The surgical data captured in the MJRR validate the successfully established registry workflow. Global alignment and harmonization of data collection and reporting has been explored in several registries, with the goal of pooling data for collaboration.1,7,14 This study demonstrates valuable early alignment with other established shoulder programs.
The MJRR reported a lower proportion of reverse TSA procedures than anatomic TSA procedures (365 v. 679), and the most frequent diagnosis for TSA procedure was degenerative osteoarthritis, followed by cuff arthropathy (Table 2). Our findings align with recent Canadian research by Pasache and colleagues from an institutional shoulder registry.23 Indications for TSA have expanded drastically over the last 15 years, in particular owing to the increasing usage of reverse TSA.6,23 Reverse TSA has shown great utility, not only in the treatment of revision arthroplasty, rotator cuff deficiency, and bone loss, but increasingly in the treatment of primary osteoarthritis.24,25 In the United States, the utilization of primary reverse TSA has increased by 191.3% between 2011 and 2017.9 Internationally, the rapid growth of TSA, and specifically reverse TSA procedures, has resulted in considerable variability in shoulder arthroplasty reporting.6 Even among high-volume databases in the United Kingdom, US, and Australia, the primary indication for procedures and the prevalence of anatomic TSA and reverse TSA procedures differ.19–21 Early data from the MJRR provide further context to these dramatic shifts in TSA usage, both within Canada and internationally.
The ISAR reviewed PROMs data collection in member registries in 2019 and released recommendations for PROMs selection and reporting in 2021. Patient-reported outcome measures (ASES and EQ-5D) in TSA are well studied in the literature and used in registry programs around the world. The EQ-5D was reported the most popular generic PROM in use by the ISAR, and the ASES score is collected by the large, national, and North American program the AAOS SER.2,21 All validated, patient-reported outcomes in the initial review of MJRR data demonstrated clinically significant improvement at the 1-year postoperative time point (ASES minimal clinically important difference [MCID] 14 to 21; EQ-5D MCID 0.027 to 0.209).15–17,26–28 The satisfaction reported for primary TSA (89.5%) patients in the MJRR was similar to reports from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) (87% to 90%) at 6 months following surgery.19 The AOANJRR is the largest shoulder arthroplasty registry worldwide.29 One-year outcomes from the MJRR were consistently lower for revision procedures than primary procedures (Table 3), in line with other reports.30–32 Very few complications were reported at the 1-year post-operative follow-up (n = 63; 1% to 3%); particularly for patients undergoing revision TSA (n = 15; 3% to 18%). Since the inception of the database, registry research indicates the presence of complications are not accurately reported by patients who underwent hip and knee arthroplasty.33 Linkage to health records or provider-reported complications would likely improve the accuracy of reported complications in the MJRR in the future. The completeness of PROMs is a factor in the representativeness of data and the voluntary nature of participation in registry PROMs data may result in bias.34 Per ISAR recommendations, the MJRR tracks PROMs response.2 The general homogeneity in characteristics of respondents and nonrespondents explored in this study speaks to the representativeness of the MJRR patient-reported data. A greater proportion of female patients, a different morbidity burden, and lower preoperative quality of life (EQ-5D) were identified in the nonrespondents group lost to 1-year follow-up, which should be considered in future clinical studies (Table 4).
Limitations
This study is limited to a single province in Canada with a population of 1.5 million people. Some of the administrative barriers and workflows discussed in the MJRR are specific to the provincial health care system and may not be applicable in other regions. The nonmandatory nature of registry data collection presented the greatest challenge in achieving robust patient and provider engagement. Further research will evaluate the role of a quality-control program at the MJRR and its influence on the capture rate and data set completeness. In particular, research is needed on a mechanism to follow-up on missed patient response in the PROMs program pre- and postoperatively.
Conclusion
The MJRR demonstrates the beginnings of a registry containing granular and useful TSA data. The greatest challenge faced in establishing the registry was patient engagement in the PROMs program, particularly at the 1-year follow-up time point. Nevertheless, the MJRR shoulder data set and early findings were both comparable to other large national programs. The MJRR experience speaks to the complexities involved in collecting and maintaining registry data. Systematic auditing and a mechanism to follow-up missed surgeon and patient responses will be next steps for the registry. The MJRR addresses an important gap in Canadian TSA data and will be a valuable starting place and model for expansion into a national joint replacement registry.
Supplementary Information
Footnotes
This work was presented at the International Society of Arthroplasty Registries annual meeting in June 2023 and the Canadian Orthopaedic Association annual meeting in June 2024.
Competing interests: Jarret Woodmass declares consulting fees from Arthrex, ConMed, Smith and Nephew, and Stryker. Sarah Harris reports employment at the Pan Am Clinic Foundation and payment for her time working on the manuscript, as well as funding to present the findings at an annual meeting. Christiaan Righolt reports institutional funding from Pfizer Canada, DePuy Synthes, Zimmer Biomet, Smith and Nephew, Stryker, and Hip Innovation Technology; payment or honoraria from Smith and Nephew; unpaid positions with the International Society of Arthroplasty Registries and the Canadian Orthopaedic Association; and employment with the Orthopaedic Innovation Centre. Peter MacDonald reports grants or contracts from Smith and Nephew, Stryker, Arthrex, and ConMed, and consulting fees from Arthrex and ConMed. Eric Bohm reports institutional funding from DePuy Synthes, Zimmer Biomet, Smith and Nephew, Stryker, Hip Innovation Technology, Canadian Institutes of Health Research SPOR (Strategy for Patient-Oriented Research) SUPPORT (Support for People and Patient-Oriented Research and Trials) Unit, Arthritis Society Canada, and the Orthopaedic Innovation Centre; consulting fees, payment, or honoraria from Stryker; unpaid positions with the International Society of Arthroplasty Registries, the Canadian Arthroplasty Society, the Canadian Joint Replacement Registry Advisory Committee, the Manitoba Joint Replacement Registry, and the Manitoba Orthopaedic Standards and Quality Committee; and receipt of gifts from Smith and Nephew and the Orthopaedic Innovation Centre.
Contributors: Jarret Woodmass and Peter MacDonald contributed to the conception and design of the study. Sarah Harris, Sheila McRae, Christiaan Righolt, and Eric Bohm contributed to the analysis and interpretation of the data. Sarah Harris contributed to writing the article. Jarret Woodmass, Sheila McRae, Christiaan Righolt, Peter MacDonald, and Eric Bohm contributed to critical review of the article. All authors gave final approval of the version to be published and agreed to be accountable for all aspects of the work.
References
- 1.Malchau H, Garellick G, Berry D, et al. Arthroplasty implant registries over the past five decades: development, current, and future impact. J Orthop Res 2018;36:2319–30. [DOI] [PubMed] [Google Scholar]
- 2.Bohm ER, Kirby S, Trepman E, et al. Collection and reporting of patient-reported outcome measures in arthroplasty registries: multinational survey and recommendations. Clin Orthop Relat Res 2021; 479:2151–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Canadian Joint Replacement Registry (CJRR). Ottawa: Canadian Institute for Health Information. Available: www.cihi.ca/en/canadian-joint-replacement-registry-cjrr (accessed 2023 Aug. 10). [Google Scholar]
- 4.Heath EL, Ackerman I, Lorimer M, et al. National implementation of an electronic patient-reported outcome measures program for joint replacement surgery: pilot study. JMIR Form Res 2022;6:e30245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wilson I, Bohm E, Lübbeke A, et al. Orthopaedic registries with patient-reported outcome measures. EFORT Open Rev 2019;4:357–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lübbeke A, Rees J, Barea C, et al. Regional variation in shoudler arthroplasty: incidence, indication, type of procedure, and outcomes evaluation in 9 countries. Acta Orthop 2017;88:592–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Page RS, Navarro RA, Salomonsson B. Establishing an international shoulder arthroplasty consortium. J Shoulder Elbow Surg 2014;23:1081–2. [DOI] [PubMed] [Google Scholar]
- 8.Rasmussen JV, Olsen BS, Fevang BTS, et al. A review of national shoulder and elbow joint replacement registries. J Shoulder Elbow Surg 2012;21:1328–35. [DOI] [PubMed] [Google Scholar]
- 9.Wagner ER, Farley KX, Higgins I, et al. The incidence of shoulder arthroplasty: rise and future projections compared with hip and knee arthroplasty. J Shoulder Elbow Surg 2020;29:2601–9. [DOI] [PubMed] [Google Scholar]
- 10.Kriechling P, Waltenspul M, Bouaicha S, et al. Establishing an institutional reverse total shoulder arthroplasty registry. Obere Extrem 2021;16:265–71. [Google Scholar]
- 11.Parada SA, Flurin PH, Wright TW, et al. Comparison of complication types and rates associated with anatomic and reverse total shoulder arthroplasty. J Shoulder Elbow Surg 2021;30:811–8. [DOI] [PubMed] [Google Scholar]
- 12.Davies AR, Sabharwal S, Liddle AD, et al. Patient-reported outcomes following total shoulder arthroplasty and hemiarthroplasty: an analysis of data from the National Joint Registry. J Shoulder Elbow Surg 2024;33:2411–20. [DOI] [PubMed] [Google Scholar]
- 13.Richard GJ, Denard PJ, Kaar SG, et al. Outcome measures reported for the management of proximal humeral fractures: a systematic review. J Shoulder Elbow Surg 2020;29:2175–84. [DOI] [PubMed] [Google Scholar]
- 14.Rolfson O. Editorial comment: 7th International Congress of Arthroplasty Registries. Clin Orthop Relat Res 2019;477:1299–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Marks M, Grobet C, Audigé L. Validity, responsiveness and minimal important change of the EQ-5D-5L in patients after rotator cuff repair, shoulder arthroplasty or thumb carpometacarpal arthroplasty. Qual Life Res 2021;30:2973–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tashjian RZ, Hung M, Keener JD, et al. Determining the minimal clinically important difference for the American Shoulder and Elbow Surgeons score, Simple Shoulder Test, and visual analog scale (VAS) measuring pain after shoulder arthroplasty. J Shoulder Elbow Surg 2017;26:144–8. [DOI] [PubMed] [Google Scholar]
- 17.Werner BC, Chang B, Nguyen JT, et al. What change in American Shoulder and Elbow Surgeons Score represents a clinically important change after shoulder arthroplasty? Clin Orthop Relat Res 2016;474: 2672–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Xie F, Pullenayegum E, Gaebel K, et al. A time trade-off-derived value set of the EQ-5D-5L for Canada. Med Care 2016;54:98–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hip, knee & shoulder arthroplasty annual report 2022. Adelaide (AU): Australian Orthopaedic Association National Joint Replacement Registry; 2022. Available: aoanjrr.sahmri.com/annual-reports-2022 (accessed 2023 Aug. 22). [Google Scholar]
- 20.NJR 20th annual report 2023. Hemel Hempstead (UK): National Joint Registry; 2023. Available: reports.njrcentre.org.uk/portals/0/PDFdownlaords/NJR/20thAnnualReport2023 (accessed 2023 Sept. 28). [Google Scholar]
- 21.Shoulder and elbow registry annual report 2022. Rosemont (IL): AAOS American Academy of Orthopaedic Surgeons; 2022. Available: https://communications.aaos.org/download-the-ser-2022-annual-report?_ga=2.168346151.740666058.1772469470-915598121.1772469470 (accessed 2023 Aug. 22). [Google Scholar]
- 22.Annual report 2022: medical device surveillance and assessment. Oakland (CA): Kaiser Permanente; 2022. Available: https://deviceassessment.kaiserpermanente.org/wp-content/uploads/2023/07/2022_ExternalAnnualReport_SCPMG_MDSA.pdf (accessed 2023 Sept. 28). [Google Scholar]
- 23.Pasache Lozano RDP, Valencia Ramón EA, Johnston DG, et al. Trends in prevalence and implant types in the Nova Scotia Joint Database Registry between 2005 and 2021. Bone Jt Open 2023;4:567–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Craig RS, Lawrence TM. Anatomic or reverse shoulder arthroplasty: indications and decision-making. Orthop Trauma 2022;36:166–74. [Google Scholar]
- 25.Mayfield CK, Korber SS, Hwang NM, et al. Volume, indications, and number of surgeons performing reverse total shoulder arthroplasty continue to expand: a nationwide cohort analysis from 2016–2020. JSES Int 2023;7:827–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Galvin JW, Kim R, Ment A, et al. Outcomes and complications of primary reverse shoulder arthroplasty with minimum of 2 years’ follow-up: a systematic review and meta-analysis. J Shoulder Elbow Surg 2022;31:e534–44. [DOI] [PubMed] [Google Scholar]
- 27.Kennedy J, Klifto CS, Ledbetter L, et al. Reverse total shoulder arthroplasty clinical and patient-reported outcomes and complications stratified by preoperative diagnosis: a systematic review. J Shoulder Elbow Surg 2021;30:929–41. [DOI] [PubMed] [Google Scholar]
- 28.Saini SS, Pettit R, Puzzitiello RN, et al. Clinical outcomes after reverse total shoulder arthroplasty in patients with primary glenohumeral osteoarthritis compared with rotator cuff tear arthropathy: Does preoperative diagnosis make a difference? J Am Acad Orthop Surg 2022;30:e415–22. [DOI] [PubMed] [Google Scholar]
- 29.Aveledo R, Holland P, Thomas M, et al. A comparison of the minimum data sets for primary shoulder arthroplasty between national shoulder arthroplasty registries. Is international harmonization feasible? Shoulder Elbow 2019;11:48–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kim K, Elbuluk A, Jia N, et al. Revision shoulder arthroplasty: patient-reported outcomes vary according to the etiology of revision. J Orthop 2018;15:922–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ravi V, Murphy RJ, Moverley R, et al. Outcome and complications following revision shoulder arthroplasty: a systematic review and meta-analysis. Bone Jt Open 2021;2:618–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sajadi KR, Kwon YW, Zuckerman JD. Revision shoulder arthroplasty: an analysis of indications and outcomes. J Shoulder Elbow Surg 2010;19:308–13. [DOI] [PubMed] [Google Scholar]
- 33.Heo SM, Naylor JM, Harris IA, et al. Reliability of patient-reported complications following hip or knee arthroplasty procedures. BMC Med Res Methodol 2019;19:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Harris IA, Cashman K, Lorimer M, et al. Are responders to patient health surveys representative of those invited to participate? An analysis of the Patient-Reported Outcome Measures Pilot from the Australian Orthopaedic Association National Joint Replacement Registry. PLoS ONE 2021;16:e0254196. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.