Abstract
Background
Since its approval, reverse total shoulder arthroplasty (rTSA) has continued to increase in usage, with expanding indications beyond rotator cuff arthropathy. Existing literature has captured further increased utilization over the last decade through 2017. However, this data has not been updated to include a contemporary cohort of patients. This study sought to determine the trends of anatomic total shoulder arthroplasty (aTSA), rTSA, and hemiarthroplasty (HA) usage based on primary diagnosis and total number of surgeons performing each procedure annually from 2016-2020.
Methods
Patients who underwent primary rTSA, aTSA, and HA from 2016-2020 were identified in the Premier Healthcare Database. Primary indication diagnoses for procedures were identified using International Classification of Diseases 10th edition codes. Temporal trends in patient and hospital demographics, primary indication, and procedure utilization were captured on an annualized basis. The number of surgeons performing each procedure annually was noted. Descriptive statistics were employed with significance set at P < .05.
Results
From 2016 to 2020, 154,499 patients undergoing primary shoulder arthroplasty were identified: 48,890 aTSA, 95,808 rTSA, and 9801 HA. In 2016, rTSA comprised a slight majority (55%) of all arthroplasty cases but increased to nearly 70% of all arthroplasty cases in 2020. The absolute numbers of aTSA and HA cases decreased over time, while rTSA volume increased from 14,781 in 2016 to a high of 23,644 cases in 2019. There was a corresponding 12% increase in the number of surgeons performing rTSA across the same time period, contrasted with a 42.1% decrease in surgeons performing HA and a 14.3% decrease for aTSA. Glenohumeral osteoarthritis remains the most common indication for rTSA and aTSA, while HA is used primarily for proximal humerus fractures or hardware complications.
Conclusion
The volume of primary rTSA in the United States has continued to increase from 2016 to 2020 with concurrent decreases in the number of primary aTSA and HA cases performed. Primary rTSA accounts for nearly 70% of all primary shoulder arthroplasty cases. The number of surgeons performing rTSA continues to increase, while there has been a decrease in the number of surgeons performing aTSA and HA.
Keywords: Total shoulder arthroplasty, Anatomic total shoulder arthroplasty, Reverse total shoulder arthroplasty, Utilization, Surgeon trends, Surgeon volume, Indications
Since its introduction in 1993,10 the reverse total shoulder arthroplasty (rTSA) has been used for rotator cuff arthropathy. rTSA has continued to gain popularity since Food and Drug Administration (FDA) approval in 2004,8 and reportedly accounts for more than 50% of all shoulder arthroplasties performed.23 The indications for rTSA have expanded to include revision, fracture, glenoid insufficiencies, and primary osteoarthritis in the elderly without rotator cuff insufficiency.1,3,4,14,16, 17, 18,21
Over the last two decades, literatures investigating the usage of both anatomic total shoulder arthroplasty (aTSA) and rTSA have noted concomitant increases in both procedures in the United States.5, 6, 7,13,20,24 Studies in the 2000s had difficulty quantifying the impact of the rTSA on that increase, as both the aTSA and rTSA were logged under the same International Classification of Diseases-9 (ICD-9) procedure code before 2010.13 The addition of a separate rTSA code in 2010 allowed subsequent studies to trend the utilization of rTSA specifically. Several of these studies demonstrated a significant increase in utilization of rTSA over a period of just a few years, with similar trends noted in several European countries.2,15,19,23 Few studies have analyzed the utilization of shoulder arthroplasty by primary diagnosis with a modern cohort through the year 2020.
The purpose of this study was to determine the trends of aTSA, rTSA, and hemiarthroplasty (HA) usage over a period of 5 years based on primary diagnosis. A secondary aim was to analyze patient comorbidities and hospital characteristics within these groups and to assess the number of surgeons performing each shoulder arthroplasty procedure. We hypothesized that both rTSA utilization would increase and indications would broaden relative to aTSA and HA over the 5-year period.
Methods
Data source and study population
The Premier Healthcare Database (PHD) was queried to identify all patients who underwent aTSA, rTSA, and HA from January 1, 2016 to December 31, 2020. The PHD is a nationally representative database of approximately 25% of the population that samples from over 1000 hospitals across the United States. It contains inpatient data regarding patient demographics, hospital factors, ICD codes, Current Procedural Terminology codes, and billing data. ICD-Tenth Revision (ICD-10) codes were utilized to identify these procedures within the PHD (rTSA: 0RRK00Z, 0RRJ00Z; aTSA: 0RRK0JZ, 0RRJ0JZ; HA: 0RRK0J6, 0RRJ0J6). All patients under 18 years old and those who underwent revision aTSA, rTSA, or HA were excluded.
This study was exempted by the institutional review board as all patient information was deidentified in accordance with the Health Insurance Portability and Accountability Act.
Patient demographics and comorbidities
Patient characteristics were extracted from PHD including age, sex, length of stay (LOS), and inpatient hospital charges. Hospital factors including geographic region, urban/rural setting, size, and teaching designation (academic/nonacademic) were collected. Rates of comorbidities were collected and compared between aTSA, rTSA, and HA patients. Lastly, the number of total operating surgeons within the PHD performing aTSA, rTSA, and HA were collected. These temporal trends were reported on an aggregated and annualized basis from 2016 to 2020.
Statistical analysis
Descriptive statistics were performed to report all patient demographics, hospital factors, and rates of comorbidities. Independent t-tests and chi-squared analyses for continuous and categorical variables, respectively, were performed to assess for differences in these factors between cohorts. Statistical significance was defined as P < .050. All statistical analyses were performed using STATA (version 16.1; StataCorp, College Station, TX, USA).
Results
Overall trends in arthroplasty and surgeon volume
In total, from 2016 to 2020, 154,499 patients were identified: 48,890 aTSA, 95,808 rTSA, and 9801 HA. The year with the highest volume of arthroplasty cases was 2019, with a 5-year high of 35,845 cases. The total number of annual arthroplasty cases has remained relatively stable since 2016, with an average of 30,860 cases per year. The number of surgeons performing shoulder arthroplasties in that timeframe has also remained steady, with an average aggregate of 2124 surgeons per year. However, within the total number of cases and surgeons performing them, there have been significant shifts in the proportion of each arthroplasty type.
In the United States, rTSA has come to account for the vast majority of shoulder arthroplasty cases. In 2016, rTSA comprised a slight majority (55%) of all arthroplasty cases. By 2020, that figure had risen to nearly 70% of all arthroplasty cases, as illustrated in Figure 1. There was a corresponding 112% increase in the number of surgeons performing rTSA across the same time period, contrasted with a 42.1% decrease in surgeons performing HA and a 14.3% decrease for aTSA. Additional data on arthroplasty and surgeon volume are given in Tables I and II.
Figure 1.
Proportion of annual shoulder arthroplasties by arthroplasty type (aTSA, rTSA, and HA). aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; HA, hemiarthroplasty.
Table I.
Annual case volume by arthroplasty type.
| 2016 |
2017 |
2018 |
2019 |
2020 |
|
|---|---|---|---|---|---|
| Annual volume (% of total annual volume) | Annual volume (% of total annual volume) | Annual volume (% of total annual volume) | Annual volume (% of total annual volume) | Annual volume (% of total annual volume) | |
| aTSA | 9810 (36.6) | 10,700 (35.1) | 10,017 (32.1) | 10,376 (29.0) | 7987 (26.7) |
| rTSA | 14,781 (55.1) | 17,435 (57.3) | 19,358 (61.9) | 23,644 (66.0) | 20,590 (68.8) |
| Hemi | 2232 (8.32) | 2312 (7.59) | 1878 (6.01) | 1825 (5.09) | 1354 (4.52) |
| Total annual volume | 26,823 | 30,447 | 31,253 | 35,845 | 29,931 |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Table II.
Annual surgeon volume by arthroplasty type.
| 2016 |
2017 |
2018 |
2019 |
2020 |
|
|---|---|---|---|---|---|
| Annual volume | Annual volume | Annual volume | Annual volume | Annual volume | |
| aTSA | 1416 | 1515 | 1387 | 1400 | 1213 |
| rTSA | 1508 | 1673 | 1661 | 1782 | 1688 |
| Hemi | 900 | 844 | 718 | 706 | 521 |
| Aggregated annual volume | 2129 | 2244 | 2124 | 2142 | 1983 |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Patient and hospital characteristics
Our three arthroplasty groups differed significantly in all demographic respects. These differences included age, sex, race, LOS, cost of care, insurance status, primary diagnoses, and comorbidities (P < .0001 in all). The rTSA group had the greatest mean age at 71.3 years. In terms of sex, there was a higher proportion of females in the rTSA group. Racial distribution was also significantly different across arthroplasty groups. Medicare users accounted for the greatest proportion of patients in all three arthroplasty groups, but this was the highest in the rTSA group at approximately 73%, relative to 56% and 50% in the aTSA and HA groups, respectively. HA patients had the longest mean hospital stay, at 2.12 days; however, costs of care were highest in the rTSA group. Each type of arthroplasty was most likely to have been performed at urban, nonacademic centers of size >500 beds. Tables III and IV provide a detailed illustration of patient and hospital characteristics. The three groups differed significantly in terms of comorbidities as well, with the three most prevalent being hypertension, obesity, and chronic pulmonary disease. Table V illustrates comorbidities in detail.
Table III.
Patient characteristics.
| aTSA (n = 48,890) | rTSA (n = 95,808) | Hemiarthroplasty (n = 9801) | P value | |
|---|---|---|---|---|
| Mean age ± SD | 66.74 ± 9.21 | 71.32 ± 8.50 | 64.12 ± 13.45 | <.0001 |
| Sex (%) | <.0001 | |||
| Female | 24,294 (49.8) | 56,273 (61.0) | 5181 (52.9) | |
| Male | 24,521 (50.2) | 37,424 (39.1) | 4618 (47.13) | |
| Race (%) | <.0001 | |||
| White | 43,317 (88.6) | 85,242 (89.0) | 8451 (87.3) | |
| Black | 2438 (5.0) | 4711 (4.9) | 557 (5.8) | |
| Asian | 2553 (5.2) | 4320 (4.5) | 534 (6.7) | |
| Other | 582 (1.2) | 1072 (1.1) | 140 (1.5) | |
| Payer category (%) | <.0001 | |||
| Private | 14,916 (36.9) | 14,955 (20.2) | 2868 (35.4) | |
| Medicare | 22,585 (55.9) | 53,886 (72.7) | 4022 (49.6) | |
| Medicaid | 1924 (4.8) | 2702 (3.6) | 774 (9.5) | |
| Workers’ | ||||
| Compensation | 868 (2.2) | 2329 (3.1) | 325 (4.0) | |
| Other | 148 (0.4) | 282 (0.4) | 123 (1.5) | |
| Mean length of stay ± SD | 1.45 ± 1.19 | 1.87 ± 2.03 | 2.14 ± 3.89 | <.0001 |
| Mean cost ± SD | 65,657 ± 35,334 | 78,622 ± 42,265 | 69,527 ± 53,941 | <.0001 |
| Primary diagnoses (%) | <.0001 | |||
| Osteoarthritis | 42,389 (86.9) | 51,150 (54.0) | 3459 (36.7) | |
| Proximal humerus hardware failure | 1812 (3.7) | 6268 (6.6) | 1857 (19.7) | |
| Inflammatory arthritis | 1545 (3.2) | 4140 (4.4) | 200 (2.1) | |
| Avascular necrosis | 1110 (2.3) | 1158 (1.2) | 757 (8.0) | |
| Posttraumatic arthritis | 633 (1.3) | 1393 (1.5) | 108 (1.2) | |
| Unspecified arthropathy | 488 (1.0) | 4760 (5.0) | 330 (3.5) | |
| Proximal humerus fracture | 383 (0.8) | 11,394 (12.0) | 2294 (24.3) | |
| Rotator cuff tear | 224 (0.5) | 12,120 (12.8) | 210 (2.2) | |
| Proximal humerus malunion / nonunion | 170 (0.4) | 2293 (2.4) | 213 (2.3) |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; SD, standard deviation.
Table IV.
Hospital characteristics.
| aTSA (n = 48,890) | rTSA (n = 93,320) | Hemiarthroplasty (n = 9855) | P value | |
|---|---|---|---|---|
| Setting (%) | <.0001 | |||
| Urban | 43,606 (89.2) | 83,889 (87.6) | 8659 (88.4) | |
| Rural | 5284 (10.8) | 11,919 (12.4) | 1142 (11.7) | |
| Teaching status (%) | <.0001 | |||
| Nonacademic | 27,771 (56.8) | 52,791 (56.1) | 5256 (53.6) | |
| Academic | 21,119 (43.2) | 42,017 (43.9) | 4545 (46.4) | |
| Size (%, in beds) | <.0001 | |||
| 0-99 | 4357 (8.9) | 8875 (9.3) | 801 (8.2) | |
| 100-199 | 9328 (19.1) | 17,257 (18.0) | 1810 (18.5) | |
| 200-299 | 8064 (16.5) | 16,699 (17.4) | 1675 (17.1) | |
| 300-399 | 7478 (15.3) | 14,710 (15.4) | 1410 (14.4) | |
| 400-499 | 6361 (13.0) | 1163 (12.3) | 1092 (11.1) | |
| >500 | 13,302 (27.2) | 26,504 (27.7) | 3013 (30.7) | |
| Region (%) | <.0001 | |||
| South | 20,843 (42.6) | 47,360 (49.4) | 4386 (44.8) | |
| Midwest | 13,083 (26.8) | 25,180 (26.3) | 2382 (24.3) | |
| West | 8014 (16.4) | 11,510 (12.0) | 1809 (18.5) | |
| Northeast | 6950 (14.2) | 11,758 (12.3) | 1224 (12.5) |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Table V.
Baseline comorbidities (%).
| aTSA (n = 48,890) | rTSA (n = 93,320) | Hemiarthroplasty (n = 9855) | P value | |
|---|---|---|---|---|
| Hypertension | 28,206 (57.7) | 57,462 (60.0) | 5169 (52.7) | <.0001 |
| Obesity | 11,381 (23.3) | 20,017 (20.9) | 1911 (19.5) | <.0001 |
| Chronic pulmonary disease | 8913 (18.2) | 20,605 (21.5) | 2007 (20.5) | <.0001 |
| Hypothyroidism | 7956 (16.3) | 18,584 (19.4) | 1524 (15.6) | <.0001 |
| Depression | 7800 (16.0) | 16,936 (17.7) | 1592 (16.2) | <.0001 |
| Uncomplicated DM | 6916 (14.2) | 15,588 (16.3) | 1405 (14.3) | <.0001 |
| Complicated hypertension | 3882 (7.9) | 12,398 (12.9) | 992 (10.1) | <.0001 |
| Renal failure | 3066 (6.3) | 9675 (10.1) | 781 (8.0) | <.0001 |
| DM with complications | 2729 (5.6) | 8396 (8.8) | 751 (7.7) | <.0001 |
| Autoimmune / collagen disorder | 2243 (4.6) | 6482 (6.8) | 571 (5.8) | <.0001 |
| Electrolyte imbalance | 2035 (4.2) | 7431 (7.8) | 837 (8.5) | <.0001 |
| Congestive heart failure | 1844 (3.8) | 6380 (6.7) | 576 (5.9) | <.0001 |
| Neurologic disorder | 1276 (2.6) | 3860 (4.0) | 643 (6.6) | <.0001 |
| Peripheral vascular disease | 1225 (2.5) | 3694 (3.9) | 308 (3.1) | <.0001 |
| Liver disease | 736 (1.5) | 1735 (1.8) | 261 (2.7) | <.0001 |
| Coagulopathy | 654 (1.3) | 1882 (2.0) | 221 (2.3) | <.0001 |
| Drug abuse | 639 (1.3) | 1158 (1.2) | 223 (2.3) | <.0001 |
| Alcohol abuse | 520 (1.1) | 1488 (1.6) | 324 (3.3) | <.0001 |
| Iron deficiency anemia | 456 (0.9) | 1867 (2.0) | 199 (2.0) | <.0001 |
| Pulmonary circulation disorder | 333 (0.7) | 1277 (1.3) | 122 (1.2) | <.0001 |
| Cardiac valve disease | 326 (0.7) | 1024 (1.1) | 78 (0.8) | <.0001 |
| Malignancy | 241 (0.5) | 645 (0.7) | 148 (1.5) | <.0001 |
| Chronic blood loss anemia | 163 (0.3) | 444 (0.5) | 58 (0.6) | <.0001 |
| Psychosis | 130 (0.3) | 333 (0.4) | 85 (0.9) | <.0001 |
| Lymphoma | 109 (0.2) | 288 (0.3) | 32 (0.3) | <.0001 |
| Peptic ulcer disease | 107 (0.2) | 326 (0.3) | 22 (0.2) | <.0001 |
| Paralysis | 73 (1.5) | 235 (0.3) | 29 (0.3) | <.0001 |
| Weight loss | 68 (0.1) | 501 (0.5) | 119 (1.2) | <.0001 |
| HIV/AIDS | 49 (0.1) | 41 (0.04) | 22 (0.2) | <.0001 |
| Metastatic cancer | 39 (0.1) | 190 (0.2) | 74 (0.8) | <.0001 |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; DM, diabetes mellitus type 2; HIV/AIDS, human immunodeficiency virus/acquired immunodeficiency syndrome.
Annualized trends in shoulder arthroplasty
In the period from 2016-2020, absolute number of aTSA cases decreased over time, from approximately 10,000 per year in 2016-2018 to 7987 in 2020. There was no significant change in the mean age or insurance status of aTSA patients over this period. However, there was a significant difference in other demographic respects, including sex (P = .005) and racial distribution (<0.0001). The mean LOS decreased steadily over time as well (P < .0001), with a 2020 mean LOS of approximately 1.3 days compared to the overall mean of 1.5 days. Mean cost increased over time as well (P < .0001), with a 2020 mean of $70,954 compared to the overall mean of $65,657. Primary diagnoses differed significantly across the five-year period (P < .0001). Osteoarthritis remained the leading diagnosis in each year, accounting for approximately 86%-87% of all aTSA cases. Table VI provides a detailed report of trends in aTSA over the 5-year period from 2016-2020.
Table VI.
Annualized trends in aTSA.
| Overall (n = 48,890) | 2016 (n = 9810) | 2017 (n = 10,700) | 2018 (n = 10,017) | 2019 (n = 10,376) | 2020 (n = 7987)) | P value | |
|---|---|---|---|---|---|---|---|
| Mean age ± SD | 66.7 ± 9.2 | 66.7 ± 9.3 | 66.9 ± 9.3 | 66.9 ± 9.3 | 66.8 ± 9.2 | 66.6 ± 9.0 | .6078 |
| Sex (%) | .005 | ||||||
| Female | 24,294 (49.8) | 4896 (49.9) | 5323 (49.8) | 5088 (50.8) | 5170 (50.0) | 3817 (48.0) | |
| Male | 24,521 (50.2) | 4914 (50.1) | 5377 (50.3) | 4929 (49.2) | 5161 (50.0) | 4140 (52.0) | |
| Race (%) | <.0001 | ||||||
| White | 43,317 (88.6) | 8847 (90.5) | 9433 (88.2) | 8824 (88.1) | 9128 (88.0) | 7085 (88.7) | |
| Black | 2438 (5.0) | 486 (5.0) | 508 (4.8) | 483 (4.8) | 547 (5.3) | 414 (5.2) | |
| Asian | 2362 (4.9) | 356 (3.6) | 651 (6.1) | 603 (6.0) | 569 (5.5) | 340 (4.3) | |
| Other | 582 (1.2) | 87 (0.9) | 108 (1.0) | 107 (1.1) | 132 (1.3) | 148 (1.9) | |
| Payer category (%) | .389 | ||||||
| Private | 14,916 (36.9) | 3125 (37.3) | 3290 (36.7) | 3115 (37.2) | 3066 (36.3) | 2320 (36.8) | |
| Medicare | 22,585 (55.9) | 4608 (55.0) | 5028 (56.1) | 4691 (56.0) | 4770 (56.5) | 3488 (55.4) | |
| Medicaid | 1924 (4.8) | 423 (5.1) | 418 (4.7) | 356 (4.3) | 387 (4.6) | 340 (5.4) | |
| Workers’ compensation | 868 (2.2) | 186 (2.2) | 194 (2.2) | 180 (2.2) | 179 (2.1) | 129 (2.1) | |
| Other | 148 (0.4) | 31 (0.4) | 30 (0.3) | 29 (0.4) | 35 (0.4) | 23 (0.4) | |
| Mean LOS ± SD | 1.45 ± 1.19 | 1.57 ± 1.12 | 1.51 ± 1.10 | 1.45 ± 1.15 | 1.39 ± 1.38 | 1.28 ± 1.09 | <.0001 |
| Mean cost ± SD (USD) | 65,657 ± 35,334 | 60,941 ± 28,351 | 63,870 ± 30,826 | 65,722 ± 35,576 | 67,818 ± 35,359 | 70,954 ± 45,884 | <.0001 |
| Primary diagnoses (%) | <.0001 | ||||||
| Osteoarthritis | 42,389 (86.9) | 8479 (86.7) | 9310 (87.3) | 8651 (86.8) | 8988 (86.8) | 6961 (87.4) | |
| Proximal humerus hardware failure | 1812 (3.7) | 318 (3.3) | 390 (3.7) | 373 (3.7) | 402 (3.9) | 329 (4.1) | |
| Inflammatory arthritis | 1545 (3.2) | 311 (3.2) | 331 (3.1) | 331 (3.4) | 329 (3.2) | 231 (2.9) | |
| Avascular necrosis | 1110 (2.3) | 223 (2.3) | 238 (2.2) | 238 (2.5) | 242 (2.3) | 161 (2.0) | |
| Posttraumatic arthritis | 633 (1.3) | 138 (1.4) | 139 (1.3) | 139 (1.2) | 140 (1.4) | 92 (1.2) | |
| Unspecified arthropathy | 488 (1.0) | 162 (1.7) | 100 (0.9) | 78 (0.8) | 81 (0.8) | 67 (0.8) | |
| Proximal humerus fracture | 383 (0.8) | 69 (0.7) | 81 (0.8) | 81 (0.7) | 85 (0.8) | 79 (1.0) | |
| Rotator cuff tear | 224 (0.5) | 44 (0.5) | 45 (0.4) | 45 (0.5) | 54 (0.5) | 34 (0.4) | |
| Proximal humerus malunion / nonunion | 170 (0.4) | 38 (0.4) | 36 (0.3) | 45 (0.5) | 37 (0.4) | 14 (0.2) |
aTSA, anatomic total shoulder arthroplasty; LOS, length of stay; SD, standard deviation; USD, United States dollar.
Absolute numbers of rTSA cases have generally increased over time. While in 2016 there were approximately 3000 more rTSA cases compared with aTSA cases, rTSAs saw a high of 23,644 cases in 2019 compared to 14,781 in 2016. There were significant differences in all demographic respects over time, including mean age (P = .0002), sex, racial distribution, and insurance status (P < .0001). Mean LOS decreased consistently over time from 2.14 in 2016 to 1.66 in 2020, as mean costs of care increased over the same timespan from $74,525 to $82,664 (P < .0001 for both). Primary diagnoses differed significantly as well (P < .0001). Though osteoarthritis was already the leading diagnosis in 2016, accounting for 48.8% of cases, the proportion of patients who underwent rTSA for osteoarthritis grew to 56.4% by 2020. Detailed trends in rTSA are reported in Table VII.
Table VII.
Annualized trends in rTSA.
| Overall (n = 95,808) | 2016 (n = 14,781) | 2017 (n = 17,435) | 2018 (n = 19,358) | 2019 (n = 23,644) | 2020 (n = 20,590) | P value | |
|---|---|---|---|---|---|---|---|
| Mean age ± SD | 71.3 ± 8.5 | 71.5 ± 8.7 | 71.5 ± 8.6 | 71.4 ± 8.5 | 71.3 ± 8.4 | 71.1 ± 8.4 | .0002 |
| Sex (%) | <.0001 | ||||||
| Female | 58,273 (60.9) | 9232 (62.5) | 10,847 (62.2) | 11,857 (61.3) | 14,120 (59.8) | 12,217 (59.6) | |
| Male | 37,424 (39.1) | 5549 (37.5) | 6588 (37.8) | 7500 (38.8) | 9487 (40.2) | 8300 (40.5) | |
| Race (%) | <.0001 | ||||||
| White | 85,242 (89.0) | 13,350 (90.8) | 15,483 (89.3) | 17,178 (89.1) | 20,879 (88.7) | 18,352 (89.6) | |
| Black | 4711 (4.9) | 725 (4.9) | 823 (4.8) | 953 (5.0) | 1167 (5.0) | 1043 (5.1) | |
| Asian | 4320 (5.5) | 484 (3.3) | 862 (5.0) | 957 (5.0) | 1206 (5.1) | 811 (4.0) | |
| Other | 1072 (1.1) | 142 (1.0) | 173 (1.0) | 185 (1.0) | 284 (1.2) | 288 (1.4) | |
| Payer category (%) | <.0001 | ||||||
| Private | 14,955 (20.2) | 2415 (19.9) | 2771 (20.0) | 2988 (19.6) | 3674 (20.6) | 3077 (20.6) | |
| Medicare | 53,886 (72.7) | 8825 (72.7) | 10,134 (73.0) | 11,164 (73.4) | 12,963 (72.5) | 10,800 (71.8) | |
| Medicaid | 2702 (3.6) | 421 (3.5) | 521 (3.8) | 521 (3.4) | 606 (3.4) | 633 (4.2) | |
| Workers’ compensation | 2329 (3.1) | 435 (3.6) | 409 (3.0) | 479 (3.2) | 566 (3.2) | 440 (2.9) | |
| Other | 282 (0.4) | 39 (0.3) | 42 (0.3) | 62 (0.4) | 66 (0.4) | 73 (0.5) | |
| Mean LOS ± SD | 1.87 ± 2.03 | 2.13 ± 2.13 | 2.02 ± 2.00 | 1.88 ± 2.04 | 1.79 ± 1.89 | 1.66 ± 2.18 | <.0001 |
| Mean cost ± SD (USD) | 78,622 ± 42,265 | 74,512 ± 38,616 | 75,897 ± 39,044 | 77,612 ± 43,186 | 80,465 ± 42,256 | 82,714 ± 45,904 | <.0001 |
| Primary diagnoses (%) | <.0001 | ||||||
| Osteoarthritis | 51,150 (54.0) | 7134 (48.8) | 9022 (52.4) | 10,463 (54.7) | 13,086 (56.0) | 11,445 (56.3) | |
| Rotator cuff tear | 12,120 (12.8) | 2161 (14.8) | 2224 (12.9) | 2273 (11.9) | 2943 (12.6) | 2519 (12.4) | |
| Proximal humerus fracture | 11,394 (12.0) | 1853 (12.7) | 2139 (12.4) | 2348 (12.3) | 2628 (11.3) | 2426 (11.9) | |
| Proximal humerus hardware failure | 6268 (6.6) | 1853 (12.7) | 2139 (12.4) | 2348 (12.3) | 2628 (11.3) | 2426 (11.9) | |
| Unspecified arthropathy | 4760 (5.0) | 971 (6.6) | 906 (5.3) | 963 (5.0) | 1040 (4.5) | 880 (4.3) | |
| Inflammatory arthritis | 4140 (4.4) | 660 (4.5) | 748 (4.3) | 840 (4.4) | 1027 (4.4) | 865 (4.3) | |
| Proximal humerus malunion / nonunion | 2293 (2.4) | 381 (2.6) | 438 (2.5) | 462 (2.4) | 552 (2.4) | 460 (2.3) | |
| Posttraumatic arthritis | 1393 (1.5) | 241 (1.7) | 286 (1.7) | 334 (1.8) | 287 (1.2) | 245 (1.2) | |
| Avascular necrosis | 1158 (1.2) | 182 (1.2) | 223 (1.3) | 241 (1.3) | 286 (1.2) | 226 (1.1) | |
LOS, length of stay; SD, standard deviation; rTSA, reverse total shoulder arthroplasty; USD, United States dollar.
Shoulder HA cases have seen a decrease in numbers from 2232 in 2016 to 1354 in 2020. Demographic characteristics, including age, sex, racial distribution, and insurance status, demonstrate no significant differences over this 5-year period. However, there was a significant increase in costs of care, with a mean of $69,527 in 2016 vs. $80,720 in 2020 (P < .0001). Primary diagnoses differ significantly across the years as well (P < .0001). Osteoarthritis, the leading diagnosis in 2016, accounting for 39.4% of HAs, was superseded by proximal humerus fracture by 2020, which accounted for the greatest proportion of cases at 23.1%. There is no linear directionality observed in comorbid conditions for the HA cohort, but significant differences exist nonetheless for obesity, complicated hypertension, renal failure, complicated DM2, and congestive heart failure. Table VIII depicts these in detail along with other characteristics of this cohort.
Table VIII.
Annualized trends in hemiarthroplasty.
| Overall (n = 9801) | 2016 (n = 2232) | 2017 (n = 2312) | 2018 (n = 1878) | 2019 (n = 1825) | 2020 (n = 1354) | P value | |
|---|---|---|---|---|---|---|---|
| Mean age ± SD | 64.1 ± 13.5 | 64.4 ± 13.0 | 64.2 ± 13.3 | 64.3 ± 13.5 | 64.2 ± 13.8 | 63.5 ± 13.8 | .421 |
| Sex (%) | .130 | ||||||
| Female | 5181 (52.9) | 1227 (55.0) | 1213 (52.5) | 1005 (53.5) | 935 (51.3) | 699 (51.7) | |
| Male | 4618 (47.1) | 1005 (45.0) | 1099 (47.5) | 873 (46.5) | 889 (48.7) | 654 (48.3) | |
| Race (%) | .061 | ||||||
| White | 8451 (87.3) | 1974 (88.4) | 1980 (85.6) | 1614 (85.9) | 1548 (84.8) | 1161 (85.8) | |
| Black | 557 (5.8) | 117 (5.2) | 141 (6.1) | 111 (5.9) | 102 (5.6) | 76 (5.6) | |
| Asian | 534 (5.52) | 112 (5.0) | 156 (6.8) | 130 (6.9) | 147 (8.1) | 93 (6.9) | |
| Other | 140 (1.5) | 29 (1.3) | 35 (1.5) | 23 (1.2) | 28 (1.5) | 24 (1.8) | |
| Payer category (%) | .178 | ||||||
| Private | 2868 (35.4) | 689 (36.0) | 693 (35.9) | 524 (33.7) | 524 (35.8) | 381 (35.6) | |
| Medicare | 4022 (49.6) | 966 (50.5) | 959 (49.6) | 796 (51.1) | 706 (48.3) | 506 (47.3) | |
| Medicaid | 774 (9.5) | 150 (7.8) | 178 (9.2) | 155 (10.0) | 154 (10.5) | 120 (11.2) | |
| Workers’ compensation | 325 (4.0) | 79 (4.1) | 79 (4.1) | 63 (4.1) | 50 (3.4) | 42 (3.9) | |
| Other | 123 (1.5) | 29 (1.5) | 24 (1.2) | 19 (1.2) | 28 (1.9) | 20 (1.9) | |
| Mean LOS ± SD | 2.14 ± 3.89 | 2.07 ± 2.68 | 2.20 ± 3.72 | 2.00 ± 3.28 | 2.12 ± 4.73 | 2.32 ± 5.26 | .181 |
| Mean cost ± SD (USD) | 69,527 ± 53,941 | 63,867 ± 43,959 | 68,158 ± 52,965 | 68,159 ± 50,081 | 71,854 ± 55,795 | 80,720 ± 70,809 | <.0001 |
| Primary diagnoses (%) | <.0001 | ||||||
| Osteoarthritis | 3459 (36.7) | 853 (39.4) | 858 (38.4) | 674 (37.6) | 604 (34.4) | 402 (21.2) | |
| Proximal humerus fracture | 2294 (24.3) | 586 (27.1) | 557 (23.4) | 420 (23.4) | 383 (21.8) | 297 (23.1) | |
| Proximal humerus hardware failure | 1857 (19.7) | 298 (13.8) | 364 (16.3) | 368 (20.5) | 436 (24.8) | 357 (2.7) | |
| Avascular necrosis | 757 (8.0) | 161 (7.4) | 192 (8.1) | 146 (8.1) | 133 (7.6) | 107 (8.3) | |
| Unspecified arthropathy | 330 (3.5) | 76 (3.5) | 86 (3.9) | 59 (3.3) | 56 (3.2) | 42 (3.3) | |
| Proximal humerus malunion / nonunion | 213 (2.3) | 49 (2.3) | 54 (2.4) | 42 (2.3) | 39 (2.2) | 27 (2.1) | |
| Rotator cuff tear | 210 (2.2) | 64 (3.0) | 43 (1.8) | 33 (1.8) | 40 (2.3) | 26 (2.0) | |
| Inflammatory arthritis | 200 (2.1) | 52 (2.4) | 47 (1.6) | 28 (1.6) | 48 (2.7) | 20 (1.6) | |
| Posttraumatic arthritis | 108 (1.2) | 26 (1.2) | 31 (1.4) | 25 (1.4) | 16 (0.9) | 9 (0.7) |
LOS, length of stay; SD, standard deviation.
Discussion
This study demonstrated in a contemporary cohort that the volume of primary rTSA in the United States has continued to increase from 2016 to 2020. Concurrently, the number of primary aTSA and HA cases has decreased over the same time period. By 2020, primary rTSA accounted for nearly 70% of all primary shoulder arthroplasty cases in this nationally sampled cohort, an increase from 55% of all cases in 2016. Glenohumeral osteoarthritis has remained the most common indication for aTSA and has increasingly become an indication for rTSA, accounting for 56% of all rTSA cases in 2020. Additionally, the number of surgeons performing rTSA has increased over the study period, while the number of surgeons performing aTSA and HA has simultaneously decreased.
The results of this study represent an updated assessment of the trends in shoulder arthroplasty in the United States. Recent studies investigating trends in shoulder arthroplasty have been limited to include data through the calendar year 2017.2,19,23 Using the National Inpatient Sample, Rabinowitz et al noted a 22% annualized increase in rTSA from 2011 to 2016, twice that of aTSA, with rTSA volume surpassing aTSA volume in 2014.19 Similarly, Wagner et al demonstrated an annual volume increase for rTSA and aTSA of 191.3% and 38.5%, respectively, from 2011 to 2017 with a similar 60% decrease in HA volume.23 Furthermore, Best et al reported a total of 64,215 shoulder arthroplasty cases in 2012 with 33.56% rTSA, 46.23% aTSA, and 18.21% HA, which increased to 108,300 total cases in 2017 with 57.90% rTSA, 37.55% aTSA and 4.55% HA.2 This data is consistent with Modor Intelligence’s Shoulder Replacement Market Analysis, which projects a continued annual growth rate of 8.16% from 2022-2027.26 These previous estimates through 2017 are similar to those of the current study, and our data provide continued expansion of these trends, demonstrating that rTSA accounts for nearly 70% of all shoulder arthroplasty in 2020.
The popularity of rTSA has grown since its approval by the FDA in 2004 and has expanded in indication greatly beyond massive irreparable rotator cuff tears and rotator cuff arthropathy.3,5,8,13 Our data demonstrate that osteoarthritis has become the most common indication for rTSA, now accounting for 56% of all cases, followed by rotator cuff arthropathy and proximal humerus fracture. Similarly, osteoarthritis remains the primary indication for approximately 87% of aTSA cases, followed by proximal humerus hardware failure and inflammatory arthropathy. Several studies using data prior to 2017 have corroborated these findings of shoulder arthroplasty indications.2,11,15 In a study spanning nine different countries, Lübbeke et al demonstrated that that the most common indication for shoulder arthroplasty was osteoarthritis, followed by rotator cuff arthropathy.15 Similarly, within the Finnish Registry, Harjula et al noted that osteoarthritis was the most common indication for both aTSA and rTSA.11 However, Best et al cited rotator cuff arthropathy as the most common indication for rTSA, while osteoarthritis remained the most common indication for aTSA.2 A possible explanation for the differences in primary diagnosis coding may be due to specificity in primary diagnosis coding, which significantly improved following the introduction of ICD-10 diagnosis and procedure codes in the third quarter of 2015. Thus, our study captures the first five years of ICD-10 coding with regards to primary indications for shoulder arthroplasty; however, continued assessment of indications and outcomes by indication for total shoulder arthroplasty is warranted.
The number of orthopedic surgeons performing rTSA increased over the study period (+11.9%), while the number of surgeons performing aTSA and HA decreased (−14.3% and −41.1%, respectively). One potential explanation for this expansion may be an increasing number of surgeons undergoing fellowship training in shoulder and elbow surgery, which has been growing rapidly and doubled from 2003 to 2013.12 However, Zmistowski et al noted that only one-third of total shoulder arthroplasty cases are performed by fellowship-trained shoulder and elbow surgeons.25 Similarly, Somerson et al noted that while the highest volume shoulder arthroplasty surgeons had shoulder and elbow fellowship training, these surgeons accounted for only 28% of providers, whereas 40% of providers had sports medicine fellowship training.22 Considering this increased prevalence of rTSA noted by Farley et al there remains concern regarding the projected revision burden in the future.9 Taken together, these findings indicate that the increase in subspecialty fellowship training following orthopedic surgery residency supports the data found within this study on the expanding number of surgeons performing rTSA.
This study is not without limitations. As with all administrative databases, the findings of this study are reliant on accurate coding entered within the database and thus are subject to potential coding errors. The primary indication for surgery is determined by ICD-10-Clinical Modification (CM) diagnosis codes; while these codes are far more specific than ICD-9-CM codes utilized prior to 2016, they do not provide specific details for the indications of each surgery. However, given the number of patients included within the study, we do not expect these to affect the results in favor of one procedure and skew the proportions of procedures performed. Furthermore, our results are in concordance with previous estimates of primary indication for shoulder arthroplasty.11,15
The strengths of this study warrant mention. This study included over 154,000 individual patient files from a nationally representative database using data up to the 2020 calendar year, thus expanding upon previous estimates available through 2017. The use of specific ICD-10 procedural and diagnosis codes allowed for more precise inclusion criteria compared to previously published literature. Additionally, we were able to assess the number of individual surgeons performing rTSA, aTSA, and HA across all study years to assess overall practice patterns within a cohort of more than 2000 surgeons. To our knowledge, this represents one of the largest contemporary cohorts investigating the changes in practice patterns for shoulder arthroplasty. The findings of this investigation provide robust evidence to support previous claims that rTSA has continued to gain popularity in its use and indications. Future studies focused on the effects of these expanding indications and utilization on long-term patient outcomes are warranted.
Conclusion
The volume of primary rTSA in the United States has continued to increase from 2016 to 2020 with concurrent decreases in the number of primary aTSA and HA cases performed. Primary rTSA accounts for nearly 70% of all primary shoulder arthroplasty cases. The number of surgeons performing rTSA continues to increase, while there has been a decrease in the number of surgeons performing aTSA and HA. These findings should provide an updated assessment of shoulder arthroplasty utilization and indication in a nationwide, representative cohort and should prompt future study on the ramifications of increased rTSA utilization.
Disclaimers
Funding: No funding was disclosed by the authors.
Conflicts of interest: Seth C. Gamradt has received royalties or licenses from Zimmer Biomet Holdings, consulting fees from Exatech Inc., support for attending meetings and/or travel from Micromed Inc., and other financial or nonfinancial interests from Arthrex Inc. Alexander E. Weber has received consulting fees from Stryker Corp. and other financial or nonfinancial interests from Arthrex Inc. Joseph N. Liu has received consulting fees and/or hospitality payments from Stryker Corp., Smith & Nephew, Merz North America Inc., and Neuracrine Biosciences and reports American Shoulder and Elbow Surgeons and Arthroscopy Association of North America (board or committee member). Frank A. Petrigliano has received consulting fees and/or hospitality payments from Stryker, Exactech, Micromed, Smith & Nephew, Zimmer Biomet, Flexion Therapeutics, and DePuy Synthes. The other authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
Footnotes
This study was exempted by the institutional review board as all patient information was deidentified in accordance with the Health Insurance Portability and Accountability Act.
References
- 1.Austin L., Zmistowski B., Chang E.S., Williams G.R., Jr. Is reverse shoulder arthroplasty a reasonable alternative for revision arthroplasty? Clin Orthop Relat Res. 2011;469:2531–2537. doi: 10.1007/s11999-010-1685-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Best M.J., Aziz K.T., Wilckens J.H., McFarland E.G., Srikumaran U. Increasing incidence of primary reverse and anatomic total shoulder arthroplasty in the United States. J Shoulder Elbow Surg. 2021;30:1159–1166. doi: 10.1016/j.jse.2020.08.010. [DOI] [PubMed] [Google Scholar]
- 3.Boileau P., Watkinson D., Hatzidakis A.M., Hovorka I. Neer Award 2005: the Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J Shoulder Elbow Surg. 2006;15:527–540. doi: 10.1016/j.jse.2006.01.003. [DOI] [PubMed] [Google Scholar]
- 4.Bufquin T., Hersan A., Hubert L., Massin P. Reverse shoulder arthroplasty for the treatment of three- and four-part fractures of the proximal humerus in the elderly. J Bone Joint Surg Br. 2007;89-B:516–520. doi: 10.1302/0301-620x.89b4.18435. [DOI] [PubMed] [Google Scholar]
- 5.Day J.S., Lau E., Ong K.L., Williams G.R., Ramsey M.L., Kurtz S.M. Prevalence and projections of total shoulder and elbow arthroplasty in the United States to 2015. J Shoulder Elbow Surg. 2010;19:1115–1120. doi: 10.1016/j.jse.2010.02.009. [DOI] [PubMed] [Google Scholar]
- 6.Day J.S., Paxton E.S., Lau E., Gordon V.A., Abboud J.A., Williams G.R. Use of reverse total shoulder arthroplasty in the Medicare population. J Shoulder Elbow Surg. 2015;24:766–772. doi: 10.1016/j.jse.2014.12.023. [DOI] [PubMed] [Google Scholar]
- 7.Dillon M.T., Chan P.H., Inacio M.C.S., Singh A., Yian E.H., Navarro R.A. Yearly trends in elective shoulder arthroplasty, 2005-2013. Arthritis Care Res. 2017;69:1574–1581. doi: 10.1002/acr.23167. [DOI] [PubMed] [Google Scholar]
- 8.Food and Drug Administration. 510 (k) summary: Delta shoulder. Rockville, MD. https://accessdata.fda.gov/cdrh_docs/pdf2/K021478.pdf.
- 9.Farley K.X., Wilson J.M., Kumar A., Gottschalk M.B., Daly C., Sanchez-Sotelo J., et al. Prevalence of shoulder arthroplasty in the United States and the increasing burden of revision shoulder arthroplasty. JB JS Open Access. 2021;6 doi: 10.2106/JBJS.OA.20.00156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Grammont P.M., Baulot E. Delta shoulder prosthesis for rotator cuff rupture. Orthopedics. 1993;16:65–68. doi: 10.3928/0147-7447-19930101-11. [DOI] [PubMed] [Google Scholar]
- 11.Harjula J.N.E., the Finnish Shoulder Arthroplasty Registry Group. Paloneva J., Haapakoski J., Kukkonen J., Äärimaa V. Increasing incidence of primary shoulder arthroplasty in Finland – a nationwide registry study. BMC Musculoskelet Disord. 2018;19:245. doi: 10.1186/s12891-018-2150-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Horst P.K., Choo K., Bharucha N., Vail T.P. Graduates of Orthopaedic residency training are increasingly Subspecialized: a review of the American board of Orthopaedic surgery Part II database. J Bone Joint Surg Am. 2015;97:869–875. doi: 10.2106/JBJS.N.00995. [DOI] [PubMed] [Google Scholar]
- 13.Kim S.H., Wise B.L., Zhang Y., Szabo R.M. Increasing incidence of shoulder arthroplasty in the United States. J Bone Joint Surg Am. 2011;93:2249–2254. doi: 10.2106/JBJS.J.01994. [DOI] [PubMed] [Google Scholar]
- 14.Levy J., Frankle M., Mighell M., Pupello D. The use of the reverse shoulder prosthesis for the treatment of failed hemiarthroplasty for proximal humeral fracture. J Bone Joint Surg Am. 2007;89:292–300. doi: 10.2106/JBJS.E.01310. [DOI] [PubMed] [Google Scholar]
- 15.Lübbeke A., Rees J.L., Barea C., Combescure C., Carr A.J., Silman A.J. International variation in shoulder arthroplasty. Acta Orthop. 2017;88:592–599. doi: 10.1080/17453674.2017.1368884. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.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]
- 17.Muh S.J., Streit J.J., Wanner J.P., Lenarz C.J., Shishani Y., Rowland D.Y., et al. Early follow-up of reverse total shoulder arthroplasty in patients sixty years of age or younger. J Bone Joint Surg Am. 2013;95:1877–1883. doi: 10.2106/JBJS.L.10005. [DOI] [PubMed] [Google Scholar]
- 18.Mulieri P., Dunning P., Klein S., Pupello D., Frankle M. Reverse shoulder arthroplasty for the treatment of irreparable rotator cuff tear without glenohumeral arthritis. J Bone Joint Surg Am. 2010;92:2544–2556. doi: 10.2106/JBJS.I.00912. [DOI] [PubMed] [Google Scholar]
- 19.Rabinowitz J., Kothandaraman V., Lin J., Li X., Friedman R.J., Eichinger J.K. Utilization of shoulder arthroplasty in the United States – an analysis of current trends and future predictions Seminars in Arthroplasty. JSES. 2020;30:200–209. doi: 10.1053/j.sart.2020.08.004. [DOI] [Google Scholar]
- 20.Schairer W.W., Nwachukwu B.U., Lyman S., Craig E.V., Gulotta L.V. National utilization of reverse total shoulder arthroplasty in the United States. J Shoulder Elbow Surg. 2015;24:91–97. doi: 10.1016/j.jse.2014.08.026. [DOI] [PubMed] [Google Scholar]
- 21.Smith C.D., Guyver P., Bunker T.D. Indications for reverse shoulder replacement: a systematic review. J Bone Joint Surg Br. 2012;94:577–583. doi: 10.1302/0301-620X.94B5.27596. [DOI] [PubMed] [Google Scholar]
- 22.Somerson J.S., Stein B.A., Wirth M.A. Distribution of high-volume shoulder arthroplasty surgeons in the United States: data from the 2014 Medicare provider utilization and payment data Release. J Bone Joint Surg Am. 2016;98:e77. doi: 10.2106/JBJS.15.00776. [DOI] [PubMed] [Google Scholar]
- 23.Wagner E.R., Farley K.X., Higgins I., Wilson J.M., Daly C.A., Gottschalk M.B. The incidence of shoulder arthroplasty: rise and future projections compared with hip and knee arthroplasty. J Shoulder Elbow Surg. 2020;29:2601–2609. doi: 10.1016/j.jse.2020.03.049. [DOI] [PubMed] [Google Scholar]
- 24.Westermann R.W., Pugely A.J., Martin C.T., Gao Y., Wolf B.R., Hettrich C.M. Reverse shoulder arthroplasty in the United States: a comparison of national volume, patient demographics, complications, and surgical indications. Iowa Orthop J. 2015;35:1–7. [PMC free article] [PubMed] [Google Scholar]
- 25.Zmistowski B., Warrender W., Livesey M., Girden A., Williams G.R., Jr., Namdari S. The characteristics of surgeons performing total shoulder arthroplasty: volume consistency, training, and specialization. Am J Orthop (Belle Mead NJ) 2018;47 doi: 10.12788/ajo.2018.0111. [DOI] [PubMed] [Google Scholar]
- 26.Shoulder replacement Market size & Share analysis - Industry research report - growth trends. https://www.mordorintelligence.com/industry-reports/shoulder-replacement-market Available from: Accessed April 18, 2023.