S101 Comparison of Outcomes Using Anatomic and Reverse Total Shoulder Arthroplasty Pierre-Henri Flurin, M.D., Yann Marczuk, M.D., Martin Janout, M.D., Thomas W. Wright, M.D., Joseph Zuckerman, M.D., and Christopher P. Roche, M.S., M.B.A. Pierre-Henri Flurin, M.D., is at the Bordeaux-Merignac Clinique du Sport, Merignac, France. Yann Marczuk, M.D., is at the Chenieux Clinic, Limoges, France. Martin Janout, M.D., is in private practice at Northwest Orthopaedic Specialists, Spokane, Washington. Thomas W. Wright, M.D., is in the Department of Orthopaedics and Rehabilitation, University of Florida, Gainesville, Florida. Joseph Zuckerman, M.D., is in the Department of Orthopaedic Surgery, Hospital for Joint Diseases, NYU Langone Medical Center, New York, New York. Christopher P. Roche, M.S., M.B.A., is employed by Exactech, Inc., Gainesville, Florida. Correspondence: Pierre-Henri Flurin, M.D., 2 rue Negrevergne, Bordeaux-Merignac 33700, France; phflurin@gmail.com. Abstract Worldwide, the usage of both anatomic total shoulder arthroplasty (atsa) and reverse total shoulder arthroplasty (rtsa) has increased significantly due, in part, to the predictability of acceptable outcomes achieved with each prosthesis type. This study quantifies outcomes using five different metrics and compares results using one platform total shoulder arthroplasty system that utilizes the same humeral component and instrumentation to perform both atsa or rtsa. Methods: 200 patients were treated by two orthopaedic surgeons using either atsa or rtsa. 73 patients received atsa for treatment of osteoarthritis (OA), and 127 patients received rtsa for treatment of rotator cuff tear arthroplasty (CTA). Each was scored preoperatively, and at latest follow-up using the SST, UCLA, ASES, Constant, and SPADI metrics, motion was also quantified. The average follow-up for all patients was 31.4 ± 9.7 months. Results: All patients demonstrated significant improvements in pain and function following treatment of OA with atsa and treatment of CTA with rtsa. No instances of instability or glenoid loosening were reported in either cohort; one instance of infection occurred in the rtsa cohort. atsa was associated with significantly higher preand postoperative outcome scores and significantly larger pre- and postoperative range of motion than rtsa. However, rtsa was demonstrated to be significantly more effective at improving outcome scores, active forward flexion, and strength than was atsa. Discussion and Conclusions: Significant improvements in outcome scores were observed for both atsa and rtsa using one platform shoulder system at a mean follow-up of 31.4 months. Significant differences were observed between prosthesis type and between scoring metrics, particularly between the Constant and ASES scoring metrics. Additional and longer term follow-up is required to confirm these observed differences. Both anatomic (atsa) and reverse (rtsa) total shoulder arthroplasty are the standard of care for various end-stage degenerative conditions of the glenohumeral joint. Osteoarthritis (OA) is the most common indication for atsa, while rotator cuff tear arthropathy (CTA) is the most common indication for rtsa. Worldwide, the usage of both atsa and rtsa has increased significantly due, in part, to the predictability of acceptable outcomes achieved with each prosthesis type for its respective indications. However, the perception of predictability of outcomes with each prosthesis type may be influenced by the outcome measure used to score each result. Standardizing the tools and methods by which healthcare professionals collect clinical outcomes is a critical component of evidence-based medicine. Numerous outcome measurement scores are available to evaluate the success of treatment of patients with debilitating conditions in the shoulder. The most commonly used scoring systems are the Simple Shoulder Test (SST), the UCLA Score, the American Shoulder and Elbow Surgery Score (ASES), the Constant-Murley (Constant), the Shoulder Pain and Disability Index (SPADI), the Disabilities of the Flurin PH, Marczuk Y, Janout M, Wright TW, Zuckerman J, Roche CP. Comparison of outcomes using anatomic and reverse total shoulder arthroplasty. Bull Hosp Jt Dis. 2013;71(Suppl 2):S101-7.
S102 Bulletin of the Hospital for Joint Diseases 2013;71(Suppl 2):S101-7 Arm, Shoulder, and Hand (DASH), the Rowe Score, and the Oxford Shoulder Score. While each metric attempts to rate the quality of care, each varies by the method in which it gauges the success of treatment (based upon the restoration of function, motion, and strength and by the reduction of pain). Additionally, each metric varies in how it distributes and weighs subjective patient responses and clinical observations from objective independent examiner assessment measurements. No standard scoring metric exists for shoulder arthroplasty. In the USA, the five most commonly used scoring systems are the SST, UCLA, ASES, Constant, and SPADI clinical metrics. The SST score is derived from a series of 12 Yes and No questions that measure the patient s ability to carry out activities of daily living; 12 is the highest or best score. The UCLA score is derived from a series of five questions that evaluates pain, satisfaction of treatment, and restoration of function, strength, and motion; 35 is the highest or best score. The ASES score is derived from a series of 11 questions that evaluates pain (50%) and restoration of function (50%); 100 is the highest or best score. The Constant score is derived from a series of 23 questions that evaluates pain (15%), restoration of function (20%), range of motion (40%), and strength and power (25%); 100 is the highest or best score. The SPADI score is derived from a series of 13 questions that evaluates pain and restoration of function; 130 is the highest score and 0 is the best score. The aim of this study is to quantify outcomes of atsa and rtsa using five different scoring metrics and compare results achieved for each indication using a single platform total shoulder arthroplasty system that utilizes the same humeral component and instrumentation to perform both atsa or rtsa. To this end, preoperative, postoperative, and pre- to postoperative improvements were collected and compared using the five aforementioned metrics for 200 patients who received either a primary atsa or rtsa for the treatment of OA or CTA, respectively. Methodology Two hundred patients (70.9 ± 7.3 years) were treated by two orthopaedic surgeons using either atsa (February 2005 to January 2011) or rtsa (May 2007 to May 2011) with the Equinoxe platform shoulder system (Exactech, Inc.; Gainesville, FL). Seventy-three patients received atsa (67.4 ± 8.0 years; 32 left and 41 rights; 54 keeled and 19 pegged glenoid implants) for treatment of OA (PHF: 64 patients; YM: 9 patients), consisting of 50 females (70.0 ± 7.6 years) and 23 males (66.6 ± 10.4 years). One hundred twenty-seven patients received rtsa (72.9 ± 6.1 years; 30 left and 97 right; 61 received 38 mm, 65 received 42 mm, and 1 received 46 mm glenosphere implants) for treatment of CTA (PHF: 53 patients; YM: 74 patients), consisting of 93 females (73.3 ± 6.0 years) and 34 males (71.8 ± 6.3 years). These patients were scored preoperatively and at latest follow-up using the SST, UCLA, ASES, Constant, and SPADI metrics; active forward flexion and internal/external rotation with the arm at the side were also measured. Internal rotation was measured by vertebral segments and was scored by the following discrete assignment: 0 degrees = 0, hip = 1, buttocks = 2, sacrum = 3, L5-L4 = 4, L3-L1 = 5, T12-T8 = 6, and T7 or higher = 7. The average follow-up for all patients was 31.4 ± 9.7 months (atsa: 32.5 ± 12.1 months; rtsa: 30.8 ± 8.0 months). A Student s two-tailed, unpaired t-test was used to identify differences in preoperative, postoperative, and pre- to postoperative improvements, where p < 0.05 denoted a significant difference. Results All patients demonstrated significant improvements in strength and function and significant decreases in pain following treatment of OA with atsa and of CTA with rtsa, respectively. Additionally, each outcome scoring measurement demonstrated significant improvements for both atsa and rtsa cohorts. No instances of instability or glenoid loosening were reported in either cohort; one instance of infection occurred in the rtsa cohort. The average pre- and postoperative outcome scores, range of motion, strength, and pain measurements associated with each treatment type are compared in Tables 1 and 2. As described in Table 1, rtsa patients were observed to have significantly lower preoperative scores as measured by three of the five metrics (SST, UCLA, and Constant), significantly less active forward flexion and strength (as measured by max weight), a significantly higher internal rotation score, and significantly less pain than the preoperative measurements associated with atsa patients. As described in Table 2, rtsa patients were observed to have significantly lower postoperative scores as measured by two of the five metrics (SST and UCLA), significantly less active forward flexion and active external rotation, a significantly lower active internal rotation score, and significantly less pain than the postoperative measurements associated with atsa patients. The average improvement in outcome scores and range of motion measurements for each treatment type is presented in Table 3. As described in Table 3, rtsa patients were observed to have significantly larger improvements in outcome as measured by two of the five metrics (UCLA and Constant), significantly larger improvements in active forward flexion and strength, and significantly less improvements in active external rotation and in the active internal rotation score than the improvements associated with atsa patients. The following figures describe the improvement associated with the ASES and Constant metrics for atsa and rtsa patients; it should be noted that each of these metrics have 100 point scales and therefore can be directly compared. Figure 1 compares improvements in atsa and rtsa patients as measured by the ASES score and demonstrates that the improvements are identical
S103 Table 1 Comparison of Average Preoperative Measurements, atsa vs. rtsa Patients Forward External Internal Rotation Max Discrete Weight Pain at Worst (10 SST UCLA ASES Constant SPADI Flexion Rotation Score (lbs) = worst) atsa 3.6 ± 2.7 14.7 ± 3.3 38.6 ± 14.9 39.6 ± 12.2 79.2 ± 27.9 115.1 ± 31.7 1.2 ± 17.3 2.5 ± 1.5 5.0 ± 4.7 8.1 ± 1.6 rtsa 2.4 ± 1.7 12.6 ± 3.6 35.2 ± 10.8 30.0 ± 11.5 82.3 ± 15.3 85.0 ± 44.7 4.5 ± 18.9 3.0 ± 1.5 0.8 ± 2.0 7.7 ± 1.5 P value 0.0001 < 0.0001 0.0674 < 0.0001 0.2358 < 0.0001 0.2125 0.0500 < 0.0001 0.0434 Table 2 Comparison of Average Postoperative Measurements, atsa vs. rtsa Patients Forward External Internal Rotation Max Discrete Weight Pain at Worst (10 SST UCLA ASES Constant SPADI Flexion Rotation Score (lbs) = worst) atsa 10.8 ± 2.0 31.8 ± 4.0 90.3 ± 14.6 76.5 ± 11.6 11.4 ± 19.4 157.3 ± 18.2 37.3 ± 16.7 5.7 ± 1.0 9.2 ± 4.2 2.3 ± 2.7 rtsa 10.3 ± 1.7 31.4 ± 2.4 86.7 ± 10.1 74.2 ± 8.6 12.3 ± 11.6 149.3 ± 16.9 32.6 ± 13.6 5.3 ± 0.9 8.5 ± 3.6 1.6 ± 1.8 P value 0.0379 0.3649 0.0451 0.1237 0.6998 0.0019 0.0319 0.0013 0.1938 0.0228 Table 3 Comparison of Average Improvement, atsa vs. rtsa Patients Forward External Internal Rotation Max Discrete Weight Pain at Worst (10 SST UCLA ASES Constant SPADI Flexion Rotation Score (lbs) = worst) atsa 7.2 ± 2.8 17.0 ± 5.0 51.7 ± 18.7 36.7 ± 15.6 67.6 ± 26.6 42.3 ± 32.7 36.1 ± 23.5 3.2 ± 1.7 4.5 ± 5.8-5.8 ± 3.1 rtsa 7.8 ± 2.3 18.7 ± 4.4 51.5 ± 15.4 44.2 ± 13.8 70.2 ± 20.1 64.3 ± 46.9 28.0 ± 24.8 2.4 ± 1.7 7.7 ± 4.0-6.1 ± 2.5 P value 0.0796 0.0128 0.9413 0.0005 0.4823 0.0005 0.0251 0.0005 < 0.0001 0.4800 Figure 1 Histogram describing the improvement in the ASES score for atsa and rtsa patients. for atsa and rtsa with this metric. Figure 2 compares improvements in atsa and rtsa patients as measured by the Constant score and demonstrates that improvements are larger for rtsa compared to atsa. Figures 3 and 4 compare improvements in atsa and rtsa patients using the ASES and Constant scores. Figure 3 demonstrates a significantly larger improvement in ASES score for atsa compared to Constant score. Figure 4 also demonstrates
S104 Bulletin of the Hospital for Joint Diseases 2013;71(Suppl 2):S101-7 Figure 2 Histogram describing the improvement in the Constant score for atsa and rtsa patients. Figure 3 Histogram comparing the improvement of atsa patients as measured by the ASES and Constant scores. Figure 4 Histogram comparing the improvement of rtsa patients as measured by the ASES and Constant scores.
S105 Table 4 Comparison of atsa Outcome Scores Reported in Literature Study Sample Size Avg Follow-up (months) Pre-op Avg Pre-op Avg ASES Score ASES Score Edwards 2003 1 601 44.0 31.1 70.3 * * Orfaly 2003 2 37 51.6 * * 37 (SSI) 91 (SSI) Godeneche 2002 3 251 30 28 71 * * Gartsman 2000 4 27 35 * * 22.7 ± 14.4 (SSI) 77.3 ±18.2 (SSI) Raiss 2012 5 39 132 27 61 * * Walch 2011 6 311 89.5 31.4 ±13.3 67.6 ±17.6 * * Present Study 73 32.5 ± 12.1 39.6 ± 12.2 75.1 ± 11.5 38.6 ± 14.9 90.3 ± 14.6 *denotes measurement not reported. Table 5 Comparison of atsa Shoulder Motion Data Reported in Literature Pre-op Avg Study (degrees) Pre-op Avg Edwards 2003 1 91.2 144.5 7.2 41.5 Orfaly 2003 2 100 147 7 39 Godeneche 2002 3 94 145 6 40 Gartsman 2000 4 86 128 36 (arm in abduction) 61 (arm in abduction) Raiss 2012 5 84 133 11 35 Walch 2011 6 94.9 ±28.2 146.6 ±27.1 9 ±16.4 35.3 ±19.5 Present Study 115.1 ± 31.7 157.3 ± 18.2 1.2 ± 17.3 37.3 ± 16.7 a greater improvement in ASES score for rtsa compared to Constant score. Discussion The results of this study demonstrate significant improvements in outcomes following treatment with both atsa and rtsa using a single platform shoulder system at a mean follow-up of 31.4 ± 9.7 months. While atsa and rtsa were used to treat different indications, each treatment method provided significant improvements in all five outcome score measurements, all motion measurements, strength, and pain for its respective indication at a similar mean follow-up of 30 or more months. These outcome scores and motion results are favorable relative to other published motion and outcome scores previously reported for both atsa 1-6 and rtsa 7-13 : Tables 4 to7 compare the pre- and postoperative outcome scores and motion results in this study to that previously reported for atsa 1-6 (Tables 4 and 5) and rtsa 7-13 (Tables 6 and 7), respectively. Several comparative differences were observed between atsa and rtsa outcome measures. atsa was associated with significantly higher preoperative outcome scores (according to three of five metrics) and significantly higher postoperative scores (according to two of the five metrics), significantly larger preoperative range of motion (according to two of three active measurements) and significantly larger postoperative range of motion (according to three of three active measurements), and lastly significantly greater preoperative strength and pain and significantly greater postoperative pain than that associated with rtsa patients. Additionally, rtsa was demonstrated to be significantly more effective at improving outcomes (as measured by two of the five metrics), active forward flexion, and strength than was atsa. However, atsa was demonstrated to be significantly more effective at improving active internal and external rotation than was rtsa. Additional and longer term follow-up is required to confirm these findings. Comparing the differences in outcome metric scores for atsa and rtsa yields some interesting observations (particularly related to the ASES and Constant metrics). The preoperative ASES score was not significantly different between atsa and rtsa patients. The preoperative Constant score was significantly higher for atsa patients compared to rtsa patients. Conversely, the postoperative ASES score was significantly higher for atsa compared to rtsa, whereas the postoperative Constant score was not significantly different for atsa and rtsa patients. That is why no statistical difference was observed in the mean ASES score improvement between atsa and rtsa patients, whereas the mean Constant score improvement was significantly higher in rtsa patients. The observed differences in improvement within scoring metrics reflect the different scoring weights for pain, function, and strength; and may also imply that each scoring metric has different sensitivities for what defines a worse patient and what defines a good outcome. Specifically, the
S106 Bulletin of the Hospital for Joint Diseases 2013;71(Suppl 2):S101-7 Table 6 Comparison of rtsa Shoulder Outcome Scores Reported in Literature Study Sample Size Avg Follow-up (months) Pre-op Avg Pre-op Avg ASES Score ASES Score Sirveaux, 2004 7 77 44 22.6 (4 to 50) 65.5 (34 to 85) * * Werner, 2005 8 44 38 29 (3 to 53); age 64 (10 to 100); * * adjusted age adjusted Frankle, 2005 9 60 33 * * 34.3 (0 to 65) 68.2 (15 to 100) Boileau, 2006 10 42 40 17 (95% CI: 14 to 19) 58 (95% CI: 51 to 64) * * Levigne, 2008 11 337 47 23 58 * * Stechel, 2010 12 59 48 15 (2 to 55) 55 (17 to 96) * * Nolan, 2011 13 71 24 27.5 (5 to 58) 61.8 (30 to 87) 26 (0 to 63) 76.1 (21 to 100) Present Study 127 30.8 ± 8.0 30.0 ± 11.5 74.2 ± 8.6 35.2 ± 10.8 86.7 ± 10.1 *denotes measurement not reported. Table 7 Comparison of rtsa Shoulder Motion Data Reported in Literature Pre-op Avg Study (degrees) Pre-op Avg Sirveaux, 2004 7 73 138 3.5 11.2 Werner, 2005 8 42 (0 to 90) 100 (0 to 145) 17 (-20 to 70) 12 (-50 to 60) Frankle, 2005 9 55.0 (0 to 120) 105.1 (30 to 180) 12.0 (-15 to 45) 35.9 (5 to 60) Boileau, 2006 10 55 (95% CI: 47 to 63) 121 (95% CI: 111 to 131) 7 (95% CI: 1 to 13) 11 (95% CI: 5 to 16) Levigne, 2008 11 70 125 7 9 Stechel, 2010 12 47 105-9 19 Nolan, 2011 13 61.2 (0 to 137) 121.3 (52 to 170) 13.8 (-35 to 60) 14.6 (-44 to 60) Present Study 85.0 ± 44.7 149.3 ± 16.9 4.5 ± 18.9 32.6 ± 13.6 most significant difference between the ASES and Constant metrics is the 25% scoring weight given to strength by the Constant score. As described in Table 1, the preoperative strength of atsa patients was significantly greater than the preoperative strength of rtsa patients. However, as described in Table 2, the postoperative measurement of strength was not statistically different between atsa and rtsa patients; therefore, rtsa patients were associated with a significantly larger improvement in strength than atsa patients. Because 25% of the Constant metric consist of strength measurement and since the ASES metric doesn t explicitly measure strength, the preoperative Constant score is lower than the ASES score in rtsa. As rtsa was demonstrated to be more effective at restoring strength than atsa, the observed improvements in strength associated with rtsa is emphasized more in the Constant metric and not reflected as much in the ASES metric. Differences in scoring weights within each metric may bias results for one type of treatment over another and as a result further indicate a need for a unified scoring system. Perhaps these observations relate to the Constant score preferentially scoring improvements with rtsa (relative to the ASES metric) may provide an explanation for the higher utilization of reverse shoulders in Europe than in the USA (as the Constant score is predominately used in Europe and the ASES score is predominately used in the USA) and may be the explanation for why indications for rtsa have expanded more in Europe than in the USA. Conclusion This comparative clinical study of anatomic and reverse total shoulder arthroplasty using a single platform shoulder system demonstrated very favorable outcomes with five different scoring metrics at a mean follow-up of 31.4 ± 9.7 months. While postoperative scores were significantly greater than preoperative scores in each of the 5 scoring metrics for both atsa and rtsa, significant differences in outcome scores between atsa and rtsa were observed; additional and longer term follow-up is required to confirm these observed differences. Differences in scoring weights/ methods within each scoring metric were also observed, particularly between the Constant and ASES scoring metrics. Future work should attempt to better understand the underlying differences between each scoring metric, identify any differences in sensitivities, and attempt to optimize scoring weights for different treatment types when attempting to
S107 create a standardized scoring metric to quantify outcomes with total shoulder arthroplasty. Disclosure Statement Pierre-Henri Flurin, M.D., Thomas W. Wright, M.D., and Joseph Zuckerman, M.D., are consultants for Exactech, Inc., and receive royalties on products related to this article. Yann Marczuk, M.D., receives consultant fees related to the topic of this research from Exactech, Inc. Christopher P. Roche, M.S., M.B.A., is employed by Exactech, Inc. Martin Janout, M.D., has no financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony. References 1. Edwards TB, Kadakia NR, Boulahia A, et al. A comparison of hemiarthroplasty and total shoulder arthroplasty in the treatment of primary glenohumeral osteoarthritis, Results of a multicenter study. J Shoulder Elbow Surg. 2003 May- Jun;12(3):207-13. 2. Orfaly RM, Rockwood CA Jr, Wirth MA. A prospective functional outcome study of shoulder arthroplasty for osteoarthritis with an intact rotator cuff. J Shoulder Elbow Surg. 2003 May-Jun;12(3):214-21. 3. Godeneche A, Boileau P, Favard L, et al. Prosthetic replacement in the treatment of osteoarthritis of the shoulder: Early results of 268 cases. J Shoulder Elbow Surg. 2002 Jan-Feb;11(1):11-8. 4. Gartsman GM, Roddey TS, Hammerman SM. Shoulder arthroplasty with or without resurfacing of the glenoid in patients who have osteoarthritis. J Bone Joint Surg Am. 2000 Jan;82(1):26-34. 5. Raiss P, Schmitt M, Bruckner T, et al. Results of cemented total shoulder replacement with a minimum follow-up of ten years. J Bone Joint Surg Am. 2012 Dec 5;94(23):e1711-10. 6. Walch G, Young AA, Melis B, et al. Results of a convex-back cemented keeled glenoid component in primary osteoarthritis: multicenter study with a follow-up greater than 5 years. J Shoulder Elbow Surg. 2011 Apr;20(3):385-94. 7. Sirveaux F, Favard L, Oudet D, et al. Grammont inverted total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff. J Bone Joint Surg Br. 2004;86-B:388-95. 8. Werner C, Steinmann PA, Gilbart M, Gerber C. Treatment of painful pseudoparaesis due to irreparable rotator cuff dysfunction with the Delta III reverse ball and socket total shoulder prosthesis. J Bone Joint Surg Am. 2005 Jul;87(7): 1476-86. 9. Frankle M, Siegal S, Pupello D, et al. The reverse shoulder prosthesis for glenohumeral arthritis associated with severe rotator cuff deficiency. A minimum two-year follow-up study of sixty patients. J Bone Joint Surg Am. 2005 Aug;87(8):1697-705. 10. Boileau P, Watkinson D, Hatzidakis AM, Hovorka I. The Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J Shoulder Elbow Surg. 2006 Sep-Oct;15(5):527-40. 11. Levigne C, Boileau P, Favard L, et al. Scapular notching in reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2008;17(6):925-35. 12. Stechel A, Fuhrmann U, Irlenbusch L, et al. Reversed shoulder arthroplasty in cuff tear arthritis, fracture sequelae, and revision arthroplasty. Acta Orthop. 2010 Jun;81(3)367-72. 13. Nolan BM, Ankerson E, Wiater JM. Reverse Total Shoulder Arthroplasty Improves Function in Cuff Tear Arthropathy. Clin Orthop Relat Res. 2011 Sep;464(9):2476-82.