Reverse Shoulder Arthroplasty for the Treatment of Rotator Cuff Deficiency

Transcription

1 1895 COPYRIGHT Ó 2017 BY THE JOURAL OF BOE AD JOIT SURGERY, ICORPORATED Reverse Shouler Arthroplasty for the Treatment of Rotator Cuff Deficiency A Concise Follow-up, at a Minimum of 10 Years, of Previous Reports* Derek J. Cuff, MD, Derek R. Pupello, MBA, Branon G. Santoni, PhD, Rachel E. Clark, BS, CCRP, an Mark A. Frankle, MD Investigation performe at the Floria Orthopaeic Institute an the Founation for Orthopaeic Research an Eucation, Tampa, Floria Abstract: We previously evaluate 94 patients (96 shoulers) who unerwent reverse shouler arthroplasty using a central compressive screw with 5.0-mm peripheral locking screws for baseplate fixation an a center of rotation lateral to the glenoi as treatment for en-stage rotator cuff eficiency. The purpose of this stuy was to report upate results at a minimum follow-up of 10 years. Forty patients (42 shoulers) were available for clinical follow-up. In the patients available for stuy, implant survivorship, with the en point being revision for any reason, was 90.7%. Since our 5-year report, 2 patients unerwent revision surgery; 1 patient sustaine a periprosthetic fracture 7 years postoperatively an 1 patient ha a islocation because of chronic shouler instability at 8 years postoperatively. At a minimum follow-up of 10 years, the patients continue to maintain their improve outcome scores an range of motion, which were comparable with earlier follow-up evaluations. Level of Evience: Therapeutic Level IV. See Instructions for Authors for a complete escription of levels of evience. Backgroun We previously reporte on a cohort of patients who were treate between February 2004 an March 2005 with a reverse shouler arthroplasty (RSA) as a part of a U.S. Foo an Drug Aministration (FDA) Investigational Device Exemption (IDE) stuy 1,2. The implant that was use ha a central 6.5-mmiameter compressive screw with 5.0-mm peripheral locking screws for baseplate fixation an a glenosphere with a center of rotation lateral to the glenoi (Reverse Shouler Prosthesis; DJO Surgical). The surgical technique positione the glenosphere with inferior tilt in relation to the face of the glenoi by preferentially reaming the inferior aspect of the glenoi. The humeral component that was use in this stuy ha a 135 neckshaft angle an was place in 30 of retroversion. The patients inclue in the stuy ha en-stage rotator cuff eficiency that was ue to a wie variety of pathological conitions, incluing primary rotator cuff eficiency, faile previous rotator cuff operations, faile previous arthroplasty, an proximal humeral nonunion. At the 5-year follow-up point, these patients ha preoperative to postoperative improvements in their clinical outcome scores an range of motion an a 94% implant survival rate 1. Raiographic analysis of this cohort at that time reveale a 9% rate of scapular notching an no glenoi baseplate loosening or baseplate failures. The purpose of this stuy was to examine the miterm results at a minimum of 10 years of follow-up focusing on survivorship of the implant, clinical outcomes, an raiographic finings. Methos Subsequent to our 5-year report, 24 patients haie an 10 patients were lost to follow-up, leaving 40 patients (42 shoulers) for evaluation (Fig. 1). With respect to inications for surgery, 19 shoulers (45.2%) ha primary rotator cuff *Original Publications Cuff D, Pupello D, Virani, Levy J, Frankle M. Reverse shouler arthroplasty for the treatment of rotator cuff eficiency. J Bone Joint Surg Am Jun;90 (6): Cuff D, Clark R, Pupello D, Frankle M. Reverse shouler arthroplasty for the treatment of rotator cuff eficiency: a concise follow-up, at a minimum of 5 years, of a previous report. J Bone Joint Surg Am ov;94(21): Disclosure: This stuy was supporte in part by a research grant for the initial U.S. Foo an Drug Aministration stuy from DJO Global, which is the manufacturer of the evice use in this stuy. On the Disclosure of Potential Conflicts of Interest forms, which are provie with the online version of the article, one or more of the authors checke yes to inicate that the author ha a relevant financial relationship in the biomeical arena outsie the submitte work an yes to inicate that the author ha a patent an/or copyright, planne, pening, or issue, for the evice that is the subject of this article ( J Bone Joint Surg Am. 2017;99:

2 1896 ROTATOR C UFF D EFICIECY Fig. 1 Flowchart illustrating patients stuie clinically an those lost to follow-up (l/t/f/u) over the course of the 10-year stuy. The istributions of operations performe accoring to the preoperative iagnosis for the 2, 5, an 10-year stuies are also provie. Pts = patients, an RC = rotator cuff. eficiency with accompanying arthritis, 13 shoulers (31.0%) ha previous faile rotator cuff operations, an 10 shoulers (23.8%) ha a faile arthroplasty an cuff eficiency. The average age of the available patients was 78 years (range, 62 to 99 years); 22 were female an 18 were male. The average time to follow-up was 132 months (range, 120 to 147 months). The surgical technique an postoperative protocol for this proceure have previously been escribe in etail 3-5. The patients were followe yearly an the same clinical outcome scores (American Shouler an Elbow Surgeons [ASES] score an Simple Shouler Test [SST]) that were utilize in our previous stuy were upate at the last follow-up evaluation. Patient questionnaires inclue a self-assessment of range of motion as inicate by a mark place at the highest attainable motion on a picture. This varie somewhat from our initial 2-year report as some patients i not have a vieo mae at the 10-year follow-up point, but all were able to fill out the assessment form. The same patient-reporte methoology for range-of-motion assessment was use in our 5-year report 1. Implant survivorship analysis, with revision surgery for any reason as the en point, was performe. Survival was efine as the percentage of shoulers in the available patients who i not require a revision surgery over the 10-year stuy perio. One of the patients who ha been lost to follow-up prior to the 10-year mark ha ha a revision 8 years after the inex proceure because of recurrent instability an was, therefore, inclue in the survivorship analysis. Raiographic analysis was performe by an inepenent observer in the same stanarize fashion as we reporte previously on our 2-year an 5-year follow-up stuies. Baseplate fixation was grae as stable (no evience of raiolucency at the baseplate-bone interface or aroun any screw), at risk (>1 mm of circumferential raiolucency at the baseplate-bone interface or aroun any 1 screw), or loose (>1 mm of raiolucency aroun the baseplate-bone interface an aroun all screws, or the existence of a shift in the position of the baseplate). Humeral loosening was measure using the graing system escribe by Sperling et al. 6. Raiographs were also evaluate for evience of islocation, scapular notching on the basis of the criteria escribe by Sirveaux et al., an screw breakage 7. Statistical Analysis Preoperative an postoperative clinical outcome ata were compare using a paire t test an commercially available statistical software (SPSS; IBM). Results At a minimum follow-up of 10 years, the patients continue to maintain their improve outcome scores an range of motion (Table I). Forty patients ha complete selfreporte functional ata (ASES score, visual analog scale [VAS] pain score, SST, an range of motion) for analysis, the majority of which was collecte either by telephone or through mail corresponence. Eight patients (20%) in the current stuy visite our facility at a minimum of 10 years postoperatively an ha raiographic evaluation. Their raiographs were evaluate for evience of harware failure an/or scapular notching. Implant survivorship at 120 months was 90.7% (39 of 43 shoulers) (Table II). A total of 4 patients in the 10-year cohort ha unergone revision surgery: 1 was inclue in the original 2-year stuy, an aitional revision was capture in the 2 to 5-year follow-up interval, an 2 aitional patients unerwent revision between 5 an 10 years after the inex proceure. The 2 more recent revisions were for recurrent instability (1 shouler)

3 1897 ROTATOR C UFF D EFICIECY TABLE I Preoperative an Postoperative Outcomes Accoring to Preoperative Diagnosis for All 42 Shoulers an 3 Subgroups* Preop. 5-Year Evaluation Latest Follow-up Evaluation P Value Total ASES score All patients 35 (0-65) 77 (7-100) 74 (45-90) Primary cuff eficiency 36 (0-63) 79 (17-100) 76 (58-90) Previous rotator cuff surgery 41 (18-65) 75 (7-100) 77 (65-85) Faile arthroplasty 27 (0-63) 72 (37-97) 68 (45-83) ASES pain score All patients 19 (0-43) 42 (0-50) 44 (35-50) Primary cuff eficiency 17 (0-30) 44 (5-50) 44 (35-50) Previous rotator cuff surgery 23 (10-43) 38 (0-50; p = 0.029) 44 (35-50) Faile arthroplasty 16 (3-43) 45 (30-50) 45 (35-50) 1.00 SST score All patients 2 (0-7) 8 (0-12) 7 (0-12) Primary cuff eficiency 2 (0-7) 8 (2-12) 7 (1-12) Previous rotator cuff surgery 2 (0-6) 7 (1-12) 7 (3-12) Faile arthroplasty 2 (0-7) 7 (0-12) 5 (0-12; p = 0.013) Forwar flexion (eg) All patients 70 (10-152) 144 (11-180) 126 (0-180) Primary cuff eficiency 85 (34-152) 157 ( ) 132 (10-180; p = 0.002) Previous rotator cuff surgery 49 (10-100) 148 (50-180) 140 (50-180) Faile arthroplasty 68 (10-126) 117 (11-180; p = 0.057) 98 (0-180; p = 0.294) Abuction (eg) All patients 65 (10-159) 129 (50-180) 117 (0-180) Primary cuff eficiency 77 (35-159) 153 (90-180) 137 (45-180) Previous rotator cuff surgery 52 (22-82) 123 (50-180) 109 (0-180) Faile arthroplasty 60 (10-158) 94 (50-150; p = 0.121) 89 (0-150; p = 0.214) External rotation (eg) All patients 18 (240-78) 53 ( ) 40 (230-90) Primary cuff eficiency 21 (240-78) 56 (260-90; p = 0.009) 53 (0-90; p = 0.002) Previous rotator cuff surgery 22 (217-50) 44 ( ; p = 0.168) 43 (230-90; p = 0.112) Faile arthroplasty 8 (221-44) 56 (10-90) 14 (230-90; p = 0.603) *Data are expresse as the mean with the range in parentheses. All comparisons of preoperative an postoperative values have a p value of <0.001 unless otherwise inicate. Comparison of 5-year follow-up with latest follow-up. an periprosthetic mishaft humeral fracture (1 shouler). There were no mechanical baseplate failures. Table I summarizes the comparison of preoperative an postoperative outcome scores at the time of final follow-up. The patients ha maintaine their gains in total average ASES scores, with an improvement from a preoperative score of 35 to a postoperative score of 74 at the last follow-up (p < 0.001). Similar gains ha been maintaine with respect to average SST scores, with an improvement from a preoperative score of 2 to a postoperative scoreof7atthelastfollow-up(p<0.001).theoutcomescores were also stratifieby the preoperative iagnosis an these ata are presente in Table I. With respect to range of motion, the average forwar flexion increase from 70 preoperatively to 126 at the last follow-up (p < 0.001); average abuction, from 65 to 117 (p < 0.001); an average external rotation, from 18 to 40 (p < 0.001). Raiographic analysis reveale no evience of raiolucency aroun the baseplate or baseplate screws in the 8 patients with 10 years of raiographic follow-up. At 10 years, 1 patient emonstrate asymptomatic humeral raiolucencies, which were initially ientifie raiographically at 6 an 7 years postoperatively. One patient with complete raiographic follow-up ha grae-1 notching at the 10-year mark. This patient ha not ha notching at 2 or 5 years as etermine from raiographic evaluation. All 40 patients ha a minimum raiographic follow-up of 2 years (average, 68 months) an, of the entire cohort available for stuy, 5 patients (5 shoulers) isplaye grae-1 notching uring the stuy perio. A single patient (1 shouler) isplaye grae-2 notching, which was note raiographically at 32 months; however, this visit to our clinic at approximately 3 years was the patient s last an all subsequent information from this

4 1898 ROTATOR C UFF D EFICIECY TABLE II Complications Requiring Revision Surgery Accoring to Preoperative Diagnosis 2-Year Follow-up (94 Patients, 96 Shoulers) 2 to 5-Year Follow-up (74 Patients, 76 Shoulers)* Since 5-Year Follow-up (41 Patients, 43 Shoulers ) Base plate failure Recurrent instability Primary cuff eficiency Previous rotator cuff surgery Faile arthroplasty Humeral loosening Primary cuff eficiency Faile arthroplasty Resorption of proximal humeral allograft Primary cuff eficiency Faile arthroplasty Periprosthetic fracture Primary cuff eficiency Faile arthroplasty Survivorship 97% 94% 91% *Three patients requiring revision at 2 years were inclue in the 5-year stuy. Two aitional revisions occurre between 2 an 5 years. One patient was inclue in the survivorship analysis because of a revision at 8 years, resulting in 43 shoulers available for survivorship analysis. Two patients requiring revision at 5 years were inclue in the 10-year stuy. Two aitional revisions in 1 patient occurre between 5 an 10 years. This patient was lost to follow-up at 10 years, an the patient s outcomes an range of motion ata were not available for analysis. patient was collecte via telephone an/or mail corresponence. Thus, the progression of the notching, if any, is unknown. The average onset to raiographic evience of scapular notching was 49.1 months (range, 25.7 to months) in the 6 of the 40 patients in whom raiographic evience was ientifie at any time uring the postoperative course. There were 2 aitional patients in the cohort who require revision surgery (Table II) since our previous 5-year report. One patient sustaine a periprosthetic mishaft humeral fracture 7 years after the inex proceure, which was performe after a faile hemiarthroplasty. The secon revision, which was for recurrent instability, was revise with another reverse prosthesis at 8 years. This patient ha a selfreporte history of chronic falling, beginning in 2011, an sustaine (1) a scapular spine fracture in 2012 that was treate nonoperatively an (2) a prosthetic islocation 4 months later that coul not be manage via close reuction. The patient ha a revision with a reverse prosthesis in July 2012 an ha a islocation again in October 2012 when rising from a chair. This patient was unreachable for follow-up at the 10-year time point. Conclusions The majority of patients who unerwent RSA in our stuy maintaine their improve function with urable clinical results at a minimum follow-up of 10 years. In our previous reports, we note that the aition of 5.0-mm peripheral locking screws for baseplate fixation ha eliminate early baseplate failures that ha been seen in an earlier series of patients in whom a glenosphere with a center of rotation lateral to the glenoi ha been utilize 1,2,4.Wenotethat the locking screws provie improve early fixation an allowe for osseous ingrowth into the baseplate in orer to achieve goo long-term fixation 2,8. The results at a minimum follow-up of 10 years in the current stuy support this fining as we observe goourability of this implant, with a survivorship of 91% using revision as an en point. Lastly, we note an increase in shouler motion between the 2 an 5-year follow-up stuies, which we attribute to the ifferent metho of assessment. Sirveaux et al. 7 an Guery et al. 9, in longer-term stuies in which the Grammontstyle RSA was use, emonstrate a functional eterioration an increase in pain levels in some patients after the 6-year follow-up point. At the 10-year mark, we i not observe any consistent tren of an increase in shouler pain, an almost 70% of the patient cohort reporte either a ecrease in pain or no change in shouler pain between their 5 an 10-year follow-up visits. However, between the 5 an 10-year stuies, we i notice a ecrease in shouler motion in all planes,

5 1899 ROTATOR C UFF D EFICIECY albeit small, for all preoperative iagnoses. In the context of ecrease motion without a concomitant increase in pain at 10 years, we attribute this fining to the avance age of our patient cohort. n Rachel E. Clark, BS, CCRP 2 Mark A. Frankle, MD 3 1 Suncoast Orthopaeic Surgery & Sports Meicine, Venice, Floria 2 Founation for Orthopaeic Research an Eucation, Tampa, Floria 3 Floria Orthopaeic Institute, Tampa, Floria Derek J. Cuff, MD 1 Derek R. Pupello, MBA 2 Branon G. Santoni, PhD 2 aress for M.A. Frankle: mfrankle@floriaortho.com ORCID id for M.A. Frankle: References 1. Cuff D, Clark R, Pupello D, Frankle M. Reverse shouler arthroplasty for the treatment of rotator cuff eficiency: a concise follow-up, at a minimum of five years, of a previous report. J Bone Joint Surg Am ov 7;94 (21): Cuff D, Pupello D, Virani, Levy J, Frankle M. Reverse shouler arthroplasty for the treatment of rotator cuff eficiency. J Bone Joint Surg Am Jun;90 (6): Frankle M, Levy JC, Pupello D, Siegal S, Saleem A, Mighell M, Vasey M. The Reverse Shouler Prosthesis for glenohumeral arthritis associate with severe rotator cuff eficiency. A minimum two-year follow-up stuy of sixty patients. Surgical technique. J Bone Joint Surg Am Sep;88(Suppl 1 Pt 2): Frankle M, Siegal S, Pupello D, Saleem A, Mighell M, Vasey M. The Reverse Shouler Prosthesis for glenohumeral arthritis associate with severe rotator cuff eficiency. A minimum two-year follow-up stuy of sixty patients. J Bone Joint Surg Am Aug;87(8): Levy J, Frankle M, Mighell M, Pupello D. The use of the Reverse Shouler Prosthesis for the treatment of faile hemiarthroplasty for proximal humeral fracture. J Bone Joint Surg Am Feb;89(2): Sperling JW, Cofiel RH, O Driscoll SW, Torchia ME, Rowlan CM. Raiographic assessment of ingrowth total shouler arthroplasty. J Shouler Elbow Surg ov-dec;9(6): Sirveaux F, Favar L, Ouet D, Huquet D, Walch G, Molé D. Grammont inverte total shouler arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff. Results of a multicentre stuy of 80 shoulers. J Bone Joint Surg Br Apr;86(3): Harman M, Frankle M, Vasey M, Banks S. Initial glenoi component fixation in reverse total shouler arthroplasty: a biomechanical evaluation. J Shouler Elbow Surg Jan-Feb;14(1)(Suppl S):162S-7S. 9. Guery J, Favar L, Sirveaux F, Ouet D, Mole D, Walch G. Reverse total shouler arthroplasty. Survivorship analysis of eighty replacements followe for five to ten years. J Bone Joint Surg Am Aug;88(8):