- Email: [email protected]
Outcomes of arthroscopic revision rotator cuff repair with acellular human dermal matrix allograft augmentation
ARTICLE IN PRESS J Shoulder Elbow Surg (2017) ■■, ■■–■■
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
Outcomes of arthroscopic revision rotator cuff repair with acellular human dermal matrix allograft augmentation Eric A. Hohn, MD*, Blake P. Gillette, MD, Joseph P. Burns, MD Southern California Orthopedic Institute, Van Nuys, CA, USA Background: The purpose was to assess the minimum 2-year patient-reported outcomes and failure rate of patients who underwent revision arthroscopic rotator cuff repair augmented with acellular human dermal matrix (AHDM) allograft for repairable retears. Methods: From 2008-2014, patients who underwent revision rotator cuff repair augmented with AHDM with greater than 2 years’ follow-up by a single surgeon were retrospectively reviewed. Data regarding surgical history, demographic characteristics, and medical comorbidities were collected. Outcome data included American Shoulder and Elbow Surgeons (ASES) and Single Assessment Numeric Evaluation (SANE) scores, as well as rotator cuff healing on magnetic resonance imaging or ultrasound. Retears and subsequent surgical procedures were characterized. Results: A total of 28 patients met our inclusion criteria, and 23 (82%) were available for follow-up at 2 years. The mean age was 60.1 ± 9.3 years (range, 43-79 years), with a mean follow-up period of 48 ± 23 months. All patients had at least 1 prior rotator cuff repair. Of the 23 patients, 13 (56%) underwent postoperative imaging, and 4 of these 13 (31%) had a retear. A reoperation was performed in 3 of 23 patients (13%). Among the 6 patients with both preoperative and postoperative outcome scores, we saw improvement in the ASES score from 56 to 85 (P = .03) and in the SANE score from 42 to 76 (P = .03). The full cohort’s mean postoperative ASES and SANE scores were 77 and 69, respectively. Conclusion: AHDM allograft augmentation is a safe and effective treatment method for patients with fullthickness rotator cuff retears. Further research is needed with larger studies to confirm these findings from our small cohort of patients. Level of evidence: Level IV; Case Series; Treatment Study © 2017 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. Keywords: Rotator cuff; revision; shoulder; arthroscopy; allograft; augmentation; acellular human dermal matrix
Institutional review board approval was obtained before the initiation of this study (Western IRB, study No. 1155091). *Reprint requests: Eric A. Hohn, MD, Southern California Orthopedic Institute, 6815 Noble Ave, Van Nuys, CA 91405, USA. E-mail address: [email protected] (E.A. Hohn).
The incidence of rotator cuff surgery continues to increase, with a recent study showing a 6-fold increase in arthroscopic repair from 1996 to 2006 in the United States.7 Moreover, a single-state surgical database showed that even with a decline in open rotator cuff surgery, the total number of repairs doubled from 2000 to 2007.14 Primary arthroscopic rotator cuff repair of small to medium tears is generally
1058-2746/$ - see front matter © 2017 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. https://doi.org/10.1016/j.jse.2017.09.026
ARTICLE IN PRESS 2 successful; however, several risk factors for failure have been identified, including large tear size, poor tendon quality, increased degree of retraction, and the presence of fatty degeneration, among others.11,18 Repair of larger tears has an increased risk of non-healing, approaching 94% in some studies.5,10 Although structural failure is not always correlated with clinical failure,20 several studies have demonstrated a clinical benefit in patients with a healed rotator cuff.13,15,17,21,28 A study by Shamsudin et al25 showed that patients undergoing revision rotator cuff repair are twice as likely to have retears and have worse postoperative pain and function compared with those undergoing primary repair. Rotator cuff revision surgery is often complicated by tendon loss or retraction, bone defects, and retained hardware. Poor tissue quality also negatively affects both the quality of surgical repair and subsequent healing.8,16,19 As these patients are less likely to achieve optimal outcomes with standard repair techniques, alternative surgical and biological approaches should be investigated. Several new techniques have been developed in an effort to improve outcomes in patients who are at high risk of failure. Biological tissue scaffolds have been introduced as a way to reduce repair tension and enhance healing of the rotator cuff. A variety of biological scaffold materials have been used with favorable results, including human acellular dermis,6 fascia lata,1 and porcine dermal collagen.12 Preliminary biomechanical and clinical evidence suggests that rotator cuff augmentation may be a safe and effective method for the treatment of massive, retracted rotator cuff tears.1-3,6,22,26 Barber et al2 showed that primary repair of large rotator cuff tears had a significantly higher healing rate when augmented with acellular human dermal matrix (AHDM) allograft (85% compared with 40%). Data regarding the use of biological scaffold augmentation in revision rotator cuff repair are lacking and are limited to open repair only.22,24 The purpose of the study was to (1) evaluate the clinical outcomes with American Shoulder and Elbow Surgeons (ASES) and Single Assessment Numeric Evaluation (SANE) scores at a minimum of 2 years’ follow-up and (2) evaluate the retear rate and need for subsequent surgery in patients with a previous failed rotator cuff repair who underwent revision arthroscopic surgery with AHDM allograft augmentation. Our hypothesis was that augmentation with AHDM allograft would be a safe and reliable treatment option for patients with failed rotator cuff repairs.
Methods Study design Between 2008 and 2014, all patients who underwent arthroscopic revision repair of full-thickness (>2 cm) rotator cuff tears augmented with AHDM allograft by a single surgeon (J.P.B.) were identified for this retrospective case-series study. The minimum subjective follow-up period was set at 2 years. The inclusion criteria included full-thickness retears of the supraspinatus and/or infraspinatus
E.A. Hohn et al. that had occurred after prior open or arthroscopic repair, measuring 2 cm or greater in size in the anterior-to-posterior plane. Any degree of tissue atrophy, degeneration, or fatty infiltration was included. Patients with concomitant pathologies such as superior labral anterior-posterior tears, chondromalacia, biceps pathology, or subscapularis tears were included. The exclusion criteria were advanced osteoarthritis according to the Samilson-Prieto classification of grade 2 or higher,23 active infection, or a rotator cuff tendon that could not be mobilized with a residual gap of less than 1 cm. Minimum 2-year follow-up data were obtained using validated shoulder questionnaires.
Surgical treatment The decision to perform rotator cuff repair with patch augmentation was based on preoperative evaluations and magnetic resonance imaging (MRI) appearance, as well as intraoperative tendon quality and mobility. Rotator cuff repair with AHDM augmentation was indicated for patients with full-thickness recurrent tearing of the supraspinatus and/or infraspinatus tendons after failed prior surgery, measuring 2 cm in size from anterior to posterior, with poor tendon quality. Any degree of muscle atrophy and fatty infiltration was acceptable, and all Goutallier grades were included in this study. If the cuff tendon could not be sufficiently mobilized with a residual gap of less than 1 cm, other options such as débridement, partial repair, tendon transfer, reverse total shoulder replacement, or a bridging allograft repair were considered, and those patients were excluded from the study. For all cases included in this series, the AHDM allograft was used for augmentation (not bridging) where the native rotator cuff tendon was secured to the medial footprint on the tuberosity. The graft was placed over the top of the rotator cuff tendon and secured circumferentially: anteriorly, posteriorly, and medially to the remaining rotator cuff tissue and laterally to the greater tuberosity, as previously described.2,27,30
Surgical technique After the administration of a regional interscalene block and induction of general anesthesia, the patient is placed in the lateral decubitus position. All bony prominences are carefully padded, and an axillary roll is placed. Ten pounds of balanced suspension in 70° of abduction is used for glenohumeral access, 15 lb in 15° of abduction is used for bursal access, and 10 lb in approximately 45° of abduction is used as a middle position for accessing the lateral aspect of the greater tuberosity in the bursal space. The operative field is prepared and draped using sterile techniques. Standard posterior and anterosuperior portals are established, a diagnostic arthroscopy is performed, and all intra-articular pathologies are addressed as necessary. An accessory lateral portal is established, and a diagnostic subacromial bursoscopy is performed. The rotator cuff tendon quality and mobility are assessed, and intra- and extra-articular releases are performed using an elevator and radiofrequency probe when necessary. The decision of whether to perform rotator cuff repair with or without augmentation is then made. Tears with normal to excellent tendon quality are repaired primarily, whereas tendons with atrophy, poor-quality suture holding, and abnormal fatty infiltration are selected for augmentation. Retained anchors and sutures from prior surgical procedures are then removed as indicated. If the
ARTICLE IN PRESS Revision rotator cuff repair with augmentation prior fixation is unlikely to interfere with the new repair or if removal will result in substantial bone loss, then the hardware is retained. If subacromial impingement is identified, an acromioplasty is performed. Rotator cuff repair with allograft augmentation is then performed. Four portals are used: 7-mm cannulas are placed anteriorly and posteriorly, an 8-mm cannula is placed anterolaterally to allow future passage of the graft, and a posterolateral portal is established for viewing with the arthroscope. The rotator cuff tear size is measured with a knotted suture-measuring device that is made with a No. 1 braided suture with knots tied 1 cm apart. The measuring suture is held with a grasper on one end while the other end is passed loaded in a standard knot pusher such that it can easily slide back and forth through the eyelet of the knot pusher. The size of the tear is measured by counting the number of knots between the instruments and is measured in both the anterior-to-posterior and medial-to-lateral directions. This measurement is used to help determine the graft size needed for augmentation. The greater tuberosity footprint is débrided to a bleeding surface using a motorized shaver. For U-shaped tears, margin convergence techniques are used to close the defects in the cuff tendon in a side-to-side fashion. L-shaped and reverse L–shaped tears are repaired using a combination of margin convergence and direct repair to the tuberosity. The native cuff is secured using a single-row repair with triple-loaded suture anchors placed on the medial footprint adjacent to the articular cartilage. Bone marrow vents are created using a 1-mm bone punch in the greater tuberosity, prior to graft placement. The graft is then prepared. The size of the allograft is determined based on the size of the tear prior to repair. The graft is intentionally oversized 3 mm on each side to allow placement of sutures; for instance, a pre-repair tear size of 3 × 3 cm will require a 3.6 × 3.6–cm allograft. The AHDM allograft is then cut to size. No. 2 simple braided permanent sutures are used for fixation of the graft. Two simple sutures are placed laterally in the graft for eventual lateral footprint fixation, and short-tail interference knots (STIKs) are placed around the remainder of the graft periphery, spaced approximately 1 cm apart. These sutures are placed through the graft while it is on the back table, and the graft is then placed adjacent to the lateral 8-mm cannula. Most grafts require 6-7 sutures. A retrograde suturing device (Spectrum; ConMed, Largo, FL, USA) is used to pass shuttling sutures and shuttle each of the STIK sutures circumferentially across the rotator cuff, progressing in a posterior-to-anterior direction, with care taken not to tangle the sutures (Fig. 1). Once all of the STIK sutures are shuttled through the anterolateral cannula and rotator cuff, the graft is folded and held with a grasper to facilitate passage through the cannula. The passed STIK sutures are pulled simultaneously with even tension in all directions as the graft is passed into position over the rotator cuff. Each suture end is sequentially tightened to unfold the graft and cover the repair site. The STIKs are sequentially retrieved and tied with an SMC (Samsung Medical Center) sliding-locking knot followed by 3 alternating half-hitches, thereby stabilizing the graft. The shoulder is abducted to 45°, and the 2 simple lateral sutures are placed and tensioned appropriately in knotless anchors just over the edge of the lateral footprint (Fig. 2). If the long head of the biceps is pathologic, tenotomy (n = 3) or subpectoral tenodesis (n = 6) is performed. The proximal portion of the long head of the biceps is resected using a radiofrequency probe. For tenodesis, the arm is abducted and slightly internally rotated, and the skin is incised in the axillary crease from 1 cm superior to 2 cm inferior to the inferior border of the pectoralis major
3
Figure 1 Human acellular dermal matrix allograft with shorttail interference knot sutures in place at edge of anterolateral cannula and ready for passage.
Figure 2 Acellular human dermal matrix allograft sutured in place over rotator cuff repair. tendon. By use of the interval between the pectoralis major superiorly and the short head of the biceps inferiorly, the long head of the biceps tendon is retrieved in the bicipital groove, externalized, and whipstitched. The tendon is passed through a unicortical bone tunnel and tied over a bone bridge. Postoperatively, patients wear an abduction sling for 6 weeks. Passive range of motion including pendulum exercises, as well as hand, wrist, and elbow exercises, is initiated immediately after surgery 3 times daily. At 6 weeks, patients begin supervised physical therapy allowing active and active-assisted range of motion and scapular exercises. Resisted biceps exercises begin at week 8 in those patients who undergo concomitant biceps tenodesis. Progressive stepwise strengthening is allowed at week 14.
Data collection Data were retrospectively retrieved for analysis. Demographic data (age, sex, smoking status, history of diabetes, and workers’ compensation status), surgical history (prior rotator cuff repairs or other surgical procedures on the operative shoulder), intraoperative data (surgical techniques, rotator cuff tear size, AHDM allograft patch size, and concomitant pathologies), and perioperative complications were collected for analysis. Preoperative MRI was reviewed for the degree of Goutallier muscle atrophy as defined by Fuchs et al.9
ARTICLE IN PRESS 4 Clinical outcome scores, including ASES and SANE scores, were obtained postoperatively. When available (n = 6), preoperative ASES and SANE scores were obtained from a prospective rotator cuff database. We did not routinely perform postoperative MRI scans or ultrasound examinations. Patients who had a postoperative traumatic event, functional weakness, or persistent pain received an MRI scan to evaluate rotator cuff repair integrity. Ultrasound examinations were performed if patients were willing. Treatment failure was defined as having a recurrent tear or undergoing subsequent surgery, such as revision rotator cuff repair or reverse total shoulder replacement.
Statistical analyses The P values for comparing binary and categorical variables in patients with versus without a retear were computed with the Fisher exact test. The P values for comparing continuous variables (eg, ASES and SANE scores) with versus without a retear were computed using the nonparametric Wilcoxon rank sum test because these variables did not follow the normal distribution. The P values for postoperative outcome scores were compared with categorical covariates using the nonparametric Kruskal-Wallis test. The association between postoperative outcome scores and continuous covariates was assessed using nonparametric Spearman correlation (rs) because relationships were monotonic but not necessarily linear. P < .05 indicated statistical significance. Statistical analyses were performed using JMP Pro (version 13; SAS, Cary, NC, USA) and the R program (version 3.3.2; R Foundation for Statistical Computing, Vienna, Austria, https:// www.r-project.org/foundation). We did not perform a power analysis prior to the initiation of this study.
Results A total of 28 patients met the inclusion criteria, and 23 (82%) (19 men and 4 women) were available for follow-up, with a mean age of 60.1 ± 9.3 years (range, 43-79 years). The median number of prior shoulder surgical procedures was 1 (range, 1-3). Of the patients, 15 underwent prior arthroscopic rotator cuff repair, 10 underwent prior open rotator cuff repair, and 2 underwent both. One patient underwent a prior incision and drainage for a superficial wound infection. The average time from prior surgery to the index operation was 49 ± 60 months (range, 2-179 months). Tobacco use was reported by 3 patients (13%), 3 patients (13%) had a history of diabetes, and 10 patients (43%) were workers’ compensation patients. The average Goutallier grade of muscle atrophy on preoperative MRI was 2.0 ± 0.7. Of the patients, 17 (74%) underwent concomitant subacromial decompression, 3 (13%) underwent biceps tenotomy, 6 (26%) underwent biceps tenodesis, 3 (13%) underwent subscapularis tendon repair, and 7 (30%) required margin convergence. The average tear size from anterior to posterior was 3.7 ± 0.7 cm. The average AHDM patch size was 10.4 ± 2.9 cm2 (Table I). In total, 13 of the 23 patients (56%) underwent postoperative imaging; 8 (35%) had an MRI scan, 10 (43%) had an ultrasound, and 5 (21%) had both. The postoperative MRI scans were obtained for a traumatic event, pain, or weakness
E.A. Hohn et al. Table I
Patient data Data
Patient factors Mean age (SD), y Male/female sex, n Smoker, n (%) Diabetes mellitus, n (%) Goutallier grade of muscle atrophy, mean (SD) Surgical factors Median No. of prior cuff repairs (range) Time from prior surgery to index surgery, mean (SD), mo Prior arthroscopic surgery, n (%) Prior open surgery, n (%) Prior arthroscopic and open surgery, n (%) Subacromial decompression, n (%) Subscapularis tears, n (%) Biceps tenotomy, n (%) Biceps tenodesis, n (%) Tear size from anterior to posterior, mean (SD), cm Patch size, mean (SD), cm2
60.1 (9.3) 19/4 3 (13) 3 (13) 2 (0.7) 1 (1-3) 49 (60) 15 (65) 10 (43) 2 (9) 17 (74) 3 (13) 3 (13) 6 (26) 3.7 (0.7) 10.4 (2.9)
SD, standard deviation.
and were taken at an average of 12 months (range, 3-64 months) postoperatively. The ultrasound examinations were obtained at a mean follow-up of 36 months (range, 16-83 months). Of the 13 patients with postoperative imaging, 9 (69%) showed intact repair constructs whereas the other 4 had evidence of a retear. In all 4 of these patients, traumatic events occurred with confirmed retears on MRI at an average of 22 months (range, 3-72 months) postoperatively. Two of these patients underwent revision rotator cuff repair and AHDM graft augmentation, one went on to undergo a reverse total shoulder arthroplasty, and one was managed nonoperatively. In total, 4 of 13 patients with postoperative imaging had a retear, which represented 17% of our entire cohort. There were no adverse effects as a result of the AHDM allografts. Minimum 2-year outcome scores were obtained at an average of 48 ± 23 months. The overall cohort postoperative ASES and SANE scores (mean ± standard deviation) were 77 ± 16 and 69 ± 21, respectively. Among the 6 patients with preoperative outcome scores, the mean ASES score significantly improved from 56 to 85 (change in score, 29; P = .03) and the mean SANE score significantly improved from 42 to 76 (change in score, 34; P = .03) (Table II). Patients without a retear had a statistically significantly higher postoperative ASES score than those with a retear: 80 versus 61 (P = .03). There was not a statistically significant difference in the postoperative SANE scores between patients with and patients without retears (Table III). Comparisons of patients with versus without a retear showed that there was a significant difference in the time from prior surgery to the index operation between the 2 groups.
ARTICLE IN PRESS Revision rotator cuff repair with augmentation
5
Table II Comparison of preoperative and postoperative patientreported outcome scores Preoperative, Postoperative, Change in P value mean (SD), mean (SD), score, mean n=6 n = 23 (SD), n = 6 ASES score 56 (16) SANE score 42 (18)
77 (16) 69 (21)
29 (20) 34 (10)
.03 .03
SD, standard deviation; ASES, American Shoulder and Elbow Surgeons; SANE, Single Assessment Numeric Evaluation.
Table III Comparison of postoperative patient-reported outcome scores between patients who had retear or failure and those who did not (success) Postoperative ASES score, mean (SD) Postoperative SANE score, mean (SD)
Success
Retear or failure
P value
80 (14)
61 (18)
.03
73 (19)
51 (25)
.08
ASES, American Shoulder and Elbow Surgeons; SD, standard deviation; SANE, Single Assessment Numeric Evaluation.
The failure group underwent index surgery at an average of 7.0 ± 3.4 months after prior rotator cuff repair compared with 57.8 ± 63.4 months in the non-failure group (P = .03). No other variables showed statistical significance between these groups (Table IV). A shorter duration between failed rotator cuff surgery and revision surgery with graft augmentation was also associated with worse postoperative ASES (rs = 0.54, P = .008) and SANE (rs = 0.52, P = .01) scores. Patients who underwent subacromial decompression had a significantly higher mean postoperative SANE score (80 vs 50, P = .009). Patients who underwent margin convergence at the time of surgery had a significantly lower mean postoperative SANE score than those who did not (60 vs 80, P = .048). There was no correlation between postoperative ASES scores and subacromial decompression or margin convergence. There were significant increases in mean postoperative ASES (rs = 0.50, P = .016) and SANE (rs = 0.45, P = .030) scores with increasing age above 62 years. There was no correlation with age in patients younger than 62 years. Smoking status, history of diabetes mellitus, workers’ compensation status, concomitant biceps tenotomy or tenodesis, subscapularis repair, Goutallier grade of muscle atrophy, tear size, and patch size did not correlate with postoperative outcome scores, although with only 23 patients, this study is likely underpowered to make these correlations.
Discussion The most important finding in this study was that arthroscopic revision rotator cuff repair with AHDM allograft
augmentation was a safe treatment method with an acceptable success rate for patients with failed rotator cuff repairs at a minimum of 2 years and mean follow-up of 4 years. The imaging-confirmed symptomatic retear rate was 17%, and 13% of patients required subsequent surgery, leaving 83% of patients without a retear and 87% who did not require additional surgery. Although only 6 patients in our study had preoperative and postoperative ASES and SANE scores, both of these outcome measures showed statistically significant clinical improvement after augmentation. Revision rotator cuff surgery presents a significant challenge and will become more ubiquitous as the number of rotator cuff repairs continues to rise.7,14 Despite an increase in the amount of pertinent research and number of relevant articles on rotator cuff surgery over the last 2 decades, the clinical and structural results have not shown significant consistent improvement.20 Large tears, even in a primary setting, when repaired, often have unacceptably high structural failure rates: ranging as high as 76%-94% in some studies.5,10 Furthermore, in the revision surgery setting, results are even poorer and are associated with longer operating times, more demanding surgical techniques, higher retear rates, and worse patient outcomes compared with primary repair.8,16,19,25 Specifically, Djurasovic et al8 evaluated 80 patients undergoing revision open rotator cuff repair, with 31% having an unsatisfactory outcome. Bigliani et al4 reported on 31 patients undergoing revision repairs of rotator cuff tendons, finding that although 81% had satisfactory pain relief, 45% reported persistent weakness that led to an unsatisfactory result. More recently, Keener et al16 showed reliable pain relief but less overall functional improvement and only 50% of repairs were intact at 1 year. New approaches and innovative techniques to facilitate the healing of the rotator cuff to the footprint are necessary, as a healed rotator cuff is associated with improved function13,15,17,21,28 and continues to be the intuitive goal of intervention. Unfortunately, most patients with rotator cuff repair failure have multiple biological barriers to healing,24 including thin or friable tendons, fatty infiltration, muscle atrophy, loss of muscle elasticity, poor bone stock, and adhesions. Some cases will be further complicated by smoking, microvascular disease, and other cellular-level processes that may not have been identified. Augmentation with a strong biological scaffold onto the rotator cuff provides biomechanical support for the healing repair and provides a structural scaffold for potential ingrowth. Van Kampen et al29 demonstrated that the addition of a reconstituted collagen scaffold to the bursal surface of the rotator cuff in a sheep model rapidly induced a thick layer of new fibroblasts and collagen over the graft. When used for the primary repair of large tears by Barber et al,2 arthroscopic graft augmentation has been shown to provide statistically significant improvement over non-augmented repairs in terms of both healing rates and clinical outcomes. However, augmenting a rotator cuff repair is currently technically demanding in the arthroscopic setting, and the augmentation may require a significant amount of additional
ARTICLE IN PRESS 6
E.A. Hohn et al. Table IV
Comparison of patient variables between patients who had retear or failure and those who did not (success)
Patient variable
Success
Failure
P value
Tobacco use, n (%) Diabetes mellitus, n (%) Age, mean (SD), y Workers’ compensation status, n (%) Previous arthroscopic surgery, n (%) Previous open surgery, n (%) Previous total No. of surgical procedures, mean (SD) Time from prior surgery to AHDM augmentation, mean (SD), mo Tear size from anterior to posterior, mean (SD), cm Patch size, mean (SD), cm2 Goutallier grade of muscle atrophy, mean (SD) Subacromial decompression, n (%) Margin convergence, n (%) Biceps tenodesis or tenotomy, n (%) Subscapularis repair, n (%)
2 (11) 3 (16) 60.3 (10.1) 9 (47) 12 (63) 8 (42) 1.5 (0.8) 58 (63) 3.7 (0.7) 10.8 (3.0) 2.1 (0.68) 15 (79) 6 (32) 9 (47) 2 (11)
2 (50) 0 59.3 (3.1) 1 (25) 3 (75) 2 (50) 1.3 (0.5) 7 (3) 4.0 (0.4) 8.8 (1.5) 1.5 (0.58) 2 (50) 2 (50) 0 1 (25)
.12 .39 .92 .60 .65 .77 .81 .03 .23 .09 .15 .23 .59 .29 .45
SD, standard deviation; AHDM, acellular human dermal matrix.
surgical time, preparation, implants, and overall cost, which is not negligible. A clinical benefit must be shown to justify this additional time and expense of the graft. This investigation clinically evaluates the results of arthroscopic augmentation of revision rotator cuff repairs using an AHDM allograft in an effort to improve the biological healing rate and outcomes. There are no other studies currently available that have evaluated outcomes of an all-arthroscopic rotator cuff repair augmentation in a revision setting. However, in a recent study by Sears et al,24 patient-reported outcomes following open revision rotator cuff repair with an extracellular matrix (ECM) patch were excellent or good in only 37% of patients, and 63% of patients reported fair or poor results with postoperative imaging (MRI or ultrasound) showing a retear. These results are comparable with reported historical outcomes in populations undergoing open revision repair without the use of ECM patches, and the authors concluded that the open approach with augmentation did not provide value.24 However, in the subset of patients who had a healed tendon on postoperative imaging, the ASES and SANE scores were 81.0 and 77.9, respectively, which were significantly better than those in patients who had retears (40 and 48.3, respectively). Similarly, in our study postoperative ASES scores were significantly lower in patients with an imaging-confirmed retear than in the rest of the cohort (61 vs 80, P = .03). Postoperative SANE scores, however, were not statistically different between these groups. In another study, Petri et al22 showed promising results following open revision repair of massive rotator cuff tears with human acellular dermal ECM patch augmentation. They had a 76% overall success rate, with functional failures in 3 patients, 1 of whom had a retear on postoperative MRI, and no patients requiring subsequent surgery at an average of 2.5
years’ follow-up. In their series they showed significant improvement in clinical outcome scores, with a postoperative ASES score of 86 and SANE score of 74.8. The latter finding is similar to our study’s postoperative outcomes (77 and 69, respectively). An interesting finding in our study was that a shorter duration between the most recent rotator cuff repair and AHDM graft augmentation was associated with significantly worse outcomes and a higher failure rate. It is not exactly clear why this may be, but it could be related to inherent patient-related factors, such as poor tissue quality or poor postoperative compliance, that put patients at increased risk of failure of both the primary repair and the revision surgical procedure. We also noted that those patients who underwent concomitant subacromial decompression had higher postoperative SANE scores than those who did not (80 vs 50, P = .009); however, the postoperative ASES scores did not differ. It may be that the decompression releases bone marrow elements that aid in healing, although this difference is also not completely understood. Patients who received a margin convergence stitch to reduce their rotator cuff had a statistically lower SANE score than those who did not (60 vs 80, P = .048). Tear patterns that are complex enough to require margin convergence for reduction to the footprint could be under more tension, leading to worse outcomes. Finally, we noted a significant increase in postoperative SANE and ASES scores as age increased above 62 years. There was no difference in outcome scores up until age 62 years, and as age increased above this, the outcome scores began to increase in a linear fashion. We attribute this to a lower demand on the shoulder as patients age. This finding does, however, reinforce the idea that older age should not be a contraindication to this procedure. Future research is needed to address these findings and evaluate them further.
ARTICLE IN PRESS Revision rotator cuff repair with augmentation We did not observe any correlation between graft size and retear or failure in this cohort. This is different from previous studies in the primary and revision setting that have shown that healing rates diminish with increasing tear size. Our data show that revision arthroscopic rotator cuff repair with AHDM allograft augmentation for selected patients is a viable and safe method by which good clinical results, low retear rates, and low reoperation rates can be achieved. Any retrospective review has inherent limitations, such as selection and information bias, in comparison with a prospective randomized study. Furthermore, in this study, without a control group, it is difficult to know whether the patients would have had similar improvements with a standard repair. Other potentially confounding variables in rotator cuff healing and/or functional outcomes were not evaluated, such as osteoporosis and the acromial index. Only 6 patients had preoperative outcome scores, which limits our ability to evaluate a change from the preoperative period to the postoperative period and is another source of potential selection bias. The improvements seen in those 6 patients may not necessarily be applicable to our entire cohort. Similarly, without postoperative imaging of all patients, asymptomatic tears may be missed, falsely lowering our reported retear rate. In addition, only 4 patients had retears, which may be too few for adequate statistical power. Only 56% of patients in our study underwent postoperative imaging; however, we believe that because postoperative imaging was obtained in all patients with unsatisfactory pain or function and in those patients with a traumatic event, we were less likely to miss retears and those that we did miss were clinically insignificant. Ultrasound was used because of the accessibility, low cost, and shorter time requirement; however, it is very user dependent, and it can be challenging to see very small tears. We did not perform a power analysis prior to the initiation of this study, and we likely did not have a large enough series of patients to adequately analyze many of the variables in our multivariate analyses. Future studies with larger cohorts could help to adequately determine which variables are predictive of outcomes.
Conclusion AHDM allograft augmentation is a safe and effective treatment method for patients with full-thickness rotator cuff retears. These preliminary results are encouraging to justify the additional surgical time and expense of the graft.
Acknowledgment We acknowledge Treny Sasyniuk, MSc, for her assistance with the manuscript and statistical insight, as well as Jeffrey Gornbein for the statistical analysis.
7
Disclaimer Joseph P. Burns receives consultant payments from ConMed Linvatec and DePuy Mitek. His institution’s research arm also receives royalties and financial or material support from Wolters Kluwer. These funds were not used for data collection, data analysis, or the preparation or editing of the manuscript. All the other authors, their immediate families, and any research foundations 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.
References 1. Baker AR, McCarron JA, Tan CD, Iannotti JP, Derwin KA. Does augmentation with a reinforced fascia patch improve rotator cuff repair outcomes? Clin Orthop Relat Res 2012;470:2513-21. http://dx.doi.org/ 10.1007/s11999-012-2348-x 2. Barber FA, Burns JP, Deutsch A, Labbé MR, Litchfield RB. A prospective, randomized evaluation of acellular human dermal matrix augmentation for arthroscopic rotator cuff repair. Arthroscopy 2012;28:815. http://dx.doi.org/10.1016/j.arthro.2011.06.038 3. Beitzel K, Chowaniec DM, McCarthy MB, Cote MP, Russell RP, Obopilwe E, et al. Stability of double-row rotator cuff repair is not adversely affected by scaffold interposition between tendon and bone. Am J Sports Med 2012;40:1148-54. http://dx.doi.org/10.1177/ 0363546512437835 4. Bigliani LU, Cordasco FA, McIlveen SJ, Musso ES. Operative treatment of failed repairs of the rotator cuff. J Bone Joint Surg Am 1992;74:150515. 5. Bishop J, Klepps S, Lo IK, Bird J, Gladstone JN, Flatow EL. Cuff integrity after arthroscopic versus open rotator cuff repair: a prospective study. J Shoulder Elbow Surg 2006;15:290-9. http://dx.doi.org/10.1016/ j.jse.2005.09.017 6. Bond JL, Dopirak RM, Higgins J, Burns J, Snyder SJ. Arthroscopic replacement of massive, irreparable rotator cuff tears using a GraftJacket allograft: technique and preliminary results. Arthroscopy 2008;24:403-9, e1. http://dx.doi.org/10.1016/j.arthro.2007.07.033 7. Colvin AC, Egorova N, Harrison AK, Moskowitz A, Flatow EL. National trends in rotator cuff repair. J Bone Joint Surg Am 2012;94:227-33. http://dx.doi.org/10.2106/JBJS.J.00739 8. Djurasovic M, Marra G, Arroyo JS, Pollock RG, Flatow EL, Bigliani LU. Revision rotator cuff repair: factors influencing results. J Bone Joint Surg Am 2001;83:1849-55. 9. Fuchs B, Weishaupt D, Zanetti M, Hodler J, Gerber C. Fatty degeneration of the muscles of the rotator cuff: assessment by computed tomography versus magnetic resonance imaging. J Shoulder Elbow Surg 1999;8:599605. 10. Galatz LM, Ball CM, Teefey SA, Middleton WD, Yamaguchi K. The outcome and repair integrity of completely arthroscopically repaired large and massive rotator cuff tears. J Bone Joint Surg Am 2004;86:21924. 11. Gerber C, Fuchs B, Hodler J. The results of repair of massive tears of the rotator cuff. J Bone Joint Surg Am 2000;82:505-15. 12. Giannotti S, Ghilardi M, Dell’osso G, Magistrelli L, Bugelli G, Di Rollo F, et al. Study of the porcine dermal collagen repair patch in morphofunctional recovery of the rotator cuff after minimum follow-up of 2.5 years. Surg Technol Int 2014;24:348-52. 13. Harryman DT, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA. Repairs of the rotator cuff. Correlation of functional
ARTICLE IN PRESS 8
14.
15.
16.
17.
18.
19.
20.
21.
E.A. Hohn et al. results with integrity of the cuff. J Bone Joint Surg Am 1991;73: 982-9. Iyengar JJ, Samagh SP, Schairer W, Singh G, Valone FH III, Feeley BT. Current trends in rotator cuff repair: surgical technique, setting, and cost. Arthroscopy 2014;30:284-8. http://dx.doi.org/10.1016/j.arthro.2013 .11.018 Jost B, Pfirrmann CW, Gerber C, Switzerland Z. Clinical outcome after structural failure of rotator cuff repairs. J Bone Joint Surg Am 2000;82:304-14. Keener JD, Wei AS, Kim HM, Paxton ES, Teefey SA, Galatz LM, et al. Revision arthroscopic rotator cuff repair: repair integrity and clinical outcome. J Bone Joint Surg Am 2010;92:590-8. http://dx.doi.org/ 10.2106/JBJS.I.00267 Kim KC, Shin HD, Lee WY. Repair integrity and functional outcomes after arthroscopic suture-bridge rotator cuff repair. J Bone Joint Surg Am 2012;94:e48. http://dx.doi.org/10.2106/JBJS.K.00158 Le BTN, Wu XL, Lam PH, Murrell GAC. Factors predicting rotator cuff retears: an analysis of 1000 consecutive rotator cuff repairs. Am J Sports Med 2014;42:1134-42. http://dx.doi.org/10.1177/ 0363546514525336 Lo IKY, Burkhart SS. Arthroscopic revision of failed rotator cuff repairs: technique and results. Arthroscopy 2004;20:250-67. http://dx.doi.org/ 10.1016/j.arthro.2004.01.006 McElvany MD, McGoldrick E, Gee AO, Neradilek MB, Matsen FA. Rotator cuff repair: published evidence on factors associated with repair integrity and clinical outcome. Am J Sports Med 2015;43:491-500. http://dx.doi.org/10.1177/0363546514529644 Oh JH, Kim SH, Ji HM, Jo KH, Bin SW, Gong HS. Prognostic factors affecting anatomic outcome of rotator cuff repair and correlation with functional outcome. Arthroscopy 2009;25:30-9. http://dx.doi.org/ 10.1016/j.arthro.2008.08.010
22. Petri M, Warth RJ, Horan MP, Greenspoon JA, Millett PJ. Outcomes after open revision repair of massive rotator cuff tears with biologic patch augmentation. Arthroscopy 2016;32:1752-60. http://dx.doi.org/10.1016/ j.arthro.2016.01.037 23. Samilson RL, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 1983;65:456-60. 24. Sears BW, Choo A, Yu A, Greis A, Lazarus M. Clinical outcomes in patients undergoing revision rotator cuff repair with extracellular matrix augmentation. Orthopedics 2015;38:e292-6. http://dx.doi.org/10.3928/ 01477447-20150402-57 25. Shamsudin A, Lam PH, Peters K, Rubenis I, Hackett L, Murrell GAC. Revision versus primary arthroscopic rotator cuff repair: a 2-year analysis of outcomes in 360 patients. Am J Sports Med 2015;43:557-64. http://dx.doi.org/10.1177/0363546514560729 26. Shea KP, Obopilwe E, Sperling JW, Iannotti JP. A biomechanical analysis of gap formation and failure mechanics of a xenograft-reinforced rotator cuff repair in a cadaveric model. J Shoulder Elbow Surg 2012;21:1072-9. http://dx.doi.org/10.1016/j.jse.2011.07.024 27. Snyder SJ. Shoulder arthroscopy. In: Rotator cuff repair with augmentation. 3rd ed. Philadelphia: Wolter Kluwers Health; 2015. p. 323-4. 28. Thomazeau H, Boukobza E, Morcet N, Chaperon J, Langlais F. Prediction of rotator cuff repair results by magnetic resonance imaging. Clin Orthop Relat Res 1997;(344):275-83. 29. Van Kampen C, Arnoczky S, Parks P, Hackett E, Ruehlman D, Turner A, et al. Tissue-engineered augmentation of a rotator cuff tendon using a reconstituted collagen scaffold: a histological evaluation in sheep. Muscles Ligaments Tendons J 2013;3:229-35. 30. Wong I, Burns J, Snyder S. Arthroscopic GraftJacket repair of rotator cuff tears. J Shoulder Elbow Surg 2010;19(Suppl):104-9. http:// dx.doi.org/10.1016/j.jse.2009.12.017
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
Outcomes of arthroscopic revision rotator cuff repair with acellular human dermal matrix allograft augmentation Eric A. Hohn, MD*, Blake P. Gillette, MD, Joseph P. Burns, MD Southern California Orthopedic Institute, Van Nuys, CA, USA Background: The purpose was to assess the minimum 2-year patient-reported outcomes and failure rate of patients who underwent revision arthroscopic rotator cuff repair augmented with acellular human dermal matrix (AHDM) allograft for repairable retears. Methods: From 2008-2014, patients who underwent revision rotator cuff repair augmented with AHDM with greater than 2 years’ follow-up by a single surgeon were retrospectively reviewed. Data regarding surgical history, demographic characteristics, and medical comorbidities were collected. Outcome data included American Shoulder and Elbow Surgeons (ASES) and Single Assessment Numeric Evaluation (SANE) scores, as well as rotator cuff healing on magnetic resonance imaging or ultrasound. Retears and subsequent surgical procedures were characterized. Results: A total of 28 patients met our inclusion criteria, and 23 (82%) were available for follow-up at 2 years. The mean age was 60.1 ± 9.3 years (range, 43-79 years), with a mean follow-up period of 48 ± 23 months. All patients had at least 1 prior rotator cuff repair. Of the 23 patients, 13 (56%) underwent postoperative imaging, and 4 of these 13 (31%) had a retear. A reoperation was performed in 3 of 23 patients (13%). Among the 6 patients with both preoperative and postoperative outcome scores, we saw improvement in the ASES score from 56 to 85 (P = .03) and in the SANE score from 42 to 76 (P = .03). The full cohort’s mean postoperative ASES and SANE scores were 77 and 69, respectively. Conclusion: AHDM allograft augmentation is a safe and effective treatment method for patients with fullthickness rotator cuff retears. Further research is needed with larger studies to confirm these findings from our small cohort of patients. Level of evidence: Level IV; Case Series; Treatment Study © 2017 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. Keywords: Rotator cuff; revision; shoulder; arthroscopy; allograft; augmentation; acellular human dermal matrix
Institutional review board approval was obtained before the initiation of this study (Western IRB, study No. 1155091). *Reprint requests: Eric A. Hohn, MD, Southern California Orthopedic Institute, 6815 Noble Ave, Van Nuys, CA 91405, USA. E-mail address: [email protected] (E.A. Hohn).
The incidence of rotator cuff surgery continues to increase, with a recent study showing a 6-fold increase in arthroscopic repair from 1996 to 2006 in the United States.7 Moreover, a single-state surgical database showed that even with a decline in open rotator cuff surgery, the total number of repairs doubled from 2000 to 2007.14 Primary arthroscopic rotator cuff repair of small to medium tears is generally
1058-2746/$ - see front matter © 2017 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. https://doi.org/10.1016/j.jse.2017.09.026
ARTICLE IN PRESS 2 successful; however, several risk factors for failure have been identified, including large tear size, poor tendon quality, increased degree of retraction, and the presence of fatty degeneration, among others.11,18 Repair of larger tears has an increased risk of non-healing, approaching 94% in some studies.5,10 Although structural failure is not always correlated with clinical failure,20 several studies have demonstrated a clinical benefit in patients with a healed rotator cuff.13,15,17,21,28 A study by Shamsudin et al25 showed that patients undergoing revision rotator cuff repair are twice as likely to have retears and have worse postoperative pain and function compared with those undergoing primary repair. Rotator cuff revision surgery is often complicated by tendon loss or retraction, bone defects, and retained hardware. Poor tissue quality also negatively affects both the quality of surgical repair and subsequent healing.8,16,19 As these patients are less likely to achieve optimal outcomes with standard repair techniques, alternative surgical and biological approaches should be investigated. Several new techniques have been developed in an effort to improve outcomes in patients who are at high risk of failure. Biological tissue scaffolds have been introduced as a way to reduce repair tension and enhance healing of the rotator cuff. A variety of biological scaffold materials have been used with favorable results, including human acellular dermis,6 fascia lata,1 and porcine dermal collagen.12 Preliminary biomechanical and clinical evidence suggests that rotator cuff augmentation may be a safe and effective method for the treatment of massive, retracted rotator cuff tears.1-3,6,22,26 Barber et al2 showed that primary repair of large rotator cuff tears had a significantly higher healing rate when augmented with acellular human dermal matrix (AHDM) allograft (85% compared with 40%). Data regarding the use of biological scaffold augmentation in revision rotator cuff repair are lacking and are limited to open repair only.22,24 The purpose of the study was to (1) evaluate the clinical outcomes with American Shoulder and Elbow Surgeons (ASES) and Single Assessment Numeric Evaluation (SANE) scores at a minimum of 2 years’ follow-up and (2) evaluate the retear rate and need for subsequent surgery in patients with a previous failed rotator cuff repair who underwent revision arthroscopic surgery with AHDM allograft augmentation. Our hypothesis was that augmentation with AHDM allograft would be a safe and reliable treatment option for patients with failed rotator cuff repairs.
Methods Study design Between 2008 and 2014, all patients who underwent arthroscopic revision repair of full-thickness (>2 cm) rotator cuff tears augmented with AHDM allograft by a single surgeon (J.P.B.) were identified for this retrospective case-series study. The minimum subjective follow-up period was set at 2 years. The inclusion criteria included full-thickness retears of the supraspinatus and/or infraspinatus
E.A. Hohn et al. that had occurred after prior open or arthroscopic repair, measuring 2 cm or greater in size in the anterior-to-posterior plane. Any degree of tissue atrophy, degeneration, or fatty infiltration was included. Patients with concomitant pathologies such as superior labral anterior-posterior tears, chondromalacia, biceps pathology, or subscapularis tears were included. The exclusion criteria were advanced osteoarthritis according to the Samilson-Prieto classification of grade 2 or higher,23 active infection, or a rotator cuff tendon that could not be mobilized with a residual gap of less than 1 cm. Minimum 2-year follow-up data were obtained using validated shoulder questionnaires.
Surgical treatment The decision to perform rotator cuff repair with patch augmentation was based on preoperative evaluations and magnetic resonance imaging (MRI) appearance, as well as intraoperative tendon quality and mobility. Rotator cuff repair with AHDM augmentation was indicated for patients with full-thickness recurrent tearing of the supraspinatus and/or infraspinatus tendons after failed prior surgery, measuring 2 cm in size from anterior to posterior, with poor tendon quality. Any degree of muscle atrophy and fatty infiltration was acceptable, and all Goutallier grades were included in this study. If the cuff tendon could not be sufficiently mobilized with a residual gap of less than 1 cm, other options such as débridement, partial repair, tendon transfer, reverse total shoulder replacement, or a bridging allograft repair were considered, and those patients were excluded from the study. For all cases included in this series, the AHDM allograft was used for augmentation (not bridging) where the native rotator cuff tendon was secured to the medial footprint on the tuberosity. The graft was placed over the top of the rotator cuff tendon and secured circumferentially: anteriorly, posteriorly, and medially to the remaining rotator cuff tissue and laterally to the greater tuberosity, as previously described.2,27,30
Surgical technique After the administration of a regional interscalene block and induction of general anesthesia, the patient is placed in the lateral decubitus position. All bony prominences are carefully padded, and an axillary roll is placed. Ten pounds of balanced suspension in 70° of abduction is used for glenohumeral access, 15 lb in 15° of abduction is used for bursal access, and 10 lb in approximately 45° of abduction is used as a middle position for accessing the lateral aspect of the greater tuberosity in the bursal space. The operative field is prepared and draped using sterile techniques. Standard posterior and anterosuperior portals are established, a diagnostic arthroscopy is performed, and all intra-articular pathologies are addressed as necessary. An accessory lateral portal is established, and a diagnostic subacromial bursoscopy is performed. The rotator cuff tendon quality and mobility are assessed, and intra- and extra-articular releases are performed using an elevator and radiofrequency probe when necessary. The decision of whether to perform rotator cuff repair with or without augmentation is then made. Tears with normal to excellent tendon quality are repaired primarily, whereas tendons with atrophy, poor-quality suture holding, and abnormal fatty infiltration are selected for augmentation. Retained anchors and sutures from prior surgical procedures are then removed as indicated. If the
ARTICLE IN PRESS Revision rotator cuff repair with augmentation prior fixation is unlikely to interfere with the new repair or if removal will result in substantial bone loss, then the hardware is retained. If subacromial impingement is identified, an acromioplasty is performed. Rotator cuff repair with allograft augmentation is then performed. Four portals are used: 7-mm cannulas are placed anteriorly and posteriorly, an 8-mm cannula is placed anterolaterally to allow future passage of the graft, and a posterolateral portal is established for viewing with the arthroscope. The rotator cuff tear size is measured with a knotted suture-measuring device that is made with a No. 1 braided suture with knots tied 1 cm apart. The measuring suture is held with a grasper on one end while the other end is passed loaded in a standard knot pusher such that it can easily slide back and forth through the eyelet of the knot pusher. The size of the tear is measured by counting the number of knots between the instruments and is measured in both the anterior-to-posterior and medial-to-lateral directions. This measurement is used to help determine the graft size needed for augmentation. The greater tuberosity footprint is débrided to a bleeding surface using a motorized shaver. For U-shaped tears, margin convergence techniques are used to close the defects in the cuff tendon in a side-to-side fashion. L-shaped and reverse L–shaped tears are repaired using a combination of margin convergence and direct repair to the tuberosity. The native cuff is secured using a single-row repair with triple-loaded suture anchors placed on the medial footprint adjacent to the articular cartilage. Bone marrow vents are created using a 1-mm bone punch in the greater tuberosity, prior to graft placement. The graft is then prepared. The size of the allograft is determined based on the size of the tear prior to repair. The graft is intentionally oversized 3 mm on each side to allow placement of sutures; for instance, a pre-repair tear size of 3 × 3 cm will require a 3.6 × 3.6–cm allograft. The AHDM allograft is then cut to size. No. 2 simple braided permanent sutures are used for fixation of the graft. Two simple sutures are placed laterally in the graft for eventual lateral footprint fixation, and short-tail interference knots (STIKs) are placed around the remainder of the graft periphery, spaced approximately 1 cm apart. These sutures are placed through the graft while it is on the back table, and the graft is then placed adjacent to the lateral 8-mm cannula. Most grafts require 6-7 sutures. A retrograde suturing device (Spectrum; ConMed, Largo, FL, USA) is used to pass shuttling sutures and shuttle each of the STIK sutures circumferentially across the rotator cuff, progressing in a posterior-to-anterior direction, with care taken not to tangle the sutures (Fig. 1). Once all of the STIK sutures are shuttled through the anterolateral cannula and rotator cuff, the graft is folded and held with a grasper to facilitate passage through the cannula. The passed STIK sutures are pulled simultaneously with even tension in all directions as the graft is passed into position over the rotator cuff. Each suture end is sequentially tightened to unfold the graft and cover the repair site. The STIKs are sequentially retrieved and tied with an SMC (Samsung Medical Center) sliding-locking knot followed by 3 alternating half-hitches, thereby stabilizing the graft. The shoulder is abducted to 45°, and the 2 simple lateral sutures are placed and tensioned appropriately in knotless anchors just over the edge of the lateral footprint (Fig. 2). If the long head of the biceps is pathologic, tenotomy (n = 3) or subpectoral tenodesis (n = 6) is performed. The proximal portion of the long head of the biceps is resected using a radiofrequency probe. For tenodesis, the arm is abducted and slightly internally rotated, and the skin is incised in the axillary crease from 1 cm superior to 2 cm inferior to the inferior border of the pectoralis major
3
Figure 1 Human acellular dermal matrix allograft with shorttail interference knot sutures in place at edge of anterolateral cannula and ready for passage.
Figure 2 Acellular human dermal matrix allograft sutured in place over rotator cuff repair. tendon. By use of the interval between the pectoralis major superiorly and the short head of the biceps inferiorly, the long head of the biceps tendon is retrieved in the bicipital groove, externalized, and whipstitched. The tendon is passed through a unicortical bone tunnel and tied over a bone bridge. Postoperatively, patients wear an abduction sling for 6 weeks. Passive range of motion including pendulum exercises, as well as hand, wrist, and elbow exercises, is initiated immediately after surgery 3 times daily. At 6 weeks, patients begin supervised physical therapy allowing active and active-assisted range of motion and scapular exercises. Resisted biceps exercises begin at week 8 in those patients who undergo concomitant biceps tenodesis. Progressive stepwise strengthening is allowed at week 14.
Data collection Data were retrospectively retrieved for analysis. Demographic data (age, sex, smoking status, history of diabetes, and workers’ compensation status), surgical history (prior rotator cuff repairs or other surgical procedures on the operative shoulder), intraoperative data (surgical techniques, rotator cuff tear size, AHDM allograft patch size, and concomitant pathologies), and perioperative complications were collected for analysis. Preoperative MRI was reviewed for the degree of Goutallier muscle atrophy as defined by Fuchs et al.9
ARTICLE IN PRESS 4 Clinical outcome scores, including ASES and SANE scores, were obtained postoperatively. When available (n = 6), preoperative ASES and SANE scores were obtained from a prospective rotator cuff database. We did not routinely perform postoperative MRI scans or ultrasound examinations. Patients who had a postoperative traumatic event, functional weakness, or persistent pain received an MRI scan to evaluate rotator cuff repair integrity. Ultrasound examinations were performed if patients were willing. Treatment failure was defined as having a recurrent tear or undergoing subsequent surgery, such as revision rotator cuff repair or reverse total shoulder replacement.
Statistical analyses The P values for comparing binary and categorical variables in patients with versus without a retear were computed with the Fisher exact test. The P values for comparing continuous variables (eg, ASES and SANE scores) with versus without a retear were computed using the nonparametric Wilcoxon rank sum test because these variables did not follow the normal distribution. The P values for postoperative outcome scores were compared with categorical covariates using the nonparametric Kruskal-Wallis test. The association between postoperative outcome scores and continuous covariates was assessed using nonparametric Spearman correlation (rs) because relationships were monotonic but not necessarily linear. P < .05 indicated statistical significance. Statistical analyses were performed using JMP Pro (version 13; SAS, Cary, NC, USA) and the R program (version 3.3.2; R Foundation for Statistical Computing, Vienna, Austria, https:// www.r-project.org/foundation). We did not perform a power analysis prior to the initiation of this study.
Results A total of 28 patients met the inclusion criteria, and 23 (82%) (19 men and 4 women) were available for follow-up, with a mean age of 60.1 ± 9.3 years (range, 43-79 years). The median number of prior shoulder surgical procedures was 1 (range, 1-3). Of the patients, 15 underwent prior arthroscopic rotator cuff repair, 10 underwent prior open rotator cuff repair, and 2 underwent both. One patient underwent a prior incision and drainage for a superficial wound infection. The average time from prior surgery to the index operation was 49 ± 60 months (range, 2-179 months). Tobacco use was reported by 3 patients (13%), 3 patients (13%) had a history of diabetes, and 10 patients (43%) were workers’ compensation patients. The average Goutallier grade of muscle atrophy on preoperative MRI was 2.0 ± 0.7. Of the patients, 17 (74%) underwent concomitant subacromial decompression, 3 (13%) underwent biceps tenotomy, 6 (26%) underwent biceps tenodesis, 3 (13%) underwent subscapularis tendon repair, and 7 (30%) required margin convergence. The average tear size from anterior to posterior was 3.7 ± 0.7 cm. The average AHDM patch size was 10.4 ± 2.9 cm2 (Table I). In total, 13 of the 23 patients (56%) underwent postoperative imaging; 8 (35%) had an MRI scan, 10 (43%) had an ultrasound, and 5 (21%) had both. The postoperative MRI scans were obtained for a traumatic event, pain, or weakness
E.A. Hohn et al. Table I
Patient data Data
Patient factors Mean age (SD), y Male/female sex, n Smoker, n (%) Diabetes mellitus, n (%) Goutallier grade of muscle atrophy, mean (SD) Surgical factors Median No. of prior cuff repairs (range) Time from prior surgery to index surgery, mean (SD), mo Prior arthroscopic surgery, n (%) Prior open surgery, n (%) Prior arthroscopic and open surgery, n (%) Subacromial decompression, n (%) Subscapularis tears, n (%) Biceps tenotomy, n (%) Biceps tenodesis, n (%) Tear size from anterior to posterior, mean (SD), cm Patch size, mean (SD), cm2
60.1 (9.3) 19/4 3 (13) 3 (13) 2 (0.7) 1 (1-3) 49 (60) 15 (65) 10 (43) 2 (9) 17 (74) 3 (13) 3 (13) 6 (26) 3.7 (0.7) 10.4 (2.9)
SD, standard deviation.
and were taken at an average of 12 months (range, 3-64 months) postoperatively. The ultrasound examinations were obtained at a mean follow-up of 36 months (range, 16-83 months). Of the 13 patients with postoperative imaging, 9 (69%) showed intact repair constructs whereas the other 4 had evidence of a retear. In all 4 of these patients, traumatic events occurred with confirmed retears on MRI at an average of 22 months (range, 3-72 months) postoperatively. Two of these patients underwent revision rotator cuff repair and AHDM graft augmentation, one went on to undergo a reverse total shoulder arthroplasty, and one was managed nonoperatively. In total, 4 of 13 patients with postoperative imaging had a retear, which represented 17% of our entire cohort. There were no adverse effects as a result of the AHDM allografts. Minimum 2-year outcome scores were obtained at an average of 48 ± 23 months. The overall cohort postoperative ASES and SANE scores (mean ± standard deviation) were 77 ± 16 and 69 ± 21, respectively. Among the 6 patients with preoperative outcome scores, the mean ASES score significantly improved from 56 to 85 (change in score, 29; P = .03) and the mean SANE score significantly improved from 42 to 76 (change in score, 34; P = .03) (Table II). Patients without a retear had a statistically significantly higher postoperative ASES score than those with a retear: 80 versus 61 (P = .03). There was not a statistically significant difference in the postoperative SANE scores between patients with and patients without retears (Table III). Comparisons of patients with versus without a retear showed that there was a significant difference in the time from prior surgery to the index operation between the 2 groups.
ARTICLE IN PRESS Revision rotator cuff repair with augmentation
5
Table II Comparison of preoperative and postoperative patientreported outcome scores Preoperative, Postoperative, Change in P value mean (SD), mean (SD), score, mean n=6 n = 23 (SD), n = 6 ASES score 56 (16) SANE score 42 (18)
77 (16) 69 (21)
29 (20) 34 (10)
.03 .03
SD, standard deviation; ASES, American Shoulder and Elbow Surgeons; SANE, Single Assessment Numeric Evaluation.
Table III Comparison of postoperative patient-reported outcome scores between patients who had retear or failure and those who did not (success) Postoperative ASES score, mean (SD) Postoperative SANE score, mean (SD)
Success
Retear or failure
P value
80 (14)
61 (18)
.03
73 (19)
51 (25)
.08
ASES, American Shoulder and Elbow Surgeons; SD, standard deviation; SANE, Single Assessment Numeric Evaluation.
The failure group underwent index surgery at an average of 7.0 ± 3.4 months after prior rotator cuff repair compared with 57.8 ± 63.4 months in the non-failure group (P = .03). No other variables showed statistical significance between these groups (Table IV). A shorter duration between failed rotator cuff surgery and revision surgery with graft augmentation was also associated with worse postoperative ASES (rs = 0.54, P = .008) and SANE (rs = 0.52, P = .01) scores. Patients who underwent subacromial decompression had a significantly higher mean postoperative SANE score (80 vs 50, P = .009). Patients who underwent margin convergence at the time of surgery had a significantly lower mean postoperative SANE score than those who did not (60 vs 80, P = .048). There was no correlation between postoperative ASES scores and subacromial decompression or margin convergence. There were significant increases in mean postoperative ASES (rs = 0.50, P = .016) and SANE (rs = 0.45, P = .030) scores with increasing age above 62 years. There was no correlation with age in patients younger than 62 years. Smoking status, history of diabetes mellitus, workers’ compensation status, concomitant biceps tenotomy or tenodesis, subscapularis repair, Goutallier grade of muscle atrophy, tear size, and patch size did not correlate with postoperative outcome scores, although with only 23 patients, this study is likely underpowered to make these correlations.
Discussion The most important finding in this study was that arthroscopic revision rotator cuff repair with AHDM allograft
augmentation was a safe treatment method with an acceptable success rate for patients with failed rotator cuff repairs at a minimum of 2 years and mean follow-up of 4 years. The imaging-confirmed symptomatic retear rate was 17%, and 13% of patients required subsequent surgery, leaving 83% of patients without a retear and 87% who did not require additional surgery. Although only 6 patients in our study had preoperative and postoperative ASES and SANE scores, both of these outcome measures showed statistically significant clinical improvement after augmentation. Revision rotator cuff surgery presents a significant challenge and will become more ubiquitous as the number of rotator cuff repairs continues to rise.7,14 Despite an increase in the amount of pertinent research and number of relevant articles on rotator cuff surgery over the last 2 decades, the clinical and structural results have not shown significant consistent improvement.20 Large tears, even in a primary setting, when repaired, often have unacceptably high structural failure rates: ranging as high as 76%-94% in some studies.5,10 Furthermore, in the revision surgery setting, results are even poorer and are associated with longer operating times, more demanding surgical techniques, higher retear rates, and worse patient outcomes compared with primary repair.8,16,19,25 Specifically, Djurasovic et al8 evaluated 80 patients undergoing revision open rotator cuff repair, with 31% having an unsatisfactory outcome. Bigliani et al4 reported on 31 patients undergoing revision repairs of rotator cuff tendons, finding that although 81% had satisfactory pain relief, 45% reported persistent weakness that led to an unsatisfactory result. More recently, Keener et al16 showed reliable pain relief but less overall functional improvement and only 50% of repairs were intact at 1 year. New approaches and innovative techniques to facilitate the healing of the rotator cuff to the footprint are necessary, as a healed rotator cuff is associated with improved function13,15,17,21,28 and continues to be the intuitive goal of intervention. Unfortunately, most patients with rotator cuff repair failure have multiple biological barriers to healing,24 including thin or friable tendons, fatty infiltration, muscle atrophy, loss of muscle elasticity, poor bone stock, and adhesions. Some cases will be further complicated by smoking, microvascular disease, and other cellular-level processes that may not have been identified. Augmentation with a strong biological scaffold onto the rotator cuff provides biomechanical support for the healing repair and provides a structural scaffold for potential ingrowth. Van Kampen et al29 demonstrated that the addition of a reconstituted collagen scaffold to the bursal surface of the rotator cuff in a sheep model rapidly induced a thick layer of new fibroblasts and collagen over the graft. When used for the primary repair of large tears by Barber et al,2 arthroscopic graft augmentation has been shown to provide statistically significant improvement over non-augmented repairs in terms of both healing rates and clinical outcomes. However, augmenting a rotator cuff repair is currently technically demanding in the arthroscopic setting, and the augmentation may require a significant amount of additional
ARTICLE IN PRESS 6
E.A. Hohn et al. Table IV
Comparison of patient variables between patients who had retear or failure and those who did not (success)
Patient variable
Success
Failure
P value
Tobacco use, n (%) Diabetes mellitus, n (%) Age, mean (SD), y Workers’ compensation status, n (%) Previous arthroscopic surgery, n (%) Previous open surgery, n (%) Previous total No. of surgical procedures, mean (SD) Time from prior surgery to AHDM augmentation, mean (SD), mo Tear size from anterior to posterior, mean (SD), cm Patch size, mean (SD), cm2 Goutallier grade of muscle atrophy, mean (SD) Subacromial decompression, n (%) Margin convergence, n (%) Biceps tenodesis or tenotomy, n (%) Subscapularis repair, n (%)
2 (11) 3 (16) 60.3 (10.1) 9 (47) 12 (63) 8 (42) 1.5 (0.8) 58 (63) 3.7 (0.7) 10.8 (3.0) 2.1 (0.68) 15 (79) 6 (32) 9 (47) 2 (11)
2 (50) 0 59.3 (3.1) 1 (25) 3 (75) 2 (50) 1.3 (0.5) 7 (3) 4.0 (0.4) 8.8 (1.5) 1.5 (0.58) 2 (50) 2 (50) 0 1 (25)
.12 .39 .92 .60 .65 .77 .81 .03 .23 .09 .15 .23 .59 .29 .45
SD, standard deviation; AHDM, acellular human dermal matrix.
surgical time, preparation, implants, and overall cost, which is not negligible. A clinical benefit must be shown to justify this additional time and expense of the graft. This investigation clinically evaluates the results of arthroscopic augmentation of revision rotator cuff repairs using an AHDM allograft in an effort to improve the biological healing rate and outcomes. There are no other studies currently available that have evaluated outcomes of an all-arthroscopic rotator cuff repair augmentation in a revision setting. However, in a recent study by Sears et al,24 patient-reported outcomes following open revision rotator cuff repair with an extracellular matrix (ECM) patch were excellent or good in only 37% of patients, and 63% of patients reported fair or poor results with postoperative imaging (MRI or ultrasound) showing a retear. These results are comparable with reported historical outcomes in populations undergoing open revision repair without the use of ECM patches, and the authors concluded that the open approach with augmentation did not provide value.24 However, in the subset of patients who had a healed tendon on postoperative imaging, the ASES and SANE scores were 81.0 and 77.9, respectively, which were significantly better than those in patients who had retears (40 and 48.3, respectively). Similarly, in our study postoperative ASES scores were significantly lower in patients with an imaging-confirmed retear than in the rest of the cohort (61 vs 80, P = .03). Postoperative SANE scores, however, were not statistically different between these groups. In another study, Petri et al22 showed promising results following open revision repair of massive rotator cuff tears with human acellular dermal ECM patch augmentation. They had a 76% overall success rate, with functional failures in 3 patients, 1 of whom had a retear on postoperative MRI, and no patients requiring subsequent surgery at an average of 2.5
years’ follow-up. In their series they showed significant improvement in clinical outcome scores, with a postoperative ASES score of 86 and SANE score of 74.8. The latter finding is similar to our study’s postoperative outcomes (77 and 69, respectively). An interesting finding in our study was that a shorter duration between the most recent rotator cuff repair and AHDM graft augmentation was associated with significantly worse outcomes and a higher failure rate. It is not exactly clear why this may be, but it could be related to inherent patient-related factors, such as poor tissue quality or poor postoperative compliance, that put patients at increased risk of failure of both the primary repair and the revision surgical procedure. We also noted that those patients who underwent concomitant subacromial decompression had higher postoperative SANE scores than those who did not (80 vs 50, P = .009); however, the postoperative ASES scores did not differ. It may be that the decompression releases bone marrow elements that aid in healing, although this difference is also not completely understood. Patients who received a margin convergence stitch to reduce their rotator cuff had a statistically lower SANE score than those who did not (60 vs 80, P = .048). Tear patterns that are complex enough to require margin convergence for reduction to the footprint could be under more tension, leading to worse outcomes. Finally, we noted a significant increase in postoperative SANE and ASES scores as age increased above 62 years. There was no difference in outcome scores up until age 62 years, and as age increased above this, the outcome scores began to increase in a linear fashion. We attribute this to a lower demand on the shoulder as patients age. This finding does, however, reinforce the idea that older age should not be a contraindication to this procedure. Future research is needed to address these findings and evaluate them further.
ARTICLE IN PRESS Revision rotator cuff repair with augmentation We did not observe any correlation between graft size and retear or failure in this cohort. This is different from previous studies in the primary and revision setting that have shown that healing rates diminish with increasing tear size. Our data show that revision arthroscopic rotator cuff repair with AHDM allograft augmentation for selected patients is a viable and safe method by which good clinical results, low retear rates, and low reoperation rates can be achieved. Any retrospective review has inherent limitations, such as selection and information bias, in comparison with a prospective randomized study. Furthermore, in this study, without a control group, it is difficult to know whether the patients would have had similar improvements with a standard repair. Other potentially confounding variables in rotator cuff healing and/or functional outcomes were not evaluated, such as osteoporosis and the acromial index. Only 6 patients had preoperative outcome scores, which limits our ability to evaluate a change from the preoperative period to the postoperative period and is another source of potential selection bias. The improvements seen in those 6 patients may not necessarily be applicable to our entire cohort. Similarly, without postoperative imaging of all patients, asymptomatic tears may be missed, falsely lowering our reported retear rate. In addition, only 4 patients had retears, which may be too few for adequate statistical power. Only 56% of patients in our study underwent postoperative imaging; however, we believe that because postoperative imaging was obtained in all patients with unsatisfactory pain or function and in those patients with a traumatic event, we were less likely to miss retears and those that we did miss were clinically insignificant. Ultrasound was used because of the accessibility, low cost, and shorter time requirement; however, it is very user dependent, and it can be challenging to see very small tears. We did not perform a power analysis prior to the initiation of this study, and we likely did not have a large enough series of patients to adequately analyze many of the variables in our multivariate analyses. Future studies with larger cohorts could help to adequately determine which variables are predictive of outcomes.
Conclusion AHDM allograft augmentation is a safe and effective treatment method for patients with full-thickness rotator cuff retears. These preliminary results are encouraging to justify the additional surgical time and expense of the graft.
Acknowledgment We acknowledge Treny Sasyniuk, MSc, for her assistance with the manuscript and statistical insight, as well as Jeffrey Gornbein for the statistical analysis.
7
Disclaimer Joseph P. Burns receives consultant payments from ConMed Linvatec and DePuy Mitek. His institution’s research arm also receives royalties and financial or material support from Wolters Kluwer. These funds were not used for data collection, data analysis, or the preparation or editing of the manuscript. All the other authors, their immediate families, and any research foundations 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.
References 1. Baker AR, McCarron JA, Tan CD, Iannotti JP, Derwin KA. Does augmentation with a reinforced fascia patch improve rotator cuff repair outcomes? Clin Orthop Relat Res 2012;470:2513-21. http://dx.doi.org/ 10.1007/s11999-012-2348-x 2. Barber FA, Burns JP, Deutsch A, Labbé MR, Litchfield RB. A prospective, randomized evaluation of acellular human dermal matrix augmentation for arthroscopic rotator cuff repair. Arthroscopy 2012;28:815. http://dx.doi.org/10.1016/j.arthro.2011.06.038 3. Beitzel K, Chowaniec DM, McCarthy MB, Cote MP, Russell RP, Obopilwe E, et al. Stability of double-row rotator cuff repair is not adversely affected by scaffold interposition between tendon and bone. Am J Sports Med 2012;40:1148-54. http://dx.doi.org/10.1177/ 0363546512437835 4. Bigliani LU, Cordasco FA, McIlveen SJ, Musso ES. Operative treatment of failed repairs of the rotator cuff. J Bone Joint Surg Am 1992;74:150515. 5. Bishop J, Klepps S, Lo IK, Bird J, Gladstone JN, Flatow EL. Cuff integrity after arthroscopic versus open rotator cuff repair: a prospective study. J Shoulder Elbow Surg 2006;15:290-9. http://dx.doi.org/10.1016/ j.jse.2005.09.017 6. Bond JL, Dopirak RM, Higgins J, Burns J, Snyder SJ. Arthroscopic replacement of massive, irreparable rotator cuff tears using a GraftJacket allograft: technique and preliminary results. Arthroscopy 2008;24:403-9, e1. http://dx.doi.org/10.1016/j.arthro.2007.07.033 7. Colvin AC, Egorova N, Harrison AK, Moskowitz A, Flatow EL. National trends in rotator cuff repair. J Bone Joint Surg Am 2012;94:227-33. http://dx.doi.org/10.2106/JBJS.J.00739 8. Djurasovic M, Marra G, Arroyo JS, Pollock RG, Flatow EL, Bigliani LU. Revision rotator cuff repair: factors influencing results. J Bone Joint Surg Am 2001;83:1849-55. 9. Fuchs B, Weishaupt D, Zanetti M, Hodler J, Gerber C. Fatty degeneration of the muscles of the rotator cuff: assessment by computed tomography versus magnetic resonance imaging. J Shoulder Elbow Surg 1999;8:599605. 10. Galatz LM, Ball CM, Teefey SA, Middleton WD, Yamaguchi K. The outcome and repair integrity of completely arthroscopically repaired large and massive rotator cuff tears. J Bone Joint Surg Am 2004;86:21924. 11. Gerber C, Fuchs B, Hodler J. The results of repair of massive tears of the rotator cuff. J Bone Joint Surg Am 2000;82:505-15. 12. Giannotti S, Ghilardi M, Dell’osso G, Magistrelli L, Bugelli G, Di Rollo F, et al. Study of the porcine dermal collagen repair patch in morphofunctional recovery of the rotator cuff after minimum follow-up of 2.5 years. Surg Technol Int 2014;24:348-52. 13. Harryman DT, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA. Repairs of the rotator cuff. Correlation of functional
ARTICLE IN PRESS 8
14.
15.
16.
17.
18.
19.
20.
21.
E.A. Hohn et al. results with integrity of the cuff. J Bone Joint Surg Am 1991;73: 982-9. Iyengar JJ, Samagh SP, Schairer W, Singh G, Valone FH III, Feeley BT. Current trends in rotator cuff repair: surgical technique, setting, and cost. Arthroscopy 2014;30:284-8. http://dx.doi.org/10.1016/j.arthro.2013 .11.018 Jost B, Pfirrmann CW, Gerber C, Switzerland Z. Clinical outcome after structural failure of rotator cuff repairs. J Bone Joint Surg Am 2000;82:304-14. Keener JD, Wei AS, Kim HM, Paxton ES, Teefey SA, Galatz LM, et al. Revision arthroscopic rotator cuff repair: repair integrity and clinical outcome. J Bone Joint Surg Am 2010;92:590-8. http://dx.doi.org/ 10.2106/JBJS.I.00267 Kim KC, Shin HD, Lee WY. Repair integrity and functional outcomes after arthroscopic suture-bridge rotator cuff repair. J Bone Joint Surg Am 2012;94:e48. http://dx.doi.org/10.2106/JBJS.K.00158 Le BTN, Wu XL, Lam PH, Murrell GAC. Factors predicting rotator cuff retears: an analysis of 1000 consecutive rotator cuff repairs. Am J Sports Med 2014;42:1134-42. http://dx.doi.org/10.1177/ 0363546514525336 Lo IKY, Burkhart SS. Arthroscopic revision of failed rotator cuff repairs: technique and results. Arthroscopy 2004;20:250-67. http://dx.doi.org/ 10.1016/j.arthro.2004.01.006 McElvany MD, McGoldrick E, Gee AO, Neradilek MB, Matsen FA. Rotator cuff repair: published evidence on factors associated with repair integrity and clinical outcome. Am J Sports Med 2015;43:491-500. http://dx.doi.org/10.1177/0363546514529644 Oh JH, Kim SH, Ji HM, Jo KH, Bin SW, Gong HS. Prognostic factors affecting anatomic outcome of rotator cuff repair and correlation with functional outcome. Arthroscopy 2009;25:30-9. http://dx.doi.org/ 10.1016/j.arthro.2008.08.010
22. Petri M, Warth RJ, Horan MP, Greenspoon JA, Millett PJ. Outcomes after open revision repair of massive rotator cuff tears with biologic patch augmentation. Arthroscopy 2016;32:1752-60. http://dx.doi.org/10.1016/ j.arthro.2016.01.037 23. Samilson RL, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 1983;65:456-60. 24. Sears BW, Choo A, Yu A, Greis A, Lazarus M. Clinical outcomes in patients undergoing revision rotator cuff repair with extracellular matrix augmentation. Orthopedics 2015;38:e292-6. http://dx.doi.org/10.3928/ 01477447-20150402-57 25. Shamsudin A, Lam PH, Peters K, Rubenis I, Hackett L, Murrell GAC. Revision versus primary arthroscopic rotator cuff repair: a 2-year analysis of outcomes in 360 patients. Am J Sports Med 2015;43:557-64. http://dx.doi.org/10.1177/0363546514560729 26. Shea KP, Obopilwe E, Sperling JW, Iannotti JP. A biomechanical analysis of gap formation and failure mechanics of a xenograft-reinforced rotator cuff repair in a cadaveric model. J Shoulder Elbow Surg 2012;21:1072-9. http://dx.doi.org/10.1016/j.jse.2011.07.024 27. Snyder SJ. Shoulder arthroscopy. In: Rotator cuff repair with augmentation. 3rd ed. Philadelphia: Wolter Kluwers Health; 2015. p. 323-4. 28. Thomazeau H, Boukobza E, Morcet N, Chaperon J, Langlais F. Prediction of rotator cuff repair results by magnetic resonance imaging. Clin Orthop Relat Res 1997;(344):275-83. 29. Van Kampen C, Arnoczky S, Parks P, Hackett E, Ruehlman D, Turner A, et al. Tissue-engineered augmentation of a rotator cuff tendon using a reconstituted collagen scaffold: a histological evaluation in sheep. Muscles Ligaments Tendons J 2013;3:229-35. 30. Wong I, Burns J, Snyder S. Arthroscopic GraftJacket repair of rotator cuff tears. J Shoulder Elbow Surg 2010;19(Suppl):104-9. http:// dx.doi.org/10.1016/j.jse.2009.12.017