“The box” allograft augmentation for at risk rotator cuff repairs Authors: Hayden Faith, Timothy Baker, Matthew Williamson, Lana Wheeler, Bryan Brown Published: 2026 ABSTRACT AbstractBackgroundRotator cuff tears are a challenging problem for orthopedic surgeons, especially those with at risk features. Because of this it is imperative to establish reliable methods to maximize patient outcomes and mitigate the risk of retears. Biologic augmentation with dermal allograft augmentation (DAA) has been proposed to improve tendon healing and structural integrity. This study describes a novel arthroscopic technique using DAA augmentation with a box-configuration transosseous-equivalent canopy construct and reports early clinical outcomes.MethodsA retrospective review was performed on patients who underwent arthroscopic rotator cuff repair (RCR) with DAA augmentation using a box-configuration construct between January 2021 and October 2025 by 2 surgeons at a single institution. Pre-operative magnetic resonance imaging was evaluated to classify tear size, tissue quality, and tendon retraction. Clinical outcomes assessed included post-operative pain, range of motion, retear rate, and complications.ResultsEleven patients underwent RCR with DAA augmentation. Mean age was 59.5 years (range, 51–72) with a mean body mass index of 27.6. Seven repairs (63.6%) were revision procedures. Ten tears (90.9%) were full thickness, with a median of 2 tendons involved and a median Patte retraction grade of 2. Mean follow-up was 7.2 months. Post-operatively, ROM remained excellent in 5 patients (45.5%) and improved in 6 patients (54.5%). Pain improved completely in 9 patients (81.8%) and partially in 2 (18.2%). No post-operative infections were observed or reoperations.ConclusionArthroscopic RCR with DAA augmentation using a box-configuration canopy construct demonstrated favorable early clinical outcomes and a low complication rate in a cohort with a high proportion of revision repairs. Larger prospective studies with longer follow-up are needed to further evaluate the durability and clinical benefits of this technique.Keywords: Rotator cuff repair, Dermal allograft augmentation, Rotator cuff retear, Revision rotator cuff repair, Transosseous-equivalent repair, Acellular dermal matrix, Biologic augmentation, Arthroscopic shoulder surgery Rotator cuff tears are among the most common upper extremity injuries, with approximately 250,000 repairs performed annually in the United States and a 141% increase in surgical volume from 1996 to 2006.3,18 Untreated tears can lead to progressive shoulder dysfunction, pain, tear propagation, and rotator cuff arthropathy. Arthroscopic rotator cuff repair (RCR) has become the gold standard for surgical management, demonstrating improved early post-operative outcomes compared to open or mini-open techniques.10 Both single-row and double-row repairs yield comparable clinical outcomes, though double-row constructs demonstrate lower retear rates in revision settings.18 Despite technical advances, primary RCR retear rates remain high (10%-94%), with patient age, tear size, tendon retraction, and medical comorbidities identified as independent risk factors.21 Revision RCR demonstrates comparable retear rates of 13.3%-54%.5,20 These findings underscore the need for techniques that address these limitations. RCR failure occurs at 1 of 3 interfaces: suture-tendon, bone-tendon, or bone-anchor.15 Augmentation strategies targeting the bone-tendon interface have been extensively investigated, including xenograft, autograft, synthetic, and allograft patches.12,16 Dermal allograft augmentation (DAA) augmentation has garnered increasing interest due to improved clinical outcomes and cadaveric biomechanical studies. In a matched-pair cadaveric model, Barber et al1 demonstrated that DAA-augmented supraspinatus repairs failed at significantly higher loads than non-augmented repairs (325 ± 74 N vs. 273 ± 116 N, P = .047), representing an approximate 19% increase in ultimate failure strength.6,7,11 The technique described herein combines a double row transosseous-equivalent RCR with a DAA using a box-configuration canopy construct. This combination maximizes footprint coverage while creating tendon-to-bone compression over the greater tuberosity on top of the underlying RCR. Cadaveric studies have demonstrated that transosseous-equivalent techniques achieve higher ultimate loads than single or double-row constructs,14 and DAA-augmented repairs exhibit significantly greater failure strength than non-augmented repairs, with Barber et al1 reporting a mean ultimate failure load of 325 N for augmented repairs compared with 273 N for controls. The combination of these approaches may confer an additive biomechanical benefit. This study describes a novel technique for RCR in patients at high risk for retear and reports early clinical outcomes from a retrospective cohort. It is hypothesized that this technique will reliably restore shoulder function while achieving low retear rates. Materials and methods Retrospective analysis was performed on patients who underwent arthroscopic RCR with acellular human dermal matrix augmentation. The associated Current Procedural Terminology code for arthroscopic RCR was used to find all cases performed between January 2021 and October 2025 at a single institution. This preliminary list was then analyzed to find all cases that utilized the acellular human dermal matrix augmentation with the described technique, which can be found below. The final cohort represented patients from 2 orthopedic surgeons with over 30 years of combined experience with all cases using an independent dermal allograft (Arthrex, FL, USA). This technique did not incorporate additional biologic augmentation (eg, PRP, bone marrow aspirate concentrate). Patients were excluded from analysis if they did not receive DAA, if the described box configuration was not utilized, or if they did not have at least 3 months of clinical follow-up. Pre-operative magnetic resonance imagings (MRIs) were analyzed (Fig. 1, A and B) to classify tear size, which was based on the greatest dimension according to Cofield,2 independent of the number of tendons involved. Tendon retraction was classified using the Patte classification.13 The outcome measures assessed were status of the repair, reoperations, post-operative range of motion (ROM), and post-operative pain. High-risk features were defined a priori as any combination of: revision repair, full-thickness tear with ≥2 tendons involved, Patte retraction grade ≥2, tear size ≥1.5 cm, or poor tissue quality noted intraoperatively. This technique was used at the surgeon's discretion when patients presented with multiple concurrent risk factors. The heterogeneity of clinical documentation precluded quantitative analysis; therefore, outcomes were limited to qualitative assessment of patient improvement in ROM and pain. (a) Coronal slice of pre-operative MRI of the shoulder demonstrating a full thickness rotator cuff tear with stage II retraction and degenerative changes to greater tuberosity footprint. (b) Sagittal slice of pre-operative MRI of the shoulder showing significant atrophy of the supraspinatus muscle. MRI, magnetic resonance imaging. The primary objective is to describe a novel fixation technique for rotator cuff allograft augmentation with a secondary objective to report associated short-term clinical outcomes. This technique was used at the surgeon's discretion for patients with multiple risk factors for repair failure. Surgical technique Either beach chair or lateral positions were utilized according to surgeon preference. Standard shoulder arthroscopy portals are used including posterior, rotator interval, anterolateral, lateral, and posterolateral portals. The posterior portal is used first to assess the glenohumeral joint along with any other pathologies that need to be addressed. The rotator interval portal is then established, and any synovitis, biceps anchor pathology, and/or labral pathology is addressed. If needed, this portal is used to perform subscapularis repair. Next, a lateral portal is made within the subacromial space where using a combination of electrocautery, shaver, and burr, the greater tuberosity is débrided to elicit a healing response. The subacromial space was also decompressed as needed. At this stage, the edge of the cuff tendons are débrided back to healthy tissue. Next, a posterolateral portal is made for a viewing portal through which the rotator cuff tendon tear morphology is assessed, and the camera is then placed in this portal for the remainder of the case. Figure 2 is a demonstration of the various working portals throughout the case. Intraoperative photo of the right shoulder showing the working portals including: rotator interval, anterolateral, direct lateral, posterolateral, and posterior. Cuff repair After the cuff tear complexity is investigated (Fig. 3), an attempt to mobilize the tear is performed using a tagging stitch in the posterior superior rotator cuff. Any adhesions medially are released until adequate tendon excursion is achieved. A passport is placed laterally and anterolaterally for suture management. Two medial row, SwiveLock suture anchors loaded with a knotless mechanism as well as fiber tape are placed via percutaneous portals off the lateral edge of the acromion. A scorpion suture passer is used to pass the tape as well as the knotless mechanisms through the torn cuff tendons and the knotless mechanism is saved for later use. The tapes of the 2 anchors medially are split and then dunked into 2 lateral anchors for a knotless speed bridge configuration which reduces the tendon back down to the greater tuberosity footprint (Fig. 4). Intraoperative arthroscopic photo demonstrating the full-thickness rotator cuff tear with retraction to the humeral head, prior to repair. Intraoperative arthroscopic photo demonstrating the repaired rotator cuff with double row configuration. Patch augmentation The distance between the 4 anchors is measured using an arthroscopic ruler to size the graft appropriately. The graft is then measured and cut to the appropriate size on the back table leaving an extra 5 mm around the periphery of the anchors. The knotless mechanisms from all 4 anchors are then brought out the lateral passport with care not to entangle them (Fig. 5). The knotless mechanisms are then passed through into the 4 respective corners of the graft (Fig. 6). The graft is then shuttled into the subacromial space through the lateral portal. The repair stitch from the posteromedial anchor is then shuttled into the anteromedial anchor to create a suture compressive “staple” across the medial row. This process is then repeated from the anteromedial anchor to anterolateral anchor, continued from anterolateral to posterolateral, and finally completed with the repair stitch from posterolateral to posteromedial anchor. This is done sequentially between all 4 anchors so that a "box "type of compression construct is created (Fig. 7). The shoulder is then gently ranged, assessing the stability of the repair and augmentation. Intraoperative photo demonstrating the 4 knotless mechanisms exiting through the lateral portal, each in their respective quadrant, before incorporation of the DAA. DAA, acellular dermal allograft. Intraoperative photo showing the DAA with the 4 knotless mechanisms shuttled through the 4 corners before the graft is transferred through the lateral portal and into the subacromial space. DAA, dermal allograft augmentation. Intraoperative arthroscopic image demonstrating the DAA laying on Top of the rotator cuff repair using the box configuration. DAA, dermal allograft augmentation. Post-operative protocol Once complete, sterile soft dressings and a simple sling were applied. The patient then began pendulum exercises only for the first 6 weeks. After 6 weeks, formal physical therapy was started working on active and passive ROM. Strengthening was generally started at 12 weeks depending on the patient's progress with physical therapy. Results A total of 11 patients underwent arthroscopic RCR with acellular DAA using the box configuration (Table I). The mean age at surgery was 59.5 year old with a range of 51-72. There were 8 males (72.7%) and 3 females (27.3%). Mean body mass index was 27.6, with a range of 20.7–40. One patient (9.1%) was an active smoker, 1 (9.1%) had diabetes mellitus, 2 (18.2%) had hypercholesterolemia, and 3 (27.3%) had heart disease. Table IPatient demographics.Mean age (yr)59.5Sex, n (%)Male: 8 (72.7%) female 3 (27.3%)Mean BMI (kg/m2)27.6Smoker, n (%)1 (9.1%)Diabetes mellitus, n (%)1 (9.1%)Hypercholesterolemia, n (%)2 (18.1%)Heart disease, n (%)3 (27.2%)BMI, body mass index. Patient demographics. BMI, body mass index. Seven patients (63.6%) underwent revision RCR, while 4 patients (36.4%) underwent primary repair. The mechanism of injury was classified as traumatic in 4 patients (36.4%) and atraumatic in 7 patients (63.6%). Of the 7 revision repairs, 2 of them were due to traumatic retears. At the time of surgery, 10 tears (90.9%) were full-thickness and 1 (9.1%) was partial-thickness. The median number of tendons involved was 2 with a range of 1-3. Tear size, as measured on pre-operative MRI, varied widely (Table II). Among the 10 full-thickness tears, the average tear size in the anteroposterior direction was 1.72 cm with a range of 0.7-2.9 cm. The average tear size in the medial to lateral direction was 1.54 cm with a range of 0.7-2.7 cm. These tear sizes corresponded to Cofield grades of 9 medium tears and 1 small tear with the average tear size being classified as medium. There was 1 partial thickness tear, which had previously been operated on and measured 0.4 × 0.4 cm. The median Patte classification for tear retraction was 2. Table IIRaw rotator cuff tear characteristics.PatientRevision RCRMechanismFull vs. partial thicknessTear size (cm)Patte classificationNumber of tendons involvedRetear1YesDegenerativepartial0.4 × 0.412No2YesDegenerativefull0.8 × 0.711No3YesDegenerativefull2.4 × 2.733No4YesDegenerativefull1.421No5YesTraumaticfull0.7 × 1.011No6YesTraumaticfull221No7YesDegenerativefull1.322No8NoDegenerativefull2.9 × 1.933No9NoTraumaticfull2.4 × 2.322Yes10NoDegenerativefull1.3 × 1.122No11NoTraumaticfull2 × 1.122NoRCR, rotator cuff repair. Raw rotator cuff tear characteristics. RCR, rotator cuff repair. The mean time from diagnosis to surgery was 106.9 days, with a range of 16–253 days. The mean duration of clinical follow-up was 7.2 months, with a range of 3-19.1 months. At final follow-up, 9 patients (81.8%) had been advanced to as-needed follow-up status with 1 (9.1%) being advanced to annual follow-up, indicating satisfactory clinical progress. There were 5/11 (45.5%) patients that had excellent ROM both pre-operatively and post-operatively. Six out of eleven (54.5%) demonstrated improved ROM post-operatively. Nine out of eleven (81.8%) had complete resolution of their pain, and 2/11 (18.2%) had partial improvement with some residual pain. Post-operative imaging (MRI or ultrasound) was not performed systematically in this cohort. Repair integrity was assessed only when clinically indicated by persistent pain, weakness, or loss of function. One patient (9.1%) experienced rotator cuff retear which was identified via MRI, which was obtained due to persistent pain. This patient was a primary RCR and the retear was discovered to have occurred atraumatically during follow-up. The retear was small and did not greatly impact the functionality of the patient's shoulder therefore non-operative management was pursued. There were no post-operative infections or reoperations reported. Discussion Rotator cuff retears can pose a challenge for primary and revision repairs, specifically in the setting of larger tears, tendon retraction, and those with poor tissue quality.21 Despite advances in anchor technology and repair constructs, reported retear rates following primary RCR remain highly variable, and outcomes following revision repair are consistently inferior.17 Because of this, using augmentation in order to improve the biologic and mechanical environment at the bone-tendon interface has gained increasing attention. DAA augmentation has gained attraction given its ability to provide immediate mechanical reinforcement while serving as a scaffold for host tissue incorporation, with several studies showing a statistically significant decrease in retear rates when compared to repairs without augmentation.4,19 In this study we present an arthroscopic RCR technique using DAA in what we have named the box-configuration and demonstrated that this technique results in promising early clinical outcomes in a collective of patients at elevated risk for repair failure. We highlight the potential additive biomechanical benefit of DAA combined with construct optimization via transosseous-equivalent repair. In previous literature, transosseous-equivalent repairs have demonstrated superior biomechanical properties compared with single-row constructs and have been associated with improved footprint restoration.8 Our box configuration employs both transosseous-equivalent repair and allograft augmentation which we propose increases footprint contact area, improves tendon-to-bone compression, and enhances load to failure which may ultimately lead to improved clinical outcomes. The clinical outcomes observed in our cohort support this proposition. With a mean follow-up of 7.2 months, over half of our patients, 6/11 (54.5%), demonstrated improved post-operative ROM with the remaining 5/11 (45.5%) patients maintaining excellent ROM both pre-operatively and post-operatively. The vast majority, 9/11 (81.8%), reported complete improvement in pain, despite a high prevalence of this technique being applied in the revision surgery setting, 7/11 (63.6%), and medium tear sizes with Patte grade II retraction.13 One diagnosed retear did occur, which occurred atraumatically during follow-up; however, given the small retear size and maintained shoulder functionality, non-operative management was continued. These findings support that this technique may provide a reproducible, biomechanically robust, and clinically significant surgical intervention for at-risk rotator cuff tears. Prior studies evaluating DAA support our findings and have demonstrated improved structural integrity compared with repair alone.4,9,19 Although factors such as tear types, tendon quality, surgical technique, and clinical outcome measures may increase variability in these findings, this study nevertheless adds to a growing body of evidence supporting the use of DAA in at-risk RCRs. Most other methods of dermal allograft use involve separate fixation strategies to incorporate the dermal allograft. This new “box” technique is unique as it utilizes anchors already placed during initial RCR, thus not requiring any additional anchors or devices. This helps to limit operative costs. Furthermore, the fixation of this graft has 360 degrees of compression around the repair. The authors theorize that this has substantial benefits to healing such as providing a barrier preventing synovial fluid from interfering with tendon healing. This also likely creates stronger fixation than medial soft tissue fixation strategies. With the described technique the dermal allograft graft is anchored to the bone versus medial soft staples or all suture devices that have a higher chance of failing by pulling out of the soft tissue. This is especially true with revision tears that have little tendon substance remaining. Our study does not come without limitations. First, our study is retrospective in nature with a small cohort of only 11 patients, limiting the ability to say with certainty this is the most efficacious technique. Second, our mean follow-up of 7.2 months falls below the conventional 1-year threshold for short-term outcomes reporting, limiting our ability to draw conclusions about mid- or long-term repair durability. Third, the heterogeneity in clinic documentation limits our ability to quantitatively analyze our patient's post-operative clinical outcome for strength and ROM. Furthermore, repair integrity was not systematically assessed via imaging in asymptomatic patients, and therefore the true retear rate in this cohort cannot be determined. Lastly, our study utilized only 2 surgeons operating at 1 facility, thus limiting generalizability. However, given these limitations, our study does describe an in-depth, reproducible technique and provides early clinical data for surgical management of a challenging at-risk patient population. Conclusion Arthroscopic RCR augmented with acellular dermal allograft using a box-configuration transosseous-equivalent construct demonstrated encouraging early clinical outcomes and a low rate of clinically apparent complications in a cohort of patients at elevated risk for repair failure. This technique may represent a valuable option for improving repair durability in high-risk rotator cuff tears. Going forward, future studies could include larger prospective cohorts with standardized functional outcome measures as well as longer-term follow-up to determine the durability of this technique and its impact on retear rates compared with other augmentation strategies versus repair without augmentation. Additionally, given emerging evidence supporting biologic adjuncts in tendon healing, future iterations of this technique could evaluate combined biologic and structural augmentation strategies. Disclaimers: Funding: No funding was disclosed by the authors. Conflicts of interest: Bryan Brown is a paid consultant and speaker for Arthrex. However, he received no financial reward for using this technique or writing this article. Any additional 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. Footnotes This study was approved by the institutional review board at Orlando Health (Reference number 25.300.10). References