INTRODUCTION
Rotator cuff tear (RCT) is one of the most common shoulder disorders and often is associated with pain, muscle weakness, and functional limitation [1,2]. Surgical repair is generally indicated when conservative treatment fails. However, primary repair is frequently unfeasible in large or massive tears with advanced tendon retraction, poor tissue quality, or severe fatty infiltration [3,4]. Even when rotator cuff repair is technically successful, structural failure remains a concern. Previous studies have reported retear rates ranging from 0% to 94% following primary repair [1-6]. For medium tears, the retear rate was reported to range from 8.5% to 40% [7-10].
Acellular dermal matrix (ADM) patch augmentation has been introduced to enhance tendon healing and reinforce structural integrity after rotator cuff repair [6,9-12]. Although widely used for large and massive tears, its application has expanded to include smaller tears with poor tendon quality or compromised healing potential [4]. Despite the potential benefits, some studies have raised concerns regarding postoperative stiffness [13-18].
Due to these considerations, establishing appropriate indications for patch augmentation remains critical for optimizing outcomes. However, previous studies have largely focused on outcomes within specific tear sizes or lacked direct comparisons between medium and large-to-massive tears. This study aimed to evaluate the clinical and radiological outcomes of arthroscopic rotator cuff repair with ADM patch augmentation and to compare these outcomes between medium and large-to-massive tears.
METHODS
The study was approved by the Institutional Review Board of Inje University Busan Paik Hospital (No. 2025-06-018), and the requirement for informed consent was waived due to the retrospective nature of the study.
Patient Selection
This retrospective multicenter study included consecutive patients who underwent arthroscopic RCT with ADM patch augmentation at two tertiary hospitals between April 2021 and April 2023. A total of 79 patients was initially reviewed. Patients were included if they underwent at least 24 months of clinical follow-up and had undergone postoperative magnetic resonance imaging (MRI) at least 6 months postoperatively. Exclusion criteria were revision rotator cuff repair, previous fracture on the ipsilateral side, and concomitant osteoarthritic changes. Tear size was determined based on the mediolateral (ML) dimension on T2-weighted coronal oblique MRI. Patients were classified based on tear size as ≤30 mm (group 1) and >30 mm (group 2). Intraoperative assessment of tear configuration and tendon mobility guided the decision for patch augmentation in both groups.
Surgical Procedure
All patients were placed in the beach chair position under general anesthesia combined with a brachial plexus block. A standard posterior portal was established for initial diagnostic arthroscopy of the glenohumeral joint, and concomitant intra-articular pathologies were evaluated. For biceps tendon lesions, <50% partial tears were managed with simple debridement. In cases with >50% partial tear, tenotomy was performed in low-demand or elderly patients, and tenodesis was performed in younger or more active individuals. Subscapularis tears were classified according to the Yoo and Rhee classification: ≤type 2A tears were treated with simple debridement, whereas ≥type 2B tears were repaired in intra-articular space using a single suture anchor. Subsequently, an anterolateral portal was established to access the subacromial space. Acromioplasty was performed selectively when needed. Preparation of the footprint and greater tuberosity and evaluation of tear size, tissue quality, and tendon mobility were performed. Two or three medial-row suture anchors (each loaded with three suture pairs) were placed at the articular margin. The suture strands were passed through the rotator cuff tendon using a penetrating suture passer and tied.
When arthroscopic evaluation revealed poor tendon quality including thinning, fraying, and reduced elasticity or incomplete footprint coverage without exposure of the articular cartilage following primary repair, ADM patch augmentation was performed (BellaCell HD; Hans Biomed). Graft size was determined intraoperatively using a calibrated probe. The ML dimension was defined as the distance from the medial suture anchors to the lateral border of the footprint, and the anteroposterior (AP) dimension was measured between the most anterior and posterior tied sutures. The patch was trimmed to match the measured ML length and exceed the AP width by approximately 2–3 mm on each margin, allowing stable coverage and secure lateral fixation without excessive graft tension.
The medial margin of the patch was marked at evenly spaced points to guide suture placement. Using a spinal needle as a guide, nonabsorbable sutures were passed through the medial quadrants of the patch (Fig. 1). The patch was then carefully introduced into the subacromial space via the anterolateral portal. After confirming proper positioning and orientation under arthroscopic visualization, the most anterior and posterior suture limbs were tied, and the central limbs were left untied. Each of the suture limbs were passed through two lateral row knotless anchors (SwiveLock; Arthrex). A suture bridge technique was used to secure the graft-incorporated repair (Fig. 2).
Rehabilitation consisted of shoulder immobility with an abduction sling for the initial 6 weeks. Passive and active-assisted motion was started at 6 weeks postoperatively. Isometric and resistive shoulder strengthening was started at 3 months postoperatively. A single senior surgeon at each institution performed all procedures using a consistent surgical technique and postoperative rehabilitation protocol.
Clinical Outcomes
Orthopedic surgeons not involved in the surgical procedures performed preoperative and final postoperative clinical assessments. Pain at rest was assessed using the visual analog scale (VAS), and functional outcomes were evaluated using the American Shoulder and Elbow Surgeons (ASES) score, Constant-Murley score, and University of California, Los Angeles (UCLA) score. Active range of motion (ROM) including forward flexion (FF), scaption, external rotation (ER) at the side, ER at 90° abduction, and internal rotation (IR) behind the back was measured using a standard goniometer with the patient in a seated position. IR was quantified by measuring the vertical distance (mm) from the C7 spinous process to the level of the patient’s thumb. Postoperative stiffness was defined using established passive ROM criteria (FF<120°, ER<30°, and IR lower than L3) assessed at the final follow-up.
Radiological Analysis
Fatty degeneration of the rotator cuff muscles was assessed using the Goutallier classification: grades 0–1 were defined as mild, grade 2 as moderate, and grades 3–4 as severe fatty infiltration [19]. Postoperative tendon integrity was evaluated based on MRI at 6 months postoperatively. Retear was diagnosed when a fluid-equivalent signal was found or when the rotator cuff tendon was not visualized on more than one T2-weighted image (Figs. 3 and 4). Integrity was classified according to the Sugaya classification. In addition, radiologic outcomes were assessed by independent observers blinded to group allocation to minimize the risk of observational bias.
Statistical Analysis
A specialized statistician performed statistical evaluation using IBM SPSS version 25.0 (IBM Corp.). The normality of continuous variables was assessed using the Shapiro-Wilk test. Preoperative ASES, Constant-Murley, and UCLA scores as well as ER (side) values were the only ones to show a normal distribution (P>0.05). All continuous variables were presented as mean±standard deviation. Changes in preoperative and postoperative scores within each group were analyzed using paired t-tests or Wilcoxon signed-rank tests. Intergroup comparisons of clinical outcomes were performed using either independent t-tests or Mann-Whitney U-tests depending on distribution. Postoperative tendon integrity (Sugaya classification) and stiffness incidence were compared between groups using Fisher's exact test. The level of statistical significance was set at P<0.05.
RESULTS
Patient Demographics
A total of 49 patients were classified into two groups based on tear size: 21 patients in group 1 and 28 patients in group 2. The mean age at surgery was 61.48±7.62 years (range, 45–73 years) in group 1 and 66.25±6.44 years (range, 51–75 years) in group 2. Sex distribution was similar between the groups (male:female=11:10 [52.4%:47.6%] in group 1 and 17:11 [60.7%:39.3%] in group 2). The mean follow-up duration was 27.24±4.36 months in group 1 and 29.39±5.09 months in group 2. The mean AP tear size was 23.10±2.00 mm in group 1 and 33.50±6.80 mm in group 2. The mean ML tear size was 23.60±2.20 mm in group 1 and 34.50±4.30 mm in group 2. Both size differences were significant (P<0.001 for each). Fatty degeneration of the rotator cuff in group 1 showed supraspinatus degeneration at a level of mild in 9.5%, moderate in 76.2%, and severe in 14.3% of patients. In group 2, supraspinatus degeneration was mild in 0.0%, moderate in 42.9%, and severe in 57.1% of patients. Infraspinatus degeneration in group 1 was mild in 90.5%, moderate in 9.5%, and severe in 0.0% of patients. In group 2, infraspinatus degeneration was mild in 46.4%, moderate in 35.7%, and severe in 17.9% of patients. The distribution of Goutallier grades for both the supraspinatus and infraspinatus differed significantly between the two groups (P=0.005 for each) (Table 1).
Clinical Outcomes
Intragroup comparisons showed significant improvements in all clinical outcomes from preoperative to final follow-up in both groups. In group 1, the VAS score improved from 4.10±0.30 to 0.81±0.60, ASES score from 61.40±5.41 to 90.66±5.75, Constant-Murley score from 54.00±11.38 to 81.24±5.45, and UCLA score from 18.05±3.56 to 30.71±3.78. In group 2, the VAS score improved from 4.50±0.51 to 1.21±0.83, ASES score from 58.14±4.07 to 87.47±7.17, Constant-Murley score from 52.82±7.28 to 80.54±3.72, and UCLA score from 17.61±2.78 to 29.39±3.42 (P<0.001 for each) (Table 2). Intergroup comparisons of the final scores showed no significant differences. Final VAS scores were 0.81±0.60 in group 1 and 1.21±0.83 in group 2, ASES scores were 90.66±5.75 and 87.47±7.17, Constant-Murley scores were 81.24±5.45 and 80.54±3.72, and UCLA scores were 30.71±3.78 and 29.39±3.42 (P=0.073, P=0.109, P=0.142, and P=0.160, respectively) (Table 3). No cases of immunologic reaction, foreign body response, or postoperative infection related to the ADM patch were observed in either group.
ROM and Stiffness
In group 1, FF improved from 134.05°±32.50° to 152.14°±14.54° and scaption improved from 132.62°±38.69° to 152.14°±14.54°; the differences were significant (P=0.006, and P=0.010, respectively). ER at the side improved from 31.90°±16.32° to 35.24°±9.42°, ER at 90° abduction improved from 58.81°±30.08° to 67.38°±25.67° and IR changed from 150.05±31.82 mm to 144.64±29.33 mm in group 1; the differences were not significant (P=0.311, P=0.146 and P=0.491, respectively). In group 2, FF changed from 136.79°±28.68° to 143.57°±18.35°, scaption from 136.79°±28.68° to 143.57°±18.35°, ER at the side changed from 25.36°±15.03° to 27.32°±8.97°, ER at 90° abduction from 31.07°±21.14° to 35.00°±21.60°, and IR from 161.34±32.09 mm to 161.27±19.82 mm; the differences were not significant (P=0.180, P=0.180, P=0.465, P=0.241, and P=0.990, respectively) (Table 4).
In intergroup comparisons of final ROM, FF was 152.14°±14.54° in group 1 and 143.57°±18.35° in group 2, and scaption was 152.14°±14.54° in group 1 and 143.57°±18.35° in group 2; the differences were not significant (P=0.094 and P=0.085, respectively). External rotation at the side was 35.24°±9.42° in group 1 and 27.32°±8.97° in group 2, ER at 90° abduction was 67.38°±25.67° in group 1 and 35.00°±21.60° in group 2, and IR was 144.64±29.30 mm in group 1 and 161.27±19.82 mm in group 2; the differences were significant (P=0.003, P=0.001, and P=0.023, respectively) (Table 5). Postoperative stiffness was observed in 1 patient (4.8%) in group 1 and 5 patients (17.9%) in group 2. However, this difference was not significant (P=0.204) (Table 6).
Radiological Outcomes
MRI performed at 6 months postoperatively revealed a retear in 8 of 49 patients (16.3%). Group 1 had 1 retear (4.8%) and group 2 had 7 retears (25.0%) but without significant difference (P=0.073) (Table 7).
DISCUSSION
In the present study, arthroscopic rotator cuff repair utilizing ADM patch augmentation resulted in clinically significant improvements in all functional outcome measures regardless of tear size. In previous studies, patch augmentation was suggested to improve clinical outcomes through both mechanical reinforcement and biological integration mechanisms. These mechanisms include reinforcement of the repair construct, increased footprint coverage, improved load sharing, and promotion of tendon healing via scaffold-mediated tissue integration [9,12]. Specifically, ADM patches may facilitate host cell infiltration, promote neovascularization, and support organized collagen matrix formation, thereby contributing to improved structural and clinical outcomes. In the present study, although both groups showed marked improvements in clinical outcomes after patch augmentation, no significant differences were observed between groups, indicating that the tear sizes achieved comparable levels of improvement following augmentation. Taken together, these findings suggest that the observed clinical improvements were due to ADM patch effects rather than tear size only, supporting its potential applicability across a range of tear severities.
Changes in postoperative ROM were generally limited in this study, particularly in patients with large-to-massive tears. Castle et al. [10] reported that, for small-to-medium tears, patients who underwent patch augmentation showed significantly greater improvement in active FF at 1-year follow-up compared with subjects treated without a patch. In contrast, for large-to-massive tears, no significant differences in any plane of motion were observed between patch and non-patch groups. These findings are consistent with the results of the present study. In particular, in the study cohort, postoperative stiffness at final follow-up was observed in both groups, although the intergroup difference was not significant (P=0.204); patients with larger tears who required larger patches tended to exhibit more frequent stiffness (4.8% in group 1 vs. 17.9% in group 2). Yeazell et al. [20] reported a postoperative stiffness rate of approximately 25% at 6 months following rotator cuff repair with patch augmentation, suggesting that an increased immunogenic or foreign body response to the graft may contribute to this outcome.
In previous studies, Cho et al. [21] reported retear rates of 22% in medium and 51.2% in large-to-massive tears after primary repair without patch augmentation. Jo et al. [22] reported retear rates of 20.6% for medium and 28.5% for large-to-massive tears. Castle et al. [10] reported a retear rate of 8.0% (4 of 50 patients) in small-to-medium tears and 11.9% (5 of 42 patients) in large-to-massive tears following rotator cuff repair with patch augmentation. The authors reported no definitive difference in retear rates based on tear size. Bushnell et al. [23] reported a prospective multicenter cohort study with a retear rate of 10.6% (7 of 66 patients) for medium and 28.6% (14 of 49 patients) for large tears at 2 years after rotator cuff repair with patch augmentation. The authors [23] emphasized that most radiographic failures occurred within the first 3 months postoperatively, before the implant could fully integrate biologically, emphasizing the importance of adequate primary fixation at the time of surgery. In the present study, retear rates were 4.8% (1 of 21 patients) in medium tears (group 1) and 25.0% (7 of 28 patients) in large-to-massive tears (group 2) at the 6-month follow-up; although this difference was not significant (P=0.073). Consistent with previous studies, the medium tear group in the present study showed a notably low retear rate of 4.8%, indicating favorable radiological outcomes following patch augmentation. In contrast, the retear rate in the large-to-massive tear group was 25.0%, which was comparable to or slightly higher than rates reported in the literature. However, direct comparisons across studies remain challenging due to heterogeneity in study design, follow-up duration, and definitions of structural failure. For example, Kim et al. [24] reported a retear rate of 24.1% at the 2-year follow-up in patients with large-to-massive RCTs, which was similar to the results of our study. However, their analysis included Sugaya type 3 as structural failure, which differed from our method and may affect interpretation of the retear rate.
Several limitations of this study should be acknowledged. First, the retrospective design and relatively small sample size may have limited the statistical power to detect significant differences in retear or stiffness rates between groups. These findings should be further investigated in larger, prospective cohorts to enhance clinical decision-making and clarify selection criteria. Second, as a multicenter study, variations in surgical technique between institutions may have influenced the outcomes despite standardized procedures. Third, the ADM patch used in this study represents only one of several commercially available products. Differences in graft composition, processing methods, and mechanical properties across manufacturers may limit the generalizability of our findings. Fourth, ROM and stiffness are influenced by various factors, including patient compliance and rehabilitation protocols, many of which could not be uniformly controlled in this study. Last, MRI follow-up was limited to 6 months, precluding long-term evaluation of tendon integrity and durability. Despite the limitations, this study offers meaningful insights into the application of ADM patch augmentation in rotator cuff repair across tear sizes. Direct comparison of outcomes between medium and large-to-massive tears using consistent surgical techniques and follow-up protocols provides results for practical guidance in clinical application.
The low retear rates suggest that ADM patch augmentation can be effective when appropriately indicated and tailored to tear characteristics. Furthermore, the observed trend toward higher stiffness, especially in larger tears, highlights the importance of patient selection, graft sizing, and tension control during surgery.
CONCLUSIONS
ADM patch augmentation after rotator cuff repair, compared with rotator cuff repair without patch augmentation, showed reduced retear rates and improved functional scores. However, consensus is lacking regarding optimal indication and graft sizing. The patch did not notably enhance postoperative ROM and was associated with stiffness, particularly in large tears requiring larger grafts. The success of ADM patch augmentation depends on patient selection and graft application.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Print
