This retrospective comparative study included patients who underwent arthroscopic RC repair with dermal allograft augmentation between February 2015 and April 2022. Of the 48 eligible patients with at least 24 months of follow-up, 42 (88%) were assessed at a median of 32.4 months (range: 24–49 months). Twenty patients (48%) had a primary repair (primary group), and 22 (52%) had revision procedures due to RC healing failure (revision group). Graft augmentation was used for RC re-tears, primary repairs involving two or three tendons (supraspinatus, infraspinatus, subscapularis), or degenerative single-tendon tears with poor tissue quality. Patients with irreparable tears, arthritis Hamada[19] grade > 2, and RC fatty infiltration Goutallier[20] > 1 were excluded.
A power analysis targeted a 6-point difference in the Constant-Murley score (CS)[21] between groups, requiring at least 40 subjects (20 per group), based on a standard deviation of 7 points and a clinically important difference of 1.7 points.
A control group of 26 consecutive patients who underwent arthroscopic RC repair without patch augmentation was matched for age, sex, body mass index (BMI), and RC injury type. Data for the control group were extracted from clinical and radiographic records. Patient selection followed Pocock’s criteria to avoid bias[22]. The study was performed in line with the principles of the Declaration of Helsinki and was approved by the local ethics committee (protocol no. cufclin.id.sirer.7340). Informed consent was obtained from all individual participants included in the study.
The demographic and anthropometric data of the 3 groups are described in Table 1.
Table 1
Demographic data of the study population
| Variable | Primary group (n = 22) | Revision group (n = 20) | Control group (n = 26) | P value |
| Age at the time of surgery, y, mean ± SD | 58 ± 1.7 | 52 ± 2.7 | 55 1.9 | 0.051 |
| Gender, male/female, n, (%) | 9/13(43/57) | 8/12(40/60) | 10/16(39/61) | 0.196 |
| Body mass index, mean | 26.9 | 27.7 | 25.2 | 0.862 |
| Follow-up duration, mo, median (range) | 33.7 (26–49) | 29.5 (24–42) | 34.6 (26–51) | 0.156 |
| Dominance in the operated shoulder, n, (%) | 15 (68.2) | 11 (54) | 17 (66) | 0.857 |
| Isolated supraspinatus tear, n, (%) | 11 (50) | 10 (50) | 14 (54) | 0.178 |
| Massive rotator cuff tear, n, (%) | | | | 0.261 |
| A | 1 (9 ) | 1 (10) | 2 (17) | |
| B | 0 | 0 | 0 | |
| C | 1 (9) | 1 (10) | 1 (8) | |
| D | 9 (82) | 8 (80) | 9 (75) | |
| E | 0 | 0 | 0 | |
| SD: standard deviation |
Preoperative Imaging
Standard radiographs (true anterior-posterior Grashey, outlet, and axillary views) were used to assess bone morphology and the gleno-humeral joint line. US was the first-line imaging technique for RC injuries, performed by an experienced orthopedic surgeon and musculoskeletal radiologist using a GE Logiq 7 machine with a 7.5- to 14-MHz linear transducer. US images included transverse and longitudinal planes of the supraspinatus, subscapularis, infraspinatus, and biceps brachii tendons[23].
The supraspinatus tendon was evaluated in its long and short axes for abnormalities such as non-visualization or abnormal echogenicity[24]. Focal tendon abnormalities were classified as intrasubstance, articular, bursal, or full-thickness tears, according to Snyder criteria[25]. The infraspinatus tendon was assessed using both transverse and longitudinal views, as described in a previous study[26]; massive RC tears were classified by tendon components per Collin et al[27].
The subscapularis tendon was evaluated along its axes during passive external and internal rotation, with full tendon width demonstrated. Infraspinatus and subscapularis tendon tears were classified as partial or full-thickness. Tendon repair integrity in the revision group was categorized into five types, following Barth et al.'s classification using frontal, sagittal, and transverse US images[28, 29].
Magnetic resonance imaging (MRI) was performed using a 1.5 T unit, with imaging in oblique coronal, parasagittal, and axial planes. RC tears were diagnosed when MRI showed cuff discontinuity, irregularity, and increased signal[30–32]. Re-tears were graded using MRI criteria from Sugaya et al[29], similar to US evaluation methods. MRI identified intact RCs based on uniform low signal intensity in T1 images and absence of increased signal in T2 images
Histological and biomechanical properties of dermal allograft
The graft used in this study was an allogenic human-derived dermis (HDM), obtained from the backs of multi-organ and/or multi-tissue donors, and decellularized using a chemico-physical method as described by Bondioli et al[33]. Mechanical experiments assessing tensile failure demonstrated that the maximum load, tensile strength, and stiffness of the decellularized dermis were significantly higher than those of cellularized dermis. Biological tests confirmed a high concentration of transforming growth factor and increased proliferation of fibroblasts in HDM compared to the cellularized control. The decellularized dermis retained its collagenous and fibrous architectural structure unchanged[33].
The thickness of the decellularized dermis ranged between 0.8 and 2 mm. The plots varied in size, ranging from 4 to 12 cm in length and 3 to 7 cm in width.
Surgical Procedure
Arthroscopic RC repair was performed by three experienced shoulder surgeons with patients in the lateral decubitus position and 5 kg of traction. Three portals (anterior, lateral, and posterior) were used. The long head of the biceps was examined for tears and instability, and subscapularis lesions were assessed. The supraspinatus insertion was explored for complete tears or re-tears (revision group). The arthroscope was then moved to the subacromial space to expose the supraspinatus and infraspinatus footprint. RC tears were classified as isolated supraspinatus tears[25] or massive tears involving two or more tendons[27].
Supraspinatus and infraspinatus tears were reduced to the greater tuberosity and fixed two triple-loaded suture anchors medially. Three mattress sutures and three single sutures were used to repair the tendon. The single sutures were cut, leaving the mattress sutures to secure the graft and reinforce the repair as a "patch augmentation technique."
Two techniques were used: all-arthroscopic or mini-open, depending on the surgeon’s preference. In the all-arthroscopic technique, the graft was prepared with free HiFi sutures and introduced into the subacromial space using the “sliding technique.” The graft was fixed using two knotless anchors laterally and secured with a double-row suture bridge technique.
Postoperative rehabilitation included immobilization in a sling for 3 weeks. Self-assisted pendulum exercises started the first day, with passive mobilization from week 3 under therapist supervision. By the 5th week, active range-of-motion exercises began in a water pool. Strengthening exercises for scapular and humeral muscles started in the 3rd month, focusing on restoring scapular-humeral rhythm. Return to heavy work or sports was allowed between 3 and 6 months.
Outcome Measures and Rotator Cuff Strength Assessment
Outcome Measures and Rotator Cuff Strength Assessment
Active shoulder range of motion (ROM) was measured using a goniometer, assessing active anterior and lateral elevation (AAE and ALE), internal rotation (IR), and external rotation (ER) with the arm at the side. Internal rotation was rated on a 0–10 scale based on the ability to reach T7 in the Appley scratch test. Supraspinatus and infraspinatus strength were evaluated using established methods[34, 35]. Supraspinatus strength was assessed with the Empty Can (EC) and Scapula Retraction Test (SRT)[36], while infraspinatus strength was measured by resisting maximum ER force in the infraspinatus strength test (IST). The infraspinatus scapular retraction test (ISRT) was also performed to assess scapular contribution to strength.
Maximum voluntary isometric contraction was recorded using a handheld dynamometer, and the average of three assessments was used for strength evaluation. Results from the affected shoulder were compared with those from the control group.
Clinical outcomes were measured using the Simple Shoulder Test (SST)[37], Subjective Shoulder Value (SSV)[38], and CS[21]. ROM and muscle strength assessments were conducted by two independent examiners who did not participate in the surgeries. Intra-rater and inter-rater agreement were calculated according to Merolla et al.'s methodology[39].
Postoperative Ultrasound Evaluation
Postoperative US assessment of the repaired RC tendons was performed by the two raters involved in the preoperative imaging evaluation, using the same technical steps. Tendon integrity after RC repair showed a homogeneously hyperechoic and fibrillar appearance. Tendon thickness (mm) was measured and recorded. A well-apposed contact between the restored tendon and the bone surface confirmed the integrity of the repair. The dermal graft appeared as an echogenic linear structure. The presence of sutures or anchors appeared as hyperechoic structures either inside or near the healed tendon. Abnormal echogenicity and thickness and disorganization of tendon fibers were found in subjects with failed healing. Complete tears were defined as an absence of continuous tendon fibers visualized over the humeral head and attaching to the greater tuberosity. Types I to III images were recorded as indicative of RC repair integrity. Full-thickness re-tears were described as minor (Type IV) and major discontinuities (Type V)[28] (Fig. 1 and Fig. 2).
US findings of graft failure included laxity due to graft detachment and hypoechoic fluid at the site of tears in the adjacent tendon. US images, both static and dynamic, were independently assessed by the two raters, and any differences were resolved through a consensus meeting.
Statistical analysis
Descriptive statistics (absolute and percentage frequency, mean, median, SD, and range) for each group were calculated for all variables. Delta scores were calculated for CS as the difference between postoperative and preoperative values. The groups were compared using the Chi-square test and Mann-Whitney test according to the type of variable. The Kruskall-Wallis test was used to investigate the relationship between preoperative RC tear variable and postoperative variables (strength and CS).
The possible association between each group and the qualitative variables, either baseline or postoperative, were analyzed with 2 Pearson χ test. The level of significance was set at 0.05.