Use of Graft Jacket as an Augmentation for Massive Rotator Cuff Tears

Indications/Contraindications 

Use of Graft Jacket augmentation with standard open rotator cuff repairs is indicated for massive rotator cuff tears, involving two or more tendons. Tear size, amount of retraction, and poor tendon tissue quality all play a role in the decision to use augmentation. Revision of a previous failed repair is also an indication for tissue augmentation. Tissue grafts are indicated to augment a repair and should not be used if the tendon is irreparable to substitute for stable repair. Use of tissue grafts is contraindicated in the case of active infection.

Preoperative Planning 

As with any repair technique for the rotator cuff, preoperative evaluation generally includes magnetic resonance imaging (MRI) or a magnetic resonance (MR) arthrogram. The degree of retraction and trophicity of the involved rotator cuff musculature should be determined before undertaking repair. In the case of revision surgery, preoperative workup should include a workup to rule out infection. Patients must be educated about the use of allograft tissue and informed consent must be obtained. The operating surgeon must be familiar with standard rotator cuff repair principles, including mobilization of the rotator cuff and appropriate management of the deltoid.

Surgical Technique 

We utilize general endotracheal anesthesia for rotator cuff repair. Patients are positioned in a modified beach chair position, with use of an articulated arm holder (McConnell Orthopaedics, Greenville, TX).

After the extremity is prepped and draped, we proceed with a diagnostic arthroscopic evaluation of the shoulder. This is important to visualize concurrent pathology such as biceps tendonosis, labral lesions, or glenohumeral arthritis. Concurrent pathology may be treated at this time. The amount of tendon retraction and mobility can also be assessed with an arthroscopic grasper. Tears that are determined to be irreparable may be treated with arthroscopic debridement at this time. At the completion of arthroscopy, the portals are closed and the shoulder is reprepped.

We proceed with a standard open approach to the rotator cuff as previously described. A very minimal acromioplasty (1-20 mm) is performed, with care taken to dissect the coracoacromial ligament from the remaining acromion for later repair. A thorough bursectomy is completed, exposing the underlying rotator cuff tear. Tear size, configuration, retraction, and tissue quality are noted. Tears are mobilized completely, including release of the coracohumeral ligament, scarring under the acromion, and juxtaglenoid adhesions. Massive tears with a significant side-to-side component “bouteneire” type deformity or an L or reverse-L component are closed with a classic McGlaughlin side-to-side, margin convergence repair using 1 mm cottony Dacron suture (Deknatel, Teleflex Medical, Research Park, NC), bringing the edge of the tear into close proximity to the greater tuberosity. As previously described, the tendon is then repaired to the sulcus and greater tuberosity footprint using the Transosseous Anchor Double Knot technique (TOAK).13

When the surgical procedure has reached the stage for final tendon-to-bone repair of the torn cuff, a very shallow (3 to 5 mm deep) but broad bony trough should be created from the sulcus to the tuberosity (Fig. 1A). The medial–lateral breadth of the trough should be approximately 10 mm. A deep trough into cancellous bone is not recommended, since the distal end of the tendon will “kink” and the footprint restoration will not be anatomic. The footprint is medialized to the extent that contact can be made with bone with native tendon so that a true augmentation is performed rather than a gap bridged. The lateral edge of the articular cartilage should be “beveled” to make the transition from cartilage to bone smooth and allow anatomic recreation of the cuff footprint (Fig. 1B).

View Large Image Download to PowerPoint

Figure 1. (A) Hand burr used to prepare greater tuberosity. (B) Final tuberosity preparation (3 to 5 mm below top of humeral head).

The senior author prefers the use of metallic suture anchors with double-loaded nonabsorbable sutures (5.0 mm Fastin RC with double-loaded OrthoCord suture; DePuy-Mitek, Boston, MA) or 5.0 mm corkscrew anchors with double loaded fiberwire (Arthrex, Naples, FL). Suture anchors, regardless of type, are inserted at an appropriate “deadman’s angle” into the medial edge of the prepared bony trough (just medial to the cartilage bone junction) (Fig. 2). We utilize one anchor per centimeter of cuff tear in the sagittal plane (anterior to posterior) dimension. All suture strands from each anchor are then passed in a horizontal mattress fashion (inferior to superior) through the cuff tendon at a distance of 1 cm from the distal tendon edge. Once all sutures are passed, the sutures are tied, securing the tendon into the trough with these medial mattress stitches. The ends of the sutures are NOT cut above the knots, since the sutures will be “reused” through the lateral portion of the tendon and used for anchoring sutures for the Graft Jacket (Fig. 3).

View Large Image Download to PowerPoint

Figure 2. Insertion of suture anchors at medial edge of prepared trough using “deadman’s angle.”

View Large Image Download to PowerPoint

Figure 3. All suture strands are passed through cuff 1 cm medial to tendon edge. These horizontal mattress sutures are tied but ends are not cut.

The greater tuberosity is now perforated for transosseous tunnels with any standard technique just lateral to the row of anchors and also on the lateral cortical margin of the tuberosity. We prefer to use either a small drill (1.8-2.0 mm) or a sharp awl. One limb from each of the suture pairs is now passed though the distal (lateral) portion of the torn tendon in a mattress fashion to begin a modified Mason–Allen stitch with two passes. That suture limb will now be on the superior portion of the tendon again after the two passes (Fig. 4A). The other suture limb is then passed from superior to inferior medial to the previous stitch—completing the Mason–Allen stitch (Fig. 4B). This limb will now be on the inferior portion of the tendon. This should be performed for one of the suture pairs from each anchor (Fig. 4C). The other suture pair from each anchor is left for use in anchoring the graft.

View Large Image Download to PowerPoint

Figure 4. (A) One suture limb (Limb A) from each pair of sutures is passed in a mattress fashion through the distal portion of the tendon. (B) The second limb (Limb B) from each pair is now passed medial to the mattress suture from Limb A—completing the modified Mason–Allen stitch. (C) Completed appearance of each pair of sutures after passing through distal portion of tendon.

The inferior limb from each suture pair is finally passed through the transosseous tunnels. A secure knot is tied on the lateral cortical margin of the greater tuberosity, and the cuff is securely restored to its anatomical footprint (Fig. 5).

View Large Image Download to PowerPoint

Figure 5. Inferior limb (Limb B) is now passed in a transosseous fashion. Knot is tied on lateral portion of greater tuberosity completing the TOAK technique.

Once the repair is completed, we proceed with tissue augmentation using a Graft Jacket. The graft is opened on the back table and hydrated for 10 min until supple (Fig. 6A). We have also used the patient’s autologous blood for hydration of the graft (Fig. 6B). While the graft is hydrating, we prepare the tendon for implantation.

View Large Image Download to PowerPoint

Figure 6. (A) Graft Jacket is hydrated for at least 10 min in normal saline. (B) The Graft Jacket may also be hydrated in a collection of the patient’s blood, with the possible benefit of autologous growth factors. (Color version of figure is available online.)

The previously tied sutures from the anchors are used in graft fixation. In addition, we place several anchoring sutures into the repaired tendon in the region that we plan to augment. For improved fixation of the graft laterally at the tuberosity, additional suture anchors may also be used as necessary. The anchoring sutures will be used to suture the graft securely overlying the repaired tendon (Fig. 7).

View Large Image Download to PowerPoint

Figure 7. Anchoring sutures in place around the circumference of the cuff repair before augmentation. (Color version of figure is available online.)

The Graft Jacket is placed over the repaired tendon with the shiny surface down (Fig. 8). The previously placed anchoring sutures are then passed through the graft around its circumference and tied securely, fixing the graft in position (Fig. 9).

View Large Image Download to PowerPoint

Figure 8. The graft is placed with the smooth, biologically active surface face down over the repair site. (Color version of figure is available online.)

View Large Image Download to PowerPoint

Figure 9. Graft Jacket augmentation securely sutured over the previous rotator cuff repair, creating a water tight surface. (Color version of figure is available online.)

Meticulous repair of the deltotrapezial fascia is performed, followed by routine wound closure. Patients are placed into an abduction pillow at the conclusion of the procedure.

Postoperative Care 

Postoperative immobilization consists of an abduction pillow 45° with no motion for 3 weeks, followed by passive motion above the pillow for an additional 3 weeks or conversion to an Ultra sling (Smith Nephew Don Joy, Carlsbad, CA) with gentle passive range of motion for 3 weeks depending on the degree of tension on the repair at the time of surgery. Active assisted range of motion is initiated at 6 weeks and slowly progressed. Strengthening is delayed until at least 12 weeks. In revision cases this is delayed 14 weeks.

Results 

Seventeen consecutive patients underwent open rotator cuff repair with Graft Jacket augmentation at our institution from May 2003 to June 2005 by a single surgeon using a previously described technique. There were 12 males and 5 females with an average age of 56.9 years. The dominant arm was involved in 9 patients (52.9%). Primary diagnosis was primary massive rotator cuff tear (11 patients; 64.7%) or recurrent massive rotator cuff tear (6 patients; 35.3%). Preoperative MRI scans were reviewed in all patients. Tears involved two, three, or four tendons and were all greater than 5 cm. Preoperative examination including range of motion, pain scores, function, and strength were documented before surgery. Concomitant procedures included distal clavicle excision (12 patients), biceps tenodesis (3 patients), repair of deltoid dehiscence (1 patient), and open reduction internal fixation os acromiale (1 patient). Patients were evaluated at regular intervals with clinical examination and subjective questionnaire. Eleven postoperative MRI scans have been obtained in and one computed tomography (CT) arthrogram performed on patients who were willing or needed to have their repair integrity evaluated.

Mean follow-up time was 1.2 years (438 days). Preoperatively, patients rated their pain as present all the time/unbearable (8 patients; 47%), present all the time/bearable (7 patients; 41%), or present during heavy or particular activity (2 patients; 12%). Postoperatively, pain improved with patients reporting no pain with normal activities (4 patients; 23%), occasional and slight pain (7 patients; 41%), present during heavy or particular activities (4 patients; 23%), or present all of the time/bearable (2 patients; 12%). Functional status improved with 10 patients (59%) unable to use their arm or able to use it for light activity only compared with 12 patients (70%) reporting normal function or only slight restriction postoperatively. Active forward flexion improved from 9 patients (53%) having <90° and 4 (24%) having 90 to 120° to 10 patients (59%) having 150° or better and 2 patients (12%) having 120 to 150°. Preoperative strength was grade 2 (poor) in 14 patients (83%) in forward flexion and 13 patients (77%) in external rotation. This improved to grade 4 (good) in 10 patients (59%) in both forward flexion and external rotation. Fourteen (82%) patients reported being satisfied by their outcome. UCLA scores improved significantly from a mean 9.06 preoperatively to 26.12 in the postoperative period (P value < 0.001).

Twelve shoulders were evaluated with advanced imaging studies. Eleven scans were done in a closed scanner 2.0 Tesla scanner without either IV or arthrographic enhancement. One patient with a pacemaker underwent CT arthrography. There were three recurrent tears of smaller size noted on the postoperative scans available for review to date. Some connection between the native tissue and the tuberosity via the graft was visible on all MRI in the postoperative period (Fig. 10). No change in muscle trophicity or fatty degeneration has been noted at this preliminary review. It appears that the graft does not increase in thickness at a 1.2 year average time frame on MRI scanning; however, the clinical thickness at the time of surgery on one look-back case exceeded that expected from the scan (Fig. 11). The look-back case showed a very quiet bursa (Fig. 12) and a robust water-tight cuff reconstitution.

View Large Image Download to PowerPoint

Figure 10. Postoperative MRI scan at 1 year follow-up. T2 coronal image shows an intact rotator cuff repair with tissue present all the way over the humeral head to the tuberosity.

View Large Image Download to PowerPoint

Figure 11. Extremely quiet fully reconstituted subacromial bursa with Graft Jacket. (Color version of figure is available online.)

View Large Image Download to PowerPoint

Figure 12. A 14-month postoperative rotator cuff showing reconstitued remodeled tendon. Histology revealed mature collagen and fibroblasts without chronic inflammatory reaction. (Color version of figure is available online.)

The thickness of the repaired tendon at surgery exceeded that of the thickness measured on MRI scanning. Both patients with preoperative anterior superior escape had retorn tendons at the time of MRI. One had persistent pain, was unsatisfied, and eventually underwent glenohumeral fusion. The another was satisfied having achieved 110° of elevation; however, she remained weak. The third patient was satisfied and had a UCLA score of 28.

Complications 

There were no other complications observed in this series. Specifically, we did not observe any infections or sterile inflammatory reactions in these patients. Subsequent to this series the senior author has had two cases with infections, neither one of which could be linked to the graft. The company was contacted and careful review of the lot numbers from which the grafts were taken revealed no reported infections in other hosts who received tissue from that specific donor.

Discussion 

Treatment of massive rotator cuff tears remains a difficult problem, with variable rates of healing reported in the literature. The use of tissue augmentation has been described to enhance healing after rotator cuff repair.6, 7, 8, 9, 10, 11, 12 A number of different tissue grafts are available for use in augmentation, including both xenograft and human allograft tissue. There is limited information in the literature regarding outcomes with the use of tissue augmentation for rotator cuff repair. Sclamberg and coworkers11 reported on minimum 6-month follow-up on 11 patients after rotator cuff repair with augmentation using porcine SIS. Follow-up MRI scans at 6 months showed recurrent tear in 10 of 11 patients. There was no significant difference between preoperative and postoperative American shoulder and elbow surgeons scores, and 5 patients actually had worse scores postoperatively. They concluded that SIS xenografting appears to be ineffective in rotator cuff augmentation. Recently, Iannotti12 reported the preliminary data from the Restore Patch Cleveland Clinic Clinical Trial. They randomized 30 patients with two tendon (supraspinatus and infraspinatus) tears to open rotator cuff repair with or without augmentation with the Restore Patch (porcine SIS). Average age was 57 years and all tears were able to be repaired with graft sewn over the repair. All patients were followed for a minimum of 1 year and underwent clinical examination and MR arthrogram follow-up. These results showed no statistical difference in PENN shoulder scores between the groups, but a trend toward less favorable results in the Restore Patch group. In addition, there was a trend toward decreased rate of healing and shoulder score in the Restore Patch group when normalized for tear size. They also observed 3 of 15 Restore patients with an excessive sterile inflammatory reaction. Our early MRI and clinical results suggests that acellurized allograft dermis is more effective and causes less inflammatory reaction than currently available xenografts.

The Graft Jacket used in this study is a processed human dermal allograft. During tissue processing, the dermis is decellularized, while maintaining its biomechanical and structural composition. The acellular nature of the tissue decreases the risk of deleterious reactions. The weaknesses of this study include the small number of patients and absence of a control group and an inability to obtain MRI scans on all of the patients. Our early results using this technique have demonstrated improvement in subjective pain and functional scores, as well as measured range of motion and strength. Importantly, we have observed no cases of inflammatory reactions with this graft material. Advanced imaging techniques on 12 of the patients willing to undergo a postoperative imaging suggests a structural healing rate superior to repair alone or repair with augmentation compared with historical controls. However, with almost one-third of the patients refusing or unable because of logistics to obtain postoperative scans, no claims of structural improvement can be made.

In summary, open rotator cuff repair with Graft Jacket augmentation for massive rotator cuff tears appears to be a safe and efficacious option in the management of this difficult problem. However a controlled blinded, multicentered randomized study with longer follow-up would be required to determine conclusively that augmentation with Graft Jacket is superior to cuff repair alone or no cuff repair altogether for that matter.