The Superior Capsular Reconstruction

Lessons Learned and Future Directions

John M. Tokish, MD; Justin L. Makovicka, MD


J Am Acad Orthop Surg. 2020;28(13):528-537. 

In This Article

Shoulder Superior Capsule

The superior capsule lies between the rotator cuff and joint space on the undersurface of the supraspinatus and infraspinatus tendons, attached medially to the superior glenoid and laterally to the greater tuberosity.[13] It acts as a static stabilizer to superior translation of the humeral head, with the rotator cuff aiding in dynamic reinforcement of the capsule and providing strength.[17] Through biomechanical analysis, it has been shown that the glenohumeral translation is increased in all directions when there is a defect in the superior capsule.[18] Evidence has shown that in addition to the massive tear of the rotator cuff, the superior capsule is often also torn in these injuries because of its intimate relationship with the cuff tissue.[19] This loss of superior stabilization resulting in humeral head malposition and improper muscle fulcrum may contribute to some of the pain seen in these patients.[17]

Early biomechanical work by Mihata et al[13] determined that this superior translation could be completely restored by using a patch graft to reconstruct the superior capsule, at least in the cadaver. In their work, fascia lata allografts were used in three scenarios: (1) as patch grafts to reconstruct the supraspinatus tendon, (2) as patch grafts to reconstruct the superior capsule, and (3) as patch grafts to reconstruct both the supraspinatus tendon and superior capsule. They found that all patch graft scenarios fully restored subacromial contact pressures to that of the intact rotator cuff, but none altered the glenohumeral joint force. Of note, the authors reported that superior translation of the humerus was fully restored to precut levels only after reconstructing the superior capsule, while isolated repair of the supraspinatus only partially restored translation. It should be noted, however, that this partial restoration was still within 2 mm of the precut levels at all arm angles, and thus, the clinical significance of this difference remains unknown. They concluded that superior capsule reconstruction with a patch graft attached medially to the superior glenoid and laterally to the greater tuberosity could completely restore stability of the glenohumeral joint by depressing the humeral head.

Based on this biomechanical work, there have been several theories put forth to explain why the SCR may have beneficial effects. One effect may be a spacer effect by the pouch itself. The graft may provide the equivalent of an interpositional spacer to dampen the compression of the humeral head against the bottom of the acromion. A second possibility is a so-called "trampoline effect," whereby the graft physically holds the head inferiorly to improve acromiohumeral clearance. A third possibility that has been purported is a force coupling effect. In this theory, when the SCR graft heals to the native rotator cuff, the direction of force generated in the native remaining posterior rotator cuff may transfer across the patch and improve the vertical force couple, thus improving the humeral head depressor effect of the graft/rotator cuff construct.[10] Although biomechanical studies have shown findings consistent with these hypotheses, no study has proven the exact mechanism of why some patients achieve improved motion and pain relief.

Autograft Reconstruction

Mihata et al[14] subsequently proposed SCR with the use of a fascia lata autograft for massive, irreparable rotator cuff tears. In this technique, the fascia lata autograft is folded over to achieve a graft thickness of 6 to 8 mm and attached medially to the glenoid, laterally to the greater tuberosity, and to residual rotator cuff tissue to improve force coupling (Figure 1, A–C). This technique has demonstrated early successful results in patients with these difficult problems, and functional outcomes have approached those seen with complete rotator cuff repair (Table 1). In their early study, at an average follow-up of 34 months, 24 shoulders in 23 patients who had undergone SCR had improved outcome scores, with patients' ASES scores improving from 24 to 93 at follow-up.[14] The acromiohumeral distance (AHD) was also found to increase an average of 4.1 mm after surgery, in addition to demonstrating an increase in shoulder muscle strength. Mean active forward elevation (FE) increased from 84° to 148°, and external rotation increased from 26° to 40°. Two follow-up studies from the same authors demonstrated similar, successful results (Table 1).[20,21] At final follow-up in both studies, patients demonstrated improvements in both ASES scores and mean active FE.

Figure 1.

Autograft SCR. A, The harvest site defect, which is generally ipsilateral tensor fascia lata done in the same prep field as the shoulder in a lateral decubitus position. B, The autograft SCR folded on itself measuring the appropriate dimensions and an 8 mm thickness. C, The completed autograft SCR demonstrating tissue coverage in a reconstructed shoulder.

An additional follow-up study by Mihata et al[20] also demonstrated high rates of return to recreational sports and physical work. In that group, all 26 patients who participated in athletics before rotator cuff tear fully returned to sport after undergoing SCR. In addition, 32 of 34 patients who had heavy physical labor jobs returned postoperatively to their previous work. Mihata et al[21] also studied whether SCR could reverse preoperative pseudoparalysis. Patients undergoing SCR were labeled and allocated to three groups based on a definition of pseudoparalysis as: none (FE > 90°), moderate (active FE < 90°, but maintained >90° of elevation after the shoulder was passively elevated); and severe (<90° of active shoulder elevation and a positive drop-arm sign). After undergoing SCR, pseudoparalysis was reversed in 96% (27 of 28) of patients with moderate pseudoparalysis and 93% (14 of 15) of patients with severe pseudoparalysis. Although most surgeries were primary (4 revisions), and there was no stratification by those whose motion was limited due to pain, which has been called for in the literature,[1,22] the reversal numbers reported by Mihata et al represent a potentially paradigm shifting advancement in the treatment of the massive irreparable rotator cuff tear.

Dermal Allograft Reconstruction

As SCR has evolved, an arthroscopic technique for SCR using an acellular dermal allograft has been developed in an effort to eliminate the need for fascia lata autograft and its associated morbidity (Figure 2, A and B).[17,23] Dermal allografts have previously been used in rotator cuff repair augmentation, and although markedly more expensive than fascia lata autografts, theoretical advantages include reduced donor site morbidity, decreased operating room time, ease of preparation, and strength of the graft. As this modification of the autograft technique has gained popularity among surgeons, early clinical results using an allograft have been favorable. Hirahara et al published results on eight patients with a mean follow-up of 32 months managed with dermal allograft SCR for irreparable rotator cuff tears and determined that SCR improves functional outcomes and improves AHD, regardless of the graft type used (Table 1).[17] Their patients showed notable improvements in ASES scores (42 to 87) and decreases in visual analog scale (VAS) pain scores (6.3 to 0.4).[23] In addition, the postoperative ASES (SCR: 87; control: 71) and VAS (SCR: 0.38; control 3.0) scores in the SCR patients were superior (only VAS significant) to postoperative scores of historical control patients who had undergone repair of massive rotator cuff tears by the same surgeon group.[17] AHD in these patients measured 8.5 and 7.7 mm immediately and 2 years postoperatively, respectively, compared with 4.5 mm before surgery. MRI was only obtained to evaluate the integrity of the graft when a failure was suspected. Five of the eight patients underwent MRI examination, in which the graft was found to be intact in four of the patients (80%).

Figure 2.

A, A human dermal allograft skin graft is prepared for passage into a rotator cuff deficient shoulder. B, The secured human dermal allograft is viewed arthroscopically from the lateral portal viewing the greater tuberosity fixation of the graft onto the humerus.

In a second multicenter prospective study, it was determined that SCR using a dermal allograft provided a successful outcome in approximately 68% of patients, as defined by postoperative ASES scores >50.[16] Denard et al reported on 59 patients with a mean follow-up of 17.7 months, who underwent SCR with a dermal allograft for massive, irreparable rotator cuff tears (Table 1). Patients after SCR demonstrated an improvement in forward flexion from 130° preoperatively to 158° postoperatively, while external rotation improved from 36° to 45°.[24] The ASES (44 to 78) and subjective shoulder value (35 to 76) scores of patients were also found to improve, while VAS scores decreased from 5.8 to 1.7. The AHD initially improved 1 mm from baseline at 2 weeks postoperatively, but subsequently decreased to only 0.1-mm improvement at final follow-up. MRI was used postoperatively to evaluate graft healing in 20 patients, and only 45% of the dermal allografts were found to be healed. These findings raise substantial concerns regarding the healing potential of dermal allografts. They also raise concern that others may not be able to replicate the results of the surgeon who designed the procedure, possibly due to differences in technique and graft type. It should be noted that this was a preliminary study, early in the evolution of the use of dermal allograft for SCR. Of the 11 unhealed grafts, two were grafts <3 mm in thickness and three were grafts used in patients with Hamada grades >3.[24] Also, it should be noted, that 100% of patients with a healed graft had a successful outcome, in contrast to only 45.5% of the unhealed grafts.

In the largest published study of SCR with the use of a dermal allograft, Pennington et al[24] reported on the results of 88 shoulders in 86 consecutive patients who were treated for irreparable rotator cuff tears, achieving a 90% patient satisfaction rate at follow-up (Table 1). At 1-year follow-up, patient improvement in both ASES (52 to 82) and VAS (4.0 to 1.5) scores was reported. After SCR, patients demonstrated notable increases in both forward flexion (121° to 160°) and abduction (103° to 159°). Also, improvements in both acromiohumeral interval (7.1 to 9.7 mm) and superior capsular distance (52.9 to 46.2 mm) were demonstrated at 1-year. The authors defined superior capsular distance as the "arc length from the medial aspect of the greater tuberosity to the superior aspect of the glenoid," with the goal of quantifying the superior translation of the humeral head. Thirty-six of their patients also had 2-year follow-up data available. These patients continued to show improvements in VAS (1.2) and ASES (85) scores at 2 years without evidence of failure. Although promising, this study did not report results stratified by pseudoparalytics, Goutallier changes, or functionally compromised rotator cuff tears.

It should be noted that the biomechanical and clinical literature for SCR should be critically evaluated before determining the procedure's effectiveness for irreparable rotator cuff tears. For example, all biomechanical and clinical outcome studies using fascia lata autograft were conducted by the same author (Mihata et al), who originally proposed the concept of SCR. In addition, the clinical studies using dermal allograft have only been published by a few groups of surgeons. The results of the procedure done by surgeons other than those who developed or championed it remain to be determined.