The results of using a tendon autograft as a new rotator cable for patients with a massive rotator cuff tear: a technical note and comparative outcome analysis

Massive rotator cuff tears (MRCTs) are defined as cuff tears involving two or more cuff tendons or a retraction of ≥ 5cm [1]. Patients frequently present with pain and loss of function (pseudoparalysis), sometimes after minimal trauma. A MRCT can have a huge impact on daily life and eventually could lead to a cuff arthropathy. Variable success rates are seen after primary cuff repair in small-to-medium-sized tears with retear rates between 14 and 25% [2]. If the repair is successful and the cuff has healed after surgery, patients have a better outcome compared with patients after conservative or failed surgical therapy [3, 4]. MRCT defects cannot be easily closed, and there is a rate of retears reaching up to 94% [5, 6]. Several surgical reconstructive options for MRCTs are available, each with its own reported advantages and complications [7]. Patches and grafts are expensive and can lead to rejection. Another technique is superior capsular reconstruction to treat these lesions [8]. Initial results for reconstruction of the superior capsule seem promising, although long-term results are awaited. Reversed shoulder arthroplasties and tendon transfers are alternative options, yet these are major surgeries with considerable morbidity [9].

A tenotomy or tenodesis of the long head of the biceps tendon is often performed as part of rotator cuff surgery. This offers the possibility of using the biceps tendon as a graft. Different studies used this tendon as a free graft or leaving the distal or proximal attachment intact, reporting significant improvement of function [10]. In the native shoulder, the rotator cable is a thickening in the rotator cuff that serves as a primary load-bearing structure between the rotator cuff and the humerus and functions as a tension bridge. Recent studies show the clinical and biomechanical importance of this structure [11, 12].

We developed a technique to treat MRCTs by reconstructing the rotator cable with a long head of biceps tendon autograft [13]. But a hamstring tendon is also suitable. This technique represents a reconstruction with several advantages: use of an autograft, ease of harvesting, no graft reactions, fixation of cuff to reconstructed rotator cable instead of footprint, and potential prevention of cuff arthropathy.

The aim of this study is to evaluate the clinical and radiological result after arthroscopic reconstruction of the rotator cuff cable with an autograft tendon in patients with irreparable massive rotator cuff tears.

A total of four patients were included in this study. Patient characteristics are depicted in Table 1. Mean age of the study population was 64 years (61–67 years), all with their dominant side affected, and one was a smoker. In all patients, a massive cuff tear of the supraspinatus and infraspinatus tendon was present with a Patte stage 3 retraction; in one patient, a partial subscapularis tendon tear was found and repaired. Fatty infiltration on the postoperative MRI scan differed between Fuchs stages 2 to 3 on the preoperative MRI. One patient discontinued the study 9 months after surgery because of lack of postsurgical improvement, and a reversed shoulder prosthesis was inserted. The final scores of this patient were considered as an endpoint.

For comparison, a similar group of patients who were treated conservatively was selected. This comparable cohort consisted of five patients, mean age was 65 years (57–72 years), and more women were present in this group. All patients had a massive cuff tear of the supraspinatus and infraspinatus tendon with a Patte stage 3 retraction; in one patient only, the supraspinatus was torn and was treated using physiotherapy.

Table 2 shows the clinical outcome of the study group after surgery and the group treated with physiotherapy. Total CMS score increased significantly after surgery in three of the four patients, with a mean improvement of 26.4 points after 12 months (p = 0.023) (Fig. 4). The WORC index, depicted as a percentage of a normal score, showed a mean improvement of 5 points (p = 0.191). SST, VAS pain, VAS disability, and VAS satisfaction scores also improved in these patients. All except for patient 4 showed improvement on the scores. In all patients, an MRI was performed 12 months after surgery. In all patients, the reconstructed rotator cable had disintegrated and the rotator cuff was detached from the rotator cable and retracted. Fatty infiltration of the infraspinatus muscle before surgery was grade 2 in two patients and grade 3 in two patients. At follow-up, all patients had fatty infiltration of the infraspinatus muscle grade 3 on MRI. The partial subscapular repair was healed on MRI in patient 3. None of the patients had any signs of osteoarthritis on MRI before or after surgery.

Compared with the patients who were treated conservatively, the CMS and VAS pain and disability scores of the surgical group were similar before treatment and at 12 months follow-up. At 3 and 6 months after surgery, the clinical results were inferior to the non-operative group.

Clinical and radiological results after tendon autograft for reconstruction of the rotator cable were collected in four patients with a MRCT. On the MRI scan, 1 year after surgery, the rotator cable reconstruction disintegrated and failed; clinical results were not superior to the results after non-operative treatment with physiotherapy. Because of disappointing preliminary results and a vanishing reconstructed rotator cable on MRI during 1 year of inclusion, no further patients were included. When developing this new surgical procedure, we hypothesized that the procedure might have several advantages. This technique results in a biomechanical reconstruction leading to transporting force couples from the rotator cuff to the humeral head. As the remaining rotator cuff was attached to the rotator cable and not to the footprint, we were able to reattach the rotator cuff to the cable in irreparable cuff tears. With this technique, tension on the massive cuff tear gap was minimized by medializing the new rotator cable. The technique represents a reconstruction with the use of an autograft. The biceps tendon is easy to harvest and gives no graft reactions. In other studies, the biceps tendon was used as an augmentation of the rotator cuff with good clinical results. Although we observed improved clinical results in three out of four patients 1 year after surgery with the CMS and WORC exceeding the minimal important clinical difference of 10.4 points [19] and 282.6 points [20] respectively, MRI showed failure of the reconstructed rotator cable in all patients.

While performing the index operation, we used margin convergence to close the cuff partially to the reconstructed rotator cable, but on MRI scan at follow-up, the cuff was re-torn and retracted compared with the preoperative situation. The improvement in pain and function could not be attributed to the reconstruction of the rotator cable as this construct failed over time [21]. The clinical improvement over time in the surgical group was similar to the improvement noted in the control group, treated non-operatively with physiotherapy. In two randomized controlled studies about cuff repair versus conservative treatment of degenerative rotator cuff tears, good clinical improvement was also found after conservative treatment. These studies showed a mean improvement of the CMS of 16.8 points after 1 year [3] and 18.4 points after 2 years [22] following conservative treatment. In both studies, this was not significantly different from the surgically treated patients. In our control group, a mean improvement of the CMS of 19.1 points was found after 1 year.

The rotator cuff showed fatty infiltration (grade 3 in two out of four patients). This may have resulted in low healing capacity and eventually a detachment of the rotator cuff from the reconstructed rotator cable on the control MRI. Previous studies also show that this is a predictor for worse outcomes in cuff repairs [23], even in cases of successful repair [24]. Reconstruction of the rotator cable failed over time in all patients. The biceps tendon may have been degenerative, thus preventing ingrowth, although histological analyses show that a degenerated biceps tendon is still rich in collagen [25]. In one patient, a hamstring graft was used because of a missing long head of biceps during surgery. This may be a bias confounding the outcome. The study group is small and consisted only of four patients. In our study protocol, we decided to include patients for this new technique for 1 year, and then evaluate the clinical and radiological results. Because the construction failed in all patients, we decided to end this pilot study. There is a possibility that this technique might be successful in some patients when a larger group of patients would be operated. Comparing the results of our study with other studies using a biceps tendon graft for rotator cuff repair, a superior outcome was found when the biceps tendon was used for augmentation with an improvement of 44.1 points on the CMS by Cho et al. [26]. Also, the long-term outcome scores of the latissimus dorsi transfer as treatment for massive rotator cuff tears were better, with an increase of 27 points on the CMS [27].

Treating MRCTs remains a significant challenge for the clinicians. Several surgical options are available for patients with a MRCT if conservative treatment fail: debridement, long head of biceps tenotomy, partial repair, rotator cuff advancement, bridging graft repair, superior capsular reconstruction, subacromial spacer, and reverse total shoulder arthroplasty. The variability in patient characteristics, co-interventions, outcome reporting, and length of follow-up in studies on MRCTs complicates sound comparison of treatments [28]. Joint preserving procedures are preferably used in young patients; in the older population, reversed shoulder arthroplasty or maybe the subacromial balloon spacer might be indicated. The superior capsular reconstruction has become a popular surgical technique. The short-term results of superior capsular reconstruction show consistent improvement in shoulder functionality and pain reduction. However, on longer follow-up, decreased acromiohumeral intervals indicate dermal allograft elongation and persistent superior migration of the humerus [29]. Placement of the subacromial balloon spacer is a minimally invasive, technically simple procedure with favorable patient-reported outcomes at limited short-term follow-up. However, inherent methodological limitations and patient heterogeneity between studies using the subacromial spacer may impair the ability to fully characterize the long-term efficacy, particularly relative to other potential surgical options [30, 31].

In conclusion, the arthroscopic reconstruction of the rotator cable using a tendon autograft failed over time in this pilot study, and showed no clinical benefit in comparison with the non-operative treatment with physiotherapy in patients with a MRCT. We therefore cannot recommend using this surgical procedure to treat patients with a massive rotator cuff tear.