1 Introduction

Treating symptomatic retracted rotator cuff tears remains a challenge for the surgeon. The mobilization of the edges and their reinsertion in the anatomical area of the humeral head can be technically difficult if the edges are highly retracted.

Numerous techniques have been used in cases of irreparable tears, such as arthroscopic debridement with or without biceps tenotomy,¹ partial convergence sutures,² tendon transfer,^3,4 superior capsule reconstruction with fascia lata,⁵ as well as tendon grafts,^6,7 allografts patches⁸ synthetic grafts⁹ or spacer-inspace balloon.¹⁰

Mihata¹¹ reported that, biomechanically, the reconstruction of the superior glenohumeral capsule with fascia lata can stabilize the shoulder, improving its functionality.

The use of an autologous semitendinosus graft to reconstruct the superior capsule of the glenohumeral joint is intended to cover, in part, the superior cuff defect with the static tenodesis effect and to prevent the humeral head lifting with deltoid muscle activity. The first results for this technique in open surgery have already been published.¹²

The aim of this study was to report the details of the arthroscopic surgical technique along with the clinical and radiographic follow-up of our first 5 patients. The working hypothesis is that this technique provide a clinical and functional improvement outcome and increase in AHD.

2 Materials and methods

2.4

2.4 Surgical technique

Under general anaesthesia and a prior interscalene block, the patient was placed in the lateral decubitus position. A tourniquet was applied to the leg where the graft was to be harvested.

We used five portals on the shoulder to perform the arthroscopic technique: posterosuperior to introduce the optic (I), posterolateral (II), lateral where the working cannula was to be positioned (III), anterior (IV) and superior (V) (Fig. 1 a).

Close

Fig. 1 Images of the plasty technique and design Fig.1a: Arthroscopic portals used for semitendinosus-plasty; Left shoulder and patient in lateral decubitus position; I. Posterosuperior portal, II. Posterolateral portal, III. Lateral portal, IV. Anterior portal, V. Superior portal. Fig. 1b: Placement of a 5 mm metal implant on the superior edge of the glenoid. Fig. 1c: Placement of the implant in the anterior area of the glenohumeral joint, posterior to the biceps slider. Fig. 1d: View of the plasty from the posterosuperior portal after surgery; the humeral head is largely covered by the graft. Fig. 1e: Didactic diagram of the semitendinosus-plasty.

Export to PPT

With the optic in the posterosuperior portal (I) and the working cannula in the lateral portal (III), the shoulder joint was inspected and assessed with regard to the type of tear and its retraction, the possibility of mobilizing the edges and the integrity of the long portion of the biceps. If, after the debridement and release of the supraspinatus tendon edges, it was found to be retracted and not mobilizable, we opted to perform the plasty. One of the indispensable requirements for its execution is subscapularis tendon integrity and an intact or partially torn (but suturable) infraspinatus tendon.

The semitendinosus graft was taken from the ipsilateral knee. The ends of the graft were referenced with Ethilon no. 2 suture threads in order to insert and manipulate them in the subacromial space of the glenohumeral joint. We introduced the plasty into the joint through the anterior portal (IV) pulling from the posterolateral portal (II) with the help of a Kocher clamp; in this way, the plasty remained intra-articular, but its ends, where the suture threads were knotted, remained extra-articular. The plasty is quite long and therefore we can mobilize it and separate it intraarticularly in the visual field to see it better. We started to fix it at the superior level of the glenoid at the 12 o'clock position by introducing a 5 mm Fastin RC metal anchor (DePuy, Mitek) with double thread through the superior portal (V) (Fig. 1 b). It was attached to the superior glenoid rim by simultaneously and lightly pulling on the two extraarticular ends of the plasty. Then, we fixed the anterior end of the plasty in a location that had to be immediately posterior to the tendon of the long portion of the biceps before it enters the slide (Fig. 1 c). We positioned a Healix 5.5 anchor (DePuy, Mitek) via the lateral portal (III). This anterior fixing point should include part of the rotator interval area, the long portion of the biceps, and the plasty. We knotted only one of the two threads attached to the anchor through the lateral portal (III) maintaining plasty traction via the anterior portal (IV). Next, we moved on to the posterior plasty fixing; the fixing point must be immediately before the infraspinatus fixing, which, if partially detached, must be brought to its anatomical position. Laterally (III) and with the help of internal arm rotation, the fixing point is slightly milled and another Healix 5.5 anchor (DePuy, Mitek) is positioned. This posterior fixing point should include the plasty and the anterior infraspinatus edge. With traction on the threads to hold the plasty via the posterolateral portal (II), we tied the anchor threads through the lateral portal (III). To finish, the free part of the posterior end of the plasty, which was distal to its fixing, was brought forward and fixed under tension to the anterior anchorage with the other unknotted thread. We ended by cutting the plasty remnants intraarticularly after their anterior and posterior fixing. Observing the plasty we performed, we could see that it formed an isosceles triangle that covered a large part of the humeral head (Fig. 1 d-e).

The patient was hospitalized for 24 h and given intravenous antibiotic cover. All patients followed the same postoperative rehabilitation protocol. The arm was immobilised in a sling for 5 weeks. Subsequently, passive range of shoulder motion exercise was started by the physiotherapist for 3 weeks. Over the following 6 weeks, isometric and resistance movements were gradually added to the programme and lasted until 4 months.

3 Results

The mean follow-up was 40.8 months (range: 28–50 months). In 4 patients, the dominant shoulder was injured.

Preoperative standard radiographs using the Hamada classification showed grade 1 in three cases and grade 2 in two cases (Fig. 2).

Close

Fig. 2 Preoperative anteroposterior radiograph: 6 mm AHD. Hamada grade 1.

Export to PPT

Preoperative MRIs showed massive supraspinatus rotator cuff tears in all patients and partial tears of the infraspinatus in four cases (Fig. 3). Grade 4 fatty supraspinatus infiltration was present in four cases and grade 3 in one; there was also grade 1 fatty infraspinatus infiltration in five cases.

Close

Fig. 3 MRI. Cuff retraction and fatty infiltration of the supraspinatus muscle, Goutallier grade 3.

Export to PPT

All patients showed postoperative improvement in their VAS, Constant and ASES scores (P < 0.001). The postoperative range of mobility showed improvements in all patients in terms of active forward flexion, abduction and external rotation (P < 0.001) (Tables 1 and 2).

At 12 months, anteroposterior radiographs showed a significant increase in AHD from 5.4 mm to 7 mm (P < 0.001) (Fig. 4) and the MRIs showed continuity and integrity of the semitendinosus ligamentoplasty in all patients (Fig. 5).

Close

Fig. 4 Postoperative anteroposterior radiograph 12 months after surgery: 8 mm AHD.

Export to PPT

Close

Fig. 5 MRI 12 months after surgery. The plasty is intact and in continuity rebuilding the superior capsule.

Export to PPT

Three patients reported feeling “much better”, and two “better”. All patients were able to return to their usual work.

4 Discussion

The treatment of massive (>5 cm in diameter) non-reinsertable rotator cuff tears in the humeral head tuberosity is undefined.

Various procedures have been used in these cases, such as arthroscopic debridement with or without biceps tenotomy,¹ medialization of the tear edges with a partial convergence suture,² tendon transfer,^3,4 tendon allograft patches,⁶ tendon autografts,⁷ allografts patches,⁸ synthe tic material grafts,⁹ spacer-inspace balloon¹⁰ or superior capsule reconstruction with fascia lata.⁵

However, in many cases the results have not been satisfactory: the use of tendon allograft patches⁶ or tendon autografts⁷ that are sutured proximally to the cuff tendon tear and laterally to the greater tuberosity of the humeral head suffer a high percentage of tears.

Audenaert et al.⁹ reported that the lifting of the humeral head with shoulder abduction movement decreases AHD and therefore causes friction between the acromion and the synthetic material patch causing it to tear.

The use of an autologous graft has several advantages, such as avoiding graft rejection and the transmission of infections, as well as reducing the cost of surgery.

Mihata published the clinical results⁵ of a biomechanical study¹¹ in which the superior capsule was reconstructed with autologous fascia lata preventing superior humeral head migration caused by abduction movement following deltoid action. In a sample of 24 patients with a mean follow-up of 34 months, flexion mobility rose from 84° to 148° and external rotation from 26° to 40°. The AHD increased from 4.6±2.2 mm to 8.7±2.6 mm. The mean ASES went from 23.5 to 92.9 points. In 4 patients, there was a plasty tear during the follow-up.

We used an autologous semitendinosus graft instead of the fascia lata to reconstruct the superior capsule for various reasons; the graft is easier to harvest, it leaves less scarring and it is also technically easier to handle within the subacromial space during the intervention. The semitendinosus is commonly used in the reconstruction of knee ligaments and coracoclavicular ligaments. As a graft, it has demonstrated that it incorporates into the recipient bone and regenerates in the donor area over the months.

Rosales et al.¹² published the results using an autologous semitendinosus graft in open surgery. In a series of 8 cases with a 12 months follow-up, they found significant clinical improvement; Constant test scores went from 49 to 77.25, mobility in active flexion went from 99.3° to 142.5° and external rotation from 32.5° to 43.7°. There was also a significant increase from the preoperative AHD of 5.25 mm up to 8.18 mm. No plasty tears were observed on the MRIs at 12 months.

The results of the arthroscopic plasty were also favourable. With an average follow-up of 40.8 months, the mean ASES scores significantly increased from 56.2 to 92 and the Constant shoulder scores from 46.8 to 82.8; the pain level (VAS) decreased from 6.2 to 1.2. The forward flexion mobility went from 100° to 154°, the abduction from 90° to 130° and the external rotation from 32° to 42°. At 12 months, the radiographic findings demonstrated a significant increase in AHD from 5.4 mm to 7 mm. The MRI assessments showed continuity and integrity of the ligamentoplasty in all cases.

Arthroscopic semitendinosus-plasty is a difficult and demanding technique; nonetheless, it avoids the complications that can occur in open surgery.

We can confirm our working hypothesis that this technique provide a clinical and functional improvement outcome and increase AHD. Like other authors,^5,11,12 we believe that this plasty is indicated in patients with pain and/or loss of shoulder flexion and abduction due to irreparable superior rotator cuff tears. However, it would not be indicated in irreparable tears of the supraspinatus associated with external rotation deficit caused by a lesion of the infraspinatus and teres minor, where is necessary to restore active external rotation with tendon transfer, latissimus dorsi³ or lower trapezius,⁴ pseudoparalytic shoulders and shoulders with glenohumeral osteoarthritis grade 4–5 of the Hamada classification.¹⁴

Limitations: The following can be cited as study limitation: the number of patients was small.

5 Conclusions

This arthroscopic technique using semitendinosus tendon autograft enabled the superior capsule reconstruction for irreparable rotator cuff tears. Given the small number of cases, few conclusions can be drawn. Nonetheless, the results regarding this technique may be considered promising.

Ethical responsibility

Safeguarding of people and animals. The authors declare that for this research no experiments were carried out on human beings or on animals.

Confidentiality and data protection. The authors declare that they have followed their work centre's protocols regarding the publication of patients' data.

Right to privacy and informed consent. The authors declare that no patient data appear in this article.

Financial interests

The authors declare that they have not received any type of financing to realise this work.

Declaration of competing interest

The authors declare that they have no conflict of interest.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.