Effect of three remplissage techniques on tendon coverage and shoulder kinematics: a navigated robotic biomechanical study

Engaging Hill-Sachs defects of the humeral head are well documented risk factors for recurrent glenohumeral shoulder instability and failed anterior shoulder stabilization [1‐3]. Several studies have shown that an engaging Hill-Sachs defect treated with Bankart repair alone is associated with a high re-dislocation rate [1, 3‐5]. Different surgical options have been described in order to address this problem: Retrograde disimpaction, osteochondral allograft, humeral osteotomy and partial or complete replacement of the humeral head [6‐9].

Connolly first described an open procedure of filling a Hill-Sachs defect with infraspinatus and capsule tenodesis, together with a Bankart repair [10]. This technique works by converting an intra-articular defect into an extra-articular one and thus preventing the defect from engaging at the anterior glenoid rim. Furthermore, there is a check-rein effect on the humeral head, effected by the fixed infraspinatus tendon and posterior capsule. Wolf et al. termed the procedure remplissage when they performed the procedure arthroscopically, placing suture anchors into the valley of the Hill-Sachs defect and passing the sutures posteriorly through the infraspinatus tendon and capsule [11]. Cadaveric studies have shown that medium sized Hill-Sachs lesions without glenoid bone loss can be successfully treated with the remplissage procedure if combined with Bankart repair [12, 13]. In recent years, clinical studies have already reported good functional outcomes with minimal loss of shoulder range of motion [14‐16]. Although, filling of the Hill-Sachs defect by infraspinatus tendon and capsule on post-operative scans has been demonstrated [17], there are no studies in the literature investigating the amount of infraspinatus tendon coverage over the defect zone produced by different techniques.

The aim of this in-vitro biomechanical study is to compare three arthroscopic remplissage techniques, each with suture anchors placed in the valley of a standardized Hill-Sachs defect, but with varying methods of fixation of suture materials. The study uses a novel parallel position-orientation and force-moment controlled navigated robotic model. Robotic parallel control was used to minimize forces in the orthogonal axis [18, 19]. At the same time, this allowed for external rotations while maintaining a constant forward flexion and abduction angle with centered humeral head within the glenoid. The effect of each remplissage technique on (I) glenohumeral engagement, (II) glenohumeral torque (Nm), and (III) area of tendon coverage over the Hill-Sachs defect zone (%) was quantified.

It was hypothesized that a knotless suture tape technique confers a larger area of tendon coverage over the defect zone compared to suture tying above the anchors and/or tying between the anchors (double-pulley) with no differences in defect engagement and glenohumeral torque.

This study was approved by the Ethics Committee of the Faculty of Medicine of the Munich University of Technology (ID number: 339/14).

The most important finding of this biomechanical study is that the proposed knotless suture tape technique provides significant superior coverage of the infraspinatus tendon over the Hill-Sachs defect zone. All three techniques revealed stable joints without engagement. There was no difference in humeral torque between groups. Therefore, the studies hypothesis was proofed. Hill-Sachs remplissage is a non-anatomic arthroscopic soft tissue procedure performed in order to add extra stability to a conventional Bankart repair. Several clinical studies have already shown good and reliable clinical results for this “add-on” technique. [14‐16, 24‐26] However, some reports have suggested, that patients undergoing a remplissage procedure might suffer from postoperative glenohumeral rotational deficits [14, 27]. In contrast, a recent systematic review indicated no significant loss of shoulder motion following the procedure [28].

Several different techniques to perform a remplissage procedure have been described in literature and a gold standard has not been defined yet. In order to better understand the effect of different remplissage procedures on joint kinematics and stability, a few biomechanical studies have been conducted [12, 29‐32].

A study by Elkinson et al. also investigated the effects of 3 different remplissage techniques on shoulder kinematics. They found that medial suture passage through the posterior soft tissue consistently led to the greatest mean restriction of range of motion and the highest stiffness values [12].

The results of the presented study show that all three investigated remplissage techniques re-stabilized the glenohumeral joint and prevented engagement of the defect sufficiently. Regarding influence on glenohumeral torque, all three intervention groups showed significantly impaired torque with 60° abduction and ≥30° external rotation compared to anatomical conditions. There was no difference observed between techniques. This in vitro finding supports current clinical, as well as biomechanical findings [12, 14, 15, 22].

Regarding area of tendon coverage over the Hill-Sachs defect, this study showed that the proposed knotless suture tape technique (T3) created a significantly larger area of tendon coverage over the defect zone compared to the two other techniques (T1, T2). The authors are not aware of any literature investigating and reporting on area of tendon coverage over the defect zone after remplissage.

For rotator cuff repair biomechanical and clinical studies have shown the critical influence of tendon coverage over the footprint for successful healing [33, 34]. In the opinion of the authors, in addition to the initial tenodesis or check-rein effect, optimal healing of posterior tissues into the defect is important for remplissage as well. Thus it is logical that anchors be placed in the valley of the defect for healing to occur. In consequence the authors propose a new technique, using knotless anchors and suture tape, aiming to increase interface of tendon coverage and yet have not more limitation of glenohumeral torque compared to other techniques.

This study has inherent weaknesses, including the limitation of applying a biomechanical model to a clinical problem. Using a simplified cadaveric model, important factors contributing to shoulder stability such as dynamic muscle activation were not considered. Nevertheless, with dynamic range of motion the joint compression force was incorporated in this in vitro model [20, 35]. Furthermore, the results from this time zero study do not account for any phenomenon of biological healing e.g. formation of scar tissue, which occurs in the clinical setting and might result in decreased ROM. Postoperative soft tissue stretching, which may minimize the restrictive effects of the remplissage over time could also not be investigated with this study. However, the setup has shown to be reproducible, reliable and capable of answering the stated hypothesis.

The strengths of the study include the use of an industrial robot and an optical measurement system, providing a high accuracy for joint stability and kinematics testing. Repairs only differed by repair technique and were all performed by a single fellowship trained orthopedic surgeon. Finally, the native specimens were used for comparison mimicking a “pre- vs. postoperative” change of shoulder kinematics.

All remplissage techniques prevented engagement of the Hill Sachs defect without luxation. With high abduction and external rotation ≥30° all techniques showed significant higher humeral torque compared to the intact specimens, while there was not one technique superior over the others. The suture tape technique conferred the largest and most effective area of tendon coverage over the Hill-Sachs defect zone. In the clinical setting, the increased tendon coverage might lead to better healing at the soft tissue—bone interface. Clinical studies will be necessary to proof potential benefits for outcome.