The results of arthroscopic anterior stabilisation of the shoulder using the bioknotless anchor system

Stability of the shoulder relies on passive and dynamic constraints. Passive constraints include the bony anatomy, augmented by the glenoid labrum, capsule, ligaments and a negative intraarticular pressure. Dynamically, the rotator cuff and scapular musculature act to centralise the humeral head within the glenoid fossa throughout the normal range of motion of the shoulder joint. Instability can be defined as repeated dislocation or subluxation of the glenohumeral joint without extraordinary external force. Shoulder instability is a very common problem which predominantly affects a young, active population [1]. The most frequent pathology is the classic Bankart lesion (figure 1) following a traumatic anterior dislocation [2], but instability can also arise without trauma and can occur in any direction. Symptoms range from mild symptomatic instability to recurrent dislocation, with some patients able to voluntarily dislocate their own shoulders. Therefore shoulder instability represents a spectrum of conditions with different aetiologies and methods of management.

The evolution of minimally invasive surgery has led to the development of arthroscopic techniques of shoulder stabilisation. However, the majority of the published literature, including 2 meta-analyses comparing arthroscopic versus open Bankart repair suggest a poorer outcome in the arthroscopic groups [1, 3‐6] with recurrence rates of up to 34% being reported. Most of this research is based on older methods of fixation such as staple capsulorrhaphy and transglenoid suturing. More recent developments in technique and instrumentation, most notably suture anchors, are meeting with greater success. There are now several studies showing comparable outcomes between open repair and stabilisation using one of the later arthroscopic methods [7‐13]. Knotless anchors arised as a solution to the difficulty of tying secure knots with reliable tension arthroscopically [14]. Bioabsorbable anchors were developed to reduce the risk of metalware infection and chronic foreign body reactions encountered with non-absorbable anchors. Bioknotless anchors are a combination of the above two technologies and consist of Poly(L-lactide) which degrades over 2–3 years, eventually being replaced by the patient's autogenous tissue [15]. The other benefits of arthroscopic surgery, in particular the lack of subscapularis damage, faster return to normal activity, improved range of motion and shorter hospital stay are now being gained without the concomitant increase in recurrence of instability.

Between January 2005 and March 2006 15 patients underwent arthroscopic stabilisation of the shoulder using BioKnotless anchors (DePuy Mitek, Raynham, MA). All patients were operated on by a single surgeon (VK) using the technique described below. No patients were excluded from the study and none were lost to follow up. 13 were male and 2 were female. Their average age at time of surgery was 27 (17 to 53). All presented as a result of recurrent traumatic anterior dislocations of which 6 experienced less than 5 dislocations and the remainder more than 5. Each patient was fully assessed by the senior author prior to surgery.

12 out of 15 (80%) patients were satisfied with their surgery (table 1). 2 patients had recurrent symptoms. 1 had a further dislocation post-operatively (7%) which occurred without major trauma within 6 months of surgery. Following reduction he had no further dislocations or symptoms of instability. Interestingly, this patient was still satisfied with the surgery as at the time of final follow-up, he felt his shoulder was better than prior to the procedure and he had returned to his usual work and recreational activities. A second patient suffered with symptomatic instability and was unable to return to his usual level of sport. Of the 3 patients who were unsatisfied, 1 was because of recurrent symptomatic instability and 2 were due to an inability to return to previous levels of activity. The range of movement measured at final follow up of all 4 patients who were either unsatisfied with the surgery or had recurrent symptoms (or both) is shown in table 2.

The average constant score was 77 ± 9.8 following surgery in the affected shoulder which was 84% ± 7.1% of that measured in the control shoulder. No patient had a score less than 50, 3 had scores between 51 and 75 and 11 had a score of 76 or above.

12 patients returned to their previous levels of work and sporting activity, 5 of whom participated in contact sports. 2 patients who were involved in heavy manual work could not return to their usual job following surgery and 2 patients curtailed their level of sporting activity post operatively.

We experienced no other major complications as a result of this surgery.

The majority of surgeons would agree that recurrent, traumatic anterior dislocation is best treated surgically. The results of conservative management are poor with 75–90% of patients re-dislocating within 18 months of the primary event [16, 17]. The timing and technique of surgery is more controversial.

Open repair of the capsulolabral complex has been the most reliable method of avoiding recurrent instability, studies reporting between 3% and 10% risk of further dislocation or instability post-operatively. This compares to recurrence rates as high as 34% following arthroscopic repair [1, 4, 5, 11, 18, 19]. New developments in arthroscopic surgery along with novel instrumentation, in particular suture anchors, are beginning to redress the balance.

Drawbacks of open surgery are a larger scar, division of subscapularis, slower recovery and rehabilitation and long term reduction in range of movement (particularly forward flexion and external rotation). Although some surgeons believe that it is this reduction in the range of movement that itself improves stability and keeps the shoulder out of the apprehension zone. However, most of the research on suture anchors has only 1 to 3 years follow up, whereas the incidence of recurrent subluxation or dislocation has been shown to increase for up to 7 years post surgery [20]. There are no prospective randomised trials with sufficient length of follow up showing any benefit of arthroscopic over open stabilisation. Furthermore, several trials have reserved arthroscopic repair for the more simple cases and revert to open surgery for the complex patients, thus not comparing like with like. There is also a definite learning curve to be overcome and other complications such as anchor failure [13] and foreign body reactions [12] have been reported.

The return to full activity, including high-risk sports is less predictable than the recurrence of instability. As few as 1/3 of competitive athletes returned to their normal level of sporting activity (especially contact and/or throwing sports) following open repair [18]. 2 of the 3 patients in this study that were dissatisfied with their surgery cited the reason as their inability to return to their former level of activity. Both of these patients suffered no further dislocations or instability post-op and had constant scores of 73 and 61 (84% and 69% of control limb) respectively. One patient played local level cricket and was unable to bowl following surgery. The second had difficulty with heavy lifting and had to change his occupation. Objectively their operations were successful but subjectively they were unhappy. This is probably due to inappropriate expectations of the surgery. Patients should be made aware that there is an excellent chance of curing their instability but a much more variable chance of their returning to normal activities, particularly if they participate in high risk occupations and/or sports.

Some advocate early repair following the first episode whereas other surgeons prefer a trial of conservative management and only operate in recurrent cases. Spatschil et al [21] found that with increasing numbers of dislocations there were significantly higher rates of glenohumeral ligament injuries and Hill-Sachs lesions. Both of these have been found to be associated with a greater risk of recurrence following arthroscopic repair [3] suggesting that early surgery would be beneficial if considering a minimally invasive procedure. In this study, all patients had dislocated at least twice prior to surgery. The mean number of dislocations was 6 (range 2 to 12). One patient was excluded from this calculation as he had had too many dislocations to remember. This patient had recurrent symptomatic instability following surgery. Both patients who had recurrence of symptoms had >5 dislocations prior to surgery.

Boileau et al [3] has identified several risks for recurrence following arthroscopic Bankart repair. Capsular hyperlaxity and a large Hill-Sachs lesion involving >25% of the gleniod articular surface were 2 factors increasing this risk significantly. Out of the 2 patients who experienced recurrence in our study, 1 had a Hill-Sachs lesion which was estimated as involving approximately 20% of the gleniod at arthroscopy. A further risk for recurrence is a bony component to the Bankart lesion. A compression fracture of the anterior glenoid rim greatly increases recurrence rates. Given that the bony anatomy can have such a large affect on outcome, a strong case for pre-operative CT evaluation can be made. One factor which did not appear to be linked with recurrence in our study was the number of anchors used. An average of 2.7 (2 to 4) anchors were used and of the patients with recurrence, 2 and 3 anchors were used respectively. Only 2 patients had 4 anchors placed. The number of anchors used varied on a case by case basis depending on the size of the labral defect to be repaired and the space available on the glenoid rim. It was felt that 2 or 3 bioknotless anchors were ample to affect a sufficiently robust repair of the capsulolabral complex until healing occurs.

This study shows that the early results of bioknotless suture anchors are in line with similar studies using other suture anchor methods [22, 23]. 80% of the patients were satisfied with their surgery and had good or excellent post-operative results with constant scores returning to 84% of that measured in their normal arm. The overall recurrence rate of dislocation was 7% and of instability was 7%. It is further confirmation that new generation arthroscopic methods have similar success rates to open techniques.