Surgical interventions for the treatment of anterior shoulder instability

Shoulder instability represents a spectrum from micro-instability through subluxation (partial dislocation) to frank dislocation and may occur as a result of different underlying pathologies. Anterior shoulder instability occurs when the humeral head moves out of its shallow joint socket in forward direction, usually following an injury such as a fall on the outstretched hand. Dislocation occurs when the head of the humerus moves completely out of the socket. As it does so, various ligaments around the joint may be torn and the bone of the humeral head or of the socket may be damaged. Recurrent episodes of dislocation are common unless complete healing of the damaged structures occurs. A 'Bankart lesion' is a type of tear involving the lining of the socket part of the shoulder joint. Such a tear predisposes the shoulder dislocate easily and in a recurrent manner. If recurrent episodes are troublesome, surgical repair using an open or minimally invasive (keyhole) technique may be advised and this principally involves the repair of 'Bankart lesion'.

This systematic review compared different techniques of surgical repair for anterior shoulder instability. Only three randomised controlled trials, involving a total of 184 people with anterior shoulder instability that usually followed a traumatic event, are included in the review. All three trials compared arthroscopic (key hole) surgery with open surgery, generally involving the repair of Bankart lesions. All three trials were inadequately reported but appeared well-conducted with minimum follow-ups of two years.

The limited data available showed no statistically significant differences between the two groups in recurrent instability or re-injury, in subsequent instability-related surgery or surgery for all reasons. Data for other outcomes, including shoulder function, also showed no significance differences between the two groups. In all the available evidence was insufficient to draw conclusions and further well designed randomised controlled trials are required.

Authors' conclusions: 

There is insufficient evidence from randomised trials comparing arthroscopic with open surgery for treating anterior shoulder instability. Further research is needed on this subject and for other surgical interventions. Sufficiently powered, good quality, well reported randomised controlled trials with validated outcome measures and long-term follow up are required.

Read the full abstract...
Background: 

The shoulder is the most common joint to develop recurrent instability. Repair of labral tears of the joint and reconstruction of damaged capsule and torn ligaments either by open or arthroscopic methods remain the cornerstone of current management.

Objectives: 

To compare the effectiveness of various surgical interventions performed to treat recurrent anterior instability of the shoulder in adults.

Search strategy: 

We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (December 2007), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2008, Issue 1), MEDLINE (1950 to March 2008), EMBASE and other databases. We searched conference proceedings and the reference lists of papers.

Selection criteria: 

Randomised or quasi-randomised controlled trials comparing different surgical interventions for treating anterior shoulder instability in adults.

Data collection and analysis: 

The authors independently selected trials, assessed methodological quality and extracted data. Only limited pooling was done.

Main results: 

Included are three randomised controlled trials involving 184 people (predominantly active young men) with unidirectional anterior shoulder instability generally following a traumatic event. All three trials compared arthroscopic versus open surgery, generally involving the repair of Bankart lesions. The three trials were inadequately reported but appeared well-conducted with minimum follow-ups of two years.

Pooled results showed no statistically significant difference between the two groups in recurrent instability or re-injury (7/92 versus 5/85, risk ratio (RR) 0.89, 95% confidence interval (CI) 0.09 to 8.72; random-effects model), in subsequent instability-related surgery (RR 0.66, 95% CI 0.05 to 8.97; random-effects model) or surgery for all reasons (RR 0.55, 95% CI 0.04 to 7.18; random-effects model). For other outcomes, including shoulder function, there were either no statistically significant differences between the two groups or the differences were clinically insignificant where statistically significant differences occurred.