An Analysis of Capsular Area in Patients with Anterior, Posterior, and Multidirectional Shoulder Instability

Abstract

Background Although increased capsular volume has been implicated in shoulder instability, there is a paucity of clinical evidence to quantify the size of the capsule with specific instability conditions of the shoulder. Hypothesis Shoulder capsular area, as measured by magnetic resonance arthrography, is increased with specific patterns of shoulder instability. Study Design Cross-sectional study; Level of evidence, 4. Methods During an 8-month period, all patients with a diagnosis of anterior (n = 19), posterior (n = 14), or multidirectional (n = 13) instability of the shoulder and who were assessed with a magnetic resonance arthrogram were reviewed. A group of 10 control patients without clinical instability were also identified. The magnetic resonance arthrograms of all groups were randomly mixed, and 5 reviewers recorded measures of capsular length and area and determined labral abnormalities. The magnetic resonance arthrogram measurements were compared between groups, and interobserver agreement was determined. Results The cross-sectional area of the capsule was increased in patients with posterior ( P = .017) or multidirectional instability ( P = .021) versus controls, but not in patients with anterior instability. Additionally, the posteroinferior cross-sectional area was increased in patients with posterior ( P = .001), multidirectional ( P = .003), and anterior ( P = .008) instability. In patients with a posterior labral tear, the mean axial ( P = .043) and mean posteroinferior sagittal cross-sectional area ( P = .011) was increased, but there were no differences in cross-sectional area for those with an anterior labral tear. The overall interobserver reliability was very good (correlation coefficient range, 0.68–0.94). Conclusion Our results reinforce the concept that capsular elongation and laxity, either preexisting or acquired, play a role in certain instability conditions of the shoulder. Additional work is needed to determine how to correlate surgical decision making with the cross-sectional area measurements demonstrated in this study.