OPEN VERSUS ARTHROSCOPIC BANKART REPAIR FOR TRAUMATIC ANTERIOR SHOULDER INSTABILITY: 1 TO 5 YEAR RESULTS

Abstract

Purpose: Although arthroscopic Bankart repair is yet an established procedure in the treatment of traumatic anterior shoulder instability, it is still not known whether it reproduces the good results of the open repair procedure. Aim of this investigation was to compare the functional and subjective outcome between open and arthroscopic Bankart repair.

Material and methods: A retrospective cross sectional study design on the comparison of open and arthoscopic Bankart treatments between 1995 and 2004 was implemented at the Dresden Orthopedic Surgery Department. In this period a total of 223 patients underwent surgery due to anterior shoulder instability, among which 212 patients had posttraumatic instability. A diagnostic arthroscopy was performed in all patients, but only 40 patients with intact capsulolabral complex and without capsular laxity were treated arthroscopically, the remaining 183 patients underwent an open Bankart procedure.

A total of 186 patients with posttraumatic anterior instability could be clinically re-examined within 1 to 5 years after initial surgery, among which 147 patients underwent an open and 39 patients an arthroscopic Bankart procedure. The median age of this sample was 27 years (interquartile range 21 – 37 years) at initial surgery, 21% of these patients were female.

Results: After open surgery 11 of 147 patients (8%) and 6 of 39 (15%) after arthroscopic surgery reported one or more re-dislocation after initial treatment, which occurred after a mean dislocation free time of 62 versus 40 months, respectively (95% confidence intervals 59 – 65 versus 36 – 44 months). Both a univariate analysis (Logrank test p=0.012) and a multivariate Cox regression analysis (Likelihood Ratio p=0.023) confirmed a statistically significant difference in the time between initial surgery to first reluxation. 4 of the 11 re-dislocations after open and 1 of 6 after arthroscopic surgery occurred after a new accident. That makes a re-dislocation rate without new adequate trauma of 5% after open and 13% after arthroscopic treatment.

ROM showed no difference between open and arthroscopic Bankart procedure for abduction and a mild difference for external rotation. 21 of 115 (18%) patients had an external rotation lag of 20° or more after open surgery versus 1 of 34 (3%) after arthroscopic treatment (Fisher p=0,027). The Rowe score demonstrated “good” or “excellent” functional results in 102 of 117 (87%) patients versus 28 of 35 (80%) patients after open versus arthroscopic treatment (Fisher p=0,285).

Conclusion: In this sample arthroscopic Bankart repair demonstrated more frequently and significantly earlier re-dislocation after initial surgery than did the open treatment alternative. External rotation lags of at least 20° occured more frequently after the open procedure.