Purpose: The purpose of this study is to examine the rate of Buford complex in a consecutive series of patients with arthroscopic Bankart repair and to evaluate the postoperative outcome compared to a group of patients without this anatomical variation.

Material and Method: The surgical findings of 138 consecutive shoulder arthroscopies for Bankart lesions were prospective evaluated between January 2005 and January 2007; the mean of age was 25 years old (range 14–52). The criteria used to establish the diagnosis of Buford Complex included: cordlike middle glenohumeral ligament inserted in the superior labrum and attached to the biceps tendon, and complete absence of anterosuperior labrum.

Postoperative outcomes were evaluated with Rowe score.

Results: Five out of 138 patients (3.6%) presented a Buford complex associated to a Bankart lesion. Four of these patients (80%) presented a type II SLAP lesion. The follow up was 28 months (range 19–40 months). In the group of patients with the Buford complex the Rowe score had a mean of 75 points (range:45–95), 2 patients (40%) presented re-dislocation whereas in the other group the Rowe score had a mean 88 points (range: 25–100) and 9 cases (6.5%) had re-dislocation (P< 0.05).

Conclusion: The rate of complex of Buford associated to Bankart lesion was of 3.6%. The presence of complex of Buford in our series was associated to worse functional postoperative outcomes.