Purpose of the study: This retrospective analysis was conducted to study the gain provided by a latissimus dorsi flap used as first-intention treatment (group 1) or secondary treatment after prior failure (group 2) for irreparable rotator cuff tears.

Material and methods: This series included ten women and nine men, mean age 58 years (range 42-64). The initial tear was a massive (> 5 cm) posterosuperior tear in 16 patients and extended to the upper third of the infra-scapularis in three. Surgery was undertaken because of persistent pain and limited joint motion despite rehabilitation. A subacromial impingement was noted in 15/20 shoulders on the arthroscan and fatty degeneration was noted as grade 3.31 on average for the supraspinatus and 3.1 for the infraspinatus (Goutallier and Bernageau classification). The latissimus dorsi flap was harvested via the superolateral approach and fixed with anchors in the superior border of the infrascapularis and on the trochiter after avivement. Tendon stumps were sutured to the medial part of the aproneurotic sheath of the latissimus dorsi.

Results: Overall outcome and outcome in group 1 patients (14 shoulders) and group 2 patients (five shoulders) were noted. Mean follow-up was 19.72 months (range 12–48). The overall Constant score progressed from 33.10 to 54.9 with a mean gain of 53° for elevation (98–151°) and 11° for external rotation (21.5–32.1°). For group 1, the Constant score progressed from 31 (15/51) to 58 (40/75) with a mean gain of 37° elevation (121–155°) and 13° external rotation (22.8–35°). For group 2, the Constant score progressed from 33 to 52 (40/75) with a mean gain of 32° elevation (88–120°) and 6° external rotation (18–24°). Pain improved from 6.3 to 11.8 on the Constant score.

Discussion and conclusion: Used as a first intention treatment for massive irreparable cuff tears with fatty degeneration scored greater than grade 3, the latissimus dorsi flap provides better results than when it is used after failure of a prior procedure. Results are good for pain relief and active elevation (45°) but modest for external rotation (6–13°) and zero for force. The two failures and the two cases of only fair subjective outcome were in group 2. We reserve the procedure for painful pseudo-paralytic shoulders in subjects aimed less than 60 years who do not respond to prolonged rehabilitation.