Does arthroscopic rotator cuff repair actually heal? Anatomic evaluation with magnetic resonance arthrography at minimum 2 years follow-up

Purpose al of this series was to evaluate the clinical and anatomical outcomes of all-arthroscopic rotator cuff tears repair at a mid-term follow-up, using MR arthrography in order to assess tendon-to-bone healing. als and methods etrospective study included 29 patients (31 shoulders) presenting, according to Cofield classification, a small or moderate-sized supraspinatus full thickness tear with (7/31) or without (24/31) infraspinatus extension. The mean size of the tear was 2.64 ± 1.61 cm. The Constant score was used for pre- and postoperative clinical evaluation. All tendons were repaired under arthroscopic control. A single row technique was used. Biceps tenotomy and subacromial decompression were systematically performed. All patients operated arm were immobilized in a sling for 4 weeks and full activity was allowed at 6 months. At last follow-up, a rotator cuff MR arthrography was performed by an independent radiologist to evaluate the anatomical status of repair. s an follow-up was 49.4 ± 17.3 months. Sixteen patients (17 shoulders) had a rotator cuff MR arthrography. Mean Constant score at last follow-up was 82.3 ± 12.4, with a mean improvement of 24%. Eighty-eight percent of repairs (15/17) showed a small or a large leakage at the MR arthrography. There was no significant correlation between the clinical and anatomical outcomes. sion terest of this series is to show, at a mid-term follow-up and using an invasive imaging technique, the low rate of tendon-to-bone healing in arthroscopic rotator cuff repair but with a minimal influence on clinical outcome.