Clinical and anatomic results of rotator cuff repair at 10 years depend on tear type.

PURPOSE: Although good short-term and mid-term outcomes are reported for rotator cuff repair, few studies have investigated long-term outcome with clinical and MRI evaluation. The hypothesis was that 10 years following repair of rotator cuff tear, the clinical and anatomic results depend on the extension of the tear.

METHODS: The records of all 965 patients who underwent repair of rotator cuff tears in 2003 were retrieved. The patients were reviewed in 2014 for evaluation at a minimum follow-up of 10 years. A total of 511 patients were evaluated clinically, of whom 397 were also evaluated using MRI. There were 289 isolated supraspinatus tears (SS), 94 tears with posterior extension (P), 92 with anterior extension (A) and 36 with anteroposterior (AP) extension.

RESULTS: The Constant score had significantly improved from 53.8 ± 14.7 preoperatively to 77.7 ± 12.1 (P < 0.0001) at 10 years, with no significant difference between the four groups. The rate of retear (Sugaya IV, V) was lower in the SS group (19%) and higher in the P (32%) and AP groups (31%). At review, infraspinatus fatty degeneration was significantly greater (Fuchs > 2) in the P (P < 0.001) and AP (P < 0.001) groups and subscapularis fatty degeneration was significantly greater (Fuchs > 2) in the A (P < 0.001) and AP (P < 0.001) groups. The rate of osteoarthritis (Samilson > 2) was significantly higher at 11% (P = 0.001) in the A group. The failure rate was significantly lower (P = 0.044) in the SS group (25%) than the massive rotator cuff tear groups (A, P and AP groups) (35%). Complications occurred in 51 shoulders (10%) and repeat surgery was required in 62 shoulders (12%), with no difference between the four groups.

CONCLUSIONS: The long follow-up period of this study, large series of patients and MRI evaluation of tendon repair allowed us to demonstrate that 10 years following rotator cuff tear repair, between 68 and 81% of tendons had healed. These findings are of value in predicting response to surgical treatment. Tears with posterior extension had a higher risk of retear. However, surgical repair appeared to give a good functional outcome whatever the type of tear, despite the overall rate of complications and repeat surgery.

LEVEL OF EVIDENCE: IV.