2015 ISAKOS Biennial Congress ePoster #2417

Primary Vs. Revision Arthroscopic Rotator Cuff Repair: An Analysis in 360 Consecutive Patients

Aminudin M. Shamsudin, MD, MS(Orth), MMed(Orth), MASurg, Selangor MALAYSIA
Patrick H. Lam, PhD, Sydney, NSW AUSTRALIA
Karin Peters, MD, Sydney, NSW AUSTRALIA
Imants Rubenis, MD, Sydney, New South Wales AUSTRALIA
Lisa Hackett, AMS, Sydney, NSW AUSTRALIA
George A. Murrell, MD, PhD, Sydney, NSW AUSTRALIA

St George Hospital Campus, UNSW, Sydney, NSW, AUSTRALIA

FDA Status Not Applicable

Summary: At 2 years patients who had revision rotator cuff repair were twice as likely to have re-torn compared to those undergoing primary repair




The aim of this study was to evaluate the outcome of revision arthroscopic rotator cuff surgery when compared with primary arthroscopic rotator cuff surgery in a large cohort of patients.


A consecutive series of 360 (310 primary and 50 revision) arthroscopic rotator cuff repairs performed by a single surgeon with minimum of two years follow-up were retrospectively reviewed using prospectively collected data. With the 50 revision cases (revision group) as a reference, three primary repair cases (primary group) were chosen immediately before and three after each revision case. Standardized, patient-ranked outcomes, examiner determined assessments, and ultrasound determined rotator cuff integrity was assessed pre-operatively at six months and two years after surgery.


The revision group were older (mean: 63 years, range 43 - 80) compared to the primary group (mean: 60 years, range 18 – 88) (p < 0.05) and had larger tear size (4.1 cm2 ± 0.5 cm2)(mean ± SEM) compared to the primary group (3.0 cm2 ± 0.2 cm2)(p < 0.05). Two years after surgery the primary group reported less pain at rest (p < 0.02), during sleep (p < 0.05) and with overhead activity (p < 0.01) compared to the revision group. The primary group had better passive forward flexion (+13o, p<0.05), abduction (+18o, p < 0.01), internal rotation (+2 vertebral levels, p < 0.001) and also significantly greater abduction strength (+15 N, p < 0.001), lift-off strength (+9.3 N, p < 0.05) and adduction strength (+22 N, p < 0.01) compared to the revision group at two years. When compared to the primary group, the revision group was less satisfied with the overall shoulder function before surgery ( p < 0.001 ) but was equally satisfied with the primary group at 6 months post operatively. However the revision group’s satisfaction deteriorated and were less satisfied with their shoulder function than the primary group at two years (p < 0.005). The re-tear rate for primary rotator cuff repair was 16% at 6 months and 21% at two years; while the re-tear rate for revision rotator cuff repair was 28% at six months and deteriorated to 40% at two years (p < 0.05). The overall shoulder satisfaction corresponded with a decline in rotator cuff integrity at two years.


The short term clinical outcomes of patients undergoing revision rotator cuff repair were similar to primary rotator cuff repair. However, these results did not persist and by two years patients who had revision rotator cuff repair were twice as likely to have re-torn compared to those undergoing primary repair. The increase re-tear rate in the revision group at two years was associated with increased pain, impaired overhead function, weaker strength, less passive motion excursion and less overall satisfaction with shoulder function.