Is rotator cuff repair appropriate in patients older than 60 years of age? Prospective, randomised trial in 103 patients with a mean four-year follow-up
These authors compared surgical rotator cuff repair to acromioplasty and biceps tenotomy in patients older than 60 years of age with a mean follow-up of 4 years.
Exclusion criteria included subscapularis tear, spontaneous long head biceps tear, and irreparable tear as determined arthroscopically.
Patients were randomly allocated to acromioplasty and tenotomy (AT group) or to acromioplasty, tenotomy, and tendon suture (CR group). In the repair group tendon suture was consistently achieved using metal anchors inserted into the tip of the greater tuberosity after abrasion of the footprint, in a single-row (n = 21) or double-row (n = 33). After surgery patients wore a sling for 4 weeks with
passive self-rehabilitation on day 1. After 4 weeks, physiotherapy sessions were prescribed if needed.
Of 130 initially included patients older than 60 years of age and having rotator cuff tears deemed amenable to surgical repair, 103 (79%) were evaluated after a mean of 4 years.
The tear was distal in 41 patients, intermediate in 40, and retracted in 22.
The complication rate was 4%. The mean Constant Score was 44 preoperatively; values after 4 years were 76 overall. The repair group had slightly greater Constant Scores (78) than the acromioplasty group (73). Less than 2/3 rds of the repairs (63%) were healed by sonography.
The Constant Score was significantly better when tendon healing was achieved (82/73, P < 0.001). In the AT group, the acromio-humeral distance was significantly smaller (6.9 mm/7.8 mm, P = 0.03) and eccentric humeral head position was more common (44%/26%,P = 0.01).
It would have been interesting to know how many of the shoulders with cuff tears examined arthroscopically were deemed ‘irreparable’ and how this determination was made. Because the constant score includes a strength component, it is expected that shoulders with healed repairs would score better than those with unhealed or unrepaired tears. It is interesting to note, however, that the difference between repair and non-repair reported in this article did not reach the level of the minimal clinically important difference for rotator cuff tears (=10).
Finally it is of note that acromioplasty sacrifices part of the coracoacromial arch and may not be benign for unrepaired cuffs. For that reason, we prefer thesmooth and move procedure when cuff repair is not possible or desirable.