Purpose of the study: A rotator cuff tear is a relatively frequent cause of pain and restricted motion of the shoulder. Some orthopaedists believe that any attempt at rotator cuff reconstruction will fail. The aim of this paper is to present our experience with arthroscopic reconstruction of rotator cuff tears.
Material and methods: Between January 1998 and December 2008, 319 patients with an early diagnosis of rotator cuff rupture were treated. The group included 67 women and 252 men, with an average age of 37 years (range, 24 to 71 years) at the time of surgery. The patients indicated for arthroscopic reconstruction had to show free motion of the shoulder, had a full thickness tear up to 3 cm in size in the sagittal plane and a Patte stage 2 tear in the frontal plane at the maximum. The outcome of surgery was evaluated at one year of follow-up and included the patient's self-assessment, modified UCLA score and incidence of complications. The probability of failure was calculated as an odds ratio of an implant failure to failure of the other implants and the probability of repeat surgery in a given implant was calculated as a relative risk in relation to the other implants.
Results: The average operative time was 52 minutes (range, 25 to 85); the average UCLA score increased from 10 to 31 points (p<0.00001). An excellent or a good result was achieved in 80% of the patients. Rotator cuff reconstruction failed in 32 patients (11%), of whom 22 (7.6%) underwent revision surgery. The failure was due to migration of rotator cuff anchors or thread failure in 14 patients (14/32; 44%). The GII anchors showed the highest risk of failure, with the odds ratio of 5.55 (95 % CI, 2.22 to 13.84) for mechanical failure of the method and a relative risk of revision surgery of 7.62 (95% CI, 2.86 to 20.27). For comparison, the RC anchors had the odds ratio for mechanical failure equal to 0.55 (95 % CI, 0.25 to 1.24) and the relative risk of repeat surgery equal to 0.41 (95% CI, 0.12 to 1.43). In addition, 18 complications were recorded. The frequency of deep wound infection was 0.7% (2/319). Six patients (2.1%) required repeat surgery for symptomatic bursitis and adhesive capsulitis.
Discussion: A recent meta-analysis has found no significant difference between the results of surgical rotator cuff reconstruction and its conservative treatment. We do not support this view but present here evidence that, when certain conditions are fulfilled, arthroscopic reconstruction can produce a very good clinical outcome.
Conclusions: The arthroscopic reconstruction of a rotator cuff tears results in a marked relief of pain and improved joint function. An ideal candidate for this treatment should show passive free motion at the shoulder joint, no clinical signs of bursitis, and mobilisable tendon stumps of the torn rotator cuff. In addition, these patients should be highly motivated for post-operative rehabilitation. A suture device was most effective in rotator cuff repair. For good fixation into the bone it is recommended to use special implants that have a minimal risk of dislodgement or anchor thread failure.