Objective: An open procedure has long been the gold standard for the repair of rotator cuff tears. Increasingly arthroscopic repair is taking the place of open surgery. This study is a comparative analysis of open and arthroscopic rotator cuff repair.

Methodology: 113 (57 female, 56 male; mean age 58 years) consecutive patients undergoing rotator cuff repair performed between 2003 and 2006 in one centre were studied prospectively. 92 were arthroscopic (48 female, 44 male; mean age 57 years) whilst 21 used a mini-open technique (9 female, 12 male; mean age 62). The determinant for which procedure was employed was solely clinical, whereby tears greater than 30mm dictated an open procedure. Assessments were made using the Oxford Shoulder Questionnaire (OSQ), DASH and Constant Scores pre-operatively, at three and six post-operative months, and six-monthly thereafter. Correlation coefficients (Pearson’s) were used to analyse and compare the post-operative course for each intervention, and the student’s t test was used to compare the mean scores for each treatment at each time point.

Results: Strong correlation was demonstrated between the rates of recovery with each surgery (Constant r=0.94; DASH r=0.96; OSQ r=0.94). Although the absolute scores were better for the arthroscopic group at each time-point with all assessment tools, these differences were statistically significant only pre-operatively with each score, and at one year with the Constant score.

Conclusion: Whilst the open repair group had poorer scores pre-operatively reflecting an initial difference in tear severity, the rates of improvement were identical with either treatment modality, as were the one year outcomes measured with the OSQ and DASH. This study demonstrates arthroscopic rotator cuff repair to be comparable with open, although we acknowledge that at one year a significant difference was evident with the Constant score.

To investigate the comparability of subjective and objective scores of shoulder function following surgery for rotator cuff pathology. From 2003 to 2006, 333 patients undergoing surgery for rotator cuff disorders were followed prospectively. 220 (66%) underwent solely subacromial decompression, whilst 113 (34%) had additional rotator cuff repair (92 arthroscopic; 21 mini-open). Assessments were made pre-operatively and six-monthly thereafter using the DASH score; Oxford Shoulder Questionnaire (OSQ); and Constant score, which was used as a reference. Standardisation calculations were used to convert all scores to a 0 to 100 scale (100 representing a normal shoulder). The student’s t test was used to compare the mean score for each subjective tool (DASH and OSQ) with the objective score (Constant). Correlation coefficients (Pearson’s) were used to analyse the post-operative course measured with subjective and objective tools for each intervention. Each statistical test was used for all surgeries collectively and the individual surgery types. There was no difference between the mean DASH and Constant scores. A significant difference was seen between the Oxford and Constant scores for at least one time point in each treatment group. Strong correlation was demonstrated between both subjective scores and the Constant. The mean Pearson correlation coefficient comparing the DASH and Constant was 0.96, whilst that for the Oxford and Constant was 0.89. The DASH and Constant scores provided identical results in terms of absolute values at a given time point, and with respect to rates of recovery. The relationship between the Oxford and Constant was less robust. In this study the DASH and Constant scores were indistinguishable, justifying the use of only the former for follow-up, obviating the need for a trained investigator required to perform a Constant score.

To compare the effectivity of arthroscopic and open stabilisation of the shoulder. Between 2003 and 2006, 100 patients (20 female, 80 male; mean age 32 years) undergoing glenohumeral stabilisation were followed prospectively. 28 were open (3 female, 25 male; mean age 30.7 years), 72 arthroscopic (17 female, 55 male; mean age 32.0 years). Assessments were made using the Constant, DASH, and Carter-Rowe (CR) scores, as well as the Oxford Shoulder Instability Questionnaire (OSIQ) pre-operatively, at three and six post-operative months, and six-monthly thereafter. The student’s t test was used to compare the mean scores at each time point. Correlation coefficients (Pearson’s) were used to compare the postoperative course with either intervention. In general the open group performed marginally less well than did the arthroscopic. However, the DASH score demonstrated less consistency both in this relationship, and the rate of post-operative recovery when compared with the other scoring systems. In the open surgery group the DASH revealed a deterioration from the pre-operative score at six months before subsequent improvement; in the arthroscopic group, this deterioration occurred at three months. However, these differences were not statistically significant regardless of the assessment tool employed. Strong correlation was demonstrated between the rates of recovery following either surgery (Constant r=0.99; OSIQ r=1.00; CR r=0.94). Again, this was not supported by the DASH (r= −0.868). The rates of improvement were identical with either treatment when measured with the Constant, OSIQ, and CR, whilst the DASH score yielded inconsistent results. No significant difference could be shown between open and arthroscopic surgery at any individual time point regardless of the assessment tool employed. We suggest that open and arthroscopic surgeries yield very similar outcomes.