ABSTRACT
INTRODUCTION: Diabetic patients have been known to have a higher incidence of frozen shoulder. Arthroscopic capsular release has been described for diabetic frozen shoulder refractory to non-surgical treatment. We describe our single-surgeon experience of this procedure done in Asian patients presenting with diabetic frozen shoulder. RESULTS: Data were prospectively collected from 25 patients from 2007 to 2008, who underwent arthroscopic capsular release for frozen shoulder with diabetes mellitus. This included the University of California-Los Angeles (UCLA) Shoulder Score by an independent observer. Mean follow-up duration was 44.3 ± 22 months with the age of 57.3 ± 10.9 years at the time of surgery. Mean duration of symptoms prior to surgery was 8.9 ± 7.5 months. Ten right shoulders and 15 left shoulders were operated on. Arthroscopic capsular release was performed in the beach chair position with the aid of a 90° electrocautery hook, from the rotator interval to posterior capsule. Significant improvement was observed in the UCLA scores post-operatively (preoperative total score 15.8 ± 3.6 vs. post-operative 33.3 ± 2.5, p < 0.00), including all domains of pain (3.9 ± 1.8 vs. 9.4 ± 0.9, p < 0.00), function (4.8 ± 1.7 vs. 9.4 ± 1.4, p < 0.00), active forward flexion (2.6 ± 0.7 vs. 4.6 ± 0.6, p < 0.00) and satisfaction (0.0 ± 0.0 vs. 5.0 ± 0.0, p < 0.00). No significant correlation between duration of symptoms and UCLA scores. Significant improvement was observed in forward flexion (preoperative 100.4 ± 18.8° vs. post-operative 157.6 ± 15.9°, p < 0.00) and external rotation (preoperative 14.4 ± 6.34° vs. post-operative 63.8 ± 8.3°, p < 0.00). Although the duration of symptoms correlated significantly with preoperative forward flexion ( r = 0.443 and p < 0.027), there was no significant correlation with post-operative forward flexion ( r = 0.023 and p < 0.91). No axillary nerve palsy observed. CONCLUSION: The diabetic frozen shoulder can be treated favourably with arthroscopic capsular release and the duration of symptoms prior to surgery does not affect the outcomes.
