The Effect of Arthroscopic Extra‐Articular Entire Coracohumeral Ligament Release for Patients with Recalcitrant Frozen Shoulder

OBJECTIVE: The thickened coracohumeral ligament (CHL) is an important part of the typical manifestations and magnetic resonance imaging of frozen shoulder. However, only a few clinical studies with limited cases on arthroscopic extra‐articular entire CHL release exist in the literature. This study was to evaluate the effect of arthroscopic extra‐articular entire CHL release for patients with recalcitrant frozen shoulder. METHODS: From February 2014 to February 2020, 81 cases of recalcitrant frozen shoulder patients treated with surgery in a single‐center shoulder department and followed for more than 2 years were analyzed. Arthroscopic 360° capsular release was performed with intra‐articular partial release (IPR group) or additional extra‐articular entire release (IPR + EER group) of CHL. The same rehabilitation program was performed after surgery in both groups. Visual analogue scale (VAS) for pain, range of motion (ROM), and the Constant–Murley scoring system was evaluated before operation, at 3 months after operation, 6 months after operation, and the final follow‐up. T‐test, Mann–Whitney U‐test and chi‐squared test were used to compared data. RESULTS: There were 39 patients in the IPR group, with an average follow‐up of 29.2 months. A total of Forty‐two patients in the IPR + EER group completed a mean follow‐up of 25.7 months. All incisions healed in stages. There were significant differences in Constant–Murley shoulder score, VAS score, and ROM before operation and at the final follow‐up in both groups (both P < 0.001). The VAS score of the IPR + EER group was lower than that of the IPR group at 3 months after surgery (P < 0.05), and 6 months after operation (P < 0.05). External rotation, internal rotation, and abduction of ROMs and Constant–Murley shoulder score were significantly greater in the IPR + EER group at 3 months (P < 0.001, P < 0.05, P < 0.001, P < 0.05, respectively) and 6 months after operation (P < 0.001, P < 0.05, P < 0.001, P < 0.05, respectively). At the last follow‐up, there was no significant difference in forward flexion, internal rotation, and abduction of ROMs, VAS, and the Constant–Murley shoulder score between the IPR and IPR + EER groups. The external rotation of the IPR + EER group was still greater than that of the IPR group at the last follow‐up (P < 0.001). CONCLUSION: Arthroscopic extra‐articular entire coracohumeral ligament release could solve early pain of shoulder joint, recover shoulder joint functions effectively, and achieve a satisfactory efficacy in the treatment of recalcitrant frozen shoulder.