Purpose: Functional outcome after shoulder arthrodesis was evaluated to assess indications for the treatment of posttraumatic partial and total brachial plexus paralysis in adults.

Material and methods: Twenty-seven patients who underwent glenohumeral arthrodesis for posttraumatic brachial plexus paralysis were reviewed. Eleven had radicular paralysis (C5, C6 and C5, C6, C7) and sixteen total paralysis. All patients recovered active elbow flexion. Shoulder reinnervation had failed in eleven patients. Before the arthrodesis, 22 patients could no use their paralysed limb. Mean time between direct neurological surgery and arthrodesis was 30 months for partial paralyses and 20 months for total paralyses. Glenohumeal screw fixation was used for the arthrodesis which was associated with an external fixation in 21.

Results: Mean postoperative follow-up was 70 months. There were two cases of non-union which fused after revision and three cases of humerus fracture which occurred during the first six months after surgery. Pain related to inferior subluxation improved in six patients. There was no significant difference between the two groups for position of the fusion, or postoperative active motion (60° flexion, 60° abduction, 45° internal rotation and 7 to −9° external rotation). There was a significant difference in force which was greater for superior paralyses (11 kgf versus 7 kgf in flexion, 12 kfg versus 7 kgf in abduction, 6 kgf versus 2 kgf in external rotation and 11 kgf versus 4 kgf in internal rotation). The same was true for hand movement. The differences were statistically correlated with force of the pectoralis major.

Conclusion: Glenohumeral arthrodesis provides significant improvement in function in patients with supraclavicular brachial plexus paralysis, even with a paralytic hand. Arthrodesis also allows reorienting surgical reinnervation to other functions such as hand movement. Shoulder force and hand movement are directly correlated with force of the pectoralis major.