Little is known about shoulder hypertonia in patients with brain injury. The principal aim of our study was to assess the results of spastic shoulder surgery in our team experience.

Sixteen patients (18 shoulders) were retrospectively reviewed. The following were studied: etiology of hypertonia, shoulder deformation schemes, initial aims of surgery, shoulder pre- and postoperative mobility, impact of surgery on limb function when possible. None limb was functional preoperatively. Spontaneously, shoulders were all in a medial rotation, in adduction in 15 cases, in abduction in 3 cases. One had spontaneous retroposition. Therapeutic aim was hygienic in all and pain-relief in 6. Ten neurotomies of the lateral pectoral nerve (LPN), 8 intramuscular lengthening of the latissimus dorsi (LD), 9 percutaneous tenotomies of the pectoralis major (PM), one neurotomy of the motor nerve for the long head of triceps brachii (LHTB) and one intramuscular lengthening of the teres major (TM) were performed.

Mean age at surgery was 52.1 (24–73). Etiology was vascular in 11, anoxic in 2, one had traumatic brain injury, one multiple sclerosis, one cervicarthrosic myelopathy. Preoperatively, the mean results for passive abduction of the shoulder was 50.6° (10–90) and for the passive lateral rotation was −5° (−70; 20). Postoperatively, there were respectively improved at 87.7° (60–160) and 6.7° (−20; 20). Only 3 muscles presented with residual tightness (two PM and 1TM) in patients where an isolated neurotomy of the lateral pectoral nerve (LPN) had been achieved. All of the initial objectives of the surgery were satisfied. One patient recovered a partial functionnal limb thanks to a recovered active shoulder flexion.

In classical medial rotation-adduction deformation schemes, surgery performed on the 3 main involved muscles (PM, LD, TM) offers good results in term of passive mobility of the shoulder and pain-relief.