Acta Neurochir (Wien) (2002) 144: 327–335 Acta Neurochirurgica > Springer-Verlag 2002 Printedin Austria Restoration of Upper Arm Function in Traction Injuries to the Brachial Plexus M. Samardz ˇic´, D. Grujic ˇic´, L. Rasulic ´, and B. Milic ˇic´ Institute of Neurosurgery, Clinical Center of Serbia, Belgrade, Yugoslavia Summary Background. Restoration of upper arm function presents the main priority in nerve repair of brachial plexus traction injuries. The re- sults are predominantly influenced by the level and extent of injury, and the type of surgical procedure. The purpose of this study is to evaluate influence of these factors on final outcome. Methods. Study included 91 surgically treated patients, including 71 patients with avulsions of one or more spinal nerve roots and20 with peripheral traction injuries. We performed120 nerve transfers, 25 nerve graftings and 29 neurolyses on di¤erent nerve elements de- pending on the type of nerve damage. Analysis of motor recovery for elbow flexion andarm abduction, isolatedor in combination, was done. Findings. Recovery of elbow flexion was obtainedin 75% nerve transfers, andin 68,7% nerve graftings in peripheral traction injuries. Recovery of arm abduction was obtained in 78,5% nerve transfers, andin 44,4% nerve graftings in peripheral traction injuries. Neu- rolysis was successful in all cases. Generally, the quality of recovery was better for the musculocutaneous nerve. Useful global upper arm function was obtainedin 49,3% of patients with avulsion of spinal nerve roots, andin 55% of patients with peripheral traction injuries. Interpretation. Regarding upper arm function the prognosis of surgically treatedpatients with traction injuriesto the brachial plexus is generally similar in cases with central or peripheral level of injury.However,nerve transfers of collateral branches seem to be superior to nerve grafting andmay be another possibility for repair in cases with extensive nerve gaps. Keywords: Brachial plexus; nerve grafting; nerve transfer; traction injury. Introduction Restoration of upper arm function presents a major aim in brachial plexus surgery. Although the recovery of even minimal function has a great value to patients, as the arm can be usedfor support, useful recovery in- cludes strong and full range elbow flexion, shoulder stabilization, anda certain range of active arm abduc- tion [1, 22]. Unfortunately, the majority of brachial plexus in- juries are causedby stretching andoften result in spi- nal nerve rootavulsion orwidespread longitudinal nerve damage [2]. In these cases results are relatively unsatisfactory due to the limited possibilities for nerve reconstructions including impossibility for direct nerve repair in cases with avulsion of spinal nerve roots, or extensive longitudinal nerve defects demanding long nerve grafts in peripheral injuries [21]. The only possi- bility for nerve repair in the first group of traction in- juries is nerve transfer. Nerve transfers have been at- tempted using a variety of donor nerves but an ideal methodhas not been established. Generally, there are two types of nerve transfer: extraplexal using intercos- tal nerves [5, 9, 11, 23, 24, 28, 33, 35], spinal accessory nerve [1, 26, 28, 33], spinal nerves andmotor branches of the cervical plexus [4, 36],phrenic nerve [10], and intraplexal using proximal spinal nerve stumps or col- lateral motor branches of the brachial plexus in cases of partialavulsion [3, 7, 22,23,24,25,27,28].The choice of donor nerves is predominantly determined by the type andextent of brachial plexus traction in- jury. The aim of this study was to analyze the results of restoration of elbow flexion andarm abduction, separately andin combination as global upper arm function in relation to the di¤erent types of surgical procedures (nerve transfers, nerve grafting, neurolysis) performedin 91 patients with traction injuries of the brachial plexus. Methods and Patients Patient Population In this study we analyzed a series of 91 patients with brachial plexus traction injuries operatedon between January 1980 andSep- tember 1999. Patient population is divided into two groups: 71 pa- tients with avulsion of one or more spinal nerve roots (group A), and 20 patients with peripheral traction injuries (group B).