Introduction: Any surgical nerve reconstruction must take into account amount of individual nerve fibres at any given level of injury. To date, however, literature on qualitative and quantitative assessment of motor axons of the peripheral nerves of the upper extremity is scarce. Furthermore, none of these studies have depicted the topography of motor fibres along the entire course of these peripheral nerves. The aim of the present study is to present the total number of motor fibres of the brachial plexus from each root down to the level of its corresponding branches.

Materials and Methods: Nerve samples have been harvested from 12 organ donors immediately after death. From 8 incisions ranging from the neck to the wrist a total of 36 nerve samples were gained per organ donor. A special immunohistochemical protocol was applied to visualize the specific structure of interest within the nerve cross section. Antibody against neurofilament served to determine the total amount of myelinated and unmyelinated axons. Antibody against choline acetyltranferase (ChAT) was used to detect cholinergic/motor fibres. Histology sections were then scanned and evaluated with a digital software program to allow quantification of each cross section. These numbers were cross checked in an animal model with standard retrograde tracing methods. Finally, the quality of this method was also cross checked with staining ventral and dorsal roots of organ donors at spinal cord level.

Results: As expected the majority of any given peripheral nerve contains afferent fibers. To our surprise, however, only around 10% of all axons in a mixed peripheral nerve are efferent fibers. In a “pure” peripheral motor nerve (thoracodorsal nerve) one third of the axons are cholinergic. In a pure cranial motor nerve the motor portion rises to about 60%(accessory nerve) but still has a significant afferent fibre population. The control experiments in a rodent animal model show good correlation between retrogradely labelled motor neurons with ChAT positive labels in the peripheral nerve section.

Conclusions: Here we present for the first time a quantitative analysis of all afferent and efferent fibres of the brachial plexus and its consecutive nerves. The surprising finding is that even “pure” motor nerves with a suspected high number of motor fibres (thoracodorsal nerve) only have a relatively small number of efferents. Since this ratio is relatively constant for motor nerves at different levels of the extremity these results challenge the traditional view of fiber distribution and innervation density in man.