Facial nerve regeneration through autologous nerve grafts: a clinical and experimental study.

Human facial nerve regeneration was compared in two systems: autologous neural cable grafts (N = 56) and direct end-to-end anastomosis (N = 34). The data were analyzed based on the following criteria: facial symmetry and muscle tone at rest, degree of total facial reinnervation, voluntary motion, synkinesis, and electrophysiologic testing. Facial tone and symmetry at rest, and electrophysiologic tests were similar, while voluntary motion and facial reinnervation were decreased and synkinesis increased in the autologous cable graft repairs. Electrophysiologic tests failed to distinguish between the two repair methods in successful regenerates. Neural cable grafts were associated with more mass movement and less fine precise emotive facial movement. Excellent cable graft repair results diminish with time because of continual progressive synkinesis, which may take up to 4 years to develop. Eight-millimeter neural cable graft repairs of the buccal division of the facial nerve in rabbits (N = 8) demonstrate, at 5 weeks, equal myelinated and more unmyelinated axons than in the normal nerve controls. Neural cable graft repairs demonstrate two features not found in the normal nerve: 1. myelin debris and 2. two subpopulations of axons--extrafascicular and intrafascicular. Myelin debris predisposes towards extrafascicular axon regeneration. Extrafascicular axons are the major neurite regenerates (63%). Intrafascicular axons are the major reinnervators of the distal transected nerve stump. Electrophysiologic studies demonstrate decreased conduction velocities which are related to the number and size of myelinated axonal regenerates. The implications for neurite reinnervation of the autologous cable graft and their role in synkinesis are discussed.