Facial nerve injury and microsurgical repair : Experimental and clinical studies

Abstract: Facial palsy is a relatively common clinical condition with a variety of causes. Irrespective of its etiology, facial palsy always represents a very serious problem for the patient. This underlines the need for more effective treatment procedures. Retrospective evaluation of a clinical material of 16 patients with facial palsy treated at the University Hospital of Linköping during the period 1990-2000 showed that to improve the results of microsurgical nerve repair experimental research - controlled studies on homogeneous materials - is imperative.To produce relevant experimental data we used a rat model. Dissections showed that the mandibular branch (MB) of the rat facial nerve is suitable for experimental studies. Electron rnicroscopy revealed that the normal rat MB contains some 2,200 axons, 1,825 of which are myelinated and show a unimodal size distribution with a mode at 4.5 µm. It was also found that the normal rat MB contains myelinated and unmyelinated sympathetic axons and that about half the C-fibers in the normal rat MB belong to capsaicin-sensitive putative polymodally nociceptive sensory neurons. Importantly, repair of the MB through transmedian grafting in one stage and in two stages, respectively, had largely similar outcomes in terms of anatomy.These data evoked questions concerning the functional outcome of the two types of repair. Electrophysiological analysis (force recordings andelectromyography) revealed that repair of the rat MB through transmedian grafting in one stage gives a somewhat better functional restoration than repair in two stages.The observations on the rat MB called for experimental studies on the effects of denervation and repair on mimic muscle. We found that the rat dilator naris muscle (DNM) is suitable for that purpose. The normal DNM contains 1,200 fast MyHC fibers with MyHC IIB fibers predominating. It is a very fast contracting muscle without static functions. A brief denervation of the DNM followed by spontaneous reinnervation by the MB did not influence fiber number, had long-term effects on fiber diameter, and had little effect on fiber types. Fiber number, fiber diameter, and occurrence of fiber types in the DNM remained abnormal both after immediate and delayed surgical repair of the MB. Long-term denervation of the DNM had severe effects on qualitative histology, fiber number, fiber diameter and fiber type distribution. Hence, a long delay between facial nerve injury and repair reduces the chances of restoring normal facial muscle function.Conventional cross facial grafting includes division of a facial nerve branch (the donor nerve) on the intact side of the face. This non-optimal situation prompted us to test cross-facial grafting with a less traumatic end-to-side procedure in the rat. After coaptation of a sural nerve graft to the rat MB in an end-to-side fashion, including opening of a perineurial window, axons in the donor nerve emitted myelinated and umnyelinated sprouts into the graft. A predegenerated graft did not stimulate sprouting more than a fresh graft. Retrograde tracing with fast blue and fluoro-ruby showed that many sprouts originate from facial motor axons. Of all traced facial motoneurons 50% projected exclusively into the graft, 45% projected into the graft and the donor MB and 5% projected exclusively into the donor MB. This shows that the end-to- side procedure can be used for cross-facial repair of the rat MB.Altogether these results provide new experimental data, which hopefully will contribute to improvements of the microsurgical treatment of patients with facial palsy.