Acute facial nerve injuries involving the total facial nerve (n = 202) and its segmental branches (n = 63) were repaired with a variety of neural (n = 225) and myofascial transfer (n = 40). A system for evaluating results based on facial symmetry and tone at rest, recovery of voluntary mimetic activity, synkinesis, and recovery of selective function in discrete facial nerve divisions is presented. The best results were achieved with immediate direct end-to-end neural-epineural anastomotic repairs. The least favorable results were seen with myofascial transpositions and long extratemporal rerouted autologous nerve grafts. A series of tissue culture and animal experiments were used to examine structural changes in neural repair. The requirements for successful motoneuron regeneration, causes of synkinesis and methods for nerve stump coaptation are elucidated. Tubulation experiments demonstrated the need for neurotrophic and neuropromotive factors. Problems associated with the use of autologous nerve grafts are discussed.

Antecedent priming lesions do not enhance axonal survival as determined by regenerating myelinated axonal counts. However, antecedent injuries enhance the efficiency of neural innervation of the affected mimetic musculature by increasing the number of myelinated intrafascicular neural regenerants in the cable graft and distal nerve stump. This is accomplished by two factors: increased perineural fibrosis and decreased intrafascicular myelin and axonal debris.

Peripheral neural regeneration, over a 10-mm transectional gap, was determined in 70 rabbit buccal divisions of the facial nerve using two entubational systems (semipermeable and impermeable silicone chambers) prefilled with three natural occurring media (serum, blood, and saline) during a 5-week period. The number of myelinated axonal regenerates at the midchamber and at 2 mm in the distal transected neural stump were counted in each group and compared to pooled myelinated axonal counts in 9 normal rabbit buccal divisions of the facial nerve. Semipermeable porous chambers had an overall greater regeneration success rate (75% vs. 42.8%) and regained, on the average, a higher number of myelinated axons (51.4% vs. 26.1%) than silicone chamber regenerates. Semipermeable chambers prefilled with serum or blood had significantly higher regeneration success rates, myelinated axonal counts, and percentages of neural innervation of the distal transected neural stump. Both entubational systems produced similar axonal counts with intraluminal saline. The highest overall success rate (93.7%) and average number of myelinated axons per chamber (3072) were achieved in semipermeable chambers prefilled with serum. The greatest variability in myelinated axonal counts (0 to 3266 axons) and percentage of distal stump innervation (5.5% to 98.1%) was seen in silicone chambers filled with saline. The percentage of myelinated axons from the midchamber that innervated the distal stump was greater in semipermeable chambers with blood (73%) and serum (54%) than in silicone saline chambers (43%). On the average, the distal stumps from semipermeable chambers filled with serum (47%) and blood (33.5%) regained a higher percentage of normal myelinated axonal counts than silicone-saline chambers (12.5%). These results suggest that both the construction of entubational chamber and the intraluminal medium can have significant influence on neurite regeneration. Semipermeable chambers prefilled with serum have a strong neurite-promoting potential in peripheral neural regeneration of rabbit facial nerves.

Thrombin and serine protease inhibitors such as protease nexin-1 (PN-1) have been implicated in neurite outgrowth activity. We compared rat sciatic nerve regeneration in 10-mm silicone tubes, bridging an 8-mm nerve gap, that were prefilled with thrombin (1.5 IU) and PN-1 (50 micrograms/mL or 1 mg/mL) to those filled with saline solution (control). Neural regeneration and fibrinoid matrix progression (deposition of extracellular matrix) were analyzed at 1, 4, 17, and 21 days after silicone tube implantation. At 1 and 4 days after implantation, thrombin reduced fibrinoid matrix length propagation from both the proximal and distal transected nerve stumps, but PN-1 and saline did not interfere with matrix progression (p < .05). Seventeen days after implantation, the number of silicone tubes containing myelinated neuronal regenerates at the mid-tube region was 1 of 6 for thrombin, 6 of 9 for PN-1, and 6 of 10 for saline solution. Twenty-one days after implantation, 11 of 11 tubes with saline solution, 9 of 11 with PN-1 at 1 mg/mL, and 7 of 9 with PN-1 at 50 micrograms/mL had myelinated neural regenerates in the mid-silicone tube region, while only 2 of 9 thrombin-containing silicone tubes contained myelinated axons. There was no statistically significant difference in myelinated neurite regenerates at 17 and 21 days after implantation among silicone tubes prefilled with saline solution and PN-1 (50 micrograms/mL or 1 mg/mL). Thrombin interfered with matrix progression and significantly reduced the number of myelinated neurite regenerates (p = .01). The PN-1 and saline solutions did not inhibit matrix progression or affect the number of myelinated axonal regenerates (p = .92).

No abstract