ORIGINAL ARTICLE Experimental Study on Facial Regeneration With or Without Geniculate Ganglionectomy

Zhengmin Wang, MD; Chunfu Dai, MD, PhD; Yuhai Zhang, MD

Objective: To investigate regeneration of the distal fa- postoperatively for histological examination by elec- cial nerve following nerve grafting within the tympanic tron microscopy. segment with preservation or dis- section. Results: The number of myelinated axons in normal fa- cial was 1819.6±535.6. In group 1, the number Design: Randomized controlled trial. of myelinated axons was 123.6±31.1, and, compared with normal facial nerves, the diameter of the regenerative Subjects: Twenty-three adult New Zealand albino rab- axons was decreased and the sheath thickness in the re- bits were used in this study. generative fiber was diminished. In group 2, the num- ber of myelinated axons was 515.1±103.1, while the my- Interventions: A 2-mm tympanic segment of the elin sheath thickness was proportionate to axon diameter. was removed, and the greater auricular (Data are given as mean±SD.) nerve was harvested for grafting in all animals. In group 1 (10 rabbits), the geniculate ganglion was pre- Conclusion: Geniculate ganglionectomy may improve served. In group 2 (13 rabbits), the geniculate gan- motor axon regeneration. glion was dissected. Mastoidal and extratemporal segments of the facial nerve were harvested 3 months Arch Otolaryngol Head Surg. 2001;127:422-425

EVERAL ATTEMPTS have been that focus on axon regeneration and func- made at morphological and tional recovery following facial nerve in- quantitative analysis of the jury. Nerve regeneration involves a com- geniculate ganglion.1-3 Previ- plex interaction of neurons, Schwann cells, ous studies2 indicated that the elements of the extracellular matrix, and Stotal number of ganglion cells in a single a host of neurotrophic substances. With temporal bone ranged from 589 to 4183 respect to surgical repair, suturing the sev- (mean, 2162 cells). In 88% of patients, ered nerve ends and nerve grafting have most of these cells were found in the ge- remained the procedures of choice.4 The niculate ganglion.2 However, in 8% of pa- immediate neural environment plays an tients, most of these cells were in the in- important role. The success of neural re- ternal acoustic meatus; in 4%, the meatus generation depends on the cellular ma- and geniculate ganglion contained an equal trix components (lamina, type IV colla- number of cells.2 There was no correla- gen, neural adhesive molecular, and tion between total ganglion cell number others) and on neurotrophic factors, such and age or sex of the patient.1 The gan- as nerve growth factor, that are produced glion cell bodies were aggregated at the by the denervated target nerve.5-7 A recent apex of the genu, close to the origin of the study8 showed that electromagnetic stimu- greater superficial petrosal nerve.2 These lation enhances early regeneration and fa- findings suggested a possible therapeutic cial movement. Most investigations have benefit from geniculate ganglionectomy in focused on motor fiber regeneration and patients with facial paralysis. facial movement. From the Department of Relative to other , the It is poorly understood how the se- Otolaryngology, Eye, Ear, facial nerve is particularly prone to in- cretomotor fibers and gustatory fibers Nose, and Throat Hospital, jury because of the long distance it regenerate after facial nerve injury and Shanghai Medical University, traverses intratemporally and extratem- whether their regeneration affects motor People’s Republic of China. porally. Studies4-8 have been conducted fiber regeneration. The purpose of the pres-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 127, APR 2001 WWW.ARCHOTO.COM 422

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 MATERIALS AND METHODS superficial petrosal nerve were excised, and the main trunk of the facial nerve was preserved. To prevent regeneration of the intermediate nerve to the distal segment of the fa- ANIMALS cial nerve, bone wax was replaced in the geniculate gan- glion. Antibiotics were administered to minimize the Twenty-three adult New Zealand albino rabbits of both sexes possibility of infectious complications. Three months post- weighing 2.5 to 3.2 kg were randomly assigned to 2 groups. operatively, all animals were placed under deep general an- All animals underwent removal of a 2-mm tympanic seg- esthesia, the otic vesicle was opened, the parotid gland was ment of the facial nerve, and the greater auricular nerve was retracted forward slightly, the was harvested for grafting. In group 1 (10 rabbits), the genic- dissected, and the mastoidal and extratemporal segments ulate ganglion was preserved. In group 2 (13 rabbits), the of the facial nerve were harvested for electron micro- geniculate ganglion was dissected. Mastoidal and extra- scopic evaluation. The methods and protocol of the study temporal segments of the facial nerve were harvested 3 were reviewed and approved by the Institutional Animal months postoperatively for histological examination. Care and Use Committee of Shanghai Medical University, People’s Republic of China. SURGICAL PROCEDURE HISTOLOGICAL EXAMINATION All operations were performed on the left side with an operating microscope (Carl Zeiss, Inc, Oberkochen, For electron microscopic evaluation, tissues were im- Germany). The right side served as a control. All surgical mersed in 3% glutaraldehyde for 2 hours and then in 1% procedures were performed under aseptic conditions. The osmium tetroxide for 1 hour. Dehydration in a series of animals were anesthetized with inhalation through intu- graded ethanols was followed by gradual infiltration with bation of 0.5% enflurane with equal parts of nitrous oxide epoxy (Embed-812; Electron Microscope Sciences, To- and oxygen. The otic vesicle was exposed postauricularly. kyo, Japan). Semithin sections were made and stained with Through the surgical fenestration, the incus was removed toluidine blue O for light microscopic examination, and ul- and the tympanic fallopian canal was identified above the trathin sections were stained with uranyl acetate and lead footplate of the stapes. The fallopian canal was opened citrate for electron microscopy (JEM-2000CX; JEOL Co, with a diamond burr, and 2.0 mm of the facial nerve was Tokyo, Japan). Axon count and size and distribution of removed. The nerve graft was carefully inserted in the unmyelinated and myelinated fibers were determined by opened canal between the cut ends of the facial nerve. electron microscopic examination. The donor nerve for grafting was the greater auricular Analyses to determine the statistical significance of the nerve. difference in the axon count of the facial nerve between In addition to the above procedures, the animals in group 1 and group 2 were performed using t tests with com- group 2 had the malleus head removed to expose the ge- mercially available software (STAT-VIEW,version 5.0; Aba- niculate ganglion. The geniculate ganglion and the greater cus Concepts Inc, Berkeley, Calif).

ent study was to observe regeneration of the distal facial elin sheath thickness was proportionate to the diameter nerve following nerve grafting within the tympanic seg- of the axons (Figure 3). No unmyelinated axons were ment with geniculate ganglion preservation or dissec- seen. In the extratemporal segments, myelinated axons tion. were evenly distributed. The number of myelinated axons was 515.1±103.1, which is about one quarter to one third RESULTS of that in normal facial nerve. Proliferation of connec- tive tissue was noted among axons. However, compar- In normal facial nerves, myelinated axons were evenly ing group 1 with group 2, the number of regenerative my- distributed. Sheath thickness was proportionate to axon elinated axons was increased significantly (PϽ.001) in diameter (Figure 1), and the number of axons was group 2. 1819.6±535.6. Data are given as mean±SD. In group 1, regenerative myelinated fibers and un- myelinated fibers were identified in the mastoidal seg- COMMENT ments. However, the number of myelinated axons was 123.6±31.1, which was much fewer than were found in It is well-known that there are somatic motor, gusta- normal facial nerves. In addition, the diameter of regen- tory, and secretomotor fibers within the facial nerve. Early erative axons was decreased, and the sheath thickness in regenerative gustatory fibers in and se- regenerative fibers was diminished (Figure 2). Extra- cretomotor fibers are unmyelinated. Mature regenera- temporal segments were almost completely composed of tive gustatory fibers and somatic motor fibers are my- connective tissue, with fewer myelinated axons than were elinated. An anatomical study9 has indicated that the total found in the mastoidal segments. Myelinated axons were number of myelinated nerve fibers in the facial nerve var- diffusely distributed throughout the extratemporal seg- ies from 7500 to 9370, depending on the anatomical level ments. It is estimated that only one eighteenth to one of the nerve segment. The greatest number of nerve axons twelfth of myelinated axons are used to innervate muscle. was found at the level of the middle of the mastoidal por- In group 2, a large number of regenerative myelin- tion. The peak diameter of the facial nerve axon was ated axons was found in the mastoidal segments. The my- between 4 and 6 µm. The number of facial nerve fibers

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 127, APR 2001 WWW.ARCHOTO.COM 423

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 matolytic technique in cats to trace the distribution of the afferent fibers through the facial nerve branches, found that 20% of fibers traversed the greater superficial pe- trosal nerve. The remainder was distributed in the chorda tympani (45%), posterior auricular rami (21%), branches to mimetic muscle (8%), deep cervical branch (5%), and nerve to the stapedius (1%). Somatic motoneurons are in the , which is located in the brain- stem. Secretomotor neurons are situated in the superior salivatory nucleus. Sensory neurons of the gustatory fibers localize in geniculate ganglia via the chorda tym- pani nerve, supplying buds in the . In this study, we demonstrated axonal regener- ation following facial nerve grafting with or without geniculate ganglionectomy. With geniculate ganglion pres- Figure 1. Electron microscopic examination of a mastoidal segment of normal facial nerve shows myelinated axons that are evenly distributed and ervation, an increase in regenerative gustatory fibers was whose sheath thickness is proportionate to the axon diameter (original identified in mastoidal segments following facial nerve magnification ϫ960). transection. There were more unmyelinated fibers than myelinated fibers. In contrast, with geniculate gangli- onectomy, a large number of myelinated axons was found in the mastoidal segments, the number of regenerative axons was similar to that of the control side, and there were no gustatory fibers. It is speculated that geniculate ganglionectomy has a positive effect on motor axon re- generation. The results may be explained by the fact that gustatory fiber is a C-type fiber and grows quickly following injury, and that gustatory neurons are in geniculate ganglia. However, compared with gustatory fibers, motor fibers must traverse a longer distance from the facial motor nucleus to the injury site. Therefore, it is advantageous for the regenerative gustatory fibers in the band of Bu¨ ngner to regenerate. After geniculate gan- glionectomy, the band of Bu¨ ngner gustatory fibers served as guides for motor fiber regeneration. Figure 2. In a mastoidal segment of facial nerve in group 1, myelinated 12 axons (arrowhead) and a greater number of unmyelinated axons (star) were Ylikoshi et al observed degenerative changes in the noted under electron microscopy (original magnification ϫ480). distal stump of the human facial nerve. Their results showed that the sensory component of the facial nerve had a normal appearance when the facial nerve was sev- ered at the internal acoustic meatus with the geniculate ganglion left intact. When the geniculate ganglion was damaged, the normal-looking sensory component of the facial nerve was absent from the distal stump. When the entire tympanic portion of the facial nerve was inter- rupted, most of the endoneural tube had thin unmyeli- nated fibers in the distal stump around the foramen, and few myelinated fibers appeared intact. It may be that re- generative sensory fibers interfere with the regeneration of the motor fibers. This observation was consistent with the findings of our study. Moreover, in group 2, no unmyelinated fibers were found in the mastoidal segments following geniculate gan- glionectomy. In addition, compared with group 1, the Figure 3. In group 2, electron miscroscopy shows a large number of number of regenerative myelinated fibers was signifi- myelinated axons in extratemporal segments of facial nerve following cantly increased in the mastoidal segments. This sug- geniculate ganglionectomy. There were fewer myelinated axons than were found on the control side. Fibers were evenly distributed, with a proliferation gests that, without the effect of nonmotor fiber, more mo- of connective tissue (arrowhead) among the axons (original magnification tor fiber may be regenerated. ϫ12000). From our experimental data, it seems reasonable to conclude that geniculate ganglionectomy is associated decreased with the age of the patient.9 Further investi- with improvement of motor axon regeneration. Earlier gation showed that a significant proportion (15%-20%) studies13,14 proposed geniculate ganglionectomy as a of the fiber composition in the facial nerve trunk and its means of excising the cell bodies of aural cutaneous pain peripheral branch is nonmotor.10 Bruesch,11 using chro- afferents in patients with geniculate neuralgia. These re-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 127, APR 2001 WWW.ARCHOTO.COM 424

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 ports showed that this surgery benefited these patients. 2. Gacek RR. On the duality of the facial nerve ganglion. Laryngoscope. 1998;108: A histological basis for the use of geniculate ganglion- 1077-1086. 3. Kitamura K, Kimura RS, Schuknecht HF. The ultrastructure of the geniculate gan- ecomy as a treatment for geniculate neuralgia was thereby glion. Acta Otolaryngol (Stockh). 1982;93:175-186. established. It is recommended that geniculate ganglio- 4. Wong BJF, Mattor DE. Experimental nerve regeneration. Otolaryngol Clin North nectomy be performed when facial nerve decompres- Am. 1991;24:739-752. sion or grafting is needed in patients with facial paraly- 5. Spector JG, Lee P, Derdy A, Frierdich GE, Neises G, Roufa DG. Rabbit facial nerve regeneration in NGF-containing Silastic tubes. Laryngscope. 1993;103:548- sis, as it may enhance motor axon regeneration and 558. improve the recovery from facial palsy. 6. Damon DH, D’Amore PA, Wagner JA. Nerve growth factor and fibroblast growth The present study only investigated motor axon re- factor regulate neurite outgrowth and gene expression in PC12 cells via both generation following geniculate ganglionectomy. Fur- protein kinase C- and c-AMP independent mechanisms. J Cell Biol. 1990;110: ther studies should be undertaken to focus on facial move- 1333-1339. 7. Ide C, Osawa T, Tohyama K. Nerve regeneration through allogeneic nerve grafts ment evaluation by electrophysiologic testing. with special reference to the role of the Schwann cell basal lamina. Prog Neu- robiol. 1990;34:1-38. Accepted for publication September 22, 2000. 8. Byers JM, Thompson GC. Effect of pulsed electromagnetic stimulation on facial We thank Christopher Lam, MD, and Dingding Xiong, nerve regeneration. Arch Otolaryngol Head Neck Surg. 1998;124:383-389. 9. Thurner KH, Egg G, Spoendlin H, Schrott-Fischer A. A quantitative study of MD, PhD, for their helpful comments on the paper. nerve fibers in the human facial nerve. Eur Arch Otorhinolaryngol. 1993;250: Corresponding author and reprints: Chunfu Dai, MD, 161-167. PhD, Department of Otolaryngology, Eye, Ear, Nose, and 10. Gasser RF. The development of the facial nerve in man. Ann Otol Rhinol Laryn- Throat Hospital, Shanghai Medical University, Shanghai gol. 1976;76:37-56. 200031, People’s Republic of China (e-mail: daichf@online 11. Bruesch SR. The distribution of myelinate afferent fiber in the branches of the cat’s facial nerve. J Comp Neurol. 1944;81:169-183. .sh.cn). 12. Ylikoshi J, Hitselberger WE, House WF, Sanna M. Degenerative changes in the distal stump of the severed human facial nerve. Acta Otolaryngol. 1981;92:239- REFERENCES 248. 13. Pulec JL. Geniculate neuralgia: diagnosis and surgical management. Laryngo- scope. 1976;86:955-964. 1. Rupa V, Weider DJ, Glasner S, Saunders RL. Geniculate ganglion: anatomic study 14. Rupa V, Saunders RL, Weider DJ. Geniculate neuralgia: the surgical manage- with surgical implications. Am J Otol. 1992;13:470-473. ment of primary otalgia. J Neurosurg. 1991;75:505-511.

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 127, APR 2001 WWW.ARCHOTO.COM 425

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021