Bone wax prevents nystagmus after labyrinthine fenestration in guinea pigs JOHN M. BROCKENBROUGH,
and M. RITA I. YOUNG,
Maywood and Hines,
Illinois OBJECTIVE: We examined the effect of 3 methods of closure of labyrinthine fenestrations. STUDY DESIGN AND SETTING: A guinea pig animal model was developed to examine the effect of 3 methods of closure of labyrinthine fenestrations. These methods included bone wax alone, fascia and fibrin glue, and fascia alone. Nystagmus, a reproducible sign of labyrinthine injury, was chosen as an end point for comparing materials commonly used in human ear surgery. RESULTS: Animals repaired with bone wax showed no postoperative nystagmus, whereas animals repaired with other materials or that underwent canal fenestration without repair showed 3 to 5 days of postoperative nystagmus. CONCLUSIONS: Although it is premature to extrapolate our results to human surgery, postoperative nystagmus can be eliminated in guinea pigs by sealing labyrinthine fenestrations with bone wax. SIGNIFICANCE: Given the intimate relationship between the vestibular and cochlear systems, it is possible that prevention of vestibular irritation might be associated with preservation of cochlear function. (Otolaryngol Head Neck Surg 2003;128: 726-31.)
V estibular injury in humans can lead to significant consequences, including deafness and balance problems. Many otologists hold the belief that entrance into the labyrinth during surgery will From the Departments of Otolaryngology–Head and Neck Surgery (Drs Brockenbrough, Marzo, and Young), Physiology (Dr Wurster), and Pathology (Dr Young), Loyola University Medical Center, Maywood , and the Department of Research Services, Hines Veterans Affairs Hospital, Hines, IL (Dr Young). Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, Denver, CO, September 9-12, 2001. Reprint requests: M. Rita I. Young, PhD, Research Service (151-Z2), Building 1, Room A319, Hines VA Hospital, Hines, IL 60141; e-mail, [email protected]
Copyright © 2003 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2003/$30.00 ⫹ 0 doi:10.1016/S1094-5998(03)23289-1 726
lead to a “dead ear.” It is also assumed that such an injury will lead to significant postoperative dizziness. The exact mechanisms accounting for the relationship between labyrinthine fenestration and these sequelae are not well explained in the literature. Recently, investigators have demonstrated that posterior semicircular canal obliteration can be performed for the treatment of refractory cases of benign positional paroxysmal vertigo while preserving hearing and minimizing postoperative vestibulopathy.1 Despite this, management of labyrinthine injury remains controversial in the otolaryngology literature. Other causes of labyrinthine injury include erosion into the semicircular canals by cholesteatoma. It is estimated that 10% of cholesteatomas are associated with a labyrinthine fistula, usually of the lateral semicircular canal (LSCC). Some authors have advocated removal of cholesteatoma matrix overlying the fistula, and a variety of repairs have been described.2-4 Other authors advocate leaving matrix over the fistula, not wishing to risk possible severe sensorineural hearing loss that sometimes results from its disruption.5,6 Regardless of one’s management philosophy, repair of labyrinthine injury should be well understood by those who perform ear surgery. Entrance into the labyrinth occasionally occurs during ear surgery, especially when normal anatomy has been destroyed and the surgeon is inexperienced. Guinea pigs have been used extensively in otologic research for a variety of reasons. Their surgical anatomy has been well described.7,8 They are relatively easy to work with, and have been shown in many studies to recover well after otologic surgery. In addition, many neurophysiologic studies have been performed, contributing to a wide body of knowledge surrounding this animal. Several investigators have attempted to describe the effect of labyrinthine injury on cochlear function in the guinea pig. In general, cochlear function in these animals has been shown to be relatively resistant to surgical vestibular injury.
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
Although many of these studies have mentioned postoperative vestibulopathy in their animals, few have attempted to further characterize this finding. Investigators who have studied postoperative vestibulopathy after labyrinthine injury in guinea pigs have described 3 characteristic physical findings. The first is circle turning behavior, or ataxia, that occurs in the acute phase of injury. This finding is inconsistent and generally subsides in the first 24 hours after injury. More commonly described are the findings of nystagmus and head tilt. Head tilt has been further subdivided into 2 components: “yaw,” or lateralization of the head toward the injured side, and “roll” or rotation of the head around the longitudinal axis of the animal toward the injured side.9 Nystagmus after labyrinthine injury has been well described in the guinea pig. It is rapid with the fast component toward the normal side and decays over an average of 3 days.10 In this study, postoperative nystagmus was recorded in guinea pigs after LSCC fenestration. Groups of animals repaired with different substances were compared with nystagmus as an end point. MATERIALS AND METHODS All procedures were approved by the Institutional Animal Care and Use Committee at Loyola University Medical Center. Forty-five male Hartley albino guinea pigs weighing 400 to 750 g were divided into 5 groups of 9 animals each. Animals were anesthetized using ketamine (Ketaject; Phoenix Pharmaceuticals, Inc, St Joseph, MO) 40 mg/kg and xylazine (Phoenix Pharmaceuticals, Inc.) 4 mg/kg injected intraperitoneally. Ketaject 20 mg/kg was injected intramuscularly for additional anesthesia if needed. Level of anesthesia was monitored by toe pinch and observation of respirations. Postauricular hair was shaved, and the area was prepped with povidine-iodine (Betadine). Lidocaine 1% with 1:100,000 epinephrine (Abbott Labs, North Chicago, IL) was injected in a volume of 0.3 mL into the left postauricular area, and the animal was placed into a holder similar to the one described by Watson et al.11 No perioperative antibiotics were used. With clean technique, a 2-cm postauricular incision was performed. The subcutaneous tissue was sharply dissected, and the muscle overlying
BROCKENBROUGH et al 727
Fig 1. Intraoperative photograph demonstrating guinea pig left ear anatomy. FN, Facial nerve; CO, cochlea; LSCC, lateral semicircular canal.
the mastoid bulla was detached. The bulla was opened using a 1.0-mm cutting bur, and the LSCC was identified, as were the facial nerve and cochlea (Fig 1). Using a 0.6-mm diamond bur, a fistula was created in the LSCC extending from the facial nerve anteriorly to the posterior semicircular canal posteriorly (Fig 2). Bone was removed until the membranous labyrinth could be well identified. In the vast majority of cases, the membranous labyrinth could be preserved using this technique. Perilymph was allowed to escape and was suctioned during the drilling. All wounds were irrigated generously throughout drilling, and care was taken to clear the operative area of bone dust. Summarized in Table 1 are the procedures that were performed in each of the 5 groups of guinea pigs. In the first group, dissection was carried out until the canal could be identified. No entrance into the LSCC was performed. In the second group, the LSCC was entered as described and no repair was attempted (Fig 2). In group 3, the LSCC was entered and the defect was sealed with bone wax. This was accomplished by completely occluding the bony canal with bone wax. In the fourth group, the defect was sealed with periosteum that had been harvested from the tympanic bulla, and covered with Beriplast HS fibrin glue
Otolaryngology– Head and Neck Surgery May 2003
728 BROCKENBROUGH et al
Fig 3. Intraoperative photograph demonstrating fistula in LSCC repaired with fascia alone.
Fig 2. Intraoperative photograph demonstrating fistula in LSCC.
Table 1. Summary of experimental animal groups Group
1 2 3 4 5
Identify canal, no fistula Fistula without repair Repair with bone wax Repair with fascia and fibrin glue Repair with fascia alone
(Marburg, Germany). In the final group, the defect was covered with periosteum alone (Fig 3). All middle ears were then gently irrigated again with saline to remove any residual bone dust and debris, taking care not to disturb the fistula repair sites. Before wound closure, all repair sites were inspected under microscopic magnification to ensure there was no evidence of perilymph leak or disruption of the fistula repair. The wounds were then closed with a combination of 4-0 Vicryl sutures and skin staples. The animals were then given buprenorphine HCl (Buprenex) (Richmond, VA) 0.3 mg/kg intramuscularly for postoperative analgesia.
Animals were weighed daily for the first week and then weekly thereafter for a total of 4 weeks. Each animal was checked for spontaneous nystagmus by holding the animal horizontally with the operated side (left) up and observing left eye movement. No attempt was made to determine nystagmus frequency. Animals with clear evidence of sustained nystagmus were recorded as “positive,” and others were recorded as “negative.” Animals were checked daily for the first week for nystagmus, and the percentage of each group with this finding was determined. Groups were compared for statistical significance using binomial tables. At postoperative day (POD) 30, each animal was humanely killed with an intraperitoneal pentobarbital overdose. Next, the wound was explored and the bulla was reopened with a 1.0-mm cutting bur. The fistula site and middle ears were then examined under the operating microscope. Observations, including bony proliferation, middle ear adhesions, and evidence of otitis media, were recorded. RESULTS In 45 operations, there were 3 deaths (deaths and complications are summarized in Table 2). One death was intraoperative, presumably due to
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
BROCKENBROUGH et al 729
Table 2. Summary of complications Group
1 2 3
Death, probable gastrointestinal None Facial nerve weakness (2 animals) Wound infection (death) Wound infection (death) Facial nerve weakness (2 animals) Wound infection (3 animals) Wound infection (3 animals) Death, anesthetic
1/9 0/9 3/0
anesthetic overdose. The second death occurred in an animal in group 1 that developed diarrhea during the second week and died on POD 12. The third death occurred in an animal from group 3; this animal developed a wound infection on POD 5, for which it received gentamicin intramuscularly and local wound therapy consisting of debridement, antibiotic irrigation, and topical antibiotic ointment. The animal died on POD 11. Three animals from group 4 developed wound infections requiring local therapy with wound debridement, irrigation, and topical antibiotic therapy. These animals were also started on cefazolin sodium (Ancef; SmithKline Beecham) 100 mg/kg intramuscularly twice daily for 5 days. All of these animals recovered and were included in the final data. There were 2 facial nerve injuries in group 4, which recovered spontaneously. In addition, 2 of the animals in group 3 had postoperative facial nerve weakness that recovered spontaneously. No obvious injury to the eyes was noted in any of these animals. The average animal weights for each group were obtained to monitor the overall health of the animals as they recovered from labyrinthine insult. Nearly all of the animals had regained surgical weight loss by the end of the first postoperative week. By the end of the second week, all animals had surpassed their original weight. There were no statistical differences in terms of weight loss or gain between the different groups. Nystagmus data are presented in Fig 4. No animal’s nystagmus persisted longer than 5 days. No animal in groups 1 or 3 showed evidence of nystagmus, and therefore the percentage for those groups was 0% for each day. P values generated
using a binomial probability model comparing these groups with group 2 ranged from 0.000 on POD 1 to 0.044 on POD 4. There was no statistically significant difference between group 2 and groups with other methods of repair. Autopsy data are as follows. All LSCC fenestrations were found to be closed. With regard to hyperostosis, this usually was manifested by bony proliferation between the LSCC fistula site and the opening created in the mastoid bulla. Group 1 showed no hyperostosis, but 25% of the animals showed middle ear adhesions. Group 2 demonstrated a 100% incidence of hyperostosis over the fistula site, and in all cases bone growth was described as moderate. No animals in this group had middle ear adhesions. Group 3 revealed a 100% incidence of hyperostosis, which was described as extensive in all except one animal. There was a 25% incidence of middle ear adhesions in this group. Group 4 demonstrated a 90% incidence of hyperostosis, described as moderate in all animals, and a 44% incidence of middle ear adhesions. Group 5 had a 38% incidence of hyperostosis and a 25% incidence of adhesions. Bone growth was described as moderate in all animals. DISCUSSION The ideal substance to repair a surgical labyrinthine injury would possess the following characteristics: (1) is inexpensive, (2) forms a perfect seal, (3) is well tolerated by the host, (4) is easy to use, and (6) preserves inner ear function. No such material has been described in the literature to date. The present study demonstrates that closing a labyrinthine fenestration with bone wax can significantly reduce the amount of postoperative nys-
730 BROCKENBROUGH et al
Otolaryngology– Head and Neck Surgery May 2003
Fig 4. Symptomatic animals. Each animal was examined for evidence of nystagmus. Shown are the percentages of animals in each experimental group with nystagmus.
tagmus in the guinea pig model. Although other investigators have suggested that vestibular compensation via central mechanisms is the primary manner by which nystagmus resolves in guinea pigs after vestibular injury,12 these data raise the possibility that mechanical obstruction of the perilymphatic fluid leak may play a role in this process. It is possible that spontaneous closure of the labyrinthine defect contributes to the resolution of nystagmus in addition to central compensatory mechanisms. There are, however, limitations to our studies. First, our conclusions were made from studies that used a guinea pig model, and it is premature to extrapolate our findings to human surgery without first conducting clinical trials in patients. Second, because the first author performed all operations and observed all animals for nystagmus, the issue of observer bias cannot be overlooked. However, the presence or absence of nystagmus was so prominent that observer bias was unlikely to reverse the statistical significance of the findings. Although spontaneous nystagmus is a useful indicator of acute labyrinthine injury, additional
studies are needed to determine the relationship between this finding and cochlear function. Several studies to date have looked at cochlear function after labyrinthine fenestration. Parnes and McClure13 demonstrated that cochlear function could be preserved in guinea pigs after fenestration and packing of the posterior semicircular canal with bone wax. Interestingly, they elected not to use bone wax in one of their animals that had no apparent intraoperative perilymphatic leak. This animal subsequently developed progressive deterioration of cochlear function as measured by brainstem evoked audiometry. They ascribed this to slow egress of inner ear fluid. Smouha and Michaelides14 performed a variety of labyrinthine fenestration procedures on guinea pigs. Their work showed hearing preservation, as measured by click-evoked auditory brain stem response, in animals after extensive labyrinthine damage. Only when the vestibule was widely opened did significant hearing loss result. Intraoperative suctioning of perilymph resulted only in a transient hearing loss, suggesting that the guinea pig cochlea may be resistant to transient perilym-
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
phatic fluid losses. Antonelli et al15 demonstrated that hearing in guinea pigs could be preserved with adequate sealing of the labyrinth after triple semicircular canal occlusion. Their results showed that cochlear function, as measured by electrocochleography, was significantly improved when bone wax packing was used compared with the use of fibrin glue and Gelfoam. They suggest that hearing may be preserved with adequate sealing of the labyrinth after labyrinthectomy.15 Other studies have confirmed these findings, including work done by Kobayashi et al16 using guinea pigs and by Gjuric et al17 using a similar rabbit model. It is possible that failure to adequately seal the labyrinth after fenestration in guinea pigs leads to subtle cochlear injury that is not easily determined by methods such as brainstem evoked audiometry and electrocochleography. An interesting study by Kokesh et al18 demonstrated that perilymphatic fistula in guinea pigs was associated with a significant alteration in distortion product otoacoustic emission (DPOAE) amplitudes. Although their model differed in that a cochleostomy was performed instead of labyrinthine fenestration, animals with healed fistulas showed alteration in DPOAE amplitude. CONCLUSION This study demonstrates that postoperative nystagmus can be eliminated in guinea pigs by sealing LSCC fenestrations with bone wax. Our results suggest that watertight closure of the labyrinth prevents perilymphatic fluid leak and subsequent vestibular irritation. Given the intimate relationship between the vestibular and cochlear systems, it is possible that the prevention of vestibular irritation might be associated with preservation of cochlear function. Further studies will be directed at determining cochlear function using more sensitive techniques such as otoacoustic emissions. The authors are grateful to Michael Raffin, PhD, for his assistance in statistical analysis of our results. We also thank Rosann Grahovac for assisting in the preparation of the manuscript.
BROCKENBROUGH et al 731
1. Parnes LS, McClure JA. Posterior semicircular canal occlusion for intractable benign paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 1990;99:330-4. 2. Parisier SC, Edelstein DR, Han JC, et al. Management of labyrinthine fistulas caused by cholesteatoma. Otolaryngol Head Neck Surg 1991;104:110-5. 3. Herzog JA, Smith PG, Kletzker GR, et al. Management of labyrinthine fistulae secondary to cholesteatoma. Am J Otol 1996;17:410-5. 4. Palva T, Ramsay H. Treatment of labyrinthine fistula. Arch Otolaryngol Head Neck Surg 1989;115:804-6. 5. Vartiainen E. What is the best method of treatment for labyrinthine fistulae caused by cholesteatoma? Clin Otolaryngol 1992;17:258-60. 6. Smyth GDL. Surgical management of chronic otitis media and labyrinthine fistula. Otolaryngol Clin North Am 1978;11:95-104. 7. Asarch R, Abramson M, Litton WB. Surgical anatomy of the guinea pig ear. Ann Otol Rhinol Laryngol 1975;84: 250-5. 8. Adamczyk M, Appleton CM, Parell GJ, et al. Stapedectomy in the guinea pig. Otolaryngol Head Neck Surg 1999;121:581-4. 9. Curthoys IS, Smith PF, Darlington CL. Postural compensation in the guinea pig following unilateral labyrinthectomy. Prog Brain Res 1988;76:375-84. 10. Ris L, Capron B, de Waele C, et al. Neck muscle activity after labyrinthectomy in the alert guinea pig. Exp Brain Res 1999;124:159-65. 11. Watson RT, Leslie WG, Jennings EH. Operating table for guinea pig ear surgery. Arch Otolaryngol 1978;104: 177-8. 12. Jeram AH, Darlington CL, Smith PF. Methylprednisone reduces spontaneous nystagmus following unilateral labyrinthectomy in guinea pig. Eur J Pharmacol 1995;275: 291-3. 13. Parnes LS, McClure JA. Effect on brainstem audiometry evoked responses of posterior semicircular canal occlusion in guinea pigs. J Otolaryngol 1995;14:145-50. 14. Smouha EE, Michaelides EM. Partial labyrinthectomy with hearing preservation: an experimental study in guinea pigs. Otolaryngol Head Neck Surg 1996:777-84. 15. Antonelli PJ, Bouchard KR, Kartush JM, et al. Triple semicircular canal occlusion in the guinea pig. Otolaryngol Head Neck Surg 1997;117:509-15. 16. Kobayashi T, Shiga N, Hozawa K, et al. Effect on cochlear potentials of lateral semicircular canal destruction. Arch Otolaryngol Head Neck Surg 1991;117:1292-5. 17. Gjuric M, Wigand ME, Hosemann W. Selective semicircular canals of rabbits with preservation of hearing. Acta Otolaryngol 1992;112:907-15. 18. Kokesh J, Norton SJ, Duckert LG. Effect of perilymphatic fistulas on evoked otoacoustic emissions in the guinea pig. Am J Otol 1994;15:466-73.