Original Research—Otology and Neurotology
Otolaryngology–
Ototoxicity of Polymyxin B, Neomycin,
and Hydrocortisone Suspension in
Tympanoplasty Surgery
James R. House III, MD1,2, and
Laura K. House, MS2
No sponsorships or competing interests have been disclosed for this article.
Abstract
Objectives. (1) To determine the safety of using a commer-
cially available suspension of polymyxin B, neomycin, and
hydrocortisone (PNH) in tympanoplasty surgery. (2) To
apply evidence-based medicine to tympanoplasty surgery
when considering potential ototoxicity.
Study Design. Case series with chart review.
Setting. Tertiary otology practice, single surgeon.
Methods. Approval for this study was obtained from the St.
Dominic–Jackson Memorial Hospital Institutional Review
Board. Data were gathered on 272 consecutive type 1,
underlay tympanoplasties for which both pre- and post-
operative audiometric data were available over a 10-year
period. In each surgery, gelatin sponge saturated in a com-
mercially available PNH suspension was placed in the middle
ear to support the graft. Patients ranged in age from 3 years
to 79 years. Preoperative and postoperative bone conduc-
tion thresholds were measured at 500, 1000, 2000, 3000,
and 4000 Hz.
Results. The average change in sensorineural hearing as mea-
sured by bone conduction thresholds was negligible, with a
slight improvement in all frequencies tested except 4000 Hz.
The changes by frequencies were as follows: 500 Hz (–1.624
dB), 1000 Hz (–1.399 dB), 2000 Hz (–0.975 dB), 3000 Hz (–
0.596 dB), and 4000 Hz (10.560 dB). The 5-frequency aver-
age change was 20.545 dB.
Conclusions. The commonly used otic solution containing
polymyxin B, neomycin, and hydrocortisone demonstrates
no ototoxicity in tympanoplasty surgery and is safe to use in
this setting.
Keywords
ototoxicity, Cortisporin, tympanoplasty, evidence-based
medicine
Received September 11, 2013; revised October 21, 2013; accepted
October 25, 2013.
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on January 22, 2016
Head and Neck Surgery
2014, Vol. 150(2) 282–284
Ó American Academy of
Otolaryngology—Head and Neck
Surgery Foundation 2013
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0194599813513007
http://otojournal.org
T
he commonly used otic solution composed of poly-
myxin B, neomycin, and hydrocortisone (PNH) or
Cortisporin has been used for more than 40 years in
the treatment of external otitis and suppurative chronic otitis
media and in surgery of the middle ear. Aminoglycosides
have long been known to be ototoxic when given systemi-
cally and when applied to the round window in treatment of
vestibular disorders.1-3 With the advent of broad-spectrum
topical otic preparations using quinolones, it has been pro-
posed that PNH be reserved for refractory cases and the qui-
nolones be used preferentially.4 It has been the senior
author’s observation that sensorineural hearing loss (SNHL)
is not observed following the use of PNH in tympanoplasty.
To evaluate this theory, a retrospective review of consecu-
tive type I tympanoplasty operations was performed.
Methods
Using a computer-generated review of CPT codes over a
10-year period in a single-otologist private practice setting,
all patients billed for a 69631 CPT code (tympanoplasty
without mastoidectomy) were reviewed. A total of 757 ears
met these criteria. Patients who had an overlay technique
were excluded because of the possibility that drilling the ear
canal might affect high-frequency hearing. Patients who did
not have pre- and postoperative bone audiometry were also
excluded. A total of 480 ears were excluded because of a
lack of complete audiologic data. Ten ears were from bilat-
eral tympanoplasty patients. One ear from each patient was
excluded by alternating between left and right ears for
patients with bilateral tympanoplasty. This left a total of
272 ears meeting the stated criteria.
The patient population included a wide age range for
subjects, from 3 years to 79 years. The average age per ear
was 24.4 years. The median age was 15 years, indicating a
1
Jackson Ear Clinic, Jackson, Mississippi, USA
2
University of Mississippi Medical Center, Jackson, Mississippi, USA
This article was presented at the 2013 AAO-HNSF Annual Meeting and
OTO EXPO; September 29–October 3, 2013; Vancouver, BC, Canada.
Corresponding Author:
James R. House III, MD, Jackson Ear Clinic, P.O. Box 16685, Jackson, MS
39236-6685, USA.
Email: jhouseiii@comcast.net
House and House
Table 1. Average threshold shifts (95% confidence interval).
500 Hz 1000 Hz 2000 Hz
–1.624 dB –1.399 dB –0.975 dB
(22.724 to 20.524) (21.989 to 20.809) (22.085 to 0.145)
Table 2. Maximum measured threshold shifts.
500 Hz 1000 Hz 2000 Hz
–20 dB –20 dB –25 dB
115 dB 115 dB 120 dB
preponderance of pediatric patients. There was 1 more right
ear than left ear. Fifty-three percent of the ears were female.
One surgeon using a temporalis fascia or tragal perichon-
drial graft in an underlay fashion performed all the surgeries.
Gelfoam, a commercially available gelatin sponge, was satu-
rated with a PNH solution and placed in the middle ear and
the external canal to support the graft. An audiologic evalua-
tion was performed no sooner than 4 months postoperatively
by licensed audiologists. Paired data for pre- and postopera-
tive audiometric bone conduction thresholds were measured
at 500, 1000, 2000, 3000, and 4000 Hz in 5-dB increments.
Results
There was no overall significant change in the audiometric
thresholds. A slight improvement was noted in the lower
frequencies, trending to a slight decline in the highest fre-
quency. The changes in bone conduction thresholds were as
follows: 500 Hz (–1.624 dB), 1000 Hz (–1.399 dB), 2000
Hz (–0.975 dB), 3000 Hz (–0.596 dB), and 4000 Hz
(10.560 dB; Table 1). There were no incidences of a dra-
matic change in sensorineural hearing. The greatest change
was observed in 1 ear with an 18-dB average improvement
in hearing. Two ears were observed to experience a 15-dB
average decline in hearing. The greatest change at any one
frequency was a 35-dB decline at 4000 Hz (Table 2).
Discussion
It is well known that aminoglycosides can cross the round
window membrane of mammals and cause vestibular and
cochlear toxicity. Numerous animal studies have shown hair
cell loss and hearing loss with topical application of neomy-
cin and polymyxin B to the round window membrane.5 Also,
hearing loss can occur in patients undergoing gentamicin per-
fusion of the inner ear via the round window to achieve a
chemical labyrinthectomy in treatment of Meniere’s disease
and other vestibular disorders.6,7 This knowledge, along with
the advent of commercially available quinolone topical otic
preparations, led to an investigation of the safety of amino-
glycosides in the treatment of otologic disease with nonintact
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on January 22, 2016
283
3000 Hz 4000 Hz 5-Frequency Average
–0.596 dB 10.560 dB –0.545 dB
(21.696 to 0.494) (20.270 to 1.400) (20.995 to 20.105)
3000 Hz 4000 Hz 5-Frequency Average
–20 dB –20 dB –18 dB
130 dB 135 dB 115 dB
tympanic membranes (NITM) through a consensus panel
commissioned by the AAO-HNS in 2004.5 Several studies
were reported by the panel, including an investigation of
animal studies,5 a review of clinical reports of SNHL with
topical solutions in NITM patients,8 and an analysis and rec-
ommendation report.4
In the review of animal studies, more than 30 investiga-
tions looking at numerous topical medications including
neomycin and polymyxin B were analyzed. These studies
were conducted on a variety of animals, including primates.
The studies demonstrated loss of inner and outer hair cells
in the cochlea and loss of hearing through electrophysiolo-
gic measurements.5
In an exhaustive review of the literature from 1966 to
2003, Matz et al8 found evidence for 11 cases of cochlear
and 2 cases of vestibular toxicity attributed to long-term use
of neomycin in mastoid cavities.
Based on these investigations, the AAO-HNS consensus
panel report discussed the poor correlation of animal studies
and clinical experience, offering several possible explana-
tions. These included differences between humans and ani-
mals in the round window and the organ of Corti. The panel
referenced studies that demonstrated no hearing loss associ-
ated with use of PNH in NITM patients. The panel admitted
that it was charged with using an evidence-based medicine
approach in the analysis. The panel could use only a grade
C for evidence. The panel concluded that potentially oto-
toxic preparations should not be used routinely despite lack
of clinical evidence that ototoxicity is likely but rather
because other preparations are available.4
In response to these recommendations, otologists at the
Lippy group in Warren, Ohio, published 2 studies based on
their observation of safety in using PNH for years in NITM
patients.9,10 One study examined 500 patients receiving
PNH at a dose of 3 drops, 3 times a day, for 5 days. There
were no cases of hearing loss.9 The second study evaluated
distortion product otoacoustic emissions in 30 patients
treated in a similar fashion compared with a control group
of 30 patients. Again, no ototoxicity was demonstrated.10
284
Other studies have also demonstrated no hearing loss in
NITM patients treated with PNH.11,12 All of these studies
reported on short-term use of PNH.
A recent study by Antonelli and colleagues13 from the
University of Florida examined this question in an ambitious
study reviewing the database for Medicaid recipients in 29
states, over a 7-year period, involving 134,598 children. The
study determined ototoxicity based on the record of tympa-
nostomy tube placement, prescriptions for PNH or quinolone
ototopicals, and an ICD code for SNHL. No audiometric
studies were examined. Patients who received quinolones
showed no increase in the SNHL diagnosis. Patients receiv-
ing 1 prescription of PNH showed a slightly less incidence of
SNHL, whereas those with 2 prescriptions for PNH demon-
strated a statistically higher rate of SNHL diagnosis.
Interestingly, the group that had 3 or more prescriptions for
PNH had a slightly lower correlation with the SNHL diagno-
sis than the group that received 2 prescriptions. The authors
concluded that repeated applications of PNH in patients with
an NITM might result in SNHL. The obvious difficulty in
this study is the absence of any audiometric studies. Another
weakness is the inexact use of diagnoses in clinical practice.
When submitting a bill for services, many possible diagnoses
can be used. Rarely are all possible diagnoses used.13
Our present investigation of possible ototoxicity using
PNH in tympanoplasty has several limitations. A retrospec-
tive analysis inherently lacks the weight of a prospective,
randomized, blinded study. Many patients did not have post-
operative audiograms to review, introducing the possibility
of missed SNHL. Also, there is no control group with which
to compare these patients.
Conclusions
This study demonstrates grade B evidence that the use of a
commercially available solution of polymyxcin B, neomy-
cin, and hydrocortisone (Cortisporin) shows no evidence of
cochlear ototoxicity when used in tympanoplasty surgery.
Author Contributions
James R. House III, data analysis, drafting, final approval; Laura K.
House, data analysis, drafting, final approval.
Disclosures
Competing interests: None.
Sponsorships: None.
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on January 22, 2016
Otolaryngology–Head and Neck Surgery 150(2)
Funding source: None.
References
1. Banerjee A, Parnes LS. The biology of intratympanic drug
administration and pharmacodynamics of round window drug
absorption. Otolaryngol Clin North Am. 2004;37:1035-1051.
2. Pullens B, van Benthem PP. Intratympanic gentamicin for
Meniere’s disease or syndrome. Cochrane Database Syst Rev.
2011;(3):CD008234.
3. Gabra N, Saliba I. The effect of intratympanic methylpredniso-
lone and gentamicin injection on Meniere’s disease.
Otolaryngol Head Neck Surg. 2013;148:642-647.
4. Roland PS, Stewart MG, Hannley M, et al. Consensus panel
on role of potentially ototoxic antibiotics for topical middle
ear use: introduction, methodology, and recommendations.
Otolaryngol Head Neck Surg. 2004;130:S51-S56.
5. Roland PS, Rybak L, Hannley M, et al. Animal ototoxicity of topi-
cal antibiotics and the relevance to clinical treatment of human
subjects. Otolaryngol Head Neck Surg. 2004;130:S57-S78.
6. Carey J. Intratympanic gentamicin for the treatment of
Meniere’s disease and other forms of peripheral vertigo.
Otolaryngol Clin North Am. 2004;37:1075-1090.
7. Suryanarayanan R, Cook JA. Long-term results of gentamicin
inner ear perfusion in Meniere’s disease. J Laryngol Otol.
2004;118:489-495.
8. Matz G, Rybak L, Roland PS, et al. Ototoxicity of ototopical
antibiotic drops in humans. Otolaryngol Head Neck Surg.
2004;130:S79-S82.
9. Berenholz LP, Burkey JM, Farmer TL, et al. Topical otic anti-
biotics: clinical cochlear ototoxicity and cost consideration.
Otolaryngol Head Neck Surg. 2006;135:291-294.
10. Berenholz LP, Rossi DL, Lippy WH, et al. Is there an ototoxi-
city risk from Cortisporin and comparable otic suspensions?
Distortion-product otoacoustic emission findings. Ear Nose
Throat J. 2012;91:106-112.
11. Pickett BP, Shinn JB, Smith MFW. Ear ototoxicity: reality or
myth? Am J Otol. 1997;18:782-791.
12. Welling DB, Forrest LA, Goll FIII. Safety of ototopical anti-
biotics. Laryngoscope. 1995;105:474-474.
13. Winterstein AG, Liu W, Xu D, Antonelli PJ. Sensorineural
hearing loss associated with neomycin eardrops and nonintact
tympanic membranes. Otolaryngol Head Neck Surg. 2013;148:
277-283.