International Journal of Pediatric Otorhinolaryngology 79 (2015) 332335

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

A pilot study investigating basic broblast growth factor for the repair
of chronic tympanic membrane perforations in pediatric patients
Aanand N. Acharya a,*, Harvey Coates b,c, Dayse Tavora-Vie`ira a, Gunesh P. Rajan a,b,c
a
Fremantle Hospital & Health Service, Department of Otolaryngology, Alma Street, Fremantle, 6160, Western Australia, Australia
b
Princess Margaret Hospital for Children, Department of Otolaryngology, Roberts Road, Perth, 6008, Western Australia, Australia
c
Otolaryngology, Head & Neck Surgery, School of Surgery, University of Western Australia, Department of Otolaryngology, 35 Stirling Highway, Perth, 6009,
Western Australia, Australia

A R T I C L E I N F O A B S T R A C T

Article history: Objective : A pilot study to investigate the utility of basic Fibroblast Growth Factor (bFGF) in tympanic
Received 21 April 2014 membrane perforation (TMP) closure in a small cohort of pediatric patients.
Received in revised form 10 December 2014 Methods : Prospective cohort study. Suitability for inclusion in the study was conrmed by the
Accepted 13 December 2014
application of dened inclusion and exclusion criteria, and informed parental consent obtained. The
Available online 22 December 2014
technique used was a modication of the bFGF-technique by Kanemaru et al. Response to treatment was
monitored with serial otoscopy and audiometric outcomes were determined. Statistical analysis of the
Keywords:
outcomes was carried out.
Tympanic membrane perforation
Results : TMPs were successfully closed in 7/12 children at the rst attempt (58%) and in 10/12 children
Otitis media
Myringoplasty overall (83%). Hearing improvement was observed in 8/10 successfully treated cases (80%). There were
Fibroblast growth factor no complications or adverse outcomes.
Regeneration Conclusions : The topical bFGF regeneration technique offers a promising, minimally invasive
Pediatric alternative to conventional myringoplasty in pediatric patients with comparable success and reduced
morbidity and cost, especially considering the option of performing repeat applications. Patients with an
active infection or inammation are not suitable for the bFGF-mediated technique.
Crown Copyright 2014 Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Finally there is no consensus on the minimum age at which

Tympanic membrane perforations (TMP) especially in the
context of chronic suppurative otitis media (CSOM) are a common
problem affecting the pediatric population and are particularly
prevalent in various rural and remote indigenous populations
worldwide [1]. The resulting hearing loss is associated with
impairment of social, emotional and academic development in
children as well an increased absence from school, lifestyle
restrictions and an increased risk of antisocial behaviour [2]. A
myringoplasty procedure may be performed, however signicant
variation in success rates between techniques and surgeons make
the outcomes unpredictable and not reliably reproducible.
Furthermore the procedure is relatively expensive as it requires
sophisticated theatre equipment, theatre time and personnel.
* Correspondence to: Department of ENT, Fremantle Hospital, P.O. Box 480,
Fremantle, 6959, Western Australia. Tel.: +61 439072407.
E-mail addresses: aanand.acharya@health.wa.gov.au, aanandacharya@me.com
(A.N. Acharya).
myringoplasty surgery is appropriate [3]. Delaying surgery can
have a signicant negative impact on a childs development
thereby making earlier, less invasive and yet cost-effective
intervention desirable.
Basic broblast growth factor (bFGF) is a polypeptide mitogen
that has been shown to stimulate the proliferation of epidermal
and connective tissue cells as well as promoting angiogenesis
[46]. The application of bFGF has been shown to signicantly
increase the overall closure rate of TMPs and reduce the average
closure time [7,8]. The bFGF regeneration technique has been
reported to be safe [9,10], simple and cost-effective, with benets
over conventional surgical treatment [9,11] and high rates of
success. [9,10,12].
Most studies examining the use of bFGF for tympanic
membrane regeneration have investigated its use in adults. Some
authors have included older children (>13 years) [8,13], there is
one study that included younger children. Zhang [8] included
patients as young as 3 years of age in his study however no
distinction was made between children and adults in the analysis
and the published results do not allow a comparison to be made.

http://dx.doi.org/10.1016/j.ijporl.2014.12.014
0165-5876/Crown Copyright 2014 Published by Elsevier Ireland Ltd. All rights reserved.
 A.N. Acharya et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 332335 333

Furthermore, other authors used up to six repeated attempts at the
treatment in quick successions where the TMP failed to completely
respond to treatment [9,10] in order to investigate the efcacy of
the technique. This requires signicantly greater resources and is
not comparable with conventional myringoplasty surgery in which
the nal outcome from treatment is measured 312 months
following the surgery [3].
The aim of this study is to investigate the effectiveness of a
single application of bFGF at closing TMPs in a pediatric
population.
2. Methods
2.1. Patient selection
The following inclusion criteria were applied:
 Age < 16 years;
 chronic perforation (persistent for >3 months);
 central pars tensa perforation;
 size < 50% of the total tympanic membrane area;
 previous history of recurrent otorrhoea;
 no otorrhoea for minimum 3 months pre-operatively;
 no evidence of cholesteatoma or other complication of CSOM.
rim of tissue, the size of the TMP measured and an appropriately sized
piece of gelatin sponge placed in the defect. The bFGF solution (at a
concentration of 21,000 IU/5 ml) was dropped onto the gelatin sponge
and once this was saturated the excess solution removed. One or two
drops of blood were then applied to the sponge to seal the gelfoam-
bFGF-plug. Initially patients had the technique as originally described
by Kanemaru et al. utilizing brin glue but this was subsequently
modied to use of one drop of cyanoacrylate in lieu of brin glue.
Patients were followed up regularly for a minimum of 12 months
following the intervention: the minimum follow-up was at 1, 2, 3 and
4 weeks post-procedure and then at 2, 3, 6 and 12 months post-
procedure. At each post-operative visit an otoscopic photograph was
acquired and where there was evidence of a residual perforation the
size of this was measured. A post-operative audiogram was acquired
from 2 months post-treatment
2.3. Patients consent/ethics
Ethical approval for the study was granted by the Institutional
Review Board and written information leaets were provided to
the parents of all study participants when informed consent was
obtained.
2.4. Statistical analyses

Exclusion criteria were:
 Marginal perforation;
 size > 50% of the total tympanic membrane area;
 failure to achieve a dry ear for 3 months pre-operatively;
 cholesteatoma or other complication of CSOM;
 inadequate view and access of the TMP;
 only hearing ear;
 failure to adhere to appropriate post-treatment care and follow-
up.
A pre-operative audiogram and video-otoscopic photograph
was taken for each case and the size of the perforation was
measured using Image J software (NIH Bethesda MD USA).
2.2. Surgical technique
The surgical technique is a modication of the technique described
by Kanemaru [9]. Under general anaesthesia and by a transcanal
(permeatal) approach the TMP edges were freshened by excision of a
Students t-test and the Wilcoxon signed rank test were utilized
to analyse the audiometric outcome in the successful group.
Differences were considered statistically signicant when P < 0.05.
3. Results
3.1. Patient proles
The parents of all patients satisfying the inclusion and exclusion
criteria for the trial were invited to participate in the trial. This
minimized selection bias on the part of the investigators.
Thirteen children (male:female ratio of 9:4) were enrolled in
the study, one of which was subsequently excluded due to failure
to adhere to post-operative care instructions and failure to attend
follow-up. The 12 children included for analysis (Table 1) had an
age range of 616 years (mean 10.9 years, median 11 years). There
were seven right and ve left TMPs. The perforation size ranged
from 6% to 40% of the tympanic membrane. Where known, the
aetiology for the TMP was secondary to insertion of ventilation
tube or previous recurrent acute otitis media. One patient had

Table 1
Patient proles.

Patient Side Gender Aetiology Subjective History of Previous attempt Age at Pre-treatment Post-treatment Successful?
of TMP hearing loss? otorrhoea? at closure? Surgery (years) FFA (dBHL) FFA (dBHL)

1 L M VT Yes Yes No 15 30 30 No
1(r) L M VT Yes Yes Yes 16 30 17.5 Yes
2 R M AOM Yes Yes Yes 6 31.25 3.75 Yes
3 L M VT No Yes No 13 27.5 22.5 Yes
4 L F AOM No Yes No 7 17.5 7.5 Yes
5 R M U No Yes No 13 20 16.25 Yes
6 R M U No Yes No 14 12.5 12.5 Yes
7 R F AOM Yes Yes No 6 27.5 10 Yes
8 L M VT No Yes No 8 10 16.25 No
8(r) L M VT No Yes Yes 8 16.25 15 No
9 R M VT Yes Yes No 12 12.5 18.75 No
10 L M U Yes Yes No 12 27.5 27.5 No
10(r) L M U Yes Yes Yes 13 27.5 16.25 Yes
11 R M VT No Yes No 10 18.75 20 No
11(r) R M VT No Yes Yes 11 20 5 Yes
12 R F VT No Yes No 10 7.5 5 Yes

(r) = revision procedure, R = right, L = left, M = male, F = female, VT = ventilation tube, AOM = acute otitis media, U = unknown, TMP = tympanic membrane perforation, FFA = four
frequency average air-conduction audiometry (air conduction thresholds at 500 Hz, 1 KHz, 2 KHz, 4 KHz).
334 A.N. Acharya et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 332335
previously undergone an attempt at conventional underlay
myringoplasty.
3.2. Closure rates and time to achieve closure
Seven out of the 12 primary procedures resulted in a successful
closure of the TMP, with success being dened as an intact
tympanic membrane 12 months post-treatment. Two out of the
ve cases of initial failure resulted from external causes indepen-
dent of the treatment procedure: in one case the patient went
swimming in the early post-treatment period (patient 10) and in
the other the patient contracted a severe upper respiratory tract
infection in the early post-treatment period (patient 11). Both
these patients underwent a second attempt at the procedure with a
successful outcome. In two other cases of initial failure (patient
1 and patient 8) it is felt that this was likely due to rapid onset
otorrhoea as a result of liquefaction of the brin glue cap,
prompting a modication of the surgical technique as described
earlier. Both these patients underwent a second attempt at the
procedure, one of which was successful (patient 1).
Thus successful closure was achieved in 7/12 children (58%) at
their rst attempt. If the outcome is considered by patients treated,
the procedure was successful in 10 out of the 12 children (83%). In
those children who were successfully treated, the mean number of
treatments required to achieve success was 1.3 (including the two
cases in which initial failure resulted from external causes).
Among the successful cases the time for complete closure to be
conrmed ranged from 2 weeks to 3 months post-treatment.
In the two children whose treatments were unsuccessful
(patient 8 and patient 9) the perforation size and position was not
statistically signicantly different from the successful cases.
Patient 8 was younger than the mean/median age but not the
youngest in the cohort; three children younger than him had a
successful outcome from their rst treatment. Both of patient 8s
treatments involved the use of a brin glue cap and rapid
liquefaction of this cap is felt to have contributed to failure of the
treatments. Following the two procedures, the size of his TMP had
not changed signicantly. Patient 9 was identied to have a
residual perforation at 3 months following his treatment.
3.3. Audiometric outcome
The four-frequency average air conduction threshold (FFA, at
0.5 KHz, 1 KHz, 2 KHz and 4 KHz) improved in eight out of the
10 patients who had their TMP closed successfully. The mean
improvement in the FFA amongst the successful closures was
9.1 dB HL with a range of 027.5 dB HL. This improvement was
statistically signicant using both the t-test (P = 0.024) and the
Wilcoxon signed rank test (P = 0.025).
4. Discussion
This is the rst study examining the effect of bFGF on
regeneration of chronic tympanic membrane perforations speci-
cally in the pediatric population. In our study the trial technique
proved successful in the two youngest patients treated (each aged
Table 2
Operating theatre cost analysis between conventional underlay myringoplasty procedure and trial FGF procedure (all gures in Australian dollars).
Cost item Conventional underlay procedure (per patient)
Instrument tray $9185
Operating theatre time $1800 (45 min at $40 per minute)
Processing (sterilization) $71
Consumables $40
Total $1911
6 years at the time of surgery) indicating that this technique is
equally appropriate for the management of younger and older
children with TMP. This is in contrast to the ndings of previous
studies that have indicated a lower likelihood of success in TMP
repair among younger patients [3]. Given that one aim of TMP
repair is to reduce morbidity and hearing loss at a young age, the
nding that this technique is successfully applicable to children as
young as 6 years of age is an important one.
Frequently conventional myringoplasty results in a 2-layer
repair rather than a regeneration of the 3-layer morphology of the
healthy, normal tympanic membrane. The healed tympanic
membrane does not recover its complete tensile strength [14]
and this can be associated with increase predisposition to
retraction and associated complications. The bFGF technique
results in a 3-layer regeneration of the tympanic membrane [15]
thereby avoiding these potential problems and presenting an
additional benet of this technique over conventional myringo-
plasty techniques.
Perforations up to 50% of the total tympanic membrane surface
area were treated in this pilot study. The effect of this treatment on
larger perforations in the pediatric population remains to be
determined although in the adult population the treatment has
been shown to be successful for perforations exceeding 2/3 of the
tympanic membrane, albeit frequently requiring more than one
application. [9]
Four of the treatments were in children who had undergone
previous attempts at TMP repair, either by conventional underlay
myringoplasty or by the trial bFGF technique. The bFGF technique
was successful in three of these patients, demonstrating that the
technique employed is suitable as both a primary and secondary
treatment option. In the case of treatment failures the use of the
bFGF technique does not preclude subsequent attempts at repair
using the bFGF technique or conventional myringoplasty techni-
ques.
Four-frequency-average audiometry improved in 80% of the
successfully treated patients. This improvement has been demon-
strated to be signicant using both the Students T-test (P = 0.024)
and the Wilcoxon signed rank test (P = 0.025). Thus it can be
concluded from this small study that there is a change towards
improvement in the hearing thresholds. Given that one of the aims
of treatment in children is to facilitate normal hearing function in
their environment, thus promoting academic and social develop-
ment, the fact that the hearing can be improved is an important
secondary potential benet to this treatment option.
The bFGF procedure is a considerably shorter procedure than a
conventional myringoplasty (average 7 min compared with
45 min) and requires considerably less sophisticated equipment
and instrumentation. Cost analysis demonstrates that in the study
centre the cost of the trial procedure is less than a third (29%) of the
cost of a conventional underlay myringoplasty (see Table 2).
Furthermore the cost of the instrument tray that is required for the
trial procedure is less than 20% of the cost of the instrument tray
required for a conventional underlay myringoplasty (see Table 2).
The 83% success rate demonstrated in this study is comparable
with published conventional myringoplasty series [3], but is lower
than that demonstrated by other authors utilizing bFGF [8,9,16]. One
Trial procedure (per patient) Cost saving (per patient)
$1700 (Not applicable)
$280 (7 min at $40 per minute) $1520
$61 $10
$219 $179
$560 $1351
 A.N. Acharya et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 332335
signicant difference between these studies and ours is that in our
study, chronic TMP were treated, whereas other studies looked at
traumatic TMP which have a known high closure rate. A second
difference lies in the number of treatments utilized: while our study
examined the response of a TMP perforation to a single attempt and a
single revision attempt using this treatment modality, other authors
investigating bFGF for TMP repair have repeated the treatment up to
six times in quick succession in any individual patient [9,10]. A nal
difference is in the length of time for which the perforation should
have remained free of otorrhoea prior to intervention. Opinions vary
on this with requirements of up to 3 years reported by some authors
[9]. Our study investigated patients with a much shorter length of
time for which the perforation remained dry: 3 months.
There were no adverse outcomes to treatment in our study
group. Most authors who have investigated the use of bFGF in TMP
repair reported no adverse outcomes. Hakuba et al. [13] reported a
5% incidence of cholesteatoma formation with ve out of the six
cholesteatomas occurring in association with perforations that
involved the anterior half of the tympanic membrane. The
incidence of iatrogenic cholesteatoma following overlay myringo-
plasty is 37%, [13] indicating that cholesteatoma formation is not
particular to this regenerative technique. In Hakubas study [13]
cholesteatoma was rst identied up to 17 months post-treatment
suggesting that long-term follow-up is indicated for patients
undergoing this technique. Lou and Wang [8] reported the
development of myringitis in 52% of cases in whom gelatin sponge
was utilized, however CT imaging conrmed the absence of any
middle ear or mastoid inammation in these cases and the
myringitis did not have any effect on hearing outcome or patient
symptoms. None of our patients treated with bFGF and gelfoam
developed the myringitis described by Lou and Wang.
5. Weaknesses
One signicant weakness in this study is the small number of
participants recruited. This has a direct impact on the likelihood of
achieving statistically signicant results and is inherent to pilot
studies. Another important weakness is that this is not a
randomized trial and is without a control arm: the benchmark
standard for success against which this technique has been
compared is the success rate for a standard myringoplasty
procedure as published in the literature.
6. Conclusions
Results from our pilot study indicate that the use of bFGF in a
single application for the repair of TMPs in children shows promise.
The technique is faster, cheaper and more readily delivered than
classic myringoplasty techniques, with rates of success that are
higher than with published conventional myringoplasty literature
with minimal morbidity. Ongoing studies with regular postoperative
bFGF application, different scaffolds and long-term follow-up are
335
underway to further investigate the utility of this treatment modality
to ascertain whether this technique is appropriate for widespread
general use.
Conict of interest
None to declare.
References
[1] H. Gunasekera, S. Knox, P. Morris, H. Britt, P. McIntyre, J. Craig, The spectrum and
management of otitis media in Australian indigenous and nonindigenous chil-
dren: a national study, Pediatr. Infect. Dis. J. 26 (2007) 689692.
[2] Australian Institute of Health and Welfare (AIHW), The health and welfare of
Australias Aboriginal and Torres Strait Islander people, an overview 2011. [inter-
net]. AIHW. Available from: hhttp://www.aihw.gov.au/publication-detail/
?id=10737418989i, 2011 (accessed 25.09.13).
[3] S. Kumar, A. Acharya, E. Hadjihannas, C. Panagamuwa, A.L. McDermott, Pediatric
myringoplasty: denition of success and factors affecting outcome,
Otol. Neurotol. 31 (2010) 14171420.
[4] M. Mondain, S. Safedine, A. Uziel, Fibroblast growth factor improves the healing
of experimental tympanic membrane perforations, Acta Otolaryngol. 111 (1991)
337341.
[5] M. Fina, A. Baird, A. Ryan, Direct application of basic broblast growth factor
improves tympanic membrane perforation healing, Laryngoscope 103 (1993)
804809.
[6] M. Mondain, A. Ryan, Histological study of the healing of traumatic tympanic
membrane perforation after basic broblast growth factor application, Laryngo-
scope 103 (1993) 312318.
[7] Q. Zhang, Z. Lou, Impact of basic broblast growth factor on healing of tympanic
membrane perforations due to direct penetrating trauma: a prospective non-
blinded/controlled trial, Clin. Otolaryngol. 37 (2012) 446451.
[8] Z.-C. Lou, Y.-B.-Z. Wang, Healing outcomes of large (>50%) traumatic membrane
perforations with inverted edges following no intervention, edge approximation
and broblast growth factor application; a sequential allocation, three-armed
trial, Clin. Otolaryngol. 38 (2013) 289296.
[9] S.-I. Kanemaru, H. Umeda, Y. Kitani, T. Nakamura, S. Hirano, J. Ito, Regenerative
treatment for tympanic membrane perforation, Otol. Neurotol. 32 (2011)
12181223.
[10] N. Hakuba, M. Taniguchi, Y. Shimizu, A. Sugimoto, Y. Shinomori, K. Gyo, A new
method for closing tympanic membrane perforations using basic broblast
growth factor, Laryngoscope 113 (2003) 13521355.
[11] Z. Lou, Healing large traumatic eardrum perforations in humans using broblast
growth factor applied directly or via gelfoam, Otol. Neurotol. 33 (2012)
15531557.
[12] Z. Lou, L. Xu, J. Yang, X. Wu, Outcome of children with edge-everted traumatic
tympanic membrane perforations following spontaneous healing versus bro-
blast growth factor-containing gelfoam patching with or without edge repair,
Int. J. Pediatr. Otorhinolaryngol. 75 (2011) 12851288.
[13] N. Hakuba, N. Hato, Y. Omotehara, M. Okada, K. Gyo, Epithelial pearl formation
following tympanic membrane regeneration therapy using an atelocollagen/
silicone membrane and basic broblast growth factor: our experience from a
retrospective study of one hundred fty eight patients, Clin. Otolaryngol. (2013),
http://dx.doi.org/10.1111/coa.12164 (Epub ahead of print).
[14] R.S. Cotran, V. Kumar, T. Collins, Tissue repair: cellular growth, brosis and wound
healing, in: Pathologic Basis of Disease, W.B. Saunders Co., Philadelphia, 1999, pp.
107109.
[15] H. Kaftan, L. Reuther, B. Miehe, W. Hosemann, A. Beule, Inhibition of broblast
growth factor receptor 1: inuence on tympanic membrane wound healing in
rats, Eur. Arch. Otorhinolaryngol. 269 (2012) 8792.
[16] N. Hakuba, M. Iwanaga, S. Tanaka, Y. Hiratsuka, Y. Kumabe, M. Konishi, et al., Basic
broblast growth factor combined with atelocollagen for closing chronic tym-
panic membrane perforations in 87 patients, Otol. Neurotol. 31 (2009) 118121.