Acta Otorrinolaringol Esp.

2015;66(2):87---91

www.elsevier.es/otorrino

ORIGINAL ARTICLE

Otoacoustic Emissions as Cochlear Function Analyser

in Children With Language Disorders
Silvia A. Aguilera-Tello,a Ileana Gutirrez-Farfn,a Laura E. Chamlati-Aguirre,a
Efrn Alatorre-Miguel,b Alfredo Durand-Riverab,

a
Divisin de Audiologa y Otoneurologa, Instituto Nacional de Rehabilitacin, Mexico City, Mexico
b
Servicio de Neurorrehabilitacin, Laboratorio de Neuroproteccin, Instituto Nacional de Rehabilitacin, Mexico City, Mexico

Received 27 February 2014; accepted 25 May 2014

KEYWORDS Abstract
Language; Introduction and objectives: Speech perception that takes place in the cochlea is involved in
Transient otoacoustic the process of language. The objective was to describe the ndings in transient otoacoustic
emissions; emissions in children with language problems before and after 6 months of speech therapy.
Audiometry; Methods: There were 17 children with language problems between 3 and 6 years of age diag-
Language disorder; nosed with anarthric language delay (expressive and mixed language disorder). They underwent
Language therapy medical history, otoscopy, intelligence level testing, initial language test, tympanometry of
226 Hz, audiometry and transient otoacoustic emission test.
Results: We evaluated the 17 patients again after 6 months of attending speech therapy. The
percentage of overall reproducibility of transient otoacoustic emissions in both ears was ade-
quate to perform frequency analysis. We found a statistically signicant difference (P.01) in
the frequency of 1 kHz reproducibility when comparing results before and after therapy in the
right ear. There was a signicant difference (P.05) when comparing the results of audiometry
at frequencies of 0.5, 1.5, 2, 4 and 8 kHz in the right ear and a highly signicant difference
(P=.001) in the frequency of 3 kHz in the left ear.
Conclusions: The analysis of sound through the cochlea is involved in the process of language
acquisition. A poor processing of speech sounds in the peripheral system could result in poor
processing at the central level. Consequently, it is important to consider our results when making
a diagnosis and carrying out rehabilitation treatment in children with language disorders.
2014 Elsevier Espa na, S.L.U. and Sociedad Espa nola de Otorrinolaringologa y Patologa
Crvico-Facial. All rights reserved.

Please cite this article as: Aguilera-Tello SA, Gutirrez-Farfn I, Chamlati-Aguirre LE, Alatorre-Miguel E, Durand-Rivera A. Emisiones
otoacsticas como analizador de la funcin coclear en ni nos con problemas de lenguaje. Acta Otorrinolaringol Esp. 2015;66:87---91.
Corresponding author.

E-mail address: alfredo.durand@gmail.com (A. Durand-Rivera).

2173-5735/ 2014 Elsevier Espa

na, S.L.U. and Sociedad Espa
nola de Otorrinolaringologa y Patologa Crvico-Facial. All rights reserved.
88 S.A. Aguilera-Tello et al.

PALABRAS CLAVE Emisiones otoacsticas como analizador de la funcin coclear en ni

nos con problemas
Lenguaje; de lenguaje
Emisiones
otoacsticas Resumen
transitorias; Introduccin y objetivos: La percepcin del sonido que se lleva a cabo en la cclea est involu-
Audiometra; crada en el lenguaje. El objetivo fue describir los hallazgos encontrados en las emisiones
Trastorno de otoacsticas transitorias (EOT) en ni nos con problemas de lenguaje antes y despus de 6 meses
lenguaje; de terapia.
Terapia de lenguaje Mtodos: A 17 ni nos diagnosticados como anrtricos, se les realiz historia clnica, escala de
inteligencia, prueba de lenguaje inicial y estudios audiolgicos.
Resultados: Al hacer la comparacin del inicio y despus de 6 meses de terapia, encontramos,
en el anlisis frecuencial de las EOT una p 0,05 en 1 kHz en el odo derecho, en la audiometra;
se encontr una p 0,05 en las frecuencias de 0,5, 1,5, 2, 3, 4 y 8 kHz en el odo derecho y en
las frecuencias de 2, 3, 4 y 8 kHz en el odo izquierdo.
Conclusiones: El anlisis del sonido a travs de la cclea est involucrado en el proceso de
adquisicin del lenguaje. Un mal procesamiento de los sonidos del lenguaje en el sistema
perifrico podra dar como resultado un mal procesamiento a nivel central. Por esto, es impor-
tante tener nuestros resultados en consideracin al realizar el diagnstico y el tratamiento
rehabilitador en los ninos con trastornos de lenguaje.
2014 Elsevier Espa na, S.L.U. and Sociedad Espa nola de Otorrinolaringologa y Patologa
Crvico-Facial. Todos los derechos reservados.

Introduction cochlear function, they could be used to verify whether

the cochlea is somehow involved in language problems and
It is generally considered that an alteration in language func- could represent an objective marker used before and after
tion reects a brain abnormality and, more specically, of language therapy to evaluate evolution.
the dominant hemisphere.1---3
Neurolinguistics assumes that the production of the
acoustic signal of speech represents the result of articula- Methods
tory maneuvers performed by the speaker, which are the
product of a series of signals transmitted by the central We conducted a cross-sectional and descriptive study
nervous system.4---6 of 25 patients, aged between 3 and 6 years, with a diag-
Ardila, modifying the model of Fant,7 proposed that nosis of anarthric language delay17 (expressive and mixed
acoustic signals reaching the cochlea undergo a tonotopic language disorder), and a normal intelligence quotient as
analysis. Subsequently, this afferent information undergoes corroborated by the WPPSI (Wechsler Preschool and Primary
processing, from the brainstem to the cortex, in order to Scale of Intelligence) and Hiskey-Nebraska tools. Subjects
produce language recognition.4,8,9 were audiologically evaluated before and 6 months after
Adequate temporal processing is one of the most impor- rehabilitation treatment.
tant factors for language intelligibility. Slower rates of Once the protocol was approved by the Ethics and
modulated amplitude, from 4 to 20 Hz, are important in Research Committee of the National Rehabilitation Institute
segmenting language into smaller units.10 and we had obtained informed consent in writing from the
Otoacoustic emissions are considered to be the product parents or tutors of the children, we elaborated a medical
of cochlear amplication, which represents a phenomenon history and proceeded to conduct the following studies:
whereby the cochlea is responsible of selecting frequencies
with great sensitivity and a wide dynamic range. The discov- Tympanometry at 226 Hz, conducted using a Madsen Zodiac
ery of the motility of outer hair cells by Brownell in 198311---15 901 tympanometer in a soundproofed chamber on both
provided a physical substrate for active processes in the ears to rule out middle ear pathology and verify the nor-
cochlea, since the measurement of the individual force gen- malization of tympanometry curves (type A). This was a
erated by outer hair cells indicates that the activation of a requirement before conducting tone audiometry and tran-
sufciently large number of cells would be capable of chang- sient otoacoustic emissions (TOAE) testing.
ing a portion of the mechanical response of the cochlea.13,16 Tone audiometry. We used a Madsen Orbiter 922 audiome-
This phenomenon could be regarded as a cochlear tuner. ter in a soundproofed chamber, to obtain the minimal
Taking the foregoing into account, we believe that infor- detection of pure tones at the frequencies of 125, 250,
mation analysis by the cochlea has a greater involvement in 500, 1000, 1500, 2000, 3000, 4000 and 8000 Hz.
the process of language acquisition than currently thought. TOAE. We used an Otodynamics ILO-96 device in a sound-
Since otoacoustic emissions provide a means of studying proofed chamber, placing a pediatric probe rst in the
Otoacoustic Emissions as Cochlear Function Analyser in Children With Language Disorders 89

Table 1 Transient Otoacoustic Emissions.

Before the After 6 months P Before the After 6 months P
therapy of therapy therapy of therapy
Right ear (%) Left ear (%)
Overall 90.9 96.2 .1843 93.9 96.4 .4889
1.0 kHz 57.7 80.8 .0500 77.5 89.9 .1581
1.5 kHz 86.4 95.4 .0755 90.6 96.5 .2340
2.0 kHz 91.3 96.8 .2490 96.5 96.2 .9166
3.0 kHz 92.1 97.2 .2457 95.7 96.7 .7158
4.0 kHz 88.7 96.9 .1970 93.0 95.3 .5817
Mean percentage of reproducibility of transient otoacoustic emissions in both ears, before and after 6 months of language therapy. There
is a statistically signicant difference (P<.05) in the right ear, at the frequency of 1 kHz.

right external auditory canal and then in the left. The The results of frequency analysis (1, 1.5, 2, 3 and 4 kHz)
reproducibility percentage results were registered into the of reproducibility of TOAE in both ears before and after
database, overall and by frequency (1, 1.5, 2, 3 and 4 kHz). 6 months of undergoing language therapy are shown in
Initial language test (ILT). Designed and standardized for Table 1.
Spanish-speaking children in Mexico aged between 3 years We performed the Student t test for related samples,
and 7 years and 11 months. The ILT measures the compo- nding a statistically signicant difference (P.05) in the
nents of language mentioned previously, proving that all frequency of 1 kHz when comparing reproducibility before
patients suffered a language problem prior to the therapy. and after therapy in the right ear (Table 1 and Fig. 1).
Tone audiometry in all the evaluated frequencies (125,
250, 500, 1000, 1500, 2000, 3000, 4000 and 8000 Hz), iden-
Results tied responses below 20 dB, considered as normal bilateral
hearing.
We ruled out four of the initial 25 patients (one due to Table 2 shows the mean values of responses obtained by
submucosal cleft palate and three due to middle ear prob- tone audiometry in each of the frequencies before and after
lems that could not be resolved by medical treatment), language therapy.
thus leaving 21 patients. Of these, four dropped out after We performed the Student t test for related samples,
not attending language therapy sessions, leaving a nal uni- nding a signicant difference (P.05) when comparing the
verse of 17 patients evaluated after 6 months of therapy. results of the audiometry at frequencies of 0.5, 1.5, 2, 3, 4
The mean age of the patients was 4.5 years, and there and 8 kHz in the right ear and a very signicant difference
were 10 males and seven females. All subjects presented (P=.05) in the frequencies of 2, 3, 4 and 8 kHz in the left ear
an intelligence quotient within normal parameters. The ITL (Table 2).
was below the level of normality before the start of lan-
guage therapy in all the children evaluated, but these results
were not analyzed in depth, as this was not the aim of the Discussion
investigation.
The overall percentage of reproducibility of TOAE in both Out of the total 25 patients evaluated, we found that 4% pre-
ears was adequate for all patients before and after language sented a submucosal cleft palate. This is a high percentage
therapy, thus conrming normal bilateral hearing. if we consider that the incidence in Mexico is of 3521 new

Figure 1 Representative images of transient otoacoustic emissions before (left) and after (right) the therapy. It is important to
note the improvement in reproducibility in the frequency of 1 kHz in the right ear.
90 S.A. Aguilera-Tello et al.

Table 2 Tone Audiometry.

Before the After 6 months P Before the After 6 months P
therapy of therapy therapy of therapy
Right ear (dB) Left ear (dB)
125 Hz 18.5 16.5 .1302 17.1 17.4 .7731
250 Hz 15.9 13.8 .0895 16.1 15.3 .4220
500 Hz 13.8 10.3 .0232 13.5 10.9 .1979
1000 Hz 12.1 10.3 .0826 11.2 8.5 .1076
1500 Hz 12.9 9.4 .0232 11.5 9.1 .1492
2000 Hz 12.1 6.8 .0051 10.0 5.6 .0235
3000 Hz 11.2 7.1 .0299 12.1 4.7 .0026
4000 Hz 9.4 4.1 .0239 9.4 4.4 .0052
8000 Hz 10.0 5.0 .0121 11.8 7.4 .0308
Mean values in dB of the results of the tone audiometry in both ears before and after 6 months of therapy. There is a statistically
signicant difference (P<.05) in the right ear at the frequencies of 0.5, 1.5, 2, 3, 4 and 8 kHz and in the left ear at the frequencies of 2,
3, 4 and 8 kHz (gures marked in bold).

cases per year, which corresponds to 9.6 new cases per day. modulator of the cochlear response. Thus, in our patients
A prevalence of 139 000 Mexicans affected by some form of with language disorders, the reproducibility in TOAE at
cleft lip or palate was reported in 2003.18 These gures indi- 1 kHz is under the normal values, with a mean of 57.6%, and
cate that this pathology is probably underdiagnosed in our improving up to 80.7% after 6 months of language therapy
country. (Table 1). From this we can infer that language disorders
Among our patients, 12% presented some form of mid- with a phonological etiology involve a malfunction of outer
dle ear problem that did not remit after medical treatment hair cells,21 which improve after specic therapy.
and 16% dropped out of language therapy after not attend- Moreover, we consider that the superior olivary complex
ing. This percentage is worrying, as it indicates desertion fullls an essential role in binaurality, being the rst loca-
of rehabilitation treatment for language problems at our tion of crossed information to reach the cochlear nuclei.
institute. This is the origin of the efferent bers that constitute the
Regarding the evaluated population, 41% were females medial olivo-cochlear bundle, creating synapses directly
and 59% were males. These percentages are not surprising, in the base of the outer hair cells.14 Therefore, we can
as several studies have reported that language disorders are assume that if TOAE evaluate the functionality of these
more common among males.17 cells,11---14 there is a malfunction at a cochlear level in one
It was also expected that the overall reproducibility of or more of the frequencies evaluated. In our study, this
TOAE would be over 70% in both ears, both before and was 1 kHz, which is the mean frequency par excellence for
6 months after language therapy. In the case of our study, correct intelligibility of spoken words. Another study found
we observed a percentage over 90%, due to the fact that alterations in other TOAE frequencies, specically 3, 4 and
all patients presented normal bilateral hearing, so we could 5 kHz, when evaluating patients with language disorders.21
conrm that the middle ear did not suffer any alterations As a result of this malfunction, there is no adequate treat-
and the function of outer hair cells was normal. The lit- ment of auditory information for the sounds of language.
erature contains reports estimating normal values for TOAE Moreover, it is well-known that the right ear is a crossed
over 70% with reproducibility among individuals with normal reex of the olivo-cochlear bundle originating in the left
hearing.19,20 We did not nd any signicant differences in the hemisphere (the hemisphere specializing in language).14
overall reproducibility in both ears when evaluating subjects When analyzing the results of the audiometry we found a
before and after undergoing language therapy, but it is worth signicant difference (P.05) when we compared the results
mentioning that the right ear presented an increase, from at middle and high frequencies in the right ear and at high
a mean value of 90.9% before the therapy to 96.15% after frequencies in the left ear (Table 2).
6 months of therapy (Table 1 and Fig. 1). It is important to highlight that there was also a statis-
An interesting part is the analysis of frequency repro- tically signicant improvement in these frequencies studied
ducibility of TOAE. We found a statistically signicant in our sample of patients with language disorders, with the
difference (P.01) at a frequency of 1 kHz when comparing right ear also reecting a greater improvement than the left,
reproducibility before and after therapy in the right ear. thus supporting our ideas.
This nding could be explained and support the theories
reported in other studies.21 We know that hair cells are
the real auditory receptors in the cochlea, specically in Conclusions
the organ of Corti. This is particularly true of outer hair
cells which, due to their morphological characteristics, We can conclude that the peripheral analysis of auditory
are capable of modifying the mechanical response of the information conducted in the cochlea is involved in the pro-
cochlea and inuence inner hair cells, probably acting as a cess of language acquisition. Therefore, a wrong processing
Otoacoustic Emissions as Cochlear Function Analyser in Children With Language Disorders 91

of language sounds by the peripheral system, as proven by 8. Dronkers NF, Pinker S, Damasio A. Lenguaje y afasias. In: Kandel
TOAE and tone audiometry, could result in auditory informa- ER, Schwartz JH, Jessell TM, editors. Principios de neurociencia.
tion arriving distorted at the level of the auditory cortex. 4.a ed Mxico: McGraw-Hill; 2001. p. 1169---87.
This makes it very important to take the results into account 9. Malmierca MS. The structure and physiology of the rat
when establishing a diagnosis and rehabilitation treatment auditory system: an overview. Int Rev Neurobiol. 2003;56:
147---211.
through language therapy in children with phonological
10. Poelmans H, Luts H, Vandermosten M, Boets B, Ghesquire P,
language disorders. These results suggest we should use Wouters J. Auditory steady state cortical responses indicate
otoacoustic emissions and audiometry as an objective deviant phonemic-rate processing in adults with dyslexia. Ear
marker before and after language therapy to evaluate Hear. 2012;33:134---43.
patient evolution. It is important to continue investigating 11. Hudspeth AJ. Transformacin sensitiva en el odo. In: Kandel ER,
in this eld in order to decipher the role of peripheral Schwartz JH, Jessell TM, editors. Principios de neurociencia.
auditory processing in language, as well as to establish the 4.a ed. Mxico: McGraw-Hill; 2001. p. 614---22.
usefulness of otoacoustic emissions and audiometry for the 12. Kemp DT. Otoacoustic emissions, their origin in cochlear func-
evaluation of patients with language disorders. tion, and use. Br Med Bull. 2002;63:223---41.
13. Robinette MS, Glattke TJ. Otoacoustic emissions clinical appli-
cations. New York: Thieme; 1997.
Conict of Interests 14. Dlano P, Robles I, Robles L. Sistema eferente auditivo. Rev
Otorrinolaringol Cir Cabeza Cuello. 2005;65:55---62.
The authors have no conict of interests to declare. 15. Oostenbrink M, Verhaagen W. Otoacoustic emissions. Am J End
Technol. 2004;44:189---98.
16. Iwasa KH, Chadwick RS. Elasticity and force generation
References of cochlear outer hair cells. J Acoust Soc Am. 1992;92:
3169---73.
1. Adams-Raymond VM. Principles of neurology. 4th ed. Singapore: 17. Azcoaga JE. Retardos del lenguaje en el ni no. Buenos Aires:
McGraw-Hill; 1989. Paids; 1984.
2. Damasio AR, Damasio H. Cerebro y lenguaje. Investigacin y 18. Trigos MI, Guzmn LFME. Anlisis de la incidencia, prevalencia
Ciencia. 1992;194:59---66. y atencin del labio y paladar hendido en Mxico. Cir Plast.
3. Mesulam MM. Large scale neurocognitive networks and dis- 2003;13:35---9.
tributed processing for attention, language, and memory. Ann 19. Harrison WA, Norton SJ. Characteristics of transient evoked
Neurol. 1990;28:597---613. otoacoustic emissions in normal hearing and hearing impaired
4. Ardila A. Psicologa del lenguaje. Mxico: Trillas; 1983. children. Ear Hear. 1999;20:75---86.
5. Ardila A. Neurolingstica. Mxico: Trillas; 1984. 20. Zimatore G, Giuliani A, Parlapiano C, Grisanti G, Colosimo A.
6. Durand-Rivera A. Breve revisin de modelos de lenguaje. Para Revealing deterministic structures in click-evoked otoacoustic
la Salud. 2002;6:32---7. emissions. J Appl Physiol. 2000;88:1431---7.
7. Fant G. Auditory patterns of speech. In: Walthem-Dunn W, 21. Durand-Rivera A, Manzano-Martnez E. El procesamiento
editor. Models for perception of speech and visual form. coclear est involucrado en los problemas de lenguaje? Rev
Cambridge: MIT Press; 1967. p. 215---24. Neurol. 2009;48:639---44.