Secretory otitis media (SOM) is a clinical entity characterized by the presence of effusion in the middle ear, without perforation in the eardrum and acute infectious process for a period of three months. It is common in children between three and nine years old. The main symptom is a hearing loss, that is usually noted by parents or teachers, due to lack of attention and interest, request for repetition of the message several times and poor performance in school. The etiology is multifactorial and the highest incidence is caused by eustachian tube dysfunction and infections of the upper airways of allergic, viral, or infectious origin. With the advancement of the age, the maturity of the immunological system is completed, as well as the growth of the auditory tube, which decreases the occurrence of the disease [
The diagnosis is done by otoscopy and confirmed by audiological evaluation. It is possible to visualize by otoscopy and, frequently, a retracted eardrum with decreased mobility, opaque appearance, and abnormal color. In the audiologic evaluation, the diagnosis is a mild to moderate conductive hearing loss, usually bilateral, with a type B tympanometric curve. The hearing loss is fluctuating, temporary, and asymmetric [
The management of treatment could be clinical or surgical and depends on the middle ear conditions and a clinical history [
The central auditory processing (CAP) battery evaluates the effectiveness of the central nervous system’s ability to process changing acoustic stimuli [
When children are deprived of normal auditory input early in life, they can face CANS changes and diminished perceptual sensitivity to process auditory information later in life [
Therefore, the main purpose of this study was to analyze the long-term effect of otitis media in the peripheral and central auditory system, through behavioral and electrophysiological tests, in children with a documented history of SOM and a bilateral tubes placement insertion in the first six years of life.
This is a prospective cross-sectional study conducted at the Laboratory of Audiology from Department of Human Development and Rehabilitation/School of Medical Sciences from the State University of Campinas (Unicamp/Brazil), after its approval by the Ethics Committee (protocol 889074). Written informed consent was obtained for all participants.
A total of 90 children, aged from 8 to 16 years old (mean 10.98 years old / 45 boys and 45 girls) from public school, participated.
The participants were divided into two groups: (i) the control group (CG) consisted of 40 children (17 boys and 23 girls, mean age of 10.7 years) with no history of otitis media and (ii) the experimental group (EG) consisted of 50 children (28 boys and 22 girls, mean age of 11.2 years) with a documented history of bilateral SOM in their first six years of life and with bilateral tympanostomy tube insertion.
The CG was recruited by the researcher at the state school and a questionnaire was filled by the parents about the child’s health history. The EG was selected from the medical records at the State Hospital between 2000 and 2009 by the researchers.
The inclusion criteria for CG were as follows: Age between 8 and 16 years old Right handed Normal otoscopy bilaterally Hearing levels bilaterally within normal limits at the time of assessment (pure-tone audiometry thresholds below 20 dBHL at 250 to 8000 Hz) [ Normal middle ear function (Type A) defined as a peak compliance within 0.3 to 1.3 mmhos and peak pressure within −100 to +20 daPa with the presence of ipsilateral and contralateral acoustic reflexes bilaterally between 70 and 100 dB for 500Hz, 1, 2, 3 and 4KHz [ Typical development: good performance at school and language development, absence of attention disorder and auditory and respiratory complains
The inclusion criteria for EG were as follows: Age between 8 and 16 years old Right handed Normal otoscopy bilaterally Hearing levels bilaterally within normal limits at the time of assessment (pure-tone audiometry thresholds below 20 dBHL at 250 to 8000 Hz) [ Normal middle ear function (Type A) defined as a peak compliance within 0.3 to 1.3 mmhos and peak pressure within −100 to +20 dPa with the presence of ipsilateral and contralateral acoustic reflexes bilaterally between 70 and 100 dB for 500Hz, 1, 2, 3 and 4KHz [ Documented history of three episodes of SOM and only one set of bilateral tympanostomy tubes placement surgery in the first six years of life Absence of middle ear infections for the last 12 months until the date of the evaluation
Children with behavioral or neurological disorders and/or genetic syndromes, including those using psychoactive medication or attending speech therapy, were excluded from the sample.
The protocol was composed based on three stages: hearing assessment, behavioral evaluation of central auditory processing, and electrophysiological evaluation.
For audiologic evaluation and CAP assessment, the audiometer AC-40-Interacoustics, TDH 39P headphones, and a Dell computer were used. In the electrophysiological evaluation, the equipment used was Biologic Navigator Pro-Natus. Immitanciometry was performed using the Interacoustics 235h. All equipment was calibrated according to ISO-389 and IEC-645 standards.
The parents were interviewed to obtain more information such as otological history and school performance. Next the hearing thresholds were assessed from 250 to 8000 Hz. Subsequently, speech recognition was assessed at 40dB HL using a list of 25 monosyllabic words from Portuguese in each ear, with a percentage of correct answers greater than 88% [
The tympanometry was obtained with the 226Hz probe. The contralateral and ipsilateral acoustic reflexes were performed in the frequencies of 500, 1000, 2000, 3000, and 4000Hz.
The tests were performed in one 45-minute session in a soundproof condition. The tests applied were dichotic digits (DD), synthetic sentence identification (SSI), gaps-in-noise (GIN), and frequency pattern test (FPT) [
It was performed in a 60-minute session in a sound proofed and electrically shielded room. Before the beginning of the collection, the skin of each subject was cleaned in the places where the electrodes were fixed through an abrasive paste. Afterwards, the electrodes were placed with an electrolytic paste and with the aid of an adhesive tape the impedance was kept below 3kΩ and the interelectrode impedance less than 2kΩ.
During the evaluation, the subjects were instructed to keep their eyes closed in order to avoid artifacts. In 50% of the patients the assessment was initiated by the right ear, while the remaining 50% by the left ear. All electrophysiological assessments were performed monoaurally.
The electrophysiological evaluation was composed by three phases in the order below. The tests started on the right ear in 50% of the participants and in the left in the others 50%. Auditory Brainstem Response with click stimulus (nonverbal) Frequency Following Response – FFR (verbal) Late Auditory Evoked Potentials with tone burst stimulus (nonverbal)
(a) Auditory Brainstem Response with click (nonverbal): the electrodes were positioned according to the 10-20 system [
(b) Frequency Following Response (FFR): the test was performed with the same electrodes positioned for click ABR. The response was elicited using a 40 ms synthetic speech syllable /da/, provided by the BioMARK software and recorded by the Biologic Navigator Pro (Natus Medical). The stimulus consists of the consonant /d/ (transient portion or onset) and the short vowel /a/ (sustained portion or following frequency response). Two traces were performed twice with 3000 stimuli and free of artifacts. Subsequently, the responses were added giving rise to a third wave composed by 6000 stimuli. In the present study the FFR evaluation was performed by time domain analysis. VA complex measures (slope, related to the temporal synchronization of the response generators and area, related to the activity that contributes to wave generation) were also performed [
The parameters used for ABR and FFR are described in Table
Parameters of acquisition of the Click-ABR and FFR.
PARAMETERS | Click-ABR | FFR |
---|---|---|
Equipment | Biologic Navigator Pro | Biologic Navigator Pro |
Stimulated Ear | RE and LE | RE and LE |
Stimulus | Not verbal | Verbal |
Type of the stimulus | Click | Speech |
Duration of the stimulus | 0.1 msec | 40 msec |
Polarity of the stimulus | Rarefaction | Alternate |
Intensity of the stimulus | 80 dBHL | 80 dB SPL |
Speed of the stimulus | 19.3/sec | 10.9 sec |
Number of sweeps | 2000 | 6000 |
Replicability | 2 collections of 2000 | 2 collections of 3000 |
Filter | 100-1500 Hz | 100-2000 Hz |
Window | 10.66 msec | 85.33 msec |
Transducer | Insert (ER-3A; Natus Medical) | Insert (ER-3A; Natus Medical) |
(c) Late auditory evoked potentials with tone burst stimulus (nonverbal) were recorded with the active electrode positioned on the vertex (Cz), the reference electrodes on the right (M2) or left (M1) mastoids and the ground electrode at the Fz position, according to the 10–20 system [
Parameters of acquisition of the LLAEP with nonverbal stimulus.
PARAMETERS | NONVERBAL |
---|---|
Equipment | Biologic Navigator Pro |
Stimulated Ear | RE and LE |
Type of stimulus | Tone burst |
Frequent stimulation | 1000Hz (80%) |
Infrequent Stimulus | 2000Hz (20%) |
Polarity of the stimulus | Alternate |
Intensity of the stimulus | 75 nHL |
Speed of the stimulus | 1.1/sec |
Number of sweeps | 300 |
Filter | 1- 30 Hz |
Window | 533 msec |
Transducer | Insert (ER-3A; Natus Medical) |
The latencies and amplitudes values of ABR, FFR and Late Auditory Evoked Potentials were viewed and marked manually by two blinded audiologists to avoid influence on the results. When there was difference in the marking, a third blinded audiologist analyzed the results and remained the mark that coincided with two equal analyses.
The groups were compared using ANOVA, for ABR, FFR and Late Auditory Evoked Potentials responses. The Wilcoxon-Mann-Whitney test was used for CAP responses. The gender and side were included in both models as a fixed effect, as well as their interactions. When the interaction effect between side and group was considered significant (p <0.05), the ears and gender were analyzed separately.
To test the homogeneity of the contingency tables, Pearson’s Chi-square test was applied, setting the significance level of 0.05.
The statistical analyses were made through the software R-project (
The distribution of the sample considering male and female gender and age group can be observed in Table
Statistical analysis of the sample considering the gender and age between groups.
CG | EG | p-valor | |
---|---|---|---|
Number of children | 40 | 50 | |
Age (Mean, years) | 10.7 | 11.2 | 0.455 |
Gender (number) | |||
Male | 17 | 28 | 0.19 |
Female | 23 | 22 | 0.19 |
Table
Mean values of hearing thresholds in the right and left ears between control and experimental groups.
250Hz | 500Hz | 1000Hz | 2000Hz | 3000Hz | 4000Hz | 6000Hz | 8000Hz | |
---|---|---|---|---|---|---|---|---|
RE-CG | 8 dB | 7.5 dB | 6.5 dB | 6 dB | 4.5 dB | 5.5 dB | 12.5 dB | 8.5 dB |
RE-EG | 8.3 dB | 7.2 dB | 5.5 dB | 5 dB | 4.4 dB | 5 dB | 12.2 dB | 7.2 dB |
p- value | 0.589 | 0.200 | 0.361 | 0.687 | 0.358 | 0.324 | 0.950 | 0.198 |
LE- CG | 8 dB | 7 dB | 5 dB | 7,5 dB | 4 dB | 7 dB | 8,8 dB | 6,5 dB |
LE-EG | 8.8 dB | 6.1 dB | 4.4 dB | 7 dB | 5 dB | 5 dB | 10 dB | 5 dB |
p- value | 0.892 | 0.301 | 0.486 | 0.154 | 0.909 | 0.150 | 0.926 | 0.672 |
Legend: RE: right ear; LE: left ear; CG: control group; EG: experimental group.
Table
Behavioral evaluation values of central auditory processing between control and experimental groups.
CG | EG | ||||||
---|---|---|---|---|---|---|---|
Test | N | Mean | SD | N | Mean | SD | P-value |
|
|||||||
RE | 40 | 98.93% | 1.86 | 50 | 95.40% | 5.16 |
|
LE | 40 | 97.93% | 4.15 | 50 | 92.55% | 7.95 |
|
|
|||||||
Humming | 80 |
93.00% | 12.4 | 100 |
83.40 |
18.5 |
|
Verbalizing | 80 |
73.50% | 21.2 | 100 |
42.7% | 22.2 |
|
|
80 |
67.5% | 13.9 | 100 |
59.8% | 16.9 |
|
|
|||||||
RE | 40 | 4.65ms | 1.00 | 50 | 6.22ms | 1.40 |
|
LE | 40 | 4.72ms | 1.06 | 50 | 6.56ms | 1.52 |
|
Legend: n: number;
In the analysis of the ears for the ABR click, FFR and Late Auditory Evoked Potentials tests, no difference was observed between groups, in the measures of latencies and amplitudes. For this reason, the data of the two ears were combined in the other analyses. Considering the analyses for gender in the ABR, FFR and Late Auditory Evoked Potentials tests no significant differences were observed, only for slope VA (p=0,021).
The Auditory Brainstem Response with click stimulus measures showed a significant increase for latencies and decrease for amplitudes of waves III (0.1ms and 0.06
Latency(ms) and amplitude((
CG (n=80) | EG (n=100) | CGxEG | |||
---|---|---|---|---|---|
Mean | SD | Mean | SD | p-value | |
|
|||||
Latency | 1.57 | 0.08 | 1.63 | 0.10 | 0.06 |
Amplitude | 0.21 | 0.11 | 0.19 | 0.09 | 0.161 |
|
|||||
Latency | 3.71 | 0.11 | 3,81 | 0.15 |
|
Amplitude | 0.32 | 0.13 | 0.26 | 0.09 |
|
|
|||||
Latency | 5.59 | 0.14 | 5.69 | 0.17 |
|
Amplitude | 0.24 | 0.10 | 0.19 | 0.12 |
|
|
|||||
Latency | 2.14 | 0.11 | 2.19 | 0.15 | 0.124 |
|
|||||
Latency | 4.01 | 0.13 | 4.06 | 0.17 | 0.246 |
|
|||||
Latency | 1.87 | 0.11 | 1.88 | 0.11 | 0.977 |
Legend: CG: control group; EG: experimental group; SD: standard deviation.
Box plots showing the median, interquartile, and range of latency (ms) of ABR for both control and EG groups.
The analysis of the Long Latency Evoked Potential between the control and experimental groups showed a statistically significant difference of P2, N2, and P300. The potential P2 had increased 9.21ms, N2 16.5ms, and P300 13.41ms in the EG compared to CG (see Table
Latency (ms) and amplitude (
CG(n=80) | EG(n=100) | CG x EG | |||
---|---|---|---|---|---|
Mean | SD | Mean | SD | p-value | |
|
|||||
Latency | 107.7 | 23.19 | 108.9 | 19.38 | 0.864 |
amplitude | 3.56 | 1.64 | 2.92 | 2.12 | 0.091 |
|
|||||
Latency | 150.45 | 25.51 | 159.66 | 23.84 |
|
amplitude | 3.47 | 1.38 | 3.71 | 2.73 | 0.288 |
|
|||||
Latency | 202.67 | 31.87 | 219.17 | 35.51 |
|
amplitude | 4.75 | 2.30 | 3.86 | 3.38 | 0.063 |
|
|||||
Latency | 317.19 | 30.75 | 330.6 | 39.27 |
|
amplitude | 5.52 | 2.13 | 5.42 | 2.42 | 0.794 |
Legend: CG: control group; EG: experimental group; SD: standard deviation.
Box plots showing the median, interquartile, and range of latency (msec) of P300 for both control and EG groups.
For the FFR, it was verified that children from the EG presented an increase in latency values of all FFR components (V, A, C, D, E, F, and O waves) associated with a decrease in slope VA, in the female gender, comparing to CG (see Table
Latency (ms), Área VA (ms x
Measure | Groups | |||||||
---|---|---|---|---|---|---|---|---|
CG | EG | |||||||
Sex | N | Mean | SD | N | Mean | SD | p value | |
V | 80 |
6.50 | 0.21 | 100 |
6.80 | 0.24 |
| |
A | 80 |
7.47 | 0.34 | 100 |
7.85 | 0.32 |
| |
C | 80 |
18.37 | 0.44 | 100 |
19.15 | 1.51 |
| |
D | 80 |
22.29 | 0.57 | 100 |
23.44 | 1.94 |
| |
E | 80 |
30.83 | 0.56 | 100 |
32.40 | 2.54 |
| |
F | 80 |
39.29 | 0.52 | 100 |
40.75 | 2.66 |
| |
O | 80 |
47.97 | 0.65 | 100 |
49.39 | 2.52 |
| |
Área VA (ms x |
80 |
0.32 | 0.13 | 100 |
0.30 | 0.28 | 0.157 | |
Slope VA (ms / |
80 |
0.35 | 0.14 | 100 |
0.28 | 0.10 |
| |
M | 30 |
0.31 | 0.11 | 56 |
0.27 | 0.09 | 0.198 | |
F | 50 |
0.39 | 0.14 | 44 |
0.29 | 0.10 |
|
Legend: n: number
Latency (ms) values of FFR waves between groups.
This study was carried out with the purpose of analyzing the functioning of CANS in children with a history of bilateral SOM in the first six years of life with tympanotomy surgery for bilateral insertion of ventilation tubes.
Analyzing the mean responses of the CG and EG based on frequencies from 250Hz to 8KHz, both groups had equal hearing thresholds at the moment of the evaluation. Thus, in the EG it was found that the SOM did not cause a long-term negative effect in the peripheral system until the VIII cranial pair. The structures, mainly of the middle ear, recovered after the end of the disease. The comparative analysis of the hearing thresholds between the groups was important to show that the peripheral portion of the auditory system, probably, did not interfere in the responses of the behavioral evaluations of the CAP and electrophysiological measures.
In the analysis of behavioral CAP responses, the EG showed a significant difference when compared to the CG. Thus, the children from EG could have difficulty processing the speech perception in the presence of background noise and combining auditory inputs from the two ears, in particular the integration of subtle timing, level, and spectral differences in the signals.
Our findings corroborate with the literature that studied the influence of OM in children and verified worse performance in auditory abilities [
The results revealed significant latency delays and reduced amplitude of waves III and V for ABR and for FFR in children with a history of otitis media. The potentials P2, N2, and P300 also showed significant latency delays in children from EG. An increase in latency for N1 would not have been expected since N1 represents acoustic perception.
Regarding ABR, several studies have also described difference in latency and amplitude values in children with a history of OM [
Maruthy and Mannarukrishnaiah [
For FFR, few studies have been found. El-Kabarity et al. [
Thus, our results demonstrated the negative effects of SOM in children, related to the maturation and functioning of the auditory pathways.
The lower results obtained in the EG in both CAP and electrophysiological behavioral tests may have been due to the fact that recurrent SOM episodes caused, in the acute phase of the disease, an auditory sensorial deprivation, fluctuating, and often asymmetric hearing loss, in a critical period for the child’s development.
As a consequence, the CANS received inconsistent, incomplete, and often different auditory information, considering the right and left ears, for an extended time, once the time between clinical treatments and the decision to perform the surgery can be long. Studies have shown that fluids remaining in an acute episode of OM remain in the middle ear for three to 12 months, and in 10 to 30% of children, the fluid remains for two to three months. Thus, a child who had three to four SOM episodes may have twelve months of conductive hearing loss at a time considered critical for their development and learning [
Another consequence of these unfavorable conditions of stimulation may be a maturational delay in the structures of the CANS, a decrease in the number of stimulated nerve fibers and transmission. These changes in the CANS can interfere in the efficiency of the analysis and interpretation of the auditory stimuli, mainly related to the auditory abilities of figure background, ordering and temporal resolution which is fundamental for the development of speech, language and school performance.
The negative effects of otitis media in the measures of long latency auditory evoked potentials and FFR in the present study lead us to hypothesize that auditory pathway is affected from the brainstem level to the cortical level.
Thus, the effective diagnosis and medical treatment are essential. The earlier intervention in cases of otitis media can avoid the length of time of auditory fluctuation and minimize the effects caused by the fluid in the middle ear in the development of the auditory abilities. Also, it is important to refer all children who had a history of otitis media in childhood to an auditory evaluation once we observed that these individuals may have a risk to have a Central Auditory Processing Disorder.
It should be emphasized that more research regarding the effects of OM on behavioral and electrophysiological assessments should be made to guide parents and health professionals about the importance of hearing care, especially in the first years of life.
From the analysis of the results, the following was concluded.
There was a negative effect of otitis media on auditory abilities and electrophysiological measures in children with a history of otitis media. Concerning auditory abilities, the alterations observed were figure-background, ordering and temporal resolution. Electrophysiological tests revealed alteration from the brainstem to the cortical level.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that there are no conflicts of interest regarding the publication of this article.
This work was supported by grant 2014/04039-1, São Paulo Research Foundation (FAPESP).