Electrophysiology assessment of auditory system in individuals with developmental persistent stuttering

Objetivo: descrever os achados dos exames eletrofisiológicos de indivíduos com gagueira e comparar com indivíduos com desenvolvimento típico. Métodos: participaram desta pesquisa 34 indivíduos, de ambos aos gêneros, com idade entre 7 e 31 anos. O grupo pesquisa foi constituído por 13 crianças (G1a) e 4 adultos (G1b) diagnosticadas com gagueira e o grupo controle por 13 crianças (G2a) com bom desempenho acadêmico e 4 adultos (G2b) com desenvolvimento típico. Foi realizada a avaliação dos potenciais auditivos, na varredura de frequência e de duração. Resultado: quando comparados os grupos de crianças gagas e controle, foi observado que as crianças gagas apresentaram na varredura de frequência, aumento da latência dos componentes P2, N2 em Cz na orelha direita e N2 e P3 em Fz na orelha esquerda, e diferença na amplitude de P2 e P3 em Cz na orelha direita. Na varredura de duração, houve redução da amplitude dos componentes N2 em Cz e P3 em Fz na orelha direita. Já no grupo de adultos gagos, observou-se na varredura de frequência, aumento da latência no componente P3 em Cz e Fz na orelha direita e amplitude reduzida no P3 em Cz na orelha esquerda, e na varredura de duração, aumento de latência de N2 e P3 em Cz na orelha direita e N2 em Cz na orelha esquerda, quando comparados ao grupo controle. Conclusão: há diferenças entre os exames eletrofisiológicos de indivíduos com gagueira comparados aqueles com desenvolvimento típico. No entanto, novos estudos neste âmbito ainda são necessários para que os achados aqui apresentados possam ser confirmados.

for at least 12 months and; (5) stuttering classified as mild according to the SSI -3 Stuttering Severity Instrument 17 .For the control group of fluent children and adults (G1a and G1b), the inclusion criteria were: (1) no complaints of current or previous stuttering; (2) negative family history of stuttering; (3) present less than 3% of stuttering disfluencies on a evaluation.
For the control group (G1a and G1b), 17 subjects were matched according to age, gender and educational level, with good performance.
Exclusion criteria for the groups included the presence of other associated impairments, such as biological, visual, and hearing complaints, audiological risk factors and/or hearing loss.These conditions were detected by the specialized team through history with the leaders and participants, inspection of the external auditory canal, tympanometry and audiometry.
In the investigation of these audiological risk factors, changes were detected in all cases, both in history and tympanometry, with type A curve in both ears and normal hearing level 18 .
Initially, the audiovisual recording of the subjects' self expressive speech sample was performed both for the experimental and control groups, consisting of 200 fluent syllables, using a Sony digital camcorder and tripod.
The transcription and analysis of the speech were carried out according to the ABFW (Child Language Test -Fluency) 7 , which considers the types of disfluencies, speech rate and the frequency of breaks.Subsequently, the Stuttering Severity Instrument (SSI-4) 19 was applied to the experimental group to classify the degree of stuttering as mild, moderate, severe or very severe.
In the second stage, for LLAEP evaluation, the Biologic Navigator Pro was used and recorded using five disposable electrodes placed at Fz and Cz in reference to the right lobe (A2) and left (A1), using two recording channels of the equipment, the ground electrode was placed at Fpz.The impedance was kept below 5 KW.
For P300 recording, the components were studied in two scans, i.e., firstly, tone stimuli (tone burst) differing in frequency (standard stimulus: 750Hz and deviation stimulus: 1000Hz), and after that, stimuli differing in duration (standard stimulus: 100ms and deviation stimulus: 50ms, both at 1000 Hz).
Both stimuli differing in frequency and duration were presented randomly in oddball paradigm, at 1.1 stimuli per second with the occurrence of deviation stimulus probability in 20% out of a total of 250 stimuli.The wave analysis time was 500ms, (auditory cortex and pathways).The central auditory processing assessment enables the diagnosis of the auditory gnosic process of an individual and it is performed through the application of tests that show the individual's performance before the solution of a difficult task.
Long Latency Auditory Evoked Potentials (LLAEP) or Cognitive Potential (CP) are electrophysiological tests that evaluate the cortical activity involved in discrimination, integration and attention skills 10 .They also reveal the integrity and capacity of the central auditory nervous system (CANS) 11,12 , therefore these tests have been increasingly used in the investigation of the auditory pathway of individuals with stuttering.
Despite the P300 does not accurately identify the brain activation sites, research results have showed, to some extent, that differences in interhemispheric activation patterns can be measured pre and post treatment.In addition to issues related to the possibility of differentiating individuals for whom therapy proves more effective 13 .
Thus, studies on the cognitive potential in this population should be encouraged in order to clarify possible correlations between auditory aspects and fluency.Given the above, this study aimed to describe the electrophysiological findings of the individuals with stuttering and to compare with fluent individuals.

METHODS
This study was approved by the Research Ethics Committee of UNESP Marilia, Sao Paulo under number 0731/2013.All participants and parents were informed about the methodological procedures and signed the Informed Consent prior to any procedure.
A cross-sectional, quantitative and qualitative study, analytical type was performed.The participants were 34 individuals with and without stuttering, of both genders, aged between 7 and 31 years.
The experimental group consisted of 13 children (G2a) and 4 adults (G2b) evaluated and treated at the Research Laboratory of Fluency -LAEF -at the Center for Education and Health Studies (CEES) of UNESP, Marilia.The control group was formed by 13 fluent children (G1a) and 4 fluent adults (G1b).
The inclusion requirements for the two experimental groups were: to be native speaker of Brazilian Portuguese and have good academic performance.The individuals with stuttering (G2a and G2b) must have presented: (1) stuttering complaint; (2) diagnosis of persistent developmental stuttering by a professional of the area; (3) at least 3% of stuttering disfluencies [14][15][16] ; (4) disfluencies observed in P2 and P3 at Cz in the right ear, in G2a when compared to G1a (Table 1).
Table 2 shows the descriptive and inferential values of the variables studied in LLAEP duration, from the children experimental group (G2a) and the children control group (G1a).
As for the results obtained in relation to the duration test, comparing the child groups (G1a and G2a), a reduction in the amplitude of components N2 in Cz and Fz P3 in the right ear in G2a (Table 2).
Table 3 shows the descriptive and inferential values of the variables studied in the LLAEP frequency from the adult experimental group (G2b) and adult control group (G1b).
In LLAEP for the adult groups (G1b and G2b), regarding the frequency test, it was observed an increase in P3 latency at Cz and Fz in the right ear, in G2b compared to G1b.Regarding the amplitude of LLAEP components, a reduction in P3 at Cz in the left ear was observed, when comparing the two groups (G1b and G2b) (Table 3).
Table 4 shows the descriptive and inferential values of the variables studied in the LLAEP duration test from the adult experimental group (G2b) and adult control group (G1b).
As for the results obtained in relation to the duration study, when comparing the groups of adults (G1b and G2b), there was an increase in N2 and P3 latency at Cz in the right ear, in G2b group (Table 4).with 0.5 filter at 30 Hz and sensitivity of 50,000 mV and alternating polarity.
The subject was instructed to perform an active task, paying attention and discriminating the stimuli, naming them as "fine" for the frequency stimuli and "short" for the duration stimuli.
The waves N1, P2, N2 and P3 were identified through the criteria set by Junqueira and Colafêmina 20 , adapted.
For the statistical analysis of data, the following was performed: descriptive statistics (mean and standard deviation) of the waves as for the latency (ms) and amplitude (microvolts), and exploratory (Student t test) to compare groups using the software STATISTICA 7.0.The adopted level of significance was 5% (p ≤ 0.05 *).

RESULTS
Table 1 depicts the descriptive and inferential values of the variables studied in the LLAEP frequency to the experimental group of children (G2a) and the control group of children(G1a).
The results obtained in LLAEP research from children's groups (G1a and G2a) in relation to the frequency showed an increase in latency of components P2, N2 at Cz in the right ear, and N2 and P3 at Fz in the left ear, in the experimental group (G2a) in comparison with the control group (G1a).Regarding LLAEP amplitude components, a difference was     and state of attention and represents the activity of the supratemporal auditory cortex [26][27][28] .
In this study we found changes in the P3 and N2 components for latency and amplitude in the stuttering children group (G1b) and stuttering adult group (G2b).When studied from non-linguistic stimuli using the oddball paradigm, the electrophysiological responses reflect the attention and auditory discrimination skills.
Especially in stuttering adult patients, the perception of the acoustic characteristics is performed with less precision and a longer reaction time to the target stimulus due to an impaired non-linguistic processing, which results in abnormal waveform patterns as observed in this study in stuttering in adults and children 29,30 .
The P3 is generated from endogenous characteristics related to cognition, more specifically to the identification task and appointment of stimuli 23 .Despite the P300 does not accurately identify the brain activation sites to be related to the associative auditory cortex 28 and to be generated from the activation of different areas of the cortex, the results of the study show the differences in the pattern of inter-hemispheric activation between stuttering and fluent people 2,25,29 .Studies 24,25 have showed differences especially regarding the latency in P3 component in stuttering and fluent.

CONCLUSION
There are differences between the electrophysiological tests of individuals with stuttering compared to fluent ones.However, further studies in this area are still needed before the findings presented here may be confirmed and allow progress to new proposals for evaluation of stuttering.

DISCUSSION
In this study, neuroaudiolpgical processes involved in fluency processing were investigated by evaluating auditory evoked potentials.When subjected to the tests that evaluated the neurological processing in the auditory and linguistic aspects 7,8 , the individuals with stuttering showed neurophysiological difficulties.
The literature has described functional differences in stutterers in the pattern of hemispheric activity in response to linguistic and non-linguistic stimuli, differences in neuro-magnetic response and P300 amplitude 21,22 .
When the fluent children group (G1a) and stuttering children group (G2a) were analyzed, differences were observed as to P3 amplitude, regardless of the stimulated ear.
As for the fluent adult group (GI1b) and stuttering adult group (G2b), differences in the P3 amplitude, shorter when right ear was stimulated, identified to capture electrical activity in the midline of the cortex (Cz).Auditory Evoked Potentials are generated from the activation of thalamic-cortical pathways, primary auditory cortex and cortex.In general, latency spikes of auditory evoked potentials of stuttering people differ from fluent ones [23][24][25] .