Mismatch Negativity and Auditory Brain Stem Response in Children with Autism Spectrum Disorders and Language Disorders

Introduction

The term ‘autism spectrum disorders’ (ASD) refers to a neurodevelopmental disorder, characterized by limitations in social communication and everyday functioning.¹ It is a developmental disability; biologically based and unclear how biological variants connect to the symptoms of this condition. In some way, the central processing of stimuli may be disturbed.² Delayed language development (DLD) occurs when children present language maturation, at least 12 months behind their chronological age due to either sensory or intellectual deficits, autism spectrum disorders, evident cerebral damage, and/or inadequate social and emotional conditions.³

Language disabilities in ASD may vary from mild to severe impairments,⁴ and consequently affect the level of functioning in autistic children. About 34% of young children with ASD are nonverbal.⁵ Early diagnosis is essential for early interventions for children who are at risk. Electrophysiological measures have the advantage of being objective tools with no need for language abilities to comply, so they are valuable for evaluating the central auditory system and identify the auditory processing deficits that may affect language development in autistic children.^6,7

It was declared that both the cerebral cortex and the subcortical area are involved in language processing.^8,9 Auditory brain stem response (ABR) is one of the electrophysiological measures that reflects the synchrony of the brain stem electrical activity in response to acoustic stimuli,¹⁰ while mismatch negativity testing (MMN) reflects the integrity of the auditory sensory memory.¹¹ This memory is essential in mapping meaning for language.^12,13

Several studies assessed the auditory pathway at the level of the brainstem in individuals with ASD using ABR.^14,15 Generally, the literature showed that those subjects may present alterations in ABR. The most common findings were increased latency of waves III and V^14,16,17 and an increase in the interpeak intervals of I–III and I–V, suggesting brainstem affection.^14,17

While ABR reflects synchronized electrical activity of the brainstem’s neural elements in response to acoustic stimulation,¹⁰ a mismatch negativity (MMN) paradigm is used to study the preconscious automatic auditory processing. Findings on MMN in children with autism are highly variable. Some studies reported larger MMN and/or shorter latencies in children with autism, whereas other studies reported smaller or even normal MMN and/or longer latencies in children with ASD.^18,19

Lepistö et al,²⁰ reported some MMN amplitude changes with no differences in regard to latency measurements in autistic children diagnosed with language impairment. Since it was declared by Green et al 2020⁶ to be the only MMN study of ASD with language impairment to typically developing children,⁶ and it did not study ABR in comparison with MMN.

Aim

This study was designed to assess auditory processing, by investigating and comparing the brain stem auditory evoked potentials (ABR) and an event-related potential mismatch negativity (MMN) in a group of children diagnosed with ASD & language impairment in spite of apparent normal peripheral hearing sensitivity.

Patients and Methods

A total of 120 children between the ages of 8 and 12 who were chosen from the pediatric and phoniatric clinics of Al Zahraa University Hospital and AL Hussein University Hospital participated in this observational case–control study. They were divided into two groups:

1. Study group: consisted of 60 children with varying degrees of autism spectrum disorders. Their mean age was 9.8 ± 1.19 ranged from 8 to 12 years and 61.7% were males. All ASD groups recruited in this study completed the tests and their language age was below 2 years.
2. Control group: consisted of 65 typically developing children. However, 5 children did not complete the evaluation so they were excluded from the analyses. Consequently, 60 typically developing children with average intellectual ability and fully developed language were included in the statistical analyses. They were matched for age and sex with the study group with mean age 10.03 ± 1.28 ranged from 8 to 12 years and 55% were males.

All participants in this study had undergone general, neurological and otorhinolaryngological examination to exclude other systemic, psychological and neurological diseases. In addition, patients with history of otological diseases, head trauma and ototoxic medications were excluded.

Diagnosis of autism was carried out according to DSM-5 (diagnostic and statistical manual – fifth edition)¹ and graded according to CARS (childhood autism rating scale).²⁰ Language evaluation was done by Arabic PLS4 (preschool language scale – fourth edition).²¹ According to DSM-5 autistic children must have persistent deficit in each area of social communication plus at least two types of repetitive behaviors. On the other hand, classification of severity was done according to CARs because it provides clear cutoff points for each level of severity (children who got score below 30% are considered non-autistic, children with score between 30 and 36.5 suffer from mild to moderate autism, and others who scored 37 up to 60 are considered severely autistic), whereas psychometric evaluation was done by SBIS5 (Stanford-Binet intelligence scale-fifth edition).²²

Audiological Examination

To verify normal peripheral hearing and intact acoustic reflexes, a basic audiological assessment was conducted:

1. Otoscopy is used to examine the tympanic membrane and external auditory canal.
2. By using an immitancemeter (Maico, Diagnostic GmbH, MI 44), 226 Hz tympanometry and ipsilateral acoustic reflex thresholds at 1, 2, and 4 kHz were obtained.
3. An audiometer was used to get pure-tone audiometry (PTA) in a sound treated room (Piano Plus INVENTIS), using calibrated TDH39 headphones. Air conduction thresholds were obtained between 0.25 and 8 kHz. Play conditioning audiometry has sometimes been employed.

A threshold of up to 20 dBnHL for all tested frequencies was considered normal.

1. Interacoustics Eclipse (EP25, Inc., Middelfart, Denmark), in a calm, dark room with the participant seated comfortably in a recliner, was used to conduct electrophysiological measurements.

Auditory brain stem response using click stimuli was obtained for all children. The subjects were tested following the oral administration of chloral hydrate to avoid interference caused by excessive muscle activity. The following parameters were applied: 100 μs, 80 dBnHL, rarefaction, and broadband click, given at a rate of 21.1/s, with 1200 total sweeps, and a 20 ms time window. The absolute latencies of waves I, III, and V on both sides as well as the IPLs of waves I–III and III–V were identified manually.

Tone burst stimuli were delivered in an odd-ball paradigm while the patients watching silent videos, to produce mismatch negative responses. The Probability for the standard was 80% and 20% for the deviant stimuli. The deviants (250 in total) were presented in an alternating polarity, with an intensity of 80 dBpeSPL and at a rate of 1.1/s. 300 sweeps for each ear were obtained and averaged. Time analysis was set at 500 ms, and included 50 ms pre-stimulus onset. Recording of two replications from each ear was made with an amplification of 50, artifact rejection at 100 μV, and a band-pass filter of 1–30 Hz.

Ethical Considerations

The study objectives, design and all details were fully explained to all participants. The study procedure was in accordance with the Declaration of Helsinki and was approved by the institutional review board (IRB), faculty of medicine, Al-Azhar University, Cairo, Egypt. Participation was voluntary; after obtaining informed consents from parents and guardians. Each participant had the right to refuse participation or withdraw from the study at any time without giving any reasons and without any interference with their rights of medical care. Data were anonymous and coded to assure confidentiality of participants.

Statistical Analysis

It was carried out using the SPSS computer package (IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp., USA). The Shapiro–Wilk test was used to examine the normality of quantitative data. For descriptive statistics: the mean ± SD or median (interquartile range 25th–75th percentile (IQR)) were used for quantitative variables while frequency and percentage was used for qualitative variables. Fisher’s exact test was used to assess the differences in frequency of qualitative variables, while Independent samples or Mann–Whitney U-tests were used to assess the differences in means of quantitative variables between both groups. The statistical methods were verified, assuming a significant level of two-sided p values <0.05.

Results

Regarding auditory brainstem evoked response, the median (mean) values of absolute and interpeak latencies in both ears were significantly delayed (prolonged) in children with autism compared to controls except for wave I in both ears, wave I–III in the right ear, and wave III in the left ear that were delayed (prolonged) but with no significant differences (Table 1 and Table 2).

Table 1 Absolute and Interpeak Latencies in the Right Ear of the Studied Groups

Table 2 Absolute and Interpeak Latencies in the Left Ear of the Studied Groups

Regarding mismatch negativity, the median (mean) values of latency measures in both ears were significantly delayed (prolonged) in autistic children in comparison to controls (P=0.035 in the right ear and P=0.005 in the left ear) while regarding amplitudes, no substantial variation was detected between both groups in both ears (P=0.995 in the right ear and P=0.644 in the left ear) (Table 3).

Table 3 Mismatch Negativity Test in Both Ears of the Studied Groups

Significantly, the ASD group showed a total of 43.3% of abnormal auditory brainstem evoked response and a total of 30% showed significant changes in mismatch negativity latency measurements on comparing with the control group (P<0.001) (Figure 1).

Figure 1 The deviant results on auditory brainstem response and mismatch negativity among the studied groups.

Discussion

The click-ABR is a noninvasive and objective test method. It does not rely on the participant’s active responses, so it became a conventional tool to evaluate auditory functions in children, who are difficult to assess behaviorally. A strong relationship between abnormalities in the click-ABR and ASD has been reported in the literatures.^23–25

In this work, we intended to study the click ABR in autistic children with language disorders and reported significantly increased absolute latencies of waves (III, V) and the interpeak intervals (Table 1 and Table 2). However, the ABR findings are variable within the scientific literature. Some studies have shown that there have been no changes to the actual wave latencies of ABR.^15,26,27 Controversely, some authors reported prolonged absolute latencies for waves III and IV,¹⁶ increased latencies for waves I and V,¹⁷ and prolonged latencies for waves V in the right ear and wave I in the left ear^28,29 in ASD. This electrophysiologic variability supports the hypothesis of perceptual inconstancy.¹⁶

Magliaro et al,¹⁴ worked on a group of autistic children and revealed that the latencies of waves III and V as well as the I–III interpeak interval were prolonged.

This inconstancy among studies can be due to methodological differences, as regarding different diagnostic criteria for autism, variability in stimulus presentation rate and differences in the studied groups’ age range.³⁰ Also, autism is a heterogeneous disorder with several subgroups,^31,32 which suggests that this latency prolongation may occur in a subgroup of autism cases. These latency differences and prolonged interpeak intervals of ABR are supposed to be due to defects in the myelination, axon diameter, and synaptic efficacy of the auditory pathways at the brainstem level.³³

Regarding the mismatch negativity, there was no statistically substantial variation between the two groups regarding amplitude measurements; however, the latencies were substantially increased in the ASD group. This agrees with Roberts et al,³⁴ who declared that on comparison with neurotypical developing children, autistic children with delayed language had a considerably longer mismatch magnetic field (magnetic analog of MMN) latencies.

Jansson-Verkasalo et al,³⁵ worked on a group of children diagnosed with Asperger’s syndrome and exhibited greater MMN latencies. They saw this as a sign of auditory discriminating issues, most likely the consequence of inadequate temporary memory storage processes. However, Kemner et al,³⁶ reported that regarding the MMN, there were no variations between the autistics and the normal controls.

Findings on MMN in ASD are quite variable.¹⁸ Some studies revealed enhanced MMN and/or shorter latencies,³⁷ while others showed smaller or even normal MMN amplitude and/or longer latencies in comparison to age-matched normally developing children.³⁴ These discrepancies between researchers may be due to variable methodology and inadequate measuring reliability, and statistical power.³⁸ Also, as mentioned above autism is a heterogeneous disorder with several subgroups.^31,32

In our study, a total of 43.3% and 30% of the ASD group showed significant changes in the ABR and MMN test results, respectively, on comparing with the control group (Figure 1).

Conclusions

Both ABR and MMN are complementary tests in evaluating language development in ASD. Our work may give some clinical evidence of remarkable dysfunction in basic auditory sound processing at the brain stem, this impacts the linguistic development of autistic children and may be partially to blame for their atypical development. The significance of this finding needs further in-depth studies with larger sample size, considering the subgroups of autistic children with language impairment and multiple settings.