The Effects of Stimulus Rate on ABR Morphology and its Relationship to P1 CAEP Responses and Auditory Speech Perception Outcomes in Children with Auditory Neuropathy Spectrum Disorder: Evidence from Case Reports

Abstract

Objective

Auditory Neuropathy Spectrum Disorder (ANSD) affects approximately 5–15% of children with sensorineural hearing loss. ANSD is characterized by the presence of otoacoustic emissions (OAE) and an absent or abnormal auditory brainstem response (ABR). The purpose of this study was to investigate the prognostic value of slow-rate ABR in predicting the auditory cortical development and auditory speech perception outcomes in case studies of children with ANSD.

Design

ABR waveform characteristics were collected at slow stimulation rates (5.1 clicks/second) and a fast stimulation rates (>11–31.1 clicks/second, rates typically used in a clinical setting) in 3 case reports of children with ANSD. P1 CAEP responses and measures of auditory speech perception using the Infant Toddler Meaningful Auditory Integration Scale (IT-MAIS) were also collected in these children. Retrospective analysis was performed to evaluate the prognostic value of slow- versus fast-rate ABR in predicting P1 CAEP responses and auditory speech perception outcomes in these children.

Study Sample

Participants included case reports of 3 pediatric participants with a clinical diagnosis of ANSD.

Results

Slow-rate ABR did not elicit significant improvements in waveform morphology compared to fast-rate ABR. P1 CAEP results were present in 2 out of 3 cases and were consistent with auditory speech perception outcomes.

Conclusions

Even when ABR stimulation rates were slowed, ABR responses in these children with ANSD did not display any characteristic or replicable pattern, and ABR responses were not predictive of cortical auditory maturation or behavioral performance. In contrast, P1 CAEP responses provided valuable information regarding the maturational status of the auditory cortex and P1 CAEP responses were consistent with behavioral measures of auditory speech perception. Overall, results highlight the high prognostic value of P1 CAEP testing when used in conjunction with behavioral measures of auditory speech perception in children with ANSD.

Introduction

Auditory neuropathy spectrum disorder (ANSD) is a diagnostic classification of hearing loss with a variable clinical presentation, which is present in 5–15% of children with sensorineural hearing loss (SNHL) [1–4]. Initially, ANSD was described in 1996 by Starr and colleagues who identified hearing impairment in ten patients which was attributed to a disorder of the auditory portion of the eighth cranial nerve (CN VIII) and was characterized by normal outer hair cell function, as indicated by the presence of otoacoustic emissions (OAE) and a cochlear microphonic (CM), but abnormal auditory pathway function, as evidenced by abnormal or absent auditory brainstem responses (ABR) [5]. Subsequently, to reflect the lack of neural synchrony of the auditory brainstem response, the descriptor auditory dys-synchrony was added to the auditory neuropathy term, leading to the creation of a further diagnostic label, auditory neuropathy/auditory dys-synchrony [6–8]. Since then, it has been widely recognized that the terms auditory neuropathy and auditory dys-synchrony actually encompass multiple etiologies and sites of lesion that result in this category of hearing loss. Therefore, the term auditory neuropathy spectrum disorder (ANSD) has been adopted to ‘unify the concept of an auditory disorder with a range of presentations secondary to a variety of etiologies’ [9].

Clinically, ANSD is diagnosed by the presence of OAE, absent or elevated acoustic reflex (ART) responses, and an absent or abnormal ABR response [4,5,10]. Clinical presentation of ANSD is quite inconsistent, with some children exhibiting permanent hearing loss and some children exhibiting fluctuating hearing loss, with greatly varying impacts of the hearing loss on auditory speech perception and language development. Most often, measures of speech perception in noise in ANSD children are poor [11–13], and unlike in sensorineural hearing loss (SNHL), perceptual deficits have been noted to be disproportionate to the behavioral hearing levels in both quiet and noisy conditions [5,11,14] in the majority of ANSD cases, with discrepancy between functional hearing ability and detection of isolated speech sounds being high in children [14].

The ABR response is an electrophysiological response elicited by an auditory stimulus which reflects the neural integrity of the nerve and brainstem pathways. In children with ANSD, an absent or grossly abnormal ABR is a key feature for diagnosis [5]. While faster stimulation rates (>10 clicks/second) are typically used in a clinical setting to elicit the ABR response in diagnosing ANSD, the clinical utility of using slower stimulation rates is largely unknown. The first purpose of this study was to evaluate whether using slower stimulation rates elicited morphologically clearer ABR responses with greater prognostic value beyond a normal or abnormal/absent ABR response. It was hypothesized that slowing stimulus rate may improve neural synchrony, thereby resulting in a more recognizable and prognostically useful ABR response in children with ANSD.

The P1 CAEP response is an electrophysiological response elicited by an auditory stimulus which provides an objective, non-invasive examination of the maturation and function at much higher centers of the brain, at the level of the auditory cortex and thalamocortical pathways. Previous studies have highlighted the clinical utility of the P1 CAEP response in predicting speech and language outcomes in children with ANSD [15–17], as well as other clinical population with hearing loss, including bilaterally deaf children receiving cochlear implants, children with single-sided deafness, and hearing impaired children with multiple disabilities [18]

The P1 response is present at birth, occurring around 300 milliseconds [19] . As a child grows and their auditory system becomes more efficient, the P1 response decreases systematically in latency until it reaches 50–70 milliseconds in adulthood [19,20]. Because the P1 response varies as a function of age, it can be used as an objective biomarker of the maturation of the central auditory pathways. The second purpose of the study was to examine the relationship between slow-rate ABR responses and P1 CAEP responses, and the whether these electrophysiological indicators were predictive of speech perception outcomes in children with ANSD.

Methods

Procedures

We conducted a retrospective case review of case histories, behavioral audiometric data (including pure tone thresholds, speech awareness thresholds, speech recognition thresholds), speech perception outcomes (questionnaires/assessments of auditory speech perception and auditory skill development), and electrophysiological test results (including OAEs, slow- and fast-rate ABR, and P1 CAEP) in children over 18 months of age with a confirmed clinical diagnosis of ANSD by their managing audiologists. Morphology of the ABR responses at each stimulation rate was compared within and across participants using an ABR Ratings Specification Index. Additionally, we related the ABR response results to P1 CAEP and behavioral speech perception to examine the prognostic value significance of slow-rate ABR and P1 CAEP testing in predicting speech perception outcomes in children with ANSD. All data was collected according to University of Colorado-Boulder Internal Review Board Procedures, and informed consent/assent was obtained by all research participants.

This study consisted of a retrospective case review of audiological, electrophysiological, and behavioral data in 3 pediatric research participants in which ABR responses at slow rates (5.1 clicks/second) and fast rates (>11–31.1 clicks/second), P1 CAEP test results, and behavioral auditory speech perception outcome data were available.

Subjects

Three case studies were selected which met our retrospective review criteria (at least 18 months of age, confirmed clinical diagnosis of ANSD by the managing audiologist, availability of slow- and fast-rate ABR results, P1 CAEP results, and IT-MAIS data). The case studies ranged in age from 1 year, 7 months (1;7) to 6 years, 2 months (6;2). Two of the subjects were male and one was female. While this case study sample is small and heterogeneous in nature, it reflects the spectrum of ANSD in the community. Further, prior evidence that the ABR is adult-like from around the age of 18–24 months [21,22] and the availability of age-based normative P1 CAEP data, our small and heterogeneous sample was deemed appropriate in this context.

Case History

Case history information for each research participant is shown in Table 1. Case history information collected included age, risk factors for ANSD, tests and results used in the clinical diagnosis of ANSD (e.g. presence/absence of a cochlear microphonic (CM), ABR morphology, and OAE), auditory thresholds (pure tone average), speech perception abilities (speech awareness threshold and/or speech reception thresholds), and hearing aid use.

Table 1. Case history information for 3 pediatric research participants with a clinical diagnosis of auditory neuropathy spectrum disorder (ANSD).

The participant age relates to the age at which slow and fast rate ABR testing was administered. These data highlight the heterogeneous nature of ANSD in terms of its highly variable clinical presentation in terms of pure tone audiometric thresholds, speech perception, and auditory skill development, and clinical choice of intervention.

Case	Age y; m	Risk Factors	Diagnosis: newborn	PTA unaided R / L (dB HL)	SAT (dB HL)	SRT R / L (dB HL)	Hearing aid use
1	3;7	NICU	OAE +	50 / 55		30 / 40 unaided	yes
			CM +
			ABR Abn			20 / 20 aided
2	6;2	NICU	CM +	65 / 70	70 unaided		yes
			OAE -		20 aided
			ABR -
3*	1;7	NICU	OAE +	20 / 20*	20*		no
			ABR -

Abbreviations: NICU = Neonatal Intensive Care Unit; OEA = Otoacoustic Emissions; ABR = auditory brainstem response; Abn = abnormal; PTA = pure tone audiometry; SAT = speech awareness threshold; SRT = speech recognition threshold; VRA = visual reinforcement audiometry.

^*Note that due to the age of participant 3, pure tone threshold and SAT were assessed using sound field testing & VRA.

ABR Recordings

ABR responses were obtained for each participant using the Bio-Logic Navigator Pro AEP System (Natus Medical Inc.). Participants were comfortably seated in an electromagnetically shielded sound booth. The ABR stimuli consisted of a standard click stimulus (0.1ms in duration) presented at slow (5.1 clicks/second) and fast (11.1–31.1 clicks/second) rates, using rarefaction and condensation polarities. Stimuli were presented at a supra-threshold level (75–95 dB HL) via insert earphones (ER3–14A or ER3–14B, depending on the age of the child) to each ear separately. At least 1 run of 500 stimulus presentations for each stimulus polarity was first collected at the slow rate (5.1 clicks/second). For children who could tolerate longer test sessions, at least 1 run of 500 stimulus presentations for each stimulus polarity was then obtained at a faster rate (11.1–31.1 clicks/second). All but the youngest participant (Case 3) were able to tolerate ABR testing at both slow and fast rates, so for Case 3 only slow-rate ABR responses are available.

ABR Analysis

Because of a lack of existing descriptive metrics to evaluate ABR morphology in the literature, an ABR Ratings Specification Index was developed and a multi-rater blinded system was employed to evaluate the morphology of the ABR waveforms at each stimulation rate for each research participant. This ABR Ratings Specification Index is described in Table 2. The ABR Ratings Specification Index provides Category 1, Category 2, or Category 3 ratings at an early interval (0–3 milliseconds) and a late interval (3–10 milliseconds) of the ABR waveform. Category 1 ratings indicate present and replicable morphology of the waveform components within the time interval of interest, Category 2 ratings indicate somewhat present and moderately replicable morphology of the waveform components within the time interval of interest, and Category 3 ratings indicate absent or very poor morphology of the ABR waveform components within the time interval of interest. The 3ms cut-off for early and late interval ratings was chosen based pre-neural (CM) and neural (ABR Waves I-V) features of the ABR waveform [5,23]. Because the CM is a pre-neural response not subject to neural fatigue that mimics the polarity of the stimulus, the early interval rating (0–3ms) focuses on the presence or absence of the CM by visualizing superimposed rarefaction and condensation ABR runs. In contrast, the late interval rating (3–10ms) evaluates the neural features of the ABR response, focusing on the morphological presence or absence of the later ABR waveform components, most importantly wave V which can be followed down to threshold to estimate hearing sensitivity. The ABR waveforms for each research participant at each stimulation rate were randomized and identifying information was removed. Next, four graduate student raters independently judged and assigned a Category 1, Category 2, or Category 3 ratings for the early and late intervals to each waveform in order to objectively and descriptively characterize morphological integrity of the ABR responses for each participant at each stimulation rate.

Table 2. Rating Index Specifications for slow rate and fast rate ABRs.

The Auditory Brainstem Response (ABR) Ratings Specification Index was created to describe and categorize abnormal ABR morphology. The 3 millisecond cut-off for early and late interval ratings was selected based on pre-neural (cochlear microphonic) and neural (ABR Waves I-V) features of the ABR waveform.

Early interval rating (≤3ms)	Late interval rating (>3-10ms)
Category 1: Inverse pattern present with minimal distortion in >50% of the early stage	Category 1: Replicable pattern is present with minimal distortion, and a definitive characteristic of an ABR waveform, e.g. Wave V
Category 2: Inverse pattern present with moderate distortion	Category 2: Moderately replicable pattern is present with some degree of ABR waveform characteristics
Category 3: No definable inverse pattern	Category 3: No definable characteristic of an ABR waveform

P1 CAEP Recordings

P1 Cortical auditory evoked potentials (CAEPs) were collected for each research participant using the Compumedics Neuroscan evoked potential system using a 5-electrode montage. Cz (midline vertex) served as the active electrode and reference electrodes were placed on the right or left mastoid(s) and forehead. Eye movements were monitored using a bipolar electrode montage (lateral outer canthus and superior outer canthus) during CAEP recordings, and eye blinks were automatically rejected online during the test session. P1 CAEP responses were elicited in response to a speech syllable /ba/ with a total duration of 90ms, at an inter-stimulus interval of 610 ms (Sharma et al., 2005). Stimuli were presented via the soundfield at a speaker located at 0° azimuth during aided testing, or via insert earphones during unaided testing. Stimuli were presented at a comfortable loudness level at least 20 dB supra-threshold from the patient’s auditory thresholds. This test paradigm has been previously used in pediatric cases to examine the maturation and physiology of the central auditory pathway [15,19,24–27]. Subjects were seated in an electromagnetically shielded sound booth. During testing, subjects watched a movie on a television monitor located at 0° azimuth with the audio muted. A minimum of 2 replicable runs of approximately 300 stimulus presentations were collected for each subject.

P1 CAEP Analysis

Individual P1 CAEP runs for each participant were post-processed off-line, where baseline correction (to entire sweep), artifact rejection (±100 μV), and averaging were performed. Grand average waveforms across the 2 average runs were then computed for each subject. P1 morphology was first evaluated to determine whether a P1 response was present, absent, or abnormal. If a P1 response was present, the latency of the P1 response was selected at the peak of the first robust positive component in the CAEP waveform. P1 CAEP latencies for each ANSD were then plotted and compared against age-based 95% confidence intervals for normal P1 latencies based upon studies of hundreds of normal hearing children from 1 month to 20 years old in order to determine whether auditory cortical development is normal, delayed, abnormal, or absent [19,24,26–29].

Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS)

The Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS) [30] is a structured interview assessment designed to assess a child’s spontaneous responses to sound in his/her everyday environment [31]. There are 10 items on the IT-MAIS assess which assess 3 primary areas: 1) vocalization behavior, 2) alerting to sounds, and 3) deriving meaning from sound. Each item is rated by the parents on a scale of 0 (Never) to 4 (Always) based upon the percentage of time that a child demonstrates specific auditory abilities. The IT-MAIS is scored out of a total of 40 possible points. One item on the assessment assumes the use of a hearing aid or cochlear implant; therefore, for Case 3 in which the child did not receive any form of audiological intervention, this section was excluded from the child’s IT-MAIS score (out of a total of 36 possible points). Expected values for the IT-MAIS related to the child’s age and degree of hearing loss have been previously identified [32,33].

Results

Case 1

Case 1 is a male child who was born prematurely at 28 weeks gestation following an intrauterine infection. Per chart review, he had a complicated birth and neonatal history, which included prematurity, mechanical ventilation, pulmonary insufficiency, hyperbilirubinemia, periventricular leukomalacia, chronic lung disease, and time spend in the neonatal intensive care unit (NICU). The child referred on several hearing screenings performed in the NICU. Testing at age 0;3 by the child’s audiologist resulted in a clinical diagnosis of ANSD, at which time the child exhibited present OAEs and CMs and an abnormal ABR with a delayed Wave V bilaterally. Pure tone audiometry revealed a moderate—moderate-severe sensorineural hearing loss (Figure 1.1). He received early intervention services from a speech language pathologist from the age of 0;6, and was fitted with hearing aids at age 3;0. The child’s aided pure tone thresholds fell in the normal-mild hearing loss range.

Figure 1. Behavioral and electrophysiological data for Case 1.

Panel 1: Pure tone audiometric results showing a moderate to moderately severe bilateral hearing loss unaided, and a slight to mild hearing loss aided; Panel 2: Longitudinal P1 latency for Case 1 plotted against 95% confidence interval normative data, showing latencies within normal limits for developmental age; Panel 3: ABR at slow click rate; Panel 4: ABR at fast click rate.

ABR responses at slow and fast click rates were obtained 3;7 years. Slowing the stimulus rate of the ABR increased the clarity of the CM pattern in the early interval, but did not improve the overall morphology of the ABR waveform (Figure 1.3 and 1.4). P1 CAEP responses were obtained when the child was 0;6, 0;10, and 2;1. At all of these test sessions, Case 1 exhibited present P1 CAEP responses with latencies falling within the 95% confidence interval for normal development of the P1 response (Figure 1.2) [19], suggesting that the child was receiving adequate quality and quantity of auditory input to allow for normal development of the auditory cortical pathways (Figure 1.2). The IT-MAIS was also administered at age 2;1, at which time the child scored 34/36 (94%). Overall, the IT-MAIS results indicated only mild delay in early pre-lingual auditory development which was consistent with the parent report of delayed language acquisition.

Case 2

Case 2 is a male child who was born prematurely at 24 weeks gestation. He was an inpatient in the NICU for 5 months, during which time he received prolonged mechanical ventilation, potentially ototoxic medications, and was diagnosed with an intraventricular hemorrhage and seizures. Testing at age 0;5 by the child’s audiologist resulted in a clinical diagnosis of ANSD, at which time the child exhibited absent OAEs, robust CMs, and an absent ABR response bilaterally. Pure tone audiometry revealed a severe sensorineural hearing loss bilaterally (Figure 2.1). He received early intervention services from a speech language pathologist and an occupational therapist from age 0;6. The child was fitted with hearing aids bilaterally age 2;0, at which time he demonstrated aided pure tone thresholds in the normal-mild hearing loss range.

Figure 2. Behavioral and electrophysiological data for Case 2.

Panel 1: Pure tone audiometric results showing a moderately severe bilateral hearing loss unaided; Panel 2: P1 latency for Case 2 plotted against 95% confidence interval normative data. No replicable P1 response for Case 2 could be identified. Panel 3: ABR at slow click rate; Panel 4: ABR at fast click rate.

ABR responses at slow and fast click rates were obtained at age 6;2. Slowing the stimulus rate of the ABR did not improve the morphology of the ABR waveform (Figure 2.3 and 2.4). P1 CAEP responses were obtained when the child was 3;4 and 3;5. At both test sessions, Case 2 exhibited absent P1 CAEP responses (Figure 2.2), suggesting that the child was not receiving adequate quality and quantity of auditory stimulation to promote normal development of the auditory cortical pathways. His IT-MAIS at age 6;2 was 20/40 (50%), consistent with severely delayed auditory skill development [32].

Case 3

Case 3 is a female child who was born prematurely at 33 weeks gestation. She had a complex history including twin-to-twin transfusion syndrome (as the donor) with resulting hypoxia, hyperbilirubinemia, necrotizing enterocolitis requiring colectomy, sepsis, exposure to potentially ototoxic medications, and extended mechanical ventilation. She received a clinical diagnosis of ANSD at age 0;3, in which her OAEs were present and her ABR was absent. Results from several hearing evaluations on this child indicated a fluctuating hearing loss; however, at age 1;4, visual reinforcement audiometry (VRA) indicated normal hearing in a sound field (500 Hz – 2000 Hz) and bilaterally at 4000 Hz – 8000 Hz (Figure 3.1).

Figure 3. Behavioral and electrophysiological data for Case 3.

Panel 1: Visual reinforcement audiometric results performed at 11 months and 16 months, showing fluctuating hearing thresholds; Panel 2: Longitudinal P1 latency for Case 3 plotted against 95% confidence interval normative data, showing borderline delayed latencies at age 2 but improvement to normal P1 latency by age 3. Panel 3: ABR at slow click rate (child was unable to tolerate ABR at fast click rate); Panel 4: IT-MAIS scores over time showing pre-lingual auditory development with increasing age. Final IT-MAIS score is within expected range for age.

Abbreviations: VRA = Visual reinforcement audiometry; IT-MAIS = Infant-Toddler Meaningful Auditory Integration Scale

ABR responses at slow and fast click rates were attempted at age 1;7. The child’s slow rate ABR had a recognizable CM, but did not have any recognizable later morphological features (Figure 3.3). Due to young age and limited attention span, fast rate ABR could not be obtained in this child. P1 CAEP responses were recorded at 0;4, 0;11, 1;1, 1;7 and 3;1. At age 0;4, the P1 CAEP latency was borderline-normal for expected age. However, the latency did not decrease with age, occurring outside the 95% confidence interval for normal development when tested at age 0;11 and 1;7 (Figure 3.2). The fluctuating P1 CAEP results we observed over time corresponded with the fluctuating auditory thresholds observed in this child over time. She was referred for a repeat measure of the P1 CAEP at age 3;1. At that time, a replicable and robust P1 occurred within a normal latency range for the patient’s age, indicating age appropriate maturation of the central auditory pathways (Figure 3.2). This coincided with substantial growth in receptive and expressive language, per parent report. The growth in language acquisition paralleled the improved latency of the P1 CAEP response. The IT-MAIS was administered on multiple occasions: at ages 0;4 (7/36; 19%), 0;11 (11/36; 31%), 1;1 months (18/36; 50%), and 1;3 (31/36; 86%). Although initially lower than for typically hearing infants [33], the increase in IT-MAIS scores over time indicated gradual improvement in functional auditory skills. By age 1;3, the IT-MAIS was consistent with age-appropriate early pre-lingual auditory development (Figure 3.4).

Results

The slow rate and fast rate ABRs from Cases 1–3 were analyzed as a group to look for any trends in replicable morphological changes associated with slowing the stimulus rate. The de-identified waveforms were rated, and then the category assigned by each rater was compared. Raters were consistent in their judgment: there was either complete agreement in the waveform rating (8 of 14 occasions), or the reviewers differed by one category only (6 of 14 occasions).

An average reviewer rating was calculated for each interval (early and late) in each rate group (slow and fast), and then the mean values for slow rate early interval, slow rate late interval, fast rate early interval and fast rate late interval, were calculated (Table 3).

Table 3. ABR Early Interval and Late Interval values for slow and fast rate ABRs.

Using the Auditory Brainstem Response (ABR) Ratings Specification Index, the ABR ratings for each participant were averaged across raters and across each participant’s waveforms for the early and late interval of both slow and fast rates. Some participants had more than one ABR available for rating. 8 of 14 intervals had complete concordance between raters; 6 of 14 intervals differed by one category.

	Slow Stimulus Rate		Fast Stimulus Rate
	Early Interval	Late Interval	Early Interval	Late Interval
	(≤3ms)	(>3ms)	(≤3ms)	(>3ms)
	Mean Category	Mean Category	Mean Category	Mean Category
	Rating	Rating	Rating	Rating
Case 1	1	2	1.25	2.38
Case 2	1.5	3	2.25	3
Case 3*	1	3
Average	1.17	2.67	1.75	2.69

^*Note: Case 3 had one slow rate set of data only due inability to tolerate the procedure.

For the slow rate ABRs, the mean early interval rating value was 1.17 and for the fast rate ABRs, the mean early interval rating value was 1.75, indicating that there was a mildly or moderately distorted inverse pattern noted in the initial 3ms of the slow and fast rate ABRs which represented a CM. Overall, the inverse pattern was more likely to be mildly distorted with the slow stimulus rate and moderately distorted with the fast rate. Considering the late interval, in the slow rate ABR the mean rating value was 2.67 and in the fast rate ABR the mean rating value was 2.69, indicating that any later waveform characteristics were consistently moderately distorted to undefinable.

When slow and fast rate waveforms across all subjects were considered, these results suggested that when slowing the click stimulus, there was a greater likelihood of observing a mildly distorted inverse pattern relating to a prominent cochlear microphonic. However, for the interval of the ABR occurring after 3ms, slowing the stimulus rate did not improve the ABR, and the late interval of the waveforms remained grossly abnormal.

Considering Cases 1–3, the scores obtained on the IT-MAIS reflected the degree of central auditory pathway maturity indicated by the P1 CAEP latency. Those scores which fell close to or within the normal range described by Zheng (2009) had a P1 latency within the expected latency range for the child’s age. Correspondingly, Case 1, who had severely delayed pre-lingual auditory development as suggested by the very low IT-MAIS score, did not have an identifiable P1 response. The differing IT-MAIS scores were not in any way reflected in the ABR waveform (Table 4).

Table 4. ABR, P1 CAEP latency and IT-MAIS results for pediatric auditory neuropathy spectrum disorder (ANSD) case studies.

In all cases, neither the slow rate nor the fast rate ABR predicted early pre-lingual auditory development (reflected by the IT-MAIS score) or central auditory pathway maturity (indicated by the P1 CAEP). However, in each case, measures of early pre-lingual auditory development and central auditory pathway maturity were linked.

Case	ABR (slow rate)	ABR (fast rate)	IT-MAIS Score	P1 CAEP Latency
#1	Absent	Absent	Very Low	Absent
#2	Absent	Absent	Mild delay	Normal
#3	Absent	Absent	Average	Delayed, then normalized

Abbreviations: ABR = auditory brainstem response; P1 CAEP = P1 cortical auditory evoked potential; IT-MAIS = Infant-Toddler Meaningful Auditory Integration Scale

Finally, central auditory pathway maturity —as indicated by P1 CAEP latency—was independent of the slow and fast stimulus rate ABR morphology. All cases had grossly abnormal ABRs, with a complete lack of recognizable Wave I-V morphology, but they each differed in their prior P1 CAEP results. Case 1 had no identifiable P1 in prior testing; Case 2 had a P1 response which fell within the normal range for expected latency; and Case 3, at the time of the ABR, had a delayed P1 response (Table 4).

Discussion

Ever since Jewett and colleagues first described auditory evoked potentials [34,35], numerous researchers have investigated the effect of the click stimulus rate on the latency, morphology, and threshold of the auditory brainstem response in humans with normal hearing [36–39]. It is widely recognized that increasing click rates above 20 CPS leads to an increase in ABR latency and a decrease in amplitude in normal-hearing adults [40], and that decreasing the stimulus rate in patients with either normal hearing or a sensorineural hearing loss leads in particular to a decrease in the latency of Wave V [41–43]. Furthermore, in certain pathologies, such as multiple sclerosis, increasing the click rate causes progressive desynchronization with characteristic and diagnostic loss of waveform morphology [36].

The body of research indicates that the ABR is sensitive to the effect of stimulus rate on neural synchrony. Despite this, there is a lack of published research on slow stimulus rates in patients with ANSD. At the time the data were being collected and analyzed, there was no published research addressing the question of whether a profoundly slowed stimulus rate would distinguish the ABR of patients with ANSD. Since increasing the rate of stimulus in ABRs taxes the neural system and affects the synchronous firing of the auditory nerve especially in those with demyelinating pathology, it was reasonable to investigate whether profoundly slowing the rate of stimulus might improve neural synchrony, resulting in a replicable waveform in the ABR of those with ANSD. If an ABR were apparent at a slow stimulus rate, it could also provide an objective measure of hearing thresholds in children with ANSD, aiding in clinical decision making such as amplification fitting.

The results from this case series were not consistent with the prediction that there would be replicable and robust ABR patterns at a slower stimulus rate. Evidence from this study found that considering the entire ABR, there were no reliable differences in ABR waveform characteristics between the stimulus rates, and no evidence of recognizable Wave I -V shapes when using the slowed rate. Though this study consisted of case studies of children with ANSD and thus should be interpreted with caution, our results suggest that slower click rates do not elicit greater neural synchrony at the level of the brainstem in children with ANSD. Due to the small size of this case series, it is not possible to say if this negative result would persist in a larger sample size. However, our results are consistent with a previous study which showed no differences in ABRs with stimulus rates ranging from 12 CPS – 30 CPS [44]. Kraus et al. (2000) suggest that the absent ABR is the result of the desynchronized high frequency biphasic action potentials, that are separated by fractions of a millisecond, being cancelled out during averaging. The present results align with Kraus, and further suggest that synchronized firing of the brainstem neurons does not improve even at very slow click rates.

While no difference in the ABR was observed at slow stimulus rates, more prominent cochlear microphonics (CM) were recognized in the early interval waveforms obtained from the slow rate ABRs compared to the fast rate ABRs. This is not surprising since the morphology (and hence clarity) of all waveform components, including the CM which reflects outer hair cell function, was expected to improve at a slower rate. Since outer hair cell function is not disrupted in ANSD, the diagnostic and prognostic relevance of this finding is unclear, and further research with a larger sample size is warranted to assess its significance.

The participants in this case series had also received P1 CAEP testing in conjunction with assessment of auditory skill development using the IT-MAIS. In each case, the P1 CAEP latency was related to the degree of auditory skill development, with normal P1 latency predicting typical development of auditory skills, which is consistent with previous research in this area [15–17].

Finally, in this case series, the absent, normal and delayed latency of the P1 CAEP was not reflected in any way in the ABR waveform. That is, all participants showed absent ABRs at slow and fast click rates, but had vastly differing patterns of central auditory maturation and speech and language development. Case 1 had P1 latencies within the normal range and a mildly delayed IT-MAIS score; Case 2 did not have an identifiable P1 and had a very low IT-MAIS score; and Case 3 initially had a borderline normal latency at 4 months of age, which became delayed by 11 months and then normalized by age 3 years and had an IT-MAIS score which increased with age and was within the expected range. Despite these differing P1 CAEP responses, none of the cases had any element other than a CM in the ABR that was discernible even when a slow rate of stimulus theoretically improved the likelihood of neural synchrony. This aligns with previous research which indicates that cortical potentials are not as sensitive to disruptions in neural synchrony as the ABR [13]. Previous studies have shown that measures such as inter-trial coherence (ITC) may be even more sensitive to measures of cortical phase synchrony in children with ANSD [45,46].

Overall the results of this case series suggest that the grossly abnormal or absent ABR is a useful tool in the diagnosis of ANSD and that using traditional ABR stimulation rates continues to be clinically appropriate for diagnosis. Considering prognosis, manipulating ABR click rate did not provide prognostic information; however, P1 CAEP latency predicted auditory skill development, as shown by the IT-MAIS scores. This is consistent with previous studies that the P1 CAEP is a reliable predictor of behavioral outcomes and is a reliable tool in clinical decision making [15–17].

Limitations of the study

This case series had the limitations inherent in a retrospective case series and further research is warranted. Caution is required in extrapolating from a small sample of an uncommon condition in a heterogeneous population. Since this study involved a retrospective review of cases, there was variation in the frequency and type of prior investigations for which results were available. Further, due to the fluctuating nature of ANSD pathology, ABR characteristics varied within and between the research participants. There was a lack of consistency in click stimulus rates since the ABRs were administered by different clinicians, with the stimulus for the “fast rate” ranging from 11.1 to 31.1 clicks per second. Additionally, participants differed in the numbers of slow and fast click rate ABRs that were collected. Research has failed to find significant differences in the ABRs of normal-hearing adults and children with ANSD with changes in the click rate from 12 – 30 CPS [44]; however, consistency in the faster rate would be preferable, and a future study would ideally specify the number of ABR runs and restrict stimulus rates to predetermined values.

This case series had only three research participants due to ANSD being a relatively rare disorder with an annual incidence of 1 – 3 new cases per 10,000 live births [47]. In a study of children whose neonatal or family histories put them at increased risk of hearing loss, Rance and colleagues determined the prevalence of ANSD to be 0.23%. [44]. Given the low prevalence of ANSD, many findings in the literature are reported in the form of case studies. The age range of the case studies in this series (1 year, 7 months – 6 years, 2 months) was not considered to be a limitation since each had an abnormal, but mature, ABR which was diagnostic of ANSD. In this respect, it was the waveform which was of interest rather than the specific clinical cases, in what is known to be a heterogeneous population. In future research, ideally more participants with ANSD would be recruited to participate in a prospective study and a larger sample size may yield more pronounced differences between cases or between differing clinical profiles.

Conclusion

This study produced a “negative” result and did not lend support to the original hypothesis that slowing the stimulus rate of the ABR would improve neural synchrony and result in a reproducible and replicable pattern within the ABR. Although the size of this study was small, it is essential to document such negative findings and further explore the auditory stimulus used to elicit diagnostic information. In contrast, P1 CAEP responses provided valuable information regarding the maturational status of the auditory cortex and P1 CAEP responses were consistent with behavioral measures of auditory speech perception. As understanding and use of electrophysiological biomarkers evolves in this heterogeneous population with ANSD, it will become possible to further individualize intervention to maximize communication function, and to predict speech and language outcomes.