|Year : 2016 | Volume
| Issue : 4 | Page : 221-230
Evaluation of otoacoustic emissions and auditory brainstem responses for hearing screening of high risk infants
Tania Nazir1, Sachin Gupta2, Ghulam Mohammad Mir3, Ashu Jamwal4, Parmod Kalsotra3, KP Singh3
1 Department of Health, Acharya Shri Chander College of Medical Sciences, Sidhra, India
2 Department of ENT and HNS, Acharya Shri Chander College of Medical Sciences, Sidhra, India
3 Department of ENT and HNS, Sri Maharaja Gulab Singh Hospital, GMC, Jammu, India
4 Department of Paediatrics, Sri Maharaja Gulab Singh Hospital, GMC, Jammu, India
|Date of Web Publication||13-Oct-2016|
Ward No. 6, Karan Nagar, Udhampur - 182 101, Jammu and Kashmir
Source of Support: None, Conflict of Interest: None
Aim: The objective of the present study is the assessment of otoacoustic emissions (OAEs) and auditory brainstem responses (ABRs) for hearing screening of high risk infants. Study Design: Prospective, hospital-based. Materials and Methods: Distortion product OAEs (DPOAEs) and brainstem evoked response audiometry (BERA) recordings were obtained for 30 controls and 100 infants with one or more high risk factors, in a sound treated room and the results were interpreted. ABR peak latencies, amplitudes, and waveform morphology in high risk infants were compared with those in control group. DPOAE as screening test was evaluated in terms of various parameters with BERA/ABR taken as gold standard. Results: Absolute latencies of Wave I and Wave V and interpeak latency of I-V were significantly prolonged in high risk group as compared to control group. The most common causes to contribute significantly for hearing impairment were found to be hyperbilirubinemia, birth asphyxia, meningitis/septicemia. DPOAE when compared with ABR taken as gold standard showed that sensitivity of the test was 87.7% (74.5%-94.9%) and specificity was 74.5% (60.0%-85.2%). Positive predictive value was 76.7% (63.2%-86.6%) and negative predictive value of the test was 86% (71.9%-94.3%). Positive likelihood ratio was 0.29 (0.18-0.46) and negative likelihood ratio was 6.08 (2.82-13.09). Conclusion : ABR/BERA, though highly reliable, is a tedious and time consuming test. DPOAE is a simple and rapid test with relatively higher acceptability but low sensitivity and specificity; therefore, limits its role as independent screening test. DPOAE-ABR test series is an effective way to screen all the high risk infants at the earliest.
Keywords: Auditory brainstem response, Brainstem evoked response audiometry, Distortion product otoacoustic emissions, Peak-latency, screening test
|How to cite this article:|
Nazir T, Gupta S, Mir GM, Jamwal A, Kalsotra P, Singh K P. Evaluation of otoacoustic emissions and auditory brainstem responses for hearing screening of high risk infants. Indian J Otol 2016;22:221-30
|How to cite this URL:|
Nazir T, Gupta S, Mir GM, Jamwal A, Kalsotra P, Singh K P. Evaluation of otoacoustic emissions and auditory brainstem responses for hearing screening of high risk infants. Indian J Otol [serial online] 2016 [cited 2021 Jul 30];22:221-30. Available from: https://www.indianjotol.org/text.asp?2016/22/4/221/192131
| Introduction|| |
Hearing is a sense essential to normal communication and consequently a normal life for all individuals. It, however, assumes greater significance in the early years of life when the faculties of speech and hearing develop concurrently and the one is dependent on the other. Both are essential for normal growth of language, cognition and behavior.
Hearing impairment is the most common form of sensory disorder in humans.  The estimated incidence of hearing loss in newborn infants ranges from 0.001% to 0.5% and may be as high as 1%-5% in high risk infants.  Hearing impairment can impose a heavy social and economic burden on individuals, families, communities, and countries. Children with profound sensorineural hearing loss (SNHL) experience delays in learning to understand the speech of others and to produce intelligible speech. The source of their delays is rooted in a lack of refined access to the spectral and temporal cues of the acoustic-phonologic components of speech.
It was observed that children whose hearing loss has been observed and managed before 6 months of age had higher scores of vocabulary, better expressive, and comprehensive language skills than those diagnosed and managed after 6 months of age emphasizing the importance of early identification and treatment. 
Hearing assessment in children is one of the dark areas in spite of the fact that two out of every 1000 children have permanent bilateral hearing loss above 60 dB.  During the past 30 years, infant hearing screening have been attempted with a number of different methods using behavioral and physiological measures.  The behavioral observation audiometry (BOA) is one of the methods of assessing hearing function. It is, however, highly insensitive besides having a low specificity.  Crib-O-Gram, another automated BOA, is also highly unreliable. 
Auditory brainstem response (ABR) and otoacoustic emissions (OAEs) have emerged as the reliable objective methods of hearing screening. OAEs can occur spontaneously or in response to acoustic stimulation. For the most clinical applications, OAEs are evoked using transients (transient evoked OAEs) or tone pairs (distortion product OAEs [DPOAEs]). Both types of OAEs are altered in the presence of mild hearing loss. 
The present study aims to evaluate the role of both modalities in screening of hearing impairment in neonates having one or more high risk factors.
| Materials and Methods|| |
This study was carried out over a period of 1 year from November 2013 to October 2014 in a group of 100 high-risk infants hospitalized in Neonatal Ward and Neonatal Intensive Care, Pediatrics Indoor and in SMGS Hospital, Government Medical College, Jammu, which is a tertiary care center. For control group, thirty normal infants with no high risk factors were included in the study.
Infants considered to be at high risk for hearing loss were based on the recommendations of Joint Committee on Infant Hearing Position Statement.  These risk factors include the following:
- Family history of hereditary childhood SNHL
- Congenital perinatal toxoplasma, rubella, cytomegalovirus, and herpes (TORCH) infection
- Craniofacial anomalies
- Birth weight < 1500 g
- Hyperbilirubinemia more than 15 mg/dl
- Ototoxic medication
- Bacterial meningitis or bacteriological proven sepsis
- Mechanical ventilation more than 5 days or longer
- Postnatal asphyxia (Apgar Score of 0-4 at 1 min or 0-6 at 5 min)
- Stigmata or other findings associated with a syndrome known to include a sensorineural and/or conductive hearing loss.
Only those infants who had one or more of the high risk factors were enrolled in this study for hearing screening using OAE and ABR.
The purpose of the study and detail protocol was discussed with one of the parents and informed consent for both the tests, i.e., ABR and OAE was obtained on the 1 st visit of screening and compared. High risk infants who failed were again retested after 3-6 months for the confirmation of hearing impairment on follow-up with ABR test. All the following information was recorded: Gestational age, sex, birth weight, admitting diagnosis, antenatal, perinatal and postnatal history, and drug history.
DPOAE and brainstem evoked response audiometry (BERA) recordings were obtained for 30 control and 100 high risk infants in a sound treated room and the results were interpreted.
These babies were tested while they were in natural sleep or in a state of quite rest. Sedation with syrup pedichloryl (trichlofos) at dose of 25 mg/kg (0.25 ml/kg) was administered as sedative. The time taken to complete DPOAE and BERA for every subject was recorded using a stopwatch.
The OAE protocol consists of hearing each ear with neurosoft neuro-audio device. The DPOAEs were performed in an acoustic chamber with background noise <30 dB and Neurosoft Neuro-audio device (OAE System Otodynamics) connected to a laptop computer with DPGRAM software and a microphone which was introduced in the external acoustic meatus and sealed with a flexible rubber mold. During the test, the child was held in his/her parent or guardian's lap, preferably while asleep. An intensity of 65 dB HL was used for the 500 Hz frequency (F1), with 55 dB HL for F2 (1000 Hz), respectively, in ascending frequency order and frequency ratio F1/F2 as 1.22.
Distortion product otoacoustic emissions
When two tones are presented simultaneously to a healthy cochlea, the response measured in the ear canal will contain several tones that are not present in the eliciting stimuli. These additional tones are called DPs and are attributable to the nonlinear processes of the normally functioning cochlea. The most robust of these DPs occurs at the frequency equal to 2f1-f2, where f1 indicates the lower frequency tone and f2 indicates the higher frequency tone of the pair. An f2/f1 ratio of 1.22 produces the largest 2f1-f2 DPs, on the average, in both adults and neonates. It is the cochlear integrity being assessed using these stimulus parameters.
Auditory brainstem evoked response/brainstem evoked response audiometry
Neurosoft neuro-audio single channel auditory evoked potentials system was used for testing infants. After adequate preparation of skin, recording silver electrodes were attached to upper forehead (recording electrode), the ipsilateral mastoid process (reference electrodes), the contralateral mastoid process (ground electrode). ABR threshold estimation was carried by neuro-audio device with the selection of following parameters:
- Stimulus - Clicks (based within 100 Hz-2500 Hz)
- Polarity - Alternate
- Number of clicks - 2000/trace
- Internal range - 100 db to 30 db (with 5 db steps)
- Electrode impedence - <5 db.
The ABR protocol consists of testing each ear at 70, 60, 50, 40, and 30 db HL. If a response would not be observed at 70 db HL, testings were performed at 80 and 90 db HL. An infant will be considered pass the ABR test if a replicable wave V response was present at 30 db HL in both ears.
Monoaural auditory stimulus consisting of rarefaction clicks of 100 μs were delivered through electrically shielded earphones at the rate of 11.1/s. Contralateral ear was masked with pure white noise of 40 dB. A band pass of 150-3000 Hz was used to filter out the undesirable frequencies in the surroundings. Responses to 2000 click presentation were averaged.
BERA threshold for each ear with absolute latencies of wave I, II, III, IV, and V waves interpeak latencies (IPL) of I-III, I-V, and III-V were considered from the recording for comparison among high risk infants and controls.
The hearing impairment was graded as follows: Normal (<30 db), mild (30-40 db), moderate (40-70 db), and severe (more than 70 db) (5).
High risk infants who failed DPOAE and BERA/ABR as screening test at the 1 st visit were asked to return for follow-up BERA/ABR evaluation as confirmatory test after 3-6 months of 1 st visit. If the results were consistent with the original study at follow-up, hearing loss was confirmed. The parents were informed of the results and were advised that further testing were necessary to confirm the exact status of their child's hearing. Specific follow-up appointment will be arranged in conjunction with the child's routine pediatric visit.
The present study was a longitudinal observational study in which the results were expressed as mean and standard deviation. Unpaired t-test was used for intergroup comparisons, P ≤ 0.05 was considered as statistically significant.
| Results|| |
The present study, a longitudinal observational study, entitled "evaluation of OAE and audiometry brainstem responses in high risk infants" was carried out in the Department of Otorhinolaryngology and Head and Neck Surgery, SMGS. Hospital, Government Medical College, Jammu, over a period of 1 year from November 2013 to October 2014. In the present study, 30 controls and 100 high risk infants were analyzed for the study. The results were expressed in mean and standard deviation. Unpaired t-test was used as test of significance. And the age group for screening of 100 high risk infants and 30 normal infants from the control group taken was 0-12 months.
In the present study, 100 high risk infants and 30 normal control infants were analyzed for the evaluation of DPOAE and ABR (ABR/BERA) for the hearing screening. All the 30 (100%) normal control infants passed the DPOAE and ABR/BERA test.
Out of 100 high risk infants tested with ABRs (BERA/ABR) at 1 st visit, 49 (49%) infants showed a clear and reproducible wave V (<30 dB HL) and 51 (51%) infants having varying degrees of hearing loss on 1 st visit [Figure 1]. As per the results of BERA/ABR test, the incidence of hearing impairment was 51%.
|Figure 1: Auditory brainstem response/brainstem evoked response audiometry test results in high risk infants on 1st visit|
Click here to view
High risk infants included a total of 100 infants and control group consisted of 30 normal infants. Out of 100 high risk infants, 49 high risk infants who passed BERA/ABR test at 1 st visit were compared with BERA/ABR test of 30 normal infants. Fifty-one high risk infants who failed BERA/ABR test during 1 st screening were not included in statistical analysis of comparison of the peak and IPL. The peak and IPL of wave I, III, and V of normal infants was compared with the peak and interpeak latency of wave I, III, and V of 49 high risk infants.
Wave I: Wave I threshold were compared in 49 high risk infants and 30 normal controls respectively [Table 1]. The latency values in the study did not show any significant difference from control at lower intensity level of 50 dB HL, but at higher intensity shows statistically significant difference in the Wave I latencies. An analysis of figures obtained revealed significant results and increased latency of Wave I in high risk group as compared to the control group (P < 0.05) at 70-30 dB HL.
|Table 1: Distribution and comparison of study population on the basis of Wave I peak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Wave III threshold were compared in 49 high risk infants and 30 normal controls respectively [Table 2]. An analysis of figures obtained did not reveal a significant or consistent alteration in the values of high risk group as compared to the control group at 70-30 dB HL, except for 50 dB HL which was statistically significant.
|Table 2: Distribution and comparison of study population on the basis of Wave III peak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Wave V: Wave V threshold in 49 high risk infants was compared with the wave V threshold of 30 normal infants [Table 3]. It showed significant increase in wave V latency in high risk group at 70-30 dB HL.
|Table 3: Distribution and comparison of study population on the basis of wave V peak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Interpeak latency wave I-III did not reveal any significant or consistent alteration in the values of high risk group at 70-30 dB HL as compared to the control group [Table 4].
|Table 4: Distribution and comparison of study population on the basis of Wave I-III interpeak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Interpeak latency wave III-V did not reveal any significant or consistent alteration in the values of high risk group from 70 to 30 dB HL as compared to control group [Table 5].
|Table 5: Distribution and comparison of study population on the basis of Wave III-V interpeak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Interpeak latency Wave I-V threshold of 49 high risk infants who passed the 1 st screening were compared with 30 normal controls [Table 6]. Prolonged I-V interpeak latency was observed in all intensities. An analysis of the figures obtained revealed significant result and the difference was statistically highly significant with P < 0.001.
|Table 6: Distribution and comparison of study population on the basis of Wave I-V interpeak of brainstem evoked response audiometry/auditory brainstem response|
Click here to view
Hearing thresholds of both groups, high risk and control, were established and compared. Abnormal ABRs (ABR/BERA) were classified based on the ABR threshold [Figure 2].  The degree of auditory deficit was estimated as mild, moderate, and severe. Normal hearing was considered as identifiable and Wave V at or <30 dB HL. Mild deficits were identified with unidentified Wave V at 30 dB HL but an identifiable and replicable Wave V at 40 dB HL. Moderate deficits were identifiable with an absent Wave V at 40-60 dB HL. Infants labeled to have severe deficits if they failed to show any identifiable Wave V at 70 dB HL.
|Figure 2: Auditory brainstem response/brainstem evoked response audiometry grading of threshold groups according to minimum stimulus-level evoking Wave V of high risk infants at 1st visit|
Click here to view
Control group of 30 normal infants showed a consistent and well recognizable response at all the stimulus intensities. Hence, 100% of control infants had hearing better than 30 dB HL.
In the present study, out of 100 high risk infants tested with ABRs (BERA/ABR) at 1 st visit, 49 infants showed a clear and reproducible Wave V (<30 dB HL) and 51 infants having varying degrees of hearing loss. Out of 51 failed infants, grading was done observed that 29 (29%) had mild hearing loss (30-40 dB HL), 17 (17%) had moderate hearing loss (40-70 dB HL), and 5 had severe hearing loss (>70 dB HL).
In the present study, follow up was done after 3-6 months after 1 st screening test of ABRs (BERA/ABR) to confirm the hearing loss. On follow up, BERA/ABRs were performed on 51 high risk infants who failed the initial BERA/ABR screening. Out of 51 high risk infants, 38 high risk infants showed reproducible Wave V (<30 dB HL) on repeat ABRs test. Further classification was based on repeat ABRs test finding of the infants who failed during initial BERA/ABR screening [Figure 3]. About 13 of 51 infants continued to have persistent hearing loss [Table 7]. Repeated ABRs confirmed an abnormal follow-up testing of 13 infants within 3-6 months.
|Figure 3: Auditory brainstem response/brainstem evoked response audiometry grading of threshold groups according to minimum stimulus-level evoking Wave V of high risk infants with abnormal auditory brainstem response/brainstem evoked response audiometry for confirmation on follow-up after 3 months|
Click here to view
|Table 7: Final result of auditory brainstem response/ brainstem evoked response audiometry on 1st screening and on follow up|
Click here to view
Out of 51, failed infants were retested on follow-up after 3-6 months of first visit and observed that 38 infants had normal hearing who passed ABR/BERA (<30 dB HL), 9 had moderate hearing loss (40-70 dB HL), and 4 had severe hearing loss (>70 dB HL).
Risk factor evaluation in affected 13 high risk infants shows that the four most frequent risk factors observed in failed screening group were hyperbilirubinemia 4 (22%), meningitis/septicemia 6 (16.6%), birth asphyxia 6 (15.3%), congenital TORCH infection 1 (33.4%) infants, family h/o congenital SNHL 1 (100%), and craniofacial anomalies 1 (100%). Among these infants with these risk factors, chances of hearing loss were higher for those who have family h/o hearing loss, having craniofacial anomalies and those with congenital TORCH infection. As there was only one infant in each criteria, they were not considered statistically evaluable. However, hyperbilirubinemia and meningitis/septicemia was related to likelihood of hearing loss.
Out of 100 high risk infants, 56 (56%) infants passed DPOAE test, 44 (44%) infants were referred (failed) DPOAE test. As per the results of DPOAE test, the incidence of hearing impairment was 44%.
In the present study, the study group of 100 high risk infant was tested with ABRs (ABR/BERA) at 1 st screening visit. Out of 100 high risk infants tested with ABRs BERA/ABR at 1 st visit, 49 infants passed and 51 failed. Similarly, DPOAE test was applied to 100 high risk infants. Out of 100 high risk infants tested with DPOAE test at 1 st visit, 56 infants passed and 44 failed. The comparison was done for those high risk infants who passed the ABR/BERA and DPOAE test versus those high risk infants who failed ABR/BERA and DPOAE auditory screening test to see the effect of risk factor on the passed or failed group.
In the current study, ABR (ABR/BERA) test was taken as gold standard and DPOAE was compared with this gold standard. The present study observed that 56 out of 100 passed the DPOAE test and 44 failed DPOAE test and 49 out of 100 passed ABR/BERA and 51 failed ABR/BERA.
When sensitivity and specificity of DPOAE was evaluated against ABR (gold standard), it was observed that DPOAE had good level of accuracy in the detection of hearing loss in infants and there was good concordance between OAE and ABR results among high-risk children referred for hearing screening.
The time taken to complete DPOAE and BERA for every subject was recorded. This includes the time required for preparation of baby, sedating the baby, for probe fitting applying electrodes for recordings their interpretation, etc., In this study, average time taken as calculated from the point of receiving baby to its discharge from audiology room was 30-40 min per neonate for BERA and 10-15 min per neonate for DPOAE test.
In the present study, the acceptability of DPOAE was higher among parents as compared to BERA because of its simple procedure of testing. At the time of taking consent, approximately 4 (4%) of parents perceived BERA as relatively invasive test.
In the present study, all the recordings were performed in a sound treated room so that DPOAE could be truly evaluated as a screening modality of hearing impairment.
| Discussion|| |
The incidence of hearing impairment in high risk infants according to different statistics varies from 1% to 40%.  There are several risk factors which precipitate hearing impairment in newborns and young infants. Most important among these are - birth asphyxia, prematurity/low birth weight, hyperbilirubinemia, bacterial meningitis, intrauterine infections, use of ototoxic drugs (Aminoglycosides), craniofacial anomalies, syndromes/stigmata, and prolonged mechanical ventilation. 
In the present study, the incidence of hearing impairment using BERA/ABR in the infants on 1 st visit who had at least one risk factor was 51% and on follow-up was 13%. These results were lower as compared with Aiyer and Parikh  and Lakshmi et al., who found in their studies the incidence of hearing impairment in the infants who had at least one risk factor to be 76.67% and 64.9%, respectively.
Similarly, Colella-Santos et al. found in their study that hearing loss was identified in 58% of the sample, conductive hearing loss represented 31.5% (12/38) and neurosensory in 28.9% of cases. This was much higher in comparison with study of Richardson et al. in which they found that 7 (6.3%) of the 110 children had abnormal ABR threshold >30 dB. The difference was perhaps due to referral nature which provides care to highly complex cases.
Distribution and comparison of study group (normal control group and high risk group infants) on the basis of peak and interpeak latencies
ABR peak latencies, amplitudes, and waveform morphology can provide information about the maturation and integrity of the auditory nerve and lower brainstem pathways. The ABR waveforms of infants and young children differ in several important ways from those recorded in adults. At higher intensities, the infant waveform shows fewer identifiable peak components than routinely observed in adults. Waves I, III, and V are easily identified. Absolute and IPL also tend to be prolonged compared with those obtained in adults. In normally developing infants, waveform morphology and peak latencies approach adult values by about the age of 18-24 months. However, latency prolongations may persist in children beyond this age range secondary to sensory loss, conductive involvement, or developmental delays, particularly at lower intensity levels. Therefore, in assessing infants and young children for the purpose of threshold prediction, it is important to extend the recording or analysis window beyond the 10 ms typically used with adults. 
The purpose of comparison of various parameters of ABR/BERA in control and high risk was to determine the presence or absence of correlation of the abnormalities of ABR/BERA with various neonatal insults, which may affect infant in early postnatal life. The study compared wave latencies and the IPL of normal and high risk infants. ,
Wave I: It indicates the cochlear function of auditory tract and increase in Wave I parameter signifies a lesion at the level of the middle ear or inner ear. The latency values in the study did not show any significant difference from control at lower intensity level except for 50 dB HL, but at higher intensity showed significant difference in the Wave I latencies. This may be explained on the basis of the observation that in increasing intensity levels, there is shortening of the wave latencies due to the rapid firing and quicker rising of synaptic potential and therefore, a decreased latency of transmission. It can be hypothesized that this increase in the firing rate increasing intensities in compromised infants does not equal that of normal infants.  Hence, though the latencies may be comparable at lower intensities due to normal cochlear function, difference occurs at the higher intensities.
Sleifer et al. found in their study that the absolute latencies of peaks I, III, and IV and the interpeak intervals I-III, I-IV, and III-V presented statistically significant differences between the groups of premature and mature babies at ages 4 and 12 months. At 20 months, only peak I failed to show a difference in absolute latency and concluded that maturation of the auditory system occurs differently between premature and full-term children and suggested that gestational age should be taken into consideration when using ABR in premature children younger than 20 months old.
Wave III: In the present study, latency of wave III was not statistically significant and did not reveal any significant alteration in the values in high risk groups as compared to control group. It was also stated by Aiyer and Parikh  in their study that ABR parameters, especially wave III did not reveal any significant or consistent alteration in the values of high risk groups as compared to control group.
Wave V: The present study showed significant increase in Wave V latency in high risk groups at 30-70 dB HL. The absolute latency of Wave V is consistent and stable parameter that is why it has received primary attention as a valuable factor in response evaluation. It is suggested that all the risk factors which bring the neonate under intensive care induce a certain amount of hypoxia of cochlea and brainstem, which leads to various cellular changes edema degeneration and necrosis. Hence, they predispose to various hearing impairment, which may be reversible following the reversal of hypoxic changes Shahnaz.  The present study demonstrated significant prolonged Wave V latency.
Jiang et al. in their study of preterm babies with perinatal complication had significant increase in Wave V latency, I-V, III-V IPL, and III-V/I-III interval ratio compared to preterm without perinatal complications and normal term babies. Aiyer and Parikh  found in their study that ABR parameters, especially Wave V were significantly prolonged in high risk infants which was also comparable to the present study. Lakshmi et al. conducted a study on 100 high risk infants who were subjected to BERA and observed that the high risk infants had increased wave V threshold when compared to the control group.
Interpeak latency I-III and III-V
In the present study, the interpeak latency I-III and III-V were statistically not significant in the values of high risk group infants as compared to control group. This finding is comparable to the results of Aiyer and Parikh,  wherein authors found that ABR parameters such as interpeak latency I-III and III-V did not reveal any significant or consistent alteration in the values of high risk groups as compared to control group.
Interpeak latency I-V
The present study demonstrated a significant increase of Wave I-V interpeak latency. The interpeak latency is a reflection of neural conduction time between the auditory nerve and the brainstem nuclei and reflects upon the efficiency of the auditory pathway. Prolonged I-V interpeak latency is a feature of neurological impairment and is indicative of delay in neurological conduction within the brainstem. Evidence that both the peripheral and central components may be abnormal during the early stages and the impairment may be temporary in some infants is provided by Kinely and Magathan. 
Jiang et al.  study with preterm babies with perinatal complication had significant increase in Wave V latency, I-V, III-V IPL, and III-V/I-III interval ratio compared to preterm without perinatal complications and normal term babies. The present study results are almost consistent with Aiyer and Parikh  who found in their study that ABR parameters interwave I-V intervals were significantly prolonged in high risk infants. Lakshmi et al.  found in their study that absolute latencies of Wave III, V, IPL of I-III and I-V were prolonged in their cases.
Distribution on the basis of risk factors for hearing impairment on 1 st hearing screening of auditory brainstem response
In the present study, the most common cause which better characterizes the group "at risk" of hearing impairment during 1 st screening were birth asphyxia 22 (56.4%) infants, hyperbilirubinemia 11 (61.2%), family h/o congenital SNHL 1 (100%), and craniofacial anomalies 1 (100%). Among the infants with these risk factors, chances of hearing loss was higher for those who had family h/o hearing loss and had craniofacial anomalies. However, as there were only one infant in each criteria, they were not considered statistically evaluable. However, hyperbilirubinemia and birth asphyxia were found to have significantly correlated with hearing impairment. The literature shows different authors described different factors as the leading causes of hearing loss in children [Table 8].
|Table 8: High risk factors as causes of hearing loss in infants by various authors|
Click here to view
The difference in the results by different authors may be because of variability in referral cases. These results suggests that it is necessary to take objective measures for assessing the serum levels of bilirubin and efficient phototherapy, which represent the measures to prevent hearing impairment resulting from hyperbilirubinemia.
On follow-up, retest with ABR/BERA after 3-6 months on 51 high risk infants who failed during 1 st ABR/BERA test revealed that 13 of 51 high risk infants continued to have persistent hearing loss. Out of 13 high risk infants, 9 had moderate hearing loss and 4 had severe hearing loss. Alberti et al. found that on follow-up, BERA confirmed hearing loss in 8% of babies in their study group. Aiyer and Parikh  found in their study that incidence of severe to profound hearing loss confirmed on follow-up ranges from 2% to 4% and they reported incidence of 3.33%. Bedajit et al.  found in their study that incidence of hearing loss in the high risk group was 3.1% (3/95).
Distribution on the basis of risk factors for hearing impairment on follow-up hearing screening of auditory brainstem response
Risk factor evaluation in affected 13 high risk infants shows [Table 9] that the four most frequent risk factors observed in failed screening group were hyperbilirubinemia 4 (22%), meningitis/septicemia 6 (16.6%), birth asphyxia 6 (15.3%), congenital TORCH infection 1 (33.4%) infants, family h/o congenital SNHL 1 (100%), and craniofacial anomalies 1 (100%). Among these infants with these risk factors, chances of hearing loss were higher for those who have family h/o hearing loss, having craniofacial anomalies, and those with congenital TORCH infection. As there were only one infant in each criteria, they were not considered statistically evaluable. However, hyperbilirubinemia and meningitis/septicemia was related to likelihood of hearing loss.
|Table 9: Risk factor evaluation in affected infants on 1st visit and follow-up|
Click here to view
Assessment of hearing impairment with distortion product otoacoustic emissions
In the present study, 100 high risk infants were analyzed for DPOAE, 56 (56%) infants passed DPOAE test, 44 (44%) infants failed to pass DPOAE test [Figure 4]. As per the results of DPOAE test, the incidence of hearing impairment was 44%.
|Figure 4: Distortion product otoacoustic emission test results in high risk infants|
Click here to view
Xu et al. in their study observed that 54 ears (27%) failed the DPOAE test during the initial screening.
Distortion product otoacoustic emissions as screening test when brainstem evoked response audiometry/auditory brainstem response is taken as gold standard
In the present study, the pass rate of DPOAE calculated was 56% and 44% high risk infants failed the DPOAE test. The incidence of hearing impairment in the infants with BERA/ABR on 1 st visit who had at least one risk factor was 51% and on follow-up was 13% [Table 10]. Xu et al. found in their study on 600 neonates (1200 ears), the incidence of ABR abnormality (78.6%, 943/1200) was remarkably higher than that of DPOAE abnormality (22.3%, 268/1200).
|Table 10: Comparison of distortion product otoacoustic emission and brainstem evoked response audiometry/auditory brainstem response with auditory brainstem response as gold standard test|
Click here to view
The sensitivity and specificity of DPOAE as screening test was calculated with BERA/ABR as gold standard. The sensitivity of DPOAE was found to be 87.7% (74.5-94.9%) and specificity of the test was 74.5% (60.0-85.2%) [Table 11].
|Table 11: Results of auditory brainstem response and distortion product otoacoustic emission results with auditory brainstem response as gold standard test|
Click here to view
The present study results are almost in agreement with the results of Llanes and Chiong  who reported in their study that OAEs had a sensitivity of 76.9% and a specificity of 90% and concluded that there was good concordance between OAE and ABR results among high-risk children referred for hearing screening.
Richardson et al. found in their study that OAE screening test had a sensitivity of 1.00, a specificity of 0.91, a positive predictive value of 0.44 and a negative predictive value of 1.00. Xu et al. found that the sensitivity was 97% for DPOAE, while Dhawan and Mathur  found in their study that sensitivity of 80% and specificity of 92.8% which is higher than the present study.
In the present study, we were able to identify two factors that significantly reduced the specificity of DPOAE. The first was blocked external auditory canal (EAC) and the second was collapsible EAC. Similar observations were reported in study of Dhawan and Mathur. 
| Conclusion|| |
The period from birth to 2 years of life is critical for the development of speech and language; therefore there is a need for early identification and assessment of hearing loss and early rehabilitation in infants and children, particularly in high risk infants. Early identification of mild to moderate hearing loss that are amenable to treatment and subsequent measures could be detrimental in preventing further deterioration of hearing. Those already having severe hearing loss can be suitably helped with hearing aids to prevent delay in speech development. The quest is still on to find a reliable, accurate, cost-effective and simple method as a universally acceptable screening tool to identify hearing loss in children. The present study evaluates the role of BERA and OAE in such a scenario.
During hearing assessment, the most common ABR/BERA change noted in the present study were absolute latencies of Wave I and Wave V and IPL of I-V which were significantly prolonged in high risk group as compared to control group. Out of high risk factors assessed in our study hyperbilirubinemia, birth asphyxia, meningitis/septicemia, craniofacial anomalies were seen to contribute significantly for hearing impairment.
DPOAE is a simple and rapid test with relatively higher acceptability and therefore has a major role as a screening tool, especially in the countries like India with very high birth rate. However, the low sensitivity and specificity as compared to ABR/BERA limit its acceptability of the status of independent screening modality for hearing impairment in infants. Therefore, DPOAE cannot completely replace ABR/BERA as screening modality for hearing impairment in neonates, however can compliment it. DPOAE-ABR test series is an effective way to screen all the high risk infants at the earliest, to prevent adverse effect on the developing auditory pathway.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Mhatre AN, Lalwani AK. Molecular genetics of deafness. Otolaryngol Clin North Am 1996;29:421-35.
Anand NK, Gupta AK, Raj H. BERA - a diagnostic tool in neonatology. Indian Pediatr 1990;27:1039-44.
Fisher NA, Curtin HD. Radiology of congenital hearing loss. Otolaryngol Clin North Am 1994;27:511-31.
Biswas A. Clinical audio-vestibulometry for otologists and neurologists. Assessing the Deaf Child. 3 rd
ed., Vol. 3. Mumbai: Bhalani Publishing House; 2002. p. 97.
Aiyer RG, Parikh B. Evaluation of auditory brainstem responses for hearing screening of high-risk infants. Indian J Otolaryngol Head Neck Surg 2009;61:47-53.
Cerber SE. Newborn hearing screening and diagnostic tests. In: Jaffe BF, editor. Hearing Loss in Children. Maryland: University Park Press; 1977. p. 78.
Durieux-Smith A, Picton T, Edwards C, Goodman JT, MacMurray B. The crib-o-gram in the NICU: An evaluation based on brain stem electric response audiometry. Ear Hear 1985;6:20-4.
Johnson KC. Audiologic assessment of children with suspected hearing loss. Otolaryngol Clin North Am 2002;35:711-32.
Joint Committee of Infant Hearing Position Statement. Am Speech Hear Assoc 1994;36:38-41.
Ansari MS. Screening programme for hearing impairment in newborns: A challenge during rehabilitation for all. Asia Pac Disabil Rehabil J 2004;15:83-9.
Lakshmi T, Sultana Z, Brid SV. Comparative study of brainstem evoked response audiometry in high risk infants. Int J Recent Trends Sci Technol 2014;12:8-12.
Colella-Santos MF, Françozo Mde F, do Couto CM, Lima MC, Tazinazzio TG, Castilho AM, et al.
Audiological and genetics studies in high-risk infants. Braz J Otorhinolaryngol 2011;77:784-90.
Richardson MP, Williamson TJ, Reid A, Tarlow MJ, Rudd PT. Otoacoustic emissions as a screening test for hearing impairment in children recovering from acute bacterial meningitis. Pediatrics 1998;102:1364-8.
Yang EY, Stuart A, Mencher GT, Mencher LS, Vincer MJ. Auditory brain stem responses to air- and bone-conducted clicks in the audiological assessment of at-risk infants. Ear Hear 1993;14:175-82.
Sleifer P, da Costa SS, Cóser PL, Goldani MZ, Dornelles C, Weiss K. Auditory brainstem response in premature and full-term children. Int J Pediatr Otorhinolaryngol 2007;71:1449-56.
Duara S, Suter CM, Bessard KK, Gutberlet RL. Neonatal screening with auditory brainstem responses: Results of follow-up audiometry and risk factor evaluation. J Pediatr 1986;108:276-81.
Jiang ZD, Brosi DM, Li ZH, Chen C, Wilkinson AR. Brainstem auditory function at term in preterm babies with and without perinatal complications. Pediatr Res 2005;58:1164-9.
Kileny PR, Magathan MG. Predictive value of ABR in infants and children with moderate to profound hearing impairment. Ear Hear 1987;8:217-21.
Alberti PW, Hyde ML, Riko K, Corbin H, Abramovich S. An evaluation of BERA for hearing screening in high-risk neonates. Laryngoscope 1983;93:1115-21.
Bedajit RK, Babu AP, Kumar JS, Mallik P, Singh MM, Devi P. Brainstem evoked response audiometry for the assessment of auditory pathway in newborns and high risk Infants. Natl J Otorhinolaryngol Head Neck Surg 2014;2:26-9.
Xu ZM, Li J, Hu TZ, Sun JH, Shen XM. Sensitivity of distortion product otoacoustic emissions and auditory brain-stem response in neonatal hearing screening, a comparative study. Zhonghua Yi Xue Za Zhi 2003;83:278-80.
Xu FL, Xing QJ, Cheng XY. A comparison of auditory brainstem responses and otoacoustic emissions in hearing screening of high-risk neonates. Zhongguo Dang Dai Er Ke Za Zhi 2008;10:460-3.
Llanes EG, Chiong CM. Evoked otoacoustic emissions and auditory brainstem responses: Concordance in hearing screening among high-risk children. Acta Otolaryngol 2004;124:387-90.
Dhawan R, Mathur N. Comparative evaluation of transient evoked otoacoustic emissions and Brainstem evoked response audiometry as screening modality for hearing impairment in neonates. Indian J Otolaryngol Head Neck Surg 2007;59:15-8.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8], [Table 9], [Table 10], [Table 11]