Indian J Otolaryngol Head Neck Surg

(July–Sept 2016) 68(3):334–338; DOI 10.1007/s12070-014-0811-6

ORIGINAL ARTICLE

Brainstem Evoked Response Audiometry (BERA) in Neonates

with Hyperbillirubinemia
Annanya Soni • Surendra Kumar Kanaujia •

Sandeep Kaushik

Received: 26 September 2014 / Accepted: 28 November 2014 / Published online: 3 December 2014
 Association of Otolaryngologists of India 2014

Abstract (1) To study the BERA changes in neonates Keywords Brainstem evoked response audiometry
with unconjugated hyperbilirubinemia. (2) To compare the (BERA)
Bilirubin toxicity
Auditory functions

BERA changes in the neonates with unconjugated hyper- Neonatal hyperbillirubinemia
bilirubinemia before and after therapy. Thirty consecutive
term appropriate for gestational age (AGA) neonates pre-
senting to NICU with total serum bilirubin requiring Introduction
intervention (using the American Academy of Pediatrics
guidelines) were included in the study as cases and thirty Neonatal jaundice is a common problem seen in the new-
normal term AGA neonates with uneventful peri-natal born. It is observed during the first week of life in
period and a maximum measured serum bilirubin \12 mg/ approximately 60 % of term and 80 % of preterm [1, 2].
dl in case of term baby were included as controls after Billirubin in healthy term neonates especially free bili-
obtaining informed consent. Initial BERA was done within rubin (Bf) could enter brain cells and cause damage which
3–24 h of hospitalization after obtaining informed consent is perceived by brainstem evoked response audiometry
from parents, at the time of discharge and at 3 month (BERA). Studies identify that Bf is more associated to
followup. Machine used for recording BERA was intelli- abnormal BERA.
gent hearing system version 3.3. In our study out of the 30 Early detection and rehabilitation of a hearing loss is
cases 10 (33.3 %) cases were found to have BERA changes important for the development of speech and language
in the form of absent wave forms, raised threshold, pro- skills in hearing impaired children [3].
longed latencies or prolonged inter peak latencies. In our BERA is an effective and non-invasive means of
study, it was observed that there was statistically significant assessing the functional status of the auditory nerve and
correlation (p value \ 0.005) between increasing bilirubin brainstem auditory sensory pathway. It is not significantly
level and BERA changes. Correlation of the findings of this altered by the state of consciousness, drugs and variety of
study with previous few studies indicates that BERA can environmental factors [4].
be used as a useful non invasive tool to determine auditory The BERA changes in response to hyperbilirubinemia
functions in the neonate especially changes of early bili- include loss of one or more peaks of waves I–V, raised
rubin toxicity. threshold, increase in latency of waves I, III or V or
increased inter peak interval [4].
It is worth mentioning that BERA can detect subclinical
bilirubin encephalopathy even before the appearance of
A. Soni (&) any signs or symptoms of kernicterus.
Department of ENT, GSVM Medical College, Room No 100, PG The present study was undertaken to evaluate the effect
Girls Hostel, Kanpur, India of hyperbilirubinemia in term newborns on BERA and
e-mail: annanyasoni1986@gmail.com
change, if any, in BERA after therapy.
S. K. Kanaujia
S. Kaushik The acute changes seen in BERA can be reversed by
Department of ENT, GSVM Medical College, Kanpur, India early biliru bin lowering interventions, thereby explaining

123
Indian J Otolaryngol Head Neck Surg (July–Sept 2016) 68(3):334–338 335

the transient nature of bilirubin encephalopathy. But per- Hospital, with total serum bilirubin requiring intervention
sistent elevation of bilirubin can cause neuronal degener- (using the American Academy of Pediatrics guidelines) [2]
ation and thereby persistent changes on BERA. were included in the study as cases and thirty normal term
Some studies have found correlation between the bili- AGA neonates with uneventful peri-natal period and a
rubin level and BERA changes, whereas some have dis- maximum measured serum bilirubin \12 mg/dl in case of
proved it. term baby were included as controls after obtaining
Thus serial BERA can be used as a tool to detect neuro informed consent.
developmental delay secondary to neonatal
hyperbilirubinemia. Exclusion Criteria

Neural Generators of the BERA [5] • Neonates born with birth asphyxia with poor APGAR
score
BERA is generated by the auditory nerve and subsequent • Intrauterine infections
structures within the auditory brainstem pathways. Infor- • Sepsis
mation regarding the origin of individual wave components • Meningitis/encephalitis
of BERA was provided by Moller and Janetta. • Drug like amino glycoside administration
Wave I It is the representation from the compound • Craniofacial malformation
action potential in the distal portion of cranial nerve VIII. • Conjugated hyperbilirubinemia
The response is believed to originate from afferent activity • External and middle ear pathology
of cranial nerve VIII fibres as they leave cochlea and enter • Kernicterus
the internal auditory canal.
Wave II It is generated by the proximal VIII nerve as it
Mode of Collection of Data
enters the brainstem.
Wave III Generated mainly in the cochlear nucleus
All neonates presenting with icterus to the NICU of Hos-
(second order neuron).
pital were subjected to total bilirubin estimation. Patient
Wave IV It arises from pontine third order neuron.
was subjected to ENT and pediatric examination prior to
Mostly located in superior olivary nucleus, but additional
test. If the neonate was awake, it was sedated by 20 mg/kg
contributions may come from cochlear nucleus and nucleus
of triclofos orally.
of lateral lemniscus.
Wave V Generation of wave V reflects activity of mul-
tiple anatomic auditory structures. Sharp positive peak of Preparation of Patient
wave V arises mainly from the lateral lemniscus, following
slow negative wave represents dendritic potential in the Skin was prepared with abrasive skin preparatory paste,
inferior colliculus. which is chloride and acetone free, electrolyte gel was
Wave V is the component analysed most often in the applied which is non staining, non irritant to skin.
clinical application of the BERA.
Wave VI and VII These waves appear to be generated in Electrode Placement
the inferior colliculus, perhaps in the medial geniculate body.
Silver electrode were used and applied in following fashion
according to double channel using four electrode montage
Materials and Methods system.
Vertex—?ve
Methodology
Left mastoid—-ve
Right mastoid—-ve
This study was conducted in the Department of ENT and
Forehead—neutral
PEDIATRICS. The study period was between December
2011–October 2013 after approval from institutional ethi-
cal committee and informed consent from guardians. Machine Setting

Inclusion Criteria Click acoustic stimuli with a click rate of 20–40/s, rare-
faction in polarity was presented by insert ear phone to
Thirty consecutive term appropriate for gestational age each ear at an intensity from 90 to 30 dBnHL. Time win-
(AGA) neonates presenting to the NICU of GSVM dow was 15 ms; with a filter setting of 30–3,000 Hz. The

123
336 Indian J Otolaryngol Head Neck Surg (July–Sept 2016) 68(3):334–338

Table 1 Distribution of cases in different bilirubin range • Inter peak intervals of I–III, III–V and I–V
Maximum measured bilirubin in mg/dl Case (n = 30) Neonates were treated for hyperbilirubinemia according
Group I 15–20 20 (66.6 %)
to the standard treatment protocol (using the American
Academy of Paediatrics guidelines) [2]. Repeat BERA was
Group II 20–25 5 (16.6 %)
done in all cases after therapy at the time of discharge and
Group III 25–30 5 (16.6 %)
after a follow-up period of 3 months.
The mean age of babies among the cases was 8 days and
presence of wave V at the Intensity of 30 dBnHL was
among the controls was 7 days in our study (Tables 1, 2).
taken as the normal threshold.
Latency of different waves and intervals decreased
BERA measures considered for diagnosis were
(returned towards normal) after therapy. When serum bil-
• Loss of one or more peaks of I–V at 90 dBnHL irubin levels came down, the abnormalities on initial
• Raised threshold BAER reverted back to normal in most cases. p value was
• Absolute latencies of wave I, III, V peaks significant for wave I, III, V and interpeak latency I–III and

Table 2 Number of cases with different BERA changes at peak levels of bilirubin
BERA changes Cases
Group I Group II Group III

Absent wave forms 0 (0 %) 0 (0 %) 4 (13.3 %)

Raised threshold 1 (3.33 %) 4 (13.3 % 5 (16.6 %
Prolonged latencies I 0 (0 %) 1 (3.33 % 2 (6.66 %)
Prolonged latencies III 0 (0 %) 1 (3.33 % 3 (10 %)
Prolonged latencies V 0 (0 %) 2 (6.66 % 2 (6.66 %)
Prolonged inter peak interval I–III 1 (3.33 %) 2 (6.66 % 3 (10 %)
Prolonged inter peak interval III–V 0 (0 %) 0 (0 %) 0 (0 %)
Prolonged inter peak Interval I–V 0 (0 %) 1 (3.33 % 3 (3.33 %)

Table 3 Comparison of latencies of I, III, V & inter peak latencies of cases at peak level of bilirubin, at the time of discharge and follow-up at
90 Dbnhl
Waves At peak level (A) in At discharge (B) in At followup (C) in p value A p value B p value A
msc mean ± S.D. msc mean ± S.D. msc mean ± S.D. versus B versus C versus C

I 2.06 ± 0.29 1.76 ± 0.01 1.58 ± 0.01 0.0001 0.0001 0.0001

III 4.55 ± 0.35 4.41 ± 0.20 4.16 ± 0.27 0.08 0.26 0.0001
V 6.79 ± 0.36 6.66 ± 0.29 6.34 ± 0.22 0.12 0.0001 0.0001
I–III 2.48 ± 0.31 2.65 ± 0.20 2.60 ± 0.35 0.011 0.09 0.16
III–V 2.23 ± 0.27 2.24 ± 0.18 2.18 ± 0.05 0.98 0.19 0.33
I–V 4.71 ± 0.37 4.89 ± 0.29 4.76 ± 0.22 0.05 0.003 0.53

Table 4 Comparison of BERA changes at peak level of bilirubin at the time of discharge and follow-up
BERA changes At peak level of bilirubin (no. of cases) At discharge (no. of cases) At follow up (no. of cases)

Absent wave forms 4 (13.3 %) 3 (10 %) 1 (3.3 %)

Raised threshold 10 (33.3 %) 8 (26.6 %) 1 (3.3 %)
Prolonged latencies I 3 (10 %) 0 0
Prolonged latencies III 4 (13.3 %) 2 (6.6 %) 2 (6.6 %)
Prolonged latencies V 4 (13.3 %) 2 (6.6 %) 2 (6.6 %)
Prolonged interpeak interval I–III 6 (20 %) 1 (3.3 %) 0
Prolonged interpeak interval III–V 0 0 0
Prolonged interpeak interval I–V 4 (13.3 %) 0 0

123
Indian J Otolaryngol Head Neck Surg (July–Sept 2016) 68(3):334–338 337

Table 5 Correlation of
Maximum measured No. of Cases with % of cases showing
bilirubin level with BERA
bilirubin in mg/dl cases BERA changes BERA changes
changes
15–20 20 1 5
20–25 5 4 80
25–30 5 5 100

I–V when latency at peak level was compared with that at study was raised threshold seen in 10 cases (33.3 %).
discharge and follow-up (Table 3). BERA changes persisted in 8 (26.6 %) cases at the time of
discharge of which in 3 (10 %) cases, changes persisted
even at follow up after 3 months. The precise blood level
Result above which indirect reacting or free bilirubin will be toxic
for an individual infant is unpredictable [10]. In our study,
In our study out of the 30 cases 10 (33.3 %) cases were it was found that all the 5 cases with bilirubin [25 mg/dl
found to have BERA changes in the form of absent wave had BERA changes and out of these 5, 3 cases had per-
forms, raised threshold, prolonged latencies or prolonged sistent BERA changes at follow up thus there was signif-
inter peak latencies (Tables 4, 5). icant correlation between BERA changes and maximum
The commonest abnormality observed in our study was measured bilirubin indicating that higher the level of bili-
raised threshold seen in 10 (33.3 %) cases which was rubin, higher the risk of toxicity to auditory pathway.
comparable with other studies. However, this study need further validation by con-
The other commonest abnormalities noted was prolonged ducting it on a large sample size.
latencies (13.3 %) and absent wave forms (13.3 %) which Studies by Sharma et al., Agrawal et al., Deorari et al.,
were also comparable with other studies. Prolonged laten- Gupta et al. [4, 6–8], also found statistically significant
cies of waves and inter peak interval indicated prolongation correlation of BERA changes with Serum Bilirubin
of nerve conduction at auditory nerve and brain stem level. [25 mg. Indicating that higher the level of bilirubin there
Among the 10 cases with raised threshold it was found is increased risk of bilirubin toxicity but with active
that 8 (26.6 %) cases had raised threshold even at discharge intervention, it is possible to reverse the changes.
but on follow up only 1 (3.3 %) case had persistently raised Correlation of the findings of this study with previous
threshold. Of the 3 cases with prolonged latencies of wave I few studies indicates that BERA can be used as a useful
on follow up all were found to have improved latencies, Of non invasive tool to determine auditory functions in the
the 4 (13.3 %) cases with prolonged latencies III and V, 2 neonate especially changes of early bilirubin toxicity.
(6.66 %) had persistent changes at discharge, as well as at
follow up whereas in 2 cases it was found that latencies Acknowledgments I wish to express my deep regards to head of the
department of pediatrics Dr. Y. K. Rao for his support, and college
improved. Similarly it was found that of the 6 cases with ethical committee to provide me clearance for the study.
prolonged I–III and 4 cases with prolonged I–V when
followed up all had normal latencies. Conflict of interest There is no conflict of interest in the prepara-
All cases were reviewed with a follow up BERA at the tion of this manuscript.
time of discharge and after a period of 3 months in our study
and in other studies by Sharma et al., Gupta et al. and
Agrawal et al. Whereas Deorari et al. followed up the cases References
till 1 year. Bhandari et al. did not follow up the cases [4, 6–9].
1. Madan A, James R et al (2005) Neonatal hyperbilirubinemia. In:
Out of the 10 cases with significant BERA changes, it Taeusch HW, Ballard RA, Gleason CA (eds) Avery’s diseases of
was found that 8 cases had persisted abnormalities at the the newborn, 8th edn. Saunders, Philadelphia, pp 1236–1240
time of discharge and 3 cases were found to have persistent 2. American Academy of Pediatrics (1994) Practice parameter:
management of hyperbilirubinemia in the healthy term newborn.
abnormalities at the time of follow up after 3 months.
Pediatrics 94:558–562
In our study, it was observed that there was statistically 3. Maisels JM (1994) Jaundice. In: Avery GB, Fletcher MA, Mac-
significant correlation (p value \ 0.005) between increas- Donald MG (eds) Neonatology, pathophysiology and manage-
ing bilirubin level and BERA changes. ment of the newborn, 4th edn. JB Lippincott Co, Philadelphia,
pp 630–725
4. Agrawal VK, Shukla R et al (1998) Brainstem auditory evoked
Conclusion response in newborns with hyperbilirubinemia. Indian Pediatr
35:513–518
5. Arnold SA (2007) The auditory brainstem response. In: Roeser
10 (33.3 %) out of 30 cases in our study had BERA RJ, Valente M, Hosford H (eds) Audiology diagnosis, 2nd edn.
changes.The most common BERA change noted in our Thieme, New York, pp 426–441

123
338 Indian J Otolaryngol Head Neck Surg (July–Sept 2016) 68(3):334–338

6. Sharma P, Chhangani NP et al (2006) Brainstem evoked response abnormality in brainstem auditory evoked responses. Indian Pe-
audiometry in neonates with hyperbilirubinemia. Indian J Pediatr diatr 31:915–921
73(5):413–416 9. Bhandari V, Narang A, Mann SBS et al (1993) Brainstem electric
7. Gupta AK, Hans R et al (1990) Auditory brainstem responses in response audiometry in neonates with hyperbilirubinemia. Indian
neonates with hyperbilirubinemia. Indian J Pediatr 57:705–711 J Pediatr 60:409–413
8. Deorari AK, Singh M et al (1994) One year outcome of babies 10. Singh M (2004) Care of the newborn, 6th edn. Sagar Publica-
with severe neonatal hyperbilirubinemia and reversible tions, New Delhi, pp 239–260

123