Auditory evoked potentials in children at neonatal risk for hypoacusis1


Saúl Garza Morales,2 Adrián Poblano,2 Alicia Robledo Galván,3 and Luis Alberto Fernández Carrocera4



ABSTRACT Brainstem auditory evoked potentials provide a simple, noninvasive method of evaluating hearing function and have been widely used for early detection of hypoacusis in children. Between April 1992 and May 1994, a study was done of 400 Mexican children who presented at least one neonatal risk factor for hearing impairment. The average age of the children studied was 6.6 months and their average gestational age at birth was 35.1 weeks. Just over half of the children had been treated with amikacin. The study found 1427 risk factors (about 3.5 per child), the most common ones being exposure to ototoxic substances, hyperbilirubinemia, and birthweight <1500 g. In 27% of the children, peripheral auditory changes were found, and 13% did not respond to auditory stimuli. Low birthweight and young gestational age at birth, high serum concentration of bilirubin, sepsis, subependymal or intraventricular hemorrhage, mechanical ventilation, and exposure to ototoxic substances were significantly associated with the presence of severe or profound hypoacusis.



The auditory evoked potentials of the brain stem afford a simple, noninvasive way of evaluating the hearing function. For this reason, a test of these potentials is very often used for early detection of hypoacusis and neural conduction irregularities in the auditory pathways of children (1). The technique does not require the cooperation of the patient and is easily reproducible (2, 3). Initial auditory responses in neonates appear in the 26th and 27th weeks of pregnancy (4, 5). Progressively over time, these follow a "maturation pattern" in which interwave intervals and latencies decrease and the amplitudes of the brainstem auditory evoked potentials increase (6).

The usefulness of this test for detecting auditory irregularities in newborns has been demonstrated in cases of prematurity (7), bacterial meningitis (8), hyperbilirubinemia (9), neonatal asphyxia (10), and prenatal exposure to lead (11). If risk factors for neonatal auditory irregularities are present, the test is deemed indispensable (12).

The purpose of the study reported here was to describe the type and frequency of irregularities in brainstem auditory evoked potentials within a population of 400 Mexican children with at least one neonatal risk factor for hypoacusis (Table 1).


v2n4a2t1.gif (8012 bytes)



Between April 1992 and May 1994, a study was conducted of the evoked potentials of 800 ears in 400 children who met the following eligibility criteria: having been born at the National Institute of Perinatology in the Federal District, residing in the Valley of Mexico (the metropolitan area of Mexico City), having at least one neonatal risk factor for hypoacusis (12), and participating by agreement in the Project to Detect Neuropsychologic Irregularities of Perinatal Origin (Proyecto de Detección de Alteraciones Neuropsicológicas de Origen Perinatal—PDANDOP). The study group included some children born before April 1992 who were participating in the PDANDOP, had not previously been studied using the evoked potentials test, and satisfied the above eligibility criteria.

All of the children showing abnormal results in the evoked potentials test were enrolled in a monitoring program. In that program, the tests and evaluations were repeated at the conclusion of a three-month period; an assessment of language development was performed; and after the child’s third birthday confirmatory hypoacusis tests were conducted using tonal audiometry and speech audiometry. The analysis reported here deals only with the results of the initial evoked potentials testing.

The potentials studied were evoked during physiologic sleep by means of monaural stimulation with clicks of alternating polarity. The stimulation was first applied during a 2-minute period of adaptation preceding the recording that progressively increased in intensity by 30 dB increments until it reached the audible level of 80 dB (audible level, sound pressure of 20 micropascals), at which point averaging was initiated. Each stimulus lasted for 100 microseconds, and the stimuli were repeated 11 times per second.

The clicks were released through TDH-49 earphones in an isolated room especially designed for this purpose. The ear not being tested was masked with white sound 30 dB below the intensity of the stimulus. Cerebral electrical activity was recorded with gold disk electrodes placed at derivations Cz (+), A1, and A2, after cleansing the area with acetone-alcohol and applying conductive gel. The study ear was negative and the subject’s other ear was ground (13).

The signal coming from the electrodes was kept below 4 kilo-ohms and was used for computer input. The band-pass filters were set between 300 and 3000 Hz in order to reduce the registering of other biologic signals (electrocardiogram and electromyogram). The time of analysis subsequent to stimulation was 10 milliseconds.

A total of 2 024 stimuli were averaged, and this process was repeated at least once in order to ensure reproducibility of the response. After making the recording, averages were obtained, the intensity of the stimulus being reduced 20 dB at a time until no response was obtained, and then subsequently raised by 10 dB increments until a clear and reproducible visualization of only the V wave (the electrophysiologic threshold) was obtained. A cursor was used to measure the latency of waves I, III, and V. The latency of the interwave intervals was obtained automatically by means of a computer program.

The results of the study were classified in accord with the criteria of Stockard et al. (14) (Table 2). In the case of premature children, the corrected postconceptual or gestational age (postnatal age in weeks plus gestational age at birth) was taken into account.


v2n4a2t2.gif (9396 bytes)


Arithmetic means, standard deviations, and value ranges were used to conduct the statistical analysis. The means were compared using Student’s t test (after verifying that the distributions of the variables were normal and the variances homogeneous), while proportions were verified using the chi-square test. The level of statistical significance selected was 0.05.



The arithmetic mean of the study children’s age was 6.6 months (range: 1 day to 6 years), with those under 6 months old accounting for 84% of the total. The mean gestational age at birth was 35.1 weeks, with one standard deviation (SD) being ±3.5 weeks and the values ranging from 24 to 42 weeks. Approximately 67% of the study children were born before 37 weeks of gestation. The children’s mean birthweight was 2 034 g (SD = 870 g; range = 700–4 575 g), with most children (234) having birthweights under 2 000 g. The mean maximum concentration of indirect serum bilirubin found during admission of the 265 children evaluated for this factor was 13.6 mg/dL (SD = 5.1 mg/dL; range = 1–30 mg/dL). Because neonatal septicemia was suspected, 231 (58%) of the children were given amikacin endovenously, 97 (24%) were given furosemide, and two (0.5%) were given gentamycin.

A total of 1 427 risk factors were identified (about 3.5 factors per child). In 114 (28%) of the children, only one risk factor was documented, while in 51 (13%) two such factors were identified, and in the rest (59%) three or more were detected (Table 3).


v2n4a2t3.gif (16927 bytes)


Auditory abnormalities were detected in 45% of the ears studied. Of these, peripheral abnormalities accounted for over half (27.5% of all the ears tested), and absence of all components (in 13.4% of all the ears tested) was very common (Table 4, Figures 1 and 2).


v2n4a2t4.gif (6241 bytes)



v2n4a2f1.gif (20705 bytes)



v2n4a2f2.gif (21721 bytes)


Regarding the various groups at risk of hypoacusis, the highest percentages of children with abnormalities were found in the groups with subependymal or intraventricular hemorrhage (79%), bacterial meningitis (76%), jaundice with blood transfusion (69%), and neonatal sepsis (68%) (Table 5).


v2n4a2t5.gif (8651 bytes)


In addition, the existence of an inverse relationship was affirmed between gestational age at birth and birthweight, on the one hand, and test results indicating abnormal auditory evoked potentials of the brain stem. By way of example, Figure 3 shows the relationship between gestational age at birth and the frequency of auditory irregularities.


v2n4a2f3.gif (5614 bytes)


In order to analyze risk factors associated with bilateral absence of all components, children with this test result (designated as Group I, n = 48) were compared to the remaining 352 participants (designated Group II). A significantly higher number of risk factors per child was detected in Group I than in Group II (c2 = 46.3; P < 0.001). In addition, significant differences were detected between the two groups with regard to birthweight (t = -3.94; P < 0.0001), gestational age at birth (t = –4.66; P < 0.0001), maximum concentration of serum bilirubin (t = –3.25; P < 0.001), use of amikacin (c2 = 41.90; P < 0.001), use of furosemide (c2 = 18.80; P < 0.01), bacterial meningitis (c2 = 31.22; P < 0.001), subependymal or intraventricular hemorrhage (c2 = 27.66; P < 0.001), and use of mechanical ventilation (c2 = 26.09; P < 0.001) (Table 6).


v2n4a2t6.gif (19371 bytes)



Children experiencing hearing loss during the initial months of life and whose auditory thresholds are above 65 dB often exhibit defective language development. Successful rehabilitation is directly dependent on early detection and treatment (4). The large number of children selected for this study reflects the high frequency of children with risk factors for auditory disorders. Many of these children (in the case of the present study, 90% of the study population) are discharged from specialized neonatal therapy facilities and could benefit from hypoacusis screening prior to reaching 1 year of age.

Most of the patients studied were born prior to the 37th week of gestation with a birthweight of less than 2 000 g and required administration of potentially ototoxic drugs (amikacin, furosemide). This helps account for the fact that abnormalities were detected in close to half of the cases tested; the causes documented include cochlear disturbance (in children exposed to furosemide and amikacin) (15, 16), tubal dysfunction (in those receiving mechanical ventilation) (17), and disturbances of the cochlear nuclei (as a consequence of hyperbilirubinemia and hypoxia) (17–19). Also noteworthy is the high frequency of cases in which abnormalities were diagnosed in children who developed subependymal or intraventricular hemorrhage, bacterial meningitis, jaundice, and septicemia—a circumstance accounted for by the effect of these developments on the acoustic nerve (20), cochlear hemorrhage (21), and exposure to other risk factors (mechanical ventilation, aminoglycosides, etc.).

The inverse relationship found between the frequency of auditory disorders and gestational age/birthweight leads us to assume, in the first place, that immature auditory neurologic pathways are particularly susceptible to such disorders. It is also possible, however, that auditory sequelae are more frequently associated with the increased survival of low birthweight neonates exposed to multiple potentially ototoxic factors (22).

An absence of potentials recorded at intensities greater than 80 dB above the audible level indicates acute auditory disturbance (14). The risk factors statistically associated with this type of abnormality make it possible to construct a high-risk profile for neonates. These risk factors are as follows: low birthweight, low gestational age at birth, acute jaundice, having been given amikacin or furosemide, having undergone mechanical ventilation, subependymal or intraventricular hemorrhage, and bacterial meningitis (23).

The statistical association found between a high number of risk factors and the presence of profound hypoacusis suggests that the above-listed risk factors may have a synergistic effect, especially in children of low birthweight and low gestational age at birth. In such children, complications are extremely frequent, and the children tend to remain in the hospital for extended periods. The risk factors identified are observed more and more frequently in children discharged from neonatology wards. Hence, these patients constitute a group at high risk of exhibiting auditory disorders. The social and economic problems posed by children with hypoacusis, care for which requires both technology and personnel specializing in rehabilitation (psychologists, teachers, language specialists), may be important in the future.



 1. Picton TW, Taylor MJ, Durieux-Smith A. Brainstem auditory evoked potentials in pediatrics. In: Aminoff MJ, ed. Electrodiagnosis in clinical neurology. New York: Churchill Livingstone; 1992:537–569.        

 2. Jewett D, Williston JS. Auditory evoked far-fields arranged from the scalp of humans. Brain 1971;94:681.        

 3. Dehkharghani F. Applications of electroencephalography and evoked potentials studies in the neonatal period. In: Sarnat HB, ed. Topics in neonatal neurology. Orlando: Grune and Stratton; 1984:257–288.        

 4. Kennedy CR. The assessment of hearing and brainstem function. In: Eyre JA. The neurophysiological examination of the newborn infant. London: Mac Keith Press; 1992:79–92.        

 5. Aicardi J. Disorders of hearing and vestibular function. In: Aicardi J, ed. Diseases of the nervous system in childhood. London: Mac Keith Press; 1992:1076–1092.        

 6. Stockard-Pope JE, Werner SS, Bickford RG. Evoked potentials. In: Stockard-Pope JE, Werner SS, Bickford RG. Atlas of neonatal electroencephalography. New York: Raven Press; 1992:53–92.        

 7. Smyth V, Scott J, Tudehope D. Auditory brainstem evoked response inter-peak latencies in very low birthweight infants. Int J Pediatr Otorhinolaryngol 1988;16:69–76.        

 8. Cohen BA, Schenk VA, Sweeney DB. Meningitis-related hearing loss evaluated with evoked potentials. Pediatr Neurol 1988;4:18–21.        

 9. Nakamura H, Takada S, Shimabuku R, Matsuo M, Matsuo T, Negishi H. Auditory nerve and brainstem responses in newborn infants with hyperbilirubinemia. Pediatrics 1985;75:703–708.        

10. Poblano A, Mendiola BH, Valdez CH, et al. Potenciales provocados auditivos del tallo cerebral en recién nacidos de bajo y alto riesgo. Bol Med Hosp Infant Mex 1993;50: 551–556.        

11. Rothenberg SJ, Poblano A, Garza-Morales S. Prenatal and perinatal low level lead exposure alters brainstem auditory evoked responses in infants. Neurotoxicology 1994;15:695–699.        

12. Bergman I, Hirsch RP, Fria TJ, Shapiro SM, Holzman I, Painter MJ. Cause of hearing loss in the high-risk premature infant. J Pediatr 1985;106:95–101.        

13. Nuwer MR, Aminoff M, Goodin D, Matsuoka S, Mauguière F, Starr A, et al. IFCN recommended standards for brainstem auditory evoked potential: report of an IFCN committee. Electroenceph Clin Neurophysiol 1994;91:12–17.        

14. Stockard JE, Stockard JJ, Kleinberg F, Westmoreland BF. Prognostic value of brainstem auditory evoked potentials in neonates. Arch Neurol 1983;40:360–365.        

15. Finitzio-Hieber T. Auditory brainstem response in assessment of infants treated with aminoglycoside antibiotics. In: Lerner SA, Matz GJ, Hawkins JE Jr, eds. Aminoglycoside ototoxicity. Boston: Little, Brown; 1981:269–280.        

16. Brown DR, Watcho JF, Sabo D. Neonatal sensorineural hearing loss associated with furosemide: a case-control study. Dev Med Child Neurol 1991;33:816–823.        

17. Streletz LJ, Graziani LJ, Branca PA, Desai HJ, Travis SF, Mikaelian DO. Brainstem auditory evoked potentials in full-term and preterm newborns with hyperbilirubinemia and hypoxemia. Neuropediatrics 1986;17:66–71.        

18. Salamy A, Eldrege L, Tooley HW. Neonatal status and hearing loss in high-risk infants. J Pediatr 1989;114:847–852.        

19. Poblano A, Garza-Morales S, Ibarra-Puig J. Utilidad de los potenciales provocados auditivos del tallo cerebral en la evaluación del recién nacido. Bol Med Hosp Infant Mex 1995;52:262–270.        

20. Jiang ZD, Liu XY, Wu YY, Zheng MS, Liu HC. Long-term impairments of brain and auditory functions of children recovered from purulent meningitis. Dev Med Child Neurol 1990;32:473–480.        

21. Spector GJ, Pettit WJ, Davis G, Strauss M, Rauchbach E. Fetal respiratory distress causing CNS and inner ear hemorrhage. Laryngoscope 1978;88:764–784.        

22. McCormick MC. ¿Se ha incrementado la prevalencia de lactantes discapacitados al mejorar la supervivencia del lactante con peso muy bajo al nacer? Clin Perinatol 1993;1:285–302. (Spanish translation by Editorial Interamericana.)        

23. Fetal trimethadione effects. In: Jones KL, ed. Smith’s recognizable patterns of human malformation, 4th ed. Philadelphia: W.B. Saunders; 1988: 500–501.        



Manuscript received on 25 October 1995. Revised version accepted for publication on 6 May 1996.




Potenciales provocados auditivos en niños con riesgo neonatal de hipoacusia

Los potenciales provocados auditivos del tallo cerebral son un método sencillo y no invasor de evaluación de la función auditiva, que se utiliza ampliamente en niños para detectar tempranamente hipoacusia. Entre abril de 1992 y mayo de 1994, se estudiaron 400 niños mexicanos que presentaban, al menos, un factor de riesgo neonatal de hipoacusia. La media de la edad de los niños estudiados fue 6,6 meses y la media de la edad gestacional al nacer, 35,1 semanas. El 51% de ellos fueron tratados con amikacina. Se registraron 1 427 factores de riesgo (3,5 por niño), entre los que predominaron la exposición a ototóxicos, la hiperbilirrubinemia y el peso al nacer <1 500 g. En 27% se encontraron alteraciones auditivas de tipo periférico y en 13%, ausencia de respuesta a estímulos auditivos. El bajo peso y la menor edad gestacional al nacer, la concentración máxima de bilirrubina en el suero, la presencia de sepsis, la hemorragia subependimaria o intraventricular, la ventilación mecánica y la exposición a ototóxicos se asociaron significativamente con la presencia de hipoacusia grave o profunda.



1    A Spanish version of this article has previously been published in this journal (Vol. 1, No. 2, 1997, pp. 119–124) under the title "Potenciales provocados auditivos en niños con riesgo neonatal de hipoacusia."

2    National Institute of Perinatology, Ministry of Health, Mexico City, Mexico. Mailing address: Instituto Nacional de Perinatología, Departamento de Neurología Perinatal, Secretaría de Salud, Montes Urales 800, Loma de Virreyes, Delegación Miguel Hidalgo, CP 11000, México DF, Mexico.

3    20th of November Medical Center, Institute of Social Security and Services for State Workers, Mexico City, Mexico.

4    National Institute of Perinatology, Ministry of Health, Mexico City, Mexico.

Organización Panamericana de la Salud Washington - Washington - United States