Identification of neonatal hearing impairment: Auditory brain stem responses in the perinatal period

Objectives: 1) To describe the auditory brain stem response (ABR) measureme nt system and optimized methods used for study of newborn hearing screening . 2) To determine how recording and infant factors related to the screening , using well-defined, specific ABR outcome measures. Design: Seven thousand one hundred seventy-nine infants, 4478 from the neon atal intensive care unit (NICU) and the remaining from the well-baby nurser y, were evaluated with an automated ABR protocol in each ear. Two channel r ecordings were obtained (vertex to mastoid or channel A and vertex to nape of neck or channel B) in response to click stimuli of 30 and 69 dB nHL in a ll infants as well as 50 dB nHL in infants who did not meet criteria for re sponse at 30 dB, Criteria for response included F-SP greater than or equal to3.1 and a tester-judgment of response. Criteria could be met in the first or repeat test with a maximum of 6144 accepted sweeps per test. Results: More than 99% of infants could complete the ABR protocol, More tha n 90% of NICU and well-baby nursery infants "passed" given the strict crite ria for response, whereas 86% of those with high risk factors met criterion for ABR response detection, The number of infants who did not meet ABR res ponse criteria in one or both ears was systematically related to stimulus l evel with the largest group not meeting criteria at 30 dB followed by 50 an d 69 dB nHL, Meeting criteria on the ABR was positively correlated with the amplitude of wave V, with low noise and low electrode impedance. Factors t hat predicted how many sweeps would be needed to reach criterion F-SP inclu ded noise level of the test site, state of the baby (for example, quiet sle ep versus crying), recording noise, electrode impedance and response latenc y, Channel A (vertex to mastoid) reached criterion more often than channel B (vertex to nape of neck) due to higher noise in channel B. Average total test time for 30 dB nHL screening in both ears was under 8 minutes. Well ba bies with risk factors took slightly longer to evaluate than other groups w ith this automated ABR procedure and have higher noise levels. Conclusions: ABR implemented with an automated detection algorithm using a 30 dB nHL click stimulus is reliable technique for rapid assessment of audi tory status in newborns. Factors other than hearing loss that influenced th e test result include infant state, electrode location and impedance, testi ng site, and infant risk status, Even so, ABRs were reliably recorded in th e vast majority of babies under circumstances in which most babies are foun d in the perinatal period.