Audiologic
Profile of Noise-Induced Hearing Loss
J. C.
Cooper, Jr, PhD, Jeffrey
H.
Owen, MA
\s=b\ Audiologic results for 450 ears presumed to have uncomplicated noiseinduced hearing loss were analyzed to describe their typical profile and to quantify the characteristic audiometric notch. Data consisted of tonal and speech thresholds, speech discrimination scores, Bekesy tracings, and short increment sensitivity index scores, and tone decay at 4,000 hertz. In each case, the pattern of results indicated a cochlear site of lesion, although the only uniformly cochlear sign was the absence of a pattern of abnormal adaptation in Bekesy tracings. In each case, the audiogram could be approximated by a quartic equation. The mean difference between actual thresholds and those predicted by the best-fit quartic did not exceed 7.72 dB and was less than 5 dB for eight of the 11 frequencies for which comparisons were made. We propose the application of this characteristic to clinical practice and to automated hearing conservation programs.
(Arch Otolaryngol 102:148-150, 1976) view of the present concern with the hazards posed by noise, it is surprising to note that the bulk of the literature has been limited to discussion of the degree of threshold shift. Except for contributions such as that of Ward et al,1 who reported performance on tone decay, speech discrimination, and recruitment tests, no report of the full breadth of audiologic findings could be found. However, a review of the existing literature permits a statement of consensus. Noiseinduced hearing loss (NIHL) is characterized by a notched audiogram, with poorest hearing in the range of 4,000 to 6,000 hertz. Continued exposure results in worsening thresholds dur-
In
for publication Nov 17, 1975. From the Department of Physical Medicine and Rehabilitation, University of Texas Health Science Center at San Antonio. Reprint requests to Department of Physical Medicine and Rehabilitation, University of Texas Health Science Center at San Antonio, San Antonio, TX 78284 (Dr Cooper).
Accepted
a period of some 10 to 15 years, after which hearing tends to stabilize or the rate of worsening decreases. Results of an audiologic site-of-lesion battery should indicate cochlear pathological features, although the findings in a particular test may be ambiguous. The present study was initiated to derive an audiologic profile of NIHL from a single sample of patients. Additionally, tonal thresholds were subjected to mathematical
ing
analysis by computer-assisted curvefitting techniques to quantify the characteristic notchedness accepted as evidence of noise exposure.
SUBJECTS AND METHODS The records of 225
patients
were
selected, using the following criteria: (1) a complaint of hearing loss; (2) a report of vocational noise exposure; (3) the presence of a bilateral, generally symmetrical sensorineural loss, more pronounced at higher
frequencies, with an improvement in threshold by 8,000 Hz; (4) absence of any indication of functional hearing loss; and
(5) absence of any report that would suggest the involvement of a cause other than noise. The patients' ages ranged from 21 to 59 years. Mean and modal ages were 42 years; the median age was 43 years. Factors related to the duration and intensity of exposure could not be reliably extracted from the interview notes. Data for both of each patient's ears were used. Results were obtained using manual and Bekesy audiometers calibrated to International Standards Organization standards (1964). Speech circuits were calibrated to 20 dB sound pressure level. Phonographic recordings of auditory test W-22 were used to measure speech discrimination. The battery of tests included air and bone conduction thresholds,2 speech reception thresholds (SRT) by monitored live voice,3 speech discrimination measures at 40 dB sensation level or 60 dB hearing level, whichever was greater, tone decay4 at 4,000 Hz, short increment sensitivity index (SISI)5 at 4,000 Hz, and Bekesy tracings. Interrupted tracings were followed by continuous tracings, both from low to high frequencies. In all cases, the tracings
Downloaded From: http://archotol.jamanetwork.com/ by a New York University User on 05/23/2015
covered the range from 1,000 to 8,000 Hz. In some cases, the range below 1,000 Hz was not traced if manual audiometry indicated it was normal.
RESULTS
The data
were
treated in three
stages. The first involved calculation of descriptive statistics for the various test results. In the second, Pearson product moment correlations were computed between various measures that have been reported to be related. In the third, tonal thresholds were
subjected
to
curve-fitting techniques
for the purpose of abstracting a mathematical statement that would adequately describe the notched configuration typical of noise-induced hearing loss. The mean results of each test are presented in Fig 1. Values are rounded to the nearest number em-
ployed clinically. Although not shown,
bone conduction approximated air conduction in all cases. Mean values for threshold shift of continuous Bekesy tracings are recorded for the range in which all cases were tested. Standard deviations for tonal thresholds increased systematically from approximately 7 dB at 250 Hz to about 22 dB at 8,000 Hz. Speech audiometry produced standard deviations of 7 dB for SRTs and of 30% for discrimination. The standard deviation for tone decay was 7 dB; for SISI, it was 9%. The width of continuous Bekesy tracings was narrower than that for the interrupted tracings in 56.5% of the cases. Accepting the stabilization of hearing loss after some 10 to 15 years of noise exposure,6 the degree of homogeneity of results supported the initial presumption that the sample represented uncomplicated noise-induced hearing loss. Of the correlations among audiologic measures, the one between SRTs and tonal thresholds in the speech frequency range has been most widely
Hz 125
250
500
1000
2000
4000
8000
8000
ID
(O
O
O
C/3
to LU
>
>
_l
_l
ce
2 DC
LU
LU
Profile of Noise-Induced Hearing Loss
J. C.
Cooper, Jr, PhD, Jeffrey
H.
Owen, MA
\s=b\ Audiologic results for 450 ears presumed to have uncomplicated noiseinduced hearing loss were analyzed to describe their typical profile and to quantify the characteristic audiometric notch. Data consisted of tonal and speech thresholds, speech discrimination scores, Bekesy tracings, and short increment sensitivity index scores, and tone decay at 4,000 hertz. In each case, the pattern of results indicated a cochlear site of lesion, although the only uniformly cochlear sign was the absence of a pattern of abnormal adaptation in Bekesy tracings. In each case, the audiogram could be approximated by a quartic equation. The mean difference between actual thresholds and those predicted by the best-fit quartic did not exceed 7.72 dB and was less than 5 dB for eight of the 11 frequencies for which comparisons were made. We propose the application of this characteristic to clinical practice and to automated hearing conservation programs.
(Arch Otolaryngol 102:148-150, 1976) view of the present concern with the hazards posed by noise, it is surprising to note that the bulk of the literature has been limited to discussion of the degree of threshold shift. Except for contributions such as that of Ward et al,1 who reported performance on tone decay, speech discrimination, and recruitment tests, no report of the full breadth of audiologic findings could be found. However, a review of the existing literature permits a statement of consensus. Noiseinduced hearing loss (NIHL) is characterized by a notched audiogram, with poorest hearing in the range of 4,000 to 6,000 hertz. Continued exposure results in worsening thresholds dur-
In
for publication Nov 17, 1975. From the Department of Physical Medicine and Rehabilitation, University of Texas Health Science Center at San Antonio. Reprint requests to Department of Physical Medicine and Rehabilitation, University of Texas Health Science Center at San Antonio, San Antonio, TX 78284 (Dr Cooper).
Accepted
a period of some 10 to 15 years, after which hearing tends to stabilize or the rate of worsening decreases. Results of an audiologic site-of-lesion battery should indicate cochlear pathological features, although the findings in a particular test may be ambiguous. The present study was initiated to derive an audiologic profile of NIHL from a single sample of patients. Additionally, tonal thresholds were subjected to mathematical
ing
analysis by computer-assisted curvefitting techniques to quantify the characteristic notchedness accepted as evidence of noise exposure.
SUBJECTS AND METHODS The records of 225
patients
were
selected, using the following criteria: (1) a complaint of hearing loss; (2) a report of vocational noise exposure; (3) the presence of a bilateral, generally symmetrical sensorineural loss, more pronounced at higher
frequencies, with an improvement in threshold by 8,000 Hz; (4) absence of any indication of functional hearing loss; and
(5) absence of any report that would suggest the involvement of a cause other than noise. The patients' ages ranged from 21 to 59 years. Mean and modal ages were 42 years; the median age was 43 years. Factors related to the duration and intensity of exposure could not be reliably extracted from the interview notes. Data for both of each patient's ears were used. Results were obtained using manual and Bekesy audiometers calibrated to International Standards Organization standards (1964). Speech circuits were calibrated to 20 dB sound pressure level. Phonographic recordings of auditory test W-22 were used to measure speech discrimination. The battery of tests included air and bone conduction thresholds,2 speech reception thresholds (SRT) by monitored live voice,3 speech discrimination measures at 40 dB sensation level or 60 dB hearing level, whichever was greater, tone decay4 at 4,000 Hz, short increment sensitivity index (SISI)5 at 4,000 Hz, and Bekesy tracings. Interrupted tracings were followed by continuous tracings, both from low to high frequencies. In all cases, the tracings
Downloaded From: http://archotol.jamanetwork.com/ by a New York University User on 05/23/2015
covered the range from 1,000 to 8,000 Hz. In some cases, the range below 1,000 Hz was not traced if manual audiometry indicated it was normal.
RESULTS
The data
were
treated in three
stages. The first involved calculation of descriptive statistics for the various test results. In the second, Pearson product moment correlations were computed between various measures that have been reported to be related. In the third, tonal thresholds were
subjected
to
curve-fitting techniques
for the purpose of abstracting a mathematical statement that would adequately describe the notched configuration typical of noise-induced hearing loss. The mean results of each test are presented in Fig 1. Values are rounded to the nearest number em-
ployed clinically. Although not shown,
bone conduction approximated air conduction in all cases. Mean values for threshold shift of continuous Bekesy tracings are recorded for the range in which all cases were tested. Standard deviations for tonal thresholds increased systematically from approximately 7 dB at 250 Hz to about 22 dB at 8,000 Hz. Speech audiometry produced standard deviations of 7 dB for SRTs and of 30% for discrimination. The standard deviation for tone decay was 7 dB; for SISI, it was 9%. The width of continuous Bekesy tracings was narrower than that for the interrupted tracings in 56.5% of the cases. Accepting the stabilization of hearing loss after some 10 to 15 years of noise exposure,6 the degree of homogeneity of results supported the initial presumption that the sample represented uncomplicated noise-induced hearing loss. Of the correlations among audiologic measures, the one between SRTs and tonal thresholds in the speech frequency range has been most widely
Hz 125
250
500
1000
2000
4000
8000
8000
ID
(O
O
O
C/3
to LU
>
>
_l
_l
ce
2 DC
LU
LU
