A new method of measuring distortion product otoacoustic emissions using multiple tone pairs: study of human adults.

Abstract

To increase the speed of a distortion product otoacoustic emissions (DPOE) test of cochlear function by employing a new multiple-tone-pair method and to validate the method. We used a new method of employing multiple tone pairs in measuring DPOEs proposed by Zurek and Rabinowitz and implemented in a modified Grason-Stadler Model GSI-60. We investigated the applicability of the multiple-pair method by comparing the 2f1-f2 DPOEs obtained with a 3-pair method with the conventional 1-pair results in human adults; f1 and f2 represent two frequencies of each tone pair, f1 < f2, f2/f1 = 1.2. We used two sets of 3-pair stimuli: 1) f2 at 1.5, 3, and 6 kHz; and 2) f2 at 2, 4, and 8 kHz. The 1-pair stimuli had f2 at each of the above six frequencies. The primary tone levels were L1 = 65 and L2 = 50 dB SPL re 20 microPa. We obtained DPOEs from 98 ears with normal hearing and 94 ears with sensorineural hearing loss in 103 human adults. We found: 1) that the DPOE levels obtained with the 3-pair method were strongly correlated with those of the 1-pair method, with correlation coefficients of 0.83 to 0.96; 2) that the mean DPOE level versus frequency functions for the normal ears were similar between the 3-pair and 1-pair methods; a 2-way analysis of variance indicated no significant difference between the two methods; 3) that the mean difference between DPOE levels obtained with the 3-pair and 1-pair methods was small (less than 1.3 dB in 11 of the 12 conditions examined, i.e., six frequencies each for normal ears and hearing-impaired ears) although the differences were significant in 6 of the 12 conditions as determined by the paired t-test; and 4) that the DPOE test performances of distinguishing normal from impaired ears were similar between the 3-pair and 1-pair tests; areas under the receiver operating characteristic curves were not significantly different between the two tests for four of the six frequencies tested as determined by a statistical procedure of Hanley and McNeil; for the remaining two frequencies where the differences were significant, each of the two tests was better than the other at one frequency. The 3-pair DPOE method yielded results generally in good agreement with those of the conventional 1-pair DPOE method. To our knowledge, this is the first systematic study of a multiple-pair DPOE method. Because a 3-pair method can be conducted approximately two to three times as fast as a 1-pair method, the former is advantageous, particularly for an auditory screening test.