The Effects of Aging on Distortion-Product Otoacoustic Emissions in Adults with Normal Hearing

Abstract

The focus of this study was to determine whether deterioration in cochlear function, as evaluated by distortion-product otoacoustic emission (DPOAE), exists before the elevation of audiometric thresholds occurs during the course of aging. In previous research, variability in normal audiometric thresholds likely contributed to the aging effect on OAE data. Therefore, in selecting subjects, we applied the stringent criterion in pure-tone thresholds (PTT) to limit dispersion among normal-hearing thresholds. We evaluated 331 subjects (136 men and 195 women) of a population-based sample of 2259 adults aged 40 to 82 yr who took part in the Longitudinal Study of Aging. We chose subjects according to the audiometric criterion that thresholds at any of five frequencies, namely 500, 1000, 2000, 4000, and 8000 Hz, did not exceed 15 dB HL. The mean age of our subjects was 48.3 +/- 7.4 yr (range, 41 to 72 yr) in men and 49.6 +/- 7.6 yr (range, 41 to 80 yr) in women. In a univariate analysis, analysis of variance was performed on DPOAE amplitudes and noise estimates at 22 test frequencies, as well as on the PTT. Age groups (40s, 50s, 60s, and above) were considered separately for men and women, without adjustment for any confounding variables. In a multivariate approach, general linear model analyses were performed to focus attention on the impact of age as a continuous variable, and on the influence of PTT on DPOAE levels. The multivariate analysis was conducted separately for men and women. DPOAE amplitudes at nine test frequencies were set as objective variables. Age (continuous variable), PTT at the corresponding test frequency, and interaction between age and PTT at the corresponding test frequency were evaluated as explanatory variables with adjustment for static admittance, history of ear disease (yes = 1), and history of occupational noise exposure (yes = 1). Of the 22 test frequencies, we found a statistically significant difference in DPOAE amplitudes among age groups at four test frequencies in men, ranging from 4761 to 6165 Hz, and at all but the 3088 Hz test frequency in women. Despite the stringent audiometric inclusion criterion, statistically significant differences in the mean PTT among the age groups were observed at 4000 Hz in men and at all five tested frequencies in women. Multivariate analyses demonstrated a significant negative effect of age on DPOAE levels at 1086 Hz f2 frequency in men and at the 1184, 2002, 2185, 4004, and 4358 Hz f2 frequencies in women. Regarding PTT, neither main nor interactive effect on DPOAE amplitude was statistically significant at any of nine test frequencies. The goodness-of-fit of the model, in terms of R2, ranged from 0.05 to 0.11 in men and from 0.11 to 0.18 in women. The present analyses substantiated the hypothesis DPOAEs deteriorate with age independently of hearing sensitivity. The aging effect on DPOAE measures was observed more in women than in men. We conclude that DPOAE measurements in audiometrical normal-hearing elderly may provide early indications of cochlear damage because of aging.