- Email: [email protected]
Auditory function in presbycusis: peripheral vs. central changes
Experimental Gerontology 38 (2003) 87–94 www.elsevier.com/locate/expgero
Auditory function in presbycusis: peripheral vs. central changes Jana Mazelova´, Jiri Popelar, Josef Syka* Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Vı´denˇska´ 1083, Prague 4 142 20, Czech Republic Received 28 May 2002; accepted 7 June 2002
Abstract The hearing abilities of a group of 30 elderly (67– 93 yr of age) subjects were compared with those of a group of 30 young (19 –27 yr of age) normal hearing volunteers with the aim of characterizing the changes in the peripheral and central parts of the auditory system. In elderly subjects the pure-tone thresholds were typically represented by a gradually sloping curve with a significantly greater decline in men than in women at frequencies of 3 and 4 kHz. In spite of pure tone threshold elevation in the elderly, the difference limen for intensity at 1 and 3 kHz were not significantly smaller than in the young subjects. The incidence and levels of spontaneous, transient and distortion product otoacoustic emissions were low, which would suggest the involvement of outer hair cell pathology. Also, contralateral suppression was less marked in elderly than in young subjects. Speech audiometry in the elderly revealed serious difficulties in understanding speech. Deteriorated temporal resolution, as demonstrated by increased gap detection thresholds, correlated significantly with increased speech recognition thresholds. The results support the view that presbycusis represents a combination of deteriorated function of the auditory periphery with deteriorated function of the central auditory system. q 2002 Elsevier Science Inc. All rights reserved. Keywords: Presbycusis; Pure tone audiometry; Otoacoustic emissions; Speech audiometry; Intensity difference limen; Gap detection
1. Introduction Increased age is often accompanied by presbycusis, which is characterized by the deterioration of many parameters of hearing function. There is a general consensus that presbycusis is the result of various types of physiological degeneration plus the accumulated effects of noise exposure, medical disorders and their treatment, as well as hereditary susceptibility (CHABA, 1988). The effects of lifetime noise exposure as well as medical treatments on auditory function are not easy to evaluate, and they have to be estimated from knowledge obtained from experimental models (Syka, 1989). Four predominant pathological types of presbycusis have been described: sensory, strial, neural and cochlear conductive presbycusis (Schuknecht and Gacek, 1993). Schuknecht and Gacek (1993) conclude that abrupt hightone loss signals sensory presbycusis, a flat threshold pattern is indicative of strial presbycusis, and a loss of word discrimination is characteristic of neural presbycusis. Cochlear conductive presbycusis is then characterized by * Corresponding author. Tel.: þ420-2-4106-2700; fax: þ 420-2-41062787. E-mail address: [email protected] (J. Syka).
a gradually decreasing linear distribution pattern on the audiometric scale without a pathologic correlate. It is evident that many individual cases of presbycusis do not fit into any one of these types, but rather show a combination of pathological changes; others show no specific changes and are considered as indeterminate presbycusis. Several studies have shown that presbycusis is a common disorder in the population over 60 yr of age (Brant and Fozard, 1990; Gates et al., 1990; Gates and Cooper, 1991; Takeda et al., 1992). Gradual loss of high frequency sensitivity in presbycusis eventually encroaches upon the range necessary for speech perception (from 250 Hz to 4 kHz) (Strickland et al., 1994) and results in a decreased quality of communication in the elderly. Several studies have been devoted to an evaluation of the audiologic profile in presbycusis, including an assessment of the cognitive abilities of the elderly (Van Rooij et al., 1989; Van Rooij and Plomp, 1990, 1992; Arlinger, 1991; Jerger et al., 1991; Frisina and Frisina, 1997). Some studies concentrated on impoverished temporal resolution (Moore et al., 1992; Gordon-Salant and Fitzgibbons, 1993; Snell and Frisina, 2000) and intensity discrimination (Abel et al., 1990) in the elderly, which might contribute to the deterioration in speech perception. Numerous studies have dealt with
0531-5565/03/$ - see front matter q 2002 Elsevier Science Inc. All rights reserved. PII: S 0 5 3 1 - 5 5 6 5 ( 0 2 ) 0 0 1 5 5 - 9
88
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
otoacoustic emissions (OAEs) and have demonstrated a lower incidence of transient OAEs (TEOAEs) in the elderly (Bonfils et al., 1988; Bertoli and Probst, 1997; Quaranta et al., 2001). TEOAE levels have been reported to decrease with age and/or hearing loss (Collet et al., 1990; Stover and Norton, 1993), with the decrease being accompanied by alterations in contralateral suppression (CS) (Castor et al., 1994). Other studies reported similar changes in DPOAEs in the aged population (Lonsbury-Martin et al., 1991; Dorn et al., 1998; Oeken et al., 2000). In summary, studies of age-related changes in the processing of speech and complex auditory signals have revealed the importance of the peripheral auditory status and have implicated non-peripherally-based auditory processing alterations and decreased temporal processing efficiency in the elderly. The aim of this study was to use several audiological methods to obtain a more complete picture of hearing abilities of the elderly and to differentiate between the deterioration of peripheral and central auditory function. Of special interest was a determination of the influence of audiometric threshold shifts and temporal resolution on speech perception in the elderly. The results obtained by testing elderly subjects were compared with the data obtained by testing a group of normal hearing young volunteers.
2. Methods 2.1. Subjects Participants of this study were 30 aged volunteers with presbycusis (17 men and 13 women) between the ages of 67 – 93 yr (mean age ¼ 75.7 ^ 4.9 yr) and 30 young volunteers (15 men and 15 women) between 19 and 27 yr (mean age ¼ 23.1 ^ 2.1 yr) with no head or neck pathology. Members of both groups had negative histories of otorhinolaryngological surgery and normal otoscopic findings. 2.2. Procedures To assess the hearing abilities of both volunteer groups, the following procedures were used: pure tone audiometry, speech audiometry, difference limen for intensity (DLI), recordings of OAEs, and testing of gap detection abilities. All investigations were performed monaurally, both ears of each subject were tested successively. Pure tone audiometry was performed using a clinical audiometer (Orbiter 922, Version 2, calibrated by Czech Institute of Metrology according to the Czech State Norm, European Norm of the International Standard Organization No. 389-1 and 398-5) with signals being delivered monaurally via headphones (Sennheiser HDA 200). Hearing thresholds at frequencies from 125 Hz to 8 kHz were
measured at standard frequencies and, in addition, at 3 and 6 kHz. In speech audiometry the speech recognition threshold (SRT) and maximum discrimination level (the lowest word presentation level at which the score was 100%) were determined using a standard CD recording of Czech word audiometry according to Seeman (Seeman, 1960). Czech word audiometry is based on the presentation of 10 groups of 10 phonetically balanced one, two and three syllable words corresponding to the functional spectra of the Czech language. The words were presented through a clinical audiometer (Orbiter 922, Version 2) using Sennheiser HDA 200 headphones. DLI was determined by the DLI test provided by the Orbiter 922 clinical audiometer. The tested frequencies were 1 and 3 kHz; the basic test tone level was 40 dB above individual hearing threshold; the steps of intensity modulation used were 5, 4, 3, 2, 1, 0.8, 0.6 and 0.4 dB. OAEs were measured with an ILO 96 Otoacoustic Analyzer system (Otodynamics, Ltd) with a B-type probe, controlled with the ILO 96 software Version 5. The probe was sealed into the volunteer’s external auditory meatus with the help of a foam rubber earplug and carefully positioned so as to avoid resonance peaks in the frequency spectrum of the stimulus. The stimulus for the spontaneous OAEs (SOAEs) and transient-evoked OAEs (TEOAEs) was the conventional, broadband non-linear click (0.5 – 6.0 kHz), with the stimulus gain adjusted to 81 dB peak. The stimuli for eliciting distortion product OAEs (DPOAEs) at 2f1 2 f2 were two primary tones ( f2/f1 ¼ 1.21; with L1 ¼ 65 and L2 ¼ 55 dB SPL). The frequency range of testing was for f2 ¼ 1.0– 6.3 kHz, at four points per octave. CS was evaluated by comparing TEOAE levels recorded under normal conditions and during contralateral stimulation with continuous broadband noise (BBN) at 62 dB SPL presented through an AKG earphone. The noise was switched on about 5 s before the testing sequence started. In the gap detection test, gaps were shaped with 1 ms cosine-squared rise-fall envelopes and placed into continuous white noise (generated by VEB Robotron Messelectronics, Otto Scho¨n Dresden Electronics No. 03-004 noise generator; 80 dB SPL, 78.5 dB SLA) in a series of five gaps, spaced by 150 ms intervals. Each series was triggered by pressing the starting button. The onset of the series varied from 0.5 to 5 s after pressing the button. One-third of the trials were ‘catch trials’, with no gaps. The volunteers scored by pressing the response button within 2 s after the onset of a gap series; scoring later or during catch trials was classified as a false alarm. The threshold of gap detection (GDT) was estimated by a method of limits: gap size gradually decreased from 30 ms in 5, 1 and 0.5 ms steps until the examined person first failed to respond. Then gap size increased again by 5 ms and gradually decreased in 1 and 0.5 ms steps until the thresholds of three succeeding series were within 1 ms. Performance for each gap duration was calculated using the formula of Swets (Swets, 1964).
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
The gap detection threshold was defined as the duration of a gap corresponding to a 0.5 hit rate corrected for false alarms. The test was controlled with custom-made software. For statistical analysis of the measured data, linear regression, unpaired two-tailed t-tests, paired two-tailed ttests and stepwise regression algorithm were used. For the purposes of computation, the data from the two ears of each subject were evaluated separately for each of the peripheral and central tests performed. Averaged data are presented in the form of mean ^ standard error of the mean (S.E.M.).
3. Results 3.1. Pure tone audiometry Average pure tone thresholds (PTT) in young and elderly subjects are shown in Fig. 1. The PTT of the young control group were within normal limits (# 20 dB HL) throughout all the tested frequencies. In the elderly subjects, audiometric thresholds at low frequencies (125, 250, 500 and 1000 Hz) were increased by 0– 40 dB and average values amounted to 19.1, 15.4, 14.7 and 19.0 dB, respectively. At frequencies above 1 kHz the threshold increased on average by 10 dB per octave to 3 kHz and further, almost linearly, by 8 dB per octave up to 63.7 dB at 8 kHz. In 47 ears, hearing thresholds were within normal limits at lower frequencies (up to 1 kHz); in 13 ears there was a flat elevation of 25 – 40 dB at these frequencies. Comparing the audiometric thresholds in elderly men and women, significantly better thresholds were found in women at 3 and 4 kHz ( p , 0.0003 and 0.0004, respectively). At higher frequencies, (6 and 8 kHz) hearing thresholds were not measurable using the tone intensities provided by the audiometer in men in two cases at 6 kHz and in six cases at 8 kHz, whereas in women only in one case at 8 kHz. For the computation of average values at these frequencies, the thresholds were
Fig. 1. Average pure tone audiograms of the elderly group and of young volunteers at frequencies 125–8000 Hz (error bars represent S.E.M.).
89
estimated to be 5 dB above the upper intensity limit of the audiometer. Comparison of audiometric thresholds in left and right ears showed no significant differences with a high degree of correlation between both ears (r . 0.7). 3.2. Otoacoustic emissions SOAEs were detected in 53 % of ears in the younger group, but only in three ears (5%) in the elderly group (in ears with relatively well preserved hearing). In young volunteers more than one SOAE per ear were frequently observed. The frequency of recorded SOAE ranged between 1 and 3 kHz, and the level was between 2 20 and 0 dB SPL (2 13.6 ^ 1.1 dB). In five cases SOAEs above 5 kHz were recorded; their levels ranged from 2 15 to 2 8 dB SPL. In the elderly, in two ears, the emission frequencies were 0.8 and 1.3 kHz and the level approximately 2 15 dB SPL; the remaining one ear was very exceptional, with a frequency of 5.5 kHz and the level 2 10 dB (in this ear, the greatest hearing loss was observed at 8 kHz but did not exceed 25 dB). TEOAEs were present (wave reproducibility $ 60%) in 97% in the young volunteers, but only in 55% of ears in the elderly group. In the elderly with a pure tone average that included 1 and 2 kHz (PTA1 – 2 kHz)) of $ 30 dB HL, TEOAEs were present in 69% of ears, but surprisingly they were still present in 29% of the elderly with a PTA1 – 2 kHz . 30 dB HL. The TEOAE overall levels (Fig. 2) were significantly higher in the young (11.11 ^ 0.61 dB) than in the group of elderly subjects (5.61 ^ 0.70 dB; p , 0.0001). CS (the difference between TEOAE levels recorded under normal conditions and during contralateral stimulation with continuous 62 dB SPL BBN)
Fig. 2. Average values of total TEOAE levels in the elderly and young groups under control conditions (dB SPL), during contralateral presentation of BBN (62 dB SPL), and for CS. Both TEOAE levels and CS are statistically different in the two groups (control TEOAEs: p , 0.0001; TEOAEs during contralateral stimulation: p , 0.0001; CS: p , 0.006).
90
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
was significantly different between both groups (1.17 ^ 0.10 dB in the young, 0.66 ^ 0.17 dB in the elderly; p , 0.0062) (Fig. 2). A comparison of TEOAE levels in left and right ears showed no significant differences and a high degree of correlation ðr ¼ 0:64Þ: As shown in Fig. 3, TEOAE frequency analysis at four points per octave showed that the emission levels in the young were reliably above background noise level in the frequency range 595– 4800 Hz, whereas in the elderly, they were recorded reliably only at lower frequencies (595 – 2000 Hz). According to Hall and Mueller (1997), TEOAE recordings should most reliably map hearing abilities in the frequency range 1 – 2 kHz. The distribution of the TEOAE overall levels as a function of a pure tone average at 1 and 2 kHz (PTA1 – 2 kHz)) is shown in Fig. 4. TEOAE amplitudes in young volunteers (empty circles) ranged from 0.8 to 20.8 dB, whereas TEOAE amplitudes in elderly subjects (full squares) did not reach as high values as in some young subjects, with a maximal value of 15.3 dB. The figure demonstrates that there is no relationship between TEOAE amplitude and PTA1 – 2 kHz in any tested group of subjects. The empty squares represent TEOAE amplitudes measured in elderly subjects with a wave reproducibility less than 60%. (TEOAE measured with wave reproducibility less than 60% are thought to be non-reproducible results). In 11 cases TEOAE were not possible to record. The data document that in elderly subjects with similar PTA1 – 2 kHz, reproducible as well as non-reproducible TEOAE can be recorded. DPOAE levels were, as shown in Fig. 5, well above background noise level in almost all young volunteers from 1 to 6.3 kHz, reaching their maximum for f2 frequencies at 4.8 and 5.7 kHz (amounting to 13.2 ^ 0.7 and 14.4 ^ 0.9 dB, respectively). In the elderly group average DPOAE amplitudes significantly exceeded background noise level (þ 2 SD) only at f2 frequencies 1700 –2400 Hz ( p , 0.05). The DPOAE levels were, however, significantly smaller in the elderly than those obtained from young
Fig. 3. Average frequency-specific TEOAE levels at (1/4) octave frequency bands from 595 to 4000 Hz for young and old subjects.
Fig. 4. The distribution of the TEOAE overall levels as a function of a pure tone average at 1 and 2 kHz (PTA1 – 2 kHz)) in the young, in the elderly with positive TEOAEs and in elderly subjects with TEOAE wave reproducibility less than 60%.
people at corresponding frequencies ( p , 0.0001). In individual ears the DPOAE levels were above background noise level only at several isolated frequencies. In five ears in the elderly (8%), DPOAE levels exceeded background noise at all evaluated f2 frequencies. In young volunteers, the presence of DPOAE levels above background noise level was more frequent. In 100% of the ears DPOAE levels were recorded reliably at least in seven frequency bands; in 77% ears DPOAE levels exceeded background noise at all the evaluated f2 frequencies. 3.3. Speech audiometry The average SRT, i.e. the lowest level at which a speech signal is intelligible enough to be recognized 50% of the time, was 15.8 ^ 0.3 dB HL in the young volunteer group, whereas in the elderly group it increased to 40.4 ^ 1.2 dB
Fig. 5. Average values of 2f1 2 f2 DPOAEs in young and old subjects. f2 increases in (1/4) octave steps from 1.0 to 6.3 kHz.
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
91
was no significant difference in DLI between the two study groups in spite of the elevated hearing thresholds in the elderly. Linear regression analysis did not reveal any significant correlation between PTT and DLI at any frequency tested. 3.5. Temporal resolution—gap detection
Fig. 6. Results of the Czech speech audiometry test in young (left curve) and old (right curve) subjects.
The ability to detect a gap in BBN (80 dB SPL, 78.5 dB SL) was within the normal range in young volunteers (2.78 ^ 0.04 ms). In the elderly the GDTs were significantly worse (6.73 ^ 0.16 ms) (Fig. 7B). The difference between the groups estimated by an unpaired t-test was highly significant ( p , 0.0001). GDT in the elderly group correlated with hearing thresholds at frequencies ranging from 1 to 6 kHz ðr 2 ¼ 0:26 2 0:50Þ; the strongest correlation was observed for 2 kHz ( p , 0.0001). 3.6. The relationships between results of individual methods
HL. Fig. 6 shows the average results (speech scores as a function of word presentation levels in dB HL) for young and old subjects. Sigmoids fitted to both data sets, which run almost parallel to each other, are shifted by 24.6 dB at the 50% discrimination level in the elderly with respect to young subjects (average audiometric thresholds PTA0.5 – 2 kHz are shifted by 17 dB).
In the young group DLI values varied from 1 to 2 dB (1.86 ^ 0.06) at 1 kHz, and from 0.8 to 2.0 dB (1.79 ^ 0.08) at 3 kHz. Values of DLI in the elderly group varied from 0.6 to 3.0 dB (1.73 ^ 0.09) at 1 kHz and from 0.4 to 4 dB (1.64 ^ 0.10) at 3 kHz (Fig. 7A). There
TPOAE vs. DPOAE. The comparison of TEOAE and DPOAE values in the same subject did not reveal any significant correlations at individual frequencies both in young and old groups. Pure-tone average vs. SRT. SRT in the young differed from the PTT0.5 – 2 kHz by 10.7 ^ 0.9 dB (ranging from 0 to 22 dB). In the elderly group this difference was significantly higher ( p , 0.0001), the average difference being 19.2 ^ 0.9 dB (ranging from 0 to 32 dB). As shown in Fig. 8, a significant correlation between PTT0.5 – 2 kHz and SRT was observed in the elderly group (r 2 ¼ 0:40; p , 0.0001), whereas in the young no correlation was found, probably due to small differences in pure-tone audiograms.
Fig. 7. Average intensity difference limen at 1 and 3 kHz (^S.E.M.): (A) and average gap detection thresholds (^S.E.M.); (B) in young and old subjects.
Fig. 8. Linear correlation of average audiometric thresholds at 0.5 –2 kHz and SRTs for young and old subjects. A significant correlation was observed (r2 ¼ 0:40; p , 0.0001) only in the elderly group.
3.4. Intensity discrimination
92
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
Fig. 9. The relationship between gap detection thresholds and speech recognition scores (linear correlation in the elderly group: r2 ¼ 0:23 p , 0.001).
Gap detection vs. SRT. In young subjects, as shown in Fig. 9, no significant correlation was found between GDT and SRT. A significant correlation was observed in the elderly ( p , 0.0016, r 2 ¼ 0:16). Gap detection thresholds significantly correlated with PTT0.5 – 2 kHz ( p , 0.0001, r 2 ¼ 0:23). These results indicate that temporal discrimination might also deteriorate together with worsening speech resolution in elderly subjects. Stepwise regression algorithm revealed that the variable, that most influences SRT in the elderly is audiometric thresholds at 1 kHz and neighboring frequencies ðr 2 ¼ 0:36Þ; less significant is the contribution of PTT at 0.25 kHz ðr 2 ¼ 0:43Þ; and the least important variable is gap detection—the linear combination of these variables accounts for 47% of the variance in SRT ðr 2 ¼ 0:47Þ:
4. Discussion The results of our study indicate that in the elderly, the deterioration of auditory receptors that manifests itself as a decreasing occurrence of OAEs combined with increased PTTs (especially at higher frequencies) is accompanied by pathological changes in the central auditory system, as demonstrated by a deterioration of speech recognition and temporal resolution. The investigation of hearing function in our elderly subjects with pure tone audiometry demonstrated that in most cases, typical gradually increasing thresholds at higher frequencies (amounting to 70 dB HL at 8 kHz) were present, usually described in the literature as the sensory type of presbycusis (Schuknecht and Gacek, 1993). In 13 elderly subjects included in this study, a combination of flat elevation at lower frequencies (up to 1 kHz) and gradual
threshold increase at higher frequencies was observed, which would suggest the participation of metabolic (strial) presbycusis (Gates et al., 2002). Previous studies have demonstrated convincingly that hearing sensitivity at higher frequencies declines faster with age in men than in women, whereas men have more sensitive hearing than women at frequencies below 1 kHz (Jerger et al., 1993; Pearson, 1995). In the present study, better thresholds were observed in women at 3 and 4 kHz; however, no significant differences were found at frequencies below 1 kHz. At 6 and 8 kHz no differences between men and women were observed, probably due to the impossibility of accurately estimating the auditory thresholds, which exceeded, mainly in men, the intensity range of the audiometer. It is a common finding that OAEs deteriorate with aging. In our study SOAEs were quite frequent in the group of young volunteers, whereas their occurrence in the elderly was quite exceptional. The frequency of SOAEs in the elderly subjects, if present, was lower than that recorded in the young ears, with one exception. This is in agreement with the finding of Kohler and Fritze (1992), who described a frequency shift of SOAEs towards lower frequencies with aging. A deterioration in the function of the outer hair cells with aging was also evident from the recording of TEOAEs. TOAEs were present in the present study in nearly all (97%) the normal hearing young volunteers; however, they were detected in the elderly subjects only in 55% of the ears. Similar results were previously found in studies investigating the effects of aging on OAEs (Bonfils et al., 1988; Collet et al., 1990; Stover and Norton, 1993; Bertoli and Probst, 1997; Matthews et al., 2001). Matthews et al. (2001) described a decline of TEOAE levels with age, but they identified hearing level as a much more significant factor influencing TEOAEs than aging per se. Bertoli and Probst (1997) tested elderly people (60 –97 yr) and found TEOAEs only in 60% of ears with a PTT0.5 – 2 kHz average better than 30 dB HL. In this study, 69% of the aged volunteers with a PTT1 – 2 kHz better than 30 dB HL had measurable TEOAEs. On the other hand, measurable TEOAEs were still present in three ears in the elderly with a mean PTA hearing loss at 1– 2 kHz worse than 30 dB (29%). The TEOAE levels were significantly lower in the elderly when compared with young volunteers, but in some old ears total TEOAE levels were similar to those recorded in young volunteers (although there was a PTT difference of about 15 dB). Moreover, great TEOAE variability was observed in both groups. Taking all these facts into account, it seems that TEOAE recordings do not appear to be of great diagnostic value in people suffering from presbycusis; however, the results unequivocally speak in favor of hair cell damage in presbycusis. Significant differences in the TEOAE levels between the elderly and young volunteers were accompanied by significant differences in the degree of CS. The decline of CS in the elderly (similarly as in the paper by Castor et al. (1994)) may be explained, however, as an aging-induced dysfunction of the outer hair cells or as
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
a deterioration in the function of the olivocochlear bundle. It should be mentioned that in a recent study (Quaranta et al., 2001), the authors did not observe an effect of age on the amplitude of efferent suppression. However, the size of their sample of ears was much smaller than in our study. In the elderly group DPOAE levels were above the background noise level (mean background noise þ standard deviation) only at lower frequencies ( f2 ¼ 1.18 –2.40 kHz) where hearing was relatively well preserved. The impact of age on DPOAE levels was not significant, but this is difficult to interpret because of the relatively small number of elderly volunteers with DPOAE levels above the background noise level. The most important factor responsible for DPOAE deterioration in the elderly seems to be the degree of hearing loss, as these two entities were significantly correlated. This is in accordance with many studies that have focused on aging and DPOAEs (Lonsbury-Martin et al., 1991; Strouse et al., 1996; Dorn et al., 1998; Oeken et al., 2000). Compared to the group of young healthy volunteers, the elderly displayed a significantly greater difference between their PTT and speech audiometry scores, which would suggest the possibility that factors other than PTT might affect speech comprehension in the elderly group. Speech audiometry scores were influenced not only by the audiometric threshold, but also by a deteriorated capacity to detect short temporal changes (evaluated by a gap detection task). These results are in good agreement with the study of van Rooij and Plomp (1990), who observed that in the elderly, a large component, mainly represented by progressive high-frequency hearing loss, accounted for approximately two-thirds of the systematic variance of the speech perception tests, and a smaller component, representing a general performance decrement due to reduced mental efficiency, accounted for one-third of the systematic variance. However, other authors (Jerger et al., 1991; Humes et al., 1994) concluded that hearing loss in the elderly is the most important factor influencing their speech recognition, whereas the impact of changes in mental processing or cognitive abilities is negligible. Speech understanding in the elderly is more adversely affected in noise than in quiet (for review see CHABA (1988) and Frisina and Frisina (1997)). Since we did not use masking noise in the speech audiometry test, our results cannot contribute substantially to the explanation of this phenomenon. In the studied elderly group, intensity discrimination assessed by DLI measurements showed no clear relationship between PTT and a just noticeable intensity difference at a certain frequency. Although the hearing thresholds were mostly elevated in the elderly, the DLI was in most cases similar to that obtained in the young, i.e. recruitment phenomenon was not clearly expressed. Similar results were reported previously by Go¨tzinger et al. (1961) and Hallermann and Plath (1971).
93
In the present study gap detection testing revealed a significant deterioration of temporal resolution abilities, with GDT in the elderly being significantly worse than in young volunteers. This finding seems to be influenced, to a large extent, by increased PTTs in the elderly, which showed a significant correlation with increased GDT. In addition, the slight correlation between GDT and SRT indicates that gap detection abilities might contribute to speech recognition. The correlation of GDT with aging does not receive unequivocal support in the literature. For example, in the study of Gordon-Salant and Fitzgibbons (1993) gap duration discrimination was found to contribute to the ability to recognize reverberant speech. In contrast to this, Strouse et al. (1998), who found higher GDT (mainly expressed at lower stimulus levels) in elderly adults with clinically normal hearing (PTTs # 20 dB HL), did not observe a significant correlation between psychoacoustic and speech measures of temporal processing. Other studies (Moore et al., 1992; Lutman, 1990) postulated that temporal resolution deteriorates with hearing threshold but not with age and concluded that reduced temporal resolution does not seem to be an inevitable consequence of aging. Recently, Snell and Frisina (2000) found significantly larger mean gap thresholds in older subjects in comparison with younger subjects, and Bertoli et al. (2002) observed reduced processing of basic temporal stimulus features in elderly subjects at a preattentive level as indicated by mismatch negativity. In summary, the results of this study support the view held by many other authors (for review, see for example Willott (1991) and Willott et al. (2001)) that problems observed in the elderly result not only from deteriorated function of the inner ear, but also from processes of central origin resulting in a deterioration of speech signal recognition and probably also temporal processing of acoustical signals. It is, however, difficult to determine to what extent peripheral, i.e., inner ear, changes and central changes contribute to the declining efficiency in speech recognition in the elderly. The present results suggest, in accordance with other authors, that the evaluation of different types of OAEs is of limited diagnostic as well as prognostic value in presbycusis. OAEs are in principle very vulnerable to, and may serve as sensitive indicators for, temporary defects of inner ear function, even in such situations as a several hour exposure to leisure noise (Mazelova´ et al., 2001). Since the elderly manifest deficiencies in their ability to process information about sound frequency, intensity and space and time parameters, the results of tests indicating changes in the central parts of the auditory system are more useful. The final outcome of such investigations may be of great help not only in precise diagnosis of presbycusis, but also in the improvement of assistive hearing devices. A long-term goal is the eventual development of pharmacological treatment for presbycusis.
94
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
Acknowledgements This research was supported by grants of the Ministry of Health NK/4747-3 and NK/6454-3. Milan Jilek helped with experimental design and calibration.
References Abel, S.M., Krever, E.M., Alberti, P.W., 1990. Auditory detection, discrimination and speech processing in ageing, noise-sensitive and hearing-impaired listeners. Scand. Audiol. 19, 43–54. Arlinger, S., 1991. Audiometric profile in presbycusis. Acta Otolaryngol. Suppl. 476, 85–90. Bertoli, S., Probst, R., 1997. The role of transient-evoked otoacoustic emission testing in the evaluation of elderly persons. Ear Hear. 18, 286– 293. Bertoli, S., Smurzynski, J., Probst, R., 2002. Temporal resolution in young and elderly subjects as measured by mismatch negativity and a psychoacoustic gap detection task. Clin. Neurophysiol. 113, 396–406. Bonfils, P., Bertrand, Y., Uziel, A., 1988. Evoked otoacoustic emissions: normative data and presbycusis. Audiology 27, 27–35. Brant, L.J., Fozard, J.L., 1990. Age changes in pure-tone hearing thresholds in a longitudinal study of normal human aging. J. Acoust. Soc. Am. 88, 813– 820. Castor, X., Veuillet, E., Morgon, A., Collet, L., 1994. Influence of aging on active cochlear micromechanical properties and on the medial olivocochlear system in humans. Hear. Res. 77, 1–8. Collet, L., Moulin, A., Gartner, M., Morgon, A., 1990. Age-related changes in evokedotoacousticemissions.Ann.Otol.Rhinol.Laryngol.99,993–997. Committee on Hearing, Bioacoustics and Biomechanics (CHABA), 1988. Speech understanding and aging. J. Acoust. Soc. Am 836, 859– 883. Dorn, P.A., Piskorski, P., Keefe, D.H., Neely, S.T., Gorga, M.P., 1998. On the existence of an age/threshold/frequency interaction in distortion product otoacoustic emissions. J. Acoust. Soc. Am. 104, 964–971. Frisina, D.R., Frisina, R.D., 1997. Speech recognition in noise and presbycusis: relations to possible neural mechanisms. Hear. Res. 106, 95–104. Gates, G.A., Cooper, J.C., 1991. Incidence of hearing decline in the elderly. Acta Otolaryngol. 111, 240–248. Gates, G.A., Cooper, J.C. Jr., Kannel, W.B., Miller, N.J., 1990. Hearing in the elderly: the Framingham cohort, 1983 – 1985. Part I. Basic audiometric test results. Ear Hear. 11, 247–256. Gates, G.A., Mills, D., Nam, B., D’Agostino, R., Rubel, E.W., 2002. Effects of age on the distortion product otoacoustic emission growth functions. Hear. Res. 163, 53 –60. Gordon-Salant, S., Fitzgibbons, P.J., 1993. Temporal factors and speech recognition performance in young and elderly listeners. J. Speech Hear. Res. 36 (1276), 1285. Go¨tzinger, C.P., Proud, G.D., Dirks, D., Evanston, M.A., Embrev, J., 1961. A study of hearing in advanced age. Arch. Otol. (Chicago) 73, 662 –674. Hall, J.W., Mueller, G., 1997. Audiologists’ Desk Reference, Diagnostic Audiology, Principles, Procedures and Practices, vol. 1. Singular Publishing Group, San Diego, pp. 1–904. Hallermann, W., Plath, P., 1971. Effect of age on the discrimination ability of the ear. HNO 9, 26–32. Humes, L.E., Watson, B.U., Christensen, L.A., Cokely, C.G., Halling, D.C., Lee, L., 1994. Factors associated with individual differences in clinical measures of speech recognition among the elderly. J. Speech Hear. Res. 37, 465 –474. Jerger, J., Jerger, S., Pirozzolo, F., 1991. Correlational analysis of speech audiometric scores, hearing loss, age, and cognitive abilities in the elderly. Ear Hear. 12, 103– 109. Jerger, J., Chmiel, R., Stach, B., Spretnjak, M., 1993. Gender affects audiometric slope in presbyacusis. J. Am. Acad. Audiol. 4, 42–49.
Kohler, W., Fritze, W., 1992. A long-term observation of spontaneous otoacoustic emissions (SOAEs). Scand. Audiol. 21, 55– 58. Lonsbury-Martin, B.L., Cutler, W.M., Martin, G.K., 1991. Evidence for the influence of aging on distortion-product otoacoustic emissions in humans. J. Acoust. Soc. Am. 89, 1749–1759. Lutman, M.E., 1990. Degradations in frequency and temporal resolution with age and their impact on speech identification. Acta Otolaryngol. Suppl. 476, 120– 125. Matthews, L.J., Mills, J.H., Dubno, J.R., 2001. Transient-evoked otoacoustic emissions (TEOAE): effects of hearing loss, age, gender and noise exposure history. ARO Abstr 24, 284. Mazelova´, J., Valvoda, J., Popela´rˇ, J., Syka, J., 2001. The influence of single exposure to amplified music on hearing of young listeners. In: Henderson, D., Prasher, D., Kopke, R., Salvi, R., Hamernik, R. (Eds.), Noise Induced Hearing Loss: Basic Mechanisms, Prevention and Control, Noise Research Network Publications, London, pp. 365 –375. Moore, B.C., Peters, R.W., Glasberg, B.R., 1992. Detection of temporal gaps in sinusoids by elderly subjects with and without hearing loss. J. Acoust. Soc. Am. 92, 1923–1932. Oeken, J., Lenk, A., Bootz, F., 2000. Influence of age and presbyacusis on DPOAE. Acta Otolaryngol. 120, 396–403. Pearson, J.D., Morrell, H.Ch., Gordon-Salant, S., Brant, L.J., Metter, E.J., Klein, L.L., Fozard, J.L., 1995. Gender differences in a longitudinal study of age-associated hearing loss. J. Acoust. Soc. Am. 97, 1196–1205. Quaranta, N., Debole, S., Di Girolamo, S., 2001. Effect of ageing on otoacoustic emissions and efferent suppression in humans. Audiology 40, 308–312. Schuknecht, H.F., Gacek, M.R., 1993. Cochlear pathology in presbycusis. Ann. Otol. Rhinol. Laryngol. 102, 1 –16. Seeman, M., 1960. Czech Word Audiometry, SZN, Prague, In Czech. Snell, K.B., Frisina, D.R., 2000. Relationship among age-related differences in gap detection and word recognition. J. Acoust. Soc. Am. 107, 1615–1626. Stover, L., Norton, S.J., 1993. The effects of aging on otoacoustic emissions. J. Acoust. Soc. Am. 94, 2670– 2681. Strickland, E.A., Viemeister, N.F., Van Tasell, D.J., Preminger, J.E., 1994. Is useful speech information carried by fibers with high characteristic frequencies? J. Acoust. Soc. Am. 95, 497–501. Strouse, A., Ashmead, D.H., Ohde, R.N., Grantham, D.W., 1998. Temporal processing in the aging auditory system. J. Acoust. Soc. Am. 104, 2385–2399. Strouse, A.L., Ochs, M.T., Hall, J.W., 1996. Evidence against the influence of aging on distortion-product otoacoustic emissions. J. Am. Acad. Audiol. 7, 339–445. Swets, J.A., 1964. Signal Detection and Recognition by Human Observers, Wiley, New York. Syka, J., 1989. Experimental models of sensorineural hearing loss—effects of noise and ototoxic drugs on hearing. In: Ottoson, D., (Ed.), Progress in Sensory Physiology, Springer, Berlin, pp. 97–170. Takeda, S., Morioka, I., Miyashita, K., Okumura, A., Yoshida, Y., Matsumoto, K., 1992. Age variation in the upper limit of hearing. Appl. Physiol. 65, 403– 408. Van Rooij, J.C., Plomp, R., 1992. Auditive and cognitive factors in speech perception by elderly listeners. III. Additional data and final discussion. J. Acoust. Soc. Am. 91, 1028–1033. Van Rooij, J.C., Plomp, R., 1990. Auditive and cognitive factors in speech perception by elderly listeners. II. Multivariate analyses. J. Acoust. Soc. Am. 88, 2611–2624. Van Rooij, J.C., Plomp, R., Orlebeke, J.F., 1989. Auditive and cognitive factors in speech perception by elderly listeners. I: Development of test battery. J. Acoust. Soc. Am. 86, 1294– 1309. Willott, J.F., 1991. Aging and the Auditory System, Singular Publishing Group, San Diego. Willott, J.F., Chisolm, T.H., Lister, J.J., 2001. Modulation of presbycusis: current status and future directions. Audiol. Neurootol. 6, 231–249.
Auditory function in presbycusis: peripheral vs. central changes Jana Mazelova´, Jiri Popelar, Josef Syka* Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Vı´denˇska´ 1083, Prague 4 142 20, Czech Republic Received 28 May 2002; accepted 7 June 2002
Abstract The hearing abilities of a group of 30 elderly (67– 93 yr of age) subjects were compared with those of a group of 30 young (19 –27 yr of age) normal hearing volunteers with the aim of characterizing the changes in the peripheral and central parts of the auditory system. In elderly subjects the pure-tone thresholds were typically represented by a gradually sloping curve with a significantly greater decline in men than in women at frequencies of 3 and 4 kHz. In spite of pure tone threshold elevation in the elderly, the difference limen for intensity at 1 and 3 kHz were not significantly smaller than in the young subjects. The incidence and levels of spontaneous, transient and distortion product otoacoustic emissions were low, which would suggest the involvement of outer hair cell pathology. Also, contralateral suppression was less marked in elderly than in young subjects. Speech audiometry in the elderly revealed serious difficulties in understanding speech. Deteriorated temporal resolution, as demonstrated by increased gap detection thresholds, correlated significantly with increased speech recognition thresholds. The results support the view that presbycusis represents a combination of deteriorated function of the auditory periphery with deteriorated function of the central auditory system. q 2002 Elsevier Science Inc. All rights reserved. Keywords: Presbycusis; Pure tone audiometry; Otoacoustic emissions; Speech audiometry; Intensity difference limen; Gap detection
1. Introduction Increased age is often accompanied by presbycusis, which is characterized by the deterioration of many parameters of hearing function. There is a general consensus that presbycusis is the result of various types of physiological degeneration plus the accumulated effects of noise exposure, medical disorders and their treatment, as well as hereditary susceptibility (CHABA, 1988). The effects of lifetime noise exposure as well as medical treatments on auditory function are not easy to evaluate, and they have to be estimated from knowledge obtained from experimental models (Syka, 1989). Four predominant pathological types of presbycusis have been described: sensory, strial, neural and cochlear conductive presbycusis (Schuknecht and Gacek, 1993). Schuknecht and Gacek (1993) conclude that abrupt hightone loss signals sensory presbycusis, a flat threshold pattern is indicative of strial presbycusis, and a loss of word discrimination is characteristic of neural presbycusis. Cochlear conductive presbycusis is then characterized by * Corresponding author. Tel.: þ420-2-4106-2700; fax: þ 420-2-41062787. E-mail address: [email protected] (J. Syka).
a gradually decreasing linear distribution pattern on the audiometric scale without a pathologic correlate. It is evident that many individual cases of presbycusis do not fit into any one of these types, but rather show a combination of pathological changes; others show no specific changes and are considered as indeterminate presbycusis. Several studies have shown that presbycusis is a common disorder in the population over 60 yr of age (Brant and Fozard, 1990; Gates et al., 1990; Gates and Cooper, 1991; Takeda et al., 1992). Gradual loss of high frequency sensitivity in presbycusis eventually encroaches upon the range necessary for speech perception (from 250 Hz to 4 kHz) (Strickland et al., 1994) and results in a decreased quality of communication in the elderly. Several studies have been devoted to an evaluation of the audiologic profile in presbycusis, including an assessment of the cognitive abilities of the elderly (Van Rooij et al., 1989; Van Rooij and Plomp, 1990, 1992; Arlinger, 1991; Jerger et al., 1991; Frisina and Frisina, 1997). Some studies concentrated on impoverished temporal resolution (Moore et al., 1992; Gordon-Salant and Fitzgibbons, 1993; Snell and Frisina, 2000) and intensity discrimination (Abel et al., 1990) in the elderly, which might contribute to the deterioration in speech perception. Numerous studies have dealt with
0531-5565/03/$ - see front matter q 2002 Elsevier Science Inc. All rights reserved. PII: S 0 5 3 1 - 5 5 6 5 ( 0 2 ) 0 0 1 5 5 - 9
88
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
otoacoustic emissions (OAEs) and have demonstrated a lower incidence of transient OAEs (TEOAEs) in the elderly (Bonfils et al., 1988; Bertoli and Probst, 1997; Quaranta et al., 2001). TEOAE levels have been reported to decrease with age and/or hearing loss (Collet et al., 1990; Stover and Norton, 1993), with the decrease being accompanied by alterations in contralateral suppression (CS) (Castor et al., 1994). Other studies reported similar changes in DPOAEs in the aged population (Lonsbury-Martin et al., 1991; Dorn et al., 1998; Oeken et al., 2000). In summary, studies of age-related changes in the processing of speech and complex auditory signals have revealed the importance of the peripheral auditory status and have implicated non-peripherally-based auditory processing alterations and decreased temporal processing efficiency in the elderly. The aim of this study was to use several audiological methods to obtain a more complete picture of hearing abilities of the elderly and to differentiate between the deterioration of peripheral and central auditory function. Of special interest was a determination of the influence of audiometric threshold shifts and temporal resolution on speech perception in the elderly. The results obtained by testing elderly subjects were compared with the data obtained by testing a group of normal hearing young volunteers.
2. Methods 2.1. Subjects Participants of this study were 30 aged volunteers with presbycusis (17 men and 13 women) between the ages of 67 – 93 yr (mean age ¼ 75.7 ^ 4.9 yr) and 30 young volunteers (15 men and 15 women) between 19 and 27 yr (mean age ¼ 23.1 ^ 2.1 yr) with no head or neck pathology. Members of both groups had negative histories of otorhinolaryngological surgery and normal otoscopic findings. 2.2. Procedures To assess the hearing abilities of both volunteer groups, the following procedures were used: pure tone audiometry, speech audiometry, difference limen for intensity (DLI), recordings of OAEs, and testing of gap detection abilities. All investigations were performed monaurally, both ears of each subject were tested successively. Pure tone audiometry was performed using a clinical audiometer (Orbiter 922, Version 2, calibrated by Czech Institute of Metrology according to the Czech State Norm, European Norm of the International Standard Organization No. 389-1 and 398-5) with signals being delivered monaurally via headphones (Sennheiser HDA 200). Hearing thresholds at frequencies from 125 Hz to 8 kHz were
measured at standard frequencies and, in addition, at 3 and 6 kHz. In speech audiometry the speech recognition threshold (SRT) and maximum discrimination level (the lowest word presentation level at which the score was 100%) were determined using a standard CD recording of Czech word audiometry according to Seeman (Seeman, 1960). Czech word audiometry is based on the presentation of 10 groups of 10 phonetically balanced one, two and three syllable words corresponding to the functional spectra of the Czech language. The words were presented through a clinical audiometer (Orbiter 922, Version 2) using Sennheiser HDA 200 headphones. DLI was determined by the DLI test provided by the Orbiter 922 clinical audiometer. The tested frequencies were 1 and 3 kHz; the basic test tone level was 40 dB above individual hearing threshold; the steps of intensity modulation used were 5, 4, 3, 2, 1, 0.8, 0.6 and 0.4 dB. OAEs were measured with an ILO 96 Otoacoustic Analyzer system (Otodynamics, Ltd) with a B-type probe, controlled with the ILO 96 software Version 5. The probe was sealed into the volunteer’s external auditory meatus with the help of a foam rubber earplug and carefully positioned so as to avoid resonance peaks in the frequency spectrum of the stimulus. The stimulus for the spontaneous OAEs (SOAEs) and transient-evoked OAEs (TEOAEs) was the conventional, broadband non-linear click (0.5 – 6.0 kHz), with the stimulus gain adjusted to 81 dB peak. The stimuli for eliciting distortion product OAEs (DPOAEs) at 2f1 2 f2 were two primary tones ( f2/f1 ¼ 1.21; with L1 ¼ 65 and L2 ¼ 55 dB SPL). The frequency range of testing was for f2 ¼ 1.0– 6.3 kHz, at four points per octave. CS was evaluated by comparing TEOAE levels recorded under normal conditions and during contralateral stimulation with continuous broadband noise (BBN) at 62 dB SPL presented through an AKG earphone. The noise was switched on about 5 s before the testing sequence started. In the gap detection test, gaps were shaped with 1 ms cosine-squared rise-fall envelopes and placed into continuous white noise (generated by VEB Robotron Messelectronics, Otto Scho¨n Dresden Electronics No. 03-004 noise generator; 80 dB SPL, 78.5 dB SLA) in a series of five gaps, spaced by 150 ms intervals. Each series was triggered by pressing the starting button. The onset of the series varied from 0.5 to 5 s after pressing the button. One-third of the trials were ‘catch trials’, with no gaps. The volunteers scored by pressing the response button within 2 s after the onset of a gap series; scoring later or during catch trials was classified as a false alarm. The threshold of gap detection (GDT) was estimated by a method of limits: gap size gradually decreased from 30 ms in 5, 1 and 0.5 ms steps until the examined person first failed to respond. Then gap size increased again by 5 ms and gradually decreased in 1 and 0.5 ms steps until the thresholds of three succeeding series were within 1 ms. Performance for each gap duration was calculated using the formula of Swets (Swets, 1964).
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
The gap detection threshold was defined as the duration of a gap corresponding to a 0.5 hit rate corrected for false alarms. The test was controlled with custom-made software. For statistical analysis of the measured data, linear regression, unpaired two-tailed t-tests, paired two-tailed ttests and stepwise regression algorithm were used. For the purposes of computation, the data from the two ears of each subject were evaluated separately for each of the peripheral and central tests performed. Averaged data are presented in the form of mean ^ standard error of the mean (S.E.M.).
3. Results 3.1. Pure tone audiometry Average pure tone thresholds (PTT) in young and elderly subjects are shown in Fig. 1. The PTT of the young control group were within normal limits (# 20 dB HL) throughout all the tested frequencies. In the elderly subjects, audiometric thresholds at low frequencies (125, 250, 500 and 1000 Hz) were increased by 0– 40 dB and average values amounted to 19.1, 15.4, 14.7 and 19.0 dB, respectively. At frequencies above 1 kHz the threshold increased on average by 10 dB per octave to 3 kHz and further, almost linearly, by 8 dB per octave up to 63.7 dB at 8 kHz. In 47 ears, hearing thresholds were within normal limits at lower frequencies (up to 1 kHz); in 13 ears there was a flat elevation of 25 – 40 dB at these frequencies. Comparing the audiometric thresholds in elderly men and women, significantly better thresholds were found in women at 3 and 4 kHz ( p , 0.0003 and 0.0004, respectively). At higher frequencies, (6 and 8 kHz) hearing thresholds were not measurable using the tone intensities provided by the audiometer in men in two cases at 6 kHz and in six cases at 8 kHz, whereas in women only in one case at 8 kHz. For the computation of average values at these frequencies, the thresholds were
Fig. 1. Average pure tone audiograms of the elderly group and of young volunteers at frequencies 125–8000 Hz (error bars represent S.E.M.).
89
estimated to be 5 dB above the upper intensity limit of the audiometer. Comparison of audiometric thresholds in left and right ears showed no significant differences with a high degree of correlation between both ears (r . 0.7). 3.2. Otoacoustic emissions SOAEs were detected in 53 % of ears in the younger group, but only in three ears (5%) in the elderly group (in ears with relatively well preserved hearing). In young volunteers more than one SOAE per ear were frequently observed. The frequency of recorded SOAE ranged between 1 and 3 kHz, and the level was between 2 20 and 0 dB SPL (2 13.6 ^ 1.1 dB). In five cases SOAEs above 5 kHz were recorded; their levels ranged from 2 15 to 2 8 dB SPL. In the elderly, in two ears, the emission frequencies were 0.8 and 1.3 kHz and the level approximately 2 15 dB SPL; the remaining one ear was very exceptional, with a frequency of 5.5 kHz and the level 2 10 dB (in this ear, the greatest hearing loss was observed at 8 kHz but did not exceed 25 dB). TEOAEs were present (wave reproducibility $ 60%) in 97% in the young volunteers, but only in 55% of ears in the elderly group. In the elderly with a pure tone average that included 1 and 2 kHz (PTA1 – 2 kHz)) of $ 30 dB HL, TEOAEs were present in 69% of ears, but surprisingly they were still present in 29% of the elderly with a PTA1 – 2 kHz . 30 dB HL. The TEOAE overall levels (Fig. 2) were significantly higher in the young (11.11 ^ 0.61 dB) than in the group of elderly subjects (5.61 ^ 0.70 dB; p , 0.0001). CS (the difference between TEOAE levels recorded under normal conditions and during contralateral stimulation with continuous 62 dB SPL BBN)
Fig. 2. Average values of total TEOAE levels in the elderly and young groups under control conditions (dB SPL), during contralateral presentation of BBN (62 dB SPL), and for CS. Both TEOAE levels and CS are statistically different in the two groups (control TEOAEs: p , 0.0001; TEOAEs during contralateral stimulation: p , 0.0001; CS: p , 0.006).
90
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
was significantly different between both groups (1.17 ^ 0.10 dB in the young, 0.66 ^ 0.17 dB in the elderly; p , 0.0062) (Fig. 2). A comparison of TEOAE levels in left and right ears showed no significant differences and a high degree of correlation ðr ¼ 0:64Þ: As shown in Fig. 3, TEOAE frequency analysis at four points per octave showed that the emission levels in the young were reliably above background noise level in the frequency range 595– 4800 Hz, whereas in the elderly, they were recorded reliably only at lower frequencies (595 – 2000 Hz). According to Hall and Mueller (1997), TEOAE recordings should most reliably map hearing abilities in the frequency range 1 – 2 kHz. The distribution of the TEOAE overall levels as a function of a pure tone average at 1 and 2 kHz (PTA1 – 2 kHz)) is shown in Fig. 4. TEOAE amplitudes in young volunteers (empty circles) ranged from 0.8 to 20.8 dB, whereas TEOAE amplitudes in elderly subjects (full squares) did not reach as high values as in some young subjects, with a maximal value of 15.3 dB. The figure demonstrates that there is no relationship between TEOAE amplitude and PTA1 – 2 kHz in any tested group of subjects. The empty squares represent TEOAE amplitudes measured in elderly subjects with a wave reproducibility less than 60%. (TEOAE measured with wave reproducibility less than 60% are thought to be non-reproducible results). In 11 cases TEOAE were not possible to record. The data document that in elderly subjects with similar PTA1 – 2 kHz, reproducible as well as non-reproducible TEOAE can be recorded. DPOAE levels were, as shown in Fig. 5, well above background noise level in almost all young volunteers from 1 to 6.3 kHz, reaching their maximum for f2 frequencies at 4.8 and 5.7 kHz (amounting to 13.2 ^ 0.7 and 14.4 ^ 0.9 dB, respectively). In the elderly group average DPOAE amplitudes significantly exceeded background noise level (þ 2 SD) only at f2 frequencies 1700 –2400 Hz ( p , 0.05). The DPOAE levels were, however, significantly smaller in the elderly than those obtained from young
Fig. 3. Average frequency-specific TEOAE levels at (1/4) octave frequency bands from 595 to 4000 Hz for young and old subjects.
Fig. 4. The distribution of the TEOAE overall levels as a function of a pure tone average at 1 and 2 kHz (PTA1 – 2 kHz)) in the young, in the elderly with positive TEOAEs and in elderly subjects with TEOAE wave reproducibility less than 60%.
people at corresponding frequencies ( p , 0.0001). In individual ears the DPOAE levels were above background noise level only at several isolated frequencies. In five ears in the elderly (8%), DPOAE levels exceeded background noise at all evaluated f2 frequencies. In young volunteers, the presence of DPOAE levels above background noise level was more frequent. In 100% of the ears DPOAE levels were recorded reliably at least in seven frequency bands; in 77% ears DPOAE levels exceeded background noise at all the evaluated f2 frequencies. 3.3. Speech audiometry The average SRT, i.e. the lowest level at which a speech signal is intelligible enough to be recognized 50% of the time, was 15.8 ^ 0.3 dB HL in the young volunteer group, whereas in the elderly group it increased to 40.4 ^ 1.2 dB
Fig. 5. Average values of 2f1 2 f2 DPOAEs in young and old subjects. f2 increases in (1/4) octave steps from 1.0 to 6.3 kHz.
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
91
was no significant difference in DLI between the two study groups in spite of the elevated hearing thresholds in the elderly. Linear regression analysis did not reveal any significant correlation between PTT and DLI at any frequency tested. 3.5. Temporal resolution—gap detection
Fig. 6. Results of the Czech speech audiometry test in young (left curve) and old (right curve) subjects.
The ability to detect a gap in BBN (80 dB SPL, 78.5 dB SL) was within the normal range in young volunteers (2.78 ^ 0.04 ms). In the elderly the GDTs were significantly worse (6.73 ^ 0.16 ms) (Fig. 7B). The difference between the groups estimated by an unpaired t-test was highly significant ( p , 0.0001). GDT in the elderly group correlated with hearing thresholds at frequencies ranging from 1 to 6 kHz ðr 2 ¼ 0:26 2 0:50Þ; the strongest correlation was observed for 2 kHz ( p , 0.0001). 3.6. The relationships between results of individual methods
HL. Fig. 6 shows the average results (speech scores as a function of word presentation levels in dB HL) for young and old subjects. Sigmoids fitted to both data sets, which run almost parallel to each other, are shifted by 24.6 dB at the 50% discrimination level in the elderly with respect to young subjects (average audiometric thresholds PTA0.5 – 2 kHz are shifted by 17 dB).
In the young group DLI values varied from 1 to 2 dB (1.86 ^ 0.06) at 1 kHz, and from 0.8 to 2.0 dB (1.79 ^ 0.08) at 3 kHz. Values of DLI in the elderly group varied from 0.6 to 3.0 dB (1.73 ^ 0.09) at 1 kHz and from 0.4 to 4 dB (1.64 ^ 0.10) at 3 kHz (Fig. 7A). There
TPOAE vs. DPOAE. The comparison of TEOAE and DPOAE values in the same subject did not reveal any significant correlations at individual frequencies both in young and old groups. Pure-tone average vs. SRT. SRT in the young differed from the PTT0.5 – 2 kHz by 10.7 ^ 0.9 dB (ranging from 0 to 22 dB). In the elderly group this difference was significantly higher ( p , 0.0001), the average difference being 19.2 ^ 0.9 dB (ranging from 0 to 32 dB). As shown in Fig. 8, a significant correlation between PTT0.5 – 2 kHz and SRT was observed in the elderly group (r 2 ¼ 0:40; p , 0.0001), whereas in the young no correlation was found, probably due to small differences in pure-tone audiograms.
Fig. 7. Average intensity difference limen at 1 and 3 kHz (^S.E.M.): (A) and average gap detection thresholds (^S.E.M.); (B) in young and old subjects.
Fig. 8. Linear correlation of average audiometric thresholds at 0.5 –2 kHz and SRTs for young and old subjects. A significant correlation was observed (r2 ¼ 0:40; p , 0.0001) only in the elderly group.
3.4. Intensity discrimination
92
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
Fig. 9. The relationship between gap detection thresholds and speech recognition scores (linear correlation in the elderly group: r2 ¼ 0:23 p , 0.001).
Gap detection vs. SRT. In young subjects, as shown in Fig. 9, no significant correlation was found between GDT and SRT. A significant correlation was observed in the elderly ( p , 0.0016, r 2 ¼ 0:16). Gap detection thresholds significantly correlated with PTT0.5 – 2 kHz ( p , 0.0001, r 2 ¼ 0:23). These results indicate that temporal discrimination might also deteriorate together with worsening speech resolution in elderly subjects. Stepwise regression algorithm revealed that the variable, that most influences SRT in the elderly is audiometric thresholds at 1 kHz and neighboring frequencies ðr 2 ¼ 0:36Þ; less significant is the contribution of PTT at 0.25 kHz ðr 2 ¼ 0:43Þ; and the least important variable is gap detection—the linear combination of these variables accounts for 47% of the variance in SRT ðr 2 ¼ 0:47Þ:
4. Discussion The results of our study indicate that in the elderly, the deterioration of auditory receptors that manifests itself as a decreasing occurrence of OAEs combined with increased PTTs (especially at higher frequencies) is accompanied by pathological changes in the central auditory system, as demonstrated by a deterioration of speech recognition and temporal resolution. The investigation of hearing function in our elderly subjects with pure tone audiometry demonstrated that in most cases, typical gradually increasing thresholds at higher frequencies (amounting to 70 dB HL at 8 kHz) were present, usually described in the literature as the sensory type of presbycusis (Schuknecht and Gacek, 1993). In 13 elderly subjects included in this study, a combination of flat elevation at lower frequencies (up to 1 kHz) and gradual
threshold increase at higher frequencies was observed, which would suggest the participation of metabolic (strial) presbycusis (Gates et al., 2002). Previous studies have demonstrated convincingly that hearing sensitivity at higher frequencies declines faster with age in men than in women, whereas men have more sensitive hearing than women at frequencies below 1 kHz (Jerger et al., 1993; Pearson, 1995). In the present study, better thresholds were observed in women at 3 and 4 kHz; however, no significant differences were found at frequencies below 1 kHz. At 6 and 8 kHz no differences between men and women were observed, probably due to the impossibility of accurately estimating the auditory thresholds, which exceeded, mainly in men, the intensity range of the audiometer. It is a common finding that OAEs deteriorate with aging. In our study SOAEs were quite frequent in the group of young volunteers, whereas their occurrence in the elderly was quite exceptional. The frequency of SOAEs in the elderly subjects, if present, was lower than that recorded in the young ears, with one exception. This is in agreement with the finding of Kohler and Fritze (1992), who described a frequency shift of SOAEs towards lower frequencies with aging. A deterioration in the function of the outer hair cells with aging was also evident from the recording of TEOAEs. TOAEs were present in the present study in nearly all (97%) the normal hearing young volunteers; however, they were detected in the elderly subjects only in 55% of the ears. Similar results were previously found in studies investigating the effects of aging on OAEs (Bonfils et al., 1988; Collet et al., 1990; Stover and Norton, 1993; Bertoli and Probst, 1997; Matthews et al., 2001). Matthews et al. (2001) described a decline of TEOAE levels with age, but they identified hearing level as a much more significant factor influencing TEOAEs than aging per se. Bertoli and Probst (1997) tested elderly people (60 –97 yr) and found TEOAEs only in 60% of ears with a PTT0.5 – 2 kHz average better than 30 dB HL. In this study, 69% of the aged volunteers with a PTT1 – 2 kHz better than 30 dB HL had measurable TEOAEs. On the other hand, measurable TEOAEs were still present in three ears in the elderly with a mean PTA hearing loss at 1– 2 kHz worse than 30 dB (29%). The TEOAE levels were significantly lower in the elderly when compared with young volunteers, but in some old ears total TEOAE levels were similar to those recorded in young volunteers (although there was a PTT difference of about 15 dB). Moreover, great TEOAE variability was observed in both groups. Taking all these facts into account, it seems that TEOAE recordings do not appear to be of great diagnostic value in people suffering from presbycusis; however, the results unequivocally speak in favor of hair cell damage in presbycusis. Significant differences in the TEOAE levels between the elderly and young volunteers were accompanied by significant differences in the degree of CS. The decline of CS in the elderly (similarly as in the paper by Castor et al. (1994)) may be explained, however, as an aging-induced dysfunction of the outer hair cells or as
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
a deterioration in the function of the olivocochlear bundle. It should be mentioned that in a recent study (Quaranta et al., 2001), the authors did not observe an effect of age on the amplitude of efferent suppression. However, the size of their sample of ears was much smaller than in our study. In the elderly group DPOAE levels were above the background noise level (mean background noise þ standard deviation) only at lower frequencies ( f2 ¼ 1.18 –2.40 kHz) where hearing was relatively well preserved. The impact of age on DPOAE levels was not significant, but this is difficult to interpret because of the relatively small number of elderly volunteers with DPOAE levels above the background noise level. The most important factor responsible for DPOAE deterioration in the elderly seems to be the degree of hearing loss, as these two entities were significantly correlated. This is in accordance with many studies that have focused on aging and DPOAEs (Lonsbury-Martin et al., 1991; Strouse et al., 1996; Dorn et al., 1998; Oeken et al., 2000). Compared to the group of young healthy volunteers, the elderly displayed a significantly greater difference between their PTT and speech audiometry scores, which would suggest the possibility that factors other than PTT might affect speech comprehension in the elderly group. Speech audiometry scores were influenced not only by the audiometric threshold, but also by a deteriorated capacity to detect short temporal changes (evaluated by a gap detection task). These results are in good agreement with the study of van Rooij and Plomp (1990), who observed that in the elderly, a large component, mainly represented by progressive high-frequency hearing loss, accounted for approximately two-thirds of the systematic variance of the speech perception tests, and a smaller component, representing a general performance decrement due to reduced mental efficiency, accounted for one-third of the systematic variance. However, other authors (Jerger et al., 1991; Humes et al., 1994) concluded that hearing loss in the elderly is the most important factor influencing their speech recognition, whereas the impact of changes in mental processing or cognitive abilities is negligible. Speech understanding in the elderly is more adversely affected in noise than in quiet (for review see CHABA (1988) and Frisina and Frisina (1997)). Since we did not use masking noise in the speech audiometry test, our results cannot contribute substantially to the explanation of this phenomenon. In the studied elderly group, intensity discrimination assessed by DLI measurements showed no clear relationship between PTT and a just noticeable intensity difference at a certain frequency. Although the hearing thresholds were mostly elevated in the elderly, the DLI was in most cases similar to that obtained in the young, i.e. recruitment phenomenon was not clearly expressed. Similar results were reported previously by Go¨tzinger et al. (1961) and Hallermann and Plath (1971).
93
In the present study gap detection testing revealed a significant deterioration of temporal resolution abilities, with GDT in the elderly being significantly worse than in young volunteers. This finding seems to be influenced, to a large extent, by increased PTTs in the elderly, which showed a significant correlation with increased GDT. In addition, the slight correlation between GDT and SRT indicates that gap detection abilities might contribute to speech recognition. The correlation of GDT with aging does not receive unequivocal support in the literature. For example, in the study of Gordon-Salant and Fitzgibbons (1993) gap duration discrimination was found to contribute to the ability to recognize reverberant speech. In contrast to this, Strouse et al. (1998), who found higher GDT (mainly expressed at lower stimulus levels) in elderly adults with clinically normal hearing (PTTs # 20 dB HL), did not observe a significant correlation between psychoacoustic and speech measures of temporal processing. Other studies (Moore et al., 1992; Lutman, 1990) postulated that temporal resolution deteriorates with hearing threshold but not with age and concluded that reduced temporal resolution does not seem to be an inevitable consequence of aging. Recently, Snell and Frisina (2000) found significantly larger mean gap thresholds in older subjects in comparison with younger subjects, and Bertoli et al. (2002) observed reduced processing of basic temporal stimulus features in elderly subjects at a preattentive level as indicated by mismatch negativity. In summary, the results of this study support the view held by many other authors (for review, see for example Willott (1991) and Willott et al. (2001)) that problems observed in the elderly result not only from deteriorated function of the inner ear, but also from processes of central origin resulting in a deterioration of speech signal recognition and probably also temporal processing of acoustical signals. It is, however, difficult to determine to what extent peripheral, i.e., inner ear, changes and central changes contribute to the declining efficiency in speech recognition in the elderly. The present results suggest, in accordance with other authors, that the evaluation of different types of OAEs is of limited diagnostic as well as prognostic value in presbycusis. OAEs are in principle very vulnerable to, and may serve as sensitive indicators for, temporary defects of inner ear function, even in such situations as a several hour exposure to leisure noise (Mazelova´ et al., 2001). Since the elderly manifest deficiencies in their ability to process information about sound frequency, intensity and space and time parameters, the results of tests indicating changes in the central parts of the auditory system are more useful. The final outcome of such investigations may be of great help not only in precise diagnosis of presbycusis, but also in the improvement of assistive hearing devices. A long-term goal is the eventual development of pharmacological treatment for presbycusis.
94
J. Mazelova´ et al. / Experimental Gerontology 38 (2003) 87–94
Acknowledgements This research was supported by grants of the Ministry of Health NK/4747-3 and NK/6454-3. Milan Jilek helped with experimental design and calibration.
References Abel, S.M., Krever, E.M., Alberti, P.W., 1990. Auditory detection, discrimination and speech processing in ageing, noise-sensitive and hearing-impaired listeners. Scand. Audiol. 19, 43–54. Arlinger, S., 1991. Audiometric profile in presbycusis. Acta Otolaryngol. Suppl. 476, 85–90. Bertoli, S., Probst, R., 1997. The role of transient-evoked otoacoustic emission testing in the evaluation of elderly persons. Ear Hear. 18, 286– 293. Bertoli, S., Smurzynski, J., Probst, R., 2002. Temporal resolution in young and elderly subjects as measured by mismatch negativity and a psychoacoustic gap detection task. Clin. Neurophysiol. 113, 396–406. Bonfils, P., Bertrand, Y., Uziel, A., 1988. Evoked otoacoustic emissions: normative data and presbycusis. Audiology 27, 27–35. Brant, L.J., Fozard, J.L., 1990. Age changes in pure-tone hearing thresholds in a longitudinal study of normal human aging. J. Acoust. Soc. Am. 88, 813– 820. Castor, X., Veuillet, E., Morgon, A., Collet, L., 1994. Influence of aging on active cochlear micromechanical properties and on the medial olivocochlear system in humans. Hear. Res. 77, 1–8. Collet, L., Moulin, A., Gartner, M., Morgon, A., 1990. Age-related changes in evokedotoacousticemissions.Ann.Otol.Rhinol.Laryngol.99,993–997. Committee on Hearing, Bioacoustics and Biomechanics (CHABA), 1988. Speech understanding and aging. J. Acoust. Soc. Am 836, 859– 883. Dorn, P.A., Piskorski, P., Keefe, D.H., Neely, S.T., Gorga, M.P., 1998. On the existence of an age/threshold/frequency interaction in distortion product otoacoustic emissions. J. Acoust. Soc. Am. 104, 964–971. Frisina, D.R., Frisina, R.D., 1997. Speech recognition in noise and presbycusis: relations to possible neural mechanisms. Hear. Res. 106, 95–104. Gates, G.A., Cooper, J.C., 1991. Incidence of hearing decline in the elderly. Acta Otolaryngol. 111, 240–248. Gates, G.A., Cooper, J.C. Jr., Kannel, W.B., Miller, N.J., 1990. Hearing in the elderly: the Framingham cohort, 1983 – 1985. Part I. Basic audiometric test results. Ear Hear. 11, 247–256. Gates, G.A., Mills, D., Nam, B., D’Agostino, R., Rubel, E.W., 2002. Effects of age on the distortion product otoacoustic emission growth functions. Hear. Res. 163, 53 –60. Gordon-Salant, S., Fitzgibbons, P.J., 1993. Temporal factors and speech recognition performance in young and elderly listeners. J. Speech Hear. Res. 36 (1276), 1285. Go¨tzinger, C.P., Proud, G.D., Dirks, D., Evanston, M.A., Embrev, J., 1961. A study of hearing in advanced age. Arch. Otol. (Chicago) 73, 662 –674. Hall, J.W., Mueller, G., 1997. Audiologists’ Desk Reference, Diagnostic Audiology, Principles, Procedures and Practices, vol. 1. Singular Publishing Group, San Diego, pp. 1–904. Hallermann, W., Plath, P., 1971. Effect of age on the discrimination ability of the ear. HNO 9, 26–32. Humes, L.E., Watson, B.U., Christensen, L.A., Cokely, C.G., Halling, D.C., Lee, L., 1994. Factors associated with individual differences in clinical measures of speech recognition among the elderly. J. Speech Hear. Res. 37, 465 –474. Jerger, J., Jerger, S., Pirozzolo, F., 1991. Correlational analysis of speech audiometric scores, hearing loss, age, and cognitive abilities in the elderly. Ear Hear. 12, 103– 109. Jerger, J., Chmiel, R., Stach, B., Spretnjak, M., 1993. Gender affects audiometric slope in presbyacusis. J. Am. Acad. Audiol. 4, 42–49.
Kohler, W., Fritze, W., 1992. A long-term observation of spontaneous otoacoustic emissions (SOAEs). Scand. Audiol. 21, 55– 58. Lonsbury-Martin, B.L., Cutler, W.M., Martin, G.K., 1991. Evidence for the influence of aging on distortion-product otoacoustic emissions in humans. J. Acoust. Soc. Am. 89, 1749–1759. Lutman, M.E., 1990. Degradations in frequency and temporal resolution with age and their impact on speech identification. Acta Otolaryngol. Suppl. 476, 120– 125. Matthews, L.J., Mills, J.H., Dubno, J.R., 2001. Transient-evoked otoacoustic emissions (TEOAE): effects of hearing loss, age, gender and noise exposure history. ARO Abstr 24, 284. Mazelova´, J., Valvoda, J., Popela´rˇ, J., Syka, J., 2001. The influence of single exposure to amplified music on hearing of young listeners. In: Henderson, D., Prasher, D., Kopke, R., Salvi, R., Hamernik, R. (Eds.), Noise Induced Hearing Loss: Basic Mechanisms, Prevention and Control, Noise Research Network Publications, London, pp. 365 –375. Moore, B.C., Peters, R.W., Glasberg, B.R., 1992. Detection of temporal gaps in sinusoids by elderly subjects with and without hearing loss. J. Acoust. Soc. Am. 92, 1923–1932. Oeken, J., Lenk, A., Bootz, F., 2000. Influence of age and presbyacusis on DPOAE. Acta Otolaryngol. 120, 396–403. Pearson, J.D., Morrell, H.Ch., Gordon-Salant, S., Brant, L.J., Metter, E.J., Klein, L.L., Fozard, J.L., 1995. Gender differences in a longitudinal study of age-associated hearing loss. J. Acoust. Soc. Am. 97, 1196–1205. Quaranta, N., Debole, S., Di Girolamo, S., 2001. Effect of ageing on otoacoustic emissions and efferent suppression in humans. Audiology 40, 308–312. Schuknecht, H.F., Gacek, M.R., 1993. Cochlear pathology in presbycusis. Ann. Otol. Rhinol. Laryngol. 102, 1 –16. Seeman, M., 1960. Czech Word Audiometry, SZN, Prague, In Czech. Snell, K.B., Frisina, D.R., 2000. Relationship among age-related differences in gap detection and word recognition. J. Acoust. Soc. Am. 107, 1615–1626. Stover, L., Norton, S.J., 1993. The effects of aging on otoacoustic emissions. J. Acoust. Soc. Am. 94, 2670– 2681. Strickland, E.A., Viemeister, N.F., Van Tasell, D.J., Preminger, J.E., 1994. Is useful speech information carried by fibers with high characteristic frequencies? J. Acoust. Soc. Am. 95, 497–501. Strouse, A., Ashmead, D.H., Ohde, R.N., Grantham, D.W., 1998. Temporal processing in the aging auditory system. J. Acoust. Soc. Am. 104, 2385–2399. Strouse, A.L., Ochs, M.T., Hall, J.W., 1996. Evidence against the influence of aging on distortion-product otoacoustic emissions. J. Am. Acad. Audiol. 7, 339–445. Swets, J.A., 1964. Signal Detection and Recognition by Human Observers, Wiley, New York. Syka, J., 1989. Experimental models of sensorineural hearing loss—effects of noise and ototoxic drugs on hearing. In: Ottoson, D., (Ed.), Progress in Sensory Physiology, Springer, Berlin, pp. 97–170. Takeda, S., Morioka, I., Miyashita, K., Okumura, A., Yoshida, Y., Matsumoto, K., 1992. Age variation in the upper limit of hearing. Appl. Physiol. 65, 403– 408. Van Rooij, J.C., Plomp, R., 1992. Auditive and cognitive factors in speech perception by elderly listeners. III. Additional data and final discussion. J. Acoust. Soc. Am. 91, 1028–1033. Van Rooij, J.C., Plomp, R., 1990. Auditive and cognitive factors in speech perception by elderly listeners. II. Multivariate analyses. J. Acoust. Soc. Am. 88, 2611–2624. Van Rooij, J.C., Plomp, R., Orlebeke, J.F., 1989. Auditive and cognitive factors in speech perception by elderly listeners. I: Development of test battery. J. Acoust. Soc. Am. 86, 1294– 1309. Willott, J.F., 1991. Aging and the Auditory System, Singular Publishing Group, San Diego. Willott, J.F., Chisolm, T.H., Lister, J.J., 2001. Modulation of presbycusis: current status and future directions. Audiol. Neurootol. 6, 231–249.