J Am Acad Audiol 26:101-108 (2015)

Air Conduction, Bone Conduction, and Soft Tissue Conduction Audiograms in Normal Hearing and Simulated Hearing Losses DOI: 10.3766/jaaa.26.1.11 Cahtia Adelman*t Adi Cohenf Adi Regev-Cohent Shai Chordekari Rachel Fraenkelt Haim Sohmer§

Abstract Background: In order to differentiate between a conductive hearing loss (CHL) and a sensorineural hearing loss (SNHL) in the hearing-impaired individual, we compared thresholds to air conduction (AC) and bone conduction (BC) auditory stimulation. The presence of a gap between these thresholds (an air-bone gap) is taken as a sign of a CHL, whereas similar threshold elevations reflect an SNHL. This is based on the assumption that BC stimulation directly excites the inner ear, bypassing the middle ear. However, several of the classic mechanisms of BC stimulation such as ossicular chain inertia and the occlusion effect involve middle ear structures. An additional mode of auditory stim­ ulation, called soft tissue conduction (STC; also called nonosseous BC) has been demonstrated, in which the clinical bone vibrator elicits hearing when it is applied to soft tissue sites on the head, neck, and thorax. Purpose: The purpose of this study was to assess the relative contributions of threshold determinations to stimulation by STC, in addition to AC and osseous BC, to the differential diagnosis between a CHL and an SNHL. Research Design: Baseline auditory thresholds were determined in normal participants to AC (supraaural earphones), BC (B71 bone vibrator at the mastoid, with 5 N application force), and STC (B71 bone vibrator) to the submental area and to the submandibular triangle with 5 N application force) stimulation in response to 0.5,1.0,2.0, and 4.0 kHz tones. A CHL was then simulated in the participants by means of an ear plug. Separately, an SNHL was simulated in these participants with 30 dB effective masking. Study sample: Study sample consisted of 10 normal-hearing participants (4 males; 6 females, aged 20-30 yr). Data Collection and Analysis: AC, BC, and STC thresholds were determined in the initial normal state and in the presence of each of the simulations. Results: The earplug-induced CHL simulation led to a mean AC threshold elevation of 21-37 dB (depending on frequency), but not of BC and STC thresholds. The masking-induced SNHL led to a mean elevation of AC, BC, and STC thresholds (23-36 dB, depending on frequency). In each type of simulation, the BC threshold shift was similar to that of the STC threshold shift. Conclusions: These results, which show a similar threshold shift for STC and for BC as a result of these simulations, together with additional clinical and laboratory findings, provide evidence that BC thresholds

‘ Speech & Hearing Center, Hadassah University Hospital, Jerusalem, Israel; tDepartment of Communication Disorders, Hadassah Academic College, Jerusalem, Israel; tDepartment of Communication Disorders, Sackler Faculty of Medicine, Tel Aviv University, The Chaim Sheba Medical Center, Tel Hashomer, Israel; §Department of Medical Neurobiology (Physiology), Hebrew University-Hadassah Medical School, Jerusalem, Israel Haim Sohmer, Department of Medical Neurobiology (Physiology), The Hebrew University-Hadassah Medical School, P.0. Box 12272, Jerusalem 91120, Israel; E-mail: [email protected] The data reported in this manuscript were presented orally at the Annual Conference of the Israeli Speech, Hearing and Language Association, Tel Aviv, Israel, February 2012, as “Clinical application of soft tissue conduction; simulation of hearing losses.”

TO*!

Journal of th e American Academy o f Audiology/Volume 26, Number 1, 2015

likely represent the threshold of the nonosseous BC (STC) component of multicomponent BC at the BC stimulation site, and thereby succeed in clinical practice to contribute to the differential diagnosis. This also provides evidence that STC (nonosseous BC) stimulation at low intensities probably does not involve components of the middle ear, represents true cochlear function, and therefore can also contrib­ ute to a differential diagnosis (e.g., in situations where the clinical bone vibrator cannot be applied to the mastoid or forehead with a 5 N force, such as in severe skull fracture). Key Words: Simulated hearing loss, conductive hearing loss, sensorineural hearing loss, bone conduction, soft tissue conduction, nonosseous bone conduction, threshold Abbreviations: AC = air conduction; ANOVA = analysis of variance; ANSI = American National Standards Institute; BC = bone conduction; CHL = conductive hearing loss; EML = effective masking level; HL = hearing loss; N = Newton; NBN = narrow band noise; SD = standard deviation; SNHL = sensorineural hearing loss; STC = soft tissue conduction

INTRODUCTION linical audiometry is based on the determina­ tion of thresholds to pure tones presented by air conduction (AC) by an earphone and by a bone vibrator applied to the skin of the head at the mas­ toid or forehead (bone conduction [BC]). The thresholds are used to classify hearing as normal or impaired, and if impaired, the degrees of impairment (i.e., threshold and type of impairment (conductive hearing loss [CHL], sensorineural HL [SNHL], or mixed) are defined, with clinical corroboration from otoscopic findings and tym ­ panometry. AC and BC thresholds are compared, and the presence of an air-bone threshold gap between them is defined as the conductive component of an HL. This is based on the assumption th a t the auditory sensation elicited in response to the BC stim ulation represents cochlear function exclusively, bypassing the middle ear and its component structures, reflecting true cochlear function (Schlauch and Nelson, 2009). However, this assum ption may not be accurate in sev­ eral cases because several of the BC mechanisms require mobility of middle ear structures such as the ossicles and the oval and round windows. These classic BC mecha­ nisms, investigated using suprathreshold sound inten­ sities, require the induction of skull bone (temporal/ petrous) vibrations, leading to vibrations of the wall of the middle ear (inertia of the ossicular chain), of the cochlear shell (compression-expansion-distortion; fluid inertia), and of the external ear (occlusion effect) (Tonndorf, 1968; Stenfelt and Goode, 2005). There­ fore, it is possible th a t BC stim ulation can be affected by lesions of the bony parts of the ear, thus not truly reflecting cochlear function exclusively. On the other hand, an additional mode of auditory stim ulation has been described in which auditory sensation can be elic­ ited by applying the bone vibrator used in the clinic at the mastoid or forehead to soft tissue sites on the skin of the head and neck, and even at skin sites that are very dis­ tant from the ear and from underlying bone, such as the thorax (Kaufmann et al, 2012b). This has been called nonosseous BC (Vento and Durrant, 2009; Ito et al,

C

1 0S

2011), soft tissue conduction (STC), or body conduction (Berger et al, 2003). In the present report, the terms STC and nonosseous BC will be used interchangeably. The three modes (AC, BC, and STC) of auditory stimu­ lation interact with each other in the inner ear; their pitches can be matched, they mutually mask each other, beat with each other (Adelman et al, 2012), and mutually cancel each other (Chordekar et al, 2012a). Also, a distor­ tion product otoacoustic emission at 2f1-f2 is produced when ft is presented by AC and f2 at the eye, an STC site (Watanabe et al, 2008). Evidence exists th at the pathway from a nonosseous BC (STC) stimulation site on the skin to the inner ear at threshold intensities does not involve middle ear structures, whereas the several mechanisms of osseous BC require the induction of skull bone (temporal/ petrous bone) vibrations (Tonndorf, 1968; Stenfelt and Goode, 2005). Furthermore, auditory sensation can be elicited at STC sites without the standard application force of 5 N, with only the minimal application force of a wet wick lightly touching soft tissues on the face (Gealdor et al, 2012). Therefore, it is likely th at threshold inten­ sity STC stimulation does not lead to vibrations of the skull bone (temporal/petrous bone), which are required for BC stimulation by inducing vibrations of the mid­ dle ear ossicles, the two windows, and of the cochlea (Tonndorf, 1968; Stenfelt and Goode, 2005). In addition, recent experiments in this laboratory have shown no change in STC and BC thresholds after immobilization and discontinuity of the middle ear ossicles and fixation of the two windows (Perez et al, 2011a,b; 2014). Such immobilization, and discontinuity and fixation, eliminat­ ing middle ear and window involvement in BC mecha­ nisms, would have been expected to lead to some degree of threshold elevation if skull vibrations (temporal/petrous bone) were involved during threshold BC stim ula­ tion. Furthermore, although auditory nerve brainstemevoked responses to STC (nonosseous BC) stimuli could be elicited, they were not accompanied by laser Doppler vibrom eter-detected vibrations of inner ear bone in Psammomys obesus (sand rat) (Chordekar et al, 2013a). STC stimulation at low intensities (threshold) probably induces vibrations along a series of soft tissues having

AC, BC, and STC in Sim ulated H earing Losses/Adelman et al

acoustic impedances (defined as the product of the density of a tissue and the velocity of sound in it) similar to that of water, so that they are transmitted to the inner ear through soft tissues and aqueous media (Gealdor et al, 2012; Adelman et al, 2013). On the other hand, the vibra­ tions in the soft tissues are reflected at the boundary between soft tissues and other tissues that have very dif­ ferent acoustic impedances such as skull bone and air, similar to ultrasound diagnostic medical imaging (Baun, 2004). This mechanism is also similar to the principle behind the functions of the middle ear as a mechanical transformer between the acoustic impedance of air and that of water (Wever and Lawrence, 1954). For all of these reasons, it is likely th a t STC stimu­ lation involves a vibration transmission pathway from the STC stimulation site directly to the inner ear, with­ out involvement of the temporal/petrous bones. There­ fore, threshold intensity STC stimulation may provide an estimation of true cochlear function, and the present study was designed as an initial assessment of the pos­ sible contributions of STC to the differential diagnosis between a CHL and an SNHL, especially in situations in which the clinical bone vibrator cannot be applied to the mastoid or forehead with an application force of 5 N. The study design involved obtaining hearing thresholds in participants with normal hearing in response to stim­ ulation by AC, BC, and STC, followed by simulating a CHL, and then a SNHL, in these same participants. Also, the AC, BC, and STC thresholds were again determined in the presence of each simulation separately. The thresh­ old shifts at each of the stimulation sites, during each of the HL simulations, were then compared. A similar shift in threshold would point to a similar mechanism acting at threshold. In addition, such a study could also provide some insight into the possible mechanisms of cochlear activation during threshold BC and STC stimulation.

METHODS he participants were four men and six women, aged 20-30 yr (mean age: 25.2 yrj with normal hearing bilaterally, defined as AC thresholds at 500, 1000, 2000, and 4000 Hz of 15 dB HL or better. Testing was conducted in a sound booth. All audiometers were calibrated accord­ ing to the American National Standards Institute (ANSI) using normal procedure. Noise in the sound booth was tested with a sound level meter (Bruel & Kjaer 2250) according to ANSI S 3.1, and it was found that external noise at 1 kHz was attenuated in the booth by 46 dB, and the ambient noise in the test booth was 3-7 dB (depend­ ing on frequency) below the maximum permitted by ANSI S 3.1. After baseline (normal) AC thresholds to warble tones at the four frequencies were determined in the sound booth using the 10 dB down-5 dB-up method, AC, BC, and STC thresholds to warble tones were ob­ tained (the baseline m easurem ents were conducted

T

with earphones in place) from the ear with the better BC thresholds. On the other hand, if thresholds were identical on both sides, the test ear was chosen so that in the final study, five right and five left ears were tested. The BC stimulation site was the skin at the mastoid (i.e., over the skull bone), but there is a wide range of possible STC sites on the skin of the head, neck, and thorax, and one must choose the most convenient STC site for study. The criteria for the choice include the availability of a wide dynamic range (i.e., low thresholds) and the ease with which the same site can be accurately identified in repeated testing (test-retest).

Baseline Measurements A GSI-61 clinical audiometer driving TDH-50P supraaural earphones was used to deliver AC stimulation, and the same audiometer, driving a standard clinical B71 bone vibrator, was used to deliver BC (mastoid) stimula­ tion with a standard headband (static application force of ~5N). STC stimulation was applied to the submental area (a bone vibrator was placed under the chin; partic­ ipants were instructed to separate the upper and lower jaw so that the upper and lower teeth were not in con­ tact). STC stimulation was also applied to the subman­ dibular triangle (below the earlobe and posterior to the body of the mandible). Warble tones at 500, 1000, 2000, and 4000 Hz were presented. The STC thresholds to these tones were expressed as the dB HL settings on the GSI-61 clinical audiometer in the BC mode. For STC testing, the bone vibrator was applied by hand to the test sites with a uniform static pressure of 500 g (5 N), achieved by having the participant press down on a spring attached to the surface of the bone vibrator opposite that in contact with the skin, with a force predetermined to present 500 g. Participants were instructed to respond to auditory stimulation only, and to ignore tactile vibrations that they might feel with the 500 Hz stimulus coming from the hand-held bone vibrator. There is no tactile sensation at the higher fre­ quencies (Hyvarinen et al 1968). Because testing would be performed with supra-aural earphones during the part of the study involving the simulated losses, baseline measurements were also conducted with such earphones already in place on the ears. In order to control for the possibility that the participants might respond to AC sound produced by the bone vibrator, we also determined thresholds with the vibrator held in the air over the BC and STC sites. The thresholds on the sites were consis­ tently lower (better) than those obtained in air.

Simulated CHL A simulated CHL was induced, as in Nia and Bance (2001), by deeply inserting classic Superfit 30 Aero Co, E-A-R foam earplugs bilaterally in the external canal

103

Journal of the American Academy of Audiology/Volume 26, Number 1, 2015

(under the earphones). AC, BC, and STC (from both STC sites) thresholds were determined. Thresholds were also determined with the vibrator held in the air over the BC and STC sites, as a control.

F re q u e n c y

Sim ulated SNHL In order to both provide a stimulus and also to present noise bilaterally, we used two audiometers sim ulta­ neously: the previously mentioned GSI-61 audiometer with TDH-50P earphones and a B71 bone vibrator, and a portable A321 Clinical Twin Channel audiometer with H7A-PTL earphones. For AC threshold testing with masking, noise and signal were both presented from the GSI-61 audiom eter to the test ear by the same TDH-50P earphone, and noise to the contralateral ear was delivered by the H7A-PTL earphone driven by the portable audiometer. For BC and STC threshold testing with masking, the bone vibrator driven by the GSI-61 audiometer was used, with bilateral masking presented through the H7A-PTL earphone. The masking utilized was 30 dB EML (effective masking level) NBN (narrow band noise). The test ear was masked in order to simu­ late an SNHL, and the contralateral ear was masked so that the participant would not respond to th at ear. Once again, thresholds were also determined with the vibrator held in the air over the BC and STC sites, as a control. The duration of the entire procedure was between 1.5-2 hr. In those participants who were unable to con­ centrate sufficiently for this duration, testing was con­ ducted in two separate sessions. Interruptions were permitted as necessary to m aintain concentration. Ran­ dom retesting of some earlier thresholds was performed toward the end of the session, to assess th a t responses were reliable and consistent (test-retest). The experimental protocol was reviewed and approved by the Hadassah Academic College Institutional Ethics Committee, and participants gave their informed consent.

AC — ♦—

BC - - - - - - - S u b m e n ta l

Subm and.

Figure 1. Mean (and SDs) AC, BC, and STC (both sites) baseline thresholds in dB HL at 0.5,1,2, and 4 kHz, from 10 normal-hearing participants, wearing earphones throughout. Submand. = subman­ dibular.

tion (AC, BC, STC-submental, and STC-submandibular with and without earplugs) and frequency. ANOVA indi­ cated a significant difference of the tested condition (fisted above) [H(7189) = 69.826, p < 0.001]. Also found was a significant frequency effect LF(3j189) = 156.571, p < 0.001) and a significant interaction between the tested condition and frequency [Hai.isg) = 11.861, p < 0.001). Pairwise multiple comparison procedures (Tukey test) revealed significant threshold shifts while using earplugs for AC (21-37 dB threshold shifts de­ pending on the frequency), whereas BC thresholds were not altered after the earplugs were used. Simi­ larly, STC thresholds at both sites also did not show

F re q u e n c y

RESULTS B aseline M easurem ents The mean (and standard deviation [SD]) thresholds (in dB HL) in the baseline audiograms for the AC, BC, and two STC sites for the 10 participants (with earphones in place) are displayed in Figure 1. Sim ulated CHL (Earplugs) The mean threshold shifts in dB (and SDs) between the baseline thresholds and those obtained after inser­ tion of earplugs (simulating a CHL) are shown in Figure 2 for each mode of stimulation, for all four frequencies tested. Two-way repeated-measures analysis of variance (ANOVA) separated the main effects of the tested condi-

104

-■ — AC

--♦ --B C

»

- s u b m e n ta l

-

a

, -s u b m a n d .

Figure 2. Mean (and SDs) AC, BC, and STC (both sites) thresh­ old shifts in dB between baseline thresholds and thresholds obtained with a simulated CHL (earplugs), at 0.5, 1, 2, and 4 kHz. Submand. = submandibular.

AC, BC, and STC in Simulated Hearing Losses/Adelman et al

a significant shift w ith earplugs compared w ith baseline, except for a small threshold elevation a t 1 kHz a t the sub­ m andibular triangle STC site ( p = 0.017). The BC threshold shift observed a t each frequency was not sig­ nificantly different from those at each of the STC sites.

Simulated SNHL (Masking) M asking w ith 30 dB NBN caused m ean threshold shifts of 23.5-36.5 dB (depending on frequency), as dis­ played in Figure 3. Two-way repeated-measures ANOVA separated the m ain effects of the tested condition (BC, STC at the subm ental and subm andibular sites) and frequency. A signif­ icant m ain effect was found for frequency [F l3f >4:) = 8.756, p < 0.001]. ANOVA found no significant effect of the test condition [F(2 j5 4 ) = 0.486, p = 0.623] nor a significant interaction between the tested condition and frequency LFfe,5 4 }= 1.635, p = 0.155]. Pairwise multiple comparison procedures (Tukey test) revealed no significant difference between the threshold shifts of BC, STC a t the submental and subm andibular sites. In all conditions w hen the vibrator was in the air over the BC and STC sites (as a control for the possibility th a t th e response was to AC sound produced by the bone vibrator), th e thresholds were poorer th a n those ob­ tained w ith bone vibrator in contact w ith skin.

DISCUSSION his study has assessed the thresholds of the responses in hum ans w ith norm al hearing to STC stim ulation, com paring them w ith AC and BC thresholds, before th e induction of the sim ulated losses,

T

Frequency 0.5kHz

1kHz

2kHz

4kHz

Figure 3. Mean AC, BC, and STC (both sites) threshold shifts in dB between baseline thresholds and thresholds obtained in a simulated SNHL (30 dB EML NBN masking) at 0.5, 1, 2, and 4 kHz. The SDs could not be clearly presented here because there is almost complete overlap of the means and SDs at each fre­ quency. Submand. = submandibular.

and again in the same participants in the presence of a sim ulated CHL and a sim ulated SNHL. The sim ulations were achieved by inserting an earplug for a CHL, sim ilar to th a t in the study hy N ia and Bance (2001), and by pre­ senting NBN for an SNHL. The CHL in this study, as in N ia and Bance (2001), sim ulated common CHLs such as otitis media, atresia of the external m eatus, and cerum en blockage, and also in general reduced the intensity of sound reaching the inner ear, as in all CHLs. The STC (nonosseous BC) sites chosen for evaluation in this study were based on our experience w ith STC sites of stim ulation and on the need for sites w ith a large dynamic range (i.e., low thresholds) and sites th a t are easily accessible and provide stable, repeatable results. It was necessary to use earphones during testing in order to reduce the possibility th a t STC thresholds would include an AC com ponent from air-conducted sounds coming from the bone vibrator. The audiogram s obtained at both of the STC sites assessed are seen to slope down at the higher frequencies (Fig. 1), being 25-35 dB poorer a t 4 kHz com pared w ith those at 500 Hz. This m ay be p artly because of the occlusion effect in th e presence of th e earphones, causing an im provem ent in thresholds a t th e lower frequencies. In addition, a t th e higher frequencies th e earphones likely blocked air-conducted sounds produced by th e bone v ibrator from reaching th e ear, resu ltin g in thresholds a t th e higher frequencies in th e presence of th e earphones, which w ere poorer th a n those u s u ­ ally obtained in the clinic, w here earphones are u s u ­ ally ab sen t. T his m ay be th e case not only w ith th e STC th resh o ld s, b u t also w ith th e BC th resh o ld s: the baseline BC results in the presence of earphones displayed in F igure 1 probably reflect th e occlusion effect a t low frequencies coupled w ith th e blocking of air-conducted sound coming from the bone vibrator at high frequencies. Although it is accepted clinical-audiological practice to assess BC thresholds w ith the bone vibrator in the absence of earphones, the BC thresholds then obtained in th a t situation without earphones may include responses to the AC components (mainly at high frequencies) coming from the bone vibrator. For the most part, these are elim inated when the BC thresholds are assessed in the presence of earphones. This trend has been documented previously in some participants (e.g., Lightfoot, 1979; Shipton et al, 1980; Matos et al, 2010). From the results, it appears th a t the combination of the occlusion effect a t low frequencies, coupled w ith the prevention of air-conducted sounds a t high frequencies, may affect STC to a greater extent than BC. An additional factor may be poorer sensitivity of STC at higher frequencies compared with lower frequencies, as reported by Ito et al ( 2011).

An im portant factor likely contributing to the sloping STC audiogram s a t baseline (Fig. 1) m ay be related to the use of the clinical bone vibrator (calibrated for

1D5

Journal o f the American Academy o f AudiologyA/olume 26, Number 1, 2015

stimulation at the mastoid, a BC site), for stimulation at STC sites also. Thus, the physical-mechanical basis for the higher dB HL threshold settings of the audiometer (in BC mode) during stimulation at the STC sites (which are relatively far from any skull bone) is the lower mechanical impedance of the soft tissue sites, so that the transmission of vibratory energy is less efficient compared with when it is applied to the mastoid (having a higher impedance). Therefore, a sensation level (threshold) at the STC site equal to that at the mastoid site is achieved when the bone vibrator is applied to STC sites with a higher dynamic force (higher audio­ meter setting). This observation has been reported and discussed earlier (Adelman and Sohmer, 2013). In addition, it is possible that, similar to the other com­ ponents of osseous BC where each is dominant in a dif­ ferent frequency range (Stenfelt and Goode, 2005), nonosseous BC may dominate at low frequencies. Also, some of the difference between the BC and STC thresh­ olds at the sites tested may be the result of greater phys­ ical distance between the STC stimulation sites and the cochlea (Sohmer, 2014; Kaufmann et al, 2012a). Thus, although the BC and STC thresholds in Figure 1 cannot be directly compared (because the audiometer settings are calibrated for bone conduction), Figures 2 and 3 showing the threshold shifts in dB in the presence of CHL and SNHL can be compared, and the statistical comparison showed no statistical difference in the threshold shifts between those at the BC and at both STC sites. In general, the STC threshold shifts obtained in the present study in normal-hearing participants with simu­ lated CHL and SNHL, compared with the AC and BC thresholds in the normal state in the same partici­ pants, in the same conditions, were as expected: in the simulated CHL, only the AC thresholds are ele­ vated, and the BC and STC thresholds are not signifi­ cantly elevated (except for a small threshold elevation at 1 kHz at the submandibular STC site). In the simu­ lated SNHL, the threshold shifts for each of the three modes of stimulation were not statistically different from each other. This study has focused on the thresholds (and the shifts in threshold) that are the standard, routine, uni­ versally accepted, and most easily evaluated quantita­ tive measure of hearing. Thus, the overall results of the present HL simulation study indicate that the STC (nonosseous BC) threshold shifts are statistically simi­ lar to the BC stimulation threshold shifts in both CHL and SNHL. Earlier parts of this report have presented evidence from previous studies (e.g., auditory sensation induced by the bone vibrator at sites on the neck and thorax not overlying skull bone) that it is likely that threshold intensity STC stimulation is not based on, nor does it require, the induction of vibrations of the temporal/

IDS

petrous bones in order to excite the cochlea. Further­ more, evidence has been presented that the threshold for BC stimulation at the mastoid or forehead is below the intensity that induces skull bone vibrations, as can be seen in other studies in which much higher BC stim­ ulation levels (from 35-70 dB HL) had to be actually delivered in order to induce bone vibrations that could be directly detected by accelerometers (Ito et al, 2011) and by very sensitive laser Doppler vibrometers (Eeg-Olofsson et al, 2008; Eeg-Olofsson et al, 2013; Chordekar et al, 2012b; 2013a). Because the BC thresh­ old shifts in the simulated CHL and SNHL were stat­ istically similar to those of the STC threshold shifts, the finding that the BC threshold shifts (which have been thought to require vibrations of the temporal/ petrous bone) were similar to those at the STC sites may be explained by the following: at low-intensity oss­ eous BC stimulation in the clinic, when the bone vibra­ tor is applied to skin sites on the head (e.g., mastoid, forehead), the vibrator is actually in initial contact with the skin. In turn, this can induce vibrations of the skin and underlying soft tissues and thereby lead to STC stimulation at the BC stimulation site, and to auditory sensation, especially at threshold intensities. Suppor­ tive evidence for this mechanism comes from studies showing th a t STC-induced auditory sensation can be achieved with very low stimulation levels on the skin and soft tissue of the head without any static applica­ tion forces. Also, the only contact with the skin is by means of gentle contact of a moistened cotton wool wick coupled to the bone vibrator (Gealdor et al, 2012), or by gentle movement of the fingertips across the skin on the head over the skull bone, in the presence of ear plugs (in order to prevent AC stimulation), or over a bandage on the head over the skull bone, as in the “scratch test” (Iacovidou et al, 2014). On the other hand, as shown, larger dynamic forces delivered by the clinical bone vibrator (audiometer settings) are required to induce actual skull bone vibrations, as detected by a sensitive laser Doppler vibrometer. In other words, a bone vibra­ tor induces threshold auditory sensation at the BC stimulation site on the head at stimulus levels that are lower than those required to induce skull bone vibrations in participants with normal hearing. There­ fore, it is suggested that the behavioral threshold thus obtained to BC (mastoid, forehead) stimulation may result from the STC stimulation to the skin and soft tis­ sues at the site overlying the bone, and represents a low-threshold nonosseous BC (STC) part of the multicomponent BC mechanisms. In addition, recent studies have presented further evidence for a limited contribution of the middle ear to osseous BC hearing. For example, Rbosli et al (2012) compared umbo velocity in five humans during AC and BC stimuli and found significant lower vibration of the umbo during BC compared with AC stimuli, leading these

AC, BC, and STC in Sim ulated H earing Losses/Adelman et al

investigators to conclude th a t the middle ear contribution in BC hearing dom inates a t frequencies of less th an 500 Hz and th a t other m echanisms are dom inant a t higher frequencies. C hhan et al (2013) reported th a t immobi­ lization of th e m iddle ea r caused a sim ilar pressure reduction in both scala tym pani and scala vestibuli du rin g AC stim ulation, b u t during BC stim ulation th e re was a sm aller p ressu re reduction in the scala vestibuli and no significant p ressu re change in the scala tym pani. This finding led C hhan and colleagues to conclude th a t BC stim ulation does not involve ossic­ u la r motion, and th a t th ere m ay be a significant in n er e a r contribution to BC a t th e frequencies studied. M oreover, additional studies (Sohm er et al 2004a; Perez et al, 2011a; Sohm er et al 2004b; Sohm er 2014; Perez et al 2011b; 2014) have shown nonsignifi­ can t BC th resh o ld shifts after th e introduction of holes in th e cochlear shell, and after ossicular chain immobili­ zation and discontinuity, and window fixation. Further­ more, it has been shown th at auditory responses can be initiated in such animals (ossicular chain immobilization or discontinuity, coupled with window fixation) by applying low-intensity click vibratory stimulation produced by a rod attached to the bone vibrator into saline added to the mid­ dle ear cavity. The inner ear is then surrounded by the sa­ line (Perez et al, 2014). Such activation of the outer hair cells probably involves fluid com m unications (not bulk fluid flow) th ro u g h th e otic capsule. For example, it h a s been shown th a t such com m unications are perm eable to re l­ atively large molecules applied to saline in th e m iddle ear, which diffuse th ro u g h the continuous fluid envi­ ronm ent su rro u n d ing the in n er ea r into the in n er ea r fluids (M ikulec et al, 2009) so th a t they m ay also tra n sm it fluid p ressu res (not bulk fluid flow). This route, called th e lacuna-canalicular system, m ay also serve as the m echanism whereby the sound pressures initiated in the soft tissues of the head, neck, and thorax reach the cochlea during STC stim ulation, inducing auditory sensations. Together, these studies therefore support the suggestion th a t during threshold intensity BC stim ulation at the mastoid, the excitation of the co­ chlea is likely dominated by a nonosseous (STC) mecha­ nism (especially at low frequencies) as a result of the direct contact and vibration of the bone vibrator with the soft tissues overlying the bone, w ith m inimal involve­ m ent of vibrations of the tem poral/petrous bone. At higher-intensity BC stim ulation, the osseous m echa­ nism s of BC, lead in g to vib ratio n s of th e o u ter and m iddle ears, are likely involved in BC transm ission modalities. From th e re su lts of threshold shift determ inations in norm al-hearing hum ans in the present study, it seems th a t the BC threshold in participants is, for the most part, the STC (nonosseous BC) threshold a t the BC stim ­ ulation site, and therefore reflects tru e cochlear function. This understanding may also provide an explanation for

the near-norm al BC thresholds in patients w ith otoscle­ rosis (Vincent et al, 2006), round-window atresia (Pappas et al, 1998; Linder et al, 2003; B orrm ann and Arnold, 2007), and postradical m astoidectom y (H ornung and Ostfeld, 1984): the thresholds determ ined w ith th e clin­ ical bone vibrator at the BC stimulation sites are probably the STC (nonosseous BC) thresholds at the BC stim ulation site. This notion also receives support from a prelim i­ n ary study in norm al-hearing p articip an ts showing th a t bone v ibrator thresholds applied w ith a 5 N appli­ cation force were, in general, sim ilar to those applied w ithout an application force (i.e., 0 N force) (Chordekar et al, 2013b). Finally, we can now understand the clinical success in the use of BC threshold determ inations to dif­ ferentiate between CHL and SNHL: the BC threshold m ay really reflect the STC (nonosseous BC) threshold at the BC stim ulation site (mastoid) and therefore repre­ sents the true cochlear function because it activates STC mechanisms, w ith minimal, if any, involvement of the tem poral/petrous bone. The present study h as dem onstrated a sim ilarity betw een the STC threshold shifts and the BC threshold shifts and has presented evidence th a t the BC th re sh ­ olds are likely dom inated by the nonosseous (STC) com­ ponent of the m ultifactorial BC m echanism s. Therefore, STC thresholds (together w ith AC thresholds) m ay be useful in a differential diagnosis of those clinical condi­ tions for which it would be problem atic to apply the bone vibrator w ith an application force of 500 g (5 N) to the mastoid or forehead (e.g., in severe skull fractures, wide­ spread head hem atom as, mastoiditis, or abscess a t these sites), especially in children.

REFERENCES Adelman C, Fraenkel R, Kriksunov L, Sohmer H. (2012) Interac­ tions in the cochlea between air conduction and osseous and non­ osseous bone conduction stimulation. Eur Arch Otorhinolaryngol 269(21:425-429. Adelman C, Perez R, Sohmer H. (2013) Effect on auditory thresh­ old of air and water pockets bordering the pathway between soft tissue stimulation sites and the cochlea. Ann Otol Rhinol Laryngol 122(8):524-528. Adelman C, Sohmer H. (2013) Thresholds to soft tissue conduction stimulation compared to bone conduction stimulation. Audiol Neurootol 18(1):31—35. Baun J. (2004) Physical Principles o f General and Vascular Sonog­ raphy. San Francisco: California Publ. Co. Berger EH, Kieper RW, Gauger D. (2003) Hearing protection: sur­ passing the limits to attenuation imposed by the bone-conduction pathways. J Acoust Soc A m 114(4 P t 1):1955-1967. Borrmann A, Arnold W. (2007) Non-syndromal round window atresia: an autosomal dominant genetic disorder with variable penetrance? Eur Arch Otorhinolaryngol 264(9):1103-1108. Chhan D, Roosli C, McKinnon ML, Rosowski JJ. (2013) Evidence of inner ear contribution in bone conduction in chinchilla. Hear Res 301:66-71.

107

J o u r n a l of th e A m erican A cadem y of Audiology/Volume 26, Number 1, 2015

Chordekar S, Adelman C, Sohmer H. (2013b) The non-osseous (soft-tissue) component of bone conduction thresholds. J Basic Clin Physiol Pharmacol 5 Abstract. Chordekar S, Kriksunov L, Adelman C, et al. (2012b) Under water hearing: By bone conduction or by soft tissue conduction. J Basic Clin Physiol Pharmacol 23(3):121-133. Chordekar S, Kriksunov L, Kishon-Rabin L, Adelman C, Sohmer H. (2012a) M utual cancellation between tones presented by air con­ duction, by bone conduction and by non-osseous (soft tissue) bone conduction. Hear Res 283(1-2): 180-184. Chordekar S, Perez R, Adelman C, Sohmer H. (2013a) Assessment of inner ear bone vibrations during auditory stimulation by bone conduction and by soft tissue conduction. J Basic Clin Physiol Pharmacol 24(3):201-204. Eeg-Olofsson M, Stenfelt S, Taghavi H, et al. (2013) Transmission of bone conducted sound - correlation between hearing perception and cochlear vibration. Hear Res 306:11-20. Eeg-Olofsson M, Stenfelt S, Tjellstrom A, Granstrom G. (2008) Transmission of bone-conducted sound in the hum an skull mea­ sured by cochlear vibrations. Int J Audiol 47(12):761-769. Geal-Dor M, Shore AF, Sohmer H. (2012) Auditory sensation via moist contact of the bone vibrator with skin at soft tissue sites. J Basic Clin Physiol Pharmacol 23(3):99-101. Hornung S, Ostfeld E. (1984) Bone conduction evaluation related to mastoid surgery. Laryngoscope 94(4):547-549. Hyvarinen J, Sakata H, Talbot WH, Mountcastle VB. (1968) Neu­ ronal coding by cortical cells of the frequency of oscillating periph­ eral stimuli. Science 162(3858):1130-1132. Iacovidou A, Giblett N, Doshi J, Jindal M. (2014) How reliable is the “scratch test” versus the Weber test after tympanomastoid sur­ gery? Otol Neurotol 35(5):762-763. Ito T, Roosli C, Kim CJ, Sim JH , Huber AM, Probst R. (2011) Bone conduction thresholds and skull vibration measured on the teeth during stimulation at different sites on the human head. Audiol Neurootol 16(l):12-22. Kauftnann M, Adelman C, Ohanuna R, et al. (2012a) Borders of areas in which hearing sensation is elicited by bone vibrator along the back, and correlation with body type. Abstracts of Annual Con­ ference o f the Israeli Speech, Hearing and Language Association, 92. Abstract. Kaufmann M, Adelman C, Sohmer H. (2012b) Mapping at sites on bone and soft tissue of the head, neck and thorax at which a bone vibrator elicits auditory sensation. Audiol Neurotol Extra 2:9-15. Lightfoot GR. (1979) Air-borne radiation from bone conduction transducers. B r J Audiol 13(2):53-56. Linder TE, Ma F, Huber A. (2003) Round window atresia and its effect on sound transmission. Otol Neurotol 24(2):259-263. Matos Rd, Valle SdeP, Dias AM, Santos TM, Leite IC. (2010) Acoustic radiation effects on bone conduction threshold measure­ ment. Braz J Otorhinolaryngol 76(5):654-658. Mikulec AA, Plontke SK, Hartsock JJ, Salt AN. (2009) Entry of substances into perilymph through the bone of the otic capsule after intratympanic applications in guinea pigs: implications for local drug delivery in humans. Otol Neurotol 30(2):131-138.

108

Nia J, Bance M. (2001) Effects of varying unilateral conductive hearing losses on speech-in-noise discrimination: an experimental study with implications for surgical correction. Otol Neurotol 22(6):737-744. Pappas DG, Jr, Pappas DG, Sr, Hedlin G. (1998) Round window atresia in association with congenital stapes fixation. Laryngo­ scope 108(8 P t 1):1115-1118. Perez R, Adelman C, Chordekar S, Ishai R, Sohmer H. (2014) Air, bone and soft tissue excitation of the cochlea in the presence of severe impediments to ossicle and window mobility. EurArch Oto­ rhinolaryngol 10.1007/s00405-014-2887-8. Perez R, Adelman C, Sohmer H. (2011a) Several mechanical manip­ ulations of the wall of the inner ear do not affect air and bone con­ duction auditory thresholds. Ann Otol Rhinol Laryngol 120(l):66-70. Perez R, Adelman C, Sohmer H. (2011b) Bone conduction activa­ tion through soft tissues following complete immobilization of the ossicular chain, stapes footplate and round window. Hear Res 280(l-2):82-85. Roosli C, Chhan D, Halpin C, Rosowski JJ. (2012) Comparison of umbo velocity in air- and bone-conduction. Hear Res 290(1-2): 83-90. Schlauch RS, Nelson P. (2009) Pure tone evaluation. In: Katz J, B urkard R, Hood L, Medwetsky L, eds. Handbook o f Clinical Audiology. 6th ed. Baltimore, MD: Lippincott, Williams & Wil­ kins. Shipton MS, John AJ, Robinson DW. (1980) Air-radiated sound from bone vibration transducers and its implications for bone con­ duction audiometry. Br J Audiol 14(3):86-99. Sohmer H. (2014) Reflections on the role of a traveling wave along the basilar membrane in view of clinical and experimental find­ ings. Eur Arch Otorhinolaryngol. Sohmer H, Freeman S, Perez R. (2004a) Semicircular canal fenes­ tration - improvement of bone- but not air-conducted auditory thresholds. Hear Res 187(1-2):105—110. Sohmer H, Sichel JY, Freeman S. (2004b) Cochlear activation at low sound intensities by a fluid pathway. J Basic Clin Physiol Pharmacol 15(1-2): 1-14. Stenfelt S, Goode RL. (2005) Bone-conducted sound: physiological and clinical aspects. Otol Neurotol 26(6):1245-1261. Tonndorf J. (1968) A new concept of bone conduction. Arch Otolar­ yngol 87(6):595-600. Vento BA, Durrant JD. (2009) Assessing bone conduction thresholds in clinical practice. In: Katz J, Burkard R, Hood L, Medwetsky L, eds. Handbook of Clinical Audiology. 6th ed. Baltimore, MD: Lippincott, Williams & Wilkins. Vincent R, Sperling NM, Oates J, Jindal M. (2006) Surgical findings and long-term hearing results in 3,050 stapedotomies for primary otosclerosis: a prospective study with the otology-neurotology data­ base. Otol Neurotol 27(8 Suppl 2):S25-S47. Watanabe T, Bertoli S, Probst R. (2008) Transmission pathways of vibratory stimulation as measured by subjective thresholds and distortion-product otoacoustic emissions. Ear Hear 29(5):667-673. Wever EG, Lawrence M. (1954) Physiological Acoustics. Prince­ ton, NJ: Princeton University Press.

Copyright of Journal of the American Academy of Audiology is the property of American Academy of Audiology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.