DPOAE Threshold Estimation is a unique and patented method to estimate the physiological hearing loss of the cochlea!

DPOAE Threshold Estimation is not to be mixed up with an Audiogram. An audiogram includes the whole pathway of hearing and the reaction/performance of the tester and patient. DPOAE Threshold Estimation focusses on the physiological place, where hearing loss most often has it's root cause - inside the cochlea!
Basing on optimized stimulus settings and implementation in Sentiero, the information derived from DPOAE Threshold Estimation is most valuable for children and non cooperative patients. But also for monitoring the status of the cochlea over a longer time period and for having a means of reliable, objective measurements for the function of the cochlear amplifier, DPOAE Threshold estimation is a valuable tool for the clinician. Diagnostics for kids and prior hearing aid fitting procedures can be sped up in time an quality!

Information about the method and it's scientific background can be downloaded HERE.

TEOAE/DPOAE stimulus setting

TEOAE responses can be evoked by two types of stimulus trains: (i) by a set of four clicks of equal magnitude (referred to as the linear protocol) or (ii) by three clicks of positive polarity followed by a fou­rth click of an inverse polarity with a relative magnitude of 9.5 dB higher than the cor­responding positive clicks (referred to as the non-linear protocol) (Kemp et al., 1986, Bray, 1989). Under the hypothesis that the TEOAE recordings originate from saturated cochlear generators, it is assumed that the non-linear protocol removes stimulus artifacts of a linear nature, i.e., the stimulus itself, because sound signals increase linearly with the stimulus level, while sound signals increasing non-linearly with the stimulus level, i.e., the emission emerging from the non-linear operation of OHCs, remain. It is generally accepted that the non-linear protocol is a practical compromise to maximize the reliability of a TEOAE recording (Kemp et al., 1990a, Kemp et al., 1990b, Grandori and Ravazzani, 1993, Von Specht et al., 2001, Hatzopoulos et al., 2003).

When using the “scissor” primary-tone setting (Fig. 2), which accounts for the different compression of the primary-tone traveling waves at the f2 place (Whitehead et al., 1995c, Whitehead et al., 1995a, Kummer et al., 2000, Boege and Janssen, 2002), the DPOAE I/O-function reflects compressive non-linear sound processing known from direct measurements of basilar membrane displacements (Ruggero et al., 1997). Due to the steep slope of the traveling wave towards the cochlear apex, the maximum interaction site is close to the f2 place in the cochlea. Thus, OHCs at the f2 place contribute most to DPOAE generation. The number of OHCs contributing to DPOAE generation depends on the size of the overlapping region, which is determined by the primary-tone levels L1 and L2, respectively, and the frequency ratio f2/f1 of the primary-tones f1 und f2. To preserve optimum overlap of the primary-tone traveling waves at a constant frequency ratio f₂/f₁=1.2 , the primary-tone level difference has to be increased with decreasing stimulus level resulting in a L1|L2 setting described by L1=0.4L2+39. This “scissor” L₁|L₂ setting (L1=65|L2=65, L1=63|L2=60, L1=61|L2=55 L1=59|L2=50, L1=57|L2=45, L1=55|L2=40, L1=53|L2=35, L1=51|L2=30 L1=49|L2=25, L1=47|L2=20) was derived from findings on the influence of the L1|L2 setting on the DPOAE level Ldp (Kummer et al., 2000). It should be emphasized that it is not the L1=L2 setting, but the “scissor” L1|L2 setting which yields compressive DPOAE growth reflecting non-linear CA sound processing (Fig. 3).

TEOAE/DPOAE stimulus calibration

Stimulus calibration is important to ensure proper data interpretation and comparability. There is a serious problem with stimulus calibration since, due to standing waves in the ear canal, the sound pressure level at the ear drum cannot be accurately set from measurements at the tip of the sound probe.

The most commonly used calibration method is the in-the-ear calibration based on the measurement of the sound pressure level at the ear probe microphone for constant voltage at the loudspeaker (Whitehead et al., 1995b). However, the ear probe microphone is located in the outer ear canal while the relevant magnitude for the generation of OAEs is the actual sound pressure level at the eardrum. Thus, dependent on ear canal length and middle-ear impedance, there is a frequency-dependent deviance of unknown quantity between the nominal sound pressure level at the tip of the ear probe and the actual sound pressure level at the eardrum due to standing wave effects (Siegel, 1994). This deviation is usually highest at frequencies corresponding to d = l/4 and l/2 (d: distance between ear probe and eardrum). Thus, problems become serious around 3 kHz and above 6 kHz in adults, but are less important in newborns and infants due to the smaller ear canal length (Keefe et al., 1993). Calibration errors may cause a change in the shape and thus the compression of DPOAE I/O functions. Further improvement of ear probe calibration is therefore necessary to enhance the accuracy of clinically relevant measures deduced from DPOAE recordings such as, for example, estimated CA threshold and compression.

In testing TEOAEs, calibration errors have less impact, because the wide-band stimulus is not influenced that strongly by standing wave problems. Apart from this, TEOAEs are usually stimulated with a click level that is relatively high, i.e., where cochlear compression already saturates cochlear motion. As opposed to DPOAEs, no level ratio between primary-tones needs to be fulfilled. This results in TEOAEs being less susceptible to stimulus calibration errors.

Relation between DPOAEs and behavioural pure-tone thresholds

The relation between DPOAE level and behavioural pure-tone threshold - or rather the lack of it – is strongly debated. Earlier, it was common to define confidence limits to determine the degree of certainty with which any measured response could be assigned to either normal or impaired hearing (Gorga et al., 1996, Gorga et al., 2000a), or to define a ‘DPOAE detection threshold’ as the stimulus level at which the response equalled the noise present in the instrument (Dorn et al., 2001).

However, since the noise is of technical origin (e.g., microphone noise) the threshold evaluated in this way does not match the behavioural threshold. A more relevant measure is the intersection point between the extrapolated DPOAE I/O-function and the primary-tone level axis at which the response’s sound pressure is zero and hence at which outer hair cell amplifiers are inactive (Boege and Janssen, 2002, Gorga et al., 2003). A linear dependency between the DPOAE sound pressure and the primary-tone sound pressure level is present (Boege and Janssen, 2002) when using the “scissor” paradigm for eliciting DPOAEs (Kummer et al., 2000). Because of the linear dependency, DPOAE data can be easily fitted by linear regression analysis in a semi-logarithmic plot, where the intersection point of the regression line with the L2 primary-tone level axis at pdp = 0 Pa can thus serve as an estimate of the DPOAE threshold. The estimated DPOAE threshold Ldpth is independent of noise and seems to be more closely related to behavioural threshold than the DPOAE detection threshold (Boege and Janssen, 2002, Gorga et al., 2003, Janssen et al., 2006).

Extrapolated DPOAE I/O-functions are used to determine DPOAE thresholds which can be displayed on in an audiogram form (DPOAE-audiogram). Circle symbol means threshold estimation by means of extrapolated DPOAE I/O functions, square symbol means simplified estimation where less than three valid responses are present and the lowest primary tone level – 15 dB is the estimate, arrow symbol means no DPOAE are measurable and thus the hearing loss is higher than 50 dB HL (Fig. 4).

Otoacoustic emissions (TEOAE and DPOAE) are widely regarded as being suitable for screening in newborns and infants, as they are not present in the case of outer hair cell dysfunction. (e.g. Kemp and Ryan, 1991, Gorga et al., 2000b, Norton et al., 2000b, Norton et al., 2000a). The premise for this approach is that inner ear hearing-loss always includes OHC damage or malfunction. However, conductive losses also cause “refer” results under screening conditions, mainly due to the attenuation of an existing OAE signal.

A major disadvantage of using OAEs in screening protocols is a lower validity as compared to ABR methods (Norton et al., 2000b, Barker et al., 2000). This is especially true in populations with a high prevalence of slight threshold elevation due to a temporary sound-conductive hearing loss, as it is found in term neonates in the first 36 hours of life because of Eustachian tube dysfunction or amniotic fluid in the tympanic cavity or due to a persisting sensory hearing-loss in premature and neonatal intensive care unit infants. In order to maintain a high sensitivity, the specificity may be reduced dramatically, making a screening procedure inefficient. To avoid high referral rates, OAE referrals should be followed up with an ABR screening before diagnostic assessment, i.e., two-stage screening (Rhodes et al., 1999, Norton et al., 2000a). DPOAE-audiograms may be an alternative means for revealing a temporary hearing loss in the early postnatal period caused by a temporary sound conductive hearing loss due to amniotic fluid or Eustachian tube dysfunction. When applying DPOAE audiograms before ABR screening, time and costs can be saved in those babies where DPOAEs are measurable and thus no additional ABR is needed.

Two case examples shall illustrate how DPOAE-audiograms can reveal temporary sound-conductive disorders in the early postnatal period (Fig. 5). In a 3 days old neonate with a “pass” response after ATEOAE screening the DPOAE-audiogram indicated a low frequency hearing loss. The DPOAE-audogram 83 days later revealed normal hearing function. In this baby, Eustachian tube dysfunction could have been the cause for the hearing loss within the first days of life where middle ear stiffness is increased and therefore low frequencies are affected. In another newborn with a “refer” ATEOAE screening response the DPOAE-audiogram indicated a hearing loss of more than 50 dB at 1.5, 2, 3, and 6 kHz (arrows) and a 40 dB hearing loss at 4 kHz. The second measurement 12 days later revealed normal hearing function. In this baby, both a low- and a high-frequency sound-conductive hearing loss might have been present where middle-ear stiffness is increased due to Eustachian tube dysfunction and thus low frequencies are affected and middle-ear mass is increased due to amniotic fluid and thus high frequencies are affected.

Fig. 6d shows mean and standard deviation of estimated DPOAE thresholds measured in 100 left and 100 right ears from 100 newborns at neonatal care unit. Mean age was 2.5 days. There was no significant difference between left and right ears. Standard deviation is partly due to different sound-conductive conditions during the early postnatal period. Results indicate that DPOAE threshold measurements can be done under hearing screening conditions.

DPOAE-audiograms can be applied in newborns to reveal a temporary sound-conductive hearing loss or to detect a persisting cochlear hearing loss. In follw-up diagnostics, DPOAE-audiograms can serve as an advanced tool for bridging the gap between screening and audiological testing.

DPOAE-audiograms are useful in pediatric audiology since they provide frequency specific information about the hearing loss in a couple of minutes. Thus, they have an advantage over TEOAEs or click-evoked ABRs (because they can not quantitatively assess cochlear hearing thresholds at distinct test frequencies) and ASSRs (because their measuring time is extremely long).

Three case examples shall demonstrate the efficacy of DPOAE-audiograms in pediatric audiology (Fig. 6 a,b,c). In a 6 years old boy, pure-tone audiogram and DPOAE-audiogram exhibited a close correspondence (Fig. 6 a). However, in the younger children there was a high discrepancy between the behavioural free-field audiograms and the DPOAE-audiograms. The free-field audiogram of a 5 month old girl (Fig. 6 b) indicated a hearing loss of 50 dB HL in the entire frequency range. However, the DPOAE-audiogram revealed a hearing loss only in the mid and low frequency region. Fig. 6 c shows the free-field audiogram and the DPOAE-audiograms of the left and the right ear. The free-field audiogram indicated a hearing loss of 40 dB HL. In the left ear no DPOAEs were measurable indicating that the hearing loss must be higher than 50 dB (arrows in Fig. 6 c). In the right ear the DPOAE-audiogram revealed normal hearing.

DPOAE-audiograms may assess cochlear hearing loss more precisely than behavioural tests, especially in infants where the conditioned free-field audiogram does not reflect the real threshold. Moreover, unilateral hearing loss can be detected. DPOAE-audiograms are an automated measuring procedure with simple handling and short measuring time. No sedative is necessary in most cases. It should be emphasized that DPOAEs only reflect outer hair cell functionality and therefore are not present at a hearing loss higher 50 dB HL. However, the incidence of a hearing loss higher than 50 dB is low. Thus, in most of the children DPOAE are measurable. In cases where DPOAEs are not measurable ABRs or ASSRs have to be applied.