Interpreting Test Results
Review of Concepts
The main outcome of interpretation of the audiologic results should be to provide information to support the goal(s) of the patient and the referring physician.
There are additional questions that the audiologic test outcomes should provide answers to, with the focus of differing depending on the purpose for the evaluation. One issue that must be determined regardless of the rationale for testing is the validity of testing outcomes. Test validity can be assessed in a number of ways, and before drawing conclusions about the function of the auditory system, the audiologist must determine whether the testing outcomes are an accurate reflection of function.
In general, there are four areas of concern for the clinician regarding function of the auditory system: middle ear function, hearing sensitivity, retrocochlear function, and communicating function. The measures collected during the audiologic exam generally contribute to understanding more than one of these areas of function and provide little useful information when considered in isolation. The role of the clinician is to incorporate all aspects of the audiologic evaluation to understand each area of audiologic function.
There are additional questions that the audiologic test outcomes should provide answers to, with the focus of differing depending on the purpose for the evaluation. One issue that must be determined regardless of the rationale for testing is the validity of testing outcomes. Test validity can be assessed in a number of ways, and before drawing conclusions about the function of the auditory system, the audiologist must determine whether the testing outcomes are an accurate reflection of function.
In general, there are four areas of concern for the clinician regarding function of the auditory system: middle ear function, hearing sensitivity, retrocochlear function, and communicating function. The measures collected during the audiologic exam generally contribute to understanding more than one of these areas of function and provide little useful information when considered in isolation. The role of the clinician is to incorporate all aspects of the audiologic evaluation to understand each area of audiologic function.
Middle Ear Function
A major area of concern for many patients and physicians is the function of the middle ear system. Tympanometric and acoustic reflex immittance results, audiometric thresholds, otoacoustic emissions and word recognition testing all add information to understanding the function of the middle ear system.
In general, middle ear dysfunction is characterized by abnomal tympanometric results, abnormal acoustic reflex immittance results, and air-bone gaps on the audiogram consistent with conductive hearing loss. In addition, in most cases of middle ear disorder otoacoustic emission responses cannot be recorded due to the attenuation of the backward transmission of sound. Therefore they will be "absent." Furthermore, word recognition testing results are typically better in the case of conductive hearing loss than they would be in the case of sensorineural hearing loss.
In some cases of middle ear dysfunction, such as otosclerosis, tympanometric results may appear normal, but acoustic reflex results will be abnormal with a corresponding conductive hearing loss indicated by air-bone gaps on the audiogram.
In the case of third-window syndromes, where there is no middle ear disorder, it is typical to find air-bone gaps on the audiogram, with normal tympanometric and acoustic reflex immittance results, and otoacoustic emissions and word recognition results consistent with the degree of air-conduction hearing thresholds.
In general, middle ear dysfunction is characterized by abnomal tympanometric results, abnormal acoustic reflex immittance results, and air-bone gaps on the audiogram consistent with conductive hearing loss. In addition, in most cases of middle ear disorder otoacoustic emission responses cannot be recorded due to the attenuation of the backward transmission of sound. Therefore they will be "absent." Furthermore, word recognition testing results are typically better in the case of conductive hearing loss than they would be in the case of sensorineural hearing loss.
In some cases of middle ear dysfunction, such as otosclerosis, tympanometric results may appear normal, but acoustic reflex results will be abnormal with a corresponding conductive hearing loss indicated by air-bone gaps on the audiogram.
In the case of third-window syndromes, where there is no middle ear disorder, it is typical to find air-bone gaps on the audiogram, with normal tympanometric and acoustic reflex immittance results, and otoacoustic emissions and word recognition results consistent with the degree of air-conduction hearing thresholds.
Hearing Sensitivity
The presence or absence of hearing loss is a central aspect of the interpretation of the audiologic evaluation. When hearing loss is present, the hearing loss should be quantified according to the type, degree and configuration of the hearing loss. The type of hearing loss is typically characterized as conductive, sensorineural, or mixed. The presence or absence of significant air-bone gaps on the audiogram provide information required to understand the type of hearing loss. In the case of significant air-bone gaps the hearing loss is typically described as conductive. In the case of absence of significant air-bone gaps the hearing loss is typically described as sensorineural. In the case of presence of air-bone gaps and elevated bone-conduction thresholds, the hearing loss is typically described as mixed.
Clinic Note: In some cases, such as third window syndromes with the presence of air-bone gaps, the hearing loss is not "conductive" in the sense that it is typically used in describing results. In such cases it may be most useful to the refering provider when interpreting results to refer to the "air-bone gap" rather than to use the term "conductive."
The degree of hearing loss should be described using common terminology (mild, severe, etc.). The configuration of the hearing loss should also be described using common terminology (high frequency, low-frequency, flat, etc.).
Clinic Note: In some cases, such as third window syndromes with the presence of air-bone gaps, the hearing loss is not "conductive" in the sense that it is typically used in describing results. In such cases it may be most useful to the refering provider when interpreting results to refer to the "air-bone gap" rather than to use the term "conductive."
The degree of hearing loss should be described using common terminology (mild, severe, etc.). The configuration of the hearing loss should also be described using common terminology (high frequency, low-frequency, flat, etc.).
Retrocochlear Function
When a sensorineural hearing loss is identified it is important to obtain information to help distinguish as much as possible between a primarily sensory (cochlear) hearing loss and a primarily neural hearing loss. In cases of sensorineural hearing loss of primarily cochlear origin, the acoustic reflex thresholds, OAE results, and word recognition scores should be consistent with the degree of hearing sensitivity. In cases where the hearing loss is of strictly neural origin, acoustic reflex thresholds may be absent at hearing sensitivity levels where they would typically be present. In contrast OAE results may be present at hearing sensitivity levels where they would typically be absent. Word recognition results may be poorer than expected based on audiometric thresholds or when there is rollover of word recognition score performance at high intensities. When there is significant asymmetry between ears with regard to pure-tone thresholds or word recognition scores, concern is raised regarding retrocochlear pathology.
Communication Function
The degree of hearing sensitivity loss as well as speech audiometric outcomes help provide an understanding of the patient's communication function. Coupled with the patient's history, these outcomes should be considered to begin to determine the need for audiologic intervention, whether by amplification or other means.
Comparing Air and Bone-Conduction Thresholds
We can deduce the location of a problem from the following principles: (1) air-conduction tests the whole ear, and (2) bone-conduction tests the sensorineural part of the ear. Thus a difference between the air and bone-conduction thresholds implies that there is a problem with the conductive system.
Sensorineural hearing loss is indicated by air and bone-conduction thresholds that are equal, or at least very close to one another. sensorineural losses can be caused by a disorder of the cochlea or auditory nerve or both. The combination term sensorineural is used to highlight the fact that we cannot distinguish between cochlear (sensory) and eighth nerve (neural) disorders from the audiogram.
Sensorineural hearing loss is indicated by air and bone-conduction thresholds that are equal, or at least very close to one another. sensorineural losses can be caused by a disorder of the cochlea or auditory nerve or both. The combination term sensorineural is used to highlight the fact that we cannot distinguish between cochlear (sensory) and eighth nerve (neural) disorders from the audiogram.
Immittance at the Plane of the Tympanic Membrane
For diagnostic purposes we are mainly concerned with the immittance of the middle ear because it provides information about (1) middle ear pathologies and (2) middle ear muscle contractions due to the acoustic reflex. However, the probe tip monitors the immittance of the ear from the perspective of its location, which is in the general vicinity of the ear canal entrance. Thus, the probe tip measures the total immittance of the ear, which includes the combined effects of the outer ear and the middle ear. This is a problem because ear canal volume (size) is usually not clinically relevant, yet its influence on the total immittance value (at the probe tip) is often big enough to cloud the effects of the clinically signifiant middle ear immittance value. For example, a patient with abnormally low middle ear admittance due to a conductive disorder may have a normal total admittance value due to a large ear canal volume. Another individual whose middle ear is normal might seem to have unusually low total admittance because their ear canal volume is quite small. A third patient might have low total admittance due to otitis media when first evaluated. The middle ear problem might be completely resolved (i.e., the middle ear immittance has returned to normal) when they return for reassessment, but the total Ya might still be abnormally low simply because her outer ear volume was made to appear smaller by a very deeply inserted probe tip. This can happen because the volume under the probe tip will be different depending on how deeply it has been inserted.
We must remove the outer ear component from the total admittance value at the probe tip to get an undistorted representation of the middle ear's admittance at the eardrum. In other words, removing the effect of the ear canal moves the measurement location from the end of the probe tip to the plane of the tympanic membrane. We can achieve this goal by taking advantage of the fact that total admittance (Ytotal) is simply the sum of the admittances of the outer ear (Yoe) and the middle ear (Yme):
Ytotal = Yoe + Yme
Consequently, subtracting the admittance of the outer ear from the total admittance leaves the middle ear admittance, which is the value that we need.
Yme = Ytotal - Yoe
This simple, additive relationship is one of the main reasons why we use measurements based on admittance rather than impedance.
The procedure for determining the admittance of the middle (i.e., at the plane of the ear drum) is simple and straightforward: The first step is to measure the total admittance (Ytotal) of the ear. The second step is to measure the ear's admittance again while pressure is being exerted on the tympanic membrane. This measurement reflects the admittance of the outer ear (or ear canal) alone. The pressure change is accomplished using the pressure pump connected to one of the tubes in the probe tip. The rational e for this tactic is that the heightened air pressure puts the eardrum under so much tension that it acts lie a hard wall, so that essentially no sound energy can be transmitted into the middle ear. In this case, we say that the middle ear has been excluded from the measurement. Hence, the admittance obtained under these conditions.
We now know the total admittance (of the outer ear and middle ear combined) from the first measurement and the admittance of the outer ear from the second measurement. The third step is to figure out the previously unknown middle ear admittance (Yme) by simply subtracting the outer ear admittance (Yoe) from the total admittance (Ytotal).
We must remove the outer ear component from the total admittance value at the probe tip to get an undistorted representation of the middle ear's admittance at the eardrum. In other words, removing the effect of the ear canal moves the measurement location from the end of the probe tip to the plane of the tympanic membrane. We can achieve this goal by taking advantage of the fact that total admittance (Ytotal) is simply the sum of the admittances of the outer ear (Yoe) and the middle ear (Yme):
Ytotal = Yoe + Yme
Consequently, subtracting the admittance of the outer ear from the total admittance leaves the middle ear admittance, which is the value that we need.
Yme = Ytotal - Yoe
This simple, additive relationship is one of the main reasons why we use measurements based on admittance rather than impedance.
The procedure for determining the admittance of the middle (i.e., at the plane of the ear drum) is simple and straightforward: The first step is to measure the total admittance (Ytotal) of the ear. The second step is to measure the ear's admittance again while pressure is being exerted on the tympanic membrane. This measurement reflects the admittance of the outer ear (or ear canal) alone. The pressure change is accomplished using the pressure pump connected to one of the tubes in the probe tip. The rational e for this tactic is that the heightened air pressure puts the eardrum under so much tension that it acts lie a hard wall, so that essentially no sound energy can be transmitted into the middle ear. In this case, we say that the middle ear has been excluded from the measurement. Hence, the admittance obtained under these conditions.
We now know the total admittance (of the outer ear and middle ear combined) from the first measurement and the admittance of the outer ear from the second measurement. The third step is to figure out the previously unknown middle ear admittance (Yme) by simply subtracting the outer ear admittance (Yoe) from the total admittance (Ytotal).
Tympanometry
Tympanometry involves measuring the acoustic admittance of the ear with various amounts of air pressure in the ear canal. We can control the amount of air pressure in the ear canal because the probe tip makes a hermetic seal with the ear canal and one of it tubes is connected to an air pump and manometer. The amount of air pressure is expressed in terms of dekapascals (daPa) relative to atmospheric pressure in the room where the test is being done. Hence, 0 daPa implies that the pressure in the ear canal is equal to the atmospheric pressure, positive pressure (+100) means that the ear canal pressure is greater than atmospheric pressure and negative pressure (-100) means it is less than atmospheric pressure. This information is shown on a diagram called a tympanogram with admittance in mmhos. Atmospheric pressure (0 daPa) is in the middle, with positive pressure increasing to the right and negative pressure increasing to the left.
The major considerations that come into play when interpreting 226 Hz tympanograms include static acoustic admittance, tympanometric gradient or width, ear volume, the pressure at which the tympanometric peak occurs, and the shape of the tympanogram. It is also possible to categorize tympanograms into a reasonably small number of types that summarize the majority of configurations found clinically. The most widely utilized tympanogram classification system was originated by Jerger. These types were based on relative tympanograms obtained with an immittance bridge, and this format is retained in the figure.
Type A tympanograms had a distinctive peak in the vicinity of atmospheric pressure and were tpical of normal patients, as well as those with otosclerosis. If the type A tympanogram had a very shallow peak, it was classified as type As, which was generally associated with otosclerosis but could also occur with otitis medai. In contrast, very high (deep) type A tympanograms were designated as type Ad. These were found in otherwise normal ears that had scarred or flaccid eardrums, or in cases of ossicular interruptions. The type Add tympanogram was so deep that the peak was off-scale, and was found in ears with ossicular discontinuities.
Type B tympanograms were essentially flat across the pressure range and were characteristic of patients with middle ear fluid and cholesteatoma. However, type B tympanograms can also be caused by entities such as eardrum perforations or impacted cerumen (or other obstructions) in the ear canal.
Type C tympanograms had negative pressure peaks beyond -100 daPa indicating negative middle ear pressure. They were associated with Eustachian tube disorders, and were also found in cases of middle ear fluid.
The major considerations that come into play when interpreting 226 Hz tympanograms include static acoustic admittance, tympanometric gradient or width, ear volume, the pressure at which the tympanometric peak occurs, and the shape of the tympanogram. It is also possible to categorize tympanograms into a reasonably small number of types that summarize the majority of configurations found clinically. The most widely utilized tympanogram classification system was originated by Jerger. These types were based on relative tympanograms obtained with an immittance bridge, and this format is retained in the figure.
Type A tympanograms had a distinctive peak in the vicinity of atmospheric pressure and were tpical of normal patients, as well as those with otosclerosis. If the type A tympanogram had a very shallow peak, it was classified as type As, which was generally associated with otosclerosis but could also occur with otitis medai. In contrast, very high (deep) type A tympanograms were designated as type Ad. These were found in otherwise normal ears that had scarred or flaccid eardrums, or in cases of ossicular interruptions. The type Add tympanogram was so deep that the peak was off-scale, and was found in ears with ossicular discontinuities.
Type B tympanograms were essentially flat across the pressure range and were characteristic of patients with middle ear fluid and cholesteatoma. However, type B tympanograms can also be caused by entities such as eardrum perforations or impacted cerumen (or other obstructions) in the ear canal.
Type C tympanograms had negative pressure peaks beyond -100 daPa indicating negative middle ear pressure. They were associated with Eustachian tube disorders, and were also found in cases of middle ear fluid.
Static Acoustic Immittance
Static acoustic immittance is the immittance of the middle ear at some "representative" air pressure.
A given static admittance measurement is considered to be (1) within normal limits if it falls within the normal range (2) abnormally low if it falls between the lower limit of the normal range and (3) abnormally high if it is above the upper limit of the normal range.
in adults, the 90% normal range is 0.37 - 1.66 mmhos
2.6mmhos is well above the upper limit of 1.66 mmhos
Another way to quantify the flatness of a tympanogram is to determine the typanometric width which is simply the width of the typanogram in daPa measured at 50% of its static acoustic admittance value.
Tympanometric widths that are too wide are associated with middle ear effusion and normative data may be used to help determine when this is the case. Representative upper cutoff values for tympanometric width are 235 daPa for infants and 200 daPa for 1 year olds through school age children. Typical 90% normal ranges for adults are 51 to 114 daPa
A given static admittance measurement is considered to be (1) within normal limits if it falls within the normal range (2) abnormally low if it falls between the lower limit of the normal range and (3) abnormally high if it is above the upper limit of the normal range.
in adults, the 90% normal range is 0.37 - 1.66 mmhos
2.6mmhos is well above the upper limit of 1.66 mmhos
Another way to quantify the flatness of a tympanogram is to determine the typanometric width which is simply the width of the typanogram in daPa measured at 50% of its static acoustic admittance value.
Tympanometric widths that are too wide are associated with middle ear effusion and normative data may be used to help determine when this is the case. Representative upper cutoff values for tympanometric width are 235 daPa for infants and 200 daPa for 1 year olds through school age children. Typical 90% normal ranges for adults are 51 to 114 daPa
Ear Volume
Tympanograms with extremely small or absent peaks are often referred to as essentially flat. These findings are usually attributed to extremely low middle ear admittance and are typically associated with middle ear pathologies such as otitis media and cholesteatoma. However, we can reach this conclusion only if the volume (admittance at 226Hz) measured at +200 daPa is attributable to the ear canal. If the volume is too large, then the flat tympanogram maybe due to such causes as (1) a perforated tympanic membrane; (2) a patent myringotomy tue, if one is present; or (3) the absence of a hermetic seal.
It is reasonable to consider the volume to be larger than normal when it exceeds 2.0 mmhos in children and 2.5 mmhos in adults. On the other hand, the following causes are associated with volumes that are too small: (1) a clogged probe tip; (2) a probe tip that is pushed against the canal wall; (3) impacted cerumen or another obstruction in the ear canal; and (4) a clogged myringotomy tube if one is present. These cases are usually identified by volumes at or close to 0 mmhos. Flat tympanograms associated with tympanic membrane perforation, otitis media with effusion and a clogged probe tip are differentiated on the basis of their ear canal volumes. It is very improtatn to keep this issue in mind when tympanograms are classified by type because the letter designation does not account for the ear volume.
It is reasonable to consider the volume to be larger than normal when it exceeds 2.0 mmhos in children and 2.5 mmhos in adults. On the other hand, the following causes are associated with volumes that are too small: (1) a clogged probe tip; (2) a probe tip that is pushed against the canal wall; (3) impacted cerumen or another obstruction in the ear canal; and (4) a clogged myringotomy tube if one is present. These cases are usually identified by volumes at or close to 0 mmhos. Flat tympanograms associated with tympanic membrane perforation, otitis media with effusion and a clogged probe tip are differentiated on the basis of their ear canal volumes. It is very improtatn to keep this issue in mind when tympanograms are classified by type because the letter designation does not account for the ear volume.
Tympanometric Peak Pressure
We know the pressure on the outer ear side of the eardrum because it is generated and measured using the air pump and manometer connected to the probe tip. In addition, we have learned that the tympanogram peak occurs when the same pressure exists on both sides of the tympanic membrane. Consequently, the ear canal pressure corresponding to the tympanogram peak is also an estimate of the pressure within the middle ear.
Abnormally negative tympanometric peak pressures are associated with Eustachian tube disorders, which can occur either with or without the presence of middle ear fluid. The amount of negative pressure needed to consider the tympanometric peak pressure abnormally negative is not clearly identifiable in the literature. Suggested cutoff values vary widely. In practice, -100 daPa appears to be a reasonable low cutoff value for tympanometric peak pressure. Lower pressures suggest the possibility of eustachian tube dysfunction. Unfortunately, there does not seem to be a particular tympanometric peak pressure cutoff value that successfully distinguishes between the presence and absence of middle ear effusion.
Abnormally negative tympanometric peak pressures are associated with Eustachian tube disorders, which can occur either with or without the presence of middle ear fluid. The amount of negative pressure needed to consider the tympanometric peak pressure abnormally negative is not clearly identifiable in the literature. Suggested cutoff values vary widely. In practice, -100 daPa appears to be a reasonable low cutoff value for tympanometric peak pressure. Lower pressures suggest the possibility of eustachian tube dysfunction. Unfortunately, there does not seem to be a particular tympanometric peak pressure cutoff value that successfully distinguishes between the presence and absence of middle ear effusion.
The Acoustic Reflex
Presenting a sufficiently intense sound to either ear results in the contraction of the stapedius muscle in both ears; and is called the acostic or stapedius reflex. This reflexive muslce contraction stiffens the conductive mechanism via the stapedius tendon and there fore changes the ear's immittance. The acoustic reflex is easily measured because the immittance change is picked up by the probe tip and displayed on the immittance device meter.
A variety of acoustic reflex tests are regularly used in clinical assessment. The two basic measurements are discussed here: the acoustic reflex threshold, which is the lowest stimulus level that produces a reflex response; and acoustic reflex decay which is a measure of how long the response lasts if the stimulus is kept on for a period of time
A variety of acoustic reflex tests are regularly used in clinical assessment. The two basic measurements are discussed here: the acoustic reflex threshold, which is the lowest stimulus level that produces a reflex response; and acoustic reflex decay which is a measure of how long the response lasts if the stimulus is kept on for a period of time
Acoustic Reflex Threshold
Acoustic reflex threshold (ART) testing involves finding the lowest level of a stimulus that causes a measurable change in acoustic immittance. We may also say that the ART is the smallest discernible immittance change that is associated in time with the presentation of a stimulus, and that responses should also be present (and generally larger) at high stimulus levels. Clinical ARTs are usually obtained using pure tone stimuli at 500, 1000 and 2000 Hz. Even though some clinicians also use 4000 Hz, it is not recommended because even young, normal-hearing persons experience elevated ARTs at this frequency due to rapid adaptation. Clinically, pure tone ARTs are obtained by changing the intensity of the stimulus in 5dB steps while watching for admittance changes caused by the stimuli. These admittance changes are observed by watching for deflections on the admittance device meter and the ART is considered to be the lowest intensity causing a deflectionthat can be distinguished from the background activity on the meter. This approach is often called "visual monitoring with 5dB steps."
Normal ARTs with a 220 or 226Hz probe tone occur between ~85 and 100 dB SPL for pure tones. Most clinical measurements involve pure tone ARTs.
Normal ARTs with a 220 or 226Hz probe tone occur between ~85 and 100 dB SPL for pure tones. Most clinical measurements involve pure tone ARTs.
Pure Tones
|
Contralateral ART Ipsilateral ART |
500 Hz
84.6 79.9 |
1000 Hz
85.9 82.0 |
2000 Hz
84.4 86.2 |
4000 hz
89.8 87.5 |
Acoustic Reflex Decay
In addition to the ART, it is also common practice to test for acoustic reflex decay (or adaptation), which is a measure of wheterh a reflex contraction is maintained or dies out during continuous stimulation. Reflex decay is tested at both 500 and 1000 Hz. Higher frequencies are not tested because even normal people can have rapid reflex decay above 1000Hz. The test involves presenting a stimulus tone continuously for 10 seconds at a level 10dB above the reflex threshold. The magnitude of the reflex response will either stay the same or decrease over the course of the 10 second stimulus. The central issue is whether the response decays to half of its original magnitude. If the magnitude of the reflex response does not decrease to 50% of it's original size during the 10 second test period, then the outcome is considered negative. The test is considered positive if the magnitude of the reflex response does decay by 50% or more within this time period.
Conductive Hearing Loss
Conductive hearing losses cause acoustic reflexes to be either "elevated" or "absent." By "elevated" we mean that the ART is higher than normal, that is, it takes more intensity to reach the reflex threshold than would have been needed if there was no conductive loss. An "absent" reflex means that a reflex response cannot be obtained, even with the most intense stimulus available (which is usually 125 dB HL on most modern immittance devices). The effects of conductive loss can be summarized by two basic rules:
- Probe-ear rule - The presence of conductive pathology in the probe ear causes the acoustic reflex to be absent. Here, even though the stapedius muscle itself may actually be contracting the presence of the pathology prevents us from being able to register any change in acoustic admittance that can be picked up by the probe tip.
- Stimulus-ear rule - A conductive loss in the stimulus ear causes the ART to be elevated by the amount of the conductive impairment. This occurs because the amount of the stimulus that actually reaches the cochlea will be reduced by the amount of the air-bone-gap. For example, suppose an otherwise normal patient develops otitis media with a 25dB air-bone-gap. The 25-dB-air-bone-gap causes the signal reaching the cochlea to be 25 dB weaker than the level presented from the earphone. If her ART would normally have been 85 dB HL (without the conductive loss) then the stimulus would now have to be raised by 25 dB to 110 dB HL to reach her cochlea at 85 dB HL. Hence, the ART would now be elevated to 110 dB HL. In addition, if the air-bone-gap is large enough, then the ART will be elevated so much that the reflex will be absent. This occurs because the highest available stimulus level cannot overcome the size of the air-bone-gap and still deliver a large enough signal to the cochlea.
Sensorineural Hearing Loss
Acoustic reflex thresholds depend on hearing sensitivity in a rather peculiar way. In this context "hearing sensitivity" represents a continuum going from normal hearing through various magnitudes of sensorineural hearing loss due to cochlear disorders.
It is well established that patients with retrocochlear pathologies have acoustic reflexes that are elevated, often to the extent that the reflex is absent. However, the decision about when and ART is "elevated" must account for the fact that the ARTs depend on the magnitude of the hearing loss in patients who do not have retrocochlear involvement. The 90th percentiles provide us with upper cutoff values for ARTs that meet this need. In fact, many prior inconsistencies about the diagnostic usefulness of the reflex were resolved by the introduction of 90th percentiles that account for the degree of hearing loss.
In practice, the patient's ARTs are compared with the respective 90th percentiles that apply to his hearing thresholds for the frequencies tested. If an ART falls on or below the relevant 90th percentile, then it is considered to be essentially within the normal and/or cochlear distribution. However, ARTs that fall above the applicable 90th percentiles are considered elevated because only a small proportion of normal and/or cochlear-impaired ears have ARTs that are so high. If the abnormally elevated or absent reflexes are not attributable to a conductive disorder, then the patient is considered to be at risk for eighth nerve pathology in the ear that receives the stimulus. In contrast, many patients with functional impairments have ARTs that are below the 10th percentiles.
Abnormal reflex decay means that the response decreases rapidly and is associated with retrocochlear disorders. It is prudent to be suspicious of retrocochlear pathology in the stimulated ear if the reflex response decays by 50% or more within 10 seconds at either 500 Hz and/or 1000 Hz. However, the student should be aware that there are differnces in approach regarding whether abnormal decay should occur during the first 5 seconds versus any time during the 10 second test as well as how to interpret reflex decay that occurs at one frequency and/or the other.
We have seen that acoustic reflex abnormalities such as elevated/absent reflexes and/or positive reflex decay are associated with retrocochlear pathologies in the ear receiving the stimulus tone. Testing both ARTs and reflex decay and considering a positive result on either or both tests as the criterion for suspecting retrocochlear pathology has been referred to as acoustic reflexes combined. Acoustic reflex thresholds and decay should be routinely tested becase ARC with 90th percentiles has a hit rate of ~85% for retrocochlear pathology and a false positive rate of only 11%.
It is well established that patients with retrocochlear pathologies have acoustic reflexes that are elevated, often to the extent that the reflex is absent. However, the decision about when and ART is "elevated" must account for the fact that the ARTs depend on the magnitude of the hearing loss in patients who do not have retrocochlear involvement. The 90th percentiles provide us with upper cutoff values for ARTs that meet this need. In fact, many prior inconsistencies about the diagnostic usefulness of the reflex were resolved by the introduction of 90th percentiles that account for the degree of hearing loss.
In practice, the patient's ARTs are compared with the respective 90th percentiles that apply to his hearing thresholds for the frequencies tested. If an ART falls on or below the relevant 90th percentile, then it is considered to be essentially within the normal and/or cochlear distribution. However, ARTs that fall above the applicable 90th percentiles are considered elevated because only a small proportion of normal and/or cochlear-impaired ears have ARTs that are so high. If the abnormally elevated or absent reflexes are not attributable to a conductive disorder, then the patient is considered to be at risk for eighth nerve pathology in the ear that receives the stimulus. In contrast, many patients with functional impairments have ARTs that are below the 10th percentiles.
Abnormal reflex decay means that the response decreases rapidly and is associated with retrocochlear disorders. It is prudent to be suspicious of retrocochlear pathology in the stimulated ear if the reflex response decays by 50% or more within 10 seconds at either 500 Hz and/or 1000 Hz. However, the student should be aware that there are differnces in approach regarding whether abnormal decay should occur during the first 5 seconds versus any time during the 10 second test as well as how to interpret reflex decay that occurs at one frequency and/or the other.
We have seen that acoustic reflex abnormalities such as elevated/absent reflexes and/or positive reflex decay are associated with retrocochlear pathologies in the ear receiving the stimulus tone. Testing both ARTs and reflex decay and considering a positive result on either or both tests as the criterion for suspecting retrocochlear pathology has been referred to as acoustic reflexes combined. Acoustic reflex thresholds and decay should be routinely tested becase ARC with 90th percentiles has a hit rate of ~85% for retrocochlear pathology and a false positive rate of only 11%.
Speech Audiometry Relation to the Pure Tone Audiogram
The pure tone thresholds in the 500 to 2000 Hz range are associated with the SRT. The lack of reasonable consistency between the SRT and the pure tone thresholds is associated with functional hearing losses. It is commonly accepted that one of the principal applications of the SRT is to corroborate the pure tone findings.
Testing Techniques
It is generally accepted that the SRT is the lowest hearing level at which a patient can repeat 50% of spondee words but there are many ways to find this point and no single technique is universally accepted. Each word is worth 2% on a 50-word test. For example, the speech recognition score would be 92% if four words are missed or repeated incorrectly, 86% if seven words are wrong. Speech recognition scores are generally expected to be ~90% to 100% in normal hearing individuals. The range of speech recognition scores is typically between 80 and 100% with most conductive losses (otitis media, otosclerosis) but has been found to be as low as 60% in cases of glomus tumor; and the range for sensorineural losses is anywhere from 0-100% depending on etiology and degree of loss. In general, speech recognition scores that are "abnormally low" are associated with retrocochlear lesions; however, there is no clear cutoff value for this decision. A lower than expected speech recognition score can also occur if the test level is not high enough. Thus, an atypically low speech recognition score often means the clinician must retest at a higher hearing level.
When is a speech recognition score "too low" with respect to the patient's audiogram? There is a rough relationship in which speech recognition scores tend to become lower as sensorineural hearing loss worsens, but it is hardly predictive and is complicated.
When is a speech recognition score "too low" with respect to the patient's audiogram? There is a rough relationship in which speech recognition scores tend to become lower as sensorineural hearing loss worsens, but it is hardly predictive and is complicated.
Performance-Level Functions
The percentage of words that are repeated correctly depends on more than just the patient's speech recognition ability. It also depends on the conditions of the test, such as the level at which the words are presented. Performance-level (PL) function shows how the patients speech recognition performance depends on the level of test materials. The more traditional term, performance-intensity (PI) function, is often used as well. The PL function is a type of psychometric function which shows performance as a function of some stimulus parameter. It is often called a PL-PB or PI-PB function when phonetically or phonemically balanced (PB) words are used.
The PL function is from a normal hearing individual and scores are low when the words are presented at very soft levels, and that the scores improve as the intensity increases. The maximum score on the PI-PB function is traditionally called PBmax. It is assumed that a patient's "speech recognition score" refers to PBmax unless otherwise indicated.
Rollover of the PL function is defined as a reduction of speech recognition scores that occurs at intensities above the level where PBmax is obtained. PBmin is the lowest speech recognition score obtained at intensity levels higher than where PBmax was obtained. These two scores are used to calculate a number called the rollover index (RI) and is calculated as follows:
The PL function is from a normal hearing individual and scores are low when the words are presented at very soft levels, and that the scores improve as the intensity increases. The maximum score on the PI-PB function is traditionally called PBmax. It is assumed that a patient's "speech recognition score" refers to PBmax unless otherwise indicated.
Rollover of the PL function is defined as a reduction of speech recognition scores that occurs at intensities above the level where PBmax is obtained. PBmin is the lowest speech recognition score obtained at intensity levels higher than where PBmax was obtained. These two scores are used to calculate a number called the rollover index (RI) and is calculated as follows:
RI = (PBmax - PBmin) / PBmax
When the NU-6 test is used, the cutoff value has been reported to be from 0.25 to 0.35. Abnormal rollover exists for the case in of the rollover of 0.41, which is significant for the NU-6 materials. Abnormal rollover also occurs in some elderly patients. These findings have been associated with neural presbycusis and exemplify the concept that audiological tests reflect the site of the disorder rather than its etiology
Testing Level
Routine speech recognition testing is often done at one hearing level for each ear. Some audiologists add a second measurement at high levels to screen for the possibility of rollover and the testing of speech recognition at more than one level is strongly encouraged. In any case it is necessary to choose a speech level that will most likely result in the highest speech recognition score for the ear being tested.
Many audiologists do routine speech recognition testing at 30, 35 or 40 dB above the patient's SRT. 40 dB SL seems to be the optimal choice among them by showing how scores change with testing level in dB SL re: SRT.
Comparisons of various presentation level methods led Guthrie and Mackersie (2009) to conclude that the highest scores tend to be obtained when the speech is presented 5 dB below the patients UCL for speech (UCL-5dB) for hearing losses ranging from mild through severe. They also found that the same results can be achieved for patients with mild to moderate sloping losses by presenting the speech at certain levels relative to the 2000Hz threshold.
Many audiologists do routine speech recognition testing at 30, 35 or 40 dB above the patient's SRT. 40 dB SL seems to be the optimal choice among them by showing how scores change with testing level in dB SL re: SRT.
Comparisons of various presentation level methods led Guthrie and Mackersie (2009) to conclude that the highest scores tend to be obtained when the speech is presented 5 dB below the patients UCL for speech (UCL-5dB) for hearing losses ranging from mild through severe. They also found that the same results can be achieved for patients with mild to moderate sloping losses by presenting the speech at certain levels relative to the 2000Hz threshold.
Acoustic Reflex
Acoustic reflex thresholds (ARTs) have been used as a physiological test for functional hearing loss since the 1960's. This is not surprising because most audiological evaluations include acoustic reflex testing as a matter of routine, so that applying ARTs to the assessment of functional impairment comes almost without cost.
A functional impairment is suspected when an ART occurs at or below the patient's hearing threshold. How high must an ART be? Recall that ARTs depend on the hearing threshold in normal and cochlear-impaird patients and that there is a range of ARTs that occurs for any given amount of loss. We consider ARTs to be atypically high if they exceeded the 90th percentiles. Now
A functional impairment is suspected when an ART occurs at or below the patient's hearing threshold. How high must an ART be? Recall that ARTs depend on the hearing threshold in normal and cochlear-impaird patients and that there is a range of ARTs that occurs for any given amount of loss. We consider ARTs to be atypically high if they exceeded the 90th percentiles. Now
Otoacoustic Emissions
Otoacoustic emissions (OAEs) have great potential as a test for at least some patients with nonorganic hearing loss because they are obliterated by many sensorineural hearing losses. The majority of adults with functional impairments also have some degree of underlying organic hearing loss.