Question
When testing infants (less than 6 months), I know it is important to use a high frequency probe tone for tympanometry. Is it better to use 678 Hz or 1000 Hz? Also, why does the volume measure change depending on what probe Hz is used. Is the volume measure inaccurate with the higher probe frequencies? Lastly, why does the compliance measure change from ml to mmho when using the 678 Hz or 1000 Hz probe tone. Any help with these 3 questions would be appreciated.
Answer
Human nature and our unwillingness to change cause us to stick with the familiar rather than adopt anything new. We continue to plot audiograms upside down, and we continue to use different units for audiograms (dB HL) and hearing aid plots (dB SPL), even though things would be much easier if these inconsistencies were corrected. We continue this tradition with acoustic immittance measures where we persist on using the word "compliance" inappropriately, even though we now know better. Some of your questions can be answered from this historical perspective.
Just as a known weight is applied to a bathroom scale to calibrate it, a known acoustic quantity is applied to an acoustic immittance device to calibrate it. An acoustic impedance device measures acoustic impedance, so a known acoustic impedance must be applied to calibrate it. Acoustic impedance consists of a resistance component ("friction") and a reactance component which is either positive (the "mass" or inertia portion) or negative (the "springiness" or compliance portion). A known acoustic resistance is difficult to create precisely. However, a known acoustic reactance is very easy to create with a hard-walled cavity of a defined volume. By definition, 1 cc volume (eg, a glass syringe set at 1 cc) has an acoustic impedance of exactly 1000 ohms at a frequency of 226 Hz. At this frequency, the acoustic impedance of this volume is comprised mostly of "compliance" (negative reactance) because this volume has virtually no resistance and no mass effects (no positive reactance). Because the normal adult ear at this low frequency also is comprised mostly of compliance effects with virtually no friction or mass effects, the early investigators assumed they were measuring "acoustic compliance" and they switched the acoustic units (ohms) to volume units (cc) that were the equivalent of the volume used to calibrate the device. Later these volume units were changed to milliliters (ml) which is essentially the same as cc.
The concepts above are the same for acoustic admittance devices except that these devices measure acoustic admittance (mmhos) which consists of a conductance component ("friction") and a susceptance component which is either negative (the "mass" or inertia portion) or positive (the "springiness" or compliance portion). At 226 Hz, the acoustic admittance of a 1 ml hard walled volume is 1.0 mmho.
Changing acoustic impedance units or acoustic admittance units to acoustic compliance units is OK only if what you are measuring does not have any mass or friction components. This is a reasonable assumption at a low frequency like 226 Hz for the normal adult ear, or during tympanometry when the ear is under high pressure (positive or negative - the ends of the tympanogram) when the ear canal is essentially a hard-walled cavity. However, with the pathologic adult ear, or with the normal or pathologic neonate ear, or for other portions of the tympanogram, or with higher probe frequencies, this change of units is incorrect because under these conditions, the ear does have significant components other that the springy component.
To summarize, the device only measures acoustic impedance (ohms) or acoustic admittance (mhos). Though we calibrate it with a known "compliance", it is still measuring ohms or mhos and should be labeled as such. The word "compliance" should be used only if you know that the ear you are measuring has no friction or mass effects. Clinically, this means that you should use the word "compliance" or "volume" only at 226 Hz and only in a normal adult ear or at the ends of a tympanogram in all other ears.
Now let's answer your specific questions:
"Why does the volume measure change depending on what probe Hz is used. Is the volume measure inaccurate with the higher probe frequencies?"Yes, the "volume" measure is inaccurate at higher frequencies. If the frequency is increased above 226 Hz, a hard-walled volume is not equivalent to the compliance part because it also has a mass part.
"Why does the compliance measure change from ml to mmho when using the 678 Hz or 1000 Hz probe tone?"The measure should not be called "compliance" in "ml" in the first place. It should always be called admittance in mmho because that is what the device is measuring. The mmho units could be changed to volume units like ml, but only if a low frequency is used and only if you know for sure that you are measuring a hard walled volume where there is only a compliance component (eg, the ear canal with a very stiff ear drum such as at the outer ends of a low frequency tympanogram).
"When testing infants (less than 6 months), I know it is important to use a high frequency probe tone for tympanometry. Is it better to use 678 Hz or 1000 Hz?"Mass and resistive components are much more significant in neonatal ears compared to the normal adult ear. Infant ear canals are cartilaginous and do not ossify until at least 4 months of age. The middle ear space has less volume and may contain amniotic fluid which makes the volume even smaller. The external ear canal wall vibration increases the resistive component. A variety of studies have shown empirically that of the three frequencies usually available (226, 678, and 1000 Hz), the 1000 Hz probe tone provides the most valid and the most reliable results for tympanometry in neonates and infants.
Here are two relevant studies:
Margolis, R. H., Bass-Ringdahl, S., Hanks, W. D., Holte, L., and Zapala, D. A. (2003). "Tympanometry in newborn infants--1 kHz norms," Journal of the American Academy of Audiology 14, 383-392.
Kei, J., Allison-Levick, J., Dockray, J., Harrys, R., Kirkegard, C., Wong, J., Maurer, M., Hegarty, J., Young, J., and Tudehope, D. (2003). "High-frequency (1000 Hz) tympanometry in normal neonates," Journal of the American Academy of Audiology 14, 20-28.
Gerald Popelka holds a PhD degree from the University of Wisconsin in Madison with an emphasis in neuroscience, and a two year post doctoral research fellowship in otolaryngology from UCLA. Prior to these he earned a masters degree in audiology from Kent State University. He was a faculty member for 24 years at Washington University inSt. Louis and joined the faculty at Stanford in 2004.
Dr. Popelka's research has been funded continuously with grants from a wide variety of agencies. He has initiated and completed successful collaborative research projects among diverse academic divisions in-cluding medicine and engineering.
Dr. Popelka is a co-inventor of the world's first all digital hearing aid. The resulting patent formed the basis of all current programmable and digital hearing aids or ~90% of all hearing aids currently sold. In 1996 he conceived and lead the development of JARO, the Journal of the Association for Research in Otolaryngology, now recognized as a premier, interna-tional auditory scientific research journal that was launched in 2000.
With over 80 research articles, many research presentations, two college textbooks, and various achievement awards, he has developed an international reputation for creating and using leading-edge technology that addresses both basic science and clinical applications. He remains in the forefront of developing innovative technology for biomedical applications, currently focusing on noninvasive measures of neonatal auditory function and hyper-bilirubinemia and real-time MRI imaging for speech and swallowing.