34 The Hearing Journal High-frequency immittance measurements in infants October 2004 • Vol. 57 • No. 10

Using the 1000-Hz probe tone to measure immittance in infantsBy Johannes Lantz, Michelle Pe t rak, and Laura Pr i g g e

Recent studies using high-frequency immittance mea-s u rements have led to clinical recommendations for mid-dle ear assessment in infants. This is especially import a n tin light of the proliferation of universal neonatal hearings c reening programs. This article examines the theory ofimmittance and the differences between infant and adultmiddle ear anatomy. It is designed to introduce readers tothe high-frequency immittance concepts and their appli-cation to infant middle ear assessment.

THE FUNDAMENTALS OF IMMITTA N C E“ Im m i t t a n c e” is a collective term for the re c i p ro c a l s“ i m p e d a n c e” and “admittance.” Impedance (measured inmmho) is the o p p o s i t i o n to energy flow into a mediumand admittance (measure din mhos) is consequentlythe e a s e with which energyf l ows into a medium. Inmodern middle ear analyz-ers, admittance is the pro p-e rty that is displayed.

Immittance measure-ments we re first deve l o p e din the 1950s by Te rk i l d s e nand co-workers to measuremiddle ear pre s s u re. Ove rtime, the contribution ofimmittance measurement toclinical diagnostics hasbecome highly valued andit is now a routine part of the audiologic test battery.

The purpose of the middle ear system is to adapt theimpedance of air to the impedance of the lymphatic flu-ids of the inner ear. Without a good impedance matchb e t ween the two media, much of the incoming acousticenergy will be reflected at the plane of the mismatch andhearing will be compromised. Immittance measure m e n t sa re used to assess middle ear function by determiningwhether a normal amount of acoustic energy is transferre dt h rough, or reflected from, the middle ear system.

An immittance instrument principally consists of anair pump, a probe with speakers, a microphone, and am a n o m e t e r. A probe tone is continuously delive red intothe ear by one of the probe speakers and the acoustic immit-tance of the ear is analyzed through monitoring of thep robe-tone sound pre s s u re level in the ear canal by meansof the probe microphone.

Under optimal conditions for sound transfer and hear-ing, little sound energy is reflected at the tympanic mem-

brane—the middle ear input. Howe ve r, under less favo r-able conditions, a greater portion of the delive red pro b etone is reflected back to the probe microphone. Du r i n gimmittance monitoring, some well-defined alterations aremade, either an air pre s s u re sweep (tympanometry) or pre-sentations of stapedius reflex-eliciting stimuli (acousticreflex measurements). The resulting probe-tone re f l e c t i o n sa re compared with normal middle ear responses to suchp re s s u re sweeps or reflex stimuli.

The resulting amplitude and phase of the reflected pro b etone depend on the extent to which the different part s / p ro p-e rties of the middle ear contribute to the transfer or re f l e c-tion of the particular sound. The mechanic/acoustic middleear system consists of anatomic stru c t u res that increase the

f o rce of the incomingsound wave vibrations tomatch the impedanceand transfer the soundsuccessfully into the fluidmedium of the cochlea.

T h ree fundamentalp ro p e rties define the sys-t e m’s response to incom-ing sound: compliance,mass, and friction.❖ Compliance elementsinclude the tympanic andround window mem-branes, the ossicular lig-aments, the middle ear

muscles, and the air in the ear canal and middlee a r. These stru c t u res transfer the vibration in spring-like movements: compressing and expanding,s t retching and slackening. Compliance is the inve r s eof stiffness.

❖ Mass elements comprise the ossicles and the air inthe middle ear mastoid air cells moving as unitswithout compression or expansion. These are stru c-t u res with mass inert i a .

❖ Fr i c t i on is the cause of energy loss through dissipa-tion into heat. All real-life mechanical systemsinclude friction. Energy loss through friction occurswhen molecules in motion collide and rub againsteach other.

Susceptance: The interaction of compliance and massFirst we will take a closer look at the compliance and masselements. The relationship between mass and compliance

Figure 1. An illustration of the interaction of compliance suscep-tance and mass susceptance.

Circle 137 on Reader Service Card

Circle 138 on Reader Service Card

38 The Hearing Journal High-frequency immittance measurements in infants October 2004 • Vol. 57 • No. 10

is best illustrated by a compliant spring with a mass attached toits end (Fi g u re 1). When this basic system is set into motion, them oving mass continues moving forw a rd due to its inertia, grad-ually transferring its kinetic energy into the spring.

Eve n t u a l l y, when all the kinetic energy is transferred into springtension, the system reaches one extreme end of its cycle. T h eenergy that is now stored in the spring will change the dire c t i o nof the movement and start a new, re versed cycle, gradually trans-ferring the energy back into the mass and so on.

If no external force is applied to influence the movement ofthe system once it has been put into motion, as described above ,the system will oscillate back and forth spontaneously at its “re s-onant fre q u e n c y.” In order to move the same system more slow l ythan the resonant fre q u e n c y, the spring will be the opposing fac-tor of movement. The system is said to be “s t i f f n e s s - c o n t ro l l e d”at these lower frequencies. For the system to be moved faster thanthe resonant fre q u e n c y, the mass would be the opposing factor. Atthese higher frequencies, the system is said to be “m a s s - c o n t ro l l e d . ”

The interaction between the compliance elements and themass elements is called “s u s c e p t a n c e” (B). Susceptance is positiveat lower frequencies when the system is stiffness-controlled andn e g a t i ve when it is controlled by mass at higher frequencies. Atresonance, susceptance is ze ro.

Co n d u c t a n c eThe description above is, of course, simplified since we cannoth a ve a system without friction. If there we re no friction, the illus-tration of mass and compliance at the resonant frequency wouldbe a perpetual motion system that, once started, would continueto oscillate fore ve r. The impact of friction elements is called con-ductance (G). Since friction cannot be negative, conductance cann e ver take a negative value.

A d m i t t a n c eThe susceptance (B) and conductance (G) components fully deter-

mine the admittance (Y) of the system. The admittance is thevector length when plotting susceptance and conductance coor-dinates in a Cartesian coordinate system (Fi g u re 2).

The conductance component is in phase with the delive re dp robe tone, whereas the susceptance is an “o u t - o f - p h a s e” com-ponent. Consequently, the susceptance can be separated from theconductance by analyzing the phase of the reflected probe tone.This has been used in some middle ear analyzers to plot B/G tym-panograms (Fi g u re 3). From the Cartesian graph, it is appare n tthat the admittance (Y) is always defined by the quantities of thefundamental components, susceptance, and conductance. T h i sis the underlying physical rationale, re g a rdless of which curve sa re presented in the middle ear analyze r.

THE 226-HZ PROBE TONEThe most commonly used probe-tone frequency is 226 Hz. T h i sp robe tone has some definitive advantages for testing the adultear because the adult middle ear system is stiffness-controlled atthis fre q u e n c y. The frequency 226 Hz is below the normal adultresonance, which lies between 650 and 1400 Hz, so the effectsof mass and friction are minor. Due to the negligible contribu-tion of mass and friction, most instruments have even labeledtheir admittance results “compliance.”

The 226-Hz curves are typically single-peaked and result in

Figure 2. Ca rtesian plot showing the admittance magnitude atg i ven susceptance and conductance va l u e s .

Figure 3. The appearance of susceptance/conductance (top) andadmittance magnitude tympanograms (bottom) from the sameadult ear, measured with a 1000-Hz probe tone and displaye don the GN Ot o Metrics OTOdiagnostics Suite software.

Circle 105 on Reader Service Card

40 The Hearing Journal High-frequency immittance measurements in infants October 2004 • Vol. 57 • No. 10

e a s y - t o - i n t e r p ret tympanograms. Mo re ove r, the acoustic re f l e xe scan be reliably assessed when using a probe tone that is well belowresonant fre q u e n c y. The admittance change associated with acti-vating the stapedius muscle is ve ry small and could easily be under-estimated with the phase shift that occurs in the vicinity of theresonant fre q u e n c y.1 These phase shifts are negligible at 226 Hz.

Another advantage of this probe tone, and the ve ry reason why226 Hz was originally chosen, is that the true compliance va l u eat this frequency is numerically equal to an enclosed volume ofa i r. This is utilized for calibration purposes, since the admittanceis 1 mmho when measured in a 1-cc cavity of air. And by usingair pre s s u re, e.g., at +200 daPa, to create an impedance mismatchat the level of the tympanic membrane and there by pro h i b i t i n gthe sound transfer through the middle ear, one can obtain thee q u i valent ear canal volume (ECV) in cubic centimeters.

Most diagnostic immittance instruments offer only a 226-Hzp robe tone, although it has been shown that other probe fre-quencies can obtain different re s u l t s .2-5

M U LTIPLE-FREQUENCY TYMPA N O M E T RYMu l t i p l e - f requency tympanometry (MFT) is a method in whichthe probe tone is swept through a series of frequencies, e.g., fro m250 to 2000 Hz. T h rough MFT it is possible to assess the re s o n a n tf requency of the middle ear system. The resonant frequency is thep robe-tone frequency at which susceptance becomes ze ro due tothe counteractive forces of its compliance and mass elements.

The frequency can be determined from the measurement wherethe notch of the baseline compensated B curve reaches ze ro. T h emean adult resonant frequency is about 900 Hz. As mentioned,the middle ear system is stiffness-controlled below and mass-con-t rolled above this fre q u e n c y, depending on which susceptance ele-ment (mass or compliance) is more pro m i n e n t .6 The changes inresonant frequency are sometimes used to assess the pathology ofthe adult middle ear system, especially of the ossicular chain.

I M M I T TANCE MEASUREMENTS IN INFA N T ST h e re have been several re p o rts of infants below the age of about6 months demonstrating what appear to be normal 226-Hz tym-panograms even with confirmed middle ear effusion.7 It is alsopossible to obtain abnormal 226-Hz tympanograms in normalears in this age gro u p.8

Tympanograms collected from infant ears clearly pro g ress dif-f e rently from those collected from adult ears. It has also beens h own that acoustic re f l e xes cannot be reliably measured using a226-Hz probe tone in this population. T h e re are many anatomicd i f f e rences in the developing infant ear and, while we do not fullyk n ow the impact of each, the sum of differences accounts for thepeculiar 226-Hz immittance findings re p o rted by different inve s-t i g a t o r s .

External and middle ear changes after birth that could accountfor the acoustic alterations include:

❖ s i ze increase of the external ear, middle ear cavity, and mas-toid

❖ a change in the orientation of the tympanic membranefusion of the tympanic ring

❖ a decrease in the overall mass of the middle ear (due tochanges in bone density, loss of mesenchyme)

❖ tightening of the ossicular joints ❖ closer coupling of the stapes to the annular ligament ❖ the formation of the bony ear canal wall Unlike the adult middle ear, which is a stiffness-controlled sys-

tem at low frequencies, the infant middle ear is a mass-dominatedsystem with a lower resonant fre q u e n c y.9 During deve l o p m e n tof the infant ear, several changes also take place that influence themechanical pro p e rties of the ear canal. Ear canal changes includethe fusing of the tympanic ring and formation of the bony re g i o n .This process invo l ves mechanical changes, which influence thetympanogram. It has been re c o g n i zed that the external and mid-dle ear systems can va ry significantly in their acoustic re s p o n s ep ro p e rties over the first 2 years after birt h .8

Tympanograms in pediatric audiology are most commonlyused to identify the presence of middle ear fluid, i.e., effusion.Acoustic re f l e xes have also proven a ve ry good complement tot y m p a n o m e t ry, since any conductive malfunction will diminishthe reflex-induced immittance change. Hence, a present re f l e xis a strong indicator that the middle ear is healthy.1 0 Because ofthe lower resonant frequency of the infant ear, higher-fre q u e n c yp robe tones have been explored for tympanometric and acousticreflex immittance measurements. T h rough the use of either MFTor single high-frequency probe tones, many re s e a rchers have con-cluded that higher probe-tone frequency tympanometry canaccurately identify middle ear effusion.3 , 5 ( For an extensive re v i ewsee Pu rdy and Wi l l i a m s .1 1) While it has been concluded that theuse of mainly 678-Hz and 1000-Hz probe tones is preferable touse of 226-Hz probe tone for infants, the 1000-Hz probe toneis pre f e r re d .1 2

The 1000-Hz probe tone has typically not been available indiagnostic immittance instruments. Yet, it is clearly re q u i red foruse in diagnostic testing when neonatal screening ABR or OA Eis abnormal. As Pu rdy and Williams note, “…low - f requency tym-p a n o m e t ry is unreliable and should not be used.”12

USING THE 1000 HZ TONEAlthough MFT may have some advantages in the amount of datacollected, the pro c e d u re has proven far too awkward and slow for

Figure 4. An infant re f e r red from a neonatal screening pro g ram is measured with a 1000-Hz probe tone using the Ma d s e nOTOflex 100 from GN Ot o m e t r i c s .

westone

Circle 145 on Reader Service Card

use with the infant population. As furt h e rre s e a rch continues to support high-fre-quency tympanometry, the literaturep resently concludes that the best choiceof a tympanometric probe frequency forinfants under about 6 months of age is1000 Hz.

Re s e a rch facilities are currently col-lecting data on normative values, andsome material has already been pub-l i s h e d .1 3 , 1 4 Still, there are many unknow n sre g a rding the sensitivity and specificity of1000-Hz tympanometry to the pre s e n c eof middle ear effusion in infants. Fu rt h e rre s e a rch is needed into more population-specific 1000-Hz susceptance, conduc-tance, and admittance n o r m a t i ve data,

as well as in assessing the sensitivity andspecificity of these measure s .

Fi n a l l y, it is recommended that high-f requency immittance measurements beincluded in a battery of tests to identifyany abnormality in an infant’s hearing,but tympanometry and acoustic re f l e xm e a s u rements are most effective whenthey are interpreted along with behavioralt h resholds and ABR and OAE re s u l t s .

Cu r rent literature re c o m m e n d a t i o n sinclude the following:

❖ Be l ow about 6 months of age, a1000-Hz probe tone should be usedfor detecting middle ear effusion.

❖ Acoustic re f l e xes complement tym-p a n o m e t ry and should be pre s e n t

in healthy ears when using a 1000-Hz probe tone and ipsilateralb road-band noise stimulus.

❖ The following 1000-Hz tym-panograms are considered normal:single- or double-peaked curve sand discernible Y, B, or G peak.

❖ If normative data are not ava i l a b l efor the specific age gro u p, a sus-ceptance (B) curve with no dis-cernible peak is considere di n d i c a t i ve of effusion.

Johannes Lantz, BSc, is an Audiologist at GN Otometrics A/S,Denmark. Michelle Petrak, PhD, is Audiologist and Product Man-ager at GN Otometrics, North America. Laura Prigge, MA, is Sup-p o rt, Education, and Training Audiologist at GN Otometrics, Nort hAmerica. Correspondence to Lantz at GN Otometrics A/S, 2 Dyben-dalsvaenget, 2630 Ta a s t rup, Denmark or e-mail to jlantz@gnoto-m e t r i c s . d k .

R E F E R E N C E S1. Wilson RH, Margolis RH: Ac o u s t i c - reflex measure m e n t s .

In Musiek ME, Rintelmann W F, eds., ContemporaryPe r s p e c t i ves in Hearing As s e s s m e n t . Boston, MA: Allynand Bacon, 1999.

2. Hunter LL, Margolis RH: Mu l t i f requency tympanom-e t ry: Cu r rent clinical application. AJA 1 9 9 2 ; 1 : 3 3 - 4 3 .

3. Ma rchant CD, Mc Millan PM, Shurin, PA: Ob j e c t i vediagnosis of otitis media in early infancy by tympa-n o m e t ry and ipsilateral acoustic reflex thresholds. JPediatr 1 9 8 4 ; 1 0 9 : 5 9 0 - 5 .

4. Shurin PA, Pelton SI, Klein JO: Otitis media in the new-born infant. Ann Otol Rhinol Lary n g o l 1976; 85 (Su p p l .25):216-222.

5. Shurin PA, Pelton SI, Finkelstein J: Ty m p a n o m e t ry inthe diagnosis of middle-ear effusion. N Engl J Me d1977;296:412-417.

6. Shanks JE, Lilly DJ: An evaluation of tympanometricestimates of ear canal volume. J Sp Hear Re s1 9 8 1 ; 2 4 , 5 5 7 - 5 6 6 .

7. Me yer SE, Ja rdine CA, De verson W: De ve l o p m e n t a lchanges in tympanometry: A case study. Brit J Au d i o l1 9 9 7 ; 3 1 : 1 8 9 - 1 9 5 .

8. Keefe DH, Levi E: Maturation of the middle and exter-nal ears: Acoustic power-based responses and re f l e c t a n c et y m p a n o m e t ry. Ear Hear 1 9 9 6 ; 1 7 : 3 6 1 - 7 3 .

9. Holte L, Margolis RH, Cavanaugh RM Jr.: De ve l o p-mental changes in multifrequency tympanograms.Audiology 1991;30:1-24.

10. Gates GA, St ew a rt IA, No rthern JL, et al.: Re c e n ta d vances in otitis media. Diagnosis and screening. An nOtol Rhinol Laryngol Suppl 1 9 9 4 ; 1 0 3 : 5 3 - 7 .

11. Pu rdy SC, Williams MJ: High frequency tympanom-e t ry: A valid and reliable immittance test protocol foryoung infants? New Z Audiol Soc Bu l l e t i n 2 0 0 0 ; 1 0 : 9 -2 4 .

12. Sutton G, Baldwin M, Brooks D, et al.: Ty m p a n o m e-t ry in neonates and infants under 4 months: A re c-ommended test protocol. 2002; The Newborn He a r i n gS c reening Programme, UK, 2002; online atw w w. n h s p. i n f o.

13. Margolis RH, Bass-Ringdahl S, Hanks WD, et al: Ty m-p a n o m e t ry in newborn infants: 1-kHz norms. JAAA2 0 0 3 ; 1 4 ( 7 ) : 3 8 3 - 3 9 2 .

14. Kei J, Allison-Levick J, Dockray J, et al.: Hi g h - f re q u e n c y(1000 Hz) tympanometry in normal neonates. JAAA2 0 0 3 ; 1 4 ( 1 ) : 2 0 - 2 8 .

42 The Hearing Journal High-frequency immittance measurements in infants October 2004 • Vol. 57 • No. 10

Circle 140 on Reader Service Card

united benefits

“...it is recommended that high-frequencyimmittance measurements be included in ab a t t e ry of tests to identify any abnormality

in an infant’s hearing...”