Tympanometry in Clinical Practice

Two “must read” articles on the development of clinical tympanometry are Terkildsen and Thomsen (1959) and Terkildsen and Scott-Nielsen (1960). Their interest in estimating middle-ear pressure and in measuring recruitment with the acoustic reflex had a profound effect on the development of clinical instruments. Each time I read these articles, I am struck first by the incredible amount of information in the articles, and second, by the lack of scientific data to support their conclusions. Most amazing of all, however, is that the principles presented in these two articles have stood up for nearly 50 years, and provide the basis for commercial instruments and tympanometry procedures still in use today. These articles laid the foundation for the use of hardwalled calibration cavities and the term “equivalent volume of air”, compensation of ear-canal volume, and the selection of a single low-frequency probe tone. You should be surprised to read that the probe tone “frequency of 220 cps was chosen partly at random” (Terkildsen and Scott-Nielsen, 1960, p. 341). A low-frequency probe tone was preferred because the current day microphones were nonlinear at high frequencies and the probe tone level could be increased without eliciting an acoustic reflex. In other words, the selection of a 220 Hz probe tone was made without any consideration of its diagnostic value in evaluating middle-ear function. Finally, these articles also set the precedent for measuring only the magnitude of complex acoustic immittance rather than both magnitude and phase angle. Phase angle measurements were abandoned because the middle ear is so stiffness controlled at 220 Hz that phase angle did not vary considerably in either normal or pathological middle ears. The focus of Terkildsen and Thomsen (1959) was on estimating middle-ear pressure. They adapted an electroacoustic impedance instrument to allow for variation in ear-canal pressure over a range of ±300 mm H2O and described the first “tympanogram” as a uniform pattern “. . . with an almost symmetrical rise and fall, attaining a maximum at pressures equaling middle ear pressures” (p. 413). They further noted that, “. . . the smallest impedance always corresponded exactly to the zone of maximal subjective perception of the test tone” (p. 413). In other words, the probe tone was the most audible and the tympanogram peaked when the pressure was equal on both sides of the eardrum. In addition, these authors recognized that although the measure of interest was the acoustic immittance in the plane of the eardrum, for obvious reasons, the measurements had to be made in the ear canal. They introduced the clinical procedure used to compensate for the volume of air enclosed between the probe tip and the eardrum. “Under conditions where the ear drum is under considerable tension, the impedance volume measured is dominated by the volume of the ear canal space itself. With decreasing tension of the ear drum the influence of the middle ear space gradually increases, attaining a maximum under conditions where the pressures in both spaces are identical”. “. . . the difference between highest and lowest impedance volumes as obtained by variations of the pressure, is thought to some extent to indicate the vibrating characteristics of the individual ear drum” (p. 414). The difference between the highest and lowest impedances, therefore, was attributed to the middle-ear system, absent the effects of the ear canal. Here, they laid the basis for estimating ear-canal volume and calculating peak compensated static acoustic admittance. Instead of reporting tympanograms in acoustic ohms, Terkildsen and Thomsen (1959) reported them in terms of an equivalent volume of air with units in milliliters (ml) or cubic centimeters (cc or cm3). This precedent was followed because at low frequencies such as 220 Hz, the ear functions