My co-workers and I have been testing the diagnostic utility of laser-Doppler vibrometry (LDV) measurements of the sound-induced velocity of the eardrum in patients with conductive hearing loss (Rosowski et al., 2003; Whittemore et al., 2004). LDV works by measuring the Doppler-shift in laser light reflected off of moving objects. Hearing researchers have successfully used LDV to measure the sound-induced velocity of ossicles and eardrums in live animals for about 25 years. In the last 10 years, LDV devices have become sensitive enough to make measurements of the motion of the eardrum of human subjects and patients practical. Figure 1 [PDF] illustrates an observer using a microscope-mounted laser while looking at the tympanic membrane through a glass-backed sound coupler. The coupler contains a probe microphone and a sound source. The observer focuses the laser on a reflective area near the center of the tympanic membrane. The light reflected back from the bright spot is captured by the laser system; the frequency of the reflected light depends on the velocity of vibration of the tympanic membrane.
Application to Ossicular Disorders
Three research groups-one at Stanford, CA (Goode et al., 1996), a second at the University of Zurich in Switzerland (Huber et al., 2001), and our group in Boston-have pioneered the use of LDV as a diagnostic device by making measurements in large populations of normal listeners and selected patients. This work has concentrated on defining the normal sound-induced eardrum velocity, and in investigating how various ossicular disorders affect the velocity response. The hope is to better diagnose the cause of conductive hearing loss prior to middle ear surgery.
The reasons behind this hope are similar to those used in the past to suggest that low-frequency or multi-frequency tympanometry could help diagnose ossicular disorders. Those older technologies measured the sound-induced motion of the entire eardrum and were greatly affected by motion of parts of the eardrum that are not well coupled to the ossicular chain. This lack of spatial selectivity contributed to the general lack of sensitivity of tympanometry to pathological alterations in the ossicular chain such as stapes fixation or ossicular fractures.
It is hoped that the higher spatial sensitivity of LDV-it measures the velocity at a point on the eardrum that is tightly coupled to the ossicular chain-will yield a superior measure of pathological changes in ossicular mobility. Such changes can result from either ossicular fixations (e.g., stapes fixation produced by otosclerosis) or interruptions of the ossicular chain that can be caused by breakdown of the ossicular bones or trauma.
Our preliminary results suggest that LDV-in combination with pure-tone audiometry-can discriminate ossicular fixations of different types and can also discriminate fixations from ossicular interruptions. Figure 2 [PDF] shows the range of velocities measured with a 500 Hz tone in normal ears and the velocities measured in 41 cases of surgically confirmed stapes fixation, six cases of surgically confirmed malleus-head fixations, and eight cases of confirmed total and partial ossicular-chain interruption. Although there is some overlap of the velocities measured with the different pathologies, the distributions are quite different. These preliminary results are promising; however, more measurements, particularly in the fixed-malleus and ossicular-chain interrupted groups, are required to better quantify the specificity and selectivity of LDV in diagnosing different forms of conductive hearing loss.