Why does the Carhart notch appear at 2000Hz in otosclerotic patients and at no other frequency?
An audiometric finding characteristic of otosclerosis is an increase in bone conduction (BC) threshold with a peak at 2,000 Hz known as Carhart's notch (Carhart, 1950). Although the notch occurs at 2,000 Hz, a reduction in bone conduction sensitivity is seen from 500 to 4,000 Hz which is, on average, 5 dB at 500 Hz, 10 dB at 1000 Hz, 15 dB at 2000 Hz, and 5 dB at 4,000 Hz (Carhart, 1971). Carhart attributed this phenomenon to "mechanical factors associated with stapedial fixation." The Carhart notch is not a true indication of "cochlear reserve" and this BC loss may be corrected by successful stapes surgery (Tonndorf, 1971). Carhart (1971) and Tonndorf (1971) provided a review of several theories to explain this phenomenon. Further, Tonndorf (1971) provided a summary of the results of his studies of stapes fixation in various mammals (cats, dogs, guinea pigs, rats, and humans). Tonndorf (1971) explained that the magnitude of the Carhart notch depended on the extent the middle ear contributed to the total BC response in each of the species tested. Further, he explained the frequency of the notch varied depending on the natural (resonance) frequency of the ossicular chain for BC signals.
It has been well established that the resonance frequency of the middle ear for air conducted (AC) stimulation is around 800-1200 Hz (e.g., Margolis et al., 1985). So why is the Carhart notch seen at 2,000 Hz? The answer is because the primary resonance of the ossicular chain for BC signals differs from the primary resonance of the chain for AC signals and is about 1600-1700 Hz (Zwislocki, 1957). Thus, for any BC disorder that decreases the mobility of the ossicular chain, the 1600-1700 Hz frequency range is affected most prominently. The reason that the Carhart notch is seen audiometrically at 2000 Hz is because 2000 Hz is the closest frequency tested in a typical hearing threshold measurement.
Recently Homma and colleagues (2009) published a fascinating study on ossicular chain resonance based on the differences in response to AC and BC stimulation of five human temporal bones. Their research offers additional clarity on the issue of vibration modes of the middle ear. According to their data, the ossicular chain has two basic modes of vibration. The first mode, with a peak around 1200 Hz, is the primary mode for AC stimulation. This mode is associated with a "hinging" motion of the ossicles caused by AC stimulation of the tympanic membrane at the umbo. The second mode, with a peak around 1700 Hz, is described as a "pivoting" motion of the malleus/incus, with an axis of rotation somewhat orthogonal to the axis of rotation associated with the "hinging" motion. The second mode is less robust than the primary mode for AC stimulation, but it is the dominant mode when excited by BC stimulation. A decreased mobility of the ossicular chain at 1700 Hz due to otosclerosis also affects the surrounding frequencies, but is seen most prominently as a BC loss at 2000 Hz in audiometric testing.
Carhart, R. (1950). Clinical application of bone conduction audiometry. Archives of Otolaryngology, 51, 798-808.
Carhart, R. (1971). Effects of stapes fixation on bone-conduction response. In I.M. Ventry, J.B. Chailkin, & R.F. Dixon (Eds.), Hearing measurement: A book of readings (pp. 116-129). New York, NY: Appleton-Century-Crofts.
Homma, K., Du, Y., Shizmu, Y., & Puria, S. (2009). Ossicular resonance modes of the human middle ear for bone and air conduction. Journal of the Acoustical Society of America, 125, 968-979.
Margolis, R. H., Van Camp, J., Wilson, R.H., & Creten, W.L. (1985). Multifrequency tympanometry in normal ears. Audiology, 24, 44-53.
Tonndorf, J. (1971). Animal experiments in bone conduction: Clinical conclusions. In I.M. Ventry, J.B. Chaiklin, & R.F. Dixon (Eds.), Hearing measurement: A book of readings (pp. 130-141). New York, NY: Appleton-Century-Crofts.
Dr. Diana Emanuel is a professor and the Audiology Graduate Program Director at Towson University. She can be reached at firstname.lastname@example.org.
Dr. Emanuel would like to acknowledge the thoughtful criticism and suggestions of Dr. Tomasz Letowski on the original posting of this ATE in 2002;his suggestions led to this revised answer.