Objective diagnosis of middle ear effusion by acoustic reflectometry

August 2004

Regular use of acoustic reflectometry, tympanometry (discussed in a previous column [October 2003]) or both should improve otoscopic diagnosis of middle-ear effusion or acute otitis media (AOM) by primary care residents, primary care pediatricians and family practice physicians. This article will review the advantages and disadvantages of acoustic reflectometry and the unique EarCheck instruments (MDI Instruments, Chester, N.J.), which are sensitive to middle-ear effusion and measure the spectrum of reflected sound. Acoustic reflectometry uses the same CPT code (92567) as impedance testing (tympanometry). Until each physician or nurse practitioner demonstrates high concordance between results of pneumatic otoscopy and acoustic reflectometry and/or tympanometry, it is suggested that they perform both procedures in tandem whenever diagnosis of middle-ear health is uncertain.

Acoustic otoscope

Acoustic pulse reflectometry was originally developed as a technique for observation of stratifications in the earth’s crust. The first commercially available instrument to measure acoustic reflectometry was produced and marketed more than 20 years ago. This Acoustic Otoscope contained an acoustic transducer that emitted a curious chirping sound, actually a subtle series of discrete sound frequencies. With the Acoustic Otoscope, it was possible to screen both ears in a cooperative child in about 15 seconds and often in less than one or two minutes for crying young children. Although several clinical studies of children with middle-ear pathology praised the technique, in actual clinical practice, the Acoustic Otoscope proved to be too dependent on user technique to be acceptable.

EarCheck models

The current acoustic reflectometers include a professional model, the EarCheck Pro Otitis Media Detector, priced at $360.00 with storage base and 20 disposable probe tips, and a less expensive screening model for parents called the EarCheck Middle Ear Monitor, priced at $50.00. Two AA alkaline batteries power either model, and low battery power prompts a warning liquid crystal display (LCD). Both models use the same acoustic physics and a similar chirping sound as the original Acoustic Otoscope. Children must be examined in an upright sitting or erect position, not in the supine position. Acoustic reflectometry is not accurate for children younger than 6 months, those with ventilation tubes, large cerumen accumulations, water or otorrhea in the ear canal or stenotic ear canals. Excessive ambient noise may also produce erroneous results. A recalcitrant toddler may be more receptive to the procedure if mock scans of mom’s ears and/or those of a doll or teddy bear are performed first. Asking the child to “listen to the funny-sounding frog singing” often encourages cooperation with spectral gradient acoustic reflectometry (SGAR).

The proper technique is as follows: The disposable plastic tip is placed adjacent to the external auditory meatus. The outer edge of the child’s auricle is gently pulled up and out to maximally straighten the ear canal. The 10-mm aperture probe tip of the instrument should overlay the external meatus and point toward the tympanic membrane (TM). The corresponding left or right ear “scan” button on the handle of the instrument is pushed in and held for 10 seconds. The scan button must be released before removal of the EarCheck instrument from the child’s ear. For each child tested, several different ear tug angles should be tried with each ear until the LCD displays the most acute angle. If the results of the EarCheck Pro are not concordant with results of pneumatic otoscopy, both procedures should be repeated. Errors in displaying a valid result are easily corrected by repeating the test procedure with each ear as many times as necessary.

Forty-four discrete frequencies in the audible range of 1.8 to 4.4 kHz constitute the 80 dB chirping sound. Waves of this chirping sound hit the TM, and some of the sound energy is reflected backward, thus the term acoustic reflectometry. By means of a microphone and microprocessor, the hand-held instrument records and analyzes the intensity of the reflected echo waveform in degrees of spectral gradient. In the normal middle ear, about 50% of the induced chirps are reflected back toward the microphone of the instrument. When there is inflammatory or infectious liquid in the middle ear space, the amount of sound energy reflected back from the tympanic membrane is increased and the corresponding spectral gradient angle decreased. Thus, the higher the probability of middle ear effusion, the smaller the spectral angle recorded. In the Pro model, acute angles less than 60° correlate well with the presence of middle ear effusion. Angles greater than 90° correlate well with a well-aerated middle ear. However, predictability of spectral angles between 76° and 90° is fair to poor, and this is a weakness of the technique. The professional model gives a direct reading of the sum total spectral angle as a direct readout or by connecting the handheld instrument to an optional but inexpensive printer. The parent model does not give a spectral angle; it simply generates a number from 1 to 5, where number 4 or 5 indicates high probability of middle-ear effusion. Recordings of 4 or 5 are usually abnormal (approximately 8% false positive) and closely correlate with middle-ear effusion or exudates. It has been demonstrated that a recording of 4 or 5 will also occur from ears with negative middle ear pressure without middle ear effusion (false positive). The intervening numbers 2 or 3 are difficult to interpret with accuracy and represent a normal middle-ear condition (particularly in children younger than 2 years old), negative middle-ear pressure or partial middle-ear effusion.

This technology is FDA-cleared for pediatric patients older than 6 months. A visit to www.earcheck.com will give more information about the device, which can be directly ordered on www.cvs.com, BabyCenter.com and other Web sites.

Advantages

The inexpensive (compared with tympanometry instruments) handheld EarCheck devices weigh only about 6 oz and are simple to learn how to use. SGAR is quicker and easier to use than tympanometry, particularly with children younger than 2 years. Failures to obtain readings because of lack of patient cooperation are much more common with tympanometry than with SGAR, and the results are more easily aborted during the testing with tympanometry. EarCheck does not require the operator to obtain an airtight seal between the probe tip and the ear canal opening, as tympanometry does. Also, unlike tympanometry, the EarCheck causes no discomfort when the probe tip is in place against the ear canal opening. With tympanometry but not with acoustic reflectometry, there is additional discomfort from the automatic variation of pressure from +200 to –400 in the sealed ear canal. Babies sometimes cry from this pressure variance.

About 10% of the time, an ear diagnosed as having AOM will produce a normal-looking tympanometric tracing. This results when there is a partial effusion and normal or minimally abnormal pressure in the middle ear. The SGAR parent model uses a numerical system of 1-5, where 1 is highly correlated with the presence of middle ear effusion. While 2 is of good positive and negative predictive value for middle ear effusion, 3 is too indeterminate to be of any help to predict effusion. SGAR may have an LCD readout of 3 or 4 in such cases.

Disadvantages

Compared with tympanometry, one disadvantage of the EarCheck Pro is the somewhat reduced sensitivity to detect middle-ear effusion, because effusions detected with certainty by pneumatic otoscopy and/or tympanometry can sometimes have a falsely normal obtuse spectral angle with the EarCheck Pro. Reliable acoustic reflectometry readings as well as tympanometry are not possible for most children younger than 6 months or children of any age with small ear canal diameters, such as children with Down syndrome. Earwax that occupies more than 50% of the ear canal diameter may produce inaccurate results. Tympanometry with the addition of the physical volume determination but not SGAR can give information about the presence of a perforation in the TM or the functional status of a pressure equalization tube. Irrigating the ear canal with water to remove excess cerumen will not affect tympanometry; however, SGAR is so sensitive to fluid that it may give a falsely high reading after canal irrigation.

Discussion

Tympanometry and SGAR are different objective technologies for confirming the diagnosis of middle-ear effusion or AOM. The two techniques complement each other and improve accuracy and precision. Tympanometry is a little more sensitive and possibly more specific than SGAR. When there is discordance between the results of pneumatic otoscopy and acoustic reflectometry, the addition of tympanometry provides improved diagnostic accuracy.

Any pediatrician interested in acquiring experience with acoustic reflectometry might first consider investing in the consumer device. The consumer device and the professional model use the same technology and give similar results. With a consumer model in the office, pediatricians can also teach parents how to use it. Neither tympanometry nor SGAR are tests for hearing acuity. While both technologies are acceptably sensitive and specific for effusion, neither gives any information as to whether the effusion is purulent, serous or mucoid. Differentiating between AOM and otitis media with effusion requires the physician’s clinical acumen and the presence of a bulging, opacified and poorly mobile TM.

Acoustic reflectometry or tympanometry does not replace skillful otoscopic examination, because these techniques are adjunctive and objective. Physicians skilled in pneumatic otoscopy should find that these techniques improve the quality of their otoscopic clinical examination.

All well-equipped pediatric offices and pediatric and family medicine training programs should have tympanometry and acoustic reflectometry easily available. Both techniques should be used frequently in the office and in the clinic.

For more information:

• Combs JT, Schuman AJ. Three technologies for taming otitis media. Contemp Pediatr. 1999;16:78-101.
• Kemololu YK, Sener T, Beder L, et al. Predictive value of acoustic reflectometry (angle and reflectivity) and tympanometry. Internat J Pediatr Otorhinolaryngol. 1999;48:137-142.
• Block SL, Pichichero ME, McLinn S et al. Spectral gradient acoustic reflectometry: detection of middle ear effusion in suppurative otitis media. Pediatr Infect Dis J. 1999;18:741-744.
• Barnett ED, Klein JO, Hawkins KA, et al. Comparison of spectral gradient acoustic reflectometry and other diagnostic techniques for detection of middle ear effusion in children with middle ear disease. Pediatr Infect Dis J. 1998; 17:556-559.
• Block SL, Mandel E, McLinn S, et al. Spectral gradient acoustic reflectometry for the detection of middle ear effusion by pediatricians and parents. Pediatr Infect Dis J. 1998; 17:560-564.
• Kemaloglu YK, Beder L, Sener T, Goksu N. Tympanometry and acoustic reflectometry in ears with chronic retraction without effusion. Int J Pediatr Otorhinolaryngol. 2000;15:21-27.

• Richard H. Schwartz, MD, is from the department of pediatrics at Inova Fairfax Hospital for Children, Falls Church, Va.