Acute otitis media: red eardrum, bulging eardrum, or neither?

June 2001

---Richard H. Schwartz, MD

I want to review the diagnostic signs of acute otitis media (AOM) and present the results of a national survey of 82 pediatric infectious disease (ID) specialists and 58 pediatric otolaryngology specialists, most in academic medicine. Many experts in pediatric ID and pediatric otolaryngology agree there should be precise criteria to accurately diagnose AOM and reduce unnecessary prescriptions for antibiotics prescribed for red eardrums. Some experts insist on pain or fussiness in addition to bulging, opacification, and limited mobility of the tympanic membrane. However, there are data that conclude that AOM may be present without pain. In my opinion, symptoms of pain or fever may be helpful to decide if antibiotic treatment is warranted, but need not be present to diagnose AOM.

Pneumatic otoscopy

When the tympanic membrane is in neutral position and there is air in the middle ear space, the tympanic membrane moves about 1 mm inward with positive pressure and 1 mm outward (toward the examiner) with application of negative pressure. It appears that many pediatricians who use the pneumatic otoscope only apply positive pressure on the tympanic membrane. Immobility of the eardrum when only positive pressure is introduced by bulb or mouth is not diagnostic of middle ear effusion. When there is excessive negative pressure in the middle ear cleft, the tympanic membrane is drawn maximally inward toward the promontory of the temporal bone. If only positive pressure is applied through the pneumatic otoscope on the retracted tympanic membrane, further inward movement cannot take place. This is not really an immobile tympanic membrane. Proper pneumatic otoscopy involves a biphasic change in air pressure in the sealed ear canal. What does this mean? Only when both negative and positive pressure is repetitively applied (3 to 4 times each), can the complete extent of tympanic membrane mobility be ascertained.

According to Siegle, the inventor of the pneumatic otoscope, initially there should be induced rarification (application of negative pressure) in the ear canal. When the tympanic membrane is retracted inward toward the middle ear, application of positive pressure on the tympanic membrane (by squeezing the rubber bulb of the pneumatic otoscope attachment, or by positive pressure by blowing into the mouthpiece), cannot move the membrane any further medially toward the middle ear because of negative middle ear pressure. On the other hand, application of negative pressure by releasing the partially squeezed bulb or by gentle suction on the mouthpiece of the pneumatic otoscope, which is the method described by Siegle, would cause the tympanic membrane to briskly move laterally toward the examiner. In other words, tympanic membrane immobility cannot only be after application of positive pressure. It must be assessed after positive and negative pressure with the pneumatic otoscope.

Criteria for AOM

The normal tympanic membrane is a translucent oval-shaped membrane approximately 10 mm in diameter, separating the ear canal from the middle ear cleft. With adequate illumination, removal of almost all cerumen and epithelial debris, and a still child, easily visible landmarks on or through the tympanic membrane include the cone of light, umbo, manubrium and lateral process of the malleus, the inco-stapedial junction, and the pars flaccida enclosed by 2 malleolar folds.

There is almost universal agreement that the best description of the tympanic membrane signs of AOM includes 3 criteria: bulging tympanic membrane, opacification of the tympanic membrane and immobility of the tympanic membrane. Usually, the area of the pars flaccida is the earliest to show bulging since the middle radial fibers are absent in this area. Application of positive pressure will often cause a bulging tympanic membrane to blanch and indent a bit. Careful observation may reveal that the indentation of the blanched bulging tympanic membrane may suddenly bring into view the hidden lateral malleolar process. Upon release of positive pressure, the blanched area returns to its former color.

Hayden reported on the variation of clinical criteria for AOM by practicing primary care physicians. The younger the child, the smaller the ear canal diameter, the greater the amount of cerumen, and the more the head moves, the greater the uncertainty regarding the diagnosis of AOM.

During my more than 30 years of pediatric experience, the “au courant” definition of AOM with an intact tympanic membrane evolved from redness of the tympanic membrane plus blurring or disappearance of the light reflex on the tympanic membrane; then to redness of the tympanic membrane plus limitation of mobility on pneumatic-otoscopy; and finally to 1 or more cardinal symptoms of pain, fussiness, or fever plus either redness, limited mobility, or bulging. Without skilled use of the pneumatic otoscope, some pediatricians and many pediatric residents remain locked in a time warp using 30-year-old definitions proven by tympanocentesis to be invalid. Many pediatricians also continue to believe that AOM is defined as whatever their viewing eye thinks it should look like, as long as the tympanic membrane is reddened. Other pediatricians insist on immobility of the reddened tympanic membrane in a symptomatic child. Thus, there continues to be a broad definition of AOM with much inter-observer disagreement.

Almost 20 years ago, 2 pediatricians and 2 pediatric otolaryngologists proposed that the criteria for AOM with an intact eardrum should be accurate and precise and include 3 cardinal signs (bulging of the tympanic membrane, opacification, including color change to red, yellow, or white, and limitation of mobility), in addition to symptoms of localized or generalized pain. The most important and consistent sign of AOM is bulging of 1 or both tympanic membrane(s). Using the gold standard of tympanocentesis plus recovery of an accepted middle ear bacterial pathogen, bulging or fullness is the most important sign that correlates with positive bacterial cultures. It is not simply that bulging is the best correlate for recovery of a bacterial pathogen from the middle ear; it is present in 90% of cases of AOM.

Total bulging is defined as a convex-appearing tympanic membrane with loss of visualization of the lateral process and/or the manubrium of the malleus bone. Often, there is a diagonal cleft in the bulging tympanic membrane where its fibers are tightly adherent to the handle of the malleus. The shape has been likened to a bagel without a central hole. Partial bulging is fullness of an opacified, convex tympanic membrane with preservation of the outline of either the manubrium or the lateral process of the malleus bone. Mobility is impaired during negative and positive pressure. U.S. investigators who frequently perform tympanocentesis and closely follow the results of middle ear cultures, are able to show that .90% of cultures taken from symptomatic bulging tympanic membrane’s contain bacterial pathogens and 6% contain only viral pathogens. All pediatricians were taught that 30% of tympanocentesis-obtained cultures of the middle ear are sterile. However, for most of these studies, bulging of the tympanic membrane was not a prerequisite, but an option for the diagnosis of AOM. AOM, according to study protocols, is defined by a reddened, opacified, immobile tympanic membrane with or without bulging.

Exceptions to the rule

Two exceptions to the requirement for bulging of the tympanic membrane account for an estimated 5% to 9% of cases of AOM in children who do not have a bulging tympanic membrane. The first exception is the acutely draining ear. Acute otorrhea (acute spontaneous drainage of purulent material out of the ear canal through a tympanostomy tube or an acute tympanic membrane perforation), associated with otalgia, crying, or frequently holding the ear prior to rupture of the intact tympanic membrane is highly predictive of bacterial acute otitis media. It is highly probable that the tympanic membrane had been bulging immediately prior to the spontaneous perforation. The second and less common exception to the requirement for bulging of the tympanic membrane is the appearance of a semicircular shaped accumulation of what appears to be yellow purulent material adjacent to the margin of the tympanic membrane. Tympanocentesis through this area of the eardrum will usually be productive of purulent fluid that contains typical otitic bacterial pathogens (observations by Stan Block).

Although diffuse tympanic membrane redness alone is said to be an early sign of AOM, there is no real gold standard to prove this. Redness of the tympanic membrane is non-specific as a sign of anything. Acceptance of this outdated definition of AOM based on the color of the tympanic membrane alone encourages imprecision in diagnosis and injudicious use of antibiotics. Many investigators now prefer the term opacification to the color of the tympanic membrane. Indeed, the primary color of the tympanic membrane in AOM is yellow, with a background of faint hyperemia, similar to the color of neonatal jaundice. In precisely defined AOM, the tympanic membrane may appear pink, red, hemorrhagic red-purple, yellow, serum-colored, off-white, or mixtures of the above colors. Check this out yourself by recording the primary color of the next 25 tympanic membranes that are bulging and opacified.

Immobility of the tympanic membrane need not be absolute. Total resistance to the application of positive pressure into the ear canal depends on the amount and consistency of liquid in the middle ear and the amount of positive pressure applied. Often when the tympanic membrane is completely bulging and contains thin purulent material, the tympanic membrane will indent and blanch with application of positive pressure through the pneumatic otoscope. Sometimes the hidden lateral malleolar process will suddenly come into view when the indented tympanic membrane presses on the malleus bone.

OM with effusion

Children with mucoid otitis media with effusion (OME) or secretory otitis media sometimes complain of acute fullness of their ear. They may tug or touch their auricle when the middle ear feels clogged up. Objectively there is an opacified and poorly mobile tympanic membrane in the neutral or retracted position. This is not AOM, even if symptoms of pain or fussiness are present. Tympanocentesis through such tympanic membranes will usually yield a large percentage of thick mucoid fluid and a lab report of “no pathogen obtained, or small numbers of colonies of Haemophilus influenzae or Moraxella catarrhalis.” The major pathogen in AOM, Streptococcus pneumoniae, is infrequently recovered when tympanocentesis is performed for this condition.

At least 75% of my national survey of 82 pediatric infectious disease and 58 pediatric otolaryngology specialists were heads of academic sections or departments, or nationally recognized and well-published leaders in their field. The simple survey asked only three questions:

1. Is redness of the tympanic membrane necessary for the diagnosis of AOM?
2. Is bulging of the tympanic membrane necessary for the diagnosis of AOM?
3. List the most important necessary signs of AOM.

Surveys were sent by fax and by regular mail with stamped, return address envelopes enclosed. More than 60% of each group returned completed surveys.

On the question of the necessity of tympanic membrane redness as a sine qua non for AOM, only 35% of 82 pediatric ID specialists and 31% of 58 pediatric otolaryngologists agreed. Fifty-seven percent of the pediatric ID specialists and 63% of the otolaryngologists were of the opinion that tympanic membrane redness was not essential for that diagnosis. The remainder selected “uncertain.” These responses were remarkably similar between the two groups surveyed.

On the requirement of tympanic membrane bulging as essential for the diagnosis of AOM, 57% of the ID respondents and 34% of the otolaryngologists agreed with the question. Thirty-nine percent of the 82 pediatric ID respondents and 59% of the 58 pediatric otolaryngologists respondents disagreed with the question about bulging of the tympanic membrane being essential for the diagnosis of AOM. The remainder noted that they were uncertain about the answer to that question (P=0.013, Fisher’s Exact, undecided doctors were excluded). What about the necessity for symptoms of fever or pain? The ID and pediatric ENT specialists were split 50-50 on this issue. Of interest, while 63% of ID respondents required immobility of the tympanic membrane as a necessary sign of AOM, only 34% of ENT respondents did so. Perhaps the availability of otomicroscopy at the outpatient ENT clinic precluded the need for pneumatic otoscopy to access the mobility of the tympanic membrane.

In a soon to be released consensus statement from an expert panel chosen by Agency for Health Care Quality and Research (AHCQR, previously known as AHCPR), AOM is best defined by a choice of signs and symptoms. Pain or irritability and/or fever, plus opacification of the tympanic membrane, fullness or bulging of the tympanic membrane, or hearing loss will soon define AOM by AHCPR criteria. Neither redness of the tympanic membrane nor bulging of the tympanic membrane is sine qua non, according to this definition.

Most pediatric ID respondents to the survey believe that bulging should be necessary for the diagnosis of AOM. Only about one-third of pediatric otolaryngologists believe that bulging of the tympanic membrane is necessary. It is clear that precision and standardization for the diagnosis of AOM is necessary for pediatric residents, primary care pediatricians and emergency medicine physicians.

Pictured here are several bulging opacified eardrums characteristic of acute otitis media. Figure 4 shows a gray eardrum; Figure 7 shows a yellow eardrum with dilated blood vessels. The diagnosis is AOM in each case.

For more information:

• Richard H. Schwartz, MD, is from the department of pediatrics at Inova Fairfax Hospital for Children, Vienna, Va.
• Rodriguez WJ, Schwartz RH. Streptococcus pneumoniae causes otitis media with higher fever and more redness of tympanic membranes than Haemophilus influenzae or Moraxella catarrhalis. Pediatric Infect Dis J. 1999;18:942-944.
• Del Deccaro MA, Mendelman PM, Inglis AF, et al. Bacteriology of acute otitis media: A new prospective. J Pediatr. 1992;120:81-4
• Karma PH, Penttila MA, Markku MS, Kataja MJ. Otoscopic diagnosis of middle ear effusion in acute and non-acute otitis media: The value of different otoscopic findings. Int J Pediatr Otorhinolaryngol. 1989;17:37-49.
• Halstead C, Leprow ML, Balassanian N, et al. Otitis Media. Am J Dis Child. 1968;115:542-551.