Acute Otitis Media

Otitis media (OM) is the most common bacterial infection in children and the most frequent indication for antimicrobial or surgical therapy in this age
group. It is also the leading cause of hearing loss in children. The morbidity associated with OM is substantial, and the costs of medical and surgical
therapy for children 5 years of age and younger are estimated at $5 billion annually in the United States.32 The number of operative procedures
performed annually in the United States is estimated at about 600,000.118 Given that prevention of OM by immunization is not widely available and
that 85% of children experience at least one episode of OM, it is fair to say that this disorder is one of the most important, unresolved clinical
problems in otolaryngology. Many factors predispose children to OM, but the impact of these factors is variable. Known risk factors are young age,
male gender, bottle feeding, crowded living conditions (including multichild day care centers), smoking in the home, heredity, and a variety of
associated conditions such as cleft palate, immunodeficiency, ciliary dyskinesia, Down syndrome, and cystic fibrosis. The genetic factors of OM
have yet to be delineated, although it is well known that OM occurs more frequently in certain families.

Treatment of OM varies worldwide, with divergent results from numerous studies and clinical trials. The indications for therapy, particularly surgical
therapy, remain controversial and often empiric. Although the widespread use of antimicrobial agents for OM has reduced the number of acute
complications (most importantly intracranial extension of infection), the number of chronic complications, particularly hearing loss due to persistent
middle-ear effusion (MEE), appears to be increasing. Because of the alarming increase in penicillin-resistant pneumococci,17 attitudes about
antibiotic use are changing. It is fair to say that acute OM (AOM) prophylaxis with long-term use of an antimicrobial agent, which has been the
standard of practice in the United States for the past two decades, currently is discouraged.88 It is likely that surgical therapy will be used more often
in children with severe recurrent OM to reduce the number of new infections and to remove persistent effusion.

This chapter reviews current knowledge about OM based on contemporary clinical research. Emphasis is on treatment in the ambulatory setting
within the context of the opposing forces of societal attitudes about cost containment and the continuing parental pressure for optimal care. As the
chief providers of surgical care for children with OM, otolaryngologists frequently find their recommendations constrained by the policies of
managed-care agents and their advisors who seek to control health-care costs by limiting surgical treatment, even though such therapy has been
proven effective in reducing OM morbidity. Such policies typically examine only the short-term costs and fail to consider the long-term value. When
the long-term impact of OM is considered, surgical therapy is more cost effective than medical therapy for severe cases.32 With the welfare of
patients as our primary concern, surgical therapy is selected from a rational, research-based, conservative protocol for patients for whom medical
therapy fails and for whom there is no other choice for remediation of symptoms and correction of hearing loss.


OM is used herein as a generic term for any inflammatory process in the middle-ear cleft beginning behind an intact tympanic membrane (TM). The
two major classes of OM are AOM and chronic OM with effusion (COME). AOM is typified by the symptoms and signs of acute infection (fever,
pain; a red, bulging TM; and MEE). COME indicates an MEE without pain, redness, or bulging of the TM. Equivalent terms are chronic secretory
OM, chronic serous OM, and ‘‘glue ear.’’ MEE denotes a liquid in the middle-ear cleft regardless of etiology. Specifically excluded from this
chapter are the various types of chronic suppurative OM associated with a permanent perforation of the TM or with cholesteatoma.

That AOM is a bacterial disorder is beyond question. Worldwide, the same pathogens are cultured with remarkably similar frequency from the
middle ears of children with AOM. Table 29-1 presents the results of cultures from children with AOM.10 Viruses have been recovered from the
middle ear in 20% of early cases, in some cases as the sole agent but more often with pathogenic bacteria.65 Thus there is now evidence that
respiratory viruses are important in AOM, both directly in the middle ear and indirectly as a cause of antecedent URI. Because the majority of cases
of AOM have a bacterial etiology, it is logical to account for how the bacteria reach the middle ear. Although the evidence is indirect, it is reasonably
persuasive. First, in usual cases of AOM, viral nasal infection precedes the ear infection.51 Second, pathogenic bacteria subsequently appear and are
found in the nasopharynges of 97% of patients with AOM, with correspondence to the organisms in the MEE in 69%.57 Third, the adenoids of
children with recurrent AOM contain pathogenic bacteria in clinically significant amounts.12,21,96 Stenfors and Raisanen119 noted a significant age
correlation with middle-ear pathogens in the nasopharynges of clinically disease-free children; 57% of those under 2 years of age were culture
positive compared with 40% of the 11- to 15-year-old children. Thus it appears that the adenoid is a bacterial reservoir in the nasopharynx in children
with in AOM.

The most likely route of entry of nasopharyngeal microorganisms into the middle-ear cleft is via the eustachian tube. Reflux from the nasopharynx
into the middle ear during swallowing has been demonstrated radiographically in OM-prone children by Bluestone and others.9 Reflux is probably
facilitated by nose blowing and closed-nose swallowing (i.e., Toynbee’s maneuver)63 or aspiration into the middle ear as a result of negative
middle-ear pressure. One source of negative middle-ear pressure is sniffing.2 Children with patulous eustachian tubes, in whom free flow of
radiographic material into the middle ear occurs, have been shown by Bluestone and others9 to be at high risk for OM. Bluestone, Beery, and
Andrus7 showed that young children have shorter, straighter, and more compliant eustachian tubes than do adults, which is an important factor in the
pathogenesis of OM in children. Sade106 was among the first to identify the open eustachian tube as a factor in AOM.

During swallowing, the adenoid is elevated by the soft palate and, when large, may obstruct the posterior choanae and contribute directly to
increased nasopharyngeal pressure and thus indirectly to reflux. Adenoid enlargement has long been postulated as a factor in OM,91 but if reflux is
indeed a sequela of adenoid enlargement, one would expect a greater prevalence of adenoid enlargement in children with OM. However, the
evidence does not support this assumption. Adenoid size in children with OM did not differ from that of control children either radiographically53 or
by weight.44 No difference was noted in the recurrence rate of effusion in children with large versus small adenoids, nor did the effect of
adenoidectomy depend on adenoid size.40,78,94

The close proximity of the adenoid to the mouth of the eustachian tube has led many to presume the existence of a cause-and-effect relationship of
adenoid-to–eustachian tube function. Under the working hypothesis that bacteria from the adenoid and nasopharynx enter the middle ear via the
eustachian tube, it is unnecessary to postulate eustachian tube obstruction as a necessary precedent for AOM, as has been held since the time of
Politzer.97 Indeed, in the early stages of AOM, bulging of the TM and positive middle-ear pressure are common, yet there are no findings associated
with eustachian tube obstruction. Furthermore, it has never been shown that the adenoid physically obstructs the eustachian tube; in fact, it has been
shown that it does not. Honjo56 studied 52 children with COME and compared eustachian tube function in those with a large adenoid, which
appeared on fiberoptic endoscopy to obstruct the pharyngeal end of the eustachian tube, with those with a clearly open tube. No difference in the
opening pressure or in positive-pressure equalization was noted between the two groups. Further, there was no difference in eustachian tube
ventilation function before and after adenoidectomy. Therefore, Honjo concluded that the adenoid does not mechanically obstruct the eustachian

Takahashi, Fujita, and Honjo120 studied 10 adult patients with COME using a thin-pressure catheter and identified the site of eustachian tube
obstruction in the distal part of the cartilaginous portion, 5 to 15 mm from the orifice, rather than at the orifice proper. Thus the role of eustachian
tube obstruction as a precursor to AOM is called into question, and the locus of obstruction in chronic COME is likely to be within the eustachian
tube rather than at its nasopharyngeal ostium.

The preceding findings suggest that eustachian tube dysfunction, which is clearly demonstrable in these children, may be the result of AOM rather
than the cause of it. Thus the classic theories that the adenoid causes mechanical obstruction of the eustachian tube and that adenoidal enlargement
is a factor in the pathogenesis of OM are not supported by current evidence. Improvement in eustachian tube function, however, does occur after
adenoidectomy.9,56 The exact mechanism for this effect is not clear, although relief from infectious stimulation may play a role (see the section on


The normal TM is gray, concave, and translucent and moves briskly on pneumatic otoscopy. The classic signs of AOM are redness and bulging of
the TM. In the beginning stages of AOM, the eardrum may bulge outward because of positive pressure in the middle ear, and it often moves
normally. As effusion develops, drum mobility is decreased. In severe cases none of the usual landmarks may be visible. If the process continues to
worsen, necrosis of the TM occurs and the effusion passes into the ear canal through a typically pinpoint perforation. Massive necrosis of the
drumhead is rare, although necrotizing streptococcal infection is a known cause of permanent perforation.

In early or mild cases of AOM, considerable judgment must be used to differentiate subtle redness of the drumhead in crying children, whose
struggles can result in vascular congestion of the TM that simulates OM. For this reason, assessment of TM mobility is vital to confirm the presence
of MEE.

The pneumatic otoscope can be used to increase and decrease the pressure in the external auditory canal, and the TM should move inward and
outward in synchrony with the pressure changes. In a healthy ear, pressures as low as 1 to 2 daPa result in detectable motion of the drumhead.35
Use of the operating microscope improves the ability to make these subtle judgments. The best method to train the eye is to examine the ears just
before myringotomy in the preoperative holding area. Comparing preoperatively recorded findings with the amount of MEE found at surgery is the
surest way to develop good otoscopic discrimination for MEE. Most examiners quickly develop good sensitivity for the presence of effusion, but
being confident about the absence of effusion (specificity) takes experience and practice.

A number of clinical variants of AOM are known and detectable otoscopically. Myringitis is an inflammation of the TM without MEE. The etiology
and pathogenesis of myringitis are not well documented. In some cases, myringitis may represent the initial stages of bacterial AOM; in others, it
may be due to a viral agent; in still others it may represent the effects of nose-blowing or sniffing. Because of the uncertainty of the clinical course
of myringitis, most practitioners use the same treatment as for AOM. Bullous myringitis is seen in both adults and children. Early reports suggested
an association with Mycoplasma pneumoniae infection. However, in most cases bullae on the TM are associated with the same pathogenic
bacteria as AOM and should be treated similarly. Pain is an outstanding feature of bullous myringitis and is not relieved by opening the bulla(e). The
classic findings of COME are a retracted, hypomobile or immobile TM and a dark, fluid-filled tympanum that obscures visualization of the long
process of the incus. Negative middle-ear pressure is also associated with a retracted TM, but the membrane moves outward briskly when pressure
in the ear canal is sharply reduced with the pneumatic otoscope. Judging the absence of MEE in the face of a retracted TM is often difficult.


Tympanometry has an important place in the evaluation of children with middle-ear disorders. Ease of use, acceptance by patients, reproducibility of
results, and availability of low-cost machines have led to widespread use of tympanometry in physicians’ offices. By plotting the amount of energy
reflected from the TM as the pressure in the external auditory canal is varied from –400 daPa to 100 daPa, a tympanogram is obtained, the shape of
which provides considerable information about the status of the middle ear. When the tympanum is completely filled or impacted with effusion,
compliance is low and the tympanogram is flat (i.e., does not vary with the pressure change). This shape was labeled type B by Jerger.62 In
air-containing ears the shape of the tympanogram is usually peaked, with the peak at –100 daPa (type A). Negative middle-ear pressure (i.e., below
150 daPa) with a sharp peak is labeled type C. Variants of these tympanogram types have been described for research purposes by Paradise, Smith,
and Bluestone,93 who pointed out that the ear canals of infants (<7 months) are hypercompliant, and normal tympanograms are possible in the
presence of MEE. In these cases, the walls of the canals move with pressure, producing a ‘‘canalogram’’ rather than a tympanogram.

The type A tympanogram, which has a sharp compliance peak at ±100 daPa, suggests an aerated tympanum. Type B curves may be completely flat
or have a very rounded compliance peak, which is often below –300 daPa. The type C tympanogram usually has a sharp peak between –150 and
–200 daPa. Such tracings reflect negative middle-ear pressure and do not indicate effusion per se. When the pressure is below –250 daPa, the
likelihood of MEE increases. Figure 29-1 shows the most common types of tympanometric curves seen in children with OM.
Antimicrobial therapy

Antimicrobial therapy is the mainstay of treatment for AOM in the United States but not worldwide.123 Modern agents are effective orally and are
widely available, safe, and generally well tolerated. Due to the emergence of b-lactamase–producing strains of Haemophilus influenzae and the
increasing role of Branhamella catarrhalis in OM,114 it is important to consider using b-lactamase–resistant agents. In the past, only a minority of
cases of AOM have been caused by b-lactamase–producing organisms, and it has not been cost-effective to use b-lactamase–resistant drugs, which
are generally more expensive, as the first line of management. In many cities, however, the incidence of b-lactamase–producing organisms in AOM
has risen to the point that amoxicillin may not be the best initial therapy.84

Duration of therapy is an important, unresolved issue. A 10-day treatment course is usually prescribed as a compromise between cost and adequacy
of therapy. In children with AOM, comparable results were found by Meistrup-Larsen and others80 after 2 versus 7 days of penicillin therapy and by
Hendrickse and others62 after 5 versus 10 days of cefaclor. The latter study noted a significantly higher failure rate with the 5-day treatment in
children whose AOM resulted in a perforation; the authors recommend a full 10-day course for these more severe cases.

In communities where b-lactamase organisms are uncommon, amoxicillin is the first-line drug of choice in nonallergic patients.
Trimethoprim–sulfamethoxazole and the combination of erythromycin estolate and sulfisoxazole (Pediazole; Ross Products Division of Abbott
Laboratories, Columbus, Ohio) are appropriate alternatives except that the sulfas are not effective against the rare case of b-hemolytic
streptococci.131 If the inflammation does not show signs of response, the initial agent is discontinued and a second-line drug (e.g., cefaclor or
amoxicillin–clavulanate) is used. Judgment and caution must be exercised if a short treatment course is chosen to ensure that the middle ear has
been sterilized and that the infection is controlled. A 10-day course of treatment is generally used to ensure adequate therapy. Many cases would
undoubtedly do well with a shorter course, but ensuring adequacy would increase costs because of the need for an interim physician visit. It is not
possible with our current technology to predict which cases would do well with a shorter course of treatment. However, patients with severe AOM,
especially those with spontaneous perforation, require full, if not prolonged, treatment.

Adjunct medical therapy

Because of the frequent association of OM and URI, use of a nasal decongestant to open the airway may be helpful, but antihistamines should be
avoided because their atropine-like effect leads to drying of secretions. The combination of an antihistamine and decongestant was found not to
affect the clearance of MEE.14 Topical decongestants may be used for short periods, but prolonged use may actually worsen rather than improve
the nasal airway because of rebound and sensitization of the nasal mucosa. A topical parasympathetic blocking agent that eliminates the problems
associated with sympathetic agonists is now available in the United States. However, the utility of this treatment in children with sinusitis and OM
remains to be explored.


Knowledge of the specific organism is important for selection of therapy in cases of AOM occurring in premature newborns, immunocompromised
patients, patients with progression of symptoms and signs while receiving an appropriate antimicrobial, patients with intracranial infection, and
research subjects. The procedure is performed with an 18-gauge spinal needle attached to a 1-ml tuberculin syringe. No anesthetic is necessary. The
needle is inserted into the anteroinferior quadrant of the TM, and the fluid is aspirated into the needle (Figure 29-2). The fluid is then flushed out of
the needle onto an agar plate containing appropriate media for middle-ear organisms. The needle hole closes too quickly for any significant drainage
to occur, although the aspiration does temporarily decompress the middle ear.


Although severe AOM has many of the clinical features of a closed-space abscess, incision and drainage (myringotomy) has proven to be of limited
value. Myringotomy promptly relieves severe pain due to AOM in patients with severe pain101 but adds little to either remission of infection or
clearance of MEE in cases of AOM treated with amoxicillin–clavulanate.25 In some European countries, myringotomy has been the mainstay of
treatment of AOM,22 with antibiotics being reserved for cases with persistent drainage. van Buchem, Dunk, and Van’t Hoff130 have shown no
difference in outcome of children with AOM whether antibiotic, myringotomy, both, or neither was used. However, the large number of
methodologic flaws and the small number of subjects in each management group created substantial problems in interpretation. Although it appears
that myringotomy does not offer sufficient benefit in the average case of AOM to justify its costs or risks, in complicated cases the procedure should
be performed without delay to provide prompt drainage and material for culture. Such cases include AOM with facial paralysis, meningitis, or other
central nervous system event; AOM not responding to antimicrobial therapy; and AOM in immunosuppressed individuals.


Follow-up examination is important for two reasons: to ensure that the infection is responding to the antimicrobial and to determine that the MEE has
resolved. In infants and young children, follow-up is also important to exclude meningitis. In most of these cases, AOM and meningitis coexist rather
than have a cause-and-effect relationship, but the signs of meningitis in infants may be subtle at first. Therefore, a 3-day check is often performed in
infants for this reason and to determine whether the child is responding to the drug. A 2-week check is often performed to determine whether the
MEE has cleared. The natural history of AOM indicates that only about half of the ears will have cleared at this point. Unfortunately, this is often
taken as an indication for a second round of antimicrobial therapy, and it is not unusual to see asymptomatic children who have received four
different antimicrobials for a single episode of AOM. There is little evidence to suggest that prolonged or repeated therapy is effective in the routine
case of AOM. If the TM has ruptured, indicating a severe episode, it is prudent to continue the antimicrobial agent until drainage has ceased and the
membrane has sealed.

Recurrent acute otitis media

Antimicrobial prophylaxis

Antimicrobial agents are effective in preventing AOM but they should not be used routinely because of the risk of promoting antimicrobial resistance
in common pathogens. Several clinical trials have documented the efficacy of chemoprevention of AOM in children with frequently recurring ear
infections95,98a,131 Because of the risk of hypersensitivity reactions to the long-term use of sulfonamides, many practitioners use a single daily dose
of amoxicillin (20 mg/kg) for 3 to 6 months to prevent recurrent AOM. Many use the fixed combination of trimethoprim–sulfamethoxazole even
though the manufacturer’s recommendations specifically prescribe against prolonged use for AOM prophylaxis. None of the studies has addressed
the long-term consequences.

Barnett and Klein4 expressed concern about the development of antimicrobial resistance in the common AOM pathogens from long-term antibiotic
use, such as for AOM prophylaxis. Recent studies have compared modified regimens for prophylaxis. In a randomized, placebo-controlled study of
76 otitis-prone children 18 months of age and younger, Prellner and others98 found that antimicrobial coverage with penicillin V during URIs reduced
the number of episodes of recurrent acute OM with effusion (rAOME) by 50%. Berman and others5 compared amoxicillin given twice daily for up
to 4 months with intermittent amoxicillin during URI in 55 otitis-prone children. The continuous group (N = 30) had significantly fewer episodes of
recurrent acute OM with effusion (eight) than did the intermittent prophylaxis group (N = 25, 12 episodes). Due to the worldwide emergence of
resistant organisms, long-term use of antimicrobial prophylaxis is being questioned. Paradise88 has argued against prophylaxis in most cases and
earlier referral for tympanostomy tubes. Administration of an antibiotic in the early stages of URI reduces the number of new infections.98
Short-term antimicrobial prophylaxis during the winter URI season has a long record of efficacy and is still being recommended in severely affected

Surgical prophylaxis

Surgical prophylaxis is being increasingly recommended for children with repeated episodes of AOM.88 The efficacy of surgical prophylaxis has
been established. Using a randomized, unblinded protocol, Gebhart43 was the first to quantify the effectiveness of tympanostomy tube insertion in
preventing recurrent AOM. Others showed a significant decrease in COME in patients undergoing tympanostomy tube placement versus placebo,
but there was no difference between tympanostomy tubes and medical therapy. Casselbrant and others16 demonstrated that tympanostomy tubes
and antibiotic prophylaxis decreased the frequency of COME as compared with placebo but did not significantly reduce the incidence of recurrent
acute OM with effusion. Le, Freeman, and Fireman67a compared unilateral tympanostomy tubes with unilateral myringotomy in treating 44 children
with recurrent acute OM with effusion and 13 children with COME. The tubed ear had less infection while the tube was present, but the infection
rate was the same for both ears after tube extrusion. Other studies5a,73 in older children specifically addressed the impact of tympanostomy tubes on

Adenoidectomy should logically also be an effective preventive treatment against recurrent AOM, but this hypothesis has not been adequately
tested. In the San Antonio study,40 the rate of new episodes of AOM in the two adenoidectomy groups did not differ from that of the two
nonadenoidectomy groups.40 However, these were older children in whom the rate of recurrent AOM is not great. In the Pittsburgh study,94 there
was a significant reduction in the number of episodes of AOM in each of the 2 years after adenoidectomy. These reports provide preliminary data
about the efficacy of adenoidectomy in preventing AOM because this was not the primary research question. A clinical study of this question would
be both appropriate and timely. Because logically there is little age difference in the outcome of surgical treatment, however, many otolaryngologists
use adenoidectomy in the younger age group. The omission of adenoidectomy from the AHCPR recommendations was based on an age technicality
that ignored the evidence of Paradise and others.94



Photo Gallery


Radiology Gallery


Video Gallery


Internet Gallery


Back Home Next