INTRODUCTION

Acute otitis media (AOM) is defined by moderate to severe bulging (picture 1 and picture 2) of the tympanic
membrane or new onset of otorrhea not due to acute otitis externa accompanied by acute signs of illness and signs
or symptoms of middle ear inflammation [1]. (See "Acute otitis media in children: Clinical manifestations and
diagnosis", section on 'Diagnosis'.)

The risk factors, microbiology, clinical features, and complications of AOM will be reviewed here. Related topics are
presented separately:

EPIDEMIOLOGY

AOM is a leading cause of acute care visits and the most common reason for administration of antibiotics in children
[2-4].

AOM is slightly more common in boys than girls [5-7]. It occurs at all ages but is most prevalent between 6 and 24
months of age, after which it begins to decline [8]. AOM is infrequent in school-age children, and adolescents.
Children who have their first episode of AOM before age six months (ie, "early-onset AOM") are at increased risk for
recurrent AOM [5,6,8]. Children who have few or no episodes of AOM before age three years are unlikely to have
subsequent severe or recurrent AOM. (See "Acute otitis media in children: Prevention of recurrence", section on
'Factors influencing choice'.)

The incidence of AOM in children in the United States decreased following universal immunization of infants with
the 7-valent pneumococcal conjugate vaccine (PCV7) in 2000 and further declined after PCV7 was replaced with the
13-valent pneumococcal conjugate vaccine (PCV13) in 2010 [3,9-13]. In a prospective longitudinal cohort of 615
children <4 years from the post-PCV era (2006 to 2016), the cumulative incidence of ≥1 episode of AOM (confirmed
by two validated otoscopists) was 23 percent at ≤1 year, 42 percent at ≤2 years, and 60 percent at ≤4 years [6]. In
the classic study from the pre-PCV era (1989), the cumulative incidence of ≥1 episode of AOM was 62 percent at ≤1
year and 83 percent at ≤3 years [5]. A shift to more stringent diagnostic criteria for AOM between the pre-PCV7 and
post-PCV13 eras also may have contributed to the decline in incidence [1,14].

The incidence of AOM in children also declined in other countries after introduction of pneumococcal conjugate
vaccines to the routine immunization schedule [15-19].

RISK AND PROTECTIVE FACTORS

●Risk factors – A number of risk factors for AOM have been established, the most of important of which is age.

•Age – The age-specific attack rate for AOM peaks between 6 and 12 months of age [6]. After that, the incidence
declines with age, although there is a small increase between five to six years (the time of school entry). AOM is
infrequent in school-age children and adolescents. (See "Acute otitis media in adults", section on 'Epidemiology of
acute otitis media'.)

The prevalence of AOM in early life is probably related to multiple factors, including immature anatomy (eg, in
infants, the Eustachian tube is shorter, more floppy, and more horizontally positioned than in adults, permitting
nasal secretions to enter the middle ear more easily), physiology, genetic predisposition, and immunologic naiveté;
other contributing factors have yet to be defined [20,21].

•Family history – Family history of AOM is a risk factor for AOM, particularly among children who have siblings [6].
In a prospective study of 615 children in the post-PCV era, the risk of AOM was increased in children with siblings
and a family history of recurrent AOM (ie, >3 episodes of AOM or tympanostomy tube insertion for
persistent/recurrent middle ear fluid in a sibling or either parent; odds ratio [OR] 3.7, 95% CI 2.5-5.8) [6].

The role of genetic factors in the development of AOM is suggested by a two-year prospective study of same-sex
twins and triplets [22]. The estimate of discordance for an episode of AOM was greater among dizygotic than
monozygotic twins (0.49 versus 0.04 percent).

Potential pathogenetic mechanisms for the heritability of AOM include anatomic, physiologic, and/or immunologic
features. Polymorphisms in proinflammatory cytokine genes and genes involved in innate and adaptive immunity
that increase susceptibility to otitis media and recurrent AOM have been identified [23-27].

•Day care – The transmission of bacterial and viral pathogens is common in day care centers. In observational
studies from the post-PCV era, day care attendance was associated with an increased risk of AOM, with ORs ranging
from 2.8 to 5.0 [6,28,29].
•Tobacco smoke and air pollution – Exposure to tobacco smoke increases the risk of AOM. In pooled analysis of
three studies (1784 children), the risk ratio (RR) of AOM was 1.66 (95% CI 1.33-2.06) among children whose parents
smoked [30]. In another pooled analysis, the OR for recurrent AOM was 1.48 (95% CI 1.08-2.04) if either parent
smoked [31]. The mechanism for this association is not entirely clear but may be related to increased
nasopharyngeal and oropharyngeal carriage of otopathogens [32,33]. (See "Secondhand smoke exposure: Effects in
children", section on 'Middle ear disease'.)

Ambient air pollution also has been associated with increased risk of AOM in observational studies, but the findings
are inconsistent for many pollutants [34].

•Pacifier use – In observational studies from the pre-PCV era, pacifier use was been associated with increased risk
of AOM [30,35]. In pooled analysis of two studies (4110 children), children who used a pacifier had a slightly higher
incidence of AOM than children who did not (RR 1.24, 95% CI 1.06-1.46) [30].

•Race and ethnicity – Native Americans, Alaskan and Canadian Inuit children, and indigenous Australian and
Greenlander children have a higher incidence of severe and recurrent AOM than do children of European descent
[36,37]. In some indigenous populations, 40 percent of the children may have chronic perforation of the tympanic
membrane by 18 months of age [38]. In a prospective study, severe otitis media was also more likely to be reported
in Bedouin than in Jewish children in Israel; severe otitis media was attributed to crowded living conditions and
colonization early in life among Bedouin children; genetic differences were not investigated [39].

•Limited resources – Lack of access to medical care and local environmental factors lead to severe suppurative
episodes of otitis media in children living in resource-limited settings [40]. Additional risk factors for chronic
suppurative otitis media are discussed separately. (See "Chronic suppurative otitis media (CSOM): Clinical features
and diagnosis", section on 'Epidemiology'.)

•Other risk factors – Other important risk factors in the development of single and recurrent episodes of AOM
include [41]:

-Season (increased incidence during the fall and winter months)

-Altered host defenses and underlying disease (eg, HIV, cleft palate, Down syndrome, allergic rhinitis)

●Protective factors

•Breastfeeding – Breastfeeding protects against AOM during the first two years of life [6,28,42-44]. Exclusive
breastfeeding for at least six months is associated with the greatest protection, but any breastfeeding compared
with no breastfeeding also appears to be beneficial. Lack of or limited breastfeeding is associated with an increased
risk of AOM.

In meta-analysis of five observational studies (including>17,700 patients), exclusive breastfeeding from birth to age
six months was associated with decreased risk of AOM before age two years (OR 0.57, 95% CI 0.44 to 0.75) [44]. In
meta-analysis of five observational studies (including >19,600 patients), any breastfeeding was associated with
decreased risk of AOM before age two years compared with no breastfeeding (OR 0.67, 95% CI 0.56-0.80). Although
most of the studies included in these meta-analyses were from the pre-PCV era, studies from the post-PCV era also
support the protective effect of breastfeeding.

Breastfeeding diminishes colonization of the nasopharynx by bacterial otopathogens (eg, Streptococcus

pneumoniae, nontypeable Haemophilus influenzae) [30,43,45]. Additional reasons for the lower incidence of AOM
among breastfed infants are uncertain but may be related to immunologic or nonimmune protective factors in
breast milk, the facial musculature associated with breastfeeding, or the position maintained during feeding from
the breast contrasted with bottle feeding [46,47]. (See "Infant benefits of breastfeeding", section on 'Biologically
active components of human milk'.)

•Xylitol – Although daily administration of oral xylitol reduces the risk of AOM, the beneficial effect requires
administration three to five times per day, which limits its practicability. The use of xylitol to prevent AOM in
children is discussed separately. (See "Acute otitis media in children: Prevention of recurrence", section on 'Xylitol'.)

PATHOGENESIS

The middle ear is a narrow chamber that is part of an aerated system that includes the nares, the Eustachian tube,
and the mastoid air cells (figure 1). The system is lined with respiratory mucosa; events affecting one area are
usually reflected in similar changes throughout the system. Extension of the suppurative process to adjacent
structures may lead to complications such as mastoiditis, labyrinthitis, petrositis, meningitis, and lateral sinus
thrombosis. (See 'Complications and sequelae' below.)
The pathogenesis of AOM in at-risk children generally involves the following sequence of events [48-51]:

●The patient has an antecedent event (usually a viral upper respiratory tract infection) while colonized with an
otopathogen(s) [42]. Some evidence suggests that co-colonization with bacterial otopathogens may be sufficient to
trigger the cascade of events in the absence of viral respiratory infection [50,52,53].

●The event results in inflammatory edema of the respiratory mucosa of the nose, nasopharynx, and Eustachian
tube.

●Inflammatory edema obstructs the narrowest portion of the Eustachian tube (the isthmus).

●Obstruction of the isthmus causes poor ventilation and negative middle ear pressure. This leads to the
accumulation of secretions produced by the middle ear mucosa.

●Viruses and bacteria that colonize the upper respiratory tract enter the middle ear via aspiration, reflux, or
insufflation.

●Microbial growth in the middle ear secretions often progresses to suppuration with clinical signs of AOM (bulging
tympanic membrane [TM], middle ear fluid, erythematous TM).

●The middle ear effusion may persist for weeks to months following sterilization of the middle ear infection. (See
"Acute otitis media in children: Treatment", section on 'Clinical course'.)

MICROBIOLOGY

The microbiology of AOM has been documented by cultures, viral studies, and molecular studies of middle ear
effusion obtained by needle aspiration [54-56]. When a variety of microbiologic methods are used, bacterial and/or
viral respiratory tract pathogens can be isolated from most middle ear aspirates from children with AOM [56-58]. In
one series of 79 children with AOM and new onset otorrhea through a tympanostomy tube, bacteria (with or
without viruses) were detected in 92 percent, viruses (with or without bacteria) in 70 percent, and both bacteria
and viruses in 66 percent [57].

Bacterial pathogens — S. pneumoniae, nontypeable H. influenzae (NTHi), and Moraxella catarrhalis are the most
common bacteria isolated from middle ear fluid in children with AOM [2,6,12,58-61].

Effect of infant immunization — Universal infant immunization with pneumococcal conjugate vaccines (PCVs) has
affected the microbiology of AOM in children by altering the relative importance of the bacterial pathogens most
frequently isolated from the middle ear fluid (ie, S. pneumoniae, H. influenzae, M. catarrhalis) as well as the most
frequently isolated serotypes of S. pneumoniae. These changes affect the choice of antimicrobial therapy when
treatment is indicated. (See "Acute otitis media in children: Treatment", section on 'Initial antimicrobial therapy'.)

●Pre-heptavalent PCV (PCV7) – Before the introduction of PCV7 (table 1), PCV7 serotypes accounted for 60 to 70
percent of AOM isolates in the 6- to 59-month age group [62].

●Post-PCV7 – In the years immediately following introduction of PCV7, prospective studies in community-based
pediatric practices noted declines in the proportion of S. pneumoniae isolates from middle ear fluid (from
approximately 45 to 30 percent) and increases in the proportion of NTHi isolates (from approximately 40 to 55
percent) [2,59,63].

Although the proportion of NTHi isolates from the middle ear fluid increased, studies of the incidence of NTHi AOM
had inconsistent results, with some suggesting an increase while others suggested declines in NTHi AOM in early
episodes but increased incidence in recurrent disease [2,61,63-66].

Following the introduction of PCV7, most pneumococcal isolates recovered from the middle ear of children with
AOM and the nasopharynx of asymptomatic children were non-PCV7 serotypes [61,63,67-69]. Multidrug-resistant
serotype 19A was reported as a cause of recalcitrant AOM [67,69,70] and coalescent mastoiditis [71]. Serotype 19A
is included in the 13-valent PCV (PCV13), which replaced PCV7 in 2010.

●Post-PCV13 – With the introduction of PCV13 in 2010, the proportion of S. pneumoniae isolates continues to be
lower than reported in the pre-PCV7 era and the proportions of M. catarrhalis and NTHi remain increased [6].

Following the introduction of the PCV13, studies of tympanocentesis-confirmed AOM in children in Israel and the
United States demonstrated declines in overall episodes of AOM and in episodes of AOM due to PCV13 serotypes
[12,72]. The Israeli study found no increase in AOM due to non-PCV13 pneumococcal serotypes or NTHi. In a study
from Italy, non-PCV13 serotypes were isolated from the majority of children with pneumococcal AOM with
spontaneous perforation [73], although it is unclear if this represents an increase in disease due to non-PCV13
serotypes or only a reduction in disease due to PCV13 serotypes. Similar findings (ie, dominance of nonvaccine
serotypes of S. pneumoniae) have been found in studies of children with AOM in Rochester, New York [74].

Surveillance after licensure of PCV13 has also documented a decline in the prevalence of nasopharyngeal
colonization with PCV13 serotypes, including 19A, in children with AOM [75-78].

●Post-PCV10 – The 10-valent pneumococcal conjugate vaccine (PCV10) includes the NTHi protein D as one of the
conjugate proteins. PCV10 became available in 2010, although it is not licensed in the United States.

Introduction of PCV10 has been associated with decreased episodes of pneumococcal AOM and vaccine serotype
AOM. In randomized clinical trials and in post-marketing, the efficacy of PCV10 against the first episode of clinical
AOM in children younger than 24 months was 24 percent (95% CI 9-37 percent); the efficacy against moderate
clinical AOM was 18 percent, (95% CI -6 to 26 percent) and the efficacy against severe clinical AOM was 33 percent
(95% CI -21 to 62 percent) [18]. A statistically significant decrease in episodes of NTHi AOM has not been
reproducibly demonstrated.

Streptococcus pneumoniae — S. pneumoniae is estimated to account for approximately 15 to 25 percent of

bacterial isolates from the middle ear fluid of young children with AOM [6,72,79,80].

The proportion of pneumococcal isolates resistant to penicillin varies worldwide, but resistance is more likely in
children with recurrent and/or persistent AOM and in those recently exposed to antimicrobial agents [48,61].
Studies of antimicrobial susceptibility among nasopharyngeal and middle ear isolates of S. pneumoniae identified
increasing resistance to penicillin, amoxicillin, and third-generation cephalosporins among nonvaccine serotypes in
2013 to 2016 compared with 2010 to 2012 [81]. Despite increases in mean inhibitory concentration for amoxicillin,
most isolates of S. pneumoniae remain susceptible, using pharmacodynamic breakpoints for high doses (90 mg/kg
per day).

S. pneumoniae frequently causes first or early otitis media episodes, although H. influenzae and M. catarrhalis may
cause early disease as well [7,82-84]. S. pneumoniae is isolated with equal frequency in bilateral and unilateral AOM
[85,86].

S. pneumoniae is associated with greater clinical severity than other otopathogens, as reflected by high fever, more
intense otalgia, and the potential for complications such as bacteremia and mastoiditis [87]. It is also associated
with a greater inflammatory response with elevated peripheral blood and middle ear fluid white blood cell counts
compared with other pathogens [82].

Haemophilus influenzae — H. influenzae is estimated to account for approximately 50 to 60 percent of bacterial

isolates from the middle ear fluid of young children with AOM [6,72,79,80].

Most H. influenzae isolates from the middle ear are NTHi [7,61]. In some settings, NTHi has become the most
common bacteria isolated from children with AOM [88].

The patterns of H. influenzae resistance vary geographically. Approximately one-third to one-half of strains of H.
influenzae recovered from middle ear fluids from children in the United States produce beta-lactamase [48,60,89].
In other parts of the world, resistance to ampicillin and cephalosporins may be associated with mutations in the
penicillin binding proteins (beta-lactamase negative, ampicillin-resistance) [69,90-92]. (See "Epidemiology, clinical
manifestations, diagnosis, and treatment of Haemophilus influenzae", section on 'Ampicillin resistance'.)

H. influenzae AOM is more commonly isolated in bilateral than unilateral AOM [85,86,93]. Otitis media caused by
NTHi is frequently associated with concurrent conjunctivitis and milder symptoms (eg, lower fever, less inflamed
tympanic membrane [TM]) than pneumococcal AOM [82,86]. NTHi infection is also associated with more complex
otitis media, including increased risk for treatment failure, recurrence despite appropriate antibiotic therapy, and
chronicity [56,93].

Moraxella catarrhalis — M. catarrhalis accounts for approximately 12 to 15 percent of bacterial middle ear isolates
in children with AOM [6,72,79,80]. It is isolated more commonly in middle ear aspirates or otorrhea specimens in
countries where children have been immunized with PCV. Virtually all strains produce beta-lactamase [6,63]. M.
catarrhalis AOM typically is less severe than pneumococcal AOM.

Group A Streptococcus — Group A Streptococcus (GAS) occasionally causes AOM (2 to 10 percent of isolates) [2].
GAS AOM tends to occur in older children and to be more frequently associated with local complications (eg, TM
perforation, mastoiditis) and less frequently associated with fever and systemic or respiratory symptoms than AOM
caused by other organisms [94-96]. (See 'Tympanic membrane abnormalities' below.)

Staphylococcus aureus — S. aureus is an uncommon cause of AOM, but its prevalence appears to have increased
after of PCV7 [2,65,97,98]. It is found often as a cause of acute otorrhea in children with tympanostomy tubes in
place. (See "Tympanostomy tube otorrhea in children: Causes, prevention, and management", section on
'Pathogens'.)

Other bacteria — Anaerobic bacteria infrequently cause AOM. Enteric gram-negative bacilli such as Escherichia coli
may cause AOM in the first months of life [99,100]. Pseudomonas aeruginosa has a special role in chronic
suppurative otitis media. (See "Pseudomonas aeruginosa infections of the eye, ear, urinary tract, gastrointestinal
tract, and central nervous system", section on 'Ear infections'.)

Viral pathogens — Microbiologic and epidemiologic data suggest that viral infection is frequently associated with
AOM [57,101-104]. With advances in microbiologic techniques, including reverse transcriptase polymerase chain
reaction, viruses are increasingly detected in the middle ear fluid of children with AOM [57,103]. The most
frequently isolated viruses are respiratory syncytial virus (RSV), picornaviruses (eg, rhinovirus, enterovirus),
coronaviruses, influenza viruses, adenoviruses, and human metapneumovirus. However, in an epidemiologic study,
only influenza, RSV, and human metapneumovirus activity in the community were associated with an increase in
AOM diagnoses [105].

Other pathogens

●Mycoplasma pneumoniae rarely has been isolated from middle ear fluids of children with AOM. (See "Mycoplasma
pneumoniae infection in children", section on 'Clinical manifestations'.)

●Chlamydia trachomatis has been associated with otitis media in infants younger than six months of age. (See
"Chlamydia trachomatis infections in the newborn".)

●Chlamydia pneumoniae has been isolated from some patients with acute and chronic otitis media [54].

Dual pathogen infection

●Dual bacterial infections – Dual bacterial infection, most commonly involving S. pneumoniae and H. influenzae, has
been identified in children with persistent AOM and/or recurrent otitis media from multiple countries [39,106-108].
The persistence and recurrence of polymicrobial otitis media probably reflects biofilms in the middle ear space. In a
prospective study, dual bacterial infection with S. pneumoniae and H. influenzae was more likely in children living in
crowded living conditions with early colonization; nonvaccine serotypes were involved most frequently [39].

●Bacterial and viral infections – Bacterial and viral coinfection is common in children with recurrent AOM. In one
series of 79 children with AOM and new onset otorrhea through a tympanostomy tube, both bacteria and viruses
were isolated in 66 percent [57].

The frequency of mixed bacterial and viral infections has important clinical implications [109]. Mixed bacterial and
viral infections may respond differently to antibiotic therapy than purely bacterial infections [110,111]. The
presence of viruses may increase middle ear inflammation [112,113], decrease neutrophil function [114], and
reduce antibiotic penetration into the middle ear [115].

PATHOGENS IN SPECIAL POPULATIONS

Young infants — Bacterial pathogens isolated from the middle ear aspirates of young infants with AOM typically are
similar to those identified in older children (eg, S. pneumoniae and H. influenzae). In infants younger than two
weeks, pathogens that cause neonatal sepsis (group B Streptococcus, enteric gram-negative bacilli, and S. aureus)
also may be found. (See "Clinical features, evaluation, and diagnosis of sepsis in term and late preterm infants",
section on 'Etiologic agents'.)

Few studies have evaluated the microbiology of AOM in young infants after the introduction of pneumococcal
conjugate vaccine (PCV). In an observational study of 182 infants <2 months of age from Israel who underwent
tympanocentesis before (2005 to 2009) and after (2010 to 2014) the introduction of PCV, the proportion of S.
pneumoniae and H. influenzae isolates decreased after the introduction of PCV (from 50 to 30 percent and 30 to 23
percent, respectively) and the proportion of culture-negative patients increased (from 19.6 to 35.6 percent) [116].
Among the 23 infants (13 percent) with concomitant serious bacterial infection, 19 had urinary tract infections
which were generally caused by pathogens different than those isolated from middle ear fluid.

Resource-limited countries — In resource-limited countries, the most common bacterial causes of AOM are the
same as in developed countries [117-119].

In addition, otitis media may be caused by Mycobacteria tuberculosis, Corynebacterium diphtheriae, and
Clostridium tetani. Parasitic (Ascaris lumbricoides) and fungal (Blastomyces dermatitidis, Candida and Aspergillus
species) middle ear infections may occur but are rare [117].
COMPLICATIONS AND SEQUELAE

Intratemporal complications

Hearing loss — Most patients with middle ear effusion have persistent or fluctuating conductive hearing loss. Fluid
filling the middle ear space prevents the tympanic membrane (TM) from vibrating adequately, thereby diminishing
movement of the ossicular chain. The hearing loss remains as long as fluid fills the middle ear space. The median
hearing threshold is 25 dB, which is equivalent to putting plugs in the child's ears [120]. (See "Hearing loss in
children: Etiology", section on 'Infection'.)

Despite treatment with appropriate antimicrobial agents, middle ear fluid may persist for weeks to months after the
onset of signs of AOM. (See "Acute otitis media in children: Treatment", section on 'Clinical course'.)

Consequences and sequelae of prolonged middle ear effusion are discussed separately. (See "Otitis media with
effusion (serous otitis media) in children: Clinical features and diagnosis", section on 'Hearing loss'.)

Sensorineural hearing loss is an uncommon consequence of AOM but may occur.

Balance and motor problems — Children with AOM may have balance, motor, or vestibular problems related to
vestibular dysfunction or labyrinthitis [121-125]. (See "Causes of vertigo", section on 'Otitis media'.)

Tympanic membrane abnormalities

●Perforation – The increased pressure in the middle ear can result in central ischemia, necrosis, and spontaneous
perforation of the TM, usually accompanied by otorrhea (picture 3) [126].

The frequency of spontaneous perforation is increased in Native Americans, Australian Aborigines, and in children in
resource-limited countries [36]. AOM caused by group A Streptococcus is associated with higher rates of TM
perforation than AOM caused by other pathogens [94-96]. The risk of perforation is also increased in children
initially observed without antimicrobial therapy and in children with dual pathogen infections, including
nontypeable H. influenzae and M. catarrhalis [96,127,128]. (See 'Group A Streptococcus' above.)

The treatment of TM perforation associated with AOM is discussed separately. (See "Acute otitis media in children:
Treatment", section on 'AOM with perforation'.)

●Myringosclerosis – Myringosclerosis (calcification of the connective tissue of TM, sometimes called

tympanosclerosis) is a complication of frequent middle ear disease [129]. It is characterized by whitish plaques in
the TM (picture 4A-B) and occasionally the middle ear. In most children, myringosclerosis is of no functional
importance, but in some, the deposits may envelop the ossicles, resulting in a conductive hearing loss.

●Retraction or collapse – Chronic or recurrent decreased pressure in the middle ear in children with recurrent AOM
may lead to retraction or collapse of the TM (middle ear atelectasis) (picture 5).

Chronic suppurative otitis media — Unresolved or complicated AOM with perforation of the TM may lead to chronic
suppurative otitis media (CSOM) or chronic otomastoiditis, which is defined as perforation of the TM with chronic
purulent drainage from the middle ear cleft for more than six weeks (picture 6). Most often CSOM is the end stage
of recurrent AOM episodes that begin in the first year of life. (See "Chronic suppurative otitis media (CSOM): Clinical
features and diagnosis".)

Children with TM perforation and suppurative discharge that persists despite oral and/or topical antibiotics should
be referred to an infectious disease specialist or otolaryngologist as soon as possible and a sample for a bacterial
culture should be obtained to ensure optimal antimicrobial treatment. Children with perforation without drainage
that persists for three months or longer should be referred to an otolaryngologist for further management [126].
Prevention of CSOM entails prompt and appropriate treatment of acute middle ear infection [36].
Chemoprophylaxis is not warranted. (See "Chronic suppurative otitis media (CSOM): Treatment, complications, and
prevention".)

Cholesteatoma — A cholesteatoma is an abnormal growth of squamous epithelium in the middle ear and mastoid
that may progressively enlarge to surround and destroy the ossicles. Recurrent AOM, which may lead to retraction
pockets in the TM, is a risk factor for cholesteatoma.

Clinical features suggestive of acquired cholesteatomas include deep retraction pockets, as in panel A of the picture
(picture 7), a white mass behind the TM (picture 8), focal granulation at the periphery of the TM (picture 9), new-
onset hearing loss, and ear drainage for more than two weeks despite treatment. The diagnosis and management of
cholesteatoma are discussed separately. (See "Cholesteatoma in children", section on 'Clinical features' and
"Cholesteatoma in children", section on 'Surgical treatment'.)
Mastoiditis — Most episodes of AOM are associated with some inflammation of the mastoid because the mastoid
air cells are connected to the distal end of the middle ear through a small canal or antrum (figure 2). In rare cases,
resolution of the mastoid infection does not occur, and acute suppurative mastoiditis develops with pus filling the
air cells.

Other intratemporal complications — Other intratemporal complications of AOM include [121]:

●Petrositis (extension of the infection into the petrous portion of the mastoid bone) (see "Acute mastoiditis in
children: Clinical features and diagnosis", section on 'Complications')

●Labyrinthitis (extension of infection into the cochlear and vestibular apparatus) (see "Causes of vertigo", section
on 'Otitis media'

●Facial paralysis (the facial nerve courses through the middle ear and mastoid); facial paralysis also may occur as a
complication of acute mastoiditis or CSOM (see "Facial nerve palsy in children", section on 'Otitis media')

Intracranial complications — Intracranial complications of AOM are uncommon in resource-rich countries. However,
they remain a concern in resource-limited countries where there is poor access to medical care [117,130].
Intracranial complications of AOM include:

●Meningitis (see "Bacterial meningitis in children older than one month: Clinical features and diagnosis", section on
'Clinical features')

●Epidural abscess (see "Intracranial epidural abscess", section on 'Clinical manifestations')

●Brain abscess (see "Pathogenesis, clinical manifestations, and diagnosis of brain abscess", section on 'Clinical
manifestations')

●Lateral sinus thrombosis (see "Septic dural sinus thrombosis", section on 'Septic lateral sinus thrombosis')

●Cavernous sinus thrombosis (see "Septic dural sinus thrombosis", section on 'Septic cavernous sinus thrombosis')

●Subdural empyema (collection of purulent material between the dura and the arachnoid membrane)

●Carotid artery thrombosis

MIDDLE EAR ANATOMY

The middle ear is a cavity between the external ear canal and the inner ear that contains the ossicular chain (figure
1). The facial nerve traverses the medial wall of the middle ear. The middle ear is contiguous with the Eustachian
tube, attic (epitympanum), and mastoid air cells.

The tympanic membrane is divided into four quadrants: a line drawn along the manubrium of the malleus divides
the anterior and posterior quadrants, and a line through the umbo divides the superior and inferior quadrants
(figure 2). The pars flaccida is the small portion of the tympanic membrane above the lateral process of the malleus;
it is thicker and less taut than the rest of the tympanic membrane. The remainder of the tympanic membrane (the
pars tensa) is thinner than the pars flaccida and is suspended from the fibrous tympanic annulus.

The normal middle ear is aerated, and the normal tympanic membrane is intact, slightly convex, translucent, and
mobile (picture 1).

CLINICAL PRESENTATION

Acute otitis media — Symptoms of AOM in children include ear pain, ear rubbing, hearing loss, and ear drainage.
Fever occurs in one- to two-thirds of children with AOM, though temperature >40°C (104°F) is unusual without
bacterema or another focus of infection [1].

Ear pain is the most common complaint and the best predictor of AOM [2-4]. However, ear pain and other ear-
related symptoms are not always present [2,3,5]. In a prospective study of 335 consecutive episodes of AOM, ear
pain was absent in 17 percent [5]. AOM without complaints of ear pain occurred more frequently in children
younger than two years (who may not be able to complain of ear pain) than in older children (25 versus 7 percent).
Other causes of ear pain, hearing loss, and otorrhea in children are discussed separately. (See "Evaluation of
earache in children" and "Hearing loss in children: Etiology" and "Evaluation of otorrhea (ear discharge) in
children".)

Young children with AOM, particularly infants, may present with nonspecific symptoms and signs (eg, fever,
fussiness, disturbed or restless sleep, poor feeding/anorexia, vomiting, diarrhea) [3,6,7]. Symptoms of AOM may
overlap with those of upper respiratory tract infection without AOM or may be subtle or absent [2,6,8-10]. The lack
of specificity of symptoms in young children makes the diagnosis challenging and underscores the importance of
otoscopy, which is also challenging in young children.

Complications of AOM — Children with complications of AOM may present with findings related to the
complication, including (see "Acute otitis media in children: Epidemiology, microbiology, and complications",
section on 'Complications and sequelae') [4,11]:

●Postauricular swelling and protrusion of the auricle (picture 2A-B) are characteristic of mastoiditis. (See "Acute
mastoiditis in children: Clinical features and diagnosis", section on 'Clinical features'.)

●Vestibular symptoms (eg, dizziness, vertigo, balance and motor problems) with or without tinnitus or nystagmus
may be related to labyrinthitis, mastoiditis, or cholesteatoma. (See "Causes of dizziness and vertigo in children and
adolescents", section on 'Otitis media' and "Evaluation of dizziness and vertigo in children and adolescents" and
"Cholesteatoma in children", section on 'Clinical features'.)

●Cranial nerve palsies (eg, facial nerve, abducens nerve) may be related to acute mastoiditis, petrositis,
cholesteatoma, or intracranial complications. (See "Facial nerve palsy in children", section on 'Otitis media' and
"Third cranial nerve (oculomotor nerve) palsy in children" and "Fourth cranial nerve (trochlear nerve) palsy" and
"Sixth cranial nerve (abducens nerve) palsy".)

●Meningeal signs, cranial nerve deficits, and/or focal neurologic findings may be related to intracranial
complications (eg, meningitis, brain abscess, epidural or subdural abscess, lateral or cavernous sinus thrombosis).
These complications are discussed in individual topic reviews. (See "Bacterial meningitis in children older than one
month: Clinical features and diagnosis", section on 'Clinical features' and "Pathogenesis, clinical manifestations, and
diagnosis of brain abscess" and "Intracranial epidural abscess" and "Septic dural sinus thrombosis".)

OTOSCOPIC EVALUATION

Otoscopic evaluation is necessary for the diagnosis of AOM.

Procedure

●Cerumen removal – Obstructing cerumen must be removed from the external canal to ensure a clear view of the
tympanic membrane. Cerumen is most practically and conveniently removed under direct vision. (See "Cerumen",
section on 'Cerumen removal'.)

●Equipment

•An otoscope with an adequate light source (eg, a halogen bulb with brightness ≥100 foot-candles [1000 lux]) and
fully charged battery [12-14]; a pneumatic otoscope with a round head is preferred because it provides the best seal
for pneumatic otoscopy (picture 3).

•Ear specula of various sizes; the largest size that fits comfortably into the cartilaginous portion of the external
auditory canal should be used (a 4 mm speculum will work for most children) [13]; the outer diameter of the
speculum may be increased by putting a small piece of rubber tubing around the tip [14,15].

•Insufflator bulb and tubing; the pneumatic system should be checked for leaks periodically by occluding the tip of
the speculum with a finger and squeezing the rubber bulb to see if resistance is felt [13].

●Assessment of tympanic membrane – Each quadrant of the tympanic membrane (figure 2) should be assessed
systematically to evaluate position, mobility, translucency, color, and other findings (eg, air-fluid levels, perforation,
retraction pockets, cholesteatoma) [13,16]. Mobility of the tympanic membrane can be assessed with pneumatic
otoscopy.

Systematic evaluation of the tympanic membrane and pneumatic otoscopy are challenging, particularly in young
children who may not cooperate with the examination.

●Pneumatic otoscopy – Pneumatic otoscopy can be painful in children with AOM. It is not necessary in children with
bulging tympanic membranes because all bulging tympanic membranes have decreased or absent mobility. (See
'Tympanic membrane findings' below.)

Pneumatic otoscopy creates positive and negative pressures within the external auditory canal by using the
insufflator bulb of the otoscope. Positive pressure is created when the insufflator bulb is compressed and negative
pressure when it is released. An airtight seal between the speculum and the external auditory canal is essential.

Mobility of the tympanic membrane is best visualized in the posterosuperior quadrant or pars flaccida, where the
tympanic membrane is most compliant (figure 2) [14]. The tympanic membrane moves away from the observer
with positive pressure and toward the observer with negative pressure. Mobility is described as normal (movie 1),
absent (movie 2), decreased, or increased.

When there is high negative pressure in the middle ear cavity, the tympanic membrane may be maximally retracted
and unable to move away from the observer with positive pressure (movie 2). Mobility of a retracted tympanic
membrane can be assessed by creating negative pressure in the external auditory canal. The otoscope should be
removed from the external canal and the bulb compressed. After the bulb is compressed, the otoscope is reinserted
into the external canal. When the seal is secured, the bulb is released, creating negative pressure, which allows the
retracted tympanic membrane to move toward the observer into a neutral position.

●Digital otoscopy – Digital otoscopy incorporates a high-resolution camera into the otoscope or endoscope to
permit photography or videography. One type of device connects to a smart phone to create a digital otoscope that
provides images similar to those viewed through a conventional otoscope; these devices typically use large specula,
require cerumen removal, and may be awkward to use if the smart phone functions as the handle of the otoscope.
Another digital otoscope uses a thin speculum to bypass or push through cerumen to provide a view of the
tympanic membrane. The recorded images can be viewed on the enlarged otoscope touchscreen or uploaded to a
computer for magnification and reviewed by single or multiple observers (including the parents). Neither device
permits pneumatic otoscopy. These devices will address one of the main challenges of the ear examination in young
children (ie, documentation) and allow study of the images. Whether they will improve diagnosis of AOM and
reduce unnecessary antibiotics remains uncertain.

Otoscopy skills, including accurate interpretation of findings, can be improved through training [17,18]. Training
modules are available online (eg, in the Children's Hospital of Pittsburgh AOM study toolkit for physicians).

Tympanic membrane findings — In children with AOM, the classic examination findings include a fluid-filled middle
ear and tympanic membrane that is bulging, opaque, yellow, or white (picture 4 and picture 5) and has decreased or
absent mobility with pneumatic otoscopy, if pneumatic otoscopy is performed [19-21]. However, this constellation
of findings is not always present.

●Bulging tympanic membrane – A bulging tympanic membrane is the hallmark of AOM and differentiates AOM
from otitis media with effusion (OME) (picture 4) [10,19,21-24]. A bulging tympanic membrane indicates both acute
inflammation and middle ear effusion (MEE, and thus decreased or absent mobility).

Bulging is first apparent in the posterosuperior area, where the tympanic membrane is most compliant (figure 2).
When the tympanic membrane is bulging, the handle of the malleus is obscured (picture 4) [14]. The tympanic
membrane may appear full rather than bulging when there are smaller amounts of infected middle ear fluid.

In a large study correlating examination findings from otoscopy with results from myringotomy (as a diagnostic tool
for AOM), the predictive value of a bulging tympanic membrane ranged from 83 to 99 percent [19]. In another
study correlating examination findings with diagnosis of AOM by experienced otoscopists, 92 percent of children
with AOM had a bulging tympanic membrane compared with none of the children with OME or no effusion [22].

In these studies, AOM was unlikely when the tympanic membrane was in a neutral position (picture 1) or retracted
(picture 6) [10,22]. When the tympanic membrane is retracted, the handle of the malleus appears to be shortened,
and the lateral process becomes more prominent and seems closer to the otoscope [13,14]. Retraction of the
tympanic membrane indicates negative pressure in the middle ear cavity (eg, Eustachian tube dysfunction).

●Acute perforation with purulent otorrhea – Acute perforation with purulent otorrhea establishes the diagnosis of
AOM provided that otitis externa is excluded [25]. (See 'Clinical diagnosis' below and "External otitis: Pathogenesis,
clinical features, and diagnosis".)

●Decreased or absent mobility – Decreased or absent (movie 2) tympanic membrane mobility is a sign of MEE,
provided that the seal is airtight during pneumatic otoscopy [13,26]. Decreased mobility also may occur with
myringosclerosis of the tympanic membrane (picture 7).

Decreased or absent mobility cannot be used in isolation to make a diagnosis of AOM because it does not
distinguish infected from uninfected middle ear fluid. In a study correlating examination findings with a diagnosis of
AOM by experienced otoscopists, decreased mobility of the tympanic membrane was present in all 50 cases of AOM
and 23 of 34 cases of OME [22].

Increased mobility of the tympanic membrane is not helpful in the diagnosis of AOM. Increased mobility of an area
of the tympanic membrane may occur at the site of a previous perforation or tympanostomy tube or may be caused
by atrophy of the tympanic membrane.

●Cloudy or opaque tympanic membrane – The tympanic membrane, or a portion of the tympanic membrane (with
an air-fluid level (picture 8)), may appear cloudy or opaque when there is fluid in the middle ear, but this finding
does not help differentiate AOM from OME. In a study correlating examination findings with a diagnosis of AOM by
experienced otoscopists, the tympanic membrane was opaque in 50 of 50 cases of AOM and 33 of 34 cases of OME
[22]. In children with opacification of the tympanic membrane without bulging, experienced otoscopists generally
diagnose OME [21,22].

●Color of the tympanic membrane

•A white or pale yellow tympanic membrane usually indicates pus in the middle ear cavity, a sign of AOM (picture
4). Middle ear fluid that is not infected (ie, OME) usually appears amber, gray, or blue (picture 9A-B).

•A red or hemorrhagic tympanic membrane may indicate acute inflammation, but it is nonspecific. Erythema of the
tympanic membrane may be caused by vasodilation related to manipulation of the canal (as occurs during removal
of cerumen), crying, or high fever. In the crying child, vascular engorgement is limited to the periphery and handle
of the malleus (picture 10) [27]. Vessels crossing the tympanic membrane suggest inflammation (picture 4).

Erythema is less important than the position and mobility of the tympanic membrane in the diagnosis of AOM
[10,19,22]. In a large study correlating otoscopic findings and AOM, a distinctly red tympanic membrane (defined as
hemorrhagic, strongly, or moderately red) in the absence of bulging or impaired mobility predicted a diagnosis of
AOM in only 15 percent of cases [19]. In a systematic review, the adjusted likelihood ratio for a distinctly red
tympanic membrane was 8.4 (95% CI 6.7-11) and for a slightly red tympanic membrane was 1.4 (95% CI 1.1-1.8)
[10].

●Other findings – Other findings may indicate inflammation (acute or chronic), MEE (infected or uninfected), and
complications or sequelae of AOM:

•Bullae are caused by inflammation of the tympanic membrane that occurs in association with AOM (picture 11).

•Bubbles or air-fluid levels (picture 8) indicate MEE and are more suggestive of OME than AOM [22]. Air-fluid levels
fluctuate with pneumatic otoscopy.

•Myringosclerosis (asymptomatic whitish plaques of calcium and phosphate crystals in the tympanic membrane
(picture 7)) may result from chronic middle ear inflammation, perforation, myringotomy with or without
tympanostomy tube placement, or trauma. Myringosclerosis moves with the tympanic membrane during
pneumatic otoscopy. (See "Otitis media with effusion (serous otitis media) in children: Clinical features and
diagnosis", section on 'Complications and sequelae'.)

•Perforation of the tympanic membrane may result from increased middle ear pressure that leads to central
ischemia and necrosis.

•Atrophic areas may result from AOM; atrophic areas may have increased mobility.

•Retraction pockets (picture 12) may be sequelae of otitis media and can predispose to the development of
cholesteatoma.

•Cholesteatomas are benign growths of desquamated, stratified, squamous epithelium. They may appear as a cyst
within the tympanic membrane, greasy white debris, or as a mass (picture 13). Children with cholesteatoma or
suspected cholesteatoma should be referred to an otolaryngologist. (See "Cholesteatoma in children".)

TYMPANOMETRY AND ACOUSTIC REFLECTOMETRY

Tympanometry and acoustic reflectometry may be used to confirm middle ear fluid.

●Tympanometry measures the compliance of the tympanic membrane, Eustachian tube function, and middle ear
function [28].

●Acoustic reflectometry measures the reflection of sound from the tympanic membrane. A smartphone application
that uses acoustic reflectometry to detect middle ear fluid has been developed [29]. In a small study, its accuracy
was comparable to that of pneumatic otoscopy and tympanometry, but additional validation is necessary before it
can be recommended.

Neither tympanometry nor acoustic reflectometry differentiates infected from uninfected middle ear fluid [30,31].
However, if tympanometry and/or acoustic reflectometry are normal, both AOM and OME are unlikely.
Tympanometry and acoustic reflectometry are discussed separately. (See "Otitis media with effusion (serous otitis
media) in children: Clinical features and diagnosis", section on 'Diagnosis'.)

DIAGNOSIS
Stringent diagnostic criteria are necessary to distinguish AOM from otitis media with effusion (OME). The
importance of accurate diagnosis of AOM cannot be overstated. Accurate diagnosis prevents overuse of antibiotics,
which leads to the development of resistant organisms [32-34].

Acute otitis media

Clinical diagnosis — The diagnosis of AOM requires middle ear effusion (MEE) and acute signs of middle ear
inflammation. Children who have MEE without evidence of acute inflammation have OME. (See 'Otitis media with
effusion' below.

A clinical diagnosis of AOM can be made in children with either [22,25]:

●Bulging of the tympanic membrane (picture 4); distinct fullness or bulging of the tympanic membrane is the most
specific and reproducible sign of acute inflammation (picture 4) [10,19,21,22]. Pneumatic otoscopy is not necessary
in children with bulging of the tympanic membrane.

●Perforation of the tympanic membrane with acute purulent otorrhea if acute otitis externa has been excluded.
(See "External otitis: Pathogenesis, clinical features, and diagnosis".)

Very infrequently, a clinical diagnosis of AOM can be made in children without bulging of the tympanic membrane
or acute purulent otorrhea if they have MEE and other signs of acute inflammation (picture 5), although
experienced otoscopists rarely make the diagnosis of AOM in the absence of tympanic membrane bulging [21,22].

MEE can be confirmed by otoscopy in children with bubbles or an air-fluid level (picture 8), or two or more of the
following: abnormal color (white, yellow, amber, gray, or blue), opacity involving all or part of the tympanic
membrane that is not due to myringosclerosis, and impaired mobility (movie 2) [35]. MEE can also be confirmed by
tympanometry, acoustic reflectometry, myringotomy, or tympanocentesis; myringotomy and tympanocentesis are
rarely performed in the primary care setting [10,33]. (See 'Tympanometry and acoustic reflectometry' above.)

In children with MEE, signs of acute inflammation are necessary to differentiate AOM from OME. Although marked
redness of the tympanic membrane is a sign of acute inflammation [35], marked redness of the tympanic
membrane without bulging is unusual in AOM [22]. In a large study correlating otoscopic findings with observations
made at the time of myringotomy, a distinctly red tympanic membrane in the absence of bulging or impaired
mobility had a positive predictive value of only 15 percent for AOM [19]. Nonotoscopic signs and symptoms of
inflammation (eg, fever, ear tugging, otalgia) must be accompanied by abnormal otoscopic findings to make a
diagnosis of AOM.

Etiologic diagnosis — Tympanocentesis (aspiration of the middle ear fluid) for culture or other microbiologic studies
is required for etiologic diagnosis. Etiologic diagnosis is not necessary in most cases of AOM because antimicrobial
therapy is chosen empirically. Tympanocentesis for etiologic diagnosis is warranted if the child appears toxic, is
immunocompromised, or has failed previous courses of antibiotic therapy. (See "Acute otitis media in children:
Treatment", section on 'Treatment failure'.)

Otitis-conjunctivitis — When AOM occurs with purulent conjunctivitis, it is called otitis-conjunctivitis (or
conjunctivitis-otitis) syndrome [36-39]. Otitis-conjunctivitis syndrome is usually caused by nontypeable
Haemophilus influenzae, although it can be caused by other organisms [39-41].

In a series of 124 patients with otitis-conjunctivitis syndrome, 50 percent were younger than two years [38]. The ear
pain typically began on the same day as, or within three days following, onset of the ophthalmologic symptoms.

Bullous myringitis — Bullous myringitis (inflammation and bullae of the tympanic membrane (picture 11)) occurs in
association with AOM. The viral and bacterial pathogens that cause bullous myringitis are similar to those that
cause AOM without bullae [42-47].

Bullous myringitis occurs in approximately 5 percent of cases of AOM in children younger than two years [48].
Children with bullous myringitis usually have more pain at the time of diagnosis than children without bullous
myringitis [49].

The treatment and prognosis for bullous myringitis are the same as for AOM without bullae. (See "Acute otitis
media in children: Treatment".)

DIFFERENTIAL DIAGNOSIS

The main consideration in the differential diagnosis of AOM is otitis media with effusion (OME).

Otitis media with effusion — Middle ear effusion with decreased mobility and opacification or cloudiness of the
tympanic membrane occurs in both AOM and OME. Although the distinction between AOM and OME may be
difficult because they are part of a continuous spectrum, other otoscopic findings can be helpful [13]. (See
'Otoscopic evaluation' above.)

●In AOM, the tympanic membrane is usually bulging and is typically white or pale yellow; pus may be seen behind it
(picture 4). Other findings associated with AOM include a perforation with purulent otorrhea or bullae (picture 11).

●In OME, the tympanic membrane may be retracted (picture 6) or in the neutral position and is typically amber,
gray, or blue (picture 9A-B); bubbles or an air-fluid level (with clear/serous fluid) may be seen behind it (picture 8).

Other conditions — Other conditions share some of the otoscopic and nonotoscopic features of AOM, but the
history and physical examination should readily distinguish these conditions from AOM. (See "External otitis:
Pathogenesis, clinical features, and diagnosis".)

●Redness of tympanic membrane – Redness of the tympanic membrane may be caused by vascular engorgement
(picture 10) due to crying, high fever, upper respiratory infection with congestion and inflammation of the mucosa
lining the entire respiratory tract, trauma, and/or cerumen removal.

●Decreased or absent mobility of the tympanic membrane – Conditions other than AOM and OME that cause
decreased or absent mobility of the tympanic membrane include myringosclerosis (picture 6) and high negative
pressure within the middle ear cavity.

●Ear pain – Ear pain may be caused by otitis externa, ear trauma, throat infections, foreign body, or
temporomandibular joint syndrome. (See "Evaluation of earache in children".)

DIAGNOSIS OF AOM

The clinical diagnosis of AOM requires 1) bulging of the tympanic membrane or 2) signs of acute inflammation (eg,
marked erythema of the tympanic membrane, fever, ear pain) and middle ear effusion (picture 1) [3,4]. Although
signs of acute inflammation and middle ear effusion without bulging may represent early AOM, localizing pain to
the ear in young children is often challenging [3,4]. (See "Acute otitis media in children: Clinical manifestations and
diagnosis", section on 'Diagnosis'.)

The importance of accurate diagnosis of AOM cannot be overstated. Accurate diagnosis ensures appropriate
treatment for children with AOM, who require antibiotic therapy, and avoidance of antibiotics in children with otitis
media with effusion, in whom antibiotics are unnecessary. Accurate diagnosis also prevents overuse of antibiotics,
which leads to an increased prevalence of resistant organisms.

CLINICAL COURSE

The systemic and local signs and symptoms of AOM usually resolve within 72 hours [5,6]. Symptoms and signs
resolve more slowly in children who are managed with analgesia and observation than in those who are prescribed
appropriate antibiotic therapy. In a 2013 meta-analysis of seven randomized trials and three observational studies
(1409 children), ear pain resolved within three days in 50 percent of children and within seven to eight days in 90
percent of children who did not receive antibiotic therapy [7].

Whether initially treated with antibiotics or not, persistence of middle ear effusion after the resolution of acute
symptoms is common. In a randomized trial in which children age 6 through 35 months with AOM were followed at
least every two weeks, the median time to resolution of middle ear effusion was 20 days in the antibiotic group and
29 days in the placebo group (a nonstatistically significant difference) [8]. In a prospective study of 2565 children
followed from birth, middle ear effusion (diagnosed via pneumatic otoscopy) persisted for weeks to months after
the onset of AOM [9]. At two weeks, 70 percent still had effusion; at one month, 40 percent still had effusion; at
two months, 20 percent still had effusion; and at three months, 10 percent still had effusion (figure 1).

The clinical features, complications, and management of persistent middle ear effusion are discussed separately.
(See "Otitis media with effusion (serous otitis media) in children: Clinical features and diagnosis" and "Otitis media
with effusion (serous otitis media) in children: Management".)

SYMPTOMATIC THERAPY

Pain is a common feature of AOM and may be severe [5]. We recommend treatment to reduce ear pain in children
with AOM whether or not they are treated with antibiotics [3].

Our approach — We suggest oral ibuprofen or acetaminophen rather than other analgesics for pain control in
children with AOM.

Topical procaine or lidocaine preparations (if available) are an alternative to oral analgesics for children ≥2 years but
should not be used in children with tympanic membrane perforation. Topical benzocaine is avoided in children <2
years because of the risk of methemoglobinemia [10]. Topical benzocaine products have been withdrawn from the
market in the United States because they have not been evaluated by the US Food and Drug Administration for
safety, effectiveness, and quality [11]. (See "Clinical use of topical anesthetics in children", section on 'Benzocaine'.)

Therapeutic tympanocentesis or myringotomy may be helpful in children with severe pain [12]. It also may have
other benefits. In a longitudinal prospective study, tympanocentesis with complete evacuation of middle ear fluid in
combination with antimicrobial therapy was associated with decreased risk of treatment failure, becoming otitis
prone, and tympanostomy tube placement [13,14]. Even if middle ear pus is not evacuated, tympanocentesis
creates a pathway that may permit drainage (otorrhea), which reduces pressure and pain in the middle ear.

In a 2016 systematic review of nonsteroidal anti-inflammatory drugs and acetaminophen for pain relief in AOM,
fewer children who received oral ibuprofen or acetaminophen than placebo had pain at 48 hours (7, 10, and 25
percent, respectively [one trial, 219 participants]) with no difference in the rate of adverse effects [15]. Ibuprofen
and acetaminophen appeared to be equally effective, although data were limited (two trials, 39 participants), and
data were insufficient to draw firm conclusions regarding the efficacy of combined ibuprofen and acetaminophen
versus acetaminophen alone.

The efficacy of topical benzocaine and lidocaine in reducing AOM-associated ear pain has also been evaluated in
randomized trials [16]. In a trial in which 54 children (5 to 19 years) who presented to an emergency department
with ear pain and AOM were randomly assigned to treatment with a topical benzocaine preparation or olive oil
placebo, more children in the treatment group reported a 25 percent reduction in ear pain score 30 minutes after
treatment (96 versus 70 percent) [17].

In a similar trial in 63 children (3 to 12 years) randomly assigned to topical aqueous lidocaine (lignocaine) or saline,
more children in the lidocaine group had a 25 percent reduction in ear pain score 30 minutes after treatment (90
versus 70 percent) [18]. No adverse effects were reported. Further studies of lidocaine are needed to optimize the
dose, as well as to identify the duration of pain relief and the effectiveness in younger children.

Unproven therapies

●Decongestants and antihistamines – We recommend not using decongestants and/or antihistamines in the
symptomatic management of AOM in children.

Studies of the efficacy of antihistamines and decongestants in treating AOM suggest a lack of benefit and a
potential for delayed resolution of middle ear fluid. A 2007 systematic review found that decongestants and
antihistamines alone or in combination were associated with increased medication side effects and did not improve
healing or prevent surgery or other complications in AOM [19]. In addition, treatment with antihistamines may
prolong the duration of middle ear effusion [20].

In children with AOM and known or suspected nasal allergy, an oral decongestant or antihistamine may provide
symptomatic relief of nasal congestion. When treating such children with decongestants or antihistamines,
practitioners should weigh the relief of nasal symptoms against the reported adverse events and potential for
prolongation of middle ear effusion. Nasal glucocorticoids may be a better option. (See "Pharmacotherapy of
allergic rhinitis", section on 'Approach to specific patient groups'.)

The American Academy of Pediatrics recommends that over-the-counter cough and cold medications not be given
to infants and children <6 years of age because of the risk of life-threatening side effects [21,22]. (See "The common
cold in children: Management and prevention", section on 'Over-the-counter medications'.)

●Other therapies – We do not suggest distraction, external application of heat or cold, or instillation of olive oil or
herbal extracts into the external auditory canal to treat pain in children with AOM. These therapies have been
proposed, but their effectiveness is unproven [3]

A topical herbal extract (containing Allium sativum [garlic], Verbascum thapsus [mullein], Calendula flores
[marigold], and Hypericum perforatum [St. John's wort] in olive oil) was compared with topical anesthetic treatment
in a randomized trial of 103 children (6 to 18 years) with AOM-associated pain [23]. Both groups experienced
comparable improvements in pain throughout the three days of the study, but there was no placebo group.

ANTIBIOTIC THERAPY VERSUS OBSERVATION

In addition to pain control, there are two strategies for initial management of AOM: 1) immediate treatment with
antibiotics and 2) observation with initiation of antibiotic therapy if the symptoms and signs worsen or fail to
improve after 48 to 72 hours.

The choice of strategy depends upon the age of the child, the severity of illness, and parental preference:

●We recommend that children <6 months with AOM be treated immediately with an appropriate antibiotic. (See
'Initial antimicrobial therapy' below.)
Febrile infants younger than 60 days who are diagnosed with AOM may require additional evaluation before
initiation of antimicrobial therapy to avoid masking an invasive bacterial infection. (See "Febrile infant (younger
than 90 days of age): Outpatient evaluation", section on 'Suggested ancillary studies by risk group'.)

●We suggest that children six months to two years with unilateral or bilateral AOM be treated immediately with an
appropriate antibiotic.

For children six months to two years with unilateral AOM and mild symptoms (ie, mild ear pain for <48 hours and
temperature <102.2°F [39°C]), the American Academy of Pediatrics (AAP) guideline (2013) permits initial
observation after shared decision-making with the parent(s)/caregiver(s) [3]. However, given the high rate of
treatment failure among children <24 months with unilateral nonsevere AOM who are initially managed with
observation and analgesia [24], we suggest that such children be treated with antimicrobial therapy. (See 'Initial
antimicrobial therapy' below.)

●We suggest that children ≥2 years who appear toxic, have persistent otalgia for more than 48 hours, have
temperature ≥102.2°F (39°C) in the past 48 hours, have bilateral AOM or otorrhea, or have uncertain access to
follow-up be immediately treated with an appropriate antibiotic. See 'Initial antimicrobial therapy' below.)

●For children ≥2 years who are normal hosts (eg, immune competent, without craniofacial abnormalities) with mild
symptoms and signs and no otorrhea, initial observation may be appropriate if the caretakers understand the risks
and benefits of such an approach. (See 'Initial observation' below.)

The treatment of otorrhea in children with tympanostomy tubes is discussed separately. (See "Tympanostomy tube
otorrhea in children: Causes, prevention, and management", section on 'Treatment'.

In a 2006 meta-analysis of individual data from six high-quality randomized trials (1643 children age 6 months to 12
years), children who were younger than two years who had bilateral AOM and children with otorrhea benefited
most from antibiotic therapy [5]. Among children younger than two years with bilateral AOM, 25 percent (95% CI
14-36 percent) fewer children treated with antibiotics than with symptomatic care continued to have pain and/or
fever on days 3 to 7 of illness. Among children with otorrhea, 36 percent (95% CI 19-53 percent) fewer children
treated with antibiotics than with symptomatic care continued to have pain and/or fever on days 3 to 7 of illness.

A 2015 meta-analysis of 13 randomized trials (3401 children, 3938 episodes of AOM) comparing antibiotics with
placebo found [6]:

●Antibiotics reduced pain at two to three days (seven trials, 2320 patients): 11.6 versus 15.9 percent (relative risk
[RR] 0.70, 95% CI 0.57-0.86)

●Antibiotics reduced tympanic membrane perforations (five trials, 1075 patients): 1.9 versus 4.8 percent (RR 0.37,
95% CI 0.18-0.76)

●Antibiotics reduced contralateral episodes of AOM (four trials, 906 patients): 10.6 versus 18.8 percent (RR 0.49,
95% CI 0.25-0.95)

●Antibiotics did not affect the rate of late recurrence (six trials, 2200 patients): approximately 20 percent

●Antibiotics increased adverse events (vomiting, diarrhea, or rash): 27.1 versus 19.6 percent (RR 1.38, 95% CI 1.19-
1.59)

●Serious complications (eg, mastoiditis, meningitis) were rare in both the treatment and placebo groups

Randomized trials comparing immediate versus delayed antibiotics have used different outcome measures (eg,
parental satisfaction, rate of filled prescriptions, etc) and types of follow-up (eg, telephone versus office
examination) but also document earlier resolution of symptoms and signs of tympanic membrane inflammation
among children who receive immediate treatment [25-30].

Systematic reviews and meta-analyses suggest that many children with AOM do well with initial observation and
analgesia only, and that the benefits of antibiotics are modest [6,31-34]. However, many of the studies included in
the meta-analyses had increased risk of bias (related to nonstringent diagnostic criteria, inclusion of children with
mild disease, exclusion of patients <2 years of age, use of an inappropriate antibiotic or inappropriate dose, etc),
making the results difficult to interpret [35-38]. Exclusion of children with severe symptoms biases toward the null
hypothesis (ie, no difference between early antimicrobial treatment and observation).

Individual randomized trials that used stringent diagnostic criteria and experienced otoscopists to make the
diagnosis of AOM and appropriate antibiotic regimens to treat AOM indicate that children younger than two years,
particularly those with marked bulging of the tympanic membrane [39], benefit from antibiotic therapy [40,41].
Pooled data from these trials indicate increased rates of treatment failure among placebo recipients <24 months
with unilateral nonsevere AOM (40 versus 14 percent among antibiotic recipients; RR 0.34, 95% CI [0.18-0.65]) [24].
These and other randomized trials suggest that children with "severe" (defined by fever and ear pain score) or
bilateral AOM also benefit from antibiotic therapy [24,42,43].

The 2013 AAP and American Academy of Family Physicians (AAFP) guideline recommends immediate antimicrobial
treatment for children <6 months, children with severe signs or symptoms (defined by moderate or severe ear pain,
ear pain for ≥48 hours, or temperature ≥39°C [102.2°F]), and bilateral AOM in children <24 months of age [3]. The
2013 AAP/AAFP guideline recommends either immediate treatment or observation (with pain control) for children
between 6 and 24 months with unilateral nonsevere AOM and for children ≥24 months with unilateral or bilateral
nonsevere AOM. However, given the additional analysis now available showing a high rate of treatment failure
among children <24 months with unilateral nonsevere AOM [24], we suggest that such children be treated at the
time of diagnosis with antimicrobial therapy.

Guidelines from many other countries (eg, the Dutch College of General Practitioners) recommend a no or delayed
antibiotic strategy for most children ≥6 months of age with AOM.

INITIAL ANTIMICROBIAL THERAPY

When the decision is made to treat AOM with antimicrobial agents, the selection among available drugs (table 1) is
based upon:

●Clinical and microbiologic efficacy

●Acceptability (taste, texture) of the oral preparation

●Absence of side effects and toxicity

●Convenience of the dosing schedule

●Cost

First-line therapy

●No recent beta-lactam therapy, no concomitant purulent conjunctivitis, and no history of recurrent AOM – We
suggest amoxicillin as the first-line therapy for children with AOM who are treated with antibiotics and at low-risk
for amoxicillin resistance (ie, have not received a beta-lactam antibiotic in the previous 30 days, do not have
concomitant purulent conjunctivitis, and have no history of recurrent AOM unresponsive to amoxicillin) (algorithm
1). The dose is 90 mg/kg per day of amoxicillin divided in two doses (we suggest a maximum of 3 g/day).

We treat children <2 years for 10 days and those ≥2 years for 5 to 7 days. (See 'Duration of therapy' below.)

●Recent beta-lactam therapy, concomitant purulent conjunctivitis, or history of recurrent AOM unresponsive to
amoxicillin – We suggest amoxicillin-clavulanate as the first-line therapy for children with AOM who are treated
with antibiotics and at increased risk of beta-lactam resistance (ie, have received a beta-lactam antibiotic in the
previous 30 days, have concomitant purulent conjunctivitis, which is usually caused by nontypeable Haemophilus
influenzae [NTHi], or have a history of recurrent AOM unresponsive to amoxicillin because NTHi is dominant in
recurrent episodes) (algorithm 1) [44-47]. The dose is 90 mg/kg per day of amoxicillin and 6.4 mg/kg per day of
clavulanate divided in two doses (we suggest a maximum daily dose of the amoxicillin component of 3 g).
Adolescents ≥16 years who can take large tablets can use extended-release amoxicillin-clavulanate 1 to 2 g of
amoxicillin and 62.5 to 125 mg of clavulanate every 12 hours.

We treat children <2 years for 10 days and those ≥2 years for 5 to 7 days.

Initial antimicrobial therapy for children with penicillin allergy is discussed below. (See 'Penicillin allergy' below.)

A 2010 systematic review concluded there is no evidence to support any particular antibiotic regimen versus
another for treatment of AOM [33,34]. Amoxicillin 90 mg/kg per day divided into two doses (we suggest a maximum
of 3 g/day) is the first-line drug of choice because it is effective, safe, relatively inexpensive, and has a narrow
microbiologic spectrum [3,48,49].

Increasing the dose of amoxicillin from 40 mg/kg per day to 90 mg/kg per day increases the concentration of
amoxicillin in the middle ear [50]. The increased concentration provides activity against most strains of
Streptococcus pneumoniae, including virtually all reported as intermediate (ie, minimum inhibitory concentration
[MIC] ≥2 and <8 mcg/mL to penicillin) [51,52]. S. pneumoniae that are highly resistant to penicillin (ie, MIC ≥8
mcg/mL), less than 2 percent of pneumococcal isolates, will not respond to high-dose amoxicillin [3,53,54].

Despite the increasing importance of NTHi, including beta-lactamase-producing strains, and Moraxella catarrhalis in
the post-pneumococcal conjugate vaccine (PCV) era [55], high-dose amoxicillin remains our preferred choice for
initial therapy in children who have not received a beta-lactam antibiotic in the previous 30 days and do not have
concurrent purulent conjunctivitis [3]. Although some experts suggest standard-dose amoxicillin-clavulanate as the
initial treatment for AOM in the post-PCV era [56], we continue to prefer high-dose amoxicillin because some
children with NTHi AOM clear spontaneously and adverse effects, particularly gastrointestinal adverse effects, are
more common with amoxicillin-clavulanate than amoxicillin [57]. In addition, there is some evidence that high-dose
amoxicillin may be more effective than standard-dose against ampicillin susceptible NTHi.

Continued monitoring of the microbiology of AOM is necessary to determine when and if a change in first-line
therapy is necessary. The increasing presence of beta-lactamase-negative, ampicillin-resistant strains in the
community is a potential concern. These strains require MICs of amoxicillin that may be at the upper limit of what is
achievable in the middle ear. Such isolates remain uncommon in the United States but have become common in
France and Japan [58]. (See "Acute otitis media in children: Epidemiology, microbiology, and complications", section
on 'Bacterial pathogens'.)

In a 2013 meta-analysis of five trials (1601 children <12 years), once or twice daily dosing with amoxicillin or
amoxicillin-clavulanate was similar to three times daily dosing in clinical cure of AOM, recurrence of AOM, and
adverse effects [59]. We suggest twice daily dosing based on pharmacokinetic and pharmacodynamic principles,
results of double tympanocentesis (ie, before and after treatment) studies demonstrating middle ear sterilization
with twice daily regimens, and the lack of any substantial data with once daily regimens (only 33 patients in the
meta-analysis received once daily therapy) [60,61].

Penicillin allergy — Acceptable alternatives to penicillin in patients with allergy to penicillin depend upon the type of
the previous reaction (algorithm 2). Clinical features and diagnosis of immediate and delayed penicillin allergy are
discussed separately. (See "Penicillin allergy: Immediate reactions" and "Allergy evaluation for immediate penicillin
allergy: Skin test-based diagnostic strategies and cross-reactivity with other beta-lactam antibiotics" and "Penicillin
allergy: Delayed hypersensitivity reactions".)

Mild delayed reaction — Mild delayed hypersensitivity reactions to penicillin appear after more than one dose,
typically after days of treatment. They lack features of immunoglobulin E (IgE)-mediated reaction (eg, anaphylaxis,
angioedema, bronchospasm, urticaria) and serious/life-threatening delayed drug reactions (eg, Stevens-Johnson
syndrome, toxic epidermal necrolysis, hemolytic anemia, etc). (See "Penicillin allergy: Delayed hypersensitivity
reactions".)

For children with mild delayed reaction to penicillin antibiotics, we suggest one of the following (algorithm 1):

●Cefdinir 14 mg/kg per day orally in one or two doses (maximum 600 mg/day) for 10 days

●Cefpodoxime 10 mg/kg per day orally in two doses (maximum 400 mg/day) for 10 days

●Cefuroxime suspension 30 mg/kg per day orally divided in two doses (maximum 1 g/day) for 10 days

●Cefuroxime tablets 250 mg orally every 12 hours for 10 days

●Ceftriaxone 50 mg/kg intramuscularly once per day (maximum 1 g/day) for one to three doses (if there is
symptomatic improvement within 48 hours of the first dose, additional doses are not necessary; if symptoms
persist, a second and, if necessary, a third dose are administered [62])

The oral regimens above do not achieve sufficient concentration in the middle ear to eradicate penicillin-resistant S.
pneumoniae and some penicillin-intermediate strains of S. pneumoniae. Cefuroxime is less potent than amoxicillin
against NTHi.

Immediate reaction or serious delayed reaction — Immediate reactions to penicillin classically begin within one
hour of the initial or last-administered dose and have features of IgE-mediated reaction (eg, anaphylaxis,
angioedema, bronchospasm, urticaria). Serious delayed reactions to penicillin include Stevens-Johnson syndrome,
toxic epidermal necrolysis, hemolytic anemia, etc. (See "Penicillin allergy: Immediate reactions" and "Penicillin
allergy: Delayed hypersensitivity reactions".)

Macrolide (eg, azithromycin, clarithromycin) or lincosamide (eg, clindamycin) antibiotics can be used to treat AOM
in children who have had an immediate hypersensitivity reaction or serious delayed reaction to amoxicillin or other
beta-lactam antimicrobial agents (algorithm 1). However, macrolide or lincosamide resistance is common
(approximately 25 to 35 percent) among isolates of S. pneumoniae, and macrolides and lincosamides generally are
not effective for eradication of H. influenzae [54,58,63,64].

Macrolides and lincosamides available for the treatment of AOM include:

●Azithromycin 10 mg/kg per day orally (maximum 500 mg/day) as a single dose on day 1 and 5 mg/kg per day
(maximum 250 mg/day) for days 2 through 5.

●Clarithromycin 15 mg/kg per day orally divided into two doses (maximum 1 g/day).
●The optimal dose for clindamycin therapy for AOM is uncertain; we suggest 20 to 30 mg/kg per day orally divided
in three doses (maximum 1.8 g/day).

Increasing the dose of macrolide antibiotics does not overcome macrolide resistance among pneumococcal isolates
(as with beta-lactam drugs) [48].

Trimethoprim-sulfamethoxazole (TMP-SMX) may be useful in regions where pneumococcal resistance to TMP-SMX

is not a concern, but TMP-SMX should not be used if group A Streptococcus (GAS; Streptococcus pyogenes) is
suspected (eg, when there is an associated otorrhea). Local pneumococcal susceptibility data may be obtained from
state or local health departments or local hospitals. (See 'AOM with perforation' below.)

AOM with perforation — For children with AOM and tympanic membrane perforation (in the absence of a
tympanostomy tube), we suggest oral rather than topical antibiotic therapy. We suggest amoxicillin 90 mg/kg per
day orally divided in two doses (we suggest a maximum of 3 g/day) as the preferred first-line oral therapy. Because
of the possibility of GAS, an agent other than TMP-SMX should be used. For patients with acute otorrhea, 10 days of
oral therapy is more effective than a shorter course [65].

Although topical therapy with quinolone otic drops (ofloxacin or ciprofloxacin) is equivalent to oral therapy for
treatment of otorrhea in children with tympanostomy tubes or chronic suppurative otitis media, topical therapy has
not been studied in children with AOM and acute perforation [66,67]. Nonantimicrobial topical agents, such as
benzocaine, other otic drops, or olive oil, should not be used in patients with perforation of the tympanic
membrane. (See "Tympanostomy tube otorrhea in children: Causes, prevention, and management", section on
'Treatment'.)

AOM in cochlear implant recipient — Management of AOM in children with cochlear implants is discussed
separately. (See "Cochlear implant infections", section on 'Otitis media'.)

Duration of therapy — Our duration of treatment varies with age, associated clinical features, and antimicrobial
agent:

●For amoxicillin, amoxicillin-clavulanate, clarithromycin, oral cephalosporins (eg, cefdinir, cefpodoxime,

cefuroxime), clindamycin, and levofloxacin:

•Ten days for children <2 years and children (of any age) with tympanic membrane perforation or history of
recurrent AOM (see 'First-line therapy' above)

•Five to seven days for children ≥2 years of age with intact tympanic membrane and no history of recurrent AOM

●For azithromycin – Five days

●For ceftriaxone – One to three doses, depending upon persistence of symptoms

Most clinical trials and standard pediatric practice provide a 10-day course of an oral antimicrobial agent for the
treatment of AOM [68-71]. However, some data suggest that a shorter course (ie, seven days) may be adequate for
older children [72].

Many of the studies comparing short- and long-term antibiotic therapy had limitations that precluded definitive
conclusions (eg, lack of blinding, failure to use strict diagnostic criteria or directly compare the same antibiotic
agent) [73]. These limitations were addressed in a trial that randomly assigned 520 children 6 through 23 months of
age with strictly defined AOM to treatment with amoxicillin-clavulanate for 10 days or for 5 days followed by 5 days
of placebo [70]. The rate of clinical failure (defined as worsening of symptoms or otoscopic signs of AOM during
treatment or lack of complete/near-complete resolution of symptoms and signs at the end of treatment) was lower
among those assigned to 10 days of treatment (16 versus 34 percent; difference of 17 percentage points [before
rounding], 95% CI 9-25). However, the trial has been criticized for using amoxicillin-clavulanate rather than
amoxicillin and having an unusually high five-day failure rate [14,74]. The rate of adverse events did not differ
between groups. The trial did not address duration of treatment in children older than two years.

INITIAL OBSERVATION

Initial observation is an alternative to antimicrobial therapy for children ≥2 years who are normal hosts (eg, immune
competent, without craniofacial abnormalities), without otorrhea, and who have mild symptoms and signs of
unilateral AOM (ie, nonsevere ear pain for <48 hours and temperature <39°C [102.2°F]). Clinicians who recommend
initial observation should exercise rigor in diagnosing AOM similar to that in the research protocols that support the
safety of this strategy. (See 'Antibiotic therapy versus observation' above and "Acute otitis media in children:
Clinical manifestations and diagnosis", section on 'Diagnosis'.)

When the initial observation strategy is chosen, caretakers must understand the risks and benefits, and appropriate
follow-up must be ensured so that antibiotic therapy can be initiated if symptoms worsen or persist after 48 to 72
hours [3]. Unilateral AOM at first observation may rapidly progress to bilateral disease during the early hours of
illness. Adequate follow-up may include a parent-initiated visit or phone contact if symptoms worsen or do not
improve at 48 to 72 hours, a scheduled follow-up appointment in 48 to 72 hours, or giving parents an antibiotic
prescription that can be filled if illness does not improve in this time frame [26,27,75-78]. Careful follow-up is critical
to the initial observation/delayed antibiotics strategy. Specific indications for urgent follow-up (eg, signs of
meningismus, cranial nerve palsy) should be provided to patients or caregivers and documented in the medical
record. (See 'Follow-up' below.)

The 2013 American Academy of Pediatrics and American Academy of Family Physicians guideline suggests initial
observation (with pain control) as an option for healthy children (ie, without conditions that predispose to AOM)
between 6 and 24 months with unilateral nonsevere AOM and for children ≥24 months with unilateral or bilateral
nonsevere AOM [3]. This is a change from the 2004 guidelines, which suggested initial observation as an option for
children ≥6 months with "uncertain" diagnosis and children ≥2 years with nonsevere illness [79].

Guidelines from many other countries (eg, the Dutch College of General Practitioners) recommend no antibiotics or
delayed antibiotics for most children ≥6 months of age with AOM.

FOLLOW-UP

Persistent symptoms — Children who fail to improve after 48 to 72 hours of antibiotic therapy should be seen in
follow-up to confirm the diagnosis of AOM, evaluate other causes of persistent symptoms, and determine whether
a change in antibacterial therapy is warranted [3]. (See 'Treatment failure' below.)

Children who worsen or fail to improve after 48 to 72 hours of initial observation and symptomatic treatment
without antibiotics should be started on antibiotics. Those who worsen should be seen before initiating antibiotics;
those who fail to improve can initiate antibiotics without an office visit. (See 'Initial antimicrobial therapy' above.)

Resolved symptoms — Follow-up for children whose symptoms have resolved depends upon the child's age and
underlying medical problems, particularly language delay or learning problems.

We suggest that:

●Children <2 years be seen 8 to 12 weeks after diagnosis (by which time middle ear effusion will have resolved in 80
to 90 percent (figure 1)); many such children will already have a routine health care visit scheduled within this time
frame

●Children ≥2 years who have language or learning problems be seen 8 to 12 weeks after diagnosis AOM

●Children ≥2 years who are without language or learning problems be followed up at their next health maintenance
visit, or sooner if there are concerns regarding persistent hearing loss

The main reason for follow-up of children with resolved symptoms is to monitor the resolution of middle ear
effusion which is associated with conductive hearing loss. Persistent middle ear effusion is common after the
resolution of acute symptoms. In a large prospective study, middle ear effusion persisted for weeks to months after
the onset of AOM in children (figure 1) [9]. (See 'Clinical course' above.)

The management of persistent middle ear effusion is discussed separately. (See "Otitis media with effusion (serous
otitis media) in children: Clinical features and diagnosis", section on 'Hearing loss' and "Otitis media with effusion
(serous otitis media) in children: Management", section on 'Approach to management'.)

Tympanic membrane perforation — Tympanic membrane perforation permits drainage of the middle ear abscess
and relieves increased middle ear pressure. With the relief of middle ear pressure, the tympanic membrane usually
heals quickly, sealing the perforation in hours to days.

If pain occurs or persists in a child with AOM and tympanic membrane perforation, causes other than AOM must be
considered. The pain is unlikely to be due to AOM because the pressure of middle ear fluid is relieved when the
tympanic membrane is perforated. Other possible causes include:

●Extension of the infection to a contiguous space, such as the mastoid (ie, mastoiditis) (see "Acute mastoiditis in
children: Clinical features and diagnosis", section on 'Diagnosis' and "Acute mastoiditis in children: Treatment and
prevention", section on 'Management approach')

●Irritation of the external canal from middle ear drainage, resulting in otitis externa, in which case treatment with a
topical quinolone may be beneficial (see "External otitis: Treatment", section on 'Topical preparations')

Nonantimicrobial topical agents, such as benzocaine or olive oil, should not be used to treat pain in patients with
perforation of the tympanic membrane. These agents are not effective and may be harmful.
Patients with perforation that persists for three months or longer (with or without suppurative drainage) should be
referred to an otolaryngologist for further management [80]. Prevention of chronic suppurative otitis media entails
prompt and appropriate treatment of AOM [81]. Chemoprophylaxis is not warranted. (See "Chronic suppurative
otitis media (CSOM): Clinical features and diagnosis" and "Chronic suppurative otitis media (CSOM): Treatment,
complications, and prevention".)

TREATMENT FAILURE

Definition and etiology — Treatment failure is defined by lack of improvement in symptoms by 48 to 72 hours in a
patient treated with antimicrobial therapy. (See "Acute otitis media in children: Clinical manifestations and
diagnosis", section on 'Clinical presentation'.)

Fluid may persist in the middle ear for prolonged periods, even when the antimicrobial agents have sterilized the
effusion and the acute signs and symptoms are no longer present. Persistent middle ear effusion after the
resolution of acute symptoms is not an indication of treatment failure or an indication for additional antibiotic
therapy [82]. (See 'Clinical course' above.)

Treatment failure suggests either the initial therapy was not adequate or another disease is present. (See
'Approach' below and "Acute otitis media in children: Clinical manifestations and diagnosis", section on 'Differential
diagnosis'.)

Inadequate therapy is usually related to infection with an organism resistant to beta-lactam antibiotics
(nontypeable H. influenzae [NTHi] and drug-resistant S. pneumoniae are becoming increasingly important), but
infection with less common organisms, such as Staphylococcus aureus, also must be considered, particularly in
children with tympanostomy tubes or perforation [83-85]. (See "Acute otitis media in children: Epidemiology,
microbiology, and complications", section on 'Microbiology' and "Tympanostomy tube otorrhea in children: Causes,
prevention, and management", section on 'Pathogens'.)

Approach

Initial treatment failure

●Patients initially treated with amoxicillin – We suggest that patients who fail treatment with high-dose amoxicillin
be treated with amoxicillin-clavulanate (algorithm 1) [3]. The dose is 90 mg/kg per day amoxicillin and 6.4 mg/kg per
day of clavulanate divided into two doses (we suggest a maximum of 3 g/day). Adolescents ≥16 years who can take
large tablets can use extended-release amoxicillin-clavulanate 1 to 2 g of the amoxicillin component and 62.5 to 125
mg of the clavulanate component every 12 hours.

We prefer amoxicillin-clavulanate for children who fail treatment with amoxicillin because of its efficacy against
beta-lactamase-producing H. influenzae and M. catarrhalis; for S. pneumoniae, amoxicillin and amoxicillin-
clavulanate have equivalent efficacy.

●Patients initially treated with amoxicillin-clavulanate or oral cephalosporins – Patients who fail initial treatment
with high-dose amoxicillin-clavulanate or oral cephalosporins may be treated with parenteral ceftriaxone
(preferred) or oral levofloxacin (algorithm 1):

•Ceftriaxone 50 mg/kg intramuscularly or intravenously once per day (maximum 1g/day) for two or three doses.

Parenteral ceftriaxone 50 mg/kg achieves high levels in the middle ear and is effective for the treatment of AOM in
children who fail amoxicillin [3]; although the US Food and Drug Administration (FDA) has approved a single dose of
parenteral ceftriaxone for the treatment of AOM in children, an open-label prospective study suggested that three
doses were superior in eradicating penicillin-resistant S. pneumoniae from the middle ear [86].

•Levofloxacin 10 mg/kg orally every 12 hours for 10 days for children six months to five years or 10 mg/kg per orally
once daily for 10 days for children ≥5 years (maximum 750 mg/day) [87].

Levofloxacin should be reserved for children with contraindications to ceftriaxone or AOM refractory to other drugs
(ideally it should only be used in children who have had serotype 19A isolated from the middle ear that is
susceptible to levofloxacin) [85]; levofloxacin is not approved by the FDA for the treatment of AOM, and
levofloxacin resistance among S. pneumoniae respiratory isolates has been described in adults and rarely in children
[88].

Trimethoprim-sulfamethoxazole (TMP-SMX), macrolides (eg, azithromycin, clarithromycin), and lincosamides (eg,

clindamycin) are not recommended for AOM that fails to respond to treatment with high-dose amoxicillin or
amoxicillin-clavulanate. Pneumococcal surveillance studies indicate that resistance to these agents is substantial
[53,89]. Macrolides and lincosamides have limited activity against NTHi, which is a more likely pathogen among
children who have failed initial amoxicillin therapy.
●Patients initially treated with macrolides, clindamycin, or parenteral ceftriaxone – The management of treatment
failure in children with AOM who were initially treated with macrolides, clindamycin, or parenteral ceftriaxone is
challenging. The pathogens most likely in this scenario are NTHi (either beta lactamase-positive or beta lactamase-
negative) or multidrug-resistant S. pneumoniae. Consultation with a pediatric infectious diseases expert and/or
pediatric otolaryngologist may be warranted (algorithm 1).

Our suggested approach is as follows:

•Tympanocentesis (if available), which will provide relief of ear pain and permit culture and susceptibility testing to
guide antimicrobial selection. Concomitant placement of a tympanostomy tube may be warranted. A discussion of
the tympanocentesis procedure is beyond the scope of this topic review but is available in the full text of reference
[90].

•If tympanocentesis is not available, we suggest levofloxacin 10 mg/kg orally every 12 hours for 10 days for children
six months to five years or 10 mg/kg per orally once daily for 10 days for children ≥5 years (maximum 750 mg/day)
[87]. Levofloxacin is not approved by the FDA for the treatment of AOM and should be reserved for AOM refractory
to other drugs.

Most NTHi are susceptible to TMP-SMX, but susceptibility of S. pneumoniae varies geographically. TMP-SMX may be
an option in areas with high rates of susceptibility.

Persistent treatment failure — Referral to a pediatric otolaryngologist and/or pediatric infectious diseases expert
may be warranted for children with persistent treatment failure. Tympanocentesis is recommended to make a
bacteriologic diagnosis if symptoms persist despite ceftriaxone or other broad spectrum therapies [85].
Tympanocentesis should be performed by an appropriately trained clinician who is comfortable performing the
procedure in an awake child (unless there is access to sedation or anesthesia). Alternatively, treatment with
levofloxacin and/or tympanostomy tube placement may be appropriate [85].

RECURRENT AOM

Recurrent AOM is defined by the development of signs and symptoms of AOM within 30 days after completion of
successful treatment. It is particularly important to establish the diagnosis of recurrent AOM with bulging of the
tympanic membrane or other signs of inflammation. Otherwise, persistent middle ear effusion in a child with a
febrile upper respiratory infection may be misinterpreted as a recurrent episode and the child may receive
antibiotics unnecessarily.

There are no randomized trials to guide treatment of recurrent AOM in children. The approach is based on our
understanding of the microbiology in children with recurrent AOM and varies depending upon the therapy that was
used for recent episodes. In the era of universal immunization with the 13-valent pneumococcal conjugate vaccine,
treatment of recurrent AOM must include coverage for resistant pathogens: S. pneumoniae as well as nontypeable
H. influenzae and M. catarrhalis.

When recurrence occurs within 15 days of completion of antimicrobial treatment for the previous episode, it is
most often due to persistence of the original pathogen; we suggest:

●Ceftriaxone 50 mg/kg per day intramuscularly (IM) or intravenously (IV) for three days, or

●Ceftriaxone 50 mg/kg per dose IM or IV every 36 hours for a total of two doses (although this regimen has not
been studied, it is suggested based on pharmacokinetics and middle ear drug levels), or

●Levofloxacin 10 mg/kg every 12 hours orally for 10 days for children 6 months to 5 years or 10 mg/kg per once
daily for 10 days for children ≥5 years (maximum 750 mg/day) [87]

When the recurrence occurs more than 15 days after completion of the treatment for the previous episode, it is
most often due to a different pathogen than the previous episode. Although the child is at higher risk for a
nonsusceptible pathogen, we suggest high-dose amoxicillin-clavulanate as initial therapy, even if the child received
amoxicillin-clavulanate for the previous episode.

Tympanostomy tube insertion may be warranted for children with ≥3 distinct and well-documented episodes of
AOM within 6 months or ≥4 episodes within 12 months if middle ear fluid is also present [91]. (See "Acute otitis
media in children: Prevention of recurrence", section on 'Tympanostomy tubes'.)

Evaluation of the immune system also may be warranted for children with ≥4 episodes of AOM within 12 months.
However, if the spectrum of infection in the child is limited to recurrent AOM, a clinically concerning immune
deficiency is unlikely. (See "Approach to the child with recurrent infections", section on 'Clinical features suggestive
of a primary immunodeficiency'.)

TREATMENT OF AOM COMPLICATIONS

AOM may be associated with intratemporal or intracranial complications. The treatment of these complications is
discussed separately:

●Cholesteatoma (see "Cholesteatoma in children", section on 'Surgical treatment')

●Mastoiditis, which may be complicaed by petrositis (osteomyelitis of the petrous bone) (see "Acute mastoiditis in
children: Treatment and prevention")

●Facial nerve palsy, which may be isolated or associated with osteomyelitis of the petrous bone (see "Facial nerve
palsy in children", section on 'Treatment' and "Acute mastoiditis in children: Treatment and prevention", section on
'Complicated disease')

●Meningitis (see "Bacterial meningitis in children older than one month: Treatment and prognosis")