Otitis media is a broad term that includes acute otitis media (AOM), otitis media with effusion (OME), chronic otitis media with effusion (COME), and chronic suppurative otitis media (CSOM). This chapter will focus on AOM and OME.
Incidence and Epidemiology of Acute Otitis Media
AOM primarily affects children in the first 3 years of life. Onset of AOM in the first 6 months of life is not common because infants are relatively protected from infection mediated by maternal antibodies acquired transplacentally. If a child experiences AOM in the first 6 months of life, then frequent AOM episodes are likely to occur throughout the first few years of life. Most AOM occurs between the age of 6 and 24 months of age. The peak incidence is between 9 and 15 months of age. AOM does occur with modest frequency between 2 and 3 years of age but then quickly diminishes in occurrence between 3 and 5 years of age.
The occurrence of otitis media in otherwise healthy infants is partly a reflection of their immature immune system and partly due to the fact that the eustachian tube of the young child is shorter, floppier, straighter, and more horizontal than the eustachian tube of the older child. By 3 years of age, the incidence of AOM decreases because of changes in the child’s anatomy and physiology and maturing immune system. Children who have had little or no otitis media by the time they reach age 3 years are unlikely to develop problems with middle ear infections later in life unless some predisposing factor, such as allergic rhinitis, tumor or fracture of the base of the skull or a facial bone, or acquired immunodeficiency, occurs.
AOM is the most common bacterial infectious disease seen in an ambulatory pediatric practice, a leading cause of health care visits, and the most frequent reason children consume antibiotics or undergo surgery. Reported AOM ambulatory visits in U.S. children younger than 2 years of age average 1244 visits per 1000 child-years, and 80% of those visits resulted in an antibiotic prescription. AOM has a high socioeconomic impact worldwide. In the United States, an estimated $4 billion is spent yearly on otitis media–related health care.
More recently several developments have impacted AOM incidence in the United States and worldwide. First, treatment guidelines endorsed higher dose amoxicillin for the empiric treatment of AOM due to penicillin-resistant pneumococcal infections. Second, conjugate vaccines against Streptococcus pneumoniae , a major pathogen of AOM, were introduced. Routine use of the 7-valent pneumococcal conjugate vaccine (PCV7) in early childhood was associated with a significant reduction in AOM visit rates as well as a decrease in pressure-equalizing tube (PET) insertion due to recurrent AOM and CSOM.
Three database analyses in three different countries demonstrated a positive effect of PCV7 on AOM episodes. Marom et al. analyzed an insurance claims database of a U.S. nationwide managed health care plan. The authors showed a decreased trend in otitis media–related health care use in children younger than 6 years. Magnus et al. studied children participating in the Norwegian Mother and Child Cohort Study and showed a decline of mother-reported AOM episodes in Norwegian children 12 to 18 months old and 18 to 36 months old of 14% and 8%, respectively. Using a national primary care database, an observational cohort study investigated trends in AOM incidence and associated antibiotic utilization in children younger than 10 years during 2002 to 2012 in the United Kingdom. The authors found that the introduction of PCV7 was associated with a 22% reduction in AOM in children younger than 10 years and an additional 19% reduction following PCV13 introduction.
Otitis media in high-risk populations continues to be a significant medical problem. For example, epidemiologic studies from Bangladesh, Nigeria, and Australia report prevalence of chronic suppurative otitis media (CSOM) of 12%, 2.5%, and 15%, respectively. Australian aboriginal children have a high rate of AOM and CSOM with tympanic membrane perforation that leads to significant long-term health impairment. Native American children have a 63% incidence of AOM by 6 months of age. A systematic review of population-based studies from multiple countries found the highest prevalence of OM in children to be 81% of Canadian Inuits and 84% of Australian Aborigines.
Risk Factors
Many significant aspects of OM epidemiology and natural history were first characterized in 1960–1985. During those years it was shown that AOM risk increased according to the child’s age at first AOM episode, male gender, Caucasian race/ethnicity (except in special high-risk populations noted earlier), low birth weight (<2500 g), preterm birth (<37 weeks’ gestation), fall season of birth, bottle-feeding, daycare attendance, higher number of young siblings, lower parent’s education/income/occupation, more frequent history of ear infections (plus other infectious illnesses or conditions) in siblings or parents, absent or poor health insurance coverage, and exposure to secondhand cigarette smoking. Table 16.1 shows results of a meta-analysis of risk factors for chronic AOM and recurrent AOM.
| Risk Factor | Odds Ratio | 95% Confidence Interval |
|---|---|---|
| Allergy/atopy | 1.36 | 1.13–1.64 |
| Upper respiratory infection | 6.59 | 3.13–13.89 |
| Chronic nasal obstruction | 1.19 | 0.84–1.69 |
| Snoring | 1.96 | 1.78–2.16 |
| Male gender | 1.24 | 0.99–1.54 |
| Daycare center attendance | 1.70 | 0.95–3.05 |
| Family history of otitis media | 1.40 | 0.86–2.28 |
| Patient history of AOM/rAOM | 11.13 | 1.06–116.44 |
| Passive smoke exposure | 1.39 | 1.02–1.89 |
| Low socioeconomic status | 3.82 | 1.11–13.15 |
| Low education level of mother | 1.68 | 0.32–8.68 |
| Mother’s smoking during pregnancy | 2.34 | 0.64–8.54 |
| Multiple siblings | 1.57 | 0.93–2.63 |
| Breastfeeding >6 mo | 0.57 | 0.17–1.93 |
| Any breastfeeding | 0.91 | 0.47–1.79 |
Microbiology of Acute Otitis Media in the Pneumococcal Conjugate Vaccine Era
The bacteria that cause AOM vary from country to country due to vaccination and antibiotic prescribing habits because these preventative and therapeutic interventions impact the otopathogens that predominate in the nasopharynx before ascending via the eustachian tube to the middle ear space. In most developed countries virtually all children are vaccinated with the PCV7 or PCV13, discussed later in this chapter. Most children are treated with antibiotics, predominantly amoxicillin in a standard dose (40 mg/kg per day divided twice daily) or a “high” dose (80 mg/kg per day divided twice daily) for 10 days. As a consequence of PCV and amoxicillin use, the etiology of AOM continues to change over time. The dynamic changes in the otopathogen mix from 1996 to 2015, based on tympanocentesis isolations from the Rochester, New York, otitis media research center are shown in Fig. 16.1 .
S. pneumoniae caused approximately 30% to 60% of episodes before the PCV era. Following the introduction of PCV7, S. pneumoniae as a cause of AOM decreased in frequency, and nontypeable Haemophilus influenzae AOM increased proportionately. Two years after PCV7 introduction, S. pneumoniae began to increase proportionately due to nonvaccine serotypes, dominated by serotype 19A. There was complete elimination of PCV7 serotypes and replacement with nonvaccine serotypes in the nasopharynx and as a cause of AOM in the countries using PCV7. The otopathogens causing AOM in young children in 2008 to 2010, the late PCV7 era in the Rochester otitis media research center, are shown in Table 16.2 and Fig. 16.1 . After licensure of PCV13, a prompt transition from PCV7 to PCV13 occurred in 2010 in the United States. A decrease in nasopharyngeal colonization caused by S. pneumoniae serotypes 6A and 19A followed. A decline in S. pneumoniae and a proportional increase in nontypeable H. influenzae and M. catarrhalis AOM occurred ( Fig. 16.1 ). Pneumococcal otopathogens from the middle ear cultures from children from 2011 to 2013 showed a significant decrease in PCV13 serotypes and in penicillin-resistant isolates. The most common non-PCV13 serotypes were 35B; 21; 23B; 15A, B, C; 11; and 23A.
| Year | Total Visits | Otopathogen Isolated From MEF | No. (%) |
|---|---|---|---|
| 2008 | 57 | S. pneumoniae | 18 (32) |
| Nontypeable H. influenzae | 16 (28) | ||
| M. catarrhalis | 10 (18) | ||
| 2009 | 78 | S. pneumoniae | 32 (41) |
| Nontypeable H. influenzae | 18 (23) | ||
| M. catarrhalis | 14 (18) | ||
| 2010 (until September 30) | 73 | S. pneumoniae | 21 (29) |
| Nontypeable H. influenzae | 25 (34) | ||
| M. catarrhalis | 5 (7) |
Pneumococcal AOM in children younger than 2 years has been studied following the sequential introduction of PCV7 and PCV13. In the PCV7 era in Israel there was a 77% decline in pneumococcal AOM episodes. In the post-PCV13 era, there was a 51% decline in PCV13 serotypes, and the most common non-PCV13 serotypes isolated from middle ear fluid (MEF) were 15B/C, 16F, and 35B. Overall the trend in AOM cases has declined since the introduction of PCVs according to other reports in the United States, the United Kingdom, and Israel. In 2015, M. catarrhalis became the most common cause of AOM in the Rochester otitis media research center.
Etiology in Neonates
Other than the more frequent occurrence of disease caused by gram-negative enteric organisms (approximately 20% of cases) and the occasional isolation of other neonatal pathogens (e.g., group B streptococci), the bacteriology of otitis media in this age group is similar to that in older children.
Pathophysiology
Tympanic Membrane
As inflammation develops in the middle ear space, changes in the tympanic membrane occur rapidly. The presence of congested blood vessels, edema (which obscures normal landmarks), and bulging occurs, indicating not only a myringitis (inflammation of the tympanic membrane) but also the presence of fluid under pressure in the middle ear space. Sometimes blebs appear on the surface epithelium as a consequence of AOM and the heat of the inflammatory process (most often associated with group A streptococcal infections of the middle ear).
Inflammation may occur on outer epithelial or inner mucosal sides of the fibrous layer (middle layer) of the tympanic membrane. In severe cases, infection may involve the fibrous layer itself. The membrane thickens as a result of edema and infiltration of polymorphonuclear leukocytes. All three layers of the tympanic membrane may undergo dissolution owing to pressure necrosis resulting from the expanding middle ear abscess or thrombophlebitis of tympanic veins, with resulting perforation. With evacuation of the contents of the middle ear abscess after perforation or therapeutic tympanocentesis, healing may be rapid, and the perforation usually seals within a few days. In the process of healing, scar formation may occur.
Occasionally, when a perforation is close to the margin of the annulus or occurs in the Shrapnell membrane, the skin of the external auditory canal and the surface squamous epithelium of the tympanic membrane may grow through the aperture and invade the middle ear. This event may lead to formation of a cholesteatoma (epidermal inclusion cyst). Even if the perforation heals, a differential in gas pressure across the tympanic membrane caused by malfunction of the eustachian tube may result in resorption of gas in the middle ear cavity and negative pressure in the middle ear, which causes retraction of the Shrapnell membrane or an atrophic scar into the middle ear or mastoid attic.
Eustachian Tube
The eustachian tube is approximately 3.8 cm long in an adult. It opens in the fossa of Rosenmüller and extends upward, backward, and laterally to open in the upper anterior wall of the tympanic cavity (protympanum). At birth, the eustachian tube is immature; that is, it is much shorter and floppier than at 1 year of age. It is composed of two portions: the cartilaginous portion extending into the nasopharynx and the bony portion originating in the middle ear. The upper third of the tube is bony; the middle ear opening is the widest, and the medial end (the part joining the cartilaginous eustachian tube), or isthmus, is the narrowest. Pneumatic peritubal air cells arising from the middle ear cavity surround it and can extend to the petrous apex. The internal carotid artery courses anteromedial to this region.
The lower two-thirds of the eustachian tube is a narrow, slit-like, fibrocartilaginous passage. It makes a 160-degree angle with the bony portion at its junction.
Three muscles are associated with the eustachian tube. The tensor tympani muscle lies on top of it; the levator veli palatini muscle lies under it; and the tensor veli palatini muscle arises on the tube, scaphoid fossa, and spine of the sphenoid and then courses around the hook of the hamulus and forms an aponeurosis with its mate (from the opposite side) in the soft palate. This muscle is the only one that acts directly on the eustachian tube.
The eustachian tube area, protympanum, and hypotympanum are lined by ciliated columnar epithelium with goblet cells or secretory cells. The epithelium is continuous with the upper airway system and paranasal sinuses. This area also contains a well-defined subepithelial connective tissue layer, which thins out and may be absent nearing the antrum and mastoid air cell system. The movement of the cilia and mucous blanket always is toward the eustachian tube and nasopharynx. The tube is surrounded by a plexus of lymphoid channels. It has an arterial supply from a branch of the middle meningeal or accessory meningeal artery and from branches of the artery of the pterygoid canal. The nerve supply is from the tympanic plexus (ninth cranial nerve; sensory) and sphenopalatine ganglion (sympathetics and parasympathetic palatine fiber).
The bony portion is rigid and patulous; the medial two-thirds normally is held closed by elastic recoil of the fibrocartilaginous tissue. Contraction of the tensor palatini muscle that inserts in the anterolateral wall opens the tube on swallowing. On average, an adult swallows once per minute while awake and once every 5 minutes while asleep. Suckling infants usually swallow five times per minute.
Mucus and ciliary action flow from the middle ear to the eustachian tube. The eustachian tube acts as a unidirectional valve that favors outflow from the middle ear to the pharynx. Reverse flow can be induced by an increase in pressure in the nasopharynx (Valsalva, barotrauma). During occlusion of the eustachian tube, oxygen and carbon dioxide (and other gases) are absorbed from the middle ear by diffusion into the rich vasculature, and a negative pressure is created. A patent eustachian tube is a crucial prerequisite for subsidence of middle ear disease.
Pathogenesis
Approximately 35% of upper respiratory infection (URI) episodes are complicated by AOM, which occurs usually within the first week of URI onset, and 90% to 95% of children with AOM have concurrent URI symptoms. Disease etiology and pathogenesis are complex and begin with colonization of mucosal surfaces in the upper respiratory tract by potential AOM bacterial pathogens ( Fig. 16.2 ). In order to colonize the nasopharynx, otopathogens must first compete with each other and with commensal bacteria in the nasopharynx. Nontypeable H. influenzae outcompetes S. pneumoniae (except serotype 19A) and M. catarrhalis for higher rates of colonization during AOM. The combination of respiratory bacteria, S. pneumoniae , nontypeable H. influenzae , and M. catarrhalis, colonizing the nasopharynx and the introduction of a viral URI results in complex interactions that set the stage for the development of AOM.
Viral URI sets the stage for AOM and OME by increasing mucus production, slowing the beat of cilia in the nasopharynx, creating an inflammatory environment in the nasopharynx, and downregulating the innate and adaptive immune response (discussed later in the section “ Immunology ”). Next the eustachian tube closes due to inflammation caused as part of the viral URI process. Negative pressure builds in the middle ear space as the air diffuses across the tympanic membrane, resulting in a retracted tympanic membrane. The middle ear goblet cells produce mucus to keep the surface of the middle ear cells moist, and, with the eustachian tube closed, the mucous builds up and causes fluid to be visible behind the tympanic membrane. This is OME. However, when OME develops, the middle ear is still sterile, and the absence of a virus or bacteria and an inflammatory process results in the tympanic membrane remaining generally translucent. Fig. 16.3 shows the continuation of OME and AOM.
As the negative pressure develops in the middle ear space, the secretions in the nasopharynx are literally sucked into the middle ear space when the eustachian tube temporarily relaxes for a part of a second. Once the secretions and accompanying virus and bacteria that were in the nasopharynx gain entry to the middle ear space, the environment is free of immune-controlling factors, and the bacteria begin to divide. In response to the local bacterial invasion, the innate immune response is activated, resulting in the influx of neutrophils. The neutrophils release mediators of inflammation, and the health care provider observes that pathogenic process when glimpsing a tympanic membrane that has become thickened with edema, perhaps red, but most important bulging from the pressure of an inflammatory response (see the section “Clinical Manifestations and Diagnosis”).
Bacterial otopathogens and respiratory viruses interact and play important roles in AOM development. However, much confusion surrounds the role of URI viruses as an etiology of AOM and OME. While there is no doubt that viral URI plays a key role in the pathogenesis of AOM and OME, the role is more facilitation of bacterial AOM than a primary etiologic role for these viruses (see the section “ Immunology ”). Respiratory syncytial virus (RSV), influenza, parainfluenza, rhinovirus, metapneumovirus, and other viruses can be detected in the nasopharynx secretions of children with URI followed by AOM. The nasopharynx secretions can reflux from the nasopharynx via the eustachian tube into the middle ear space, so the detection of the virus in MEF by tympanocentesis does not confirm that virus is the etiology of AOM. With modern polymerase chain reaction (PCR) techniques, the DNA or RNA of viruses can be detected in the nasopharynx and MEF of many children with AOM and OME. The presence of a respiratory virus without a bacterial otopathogen simultaneously detected is uncommon, probably around 2% to 10% of all cases of AOM.
OME can persist for some time after an AOM. About half of children who experience AOM will have OME 1 month after initial diagnosis, one-third have OME 2 months after AOM, and 10% have OME 3 months after AOM ( Fig. 16.4 ).
Diagnosis
Symptoms and Clinical Manifestations of AOM
Symptoms associated with AOM and its complications and sequelae include the following:
- 1.
Otalgia, or ear pain, is the most common feature of infants and children with AOM. The symptom is suggested in young infants who are pulling at the ear or excessively irritable. Some infants do not have earache.
- 2.
Otorrhea is drainage from the middle ear through a perforation in the tympanic membrane. Relief of the pressure on the tympanic membrane results in immediate decrease in pain. Because the tympanic membrane has a dense network of blood vessels, rapid repair of the membrane occurs, and the perforation usually closes within 24 to 72 hours. If the tympanic membrane seals and the infection still is present, fluid may reaccumulate with renewed acute signs of AOM.
- 3.
Hearing loss occurs whenever fluid fills the middle ear space, whether the fluid is associated with acute infection or with OME. When fluid fills the middle ear space, the median hearing loss is 25 dB (the equivalent of having plugs in the ear canals).
- 4.
Vertigo is not a common complaint of children with AOM. Vertigo occurs more commonly in unilateral than bilateral disease and may be caused by labyrinthitis. Older children describe a feeling of spinning, whereas younger children may not be able to verbalize these symptoms but manifest disequilibrium by falling or stumbling.
- 5.
Tinnitus is an uncommon complaint in children, but when it does occur, the symptom often is caused by OME and eustachian tube dysfunction.
- 6.
Swelling around the ear, especially in the postauricular area, may be a sign of mastoiditis.
- 7.
Facial paralysis in children occurs as a complication of AOM or chronic otitis media with perforation of the tympanic membrane or as a result of an enlarging cholesteatoma.
- 8.
Conjunctivitis has been associated with AOM because the organisms that cause AOM originate in the nasopharynx and may simultaneously cause AOM and conjunctival infection.
There is ample evidence to confirm that the diagnosis of AOM cannot rely on clinical manifestations. Upper respiratory viral infections and AOM occur simultaneously and share many nonspecific symptoms. Table 16.3 shows the sensitivity and specificity of a number of symptoms young children have at the time of evaluation for possible AOM. Parents who suspected that their child aged 6 to 35 months old had AOM were correct only half of the time when the clinicians used strict otoscopic criteria. Half had a URI but no AOM. Ear pain, ear rubbing, fever, irritability, restless sleep, and severe or prolonged rhinitis/cough does not increase the probability of AOM.
| Symptoms | Sensitivity (%) | Specificity (%) |
|---|---|---|
| Cough | 47–84 | 45–83 |
| Ear pain | 54–60 | 82–92 |
| Ear rubbing | 42 | 87 |
| Excessive crying | 55 | 69 |
| Fever | 40–69 | 23–48 |
| Headache | 9 | 76 |
| Poor appetite | 36 | 66 |
| Restless sleep | 64 | 51 |
| Rhinitis | 75–96 | 43–92 |
| Sore throat | 13 | 74 |
| Upper respiratory tract infection | 96 | 29 |
| Vomiting | 11 | 89 |
AOM symptom scoring scales have been developed to aid in the diagnosis of AOM, to assign severity scores, and to assist in the evaluation of response to treatment or watchful waiting in young children with AOM. Table 16.4 shows AOM scoring systems and the items assessed. A 3-item score (OM-3), consisting of physical symptoms including ear pain, fever, emotional distress (irritability or poor appetite), and limitation in activity, and a 5-item score (Ear Treatment Group Symptom Questionnaire, 5 Items; ETG-5), including fever, earache, irritability, decreased appetite, and sleep disturbance, have been validated. A parental assessment of symptom severity using a visual scale of faces, Acute Otitis Media–Faces Scale (AOM-FS), also has been developed. None of these three scales was adequately sensitive for making the diagnosis of AOM based only on symptoms. However, the OS-8 scale, consisting of a set of tympanic membrane photographs that the clinician could use to grade the severity of tympanic membrane inflammation and combined with the faces scale (OM-FS) to assess the parental perception of severity of symptoms, was able to show changes in the child’s symptoms and signs of AOM. Another validated scale included a 7-item parent-reported symptom score (Acute Otitis Media Severity of Symptom Scale–AOM-SOS) for children with AOM to aid in assessing patient-reported outcome measures during clinical trials. Symptoms included ear tugging/rubbing/holding, excessive crying, irritability, difficulty sleeping, decreased activity or appetite, and fever. The AOM-SOS scale, when compared with otoscopic diagnoses (AOM, OME, and normal), showed that it, too, changed in response to clinical improvement or deterioration similar to the OS-8 scale.
| 3-Item Otitis Media Score (OM-3) | Ear Treatment Group Symptom Questionnaire (ETG-5) | Acute Otitis Media Faces Scale (AOM-FS) | Otoscopic Severity Scale (OS-8) | Acute Otitis Media Severity of Symptom Scale (AOM-SOS) | Otitis Media Clinical Severity Index (OM-CSI) 30-Point Scale a | Otitis Media Clinical Severity Index (OM-CSI) 10-Point Scale a |
|---|---|---|---|---|---|---|
| Physical suffering | Ear pain | Seven facial expressions ranging from no problem to extreme problem | Eight categories of TM inflammation b | Ear pain | Ear pain | Ear pain |
| Emotional distress | Fever | Ear tugging | Fever | Fever | ||
| Limitation of activities | Irritability | Irritability | Irritability | Irritability | ||
| Appetite | Decreased play | Fever at examination | Fever at examination | |||
| Sleep quality | Decreased appetite | TM erythema | TM erythema | |||
| Difficulty sleeping | TM mobility | TM mobility | ||||
| Fever | TM position | TM position | ||||
| Effusion color | Effusion color | |||||
| Otorrhea | Otorrhea |
a The 30-point scale used a 2- to 5-point Likert scale and the 10-point scale used a 2- to 3-point Likert scale.
b 0 = normal; 1 = erythema only; 2 = erythema, air-fluid level, clear fluid; 3 = erythema, complete effusion, no opacification; 4 = erythema, opacification with air-fluid level or air bubbles, no bulging; 5 = erythema, complete effusion, opacification, no bulging; 6 = erythema, bulging rounded doughnut appearance of the tympanic membrane; 7 = erythema, bulging, complete effusion and opacification with bulla formation.
A 30-point symptom scoring instrument (Otitis Media Clinical Severity Index, OM-CSI) was developed to determine the severity of AOM and to measure the treatment outcome of AOM in young children. A shorter, easier-to-use 10-point scoring instrument was developed and compared to the 30-point scoring instrument; both the 10-point and 30-point scoring systems were able to measure treatment outcome and differentiated between clinical cure and failure at the follow-up test-of-cure visit.
Diagnostic Signs of Acute Otitis Media
AOM is a visual diagnosis based on viewing the tympanic membrane. The accurate diagnosis of AOM in infants and young children can be difficult. Having the proper equipment and the proper positioning of the child are critical to achieve the correct diagnosis. It is not possible to diagnose AOM accurately without complete visualization of the tympanic membrane. A glimpse of a small portion of the tympanic membrane is not sufficient to make the diagnosis of AOM. It is recommended that the clinician take the time and make the effort to clear all or nearly all of the ear canal cerumen, use optimal lighting, and use the largest ear speculum that can fit snuggly into the child’s ear canal so that a seal can be made for pneumatic otoscopy.
Examination of the Ear
Otoscopy
Examination of the ear should begin with observation of the auricle and the external auditory meatus. Palpation of the periauricular areas should be done to indicate presence of periostitis or diffuse external otitis. The ear canal should be examined for inflammation or cerumen that obstructs vision of the tympanic membrane. Proper positioning of the child for optimal examination of the tympanic membrane is shown in Fig. 16.5 .
For proper assessment of the tympanic membrane and its mobility, a pneumatic otoscope in which the diagnostic head has a secure seal should be used. The speculum should have the largest lumen that can fit comfortably into the child’s cartilaginous external auditory meatus ( Fig. 16.6 ). The important landmarks of the tympanic membrane that can be visualized with the otoscope are indicated in Fig. 16.7 .
The otoscopic examination should include observation of the following conditions of the tympanic membrane:
- •
Position: Normal is slightly convex; bulging indicates increased pressure from positive air pressure or fluid; a retracted eardrum indicates negative pressure with or without effusion; fullness of the tympanic membrane is apparent initially in the posterosuperior portion of the pars tensa and the pars flaccida because these two areas are the most highly compliant parts of the membrane.
- •
Appearance and color: The normal color is pearly gray and translucent; any congestion of the mucous membrane of the middle ear would be reflected in congestion of the vessels of the tympanic membrane and appear pink; a blue discoloration suggests blood in the middle ear associated sometimes with basal skull fracture; the inflamed middle ear mucosa usually is reflected in a bright red tympanic membrane.
- •
Integrity of the membrane: All four quadrants of the tympanic membrane should be inspected for presence or absence of perforation, retraction pockets, or cholesteatoma.
- •
Mobility: Application of positive and negative pressures by the pneumatic otoscope enables the viewer to determine the presence of an air-filled space (rapid movement of the membrane on positive and negative pressures) or a fluid-filled space (limited or no movement of the membrane). Fig. 16.8 shows a normal tympanic membrane, a retracted tympanic membrane, and a bulging tympanic membrane.
FIG. 16.8
(A) Normal tympanic membrane with a neutral position, pearly gray in color and translucent. Normal landmarks are easily visible. (B) Otitis media with effusion: a retracted tympanic membrane with a very prominent head of the malleolus, gray/yellow in color and semiopaque, with air-fluid level and air bubble visible. (C) Acute otitis media: bulging tympanic membrane with complete loss of normal landmarks, white/yellow in color, and fully opaque with injected vasculature.
(Courtesy Dr. Hoberman, University of Pittsburgh.)
In 2013 the American Academy of Pediatrics (AAP) and the American Academy of Family Physicians (AAFP) released a revision and update of their 2004 AOM guidelines. The 2013 updated guidelines used a more stringent diagnostic definition of AOM. Bulging of the tympanic membrane was emphasized as the most specific otoscopic sign of AOM. Three studies evaluated otoscopic signs—position, color, mobility of the tympanic membrane—and correlated the signs with the presence of middle ear effusion or the isolation of an otopathogen following myringotomy or tympanocentesis. Karma et al. looked at the tympanic membrane physical findings of color, position, and mobility in children with and without acute symptoms suggestive of AOM. Of all the tympanic membrane characteristics, a bulging tympanic membrane reliably predicted the presence of middle ear effusion. Halsted et al. described the likelihood of a positive tympanocentesis culture given different tympanic membrane characteristics. The authors found a positive bacterial culture in 91% of children who had a bulging tympanic membrane, in 21% with a minimally bulging or full tympanic membrane, and in 0% of children without a bulging tympanic membrane. Last, McCormick et al. studied the association between certain signs and symptoms of AOM and the presence of bacterial or viral isolates from middle ear fluid. The authors concluded that the single most important clinical feature associated with a pathogen was fullness and bulging of the tympanic membrane. In all these studies, erythema or lack of mobility of the tympanic membrane was less likely to be associated with middle ear effusion or the isolation of a bacterial pathogen from the middle ear fluid. From those and other studies the key diagnostic feature of AOM was established as a bulging or full appearance to the tympanic membrane ( Fig. 16.8C ). The bulging of the tympanic membrane is due to positive pressure behind the tympanic membrane caused by inflammation in the middle ear space. AOM is not associated with a retracted tympanic membrane, so a determination of retraction of the tympanic membrane is a viral-mediated phenomenon or associated with OME ( Fig. 16.8B ). A bulging compared to a retracted tympanic membrane can be difficult to distinguish. Use of pneumatic otoscopy is very helpful in distinguishing the two because positive pressure on insufflation will result in movement backward by a bulging tympanic membrane, and negative pressure will result in movement forward by a retracted tympanic membrane.
Because of the inflammation in the middle ear space during AOM, the tympanic membrane becomes thickened and semi-translucent or completely opaque. A translucent tympanic membrane is very unlikely to be associated with AOM. A translucent or semi-translucent tympanic membrane and middle ear fluid visualized behind the tympanic membrane point to a likely diagnosis of OME (see Fig. 16.8 ).
Redness of the tympanic membrane is not considered a valuable diagnostic sign of AOM. Redness occurs from inflammation, but a red tympanic membrane can also occur from fever or from a crying child (and most children of the age who experience AOM cry during the examination). When an otoscope speculum is inserted into the external auditory canal, the tympanic membrane will sometimes turn red. The presence of one red tympanic membrane and the other tympanic membrane not red suggests inflammation of the tympanic membrane and is consistent with the diagnosis of AOM if there is fluid visualized behind the tympanic membrane. Such an examination most likely represents an early AOM before inflammation has persisted long enough to cause the tympanic membrane to become thickened and more yellowish and opaque.
OME is not associated with inflammation; however, a child with OME may feel discomfort and may feel popping as air gains entry via the eustachian tube into the middle ear space as well as hearing loss. This may cause the child to pull and tug at the ear or even cry. But the visual examination is distinct from AOM, as just outlined. Fig. 16.9 shows the transition from OME to AOM.
Tympanometry
The tympanogram identifies the movement of the tympanic membrane in response to positive and negative pressure. It utilizes the concept that sound energy in a closed space is diluted or concentrated as the volume increases or decreases ( Fig. 16.10 ). The movement or compliance of the tympanic membrane that occurs following a pulse of air pressure is recorded graphically. The device sweeps pressure from 200 dekapascals (daPa, a unit of pressure equal to 1.04 mm of water) to −300 to 600 daPa. A number of different curves are generated by the tympanometer depending on the state of the middle ear and the patency of the tympanic membrane. A sharply peaked curve with maximum compliance at normal pressure is designated as an A curve and is usually seen with a normal tympanic membrane and middle ear space. A flat curve, referred to as a B curve, is found when there is no movement of the tympanic membrane, when the middle ear is full of fluid, or when there is a perforation of the tympanic membrane. A curve with a peak that occurs at an abnormal amount of negative pressure is referred to as a C curve. A C curve is often seen when there is negative pressure within the middle ear space and the tympanic membrane is retracted ( Fig. 16.11 ). In some cases, the tympanic membrane will be full or bulging, and the tympanic reading is then called a positive pressure tympanogram readout . Tympanometry requires a complete seal of the inserted speculum-like device in the external auditory canal and a few seconds of patient cooperation. Children younger than 2 years of age, when AOM is most common, often move, and a seal cannot be obtained. Also if a child is crying a tympanogram reading cannot be obtained.
Acoustic Reflectometry
Spectral gradient acoustic reflectometry (SGAR) has been validated as an accurate method of testing for middle ear effusion in children and infants older than 3 months of age ( Fig. 16.12 ). An 80-dB spectrum of sound, 1.6 to 4.7 kH, is emitted from the device and directed toward the tympanic membrane. The device measures the sum of the reflected sound energy. A normal tympanic membrane and air-filled middle ear space absorb the majority of the sound, and very little acoustic energy reflects back to the device, giving a high numeric readout. If the sound waves hit a thickened tympanic membrane and fluid in the middle ear space, less sound is absorbed and more sound is reflected back to the device, giving a lower numeric readout. This instrument does not require a seal in the external auditory canal, and readings can be obtained in the crying child. The main limitation is the presence of cerumen in the external auditory canal.
A comparison of tympanometry and SGAR in the ability to assess the presence of middle ear effusion using pneumatic otoscopy as the gold standard showed an agreement between the two methods of 86%. The detection of middle ear effusion by tympanometry and SGAR compared to diagnosis by pneumatic otoscopy showed excellent agreement. Table 16.5 shows the percentage of ears with middle ear effusion documented by pneumatic otoscopy versus the level of SGAR. SGAR instruments detected middle ear effusion with 67% sensitivity and 87% specificity and performed similarly to pneumatic otoscopy. The ease of use, easy portability, absence of need for an airtight seal, and low cost make the use of SGAR an excellent aid in the diagnosis of middle ear effusion.
| Level | Predicted Risk of MEE | N | Ears With Documented MEE Based on Pneumatic Otoscopy Examination (%) |
|---|---|---|---|
| 1 | Low | 383 | 3 |
| 2 | Low-moderate | 279 | 16 |
| 3 | Moderate | 82 | 34 |
| 4 | Moderate-high | 76 | 58 |
| 5 | High | 50 | 92 |
Audiometric Testing
When fluid is present in the middle ear space, it can cause diminished hearing. An audiogram can assist in establishing the extent of hearing loss for both AOM and OME. Testing the hearing of young children is difficult in clinical settings other than audiology practices that can utilize brainstem evoked responses. Thus, although a potentially useful tool to quantify hearing loss, audiometry is not often used in the first few years of life when AOM and OME most frequently occur. This paradox is problematic for compliance with national recommendations for management of OME, where the presence of unilateral hearing loss of 6 months’ duration or bilateral hearing loss of 3 months’ duration to greater than 30-decibel thresholds in the speech range (500–2000 Hz) is a primary criterion for the recommendation of insertion of tympanostomy tubes.
Tympanocentesis and Myringotomy
Tympanocentesis, a needle aspiration of the middle ear effusion, is used primarily for establishing the presence or absence of an effusion and for microbiologic diagnosis. The 2013 AAP/AAFP AOM guidelines endorse the use of tympanocentesis in the hands of a person skilled in the procedure for treating children with recurrent AOM or treatment failure. Because cultures of the upper respiratory tract are of limited value in providing specific microbiologic diagnosis of AOM, only bacterial cultures obtained by aspiration of the middle ear abscess can be considered a true reflection of the etiology of AOM.
Myringotomy is an incision in the anterior lower quadrant of the tympanic membrane for therapeutic drainage. Tympanocentesis or myringotomy should be considered in patients who at onset of AOM appear toxic or are seriously ill, in patients who are not clinically responding after initiation of empiric antimicrobial therapy, in the presence of suppurative complications (including mastoiditis and meningitis), in immunologically deficient patients in whom an unusual organism may be present, and for recurrent AOM in the otitis-prone child.
Radiography
Radiographic evaluation of the temporal bone is indicated when complications or sequelae of otitis media are suspected or present. Plain radiographs are of limited value in the diagnosis of osteitis of the mastoid or cholesteatoma; computed tomography and magnetic resonance imaging are more precise and should be obtained if a suppurative intratemporal or intracranial complication is suspected.
Management of Acute Otitis Media
In the United States the principal treatment of AOM is antibiotics, whereas in other countries watchful waiting with repeat examinations for mild AOM is the norm. Table 16.6 shows the dosing schedule for the antibiotics with an indication for the treatment of AOM. The antibiotic chosen to treat AOM should have a spectrum of activity that includes coverage of the most common otopathogens S. pneumoniae, nontypeable H. influenzae, and M. catarrhalis and have documented clinical and microbiologic efficacy, limited side effects, availability in a convenient dosage schedule, palatability when provided in suspension, and reasonable cost. When treated with appropriate antimicrobial therapy, a patient should have substantial resolution of signs and symptoms within 72 hours and absence of signs of relapse, recurrence, or suppurative sequelae.
| Agent | 24-Hour Dosage |
|---|---|
| Amoxicillin | 40–80 mg/kg in 2–3 doses |
| Amoxicillin-clavulanate | 40–100 mg/kg in 2 doses |
| Cefprozil | 30 mg/kg in 2 doses |
| Cefpodoxime | 10 mg/kg in 2 doses |
| Cefaclor | 40 mg/kg in 2–3 doses |
| Cefixime | 8 mg/kg in 1 dose |
| Cefuroxime axetil | 30 mg/kg in 2 doses |
| Loracarbef | 30 mg/kg in 2 doses |
| Ceftriaxone | 50 mg/kg in 1 dose (1–3 days) |
| Ceftibuten | 9 mg/kg in 1 dose |
| Cefdinir | 14 mg/kg in 1–2 doses |
| Erythromycin-sulfisoxazole | 50 mg/kg erythromycin, 150 mg/kg sulfisoxazole in 4 doses |
| Clarithromycin | 15 mg in 2 doses |
| Azithromycin | 30 mg/kg in 1 dose (1 day) |
| 10 mg/kg in 1 dose (3 days) | |
| 10 mg/kg in 1 dose (day 1); 5 mg/kg in 1 dose (days 2–5) | |
| Trimethoprim-sulfamethoxazole | 8 mg trimethoprim, 40 mg sulfamethoxazole in 2 doses |
The 2013 AAP/AAFP AOM guidelines for initial management of uncomplicated AOM was updated. The updated guidelines differentiate treatment based on the child’s age, unilateral or bilateral AOM, and severity of the child’s symptoms. High-dose amoxicillin (80–100 mg/kg per day, with a maximum dose of 3 g) is the recommended first-line treatment in most patients. The selection of high-dose amoxicillin has been made on the basis of the long-term safety of the drug; a first intention to treat penicillin-resistant S. pneumoniae since that organism can cause the most morbidity; and the recognition that overdiagnosis of AOM is commonplace, and, as a consequence, emergence of antibiotic-resistant microbes occurs more frequently with broader spectrum antibiotics. However, the guidelines point out that a number of antibiotics are clinically effective. In children who have been treated with amoxicillin in the prior 30 days or who have purulent conjunctivitis or in those children who are more likely to have β-lactamase–producing nontypeable H. influenzae or M. catarrhalis , therapy should be initiated with high-dose amoxicillin-clavulanate (80–100 mg/kg per day of amoxicillin, with 6.4 mg/kg per day of clavulanate), a ratio of amoxicillin to clavulanate of 14 : 1, given in two divided doses, which is less likely to cause diarrhea than other amoxicillin-clavulanate preparations.
Following the introduction and wide use of PCV7 and now PCV13 in the United States, nontypeable H. influenzae and M. catarrhalis are more common causes of AOM than penicillin-nonsusceptible S. pneumoniae . Given the current otopathogen mix and antibiotic resistance, the best choice of antibiotic would be high-dose amoxicillin-clavulanate (90 mg/kg per day divided twice daily). Amoxicillin would not be considered the most effective antibiotic for empiric selection because it has no activity against β-lactamase–producing bacteria, currently far more common than penicillin-resistant S. pneumoniae . The addition of clavulanate as a β-lactamase–neutralizing product would provide anticipated efficacy against β-lactamase–producing H. influenzae and M. catarrhalis while maintaining excellent antimicrobial activity against penicillin-resistant S. pneumoniae . There are two disadvantages to amoxicillin-clavulanate: the suspension formulation has a marginal taste that can cause nonadherence to the prescribed regimen, and it causes more loose stools and diarrhea than amoxicillin and many cephalosporin alternatives.
The oral cephalosporins of choice for treatment of AOM, as designated by the AAP, are cefdinir, cefuroxime axetil, and cefpodoxime proxetil. Among these choices in the United States, cefdinir has emerged as the most frequently used, largely because the bitter taste of the other two drugs can be an obstacle for adherence. Cefdinir can be dosed once or divided twice daily (14 mg/kg per day). The duration of treatment with cefdinir can be 5 days with twice-daily dosing or 10 days with once-daily dosing. In a head-to-head comparison of amoxicillin-clavulanate 80 mg/kg per day divided twice daily for 10 days versus cefdinir 14 mg/kg per day divided twice daily for 5 days, amoxicillin-clavulanate demonstrated superior efficacy. However, outside the context of a clinical trial, the adherence characteristics favor cefdinir (better taste and less diarrhea).
The taste of cefuroxime and cefpodoxime can be masked with chocolate syrup, but the addition of flavorings at the pharmacy should not be advised since these antibiotics have been shown to precipitate with changes in pH and chemical reactions between the active drug and the flavoring ingredients.
Ceftriaxone by injection (50 mg/kg per dose) is among the preferred antibiotics in the case of initial antibiotic treatment failure in AOM. Ceftriaxone is effective as a single injection against all penicillin-susceptible S. pneumoniae and against β-lactamase–producing H. influenzae and M. catarrhalis. Three doses of ceftriaxone are needed for penicillin-resistant S. pneumoniae . In light of the long half-life of ceftriaxone, it may be possible to administer the sequential doses spaced every other day or even every third day if weekends or holidays dictate an alternative regimen.
Macrolides, such as erythromycin and azithromycin, have poor efficacy against both S. pneumoniae and H. influenzae . Clindamycin is also included as a treatment alternative for AOM following failure of other preferred first- and second-line agents. Clindamycin is ineffective against β-lactamase–producing H. influenzae and M. catarrhalis and only effective against penicillin-susceptible and penicillin-resistant S. pneumoniae . Its use might best be limited to cases where a tympanocentesis has been performed and the persisting bacteria and antimicrobial susceptibilities are identified.
Antibiotics not listed in the AAP/AAFP guidelines are omitted because of concerns about poor efficacy, poor adherence, or safety. Therefore, the use of alternatives not included in the guideline should be undertaken after due consideration.
Tympanocentesis as Treatment
Tympanocentesis is a treatment for AOM if performed with the intention of evacuating pus, microbes, and proinflammatory fluid from the middle ear space. Not all tympanocenteses are performed with evacuation of middle ear fluid, and that fact complicates the interpretation of available clinical studies of the therapeutic benefit of the procedure. Tympanocentesis can be performed in an office practice setting without anesthesia or conscious sedation in the hands of a person skilled in the procedure. Instillation of 8% tetracaine into the external canal and placement of an otowick to assure the anesthetic reaches the tympanic membrane are effective to allow the procedure to occur without the child experiencing pain. However, the child must be restrained to avoid head movement during the procedure, and children typically cry when they are restrained. Tympanocentesis is not a frequently performed procedure because most clinicians have not been trained how to do it, and those who have been trained become concerned about its use as a standard of care despite the AAP and U.S. Centers for Disease Control recommendation for selected use in AOM management. Tympanocentesis alone is not as effective as tympanocentesis followed by antibiotics if evacuation of middle ear pus is not achieved.
Using a stringent clinical diagnosis (bulging or full tympanic membrane, cloudy or purulent effusion, and reduced or absent tympanic membrane mobility) followed by tympanocentesis for microbiologic diagnosis of AOM, the effects on recurrent AOM and tympanostomy tube placement in children younger than 3 years of age have been studied. The use of strict diagnostic criteria, tympanocentesis, and empiric antibiotic treatment using evidence-based knowledge of circulating otopathogens and their antimicrobial susceptibility profile reduces the frequency of recurrent AOM and tympanostomy tube surgery.
Watchful Waiting
In 2004, the AAP/AAFP endorsed a recommendation for watchful waiting as an option for management of AOM in selected cases. The main concept was to allow observation as an option when the diagnosis is uncertain (unlikely in AOM) as long as the child was older than 2 years and reevaluation could occur if the child did not improve in 48 to 72 hours or worsened at any time. The AAP guidelines may have resulted in a reduction of AOM by establishing precise criteria for diagnosis, but a study indicated that there was no evidence that the pattern of prescribing antibiotics for AOM was altered. Because of the limitations of the clinical trials that supported the strategy of watchful waiting, investigators in Pittsburgh, Pennsylvania, and Turku, Finland, independently conducted a randomized blinded trial of amoxicillin-clavulanate compared with placebo in children younger than 2 and 3 years, respectively. Amoxicillin-clavulanate showed a significant benefit in the duration of acute signs of illness as compared with placebo. The 2013 AAP/AAFP AOM guidelines removed the “uncertain” diagnosis and encouraged practitioners to make every effort to use more stringent criteria for the diagnosis of AOM. The guidelines again endorsed the watchful waiting recommendation for children older than 2 years of age with bilateral or unilateral AOM without otorrhea and in children of any age with unilateral AOM without otorrhea and severe symptoms. Again there must be a mechanism in place for prompt reassessment if the child does not improve within 48 to 72 hours or if the child worsens at any time.
Pain Management
There can be considerable pain with AOM, especially in the first 24 hours, and pain management for AOM has received more attention with the publication of the AAP’s policy statement on the assessment and management of acute pain in infants, children, and adolescents. A child with suspected or confirmed AOM should be given pain treatment regardless of the use of antibiotics. Various treatments of otalgia have been used; however, none has been well studied. Typically pain control is managed with oral analgesics such as acetaminophen or ibuprofen in weight-appropriate doses. The use of ototopic analgesic ear drops is also to be considered, although the evidence of efficacy is limited.
Antihistamines and decongestants taken orally or intranasally are not recommended as treatment for AOM or OME because they either have been shown to be nonefficacious or have not been studied at all. They may provide some symptomatic treatment of nasal allergies or nasal congestion.
Duration of Treatment
The optimal duration of antibiotic treatment is generally considered to be 10 days in the United States; however, there is scant evidence for that recommendation. Instead, the 10-day treatment course for AOM was derived from the 10-day treatment course for streptococcal pharyngitis, which is evidence based. Treatment regimens of 1 day, 3 days, 5 days, 7 days, and 10 days are all standard in different countries. The 2013 AAP/AAFP guidelines endorse 10 days of treatment duration as the standard for most AOM but state that shorter treatment regimens may be as effective. A systematic analysis and a meta-analysis have concluded that 5 days’ duration of antibiotics is as effective as 10 days of treatment for all children older than age 2 years and only marginally inferior to 10 days for children under the age of 2 years. A comparison of 5-, 7-, and 10-day treatments of AOM concluded that 5-day treatment was equivalent to 7- and 10-day treatment for all ages unless the child had a perforated tympanic membrane or had been treated for AOM within the preceding month (because recently treated AOM was associated with more frequent causation of AOM by resistant bacteria and with a continued inflamed middle ear mucosa).
Treatment in the Penicillin-Allergic Child
The use of cephalosporins in penicillin-allergic children has been recently reevaluated. The cephalosporins selected by the AAP took consideration of the likelihood of cross-reaction with penicillin. Second- and third-generation cephalosporins have chemical side chain structures sufficiently distinct from penicillin and amoxicillin that they may be used in penicillin- and amoxicillin-allergic children. The β-lactam ring is not the structure accountable for allergy; it is the side chain. There is some cross-reactivity among cephalosporins based on their shared chemical side chain structure ( Tables 16.7 and 16.8 ). The possibility of allergy to all penicillins and cephalosporins has never been confirmed as a reality, and statistically the likelihood of such a case is indeed very small.
