After completing this article, readers should be able to:
Break down the percentage of cases of chronic otitis media with effusion (OME) that yield bacterial growth on culture and which bacteria they yield.
Describe the effect of antibiotic treatment for acute OM on long-term resolution of middle ear effusion.
List the risk factors that increase the duration of OME and the risk of chronic OME.
Break down the percentage of children who experience hearing loss following chronic OME.
Describe the child who requires referral to an otolayrngologist for myringotomy with insertion of ventilation tubes.
Otitis media is a common childhood disease that accounts for nearly $4 billion in health-care expenditures annually. The term otitis media often is used to describe any one of a continuum of related diseases: acute otitis media, recurrent acute otitis media, otitis media with effusion (OME), and chronic otitis media with effusion. OME involves middle ear effusion (MEE) behind an intact tympanic membrane without signs or symptoms of acute infection. The effusion may be serous, mucoid, purulent, or some combination of these (Fig 1⇓ ). OME is defined as chronic when MEE has been present for at least 3 months. The Agency for Health Care Policy and Research (AHCPR) guidelines for management of OME defined chronic OME as MEE persisting for 4 or more months. The term“ chronic otitis media” refers to intractable middle ear or mastoid tissue pathology (eg, granulation tissue or cholesteatoma) behind an intact or perforated tympanic membrane rather than simple chronic OME. Chronic OME and recurrent acute otitis media frequently overlap, and children who have one condition are much more likely to develop the other. Whatever the label, one of the principal concerns associated with unremitting MEE is reduced hearing, which may interfere with speech and language development.
Otitis media is an inflammatory condition of the middle ear that is initiated by functional or mechanical obstruction of the eustachian tube, which leads to alteration in the partial pressure of middle ear gases, resulting in negative middle ear pressure (Fig 2⇓ ). Transudation of capillary serous fluid into the middle ear space follows. Nasopharyngeal bacteria may invade the middle ear space via the obstructed eustachian tube and replicate within the serous middle ear fluid. Bacterial and inflammatory host cell products are released into MEE and tissues, attracting peripheral blood leukocytes and leading to a cascade of acute inflammatory events that result in symptomatic acute otitis media.
Eustachian tube obstruction is caused most commonly by events in the upper respiratory tract that injure epithelial cells in the nasopharyngeal portion of the tube, which leads to increased secretion of mucus by distal tubal epithelial cells and obstruction of the tubal lumen by these mucoid secretions and sloughed cellular debris (Fig 3⇓ ). The chinchilla has been used as an animal model to understand the pathogenesis of acute otitis media and OME. Infection with influenza virus and adenoviruses has been shown to cause eustachian tube dysfunction. Further, respiratory syncytial virus and parainfluenza virus infections in children often are complicated by otitis media, suggesting that these viruses also have an influenza-like effect on tubal epithelium. It seems reasonable that other mucosal irritants, such as passive smoke exposure, might have a similar deleterious effect on tubal epithelial cells. This may explain the increased risk of otitis media in young children exposed to passive smoke.
Cleft palate and certain congenital malformations of craniofacial development also cause dysfunction in the tubal opening. Abnormal patulence of the tube has been described, and this condition may be associated with more frequent regurgitation of nasopharyngeal secretions into the middle ear because there is no protective effect of the usually closed tube at rest. The increased frequency of otitis media among young infants may be associated with the shorter length and flatter position of the eustachian tube relative to the posterior nasopharyngeal wall compared with that of older children and adults. The eustachian tube increases in length and diameter and develops a more acute angle relative to the posterior nasopharyngeal wall with increasing age.
Middle ear gas pressure relationships are illustrated in Figure 4⇓ . With normal eustachian tube function, the partial pressure of oxygen, carbon dioxide, and nitrogen are equivalent in middle ear and atmospheric air, but the nitrogen content of the surrounding middle ear tissues is at a lower partial pressure than in atmospheric or middle ear air. With tubal obstruction, middle ear nitrogen is absorbed into mixed venous blood in the subepithelial tissues, which lowers total middle ear gas pressure, resulting in negative middle ear pressure and a retracted tympanic membrane. Transudation of serous fluid from subepithelial capillaries into the obstructed middle ear space reduces the middle ear air volume, thereby increasing middle ear pressure.
INCIDENCE AND PREVALENCE
OME occurs most commonly in the fall, winter, and spring, with point prevalence estimates of 3% to 25% among United States and Scandinavian children ages 6 months to 11 years. Studies of clearance of acute otitis media with antimicrobial treatment show that MEE persists for 1 month in 30% to 50% of children, for 2 months in 15% to 25%, and for 3 months in 8% to 15% (Fig ⇓ ). Antibiotic treatment for acute otitis media has a negligible effect on the long-term resolution of MEE. A meta-analysis of randomized antibiotic trials conducted among more than 5,400 children revealed that the resolution rate among placebo-treated or untreated children 7 to 14 days after the onset of acute otitis media was 81%, and it improved to 95% with antibiotic treatment. However, MEE had resolved 30 days after initiation of treatment in 65% of controls and 60% of children treated with antibiotics. Resolution was not related to the type of antimicrobial agent employed.
Because MEE is a labile condition, widely spaced examinations are not useful in estimating the continuous duration of OME. Studies reporting the incidence of chronic OME typically have evaluated persistence of MEE after an episode of acute otitis media or identified children who had OME and examined them at short intervals to determine resolution or persistence of the condition. Using these methods, 4% to 65% of children who had otitis media developed chronic OME, with the majority of estimates ranging between 4% and 20% (Fig 6A⇓ ). The 4% estimate is derived from a Finnish study that used the stringent criteria of OME of 2 months’ duration diagnosed with pneumatic otoscopy, tympanometry, and tympanocentesis and verified every 2 to 3 weeks plus confirmation of mucoid effusion, thick tympanic membrane, or thick mucosa at surgery. The 65% estimate is from a study of North Carolina children who were examined every 1 to 2 weeks with pneumatic otoscopy and tympanometry in child care centers. Studies in which frequent examinations were employed have reported mean or median OME duration in the first 2 years of life to be 58 to 72 days per year (15% to 20% of a child’s life), which represents significant morbidity.
TRENDS OVER TIME
Although there is evidence that the rate of otitis media has increased over the past 2 decades, there are no surveillance-based estimates to monitor changes in the rates of chronic OME. However, rates of myringotomy from the National Hospital Discharge Survey rose from 22.3 to 41.1 per 10,000 children younger than 15 years between 1970 and 1977 and then began declining as tube surgeries moved from hospitals to ambulatory surgery centers. Data from the 1995 National Ambulatory Survey showed that 90.2 per 10,000 children had tube surgery in an ambulatory setting. This rate is equivalent to the expected 1995 rate, based on the observed rates of increase in tube surgeries that occurred in the 1970s. A comparison of United States studies of children 3 years of age and younger also suggests an increase in chronic OME over time (Fig 6B⇑ ), a finding that must be interpreted cautiously because diagnostic methods, disease definitions, frequency of examination, and group characteristics varied across studies.
Several studies have evaluated the role of demographic and environmental variables on the duration, persistence, and chronicity of OME. Studies that have large sample sizes, involve frequent examinations, and include multivariate analyses to control for confounding elements provide the best information about risk factors for duration and chronicity of OME. These studies have reported an increased risk of chronic OME or a longer duration of OME associated with the presence of environmental factors (group child care, number of hours in child care, exposure to children at home or in child care, number of smokers and cigarettes smoked in the household, feeding in the supine position, shorter duration of breastfeeding, autumn season) and characteristics of the child or the specific disease history (early onset of otitis media, several prior episodes, bilateral OME, male gender, lower socioeconomic status, having a sibling with a history of otitis media). Many of these factors also have been implicated in increasing the risk of otitis media and recurrent otitis media.
SIGNS AND SYMPTOMS
Clinically, eustachian tube obstruction and resulting negative middle ear pressure can be diagnosed by examination with the pneumatic otoscope; the tympanic membrane is retracted and mobility is reduced. Tympanometry reveals a negative peak pressure. OME is diagnosed when tympanic membrane mobility to pneumatic insufflation is reduced and the tympanogram has very low static admittance or is flat. In such a case, the membrane usually is white or pink and opaque. With bacterial replication and acute inflammation, the tympanic membrane becomes injected with capillary dilatation and diffuse erythema. Although otoscopes have been marketed for home use, there is no evidence that they are useful in detecting chronic OME because middle ear effusion is not apparent by visual inspection without testing the mobility of the tympanic membrane.
Pneumatic otoscopy requires skill, experience, and the proper equipment. The sensitivity and specificity of pneumatic otoscopy have been established at academic centers with validated otoscopists, who primarily were otolaryngologists. In these ideal settings, pneumatic otoscopy is about 95% sensitive and 80% specific for MEE, using myringotomy as the gold standard. Otoscopy in the clinic probably is less reliable in detecting OME, particularly if pneumatic otoscopy is not performed, because fluid often cannot be seen behind the tympanic membrane. For this reason, tympanometry, which has higher specificity but is slightly less sensitive than pneumatic otoscopy, is a very useful supplement, particularly when the diagnosis of OME is in question.
If the presence of fluid is identified by pneumatic otoscopy, tympanometry can provide additional information about degree of membrane mobility, but it probably will not add diagnostic value to the identification of OME. It is important to understand that results of the two procedures (pneumatic otoscopy and tympanometry) will not always agree and that the combination provides a more accurate diagnosis. In one study, the combined sensitivity of pneumatic otoscopy and tympanometry was 97%, with a specificity of 90%.
Recently, the value of tympanometry in healthy preschool children has been demonstrated. Use of combined static admittance width and height criteria for tympanometry in this population has high sensitivity (84%) and specificity (95%), with a positive predictive value of 69% and a negative predictive value of 98%. Another recent study of tympanometry with validated pneumatic otoscopy as the gold standard showed that the single best criterion for detection of OME was a tympanometric width greater than 250 dekaPascals (daPa).
An example of a normal tympanogram and the measurements typically made are shown in Figure 7⇓ . Many instruments automatically calculate admittance, width, and tympanometric peak pressure (TPP), although the values may not be reliable if there are multiple peaks due to the child moving or crying. A template also may be used to measure admittance and width easily. TPP, the position of the peak on the pressure axis, is measured in daPa (0 daPa for this example), although some instruments measure in mm H2O. Compensated admittance is the height of the peak determined by subtracting the admittance value obtained at +200 daPa from the admittance value at TPP (0.8 for this example). Admittance is measured in mmho, although some instruments measure in mL. Tympanometric width in daPa is measured halfway down from the peak to the point at +200 daPa. For example, the vertical lines show the width at the halfway point down from the peak (50 daPa for this example).
Chronic OME can be extremely difficult to detect because symptoms often are absent or subtle and may be mistaken for developmental problems or changes. Parents should be given information about symptoms of hearing loss (Table 1⇓ ), particularly if their children have experienced recurrent otitis media (three or more episodes in 12 months) or previous chronic OME. Parents should be encouraged to have their children’s ears examined whenever there is concern about the presence of OME or hearing loss.
MICROBIOLOGY AND BIOCHEMISTRY
The most common bacteria invading the acutely obstructed middle ear space are Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis; these three bacteria account for 85% of acute middle ear infections. The action of natural immune defense mechanisms, including the diffusion of bacteria-specific immunoglobulin G (IgG) antibodies into the serous transudate, plus antimicrobial treatment alter the bacteriology of chronic OME. In chronic OME, approximately 70% of MEE yield bacterial growth on culture, but only 30% to 35% yield S pneumoniae, H influenzae, or M catarrhalis. The remaining 35% to 40% of effusions yield other bacteria that have reduced virulence, including coagulase-negative staphylococci.
MEE and middle ear mucosal characteristics change with the evolution of middle ear inflammation. The biochemistry of serous transudate during the initial inflammatory events following tubal obstruction is very similar to that of serum except for an increase in oxidative and hydrolytic enzymes. Purulent effusion of acute otitis media is an acute inflammatory exudate that contains abundant polymorphonuclear leukocytes, increased protein content relative to serum, and extremely high concentrations of inflammatory cytokines and eicosanoids (leukotrienes and prostaglandins). Evolution of purulent effusion in some cases is associated with metaplasia of middle ear epithelium and a dramatic alteration in epithelium from the normally squamous and cuboidal to pseudostratified columnar epithelium and goblet cells secreting a mucous glycoprotein. The molecular characteristics of the secretory process currently are being explored in animal models, and results suggest that several of the inflammatory cytokines trigger the secretory metaplasia of these epithelial cells.
Managing OME requires a knowledge of the pathogenesis, epidemiology, and natural history of otitis media. Because upper respiratory viral infection is common among infants and young children, episodes of eustachian tube dysfunction with resulting negative middle ear pressure occur frequently, and serous transudate is a common accompanying event. Prospective epidemiologic studies have shown that many episodes of negative middle ear pressure with or without serous MEE are not complicated by bacterial invasion and acute otitis media. Only a small proportion of these episodes leads to chronic OME. Therefore, managing the child who has OME requires several examinations over a period of time to document that the MEE is, in fact, chronic (ie, persisting for 3 months). Because serous MEE may be transient, medical or surgical intervention for this condition usually is not warranted. In contrast, surgical intervention with placement of tympanostomy tubes, adenoidectomy, or both procedures generally is indicated in managing chronic OME.
In 1994, the Agency for Health Care Policy Research (AHCPR), in collaboration with the American Academy of Pediatrics, the American Academy of Family Physicians, and the American Academy of Otolaryngology-Head and Neck Surgery, extensively evaluated published clinical trials of chronic OME management to develop guidelines for managing this condition. Their target child was healthy, was 1 to 3 years of age, and had no cranio-facial or neurologic abnormalities. The panel concluded that whenever OME is suspected, its presence should be confirmed by pneumatic otoscopy or tympanometry. When OME is documented, parents should be counseled regarding OME risk factors (passive smoke exposure, attendance at child care), and the child should be re-examined in 6 weeks (Fig 8⇓ ). A child who has bilateral fluid for a total of 3 months should undergo hearing evaluation, which also is an option for children who have effusion of shorter duration, according to the AHCPR panel. Several large prospective studies have shown that 70% of children in whom chronic effusion lasts 3 or more months suffer mild or moderate hearing loss. Since the guidelines for treatment of OME were published, evidence has mounted that unilateral hearing loss may be associated with a decreased ability to process sounds binaurally (using information from both ears), which may affect the ability to detect speech in background noise. Further research will be needed to determine if unilateral hearing loss also should be treated with myringotomy and a tympanostomy tube in the affected ear.
Medical treatment of this condition has been unsatisfactory. Meta-analysis of controlled studies using antimicrobial drugs to treat acute otitis media and OME shows a small increase (14% to 23%) in resolution of OME. A few clinical trials that have evaluated corticosteroid medications administered systemically or intranasally have documented a small transient therapeutic benefit (18% to 21%), but no long-lasting relief. An even smaller number of combined antibiotic and corticosteroid trials suggest a larger therapeutic benefit (25% to 31%), but these studies have not been of sufficient duration to show sustained relief.
For the child who has bilateral middle ear effusion for 3 months that involves hearing impairment, referral to an otolaryngologist for myringotomy with tympanostomy tube insertion becomes a treatment option. Referral for tympanostomy tube insertion was recommended by the AHCPR panel for children who have had 4 months of effusion with hearing loss (Fig 9⇓ ). Prospective studies have shown that tube placement for chronic OME results in improved hearing and a reduced frequency of symptomatic acute otitis media episodes while the tubes remain in place. Long-term studies of tympanostomy tubes have not been designed to study language outcomes as a result of tube treatment, although a study with this design is in progress in Pittsburgh, Pennsylvania.
Antihistamine and decongestant therapy is not recommended for treatment of OME because these agents are not effective for the condition either separately or together.
Adenoidectomy generally is not an appropriate treatment for uncomplicated OME, but it may be considered in an older child who has chronic OME. Although this surgical procedure has not been advocated for children younger than 4 years of age because of inadequate clinical trials, such a clinical investigation is in progress in Pittsburgh. Tonsillectomy either alone or with adenoidectomy has not been found to be effective for treating chronic OME.
The possible association between allergy and chronic OME has been studied epidemiologically and biochemically. A strong association has not been found, and biochemical studies of MEE in most cases have shown an absence of biochemical markers of type I hypersensitivity reactions. Thus, allergy management generally is not pursued in children who have chronic OME, and most experts do not recommend allergy evaluation in such children.
Evidence of the efficacy of other therapies for treatment of otitis media, such as chiropractic, holistic, naturopathic, traditional, homeopathic, and other treatments, were sought by the AHCPR review panel. Because no information has been published in randomized controlled studies of these therapies, the panel made no recommendations about these therapies for the treatment of OME in children.
The American Academy of Pediatrics has developed a practice parameter for managing OME in young children. This document can be accessed on the web at http://www.aap.org/policy/otitis.htm.
ASSESSMENT OF HEARING
Hearing loss associated with OME is usually conductive, meaning that sound is not conducted normally through the middle ear system to the inner ear. The degree of hearing loss typically is mild to moderate (15 to 50 dB averaged across 500 to 4,000 Hz), which is the most important frequency range for speech perception. The degree of hearing loss exceeds 20 dB in at least one ear in more than 50% of cases. The slope of the hearing loss may be uprising, meaning that low frequency sounds are diminished, flat across frequencies, or peaked, with relatively better hearing in the region of 2,000 Hz. Because the degree and shape of the hearing loss is highly variable, audiometry is necessary to determine the impact of OME on the child’s speech perception. It is not possible to predict the degree of hearing loss from otoscopy or tympanometry.
Screening of the child’s speech and language development also can detect possible developmental effects. Screening tools such as the Early Language Milestones (ELM) test are effective for detecting speech or language delay that may accompany hearing loss. Children who have head and neck anomalies that may affect eustachian tube function, such as cleft lip or palate, submucous cleft palate, Down syndrome, fetal alcohol syndrome, and metabolic disorders such as muco-polysaccharidosis, are at very high risk for hearing loss due to chronic OME. According to the Joint Committee on Infant Hearing, hearing in these children should be screened at birth, every 6 months until age 3 years, and at appropriate intervals thereafter. Speech and language outcomes in affected children are highly dependent on hearing status.
Several options are available to screen or test hearing of infants and young children. Table 2⇓ lists appropriate tests by age and advantages and disadvantages of each method.
Complications, Sequelae, and Prognosis
The middle ear space includes the area bounded by the tympanic membrane laterally, the inner ear medially, the mastoid air cells posteriorly, and the eustachian tube anteriorly. The primary purpose of this small, air-filled space is to conduct sound striking the tympanic membrane efficiently to the inner ear. This is accomplished through vibration of the ossicles-malleus, incus, and stapes-which are the smallest and lightest bones in the human body. The small size and light mass of the ossicles contribute to acute hearing sensitivity, particularly for high-frequency speech sounds. When vibration of the ossicles is impeded, hearing sensitivity is diminished.
In addition to the temporary conductive hearing loss caused by OME, there also is an increased risk of permanent, high-frequency sensorineural hearing loss due to inner ear damage. The mechanism for damage is believed to arise from passage of bacterial and host inflammatory mediators from MEE through the round window membrane into the inner ear. A prospective, longitudinal study of high-frequency hearing in children who had chronic OME showed that progressive loss of hearing sensitivity above 4,000 Hz was associated with the number of OME episodes and the number of tympanostomy tube placements required to treat chronic and recurrent OME. Therefore, hearing assessment should include high frequencies (above 4,000 Hz) in children who require repeated placement of tubes.
The impact of chronic OME on speech, language, and intellectual and social sequelae is controversial. The AHCPR panel of experts reviewed more than 100 studies; 9 of 14 studies (64%) that were technically adequate found significant effects in one or more areas of development. Expressive language skills (what the child expresses verbally) often were found to be deficient compared with receptive language skills (what the child understands). Poorer attention skills also were found. Studies demonstrated that the ability to process sounds binaurally is affected for many months after the hearing status returns to normal.
Speech perception is an obvious area of possible difficulty due to fluctuating hearing loss in OME, but only a few studies have addressed this domain. They have shown that OME is associated with poorer discrimination of short, similar sounds such as “da” and “ta”. Young infants who have OME have shown less diversity in their babbling. Studies of sound production have had mixed results; some showed generally poorer speech production among those who have OME, and others showed no difference compared with peers who do not have OME. Researchers at the University of North Carolina reported that speech error patterns at age 4½ years were more common among children who had histories of OME, and the richness of the caregiving environment positively affected language outcomes among impoverished children. The caregiving environment was more important in predicting language than was time spent having OME. Therefore, parents should be taught about the dual importance of medical management and positive verbal interchange with their infants and young children who have OME.
Other otitis media sequelae can be classified as intratemporal, meaning occurring within the ear and temporal bone, or intracranial. Common intratemporal complications include chronic suppurative otitis media, tympanosclerosis, atrophy, perforation, retraction of the tympanic membrane, atelectasis, adhesive otitis media, and cholesteatoma. Drainage is a common complication, particularly after myringotomy and tympanostomy tube insertion, but it may persist and require aggressive management in only 2% to 5% of children. Several small, long-term, prospective studies have been conducted following tympanostomy tube placement to assess anatomic sequelae associated with chronic OME. The most convincing evidence of relationships between tympanostomy tubes and sequelae comes from prospective studies of children who had bilateral chronic OME in which one ear randomly was assigned tympanostomy tube treatment and the control ear received no treatment or myringotomy. Two to seven years after treatment, tympanosclerosis (a whitish plaque produced in the tympanic membrane interstitium) was significantly more common in ears treated with tympanostomy tubes (50% to 70%) than in control ears, but this condition does not significantly affect hearing sensitivity. Rates of tympanic membrane atrophy and retraction were similar in treated and control ears. Although tympanic membrane atelectasis and cholesteatoma also have been described after insertion of tympanostomy tubes, these conditions are found commonly in ears affected by chronic OME but not treated with tubes. A long-term prospective study in the Netherlands reported that significant tympanic membrane retraction was a more frequent complication among children in whom tympanostomy tubes had previously been in place compared with children who never had tympanostomy tubes, but the study was not randomized.
Tympanic membrane retraction is associated both with hearing loss and with the pathogenesis of cholesteatoma, a benign but invasive neoplasm. Cholesteatoma occurs in approximately 1% of ears that have histories of chronic OME, but it may take 10 to 20 years to develop. Rare middle ear complications include facial nerve paralysis and destruction of the ossicular chain. Complications that can occur within the mastoid process are mastoiditis, very rarely Bezold abscess (infection in the sternocleidomastoid), and petrositis. Intracranial complications are very uncommon today, but the rise of antibiotic-resistant bacterial strains may result in a concomitant increase in the incidence of these conditions in the future. Meningitis, lateral sinus thrombosis, and extradural abscess are among these rare complications. Warning symptoms of intracranial complication, especially if the child is receiving antibiotics, include persistent headache, lethargy, malaise, irritability, severe otalgia, persistent fever, nausea and vomiting, or central nervous system signs such as stiff neck, focal seizures, ataxia, and blurred vision.
Chronic OME, which arises from a complex series of inflammatory events in the middle ear, affects approximately 5% to 30% of children. The mean duration of MEE is 16 to 20 weeks during the first 2 years of life. This condition is diagnosed best with pneumatic otoscopy and tympanometry. The risk of chronic OME is increased by environmental factors and characteristics of the child, including disease history. Approximately 70% of MEE are culture-positive, with approximately 50% of these yielding S pneumoniae, H influenzae, or M catarrhalis. However, antibiotic treatment of acute otitis media and OME has only a minimal effect on the long-term resolution of MEE. Research has shown that 70% of children who have chronic OME suffer mild-to-moderate hearing loss, so a child who has bilateral MEE for 3 months should undergo hearing evaluation. If the child has hearing impairment, referral to an otolaryngologist for myringotomy and tympanostomy tube insertion is a treatment option that the AHCPR recommends after 4 months of effusion with hearing loss. Sequelae of chronic OME include deficient expressive language and poorer attention skills due to the temporary hearing loss associated with OME, high-frequency sensorineural hearing loss, tympanic membrane atrophy, perforation, retraction, atelectasis, and cholesteatoma.
American Academy of Pediatrics. Position Statement 1994. Joint Committee on Infant Hearing. Pediatrics. 1995;95:152-156
Bluestone CD, Klein JO. Otitis Media in Infants and Young Children. 2nd ed. 1995 Philadelphia, Penn.: WB Saunders Co
Caplan J. Early Language Milestone Scale. 2nd ed. 1993 Austin.: Tex: Pro-Ed
Goycoolea MV. Otitis media: the pathogenesis approach. Otolaryngol Clin North Am. 1991;24:757-983
Marchant CD, Shurin PA, Turcyzk VA, et al. A randomized controlled trial of cefaclor compared with trimethoprim-sulfamethoxazole for treatment of acute otitis media. J. Pediatr. 1984;105:633-638
Roberts JE, Wallace IF, Henderson FW, ed. Otitis Media in Young Children. Medical, Developmental and Educational Considerations. 1997 Baltimore, Md:.: Paul H. Brookes Publishing Co
Rosenfeld RM, Mandel EM, Bluestone CD. Systemic steroids for otitis media with effusion in children. Arch Otolaryngol Head Neck Surg. 1991;117:984-989
Rosenfeld RM, Post JC. Meta-analysis of antibiotics for the treatment of otitis media with effusion. Otolaryngol Head Neck Surg. 1992;106:378-386
Rosenfeld RM, Vertrees JE, Carr J, et al. Clinical efficacy of antimicrobial drugs for acute otitis media: meta-analysis of 5400 children from thirty-three randomized trials. J Pediatr. 1994;124:355-367
Schwartz RH, Rodriguez WJ, Grundfast KM. Duration of middle ear effusion after acute otitis media. Pediatr Infect Dis. 1984;3:204-207
Stool SE, Berg AO, Berman S, et al. Otitis Media with Effusion in Young Children. Clinical Practice Guideline, Number 12, AHCPR Publication No. 94-0622. July 1994 Rockville, Md.: Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services
Teele DW, Klein JO, Rosner B. Epidemiology of otitis media during the first seven years of life in children in greater Boston: a prospective, cohort study. J Infect Dis. 1989;160:83-94
Williams RL, Chalmers TC, Stange KC, Chalmers FT, Bowlin SJ. Use of antibiotics in preventing recurrent acute otitis media and in treating otitis media with effusion. JAMA. 1993;270:1344-1351
- Copyright © 1999 by the American Academy of Pediatrics