|Home | About | Journals | Submit | Contact Us | Français|
The present position statement updates a previous document released in 1998 (1). It is based on published evidence, and is intended to be a guide for sound clinical decision making. The present document is not intended for children with craniofacial abnormalities, immunocompromising conditions or complicated acute otitis media (AOM), or newborns younger than eight weeks of age.
AOM is extremely common in children – in fact, 75% of children have at least one episode by one year of age (2). The primary defect leading to AOM is eustachian tube dysfunction and obstruction. Compared with adults, children are predisposed to AOM because their eustachian tubes are shorter, more horizontal and more prone to obstruction by enlarged adenoids (3,4). Furthermore, viral infections and allergies are common in young children, and both can cause eustachian tube inflammation (5,6). Finally, children (especially those with recurrent otitis media) may have decreased levels of secretory immunoglobulin A – an antibody that decreases bacterial adherence in the nasopharynx (7).
Once the eustachian tube is obstructed, two things happen. First, mucociliary clearance is impaired, trapping mucus in the middle ear space (8). Second, resorption of gases within the middle ear space creates a pressure differential, akin to a vacuum, which pulls bacteria from the nasopharynx into the middle ear space. Once introduced into this space, bacteria can proliferate and may cause a secondary infection. Thus, it is rare to develop AOM without an antecedent viral upper respiratory tract infection, with AOM typically developing after several days of viral symptoms.
The major risk factors for AOM are young age and daycare attendance. The former is likely related to the anatomy of the eustachian tube and low secretory immunoglobulin A levels, while the latter is related to increased exposures to viral infections, coupled with an increased incidence of nasopharyngeal colonization with pathogenic bacteria. Other risk factors include orofacial abnormalities (such as cleft palate), household crowding, exposure to cigarette smoke, premature birth, not being breastfed, immunodeficiency and a positive family history of otitis media (9,10). Children of First Nations or Inuit ethnicity are also at higher risk for AOM (11).
To properly diagnose AOM, there must be fluid behind the tympanic membrane (a middle ear effusion) and specific signs and symptoms of middle ear inflammation (Table 1) (12–22) – indicating that this fluid is pus.
Understanding the etiology of acute middle ear effusion and inflammation is the key to answering this question. Viruses play an important role in the pathogenesis of AOM and may be a direct cause of spontaneously resolving AOM, because they have been found in middle ear fluid in the absence of bacteria (23). However, studies (24,25) using tympanocentesis show bacteria are present most of the time. The strains of bacteria have changed over time. Before the introduction of the pneumococcal conjugate vaccine, the most common bacteria isolated from AOM were Streptococcus pneumoniae (median 42% of cases), Haemophilus influenzae (median 31% of cases) and Moraxella catarrhalis (median 16% of cases) (26). Other bacteria such as group A streptococci and Staphylococcus aureus were rare, as were polymicrobial infections (27,28). After the introduction of the conjugated pneumococcal vaccine, American studies assessed bacterial isolates from vaccinated children younger than two years of age with severe or refractory AOM. They found that the proportion of AOM cases caused by S pneumoniae decreased from 48% to 31% and the proportion of cases caused by nontypeable H influenzae increased from 41% to 56% (29–31).
Several meta-analyses (32–34) have examined the role of antimicrobials in the treatment of AOM. As one might predict for what is primarily a bacterial infection, the cumulative evidence demonstrates more rapid resolution of symptoms with the use of antimicrobials. However, the treatment effect for antimicrobials is small – approximately 15 children have to be treated for one child to have resolution of symptoms (clinical cure) at 48 h (32). There have been criticisms of the studies that led to this conclusion (35,36). First, in most of the studies, the diagnosis of AOM was made clinically, which suggests the possibility of a misdiagnosis, but the same applies to diagnosing AOM in Canadian children today. Second, clinical cure rather than bacteriological cure was chosen as the primary outcome because of the difficulty of performing tympanocentesis initially and at follow-up. Children with early bacteriological cure are at lower risk of early recurrence of AOM with the same organism, but it appears that approximately five children need to achieve bacteriological cure to prevent one recurrence (37). Finally, a placebo was not always administered to the control group. Despite these criticisms, spontaneous resolution occurs in most cases. Not all children with AOM should receive immediate treatment with antimicrobials, and a watchful waiting approach with analgesia can be used in many cases.
If the child is older than six months of age with mild signs and symptoms, observation without the use of antimicrobials for 48 h to 72 h may be an option if follow-up can be assured (Table 2, Figure 1) (38,39). Six months was chosen as the lower age limit because there are limited data on this approach in younger children and severe illness is more difficult to recognize (40,41). If the watchful waiting approach is used, it is vital to provide appropriate advice about analgesics, with acetaminophen or ibuprofen being the usual choices. It is recommended to either have the family return if the child does not improve or to provide a prescription for antimicrobials that can be filled at the parents’ discretion (deferred prescription). Studies (42,43) have shown that although symptom resolution may take slightly longer with a watchful waiting approach, parents are generally satisfied with this option, and only approximately one-third of those children eventually receive antimicrobials.
The watchful waiting option is not appropriate for children who have severe symptoms (appear toxic, have severe otalgia and/or high fever [greater than 39°C, orally]) (44). Aboriginal children have been found to have a high incidence of chronic suppurative otitis media, but it is not known whether a watchful waiting approach in these children increases the risk of this complication (11,45). Nonetheless, it would seem prudent to prescribe antimicrobials sooner to Aboriginal children.
It seems intuitive that the early use of antimicrobials will reduce the incidence of serious complications of AOM, such as mastoiditis, meningitis and intracranial abscesses. In the Netherlands, where antimicrobial prescription rates for AOM are approximately 30%, the incidence of paediatric mastoiditis was approximately double the incidence in countries where prescription rates were greater than 90% (46). Nonetheless, given the rarity of mastoiditis, the authors calculated that at least 2500 prescriptions would have to be filled to prevent one case. They also point out that only approximately 25% of mastoiditis cases require a mastoidectomy, and that approximately one-half of children with mastoiditis develop this complication despite previously taking antimicrobials for AOM (47). There are no comparable studies for other severe suppurative complications of AOM, but again, it seems likely that thousands of children would have to be treated to prevent one complication.
There are also risks associated with the use of antimicrobials. Approximately 20% of children develop diarrhea, with complications such as Stevens-Johnson syndrome or anaphylaxis being very rare but sometimes life-threatening. In addition, the development of antibiotic-resistant organisms is primarily driven by the over-use of antibiotics.
S pneumoniae antimicrobial resistance is an issue in Canada and around the world. Furthermore, AOM caused by S pneumoniae is the least likely to spontaneously resolve (only 20% of cases spontaneously resolve versus 50% with H influenzae) (48). In some cases, it is possible to identify children at risk for infections with antimicrobial resistant S pneumoniae (49). Risk factors include children younger than two years of age, who attend daycare (defined as greater than 4 h per week with at least two unrelated children), who have frequent otitis media and/or recent antimicrobial use (within the past three months), or who have failed initial antimicrobial therapy for AOM.
Other organisms associated with AOM include H influenzae and M catarrhalis. Almost all M catarrhalis isolates and approximately one-quarter of H influenzae produce beta-lactamases (50). Some beta-lactam antimicrobials are still effective against these organisms, including second- and third-generation cephalosporins and amoxicillin with a beta-lactamase inhibitor (such as clavulanate) added. The activity of clarithromycin or azithromycin is unaffected by the presence of beta-lactamase production.
First-line therapy in a child with no beta-lactam allergies is amoxicillin (Figure 2). No other oral antimicrobial has been shown to have superior efficacy for AOM in a randomized trial. This drug has excellent middle ear penetration (which may be effective despite in vitro resistance), is inexpensive and well tolerated, and has a relatively narrow antimicrobial spectrum (51). Given in an adequate dose, it is the oral drug that is most likely to be effective against penicillin-resistant S pneumoniae. Because it is not always apparent whether a child has risk factors for resistant S pneumoniae, we recommend that physicians treat with high-dose amoxicillin at 75 mg/kg/day to 90 mg/kg/day. At this dose, the drug will be effective against penicillin-intermediate strains and possibly even high-level penicillin-resistant strains.
If the child has had a type 1 hypersensitivity reaction to amoxicillin or to another beta-lactam antimicrobial (urticaria and/or systemic anaphylaxis), then use of a macrolide (clarithromycin or azithromycin) is an option. If the previous reaction to amoxicillin is not type 1, second-generation cephalosporins are appropriate choices (Table 3) (52). If the child has had a type 1 hypersensitivity reaction to a beta-lactam antimicrobial and has failed macrolides, clindamycin or a quinolone should be considered in consultation with an infectious diseases physician. Alternatively, a referral to otolaryngology for tympanocentesis may be considered to determine the etiological agent and to guide therapy (53).
Symptoms should improve within one to two days and resolve within two to three days of starting antimicrobials. Middle ear effusions, on the other hand, may persist for months, despite clinical and bacteriological resolution. Therefore, the presence of middle ear effusion does not necessitate a change in antimicrobials. However, if symptoms have not improved after two days, the antimicrobial should be changed to one that targets both penicillin-resistant S pneumoniae and beta-lactamase-producing organisms – two choices are amoxicillin/clavulanate or parenteral ceftriaxone (Table 3) (54).
Amoxicillin/clavulanate should be calculated at 90 mg/kg/day of the amoxicillin component and 6.4 mg/kg/day of the clavulanate component divided into two doses. Methods for achieving this dose, which require combining amoxicillin/clavulanate with amoxicillin, are shown in Table 4. A 14:1 preparation of amoxicillin/clavulanate is licensed in the United States, allowing for the use of a single medication, but it is not yet available in Canada (55).
Five days of antimicrobial treatment with amoxicillin or second-generation cephalosporins are at least as effective as 10 days of therapy in children older than two years of age with uncomplicated AOM (56,57).
Ten-day antimicrobial treatment courses are appropriate for children younger than two years of age, children with frequent recurrent AOM or otitis media with perforated tympanic membrane, and in children who failed their initial antimicrobial, because these children are at increased risk of treatment failure (exceptions to this rule are azithromycin for which a five-day course is the maximum, and ceftriaxone for which one dose is usually given for uncomplicated cases and three doses for cases that failed initial therapy) (58–65). The benefit of the longer course may partly come from the child being in a ‘prophylaxed’ state should he or she develop a new upper respiratory tract infection within 10 days of AOM diagnosis. However, should the child develop antimicrobial-related adverse events between day 5 and day 10, it is reasonable to stop antimicrobials rather than prescribing an alternative antimicrobial.
Parents can reduce their child’s risk for AOM by implementing practices that reduce the chances of contracting viral respiratory tract infections or by preventing other factors that promote inflammation of the eustachian tube:
Use of the influenza vaccine is highly encouraged for healthy children older than six months of age and for their parents and caregivers (77–79). Influenza plays an important role in the pathogenesis of AOM, and the killed influenza vaccine has been shown to provide some protection against AOM in toddlers (80–82). Although not yet available in Canada, the live attenuated intranasal vaccine showed high efficacy (94% to 98%) in preventing influenza-associated AOM in children 15 to 71 months of age (83,84).
The pneumococcal conjugate vaccine is part of the routine schedule for all Canadian children. This vaccine has limited efficacy against AOM because only seven pneumococcal serotypes are contained in the current vaccine, and there is increasing evidence of ‘replacement disease’ with nonvaccine serotypes (85,86). Preliminary studies of upcoming conjugated pneumococcal vaccines show a greater effect against AOM. They contain more pneumococcal serotypes and some are conjugated to carriers such as protein D from H influenzae and, thus, are efficacious in preventing AOM from pneumococci and nontypeable H influenzae (87,88).
The levels of evidence reported in the recommendations have been described using the evaluation of evidence criteria outlined by the Canadian Task Force on Preventive Health Care (89).
INFECTIOUS DISEASES AND IMMUNIZATION COMMITTEE
Members: Drs Robert Bortolussi, IWK Health Centre, Halifax, Nova Scotia (Chair); Jane Finlay, Richmond, British Columbia; Joan L Robinson, Edmonton, Alberta; Élisabeth Rousseau-Harsany, Sainte-Justine UHC, Montreal, Quebec (Board Representative); Lindy M Samson, Children’s Hospital of Eastern Ontario, Ottawa, Ontario
Consultants: Drs Noni E MacDonald, IWK Health Centre, Halifax, Nova Scotia; Dorothy L Moore, The Montreal Children’s Hospital, Montreal, Quebec
Liaisons: Drs Upton D Allen, The Hospital for Sick Children, Toronto, Ontario (Canadian Pediatric AIDS Research Group); Charles PS Hui, Children’s Hospital of Eastern Ontario, Ottawa, Ontario (CPS Liaison to Health Canada, Committee to Advise on Tropical Medicine and Travel); Nicole Le Saux, Children’s Hospital of Eastern Ontario, Ottawa, Ontario (Immunization Program, ACTive); Larry Pickering, Elk Grove, Illinois, USA (American Academy of Pediatrics); Marina I Salvadori, Children’s Hospital of Western Ontario, Ottawa, Ontario (CPS Liaison to Health Canada, National Advisory Committee on Immunization)
Principal authors: Drs Sarah ED Forgie, University of Alberta, Edmonton, Alberta; George Zhanel; University of Manitoba, Winnipeg, Manitoba; Joan L Robinson, Edmonton, Alberta
The recommendations in this statement do not indicate an exclusive course of treatment or procedure to be followed.
Variations, taking into account individual circumstances, may be appropriate.
All Canadian Paediatric Society position statements/notes are reviewed, revised or retired as needed on a regular basis.
For the most current version, please consult the “Position Statements” section of the CPS website (www.cps.ca/english/publications/statementsindex.htm).