Antibiotic management of acute otitis media

Otitis media is the most common childhood infection for which antibiotics are prescribed. Nonetheless, there are a number of important questions about the optimal management of acute otitis media (AOM), and opinion is divided within the medical community on a range of fundamental issues. The purpose of this statement is to address several controversial questions regarding antimicrobial management of AOM and to present a consensus opinion on each. It must be emphasised that much remains to be learned about management of this common childhood problem and that ongoing research may mandate revision of these opinions in the near future; these recommendations must, therefore, be considered provisional and dependent upon the current ‘state of the art’.

Evaluation of studies to determine whether antibiotic therapy influences the outcome of AOM has been difficult to interpret because of the high spontaneous recovery rate in children with the disease (1). Furthermore, many of the studies have been designed to try to demonstrate whether new antimicrobial agents are as good as conventional therapy, but they have employed small sample sizes. For a disease with a high spontaneous rate of improvement, only studies with a very large sample size provide enough power to demonstrate whether two different interventions are indeed truly comparable. Thus, the published literature provides little guidance regarding superiority of one antimicrobial agent over another.

For the purpose of this paper, AOM is defined as the sudden onset of inflammation of the middle ear associated with an effusion and one or more of the following: pain, fever and irritability. The diagnosis must be established by careful examination with pneumatic otoscopy.

SHOULD ANTIBIOTICS BE USED TO TREAT AOM?

Antimicrobial therapy is one of the cornerstones in the management of AOM but some studies have suggested that its routine use is not indicated (2–4). Because the majority of cases of AOM resolve spontaneously (1), it might appear that antimicrobial therapy is not necessary. Nonetheless, in the preantibiotic era, complications of AOM such as mastoiditis were far more common than they are today (5,6); this difference may be due to the current routine use of antibiotics. A recent meta-analysis of 5400 children with AOM indicated that antimicrobial therapy enhanced the ‘primary control’ by 13.7% despite a spontaneous recovery in 81% of cases (1). Because it is probably not possible to determine a priori which cases of AOM will result in suppurative complications, it is likewise not possible to determine which cases require antimicrobial therapy and which will resolve spontaneously. Therefore, it appears prudent to consider all cases of AOM candidates for antimicrobial therapy in order to minimize the likelihood of complications. This is clearly an area where more research is required to identify which patients with AOM require therapy and which would improve spontaneously without it. Some experts recommend watchful waiting for 48 to 72 h before initiating antibiotic therapy (4). This approach may be feasible in children over two years of age if good follow-up can be assured; therefore, decisions about whether to withhold antibiotics therapy initially must be made on a patient-by-patient basis.

WHICH ANTIBIOTIC IS THE DRUG OF CHOICE FOR AOM?

Bacteria cause the majority of cases of AOM, and the most frequent etiological agents are Streptococcus pneumoniae, nontypeable Haemophilus influenzae, Moraxella catarrhalis, group A streptococcus and Staphylococcus aureus. Viruses continue to cause a substantial minority of cases (7), and antibiotic therapy would not be expected to affect the outcome. With the increasing prevalence of beta-lactamase-producing (penicillin-resistant) strains of H influenzae and M catarrhalis, alarms have been sounded about the wisdom of routinely using aminopenicillins (such as amoxicillin) as the standard first-line antimicrobial for uncomplicated AOM. Despite theoretical concerns about the diminishing usefulness of amoxicillin, it continues to be as effective as any other oral antimicrobial agent for childhood AOM. In fact, it works as well as extended spectrum, penicillinase-resistant oral agents for otitis media caused by either penicillin-susceptible or -resistant bacteria (1). Most comparative trials of antimicrobial therapy in AOM have failed to demonstrate a difference in effectiveness between amoxicillin and any other agent. Furthermore, the newer, broader spectrum, penicillinase-stable antimicrobial agents are substantially more expensive than amoxicillin (Table 1), and their use may be associated with relatively high rates of side effects and may increase the pressure for selection of multiply antibiotic-resistant strains of bacteria. Therefore, because of its excellent ‘track record’ (for infections due to penicillin-susceptible and -resistant bacteria), low cost, safety and acceptability to patients, amoxicillin remains the drug of choice for uncomplicated AOM.

TABLE 1:

Cost of a 10-day course of antibiotics for therapy of a 10 kg child with otitis media

Antibiotic	Dose	Frequency	Cost by location ($)*
			Vancouver, BC	Quebec City, QC
Amoxicillin	125 mg	tid	10.13	11.99
Amoxicillin-clavulanate	125 mg	tid	23.25	27.25
Trimethoprim-sulfamethoxazole	5 cc	bid	9.07	9.43
Erythromycin-sulfisoxazole	2.5 cc	qid	19.31	25.72
Cefaclor suspension	125 mg	tid	25.32	30.53
Cefixime	100 mg	od	25.59	25.07
Cefuroxime axetil	125 or 250 mg	bid	23.46	23.45
Cefprozil suspension	125 mg	bid	22.37	22.30
Clarithromycin suspension	75 mg	bid	33.70	24.37
Azithromycin†	day 1: 100 mg	od	24.42	23.85
	days 2–5: 50 mg

^*Total cost to patient, including dispensing fee; prices obtained from private pharmacies in Vancouver and in Quebec City;

^†A five-day course of azithromycin is recommended because of its unique pharmacokinetics

WHAT IS A REASONABLE EXPECTATION FOR RESPONSE TO THERAPY?

One can reasonably expect that the symptoms of AOM (fever, irritability and ear pain) will resolve within 72 h of initiation of antimicrobial therapy. If, after this length of therapy symptoms persist, the child should be reevaluated to determine if the infection is persisting or has evolved into one of the suppurative complications. If the patient has complied with the prescribed therapy and the symptoms, such as pain and fever, have persisted, a change in antimicrobial regimen is appropriate. The different second-line agents from which to choose are listed in Table 1. Whereas the symptoms of AOM listed above should resolve promptly with antimicrobial therapy, the middle ear effusion may persist for up to three months despite bacteriological cure (8). Therefore, persistence of middle ear fluid after a full course of antibiotic therapy of AOM is not an indication for continued therapy or for institution of treatment with a second-line drug.

WHAT IS THE OPTIMAL DURATION OF ORAL THERAPY OF AOM?

Because there is such a high spontaneous cure rate for AOM, it has been difficult to determine the optimum duration of antimicrobial therapy. Furthermore, without appropriate bacteriological assessment (9) and follow-up (with repeated tympanocentesis) response to therapy is impossible to gauge. Therefore, studies with sufficient power and appropriate design to determine the efficacy of short-course therapy have been difficult to perform. Studies that purport to demonstrate equivalence of five-and 10-day duration (10,11) may simply have indicated that the majority of patients will improve whether they are treated or not. Until it can be clearly demonstrated that a short course of therapy is effective, patients should be treated for 10 days. The only exception is azithromycin, for which a five-day course is recommended because of its unique pharmacokinetics.

WHAT IS THE ROLE OF PARENTERAL THERAPY FOR AOM?

With the advent of extended spectrum cephalosporins with prolonged half-life (eg, ceftriaxone), the option of parenteral therapy with a single dose has become feasible. There is, at present, little published evidence that parenteral therapy provides any advantage to the conventional 10-day oral therapy. Furthermore, the use of such broad-spectrum agents may hasten the emergence of antibiotic-resistant organisms. Except in extraordinary situations, parenteral therapy should not be employed for simple uncomplicated childhood AOM. If a child appears to be too ill to comply with standard oral therapy, a diagnosis other than AOM should be entertained and admission to hospital should be considered.

WHAT IS THE OPTIMAL ANTIMICROBIAL MANAGEMENT OF TREATMENT FAILURES?

As stated above, the symptoms of AOM (fever, irritability and otalgia) should resolve within 72 h of initiating antimicrobial therapy. Failure of symptomatic response to appropriate therapy (with evidence of compliance) constitutes a treatment failure. The optimal management of such patients is controversial and various approaches have been advocated (12). A tympanocentesis should be considered for both therapeutic (relief of pressure and pain) and for diagnostic (recovery of the etiologic agent) purposes. If a tympanocentesis is not practical, consideration should be given to adding amoxicillin-clavulanate or selecting one of the alternative agents from Table 1. If a tympanocentesis is performed, the antibiotic choice should ultimately be guided by the etiological agent and antimicrobial susceptibility.

WHAT IS THE BEST THERAPY FOR AOM IN PATIENTS ALLERGIC TO PENICILLIN?

In patients with documented allergy to penicillin, an alternative to amoxicillin is required. Although there is a risk of cross-reaction to other beta-lactam agents, this occurs rarely and therapy with a cephalosporin is generally safe. Agents from which to choose and their costs are listed in Table 1. The choice should be guided by various considerations including cost, frequency of adverse side-effects and patient tolerability. A reasonable choice is either trimethoprim/sulfamethoxazole or erythromycin/sulfisoxazole.