Should antibiotics be prescribed for acute otitis media?


  • The authors have no funding, financial relationships, or conflicts of interest to disclose.


Acute otitis media (AOM) accounts for nearly 15 million antibiotic prescriptions every year and has become the most commonly cited reason for antimicrobial therapy among children in the United States. With the continuing rise of healthcare costs and the emergence of multi–drug-resistant bacteria, overuse of antibiotics has become a major public health concern worldwide. As a result, experts have called for the judicious use of antimicrobials in the treatment of AOM. However, despite years of clinical experience with the disease, an optimal treatment strategy is still unclear, as the clinical decision to treat or not to treat children with AOM remains controversial.


Several European nations have long used a watchful waiting approach for the treatment of AOM in children. In this treatment strategy, antibiotics are withheld from patients unless they do not improve spontaneously. The widespread adoption of this strategy has been based on studies that have shown that AOM has a favorable natural history with high rates of spontaneous resolution. One such study was a randomized controlled trial by Little et al. examining the efficacy of different prescribing strategies for AOM. In this trial, 315 children between the ages of 6 months to 10 years presenting with AOM were randomized to either begin amoxicillin immediately or receive delayed antibiotic therapy, in which a prescription was given to parents with instructions to initiate amoxicillin only if the child was not improving after 72 hours. This study found that, although children receiving immediate antibiotic therapy had shorter illness duration and slightly less acetaminophen use, they did not have any significant difference in school absence or pain scores and suffered from more incidents of diarrhea (19% vs. 9%) as compared to those with delayed or no treatment. In addition, 77% of the parents whose children were assigned to the delayed antibiotic group reported being satisfied with the watchful waiting approach. Thus, the study concluded that a wait-and-see strategy is not only feasible but also acceptable to parents of children suffering from AOM.1

Another similar randomized, controlled trial by Spiro et al. examined 283 children between the ages of 6 months and 12 years who were diagnosed with AOM in an emergency department. The diagnosis was made at the discretion of the clinician, and the children were randomly assigned to begin antibiotics, which were chosen and dosed independently by each clinician immediately after the visit, or given a wait-and-see prescription (WASP) and instructed not to fill the prescription unless the child's symptoms did not improve after 48 hours. This study found that the WASP approach reduced the use of antibiotics by 56%. Even more, there was no statistically significant difference between the groups in the rates of subsequent fever, otalgia, or unscheduled medical visits.2

As support for this watchful waiting strategy grew, the American Academy of Pediatrics and the American Academy of Family Physicians, with input from the American Academy of Otolaryngology–Head and Neck Surgery, released a clinical practice guideline in 2004 for the management of AOM in children aged 6 months to 12 years.3 The guidelines not only included a watchful waiting strategy in children older than 2 years, but more notably, it also endorsed a similar watchful waiting approach in younger children from ages 6 to 23 months. These specific guidelines, which are still in effect today, are detailed in Table I.

Table I. Current Management of Acute Otitis Media (U.S. Guidelines)
AgeAntibioticsWatchful Waiting
  • *

    Severe illness: moderate to severe otalgia or fever ≥39°C in the past 24 hours.

  • Certain diagnosis meets all of the following three criteria: 1) rapid onset, 2) signs of middle ear effusion, and 3) signs and symptoms of middle ear inflammation.

  • Nonsevere illness: mild otalgia and fever <39°C.

<6 monthsAmoxicillin 80–90 mg/kg/day for 10 days; if severe illness,* amoxicillin (90 mg/kg/day)—clavulanate (6.4 mg/kg/day)No watchful waiting option
6 months–2 yearsAntibiotics as above if certain diagnosis or severe illnessWatchful waiting if uncertain diagnosis and nonsevere illness
≥2 yearsAntibiotics as above if severe illnessWatchful waiting if uncertain diagnosis or nonsevere illness

Although several additional studies have appeared to appropriately justify this watchful waiting approach, some experts have questioned its use, especially in younger children, whereas others have raised the concern that this approach may be responsible for the increased incidence of mastoiditis and other complications of AOM. These authors argue that many of the previous clinical trials suffered from significant limitations, such as the inclusion of only a small number of young children, the exclusion of children with more severe disease, and perhaps most importantly, a lack of stringent diagnostic criteria. These limitations may have biased the studies to include patients who were either misdiagnosed or more likely to recover spontaneously.

To avoid these flaws, a recently published randomized controlled trial by Tähtinen et al. used stringent diagnostic criteria for AOM to ensure appropriate patient selection. The criteria included specific pneumatic otoscopic findings suggestive of a middle ear effusion, such as decreased or absent tympanic membrane mobility or air-fluid interfaces, explicit signs of acute middle ear inflammation, and fever. The authors then compared the use of amoxicillin-clavulanate or placebo in 319 young children, 6 to 35 months of age, who specifically met these diagnostic standards. The study found that antibiotic therapy reduced the risk of treatment failure by 62%. In addition, there were significantly better outcomes with antibiotic therapy in terms of both overall condition, which was based on the parent's assessment of their child's general state of health (improved, no change, or worse) before and after treatment, and otoscopic signs, such as changes in the mobility and color of the tympanic membrane post-treatment, when compared with placebo (P < .001 for both outcomes).4

Results from another recent clinical trial by Hoberman et al. appeared to further question the use of the watchful waiting approach in younger children. This study used a similarly strict diagnostic criteria and found that among 291 children 6 to 23 months of age with AOM, treatment with amoxicillin-clavulanate resulted in significantly lower rates of clinical failure when compared with placebo; 4% versus 23% at or before day 4 or 5 of treatment (P < .001) and 16% versus 51% at or before day 10 to 12 (P < .001). In addition, the study used a symptom severity scale and found that antimicrobial therapy was superior to placebo even in children with less severe disease.5 Nevertheless, the authors of this study, as well as the study by Tähtinen et al., emphasized that the greater beneficial effect of antibiotics that was seen in these studies as compared with previous trials was likely due to the use of a rigorous diagnostic standard that ensured only children with a definitive diagnosis of AOM were included in the studies.


When medical therapy is deemed necessary for AOM, amoxicillin is considered the first-line treatment due to its safety, low cost, and narrow microbiologic spectrum. In patients who have severe illness (moderate to severe otalgia or fever of 39°C or higher) amoxicillin-clavulanate should be used.3 Currently, immediate antibiotic therapy is recommended in children younger than 6 months of age, whereas watchful waiting is advocated in children older than 6 months of age with nonsevere illness. However, two recent studies using more stringent diagnostic criteria for AOM have shown significantly better outcomes with immediate antibiotic therapy in children between 6 to 35 months of age.4, 5 These studies highlight the importance of accurate diagnosis of AOM in choosing therapeutic intervention, and its findings may lead to a reevaluation of the watchful waiting strategy.


In this review, four level 1 studies (randomized controlled trials) and a national clinical guideline were reviewed.