Journal reference: Tähtinen PA, Laine MK, Huovinen P, et al. A placebo-controlled trial of antimicrobial treatment for acute otitis media. N Engl J Med 2011; 364: 116-26 ^[1]

Link: http://dx.doi.org/10.1056/NEJMoa1007174

Evidence cookie says...

Children aged 6-35 months with acute otitis media may benefit from amoxicillin-clavulanate but at the cost of adverse events. The likelihood of benefit from antibiotics is similar to the likelihood of harm. The Antibiotic Therapeutic Guidelines recommendations remain a useful guide [2] immediate antibiotic therapy for AOM with systemic features symptomatic treatment with potential delayed antibiotic therapy for AOM without systemic features Note: it is likely that the favourable results found for amoxicillin-clavulanate in this study cannot be generalised to the Australian setting. See below for more details.

More details:

Article details

Study design:

double-blind randomised controlled trial

Study aim:

to determine the efficacy of antimicrobial treatment for acute otitis media in children aged 6 to 35 months of age

Methods summary:

Inclusion criteria:

• children aged 6 to 35 months
• diagnosis of acute otitis media by strict protocol
  • middle-ear fluid had to be detected by means of pneumatic otoscopic examination that showed at least two of the following findings:
    • bulging position
    • decreased or absent mobility
    • abnormal colour or opacity not due to scarring
    • air-fluid interfaces
  • at least one of the acute inflammatory signs in the tympanic membrane:
    • distinct erythematous patches
    • streaks
    • increased vascularity over full, bulging, or yellow tympanic membrane
  • had to have acute symptoms including:
    • fever
    • ear pain
    • respiratory symptoms

Exclusion criteria

• ongoing antimicrobial treatment
• acute otitis media with spontaneous perforation of the tympanic membrane
• systemic or nasal steroid therapy within the preceding 3 days
• antihistamine therapy within the preceding 3 days
• oseltamivir (Tamiflu) therapy within the preceding 3 days
• allergy to penicillin or amoxicillin
• presence of tympanostomy tube
• severe infection requiring systemic antimicrobial treatment
• documented Epstein-Barr virus infection within the preceding 7 days
• Down syndrome or other condition affecting middle ear diseases (e.g., cleft palate)
• known immunodeficiency
• severe vomiting or another symptom that affects oral dosing
• parents with insufficient fluency in Finnish
• use of any investigational drugs during the preceding 4 weeks

Study arms:

• active treatment:
  • amoxicillin–clavulanate, 40 mg + 5.7 mg /kg/day divided into two doses x 7 days
• placebo controlled treatment:
  • similar to active treatment in appearance and taste x 7 days
• parents were given a diary and asked to record symptoms, doses of study drugs and any other medications
• use of OTC medications for symptom control (e.g., analgesics, antipyretic agents, ear and nose drops) were encouraged
• follow up was on day 3 and day 8

Primary outcome:

• time to treatment failure (composite outcome):
  • no improvement in overall condition by first scheduled visit
  • a worsening of the child’s overall condition at any time
  • no improvement in otoscopic signs by the end-of-treatment visit on day 8
  • perforation of the tympanic membrane at any time
  • severe infection necessitating systemic open-label antimicrobial treatment at any time
  • any other reason for stopping the study drug

Secondary outcomes:

• time to the initiation of rescue treatment
• development of contralateral acute otitis media
• use of analgesics or antipyretics agents
• treatment result
• adverse events

Analysis:

• a priori calculation that with 260 patients, the study would have 90% power to detect an absolute reduction of 15% in the rate of treatment failure between the groups (assuming 25% rate of treatment failure in the placebo group)
• planned to enrol 320 patients to account for a possible 20% rate of withdrawal
• intention to treat analysis

Results summary:

Primary outcome, treatment failure:

• received amoxicillin-clavulanate: 18.6%
• received placebo: 44.9%
• i.e., amoxicillin-clavulanate reduced the risk of treatment failure by 62%
  • hazard ratio (HR) = 0.38 (95% CI 0.25-0.59), P < 0.001
  • Number needed to treat (NNT) = 3.8 (95% CI 2.7-6.2)

Secondary outcomes:

• rescue treatment (antimicrobial group compared to placebo group):
  • HR = 0.19 (95% CI 0.10-0.36), P < 0.001
• contralateral acute otitis media:
  • amoxicillin-clavulanate group: 8.2%
  • placebo group: 18.6%
  • P = 0.007
• no significant differences between groups on use of analgesics and antipyretics
• treatment result at day 8, those not improved or worsened:
  • amoxicillin-clavulanate group: 6.8%
  • placebo group: 29.7%
  • i.e., absolute risk reduction compared to placebo = 22.9% (95% CI 14.4 to 31.4)
  • NNT = 4.4 (95% CI 3.2-6.9), note: this result was not in the original article, it was calculated by the Morsels of Evidence author

Adverse events:

• Overall:
  • amoxicillin-clavulanate group: 52.8%
  • placebo group: 36.1%
  • i.e., absolute risk increased = 16.7% (95% CI 5.8-27.6), P = 0.003
  • Number needed to harm (NNH) = 6.0 (95% CI 3.6-17)

Study conclusion:

Children with acute otitis media benefit from antimicrobial treatment as compared with placebo, although they have more side effects. Future studies should identify patients who may derive the greatest benefit, in order to minimise unnecessary antimicrobial treatment and the development of bacterial resistance.

Participants:

• 322 Finnish children aged 6 to 35 months of age underwent randomisation
• recruitment March 2006-December 2008
• recruited in the Turku city area, Finland, by nurse and physician suggestion and by parental self-selection
  • parents had to contact the study clinic by phone
• adherence to the study drug was approximately 94% as assessed according to diary entries and 99% as assessed according to the amount of returned study drug
• only 3 participants dropped out → 319 included in analysis
• baseline characteristics:
  • mean age: 16 months
  • mean previous episodes of acute otitis media: 2
  • mean time since last episode of acute otitis media: 3 months
  • 57% of the children were male
  • all except 2 children were Caucasian
  • immunisations:
    • 1 or more doses of pneumococcal vaccine: only 2.2%
    • 1 or more doses of influenza vaccine: 12.9%
    • 1 or more doses of Hib vaccine: 100%
• ~ 60% had Streptococcus pneumoniae on the nasopharyngeal sample

Methodological weaknesses

• the primary outcome “time to treatment failure”, which is a composite outcome, is somewhat unusual and possibly biases in favour of antimicrobial treatment
  • specifically, some elements in the composite that favoured antimicrobial treatment have limited clinical relevance, e.g., worsening of overall condition at any time, no improvement in otoscopic signs by day 8
• it is curious that the overall treatment result at day 8 was relegated to a secondary outcome rather than as the primary outcome
• homogeneous population; almost all the participants were ethnically Caucasian Finns
  • this is contrasted with the ethnically diverse population in Australia
• the study population was not immunised against Streptococcus pneumoniae (only 2.2% had received one or more doses)
  • this is in contrast to Australia children who receive S. pneumoniae immunisation as part of the standard schedule
  • S. pneumoniae was the most common pathogen isolated on nasopharyngeal samples
  • this is a serious threat to external validity
  • it might not be valid to generalise these results to Australian children
• although the strict diagnostic criteria for acute otitis media improves the study’s internal validity, the results favouring antimicrobial therapy may be an overestimate when applied to the laxer diagnostic standards in normal practice

Methodological strengths

• generally well conducted randomised controlled trial
• well performed statistical analyses

Biases and conflicts of interests

• nil substantial declared and nil seem obvious

Clinical relevance to primary health care

Acute otitis media (AOM) is the most common bacterial infection in early childhood.  There is some controversy regarding the best management strategy: antibiotics, judicious use of antibiotics, antibiotics only after an observation period have all be proposed.

It is likely that the results of this study are an overestimate in favour of amoxillin-clavulanate (Augmentin). The study population was not immunised against Streptococcus pneumoniae and the diagnosis of acute otitis media followed a strict protocol. Both factors are substantially different to the Australian primary health care setting; young children receive S. pneumoniae immunisation as part of the standard schedule and general practitioners are not experts in pneumatic otoscopy. Nevertheless, the results confirm that treatment with amoxicillin-clavulanate in young children where there is little doubt of the diagnosis, is clearly superior in terms of resolution of the symptoms.  This must be balanced, however, with the rate of adverse events.  For every 6 children treated with antimicrobial therapy rather than placebo, one extra child will have adverse effects, mostly diarrhoea.  That is, the number needed to harm (NNH) is 6.0.

What is the number needed to treat (NNT)?  This study found an NNT of 4.4 on antibiotics when looking at treatment results at day 8 (for every 4.4 children placed on antibiotics rather than placebo, one extra child will have benefit). However, it is unlikely that this value is valid in the Australian context due to the aforementioned limitations. The same issue of the New England Journal of Medicine published a randomised controlled trial on the treatment of acute otitis media with amoxicillin-clavulanate versus placebo in children under 2 years of age in the United States ^[3].  Although they too found superior results with antibiotics, the results were more modest; the NNT was 7.1 for sustained resolution of symptoms by day 7; an NNT of 17 was found for initial resolution of symptoms by day 7 ^[2].  Interestingly, these results from the US study are similar to estimates from prior meta-analyses (see “discussion”, p123) ^[1].  The US study found similar rates of adverse events as the Finnish trial; NNH = 6.2 (diaper-area dermatitis) and NNH = 10 (diarrhoea).

Moreover, a useful finding is that around 70% of children diagnosed with AOM according to the strict criteria who were treated with placebo were improved by day 8.

A practical interpretation of these results is that antibiotic treatment may improve outcomes in acute otitis media, but the effect may be modest in many children aged 6 to 35 months with adverse events as likely as benefits. Children most likely to benefit from antibiotics are those with a clear diagnosis of acute otitis media:

• history of acute onset of symptoms and signs
• demonstrable middle ear effusion
• signs and symptoms of middle ear inflammation

The Antibiotic Therapeutic Guidelines (2010) recommends immediate antibiotic therapy for AOM with systemic features (e.g., fever, vomiting), and symptomatic treatment with potential delayed antibiotic therapy for AOM without systemic features ^[2].

References

1. Tähtinen PA, Laine MK, Huovinen P, et al. A placebo-controlled trial of antimicrobial treatment for acute otitis media. N Engl J Med 2011; 364: 116-26
2. “Otitis media” in Antibiotics, version 14. Therapeutic Guidelines Limited, November 2010.
3. Hoberman A, Paradise JL, Rockette HE, et al. Treatment of acute otitis media in children under 2 years of age. N Engl J Med 2011; 364: 105-15

Reviewer: Angie Pinto