Modern treatment of otitis media in children

Acute otitis media (AOM) is a bacterial infection that occurs almost exclusively after a viral upper respiratory tract infection; the main pathogens are Streptococcus pneumoniae, nontypeable Haemophilus influenzae and Moraxella catarrhalis. Its incidence is falling thanks to pneumococcal conjugate vaccines and stricter diagnostic criteria: by the age of 2, 41% of children have had at least one episode of AOM and 13% have had three or more. ^[1]

Diagnosis

The key diagnostic sign of AOM is a bulging eardrum — this is the sign that correlates most reliably with the presence of bacteria in the middle ear. Otitis media with effusion (OME) is diagnosed when there are signs of fluid (cloudiness, reduced mobility) without bulging. This distinction is essential, because AOM may require antibiotics, whereas OME usually resolves on its own. ^[1]

Treatment approach: watchful waiting vs antibiotics

Children with mild or moderate symptoms can either be treated with antibiotics or observed with a “safety-net prescription”. A meta-analysis of 13 RCTs showed that antibiotics reduce the risk of persistent symptoms by 29% by day 2–3 (NNT=20), by 24% by day 4–7 (NNT=17) and by 67% by day 10–12 (NNT=7). Antibiotics also halve the risk of AOM in the other ear (from 19% to 10%) and reduce the risk of eardrum perforation by two thirds (from 5% to 2%). ^[1]

Importantly, in the only RCT that used a validated symptom scale, the risk of persistent symptoms was 67% lower in the amoxicillin-clavulanate group than with placebo (NNT=7). ^[1]

First-line antibiotic therapy

High-dose amoxicillin (80–90 mg/kg/day in 2 doses) remains the first-line drug. Amoxicillin-clavulanate is recommended when H. influenzae is likely to predominate: antibiotic use in the previous 30 days, otitis-conjunctivitis syndrome or spontaneous eardrum perforation. Oral cephalosporins are considerably less effective against penicillin-nonsusceptible S. pneumoniae and should generally be avoided. ^[1]

Duration of treatment

In children under 24 months, a 5-day course is ineffective: the treatment failure rate was twice as high with a 5-day course as with a 10-day course (34% vs 16%), while the rate of side effects did not differ significantly. ^[1]

Refractory AOM

When several courses of oral antibiotics have failed, intramuscular ceftriaxone (50 mg/kg/day × 3 days) and tympanocentesis to identify the pathogen and drain the ear are indicated. ^[1]

Tympanostomy for recurrent AOM

A large RCT showed that in recurrent AOM the rate of episodes over 2 years of follow-up was similar between the tympanostomy group and the episodic antibiotic treatment group. ^[1]

II. Watchful waiting: selection criteria

The AAP guideline (2013) sets out specific criteria for observation without immediate antibiotics: ^[2]

• Children 24 months and older with unilateral or bilateral non-severe AOM (pain for less than 48 h, temperature below 39°C) — observation is an equally valid option
• Children 6–23 months with unilateral non-severe AOM — observation is acceptable as a shared decision with the parents
• Mandatory conditions: a follow-up mechanism (a repeat visit or a “safety-net prescription”), and starting antibiotics if the child gets worse or does not improve within 48–72 hours

About two thirds of children in the observation group do not need “rescue” antibiotics, which can potentially reduce antibiotic use by 65%. ^[2]

III. Pain relief — the cornerstone of treatment

The Cochrane review by de Sévaux et al. (2023) stresses that analgesic therapy is the basis of AOM treatment, whether or not antibiotics are prescribed: ^[3]

Antihistamines and decongestants are not effective in AOM and are associated with side effects (drowsiness, diarrhea, rash, persistent effusion). ^[1][8]

IV. Antibiotic therapy: choice of drug and duration

First line

The retrospective cohort study by Frost et al. (2022, n=1,051,007 children) confirmed that despite microbiological changes, amoxicillin remains the most effective oral drug: the combined rate of failure and recurrence was 1.7% for amoxicillin vs 10–11% for amoxicillin-clavulanate, cefdinir and azithromycin. ^[9]

Second line and refractory cases

If amoxicillin fails — amoxicillin-clavulanate (90 mg/kg/day of amoxicillin + 6.4 mg/kg/day of clavulanate, a 14:1 ratio). Alternatives: cefdinir (14 mg/kg/day), cefuroxime (30 mg/kg/day), cefpodoxime (10 mg/kg/day). If oral intake is not possible — intramuscular ceftriaxone 50 mg/kg. ^[2]

Length of the course

The RCT by Hoberman et al. (NEJM, 2016, n=520, children aged 6–23 months) convincingly showed that a 5-day course is not noninferior to a 10-day course: clinical failure was 34% vs 16% (NNT=6 in favor of the 10-day course). ^[10]

“In this noninferiority trial involving children 6 to 23 months of age with acute otitis media, reduced-duration treatment with amoxicillin–clavulanate for 5 days was less effective than standard-duration treatment for 10 days”. — Alejandro Hoberman, M.D., et al., University of Pittsburgh School of Medicine and other institutions. Shortened Antimicrobial Treatment for Acute Otitis Media in Young Children. N Engl J Med. December 22, 2016. Used under license from The New England Journal of Medicine.

Recommendations on duration: ^[2]

• Under 2 years and severe course → 10 days
• 2–5 years, mild/moderate course → 7 days
• 6 years and older, mild/moderate course → 5–7 days

V. Tympanocentesis

Tympanocentesis is indicated for: ^[1][11][12]

• Failure of several courses of antibiotics (refractory AOM)
• A severely ill/toxic child with suspected sepsis
• The need to identify the pathogen for targeted therapy

The procedure can be performed safely in an outpatient setting without general anesthesia. ^[12]

VI. Otitis media with effusion (OME): management

The AAO-HNS (2016) and NICE (2024) guidelines recommend: ^[13][14][15]

• Watchful waiting for 3 months or more — high rate of spontaneous resolution
• Autoinflation (nasal balloon) — low-quality evidence of short-term improvement in hearing; suitable for children aged 4–11 ^[13]
• Antibiotics are not recommended (clinically insignificant improvement) ^[13][16]
• Oral and intranasal steroids — a Cochrane review (2023) showed that oral steroids are probably no better than placebo for restoring hearing at 1 year (moderate certainty); intranasal steroids may reduce persistence of effusion in the short term, but the evidence is of very low certainty. The OSTRICH study (Lancet, 2018, n=389) found no clinically meaningful effect of oral steroids ^[17][18]
• Decongestants, antihistamines, mucolytics, anti-reflux drugs — do not improve outcomes ^[13]
• Tympanostomy — for persistent bilateral OME for 3 months or more with hearing loss of 16 dB or more ^[14]
• Hearing aids — an alternative to surgery for fluctuating hearing loss or narrow ear canals ^[13]

VII. Prevention

VIII. Summary table: AOM treatment algorithm