Acute otitis media

Probably 60% of infections are mixed viral/bacterial! Pneumococcus, Haemophilus (non capsulated), Moraxella catarrhalis. Group A Strep is characterized by older age, higher local aggressiveness (ie tympanic perforation and mastoiditis) but lower rates of fever and respiratory symptoms.

AOM is associated with dummy use, adenoids/tonsillitis.

60% of placebo treated children were pain free within 24 hours of presentation [Cochrane 2004, PMID 14973951]. Antibiotics do not increase this proportion. At 2-7 days after presentation, antibiotics reduce by a third the number of children who still have pain (only 14% of the original number), giving a NNT of 15. Too few cases of complications eg mastoiditis to be able to comment on whether antibiotics are useful for preventing such complications.NICE clinical knowledge summary gives NNT of 4000 to prevent 1 case of mastoiditis!

Management is therefore primarily good analgesia. No role for decongestants/antihistamines (Cochrane/NICE).

SIGN guideline 66 on AOM in primary care does not recommend routine antibiotics (but if used, Amoxicillin or co-amox for 5 days recommended). SIGN warns that evidence is poor in infants or in severe disease.

NICE clinical knowledge summary states that at initial presentation, pain and fever should be treated with paracetamol or ibuprofen, at maximum doses if necessary.  No benefit from using both (though poor quality evidence).  For most children antibiotics can be delayed until day 4 of illness (mean duration of illness is 4 days).

Eardrops containing analgesia and anaesthetic (phenazone and lidocaine = “Otigo”) work within 10 minutes – they also significantly reduce the number of people who go on to have antibiotics. Not to be used instead, however, and not to be used where perforation/discharge.

Antibiotics should be offered however at presentation to people who are systemically unwell. Depending on severity, antibiotics should be considered where child is under 2yrs with bilateral otitis media, or if there is perforation and/or discharge in the ear canal.  This advice is also reflected in the British National Formulary for children.

Delaying antibiotics certainly reduces prescriptions, metanalysis of 4 studies in Cochrane did not find any significant difference in pain at 3-7 days, although all but 1 reported some benefit in the immediate treatment group.

A study comparing delayed antibiotics with vs without a prescription (ER based) found high and comparable parental satisfaction rates with both approaches [Chao Peds 2008 PMID 18450878]. In the Cochrane review of delayed antibiotics for URTI, which looked at both adults and children, immediate antibiotics were felt to be more likely to confer modest benefits than delayed antibiotics, with no differences in complication rates between immediate vs delayed antibiotics. Immediate antibiotics had slightly higher levels of patient satisfaction than delayed antibiotics but of marginal clinical significance (92% versus 87%). Concluded that as no evidence for benefit of delayed vs no prescription, best to offer nothing (likely to result in the least antibiotic use). [Cochrane 2011 Delayed antibiotics in URTI, PMID 17636757]

BNFc does suggest antibiotics if:

  • no improvement after 72 hours,
  • clinical deterioration,
  • systemically unwell,
  • at high risk of serious complications (eg in immunosuppression, cystic fibrosis),
  • mastoiditis is present,
  • under 2 years of age with bilateral otitis media.

BNFc also suggests that perforation of the tympanic membrane usually heals spontaneously without treatment; but treat if there is no improvement (eg pain or discharge persists).

Lancet 2006 [pmid 17055944] meta-analysis supports the idea of treating under 2s with bilateral signs (NNT=4), but unlike the BNFc supports treating otorrhoea (NNT=3), as does Cochrane!

Exponential increase in drug resistance and multiresistance. Given how effective placebo is, an effective drug has to do considerably better! Amoxicillin no longer useful (but still recommended by NICE), cefaclor and TMP-SMX not good for borderline resistant pneumococci, azithromycin does not achieve MIC for Hib/Pneumo in ear (although cure rates may not be all that bad…). Co-amox in double dose ie 90 mg/kg/d in 2 divided doses is effective, but increasing resistance.

Treatment leads to higher numbers of resistant species in nasopharynx, esp dually resistant bugs. Azithromycin persists in body for several weeks so is excellent for inducing resistance. Despite overall trend towards reducing antibiotic usage in AOM, most reduction in amoxicillin, with increased prescribing of quinolones and azithromycin. So don’t treat at all unless added features (unless under a year). [Ron Dagan, Beersheva]

Lots of potential complications:

  • Acute mastoiditis = a type of osteomyelitis, with potential for intracranial spread and meningitis/cerebral abscess formation.  Classic signs are erythema, swelling and tenderness behind the ear, with deviation of the pinna.
  • Gradenigo syndrome = intratemporal extension of AOM, causing VI nerve palsy (via apex of petrous temporal bone).
  • Grisel’s syndrome = non-traumatic subluxation of the atlanto-axial joint caused by inflammation of the adjacent tissues.  Clue is trismus and torticollis.