Antibiotics for acute otitis media in children

摘要: Background Acute otitis media (AOM) is one of the most common diseases in early infancy and childhood. Antibiotic use for AOM varies from 56% Netherlands to 95% USA, Canada Australia. This an update a Cochrane review first published The Library Issue 1, 1997 previously updated 1999, 2005, 2009 2013. Objectives To assess effects antibiotics children with AOM. Search methods We searched CENTRAL (2015, 3), MEDLINE (1966 April week 3, 2015), OLDMEDLINE (1958 1965), EMBASE (January 1990 Current Contents CINAHL (2008 2015) LILACS 2015). Selection criteria Randomised controlled trials (RCTs) comparing 1) antimicrobial drugs placebo 2) immediate antibiotic treatment expectant observation (including delayed prescribing) AOM. Data collection analysis Two authors independently assessed trial quality extracted data. Main results For against placebo, 13 RCTs (3401 3938 episodes) high-income countries were eligible had generally low risk bias. combined results revealed that by 24 hours start treatment, 60% recovered whether or not they antibiotics. Pain was reduced at (risk ratio (RR) 0.89, confidence interval (CI) 0.78 1.01) but almost third fewer residual pain two three days (RR 0.70, CI 0.57 0.86; number needed treat additional beneficial outcome (NNTB) 20). A quarter four seven 0.76, 0.63 0.91; NNTB 16) two-thirds 10 12 0.33, 0.17 0.66; 7) compared placebo. Antibiotics did reduce abnormal tympanometry findings weeks 0.82, 0.74 0.90; 11), six eight 0.88, 1.00; tympanic membrane perforations 0.37, 0.18 0.76; 33) halved contralateral episodes 0.49, 0.25 0.95; 11) However, neither months 0.97, 0.76 1.24) nor late recurrences 0.93, 1.10) when Severe complications rare differ between treated those Adverse events (such as vomiting, diarrhoea rash) occurred more often taking 1.38, 1.19 1.59; harmful (NNTH) 14). Funnel plots do suggest publication Individual patient data meta-analysis subset included found be aged less than years bilateral AOM, both otorrhoea. For observation, five (1149 children) moderate Four (1007 reported could used this review. From these trials, 959 days. No difference detectable 0.75, 0.50 1.12). One (247 on 11 14 Immediate associated reduction 0.91, 0.75 observation. Additionally, no differences weeks, recurrence observed groups. serious either group. substantial increased rash 1.71, 1.24 2.36; NNTH 9). Results individual including high-quality (1643 also our showed seem younger age (NNTB 4) otorrhoea 3). Authors' conclusions This reveals have effect pain, slight following only modest perforations, AOM. In countries, cases spontaneously remit without complications. benefits must weighed possible harms: every child experienced adverse event would if withheld. Therefore clinical management should emphasise advice about adequate analgesia limited role are useful under otorrhoea. For other mild disease observational approach seems justified.