Nine RCTs (n=2,549) were included in the assessment of whether antibiotics were effective. Seventy-two RCTs (n=18,242) were included in the assessment of the relative effectiveness of different antibiotic regimens.
About half of the studies were not of adequate quality; diagnostic criteria for AOM were not uniform across the studies; the outcomes and outcome definitions were not uniform across the studies; most studies seemed to have insufficient statistical power; subgroup analyses were not possible. Based on the exclusion factors of the investigations used in this analysis, the study findings are most applicable to children without co-morbidities and with AOM of lesser severity.
The meta-analysis demonstrated a reduction in the clinical failure rate within 2 to 7 days of 12.3% (95% confidence interval, CI: 2.8, 21.8) in favour of ampicillin or amoxicillin therapy, compared with placebo or observational treatment. This result was generally robust to a sensitivity analysis. Eight children with AOM would need to be treated with ampicillin or amoxicillin, rather than no antibiotic treatment, to avoid a case of clinical failure.
The meta-analyses did not demonstrate a significant difference in the clinical failure rates in children with AOM treated with ampicillin or amoxicillin, compared with children treated with penicillin, cefaclor or cefixime.
The meta-analysis did not demonstrate a significant difference in the clinical failure rates in children treated with trimethoprim-sulfamethoxazole, compared with children treated with cefaclor for AOM.
The meta-analysis demonstrated that children treated with cefixime had an 8.4% greater rate of diarrhoea than children treated with ampicillin or amoxicillin. Twelve children with AOM would need to be treated with ampicillin or amoxicillin, rather than cefixime, to avoid one case of diarrhoea.
No RCTs were found that compared oral fluoroquinolones with other antibiotics.
A single study demonstrated no difference in the clinical effect of high-dose versus standard-dose amoxicillin-clavulanate.
A single study did not demonstrate a difference in the clinical effect of taking high-dose amoxicillin twice daily versus three times a day.
The meta-analysis did not demonstrate a difference in the clinical effect between short- and long-duration therapy when comparing single-dose ceftriaxone therapy with 7 to 10 days' amoxicillin therapy, and azithromycin therapy for less than 5 days with 7 to 10 days' amoxicillin-clavulanate therapy.
The meta-analysis demonstrated that children treated with 7 to 10 days' amoxicillin-clavulanate had a 19.2% (95% CI: 9.2, 29.2) greater rate of overall adverse effects, and a 12.9% (95% CI: 4.5, 21.2) greater rate of gastrointestinal adverse effects, than children treated with 5 days' azithromycin. Eight children would need to be treated with azithromycin, rather than amoxicillin-clavulanate, to avoid a gastrointestinal adverse event.
The authors also reported the results of previous meta-analyses, but these are not reproduced here on the grounds that they do not meet the stated inclusion criteria for this review.