Acute otitis media is one of the most common childhood infections, yet there is considerable debate over optimal treatment. The rate of antibiotic prescription varies enormously between countries and between groups within countries. Systematic review suggests that antibiotic therapy is only marginally beneficial and is associated with potential adverse effects and increased antibiotic resistance. Little and colleagues studied 315 children between six months and 10 years of age to identify factors associated with poor outcome in otitis media. They also sought to establish which children are most likely to benefit from early antibiotic therapy.

They studied children brought to family physicians in southern England with acute otalgia. Children with otoscopic signs of acute inflammation were eligible for the study if they did not have symptoms of serious disease or a history of significant complications from otitis. Patients who had used antibiotics within the previous two weeks and those who were too sick to delay treatment were excluded. Other exclusion criteria were a diagnosis of chronic suppurative otitis media or otitis with effusion, and otoscopic appearance consistent with crying or fever alone. The children were randomly assigned to immediate-antibiotic therapy or supportive therapy for 72 hours with a prescription for antibiotics to be taken at the end of that period if the child was still symptomatic (delayed-antibiotic therapy). Parents rated the perceived severity of pain and recorded daily symptoms, temperature, number of episodes of distress, and use of acetaminophen. Family physicians recorded physical signs, days of illness, and prescriptions for antibiotics. The outcome assessment concerned symptoms at 72 hours after the initial consultation.

The predictors of persistent earache at 72 hours were ear discharge (adjusted odds ratio [OR], 2.6), history of at least three courses of antibiotics over the previous year (OR, 0.2), and dissatisfaction with the consultation (OR, 3.0). Most children with persistent earache on day 3 had mild pain, with a mean score of 2.6 on a scale of zero to 10.

Parents indicated that disturbed sleep and episodes of distress were more relevant outcomes. Disturbed sleep was predicted by high initial temperature (greater than 99.5°F [37.5°C]; OR, 2.4); vomiting (OR, 2.1); cough (OR, 2.3); and ear discharge (OR, 2.1). Distress was predicted by initial high temperature (OR, 4.5), vomiting (OR, 2.6), and cough (OR, 2.0).

Subgroup analysis showed that children with high temperature or vomiting who were given immediate antibiotics were significantly less likely to report distress or disturbed sleep by day 3 than children with the same initial symptoms who were assigned to the delayed-antibiotic group. In children without initial high temperature or vomiting, no significant difference was reported in distress or sleep disturbance. In the less-severely ill children, 15 percent of those treated immediately and 19 percent of those in the delayed-treatment group reported distress on day 3. Disturbed sleep was reported by 20 percent of the immediate-treatment group and by 27 percent of the delayed-treatment group.

The authors conclude that antibiotic therapy has little benefit in children with otitis media who do not have fever or vomiting. Conversely, the greatest benefit from immediate-antibiotic therapy is likely in children who have high temperature or vomiting. Even in children with more severe symptoms, approximately one half improve by day 3 without antibiotic prescription (see the accompanying table).

editor's note: The hidden gem in this study is the powerful effect of “satisfaction with the consultation.” What a validation for good doctoring! Taking extra time and care to give families confidence in the management plan is a good investment. It provides better outcomes for the index episode and is likely to influence parental behavior with future illnesses. The other gems are the statistics that we can use to help parents make good choices and have realistic expectations for the course of the illness. Even if antibiotics are prescribed, about one third of the most ill and 15 to 20 percent of the less-symptomatic children remain distressed after three days of therapy. Parents must be coached not to expect complete and immediate resolution of symptoms. Conversely, many more patients than we realize are willing to try symptomatic care if we give them the evidence from the studies plus the reassurance of prompt medical attention if things do not go according to plan.—a.d.w.