Clinical Question: Are three days of treatment with amoxicillin as effective as five days of the same therapy in children with nonsevere pneumonia?

Setting: Outpatient (primary care)

Study Design: Randomized controlled trial (double-blinded)

Synopsis: More than 2,100 children between two months and six years of age with non-severe pneumonia were enrolled in this study. The diagnostic criteria were clinical: cough, rapid respiration, or difficulty breathing that was not deemed to be indicative of severe pneumonia. Fever was not a criterion. Children were randomized (using concealed allocation) to receive three or five days of oral amoxicillin in dosages of 30 to 50 mg per kg per day. Treatment failure was defined as retractions (i.e., chest indrawing), convulsions, drowsiness, the inability to drink, or a cutoff respiratory rate that was too high or an oxygen saturation that was too low on day 5.

In the intention-to-treat analysis, cure rates were similar between the groups (90 percent), as were relapse rates within the next seven to nine days (5 percent). Cure rates were the same whether or not the children had wheezes and did not vary by age. Hospitalizations were infrequent but similar in the two groups (2 percent). The study had the power to find a 5 percent difference in treatment failures, assuming a 12 percent failure rate with the longer therapy.

Bottom Line: Outpatient treatment with three days of amoxicillin produces results similar to five days of treatment in children with presumed pneumonia. In both study groups, treatment failed in 10 percent of children. (Level of Evidence: 1b)