Randomized trial of amoxicillin for pneumonia in Pakistan

Document Type



Paediatrics and Child Health


Background: The World Health Organization (WHO) recommends oral amoxicillin for patients who have pneumonia with tachypnea, yet trial data indicate that not using amoxicillin to treat this condition may be noninferior to using amoxicillin.
Methods: We conducted a double-blind, randomized, placebo-controlled noninferiority trial involving children at primary health care centers in low-income communities in Karachi, Pakistan. Children who were 2 to 59 months of age and who met WHO criteria for nonsevere pneumonia with tachypnea were randomly assigned to a 3-day course of a suspension of amoxicillin (the active control) of 50 mg per milliliter or matched volume of placebo (the test regimen), according to WHO weight bands (500 mg every 12 hours for a weight of 4 to <10 kg, 1000 mg every 12 hours for a weight of 10 to <14 kg, or 1500 mg every 12 hours for a weight of 14 to <20 kg). The primary outcome was treatment failure during the 3-day course of amoxicillin or placebo. The prespecified noninferiority margin was 1.75 percentage points.
Results: From November 9, 2014, through November 30, 2017, a total of 4002 children underwent randomization (1999 in the placebo group and 2003 in the amoxicillin group). In the per-protocol analysis, the incidence of treatment failure was 4.9% among placebo recipients (95 of 1927 children) and 2.6% among amoxicillin recipients (51 of 1929 children) (between-group difference, 2.3 percentage points; 95% confidence interval [CI], 0.9 to 3.7). Results were similar in the intention-to-treat analysis. The presence of fever and wheeze predicted treatment failure. The number needed to treat to prevent one treatment failure was 44 (95% CI, 31 to 80). One patient (<0.1%) in each group died. Relapse occurred in 40 children (2.2%) in the placebo group and in 58 children (3.1%) in the amoxicillin group.
Conclusions: Among children younger than 5 years of age with nonsevere pneumonia, the frequency of treatment failure was higher in the placebo group than in the amoxicillin group, a difference that did not meet the noninferiority margin for placebo. (Funded by the Joint Global Health Trials Scheme [of the Department for International Development, Medical Research Council, and Wellcome] and others; RETAPP ClinicalTrials.gov number, NCT02372461.).

Publication (Name of Journal)

New England Journal of Medicine