Three days of amoxicillin is effective for pneumonia with tachypnea in young children

Référence

Jehan F, Nisar I, Kerai S, et al. Randomized trial of amoxicillin for pneumonia in Pakistan. N Engl J Med 2020;383(1):24-34.

Sommaire

In low-resource health settings, pneumonia is often diagnosed clinically based on typical signs and symptoms accompanied by tachypnea. If the prevalence of HIV disease is low, a 3-day course of amoxicillin is recommended. This study identified 4002 children, ages 2 months to 59 months in Pakistan who presented with cough or tachypnea and were included if the respiratory rate was greater than 50 in children 2 months to 11 months old and greater than 40 in children 12 months to 59 months old. If wheezing was present, the children were given a bronchodilator and if tachypnea persisted, they were still included in the study. Patients with retractions or any red flags for severe pneumonia, such as oxygen saturation of less than 90%, were excluded. Children were randomized to receive a 3-day weight-dosed course of amoxicillin given twice daily or matching placebo. This was designed as a noninferiority trial with an inferiority margin of 1.75%, assuming a treatment failure rate of 3.5% with amoxicillin. Analysis was by intention to treat and groups were balanced at baseline. The median age of patients was 13 months, 59% had fever, and more than 99% presented with cough. The likelihood of treatment failure was low overall, but was higher in the placebo group (4.8% vs 2.5%; P < .05; NNT = 43). The likelihood of relapse at 4 days to 14 days was also low, with no significant difference between groups. A multivariate regression analysis found that treatment failure was more likely in patients with subjective or measured fever, those with wheeze, those with a respiratory rate in the top tertile, and those with poor household air quality. This provides the basis for further research to identify groups most likely to benefit from an antibiotic, and others in whom it might be avoided. Another study in the same issue found that 3 days was noninferior to 5 days of amoxicillin in Malawian children with pneumonia and retractions.