Injectable penicillin is the recommended treatment for WHO-defined severe pneumonia (lower chest indrawing). If oral amoxicillin proves equally effective, it could reduce referral, admission, and treatment costs. We aimed to determine whether oral amoxicillin and parenteral penicillin were equivalent in the treatment of severe pneumonia in children aged 3–59 months.

This multicentre, randomised, open-label equivalency study was undertaken at tertiary-care centres in eight developing countries in Africa, Asia, and South America. Children aged 3–59 months with severe pneumonia were admitted for 48 h and, if symptoms improved, were discharged with a 5-day course of oral amoxicillin. 1702 children were randomly allocated to receive either oral amoxicillin (n=857) or parenteral penicillin (n=845) for 48 h. Follow-up assessments were done at 5 and 14 days after enrolment. Primary outcome was treatment failure (persistence of lower chest indrawing or new danger signs) at 48 h. Analyses were by intention-to-treat and per protocol.

Treatment failure was 19% in each group (161 patients, pencillin; 167 amoxillin; risk difference –0·4%; 95% CI –4·2 to 3·3) at 48 h. Infancy (age 3–11 months; odds ratio 2·72, 95% CI 1·95 to 3·79), very fast breathing (1·94, 1·42 to 2·65), and hypoxia (1·95, 1·34 to 2·82) at baseline predicted treatment failure by multivariate analysis.

Injectable penicillin and oral amoxicillin are equivalent for severe pneumonia treatment in controlled settings. Potential benefits of oral treatment include decreases in (1) risk of needle-borne infections; (2) need for referral or admission; (3) administration costs; and (4) costs to the family.