METHODS: We undertook a multicenter, double-blind, superiority, randomized controlled trial involving 7 Australian, New Zealand, and Malaysian hospitals. Children aged 3 months to ≤5 years hospitalized with radiographic-confirmed CAP who received 1-3 days of intravenous antibiotics, then 3 days of oral amoxicillin-clavulanate, were randomized to either extended-course (8-day oral amoxicillin-clavulanate) or standard-course (8-day oral placebo) arms. Children were reviewed at 12 and 24 months. The primary outcome was children with the composite endpoint of chronic respiratory symptoms/signs (chronic cough at 12 and 24 months; ≥1 subsequent hospitalized acute lower respiratory infection by 24 months; or persistent and/or new chest radiographic signs at 12-months) at 24-months postdischarge, analyzed by intention-to-treat, where children with incomplete follow-up were assumed to have chronic respiratory symptoms/signs ("worst-case" scenario).

RESULTS: A total of 324 children were randomized [extended-course (n = 163), standard-course (n = 161)]. For our primary outcome, chronic respiratory symptoms/signs occurred in 97/163 (60%) and 94/161 (58%) children in the extended-courses and standard-courses, respectively [relative risk (RR) = 1.02, 95% confidence interval (CI): 0.85-1.22]. Among children where all sub-composite outcomes were known, chronic respiratory symptoms/signs between groups, RR = 1.10, 95% CI: 0.69-1.76 [extended-course = 27/93 (29%) and standard-course = 24/91 (26%)]. Additional sensitivity analyses also revealed no between-group differences.