A Pragmatic Trial of Antibiotics and Supportive Care for Severe Pneumonia in Hospitalized Children

Le résumé fourni par la source

Abstract Background Evidence to guide the choice of injectable antibiotics and supportive care for children with severe pneumonia is limited and may not reflect changes in epidemiology associated with vaccination and antimicrobial resistance. Methods In this pragmatic, open-label, factorial, randomized trial conducted in 16 hospitals in Kenya, children aged 2-59 months with World Health Organization-defined severe pneumonia were assigned to receive one of three injectable antibiotic regimens: benzylpenicillin plus gentamicin (standard care), ceftriaxone, or amoxicillin-clavulanic acid. Eligible children were also randomly assigned to receive nasogastric tube feeding or intravenous fluids. The primary outcome was death from any cause by day 5 after enrollment. Results A total of 4393 children underwent randomization to the antibiotic groups, and 1064 to the supportive care groups. By day 5, deaths occurred in 87/1463 children (6.0%) receiving benzylpenicillin plus gentamicin, 82/1458 (5.6%) receiving amoxicillin-clavulanic acid (adjusted risk ratio [aRR], 0.94; 97.5% confidence interval [CI], 0.67 to 1.31), and 81/1462 (5.5%) receiving ceftriaxone (aRR vs. benzylpenicillin plus gentamicin, 0.95; 97.5% CI, 0.68 to 1.33). Death by day 5 occurred in 30/531 children (5.7%) receiving nasogastric tube feeding and 35/532 (6.7%) receiving intravenous fluids (aRR, 1.13; 97.5% CI, 0.71 to 1.79). Secondary outcomes were similar across groups. Conclusions Among children hospitalized with severe pneumonia, outcomes with benzylpenicillin plus gentamicin were similar to those with ceftriaxone or amoxicillin-clavulanic acid, and nasogastric tube feeding was similar to intravenous fluids with respect to mortality and safety. Trial registration ClinicalTrials.gov: NCT04041791 ; Pan African Clinical Trial Registry: PACTR202106720981298

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.