Short- versus long-course antibacterial therapy for community-acquired pneumonia : a meta-analysis.

BACKGROUND The evidence for traditionally recommended 7- to 14-day duration of antibacterial therapy for community-acquired pneumonia (CAP) is not well established. OBJECTIVES We endeavoured to assess the effectiveness and safety of shorter than traditionally recommended antibacterial therapy for CAP. METHODS We performed a meta-analysis of randomized controlled trials (RCTs) comparing short- (< or = 7 days) versus long- (> or = 2 days difference) course therapy for CAP with the same antibacterial regimens, in the same daily dosages. RESULTS Five RCTs involving adults (including outpatients and inpatients who did not require intensive care) and two RCTs involving children (aged 2-59 months, residing in developing countries) were included. All RCTs were double-blind and assessed patients with CAP of mild to moderate severity. No differences were found between short- (adults 3-7 days; children 3 days) and long- (adults 7-10 days; children 5 days) course regimens (adults - amoxicillin, cefuroxime, ceftriaxone, telithromycin and gemifloxacin; children - amoxicillin) regarding clinical success at end-of-therapy (six RCTs; 5107 patients [1095 adults, 4012 children]; fixed-effect model [FEM]; odds ratio [OR] = 0.89; 95% CI 0.74, 1.07), clinical success at late follow-up, microbiological success, relapses, mortality (seven RCTs; 5438 patients; FEM; OR = 0.57; 95% CI 0.23, 1.43), adverse events (five RCTs; 3214 patients; FEM; OR = 0. 90; 95% CI 0.72, 1.13) or withdrawals as a result of adverse events. No differences were found in subset analyses of adults or children, and of patients treated with no more than 5-day short-course regimens versus at least 7-day long-course regimens. CONCLUSION No difference was found in the effectiveness and safety of short- versus long-course antimicrobial treatment of adult and paediatric patients with CAP of mild to moderate severity.