Antibiotic treatment duration for community-acquired pneumonia in children

Length of treatment

The Infectious Diseases Society of America and Pediatric Infectious Diseases Society guidelines from 2011 state that the most frequently studied treatment duration of pediatric CAP is 10 days; however, shorter durations may be effective for patients with mild disease treated in outpatient settings.³ Similarly, the Canadian Paediatric Society statement on uncomplicated CAP from 2015 highlights that hospitalized children require 7 to 10 days of antibiotic therapy; however, 5 days may be appropriate for outpatients.¹ The optimal duration of antibiotic treatment for outpatient management of uncomplicated CAP has been a topic of interest recently, as new evidence reveals that the historically used 10-day duration may be unnecessary.^4-11

Five- and 10-day treatment courses

Three studies compared 5 and 10 days of treatment. A double-blind, randomized, placebo-controlled trial by researchers in Israel published in 2014 started comparing short (3-day) and long (10-day) durations of high-dose amoxicillin (80 mg/kg/day divided into 3 doses) for radiographically confirmed alveolar CAP in 108 febrile children with leukocytosis aged 6 to 59 months who were suitable for outpatient management.⁶ The primary outcome, treatment failure, was described as “a situation assessed by the study physicians to be nonresponsive or deteriorating to the point that the study drug needed to be replaced, or if the patient was hospitalized due to deterioration in medical condition.”⁶ Statistically significant treatment failure rates early in the 3-day arm compared with the 10-day arm (40% vs 0%, respectively) led to a change in methods and the investigators evaluated 5 days versus 10 days of therapy. With no treatment failure in either arm, the authors concluded that 5 days of treatment was not inferior to 10 days of treatment of uncomplicated pneumonia. The relatively small number of patients included in this study, the strict criteria of diagnosis, and the restricted age of the participants limit the generalizability of these findings.

Another placebo-controlled noninferiority trial in 2 Canadian emergency departments also compared 5 days and 10 days of high-dose amoxicillin (90 mg/kg/day divided into 3 doses) in children 6 months to 10 years old presenting with CAP (radiologic and clinical diagnosis).¹⁰ The primary outcome of this study was clinical cure with strict criteria, which led many participants whose CAP responded to antibiotics to have been categorized as having treatment failure, who outside of the trial would have been considered to have had an adequate response. Therefore, a secondary definition of clinical cure was developed: clinical cure not requiring additional intervention. Results for 252 patients (mean age 2.5 years) were analyzed. For the clinical cure not requiring additional intervention outcome, the intention-to-treat analysis showed a rate of 93.5% in the 5-day group compared with 90.4% in the 10-day group (risk difference [RD]=0.028; 97.5% confidence limit [CL]=−0.038). The per protocol analysis rates were 95.5% and 95.4% (RD=−0.006; 97.5% CL=−0.055) and the strict per protocol analysis rates (including only patients with radiographic CAP) were 95.0% and 94.0%, respectively (RD=−0.004; 97.5% CL=−0.071). Hence, 5-day treatment was statistically noninferior to 10-day treatment.

A more recent multicentre, randomized, double-blind, placebo-controlled superiority trial compared 5 days and 10 days of high-dose amoxicillin, amoxicillin–clavulanic acid combination (80 to 100 mg/kg/day divided into 2 doses), or cefdinir (12 to 16 mg/kg/day divided into 2 doses) in 380 patients (mean age 35.7 months) with uncomplicated CAP.¹¹ If patients showed clinical improvement after 5 days of antibiotic treatment, they were randomized to an additional 5 days of antibiotic treatment or placebo. The primary outcome was the response adjusted for duration of antibiotic risk at days 6 to 10. The estimated probabilities of a more desirable response adjusted for duration of antibiotic risk for 5 days were 0.69 (95% CI 0.63 to 0.75) at days 6 to 10 and 0.63 (95% CI 0.57 to 0.69) at days 19 to 25. Hence, the shorter duration was superior to the longer one. Limitations of this study include variable clinical diagnosis of CAP, lack of microbiological testing, and applicability limited only to children younger than 6 years.

Even shorter duration?

With a 5-day treatment duration proving to be sufficient for most children, investigation of an even shorter duration gained momentum. A multicentre, randomized, placebo-controlled, 4-arm noninferiority trial conducted in 814 children older than 6 months in the United Kingdom and Ireland in 2021 compared low-dose (35 to 50 mg/kg/day divided into 2 doses) and high-dose (70 to 90 mg/kg/day divided into 2 doses) oral amoxicillin and a 3-day versus a 7-day treatment duration for uncomplicated clinically diagnosed CAP.⁵ The noninferiority margin for the primary outcome of clinically indicated treatment within 28 days of starting the study drug was met for both the 3-day and 7-day arms (12.5% vs 12.5%) and for the low-dose versus high-dose comparison (12.6% vs 12.4%). At baseline and at end of treatment, 42% and 29.5% of children, respectively, had nasopharyngeal samples found to be colonized with S pneumoniae. No samples were identified as having penicillin-resistant S pneumoniae and the rates of antibiotic-associated adverse effects were similar. While the study did not analyze the data based on whether patients were already receiving antibiotics when initially admitted to the hospital, 3 days of treatment was noninferior to 7 days of treatment in regard to nasopharyngeal swab findings.

Systematic assessment of studies

A systematic review and meta-analysis published in 2023 supported short-course antibiotic therapy as noninferior to a standard course of therapy when evaluating antibiotic retreatment, hospitalization, treatment failure, and antibiotic-related adverse effects.⁷ Another systematic review and meta-analysis of randomized controlled trials comparing a shorter course with a longer course of antimicrobial therapy using the same antibiotic medication in children with nonsevere CAP was completed by Li et al in 2022.⁸ Based on 9 studies, 7 of which used 3 days as the short duration, the study reported that for treatment failure and relapse rates short-course treatment was noninferior to longer-course treatment. A subgroup analysis of children aged 2 to 59 months also had noninferior findings for the short course; however, in older children (aged 5 to 10 years) the short duration of treatment failed to be noninferior. The authors concluded that for children aged 2 to 59 months with nonsevere CAP, a short course of antibiotic treatment is noninferior to longer treatment and can be implemented safely.⁸