Duration of amoxicillin-clavulanate for protracted bacterial bronchitis in children (DACS): a multi-centre, double blind, randomised controlled trial

Background Protracted bacterial bronchitis (PBB) is a leading cause of chronic wet cough in children. The current standard treatment in European and American guidelines is 2 weeks of antibiotics, but the optimal duration of therapy is unknown. We describe the first randomised controlled trial to assess the duration of antibiotic treatment in children with chronic wet cough and suspected PBB. We hypothesise that 4 weeks of amoxicillin–clavulanate is superior to 2 weeks for improving clinical outcomes. Methods Our parallel, double-blind, placebo-controlled, randomised controlled trial was completed in four Australian hospitals. Children aged 2 months to 19 years with chronic (>4 weeks duration) wet cough, and suspected PBB were randomly assigned (1:1) using permuted block randomisation (stratified by age and site) to 4 weeks of amoxicillin–clavulanate (25–35 mg/kg twice daily oral suspension; 4-week group) or 2 weeks of amoxicillin–clavulanate followed by 2 weeks of placebo (2-week group). The children, caregivers, all the study coordinators, and investigators were masked to treatment assignment until data analysis was completed. The primary outcome was clinical cure (cough resolution) by day 28. Secondary outcomes were recurrence of PBB at 6 months, time to next exacerbation, change in Parent-proxy Cough-Specific Quality-of-Life (PC-QoL) score from baseline to day 28 and from day 28 to 7 months, adverse events, nasal swab bacteriology, and antimicrobial resistance. Analyses followed the intention-to-treat principle. This trial is complete and registered with Australian/New Zealand Registry, ACTRN12616001725459. Findings Between March 8, 2017, and Sept 30, 2019, 106 children were randomly assigned (52 in the 4-week group, median age 2·2 years [IQR 1·3–4·1]; 54 in the 2-week group, median age 1·7 years [1·2–3·8]) with 90 children completing the 4-week treatment. By day 28, the primary endpoint of clinical cure in the 4-week group (32 [62%] of 52 patients) was not significantly different to the 2-week group (38 [70%] of 54 patients; adjusted relative risk 0·87 [95% CI 0·60 to 1·28]; p=0·49). Time to next wet cough exacerbation was significantly longer in the 4-week group than the 2-week group (median 150 days [IQR 38–181] vs 36 days [15–181]; adjusted hazard ratio 0·47 [0·25 to 0·90]; p=0·02). The rate of recurrence of PBB at 6 months was 17 (53%) of 32 patients in the 4-week group vs 28 (74%) of 38 patients in the 2-week group, but the difference between the groups was not significant (adjusted odds ratio 0·39 [0·14 to 1·04]; p=0·07). PC-QoL significantly improved from baseline to day 28 in both groups, but there was no significant difference between them (mean difference in change -0·2 [95% CI -1·0 to 0·6]; p=0·64). From day 28 to 7 months, median PC-QoL remained stable in both groups with no difference in change between them. Data on respiratory pathogens and antimicrobial resistance (paired swabs available for 48 children) were similar between groups. Adverse events occurred in 13 (25%) children in the 2-week group and ten (19%) in the 4-week group (p=0·57). Interpretation A 4-week course of amoxicillin–clavulanate for treating children with chronic wet cough and suspected PBB confers little advantage compared with a 2-week course in achieving clinical cure by 28 days. However, as a 4-week duration led to a longer cough-free period, identifying children who would benefit from a longer antibiotic course is a priority.