Effect of placebo versus prophylactic postoperative amoxicillin on post-(adeno) tonsillectomy morbidity in Tanzanian children: a two-centre, double-blind randomized controlled non-inferiority trial

INTRODUCTION: to date in Africa, there is limited evidence regarding the role of prophylactic antibiotics to prevent post (adeno) tonsillectomy ((A)TE) morbidity in children. As (A)TE is the most performed surgery in the pediatric population, the use of prophylactic antibiotics is likely a major factor in the development of AMR. In Tanzania, as in many other settings with limited resources antibiotics are misused and overprescribed. Potential reasons include limited stewardship and widespread use of postsurgical prophylactic antibiotics. Misuse of antibiotics might contribute significantly to the development of antimicrobial resistance (AMR). METHODS: a two-centre, double-blinded randomized controlled non-inferiority trial. Subjects included children from 2-14 years of age with recurrent chronic tonsillitis and/or obstructive sleep apnoea due to adenotonsillar hypertrophy who were electively scheduled for (A)TE in two tertiary hospitals. Participants were randomly allocated to receive either placebo or amoxicillin for five days postoperatively. Primary outcome was non-inferiority of placebo compared with amoxicillin for postoperative haemorrhage (margin 5%; at 14 days) postoperative fever (margin 5%; at 7 days), and pain (margin 1 point; at 7 days). Secondary outcomes included the times required for resumption of normal diet and normal activities, and microbial recolonization of the tonsillar beds. Data were analysed according to intention-to-treat principle. Follow-up was 14 days. RESULTS: between March 13, 2019 and September 20, 2019 270 children were enrolled. All children were randomly assigned to receive placebo (n = 136) or amoxicillin (n = 134). By 14(t) (h)day post-operatively, total of 8 children were lost to follow-up in each arm. No major postoperative haemorrhage was registered. By 14(th) day post-operatively, 22 (17.5%) children in the amoxicillin arm and 19 (14.8%) children in the placebo arm had reported minor haemorrhage (risk difference (RD) -2.6% (95%CI -10.2 - 5.0); p(non-inferiority) = 0.045). By 7(th) day post operatively, 8 (6.3%) children in amoxicillin arm and 4 (3.1%) children in placebo arm reported fever during the first week (RD -3.2% (95%CI -7.6 - 1.2); p(non-inferiority) = 0.001). By 7(th) day post operatively, mean pain score (mean (SD)) was 3.25 (1.53) in the amoxicillin arm and 3.56 (1.68) in the placebo arm (mean difference 0.31, (95% CI -0.02 - 0.65); p(non-inferiority) < 0.001). No statistically significant differences between the two groups were found in any of the secondary outcomes. Findings shows, placebo is non-inferior to amoxicillin for post-operative morbidities in Tanzanian children undergoing (A)TE. CONCLUSION: it is recommended that antibiotics should only be used when clinically necessary to treat a specific infection. Unnecessary use of antibiotics contributes to the development of antimicrobial resistance. Trial Registration: Pan African Clinical Trials Registry PACTR201905466349317. Retrospectively registered on 15 May 2019.