Efficacy of Antibiotic and Non-antibiotic Interventions in Preventing and Treating Necrotic Enteritis in Broiler Chickens: A Systematic Review

The objective of this systematic review was to compare the efficacy of antibiotic and non-antibiotic alternatives in the prevention and treatment of necrotic enteritis (NE) in broiler chickens. In vivo experimental and observational studies that compared the administration of non-antibiotic compounds with antibiotics to prevent or treat NE in broiler chickens and that evaluated mortality and/or clinical or subclinical NE outcome measures were eligible. Four electronic databases were searched in December 2019 and updated in October 2021. Retrieved studies were evaluated in two phases: abstract and design screening. Data were then extracted from included studies. Risk of bias was assessed by outcome following the Cochrane Risk of Bias 2.0 tool. A meta-analysis was not conducted due to heterogeneity across interventions and outcomes. The non-antibiotic and antibiotic groups were compared at the outcome level for individual studies using the mean difference and 95% confidence interval (CI) calculated post hoc from raw data. In total, 1282 studies were originally identified, and 40 were included in the final review. The overall risk of bias for the 89 outcomes was either "high'' (n = 34) or ``some concerns'' (n = 55). Individual study comparisons showed a beneficial trend toward the antibiotic group for reduced mortality, NE lesion scores (overall, jejunum, and ileum), Clostridium perfringens counts, and for most histologic measurements (duodenum, jejunum, and ileum villi height, and jejunum and ileum crypt depth). The non-antibiotic groups showed a beneficial trend for NE duodenum lesion scores and duodenum crypt depth measurements. Based on this review, there is a trend that mostly favors antibiotic compounds in preventing and/or treating NE, but the evidence also suggests no difference when comparing them with non-antibiotic alternatives. Studies assessing this research question were heterogeneous in their intervention conditions and outcomes measured, and there were key aspects of the experimental design not reported in some of the studies.