Background: Metronidazole is the preferred empirical anti-anaerobic agent for patients with suspected anaerobic infection. The balance between benefits and harms of empirical metronidazole is unclear. We aimed to assess patient-important benefits and harms of empirical metronidazole vs placebo/no treatment in adult patients with severe bacterial infection of any origin. Methods: We conducted a systematic review with meta-analysis and trial sequential analysis of randomized clinical trials assessing empirical metronidazole vs placebo/no treatment in adult hospitalized patients with severe bacterial infection. The review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement, the Cochrane Handbook and the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology. A protocol and statistical analysis plan was published prior to conducting the review. Results: We included a total of nine trials (n = 1753 patients), all of which were adjudicated as having high risk of bias. We found no difference in the primary outcome mortality within 90 days (relative risk 1.56, 95% confidence interval 0.39-6.25). Fewer patients receiving metronidazole had secondary infections (relative risk 0.43, 95% CI: 0.27-0.68). Trial sequential analysis indicated high risk of random errors due to lack of data, and the quality of evidence was very low for all outcomes. Conclusions: There is low quantity and quality of evidence supporting the use of empirical metronidazole in adult patients with severe bacterial infections of any origin, and no firm evidence for benefit or harm.