The effect of antibiotic therapy for Clostridioides difficile infection on mortality and other patient-relevant outcomes: a systematic review and meta-analysis.

BACKGROUND: Current practice guidelines favour fidaxomicin over vancomycin and exclude metronidazole from the recommended standard regimen for Clostridioides difficile infection (CDI), based on lower recurrence rates with fidaxomicin, giving little weight to mortality or the clinical implications of recurrences. OBJECTIVES: To compile the effects of metronidazole, glycopeptides (vancomycin or teicoplanin), and fidaxomicin for CDI on mortality and other patient-relevant outcomes. DATA SOURCES: PubMed, the Cochrane Library, ClinicalTrials.gov, conference proceedings, and Google Scholar, until August 2023. STUDY ELIGIBILITY CRITERIA: Randomized controlled trials (RCTs). PARTICIPANTS: Adult patients experiencing primary or recurrent CDI. INTERVENTIONS: Glycopeptides versus fidaxomicin or metronidazole (comparators). ASSESSMENT OF RISK OF BIAS: We used the Risk of Bias 2 (RoB 2) tool for randomized trials, focusing on the outcome of all-cause mortality. METHODS OF DATA SYNTHESIS: Random effects meta-analyses were performed for dichotomous outcomes. Outcomes were summarized preferentially for all randomly assigned patients. RESULTS: Thirteen trials were included. There was no significant difference in all-cause mortality (risk ratio [RR] < 1 favouring the comparator) between vancomycin and fidaxomicin (RR 0.86, 95% CI 0.64-1.14, 8 RCTs, 1951 patients) or metronidazole (RR 0.78, 95% CI 0.46-1.32, 4 RCTs, 808 patients), with low and very low certainty of evidence, respectively. No significant difference in initial treatment failure between fidaxomicin and vancomycin was found, however, initial treatment failure was higher with metronidazole (RR 1.58, 95% CI 1.10-2.27, 5 RCTs, 843 patients). No study reported on symptomatic recurrence necessitating re-treatment among all randomly assigned patients. Among initially cured patients, symptomatic recurrence necessitating re-treatment was lower with fidaxomicin than with vancomycin (RR 0.54, 95% CI 0.42-0.71, 6 RCTs, 1617 patients). None of the studies reported on other CDI complications or the burden of infection on daily activities. CONCLUSIONS: Setting patient-relevant outcomes for CDI independently of the RCT definitions and results might lead to less confidence in the guidance for CDI management.