RESUMEN
INTRODUCTION: Understanding the proportion of patients with COVID-19 who have respiratory bacterial co-infections and the responsible pathogens is important for managing COVID-19 effectively while ensuring responsible antibiotic use. OBJECTIVE: To estimate the frequency of bacterial co-infection in COVID-19 hospitalized patients and of antibiotic prescribing during the early pandemic period and to appraise the use of antibiotic stewardship criteria. METHODS: Systematic review and meta-analysis was performed using major databases up to May 5, 2021. We included studies that reported proportion/prevalence of bacterial co-infection in hospitalized COVID-19 patients and use of antibiotics. Where available, data on duration and type of antibiotics, adverse events, and any information about antibiotic stewardship policies were also collected. RESULTS: We retrieved 6,798 studies and included 85 studies with data from more than 30,000 patients. The overall prevalence of bacterial co-infection was 11% (95% CI 8% to 16%; 70 studies). When only confirmed bacterial co-infections were included the prevalence was 4% (95% CI 3% to 6%; 20 studies). Overall antibiotic use was 60% (95% CI 52% to 68%; 52 studies). Empirical antibiotic use rate was 62% (95% CI 55% to 69%; 11 studies). Few studies described criteria for stopping antibiotics. CONCLUSION: There is currently insufficient evidence to support widespread empirical use of antibiotics in most hospitalised patients with COVID-19, as the overall proportion of bacterial co-infection is low. Furthermore, as the use of antibiotics during the study period appears to have been largely empirical, clinical guidelines to promote and support more targeted administration of antibiotics in patients admitted to hospital with COVID-19 are required.