Predictors and microbiology of respiratory and bloodstream bacterial infection in patients with COVID-19: living rapid review update and meta-regression

Abstract Objective To identify risk factors and microbiology associated with respiratory and bloodstream bacterial infection in patients with COVID-19. Methods Data Sources: We searched MEDLINE, OVID Epub and EMBASE for published literature up to February 5, 2021. Study eligibility criteria: Studies including at least 50 patients with COVID-19 in any healthcare setting. Assessment of risk of Bias: We used a validated 10-item risk of bias tool for disease prevalence Methods of data synthesis: The main outcome of interest was the proportion of COVID-19 patients with bloodstream and/or respiratory bacterial co-infection and secondary infection. We performed meta-regression to identify study population factors associated with bacterial infection including healthcare setting, age, comorbidities and COVID-19 medication. Results Out of 33,345 studies screened, 171 were included in the final analysis. Bacterial infection data were available from 171,262 patients. The prevalence of co-infection was 5.1% (95% CI: 3.6% to 7.1%) and secondary infection was 13.1% (95% CI: 9.8% to 17.2%). There was a higher odds of bacterial infection in studies with a higher proportion of patients in the intensive care unit (ICU) (adjusted OR 18.8, 95% CI: 6.5 to 54.8). Female sex was associated with a lower odds of secondary infection (adjusted OR 0.73, 95% CI: 0.55 to 0.97) but not co-infection (adjusted OR 1.05, 95% CI: 0.80 to 1.37). The most common organisms isolated included Staphylococcus aureus, coagulase-negative Staphylococci, and Klebsiella species. Conclusions While the odds of respiratory and bloodstream bacterial infection are low in patients with COVID-19, meta-regression revealed potential risk factors for infection, including ICU setting and mechanical ventilation. The risk for secondary infection is substantially greater than the risk for co-infection in patients with COVID-19. Understanding predictors of co-infection and secondary infection may help to support improved antibiotic stewardship in patients with COVID-19. Registration PROSPERO (ID CRD42021241098).