Few bacterial co-infections but frequent empiric antibiotic use in the early phase of hospitalized patients with COVID-19: results from a multicentre retrospective cohort study in The Netherlands

Abstract Background Knowledge on bacterial co-infections in COVID-19 is crucial to use antibiotics appropriately. Therefore, we aimed to determine the incidence of bacterial co-infections, antibiotic use and application of antimicrobial stewardship principles in hospitalized patients with COVID-19. Methods We performed a retrospective observational study in four hospitals (1 university, 2 non-university teaching, 1 non-teaching hospital) in the Netherlands from March to May 2020 including consecutive patients with PCR-confirmed COVID-19. Data on first microbiological investigations obtained at the discretion of the physician and antibiotic use in the first week of hospital admission were collected. Results Twelve (1.2%) of the 925 patients included had a documented bacterial co-infection (75.0% pneumonia) within the first week. Microbiological testing was performed in 749 (81%) patients: sputum cultures in 105 (11.4%), blood cultures in 711 (76.9%), pneumococcal urinary antigen testing in 202 (21.8%), and Legionella urinary antigen testing in 199 (21.5%) patients, with clear variation between hospitals. On presentation 556 (60.1%; range 33.3–73.4%) patients received antibiotics for a median duration of 2 days (IQR 1–4). Intravenous to oral switch was performed in 41 of 413 (9.9%) patients who received intravenous treatment >48 h. Mean adherence to the local guideline on empiric antibiotic therapy on day 1 was on average 60.3% (range 45.3%–74.7%). Conclusions On presentation to the hospital bacterial co-infections are rare, while empiric antibiotic use is abundant. This implies that in patients with COVID-19 empiric antibiotic should be withheld. This has the potential to dramatically reduce the current overuse of antibiotics in the COVID-19 pandemic.