International guidelines suggest the administration of empirical broad-spectrum antibiotics for suspected bacterial co-infection in COVID-19 critically ill. However, data on associated respiratory infections is rare and antimicrobial stewardship interventions promoting antibiotic savings are non-existent in this context.

The main objectives of the trial are:

to evaluate the rate of co-infections among COVID-19 critically ill to evaluate the added value of a a rapid molecular diagnostic tool (FA-PNEU) to detect the presence of co-infecting pathogens in order to rapidly tailor the patient's antibiotic treatment

At the end of April 2020, the World Health Organisation reported that nearly three million people worldwide were infected with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) of whom 202 000 people already died from the associated disease named COVID-19. Clinical presentation varies from asymptomatic carriage to a severe respiratory infection with an acute respiratory distress syndrome and it has been calculated that approximately 5% of COVID-19 patients require intensive care. Dramatically, the reported mortality rate 28 days following admission upon the intensive care unit (ICU) is as high as 61.5%.

The 23th of March, an expert team of the Surviving Sepsis Campaign (SSC) has released a series of guidelines on the management of COVID-19 in critically ill in which they address recommendations on infection control and testing, hemodynamic control, ventilatory support and therapy. Considering the latter, the use of empiric antimicrobial/antibacterial agents is suggested among patients who have respiratory failure requiring mechanical ventilation because of suspected co-infection. Reported rates of bacterial co-infection in Influenza A patients range from 6 to 65%. It could be thought that SARS-CoV-2 destroys the respiratory epithelium in a similar way thereby facilitating bacterial invasion. SSC experts further suggest to assess for de-escalation. Rationally the adjustment to a targeted antibiotherapy requires a microbiological diagnosis of the suspected bacterial co-infection. Contemporary laboratory culture-based testing on lower respiratory tract samples requires a minimum of 24-48 hours to obtain respectively identification and antimicrobial susceptibility results. During this incompressible time interval, administration of broad-spectrum antibiotics is maintained with a risk of selecting multi-resistant bacteria potentially causing nosocomial infections. Aiming to reduce time of microbiological analysis, molecular tools have been very recently developed allowing the detection of a panel of microorganisms including bacteria and viruses directly from a lower respiratory tract sample within a few hours. Recent evaluations reported satisfying performances compared to routine microbiological methods and forecasted a substantial benefit of molecular testing in the speeded-up instauration of a targeted antibiotherapy and isolation precautions in patients with lower respiratory tract infections.

In this prospective clinical trial, the investigators performed a rapid molecular test on the initial lower respiratory sample of each COVID-19 patient admitted to ICU. The primary objective of the study was to determine the rate of co-infections in COVID-19 critically ill and the secondary objective was to measure the impact of the rapid molecular diagnostic test on the antimicrobial management of all included patients. The investigators hypothesized the rapid results of the detected pathogens will support the intensive care physician in the initiation of a targeted antibiotic treatment but also in the weaning of any antibiotherapy if all microorganisms remain undetected.

Inclusion Criteria:

• Adult patients remaining at the intensive care (SIM, SIT) of the CUSL with a confirmed COVID-19 infection

Exclusion Criteria:

• Patients from whom no respiratory sample can be obtained or
• Patients benefitting from palliative care or
• Insufficient volume of the respiratory sample after routine testing to perform the FA-PNEU test.