Use of a multiplex PCR assay for detection of respiratory co-infections in COVID-19 hospitalized non-ICU patients: an ace up a clinician's sleeve

ABSTRACT

Background. Bacterial co-infections in COVID-19 patients represent a significant challenge for clinicians and can impact outcomes. Rapid identification of bacterial co-pathogens improves management, and is crucial to avoid inappropriately administered antimicrobials. We evaluated the use of a multiplex PCR assay in the early detection of respiratory co-infections in COVID-19 hospitalized patients. Methods. The study included non-ICU patients with high clinical suspicion of respiratory co-infection. Lower respiratory tract samples (sputum/bronchial secretions), were analyzed using BIOFIRE FILMARRAY pneumonia panel plus [(bioMerieux, USA), FA]. Specimens were considered as acceptable based on Gram stain. Conventional cultures were also performed. Results. A total of 28 samples from 27 patients (20 males, median age 60 years, IQR 49-71) were analyzed. 8/27 patients were intubated, 5 were treated with high flow nasal canula oxygenation and 11 with high or low oxygen mask. 18 patients received dexamethasone. Co-infection was detected in 17/27 patients (62.9%). 11 specimens were collected in less than 48 hours from admission and no target was identified in 6/11. Detected pathogens and AMR genes, per sample tested, are presented in the table. Bacterial pathogens and any AMR genes were detected in 15 and 4 samples, respectively, leading to modifications of antimicrobial treatment. The semi-quantitative results along with patients' clinical presentation assisted with differentiation of bacterial colonization versus infection, especially in cases where multiple targets were identified. De-escalation was implemented for 12 patients, for whom no co-infection or a viral co-pathogen (n=2) was detected. One patient was co-infected with another Coronavirus;further analysis, using the FA respiratory panel, detected Human coronavirus HKU-1 along with SARS-CoV-2. Median time until discharge or death was 13.5 days (IQR 9-24). Overall mortality was 33.3%. Detected pathogens, microbial load and AMR genes, per sample tested. Conclusion. Implementation of FA assay proved effective for the rapid detection of respiratory co-infections in COVID-19 non-ICU patients. Molecular panel-based assays can contribute to timely adaptation of antimicrobial treatment, benefiting patient management and antibiotic stewardship strategies.