Evaluation of a Direct-from-Blood Molecular Diagnostic Platform for Pathogen Detection in Critically Ill Patients: A Prospective Observational Cohort Study

Le résumé fourni par la source

Rationale: Defining the microbial etiology of sepsis is difficult but has important implications for antibiotic stewardship and prognosis. We explore the diagnostic yield of a novel, direct-from-blood multiplex polymerase chain reaction in intensive care unit patients with suspected sepsis. Methods: From March 2022 to March 2024, patients were included if the clinical team ordered a blood culture. In study participants, an additional 6 mL of whole blood was collected by the phlebotomist. Samples were analyzed by the RaPID/BSI platform, a polymerase chain reaction-based assay that detects 19 bacterial and fungal species direct from whole blood. Results: A total of 165 samples were included in the final cohort. Of these, 27 patients had a positive RaPID/BSI result and 8 patients had positive blood cultures with 1 pathogen not included in the RaPID/BSI panel. Of the 27 patients with positive RaPID/BSI results, 5 patients also had positive blood cultures. Pathogen identification from RaPID/BSI and blood cultures were concordant for 4 of 5 of these pathogens. Twenty-five of the 27 RaPID/BSI positive patients had clinically defined infections. Eighteen patients of the 27 RaPID/BSI-positive patients had additional positive microbial culture data that corroborated pathogen identification. Of the 27 patients with bloodstream infections detected by RaPID/BSI and/or blood cultures, 20 had received antibiotics in the 48 hours before their blood draw. The RaPID/BSI assay identified the pathogen in all 20 of these patients, whereas only 4 of these 20 patients had positive blood cultures. Conclusions: The diagnostic yield of a direct-from-blood molecular diagnostic assay for the detection of bloodstream infection was 15.2% compared to 4.3% from blood cultures, providing higher detection rate especially in the setting of recent antibiotic administration.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.