Accelerating Time to Pathogen-adapted Antibiotic Treatment through Culture-independent Antimicrobial Susceptibility Testing in Patients Suffering from Sepsis

Abstract Background Accurate and fast pathogen identification and consecutive antimicrobial susceptibility testing (AST) is of vital importance for patient outcome in patients suffering from sepsis. Methods The Accelerate Pheno™ system is a new, fully automated, culture-independent diagnostic method for both pathogen identification (ID) and antimicrobial susceptibility testing (AST). We analyzed positive blood cultures from critically ill patients with new onset of sepsis according to the new sepsis guidelines, using both conventional standard methods (VITEK, MALDI-TOF) and Accelerate Pheno™ system. ID/AST results of the Accelerate Pheno™ system were not reported to treating physicians as part of our internal evaluation process. Results Accelerate Pheno™ system correctly detected 74 pathogens [Gram-negative (GN) (n = 27), Gram-positive (GP) (n = 47)] straight out of 84 positive blood culture bottles. Gram-negative (GN) pathogens were identified as E. coli (n = 15; concordance rate 100%), K. pneumonia (n = 7; 71,4 %), S. marcescens (n = 3; 100%), E. cloacae (n = 2; 50%), P. mirabilis (n = 1; 100%) and P. aeruginosa (n = 1; 33%). Gram-positive pathogens were identified as CNS (n = 24; 82,6%), S. aureus (n = 15; 88,2%), E. faecium (n = 6; 100%) and E. faecalis (n = 2; 100%). The Accelerate Pheno™ system generated a GN-AST result in 70,4 % (19 of 27 samples) and a GP-AST result in 61,7 % (29 of 47 samples) when compared with routine AST. Growth control, analysis and mechanical failure led to reduced results in comparison to conventional ID/AST. Accelerate PhenoTM delivered correct MIC-results for most of the panel antibiotics [e.g., meropenem: 83,3%, gentamicin: 88,9%, ertapenem: 100%]. Conclusion The use of the Accelerate Pheno™ system significantly improved time-to-ID/AST and would have led to a reduced time-to-treatment in patients suffering from sepsis if results would have been reported. The system currently has some weakness in the detection of polymicrobial and streptococcal infections but due to the short hands-on-time, culture-independence and fast generation of results, it represents a promising new diagnostic method for the consecutive antibiotic treatment of septic patients. Disclosures All authors: No reported disclosures.