Evaluation of the LifeScale system for rapid phenotypic antimicrobial susceptibility testing from positive blood cultures

Abstract

Abstract Objective The spread of antimicrobial resistance is a growing threat to public health, thus there is a critical need for prompt and reliable antimicrobial susceptibility testing (AST). Rapid AST may improve patient outcomes by allowing for quicker start of optimal therapy and reducing hospital length of stay, therefore decreasing overall healthcare costs. However, rapid results may not improve patient care if they are not considered equivalent to standard AST. Several platforms are under development that can provide rapid AST results; the objective of this work is to assess the performance of the new LifeScale rapid AST platform (Affinity Biosensors) that uses resonant mass measurements for calculation of antimicrobial susceptibility results. Methods Positive BACTEC blood-cultures collected from patients were included if the gram stain revealed only Gram-negative rods. Standard of care (SOC) AST results, as determined by a combination of BD Phoenix automated, Sensititre broth microdilution, and/or standard disk diffusion methods, were compared to results from direct positive blood culture testing on the LifeScale system. Minimum inhibitory concentrations (MICs) and categorical interpretations based on CLSI and/or FDA AST breakpoints were compared by microorganism and drug to calculate essential and categorical agreement for five species with pending claims on the LifeScale system: Escherichia coli (N=33), Klebsiella pneumoniae (N=10), K. oxytoca (N=5), Pseudomonas aeruginosa (N=11) and Acinetobacter sp. (N=1); tested against 13 relevant antibiotics: Amikacin, ampicillin, aztreonam, cefazolin, cefepime, ceftazidime, ceftazidime/avibactam ertapenem, gentamicin, levofloxacin, meropenem, piperacillin/tazobactam, and trimethoprim/sulfamethoxazole. Results LifeScale had 95.3% essential agreement with the SOC results, and a 93.1% categorical agreement based on CLSI M52 criteria. Of the 665 organism-drug combinations evaluated, agreement was high, with only 1 very major discrepancy, 8 major discrepancies and 37 minor discrepancies, of which 22 (47.8%) were within ±1 doubling dilution. Among these discrepancies, LifeScale MICs averaged 1 dilution higher than SOC MICs, which may reduce the risk of false-susceptible interpretations. Conclusion s: The accuracy and reliability of results from rapid AST is of utmost importance. The LifeScale technology is unique among AST platforms as it measures bacterial cell mass and counts to determine MICs for up to 14 antibiotics in ~4 h. These results suggest this platform is able to deliver rapid, actionable results to guide antimicrobial therapy for patients with bacteremia.