Scientific Reports (Nov 2022)

Rapid pathogen identification and phenotypic antimicrobial susceptibility directly from urine specimens

  • Larry Burg,
  • Gretel Crewe,
  • James DiMeo,
  • Xin Guo,
  • Carmen G. Li,
  • Melissa Mayol,
  • Andrew Tempesta,
  • William Lauzier,
  • Rachelle Markham,
  • Katarzyna Crissy,
  • Colleen Barry,
  • Bruce Walsh,
  • James E. Kirby,
  • Don Straus

DOI
https://doi.org/10.1038/s41598-022-22792-y
Journal volume & issue
Vol. 12, no. 1
pp. 1 – 16

Abstract

Read online

Abstract Implementing effective antimicrobial therapy close to the onset of infection lowers morbidity and mortality and attenuates the spread of antimicrobial resistance. Current antimicrobial susceptibility testing (AST) methods, however, require several days to determine optimal therapies. We present technology and an automated platform that identify (ID) Urinary Tract Infection pathogens in 45 min and provide phenotypic AST results in less than 5 h from urine specimens without colony isolation. The ID and AST tests count cells fluorescently labeled with specific rRNA probes using non-magnified digital imaging. The ID test detected five pathogens at ≤ 7,000 CFU/mL and had a linear range of ~ 4 orders of magnitude. For contrived specimens, AST tests gave 93.1% categorical agreement with 1.3% Very Major Errors (VME), 0.3% Major Errors (ME), and 6.3% minor Errors (mE) compared to the broth microdilution (BMD) reference method. For clinical specimens, the ID test had 98.6% agreement and the AST test had 92.3% categorical agreement with 4.2% mE, 3.4% ME and 4.0% VME compared to BMD. Data presented demonstrates that direct-from-specimen AST tests can accurately determine antimicrobial susceptibility/resistance for each pathogen in a specimen containing two pathogens. The method is robust to urine matrix effects and off-target commensal and contaminating bacteria.