Symposium Presentation Australian Society for Microbiology Annual Scientific Meeting 2024

An innovative commercial diagnostic test for sepsis directly from blood – correlation with clinical metadata and sepsis criteria (102887)

Flavia Huygens 1 , Lisa Simms 1 , Corey Davies 1 , Nadeesha Jayasundara 1 , Sumeet Sandhu 1 , Alexander Pintara 1 , Raffaella Giardino 1 , Anton Lord 2 , David Farlow 3
  1. Microbio Ltd, Brisbane, QLD, Australia
  2. Specdata Consultants, Risbane
  3. Mackay Hospital and Health Service, Mackay

Background:

Early and accurate sepsis diagnosis can save lives and billions in healthcare costs. The sepsis burden in Australia is 55,000 cases/year and 8,700 deaths/year. InfectID-BSI test detects sepsis-causing pathogens directly from blood. The test is designed to identify 26 of the most prevalent sepsis-associated pathogens in less than 3 hrs. Published literature shows under-optimal correlation between blood culture and clinical sepsis criteria.

Methods:

A 4-ml EDTA  blood tube was collected at the same time as blood culture from patients with suspected blood stream infection (BSI)/sepsis. DNA was extracted from 374 blood samples (0.35 mL) and tested using InfectID-BSI. Clinical and diagnostic laboratory data fields (n=50/patient) were collected from patient medical records. Seven objective sepsis risk criteria stratified patients as high, moderate or low risk for sepsis/septic shock (Queensland Adult Sepsis Pathway guidelines). To determine correlation of InfectID-BSI with clinical sepsis criteria, 39/202 patients were selected that returned blood culture negative and InfectID-BSI positive results.

Results:

Of the 39 patients, 18 (46.2%) were identified by clinicians as high risk of sepsis, 13 (33.3%) were moderate risk and 8 (20.5%) were low risk. Notably, of the patients who returned a positive InfectID-BSI result and a negative blood culture, 18/39 (46.2%) were infected with a fastidious bacterium.

Discussion:

InfectID-BSI uses a bioinformatics approach to identify highly discriminatory SNPs in conserved regions of the microbial genomes, resulting in InfectID-BSI’s high sensitivity and specificity. This is demonstrated by its high concordance rate with blood culture and its ability to detect additional sepsis-associated bacteria where blood culture doesn’t.

Conclusions:

InfectID-BSI outperforms traditional blood culture in true positivity rate (36.5% vs 18%), detection of mixed infections (4.3% vs 1.6%), detection of fastidious bacteria (100% vs 0%), and time to result (<3 hrs vs ~48 hrs). InfectID-BSI’s greater positivity rate compared to blood culture means that for every 100 patients with a positive blood culture, InfectID-BSI identifies an additional 105 patients with bacteria. Our results show excellent correlation between InfectID-BSI and clinical sepsis criteria which enables more precise and targeted patient management strategies.