Optimising treatment outcomes for children and adults through rapid genome sequencing of sepsis pathogens (DIRECT)

We are researching a blood test which may reduce the time to detect the bacteria causing sepsis. We are investigating the use of ‘genome sequencing’ to detect the bacteria causing sepsis to decide whether we should use this new technology in the future. In addition, we would like to measure the levels of antibiotics in the blood to see if monitoring these levels improves the dosing of antibiotics that doctors give to treat sepsis. Each of these tests could lead to better treatment of infections and reduce unnecessary antibiotic use.

Information for parents and guardians of our patients

What is the proposed test?

The first test uses ‘genome sequencing’ to detect the bacteria causing sepsis. Like humans, bacteria have a DNA code which can be used to identify them. ‘Genome sequencing’ is a method to read the DNA code of the bacteria causing sepsis, identifying exactly which bacteria are present and helping doctors choose the best antibiotic treatment. Current methods to detect bacteria require the bacteria to grow in culture, and this may take days. The genome sequencing approach may provide a quicker method for bacterial identification.

We will not analyse genetic data not related to the risk of infection. The study is not designed to provide information on genetic disease in children or their families. Neither will the study be able to look at paternity.

The second test we propose to do will measure the level of antibiotics in your child’s blood. Doctors generally prescribe doses of antibiotics according to guidelines, and use the same doses for all children. We would like to see how often these doses provide effective levels in the blood and whether doses calculated for each patient by a computer could improve antibiotic levels in the blood and the treatment of sepsis.

Why is the study being done?

Sepsis is one of the most common reasons for children to present to Emergency Departments and Intensive Care Units. It can be difficult to diagnose sepsis and to identify the infection causing it. The current tests are slow, and this causes delays in providing effective treatment. Bacterial genome sequencing is a quicker method than the existing culture methods and may be superior but has not yet been trialled in a large study. This is the reason why we need to do this study.

How is the study being achieved?

Your child has had or is having routine blood samples taken as part of the management of sepsis. Some of this blood (less than 5ml) collected during routine testing has been stored. If you agree to your child’s involvement in this study, then we can use this already stored blood to evaluate the new study tests. A small blood sample (less than 2ml each) will be collected on day 1, day 2, day 3 and day 4 of antibiotic therapy. Each sample of blood (less than a teaspoon) will be collected from an existing cannula already inserted as part of standard care in the ICU. In addition, we will ask you about what led you to seek medical care. Data from your child’s medical record will also be collected.

All of the medical decisions about your child will still be based on current clinical tests.  The sequencing testing will be conducted purely for research purposes and will not affect your child’s treatment in any way. Participation in this study will not change the way your child is treated. The study results will take months to years to become available.

Information for participating study centres


Sepsis contributes significantly to morbidity and mortality of Australian children and adults with an incidence that is increasing in line with global estimates. There are approximately 5,000 deaths from sepsis in Australia each year, substantially more than from road traffic accidents. Close to 20,000 Australians develop sepsis every year. The cost to the Australian healthcare system has been estimated to exceed AUD$846 million annually, with ICU treatment required for ongoing organ dysfunction resulting in the majority of costs (https://www.georgeinstitute.org/sites/default/files/documents/stopping-sepsis-national-action-plan.pdf).

Appropriate antibiotic therapy is demonstrably effective in improving outcomes in sepsis. Each hour delay in appropriate antibiotics is associated with an estimated increase in mortality of 8%, a finding consistent in children and adults. However, currently, it often takes 48 to 72 hours before antibiotic susceptibility of pathogens are known, leading to substantial delays in optimizing antibiotic therapy, and to overuse of very broad spectrum antibiotics. The appropriateness of antibiotic therapy in sepsis is still determined by traditional susceptibility testing of cultured organisms. These conventional techniques have a yield of approximately 50% and often take days to complete. In an era of emerging antibiotic resistance, this timeframe for the identification of a causative organism and its susceptibility profile may result in a prolonged period of inappropriate empirical treatment. In adults, it has been observed that Gram-negative healthcare-associated infections are associated with significant delays in effective antibiotic therapy. Such delays prolong the requirement for intensive care and lead to increased morbidity, mortality and cost. A day in ICU costs approximatively $6,000 for ICU infrastructure alone. Hence strategies with the potential to improve patient recovery and shorten ICU length of stay have great promise to be cost-effective. A rapid, sensitive technique capable of identifying the pathogen responsible for sepsis, along with its susceptibility profile has the potential to transform healthcare and substantially improve clinical outcomes. This can be further enhanced by optimising antibiotic concentrations through individualized therapy.

We propose an integrated approach to personalize antimicrobial therapy in children and adults with sepsis. We will aim to improve effective antibiotic therapy in adult and paediatric patients with sepsis by incorporating the following innovations into routine clinical care in ICU:

  • Real-time, portable nanopore sequencing using the Oxford Nanopore Techologies MinION device coupled with streaming bioinformatics tools to identify causative bacterial pathogens and predict antibiotic susceptibility direct from blood
  • The application of dosing software to ensure achievement of maximally effective personalised antibiotic exposures.

This approach uses cutting edge nanopore sequencing technology offering the potential to impact clinical decision making through real-time sequencing. It advances ongoing work supported by QGHA to track healthcare-associated infection using cultured isolates, and promises a culture-independent method of identification, which shifts the potential of sequencing to the bedside. It also builds on existing program quantifying the clinical value of dosing software for more accurate therapy in critically ill patients. The proposed program represents a novel approach, not currently used in Queensland, nor elsewhere in the world, and directly applicable to patients with sepsis.


Real-time pathogen sequencing combined with personalised antimicrobial therapy will reduce the time to optimal antimicrobial concentrations in patients with sepsis leading to improved patient outcomes.

Aim of the Study

To demonstrate the feasibility and diagnostic accuracy of real-time pathogen sequencing from clinical samples, and estimate the impact of this diagnostic approach when integrated with personalised antimicrobial dosing in children and adults on ICU with sepsis.


This is a multi-centre prospective study of real-time pathogen sequencing in patients with sepsis admitted to intensive care units in Brisbane, Queensland incorporating:

  1. DIRECT:
    • A diagnostic accuracy study of real-time pathogen sequencing from patient samples using the Oxford Nanopore Technologies MinION sequencer, which will be compared to conventional pathogen sequencing, and conventional blood culture testing.
    • A diagnostic accuracy study of rapid pathogen identification and anti-microbial resistance detection methods (such as T2 Biosystems) from residual patient samples salvaged from the Nanopore sequencing study.
  2. DIRECT 2:
    • An evaluation of the impact of an integrated diagnostic and therapeutic algorithm combining rapid direct pathogen sequencing and software-guided, personalised antibiotic dosing

Sample size:

In this feasibility study, the target sample size is 100 patients, including 30 children from the Queensland Children’s Hospital PICU.

Participating sites

  • ICU at Royal Brisbane and Women’s Hospital, Queensland
  • ICU at The Prince Charles Hospital, Queensland
  • ICU at Princess Alexandra Hospital, Queensland
  • PICU at Queensland Children’s Hospital, Queensland

Participant Population:

Patients admitted to participating ICUs with suspected sepsis will be screened at admission or at the time of suspicion of sepsis for inclusion in the study.

Inclusion Criteria:

  • Age >1month
  • Admitted to paediatric or adult ICU at one of the participating centres
  • Suspected sepsis, defined as suspected or proven infection with organ dysfunction. Organ dysfunction in children and adults is defined in Appendix 1.
  • Recently commenced (within 24h) on intravenous broad-spectrum antibiotics, including a change to new antibiotics for a suspected new episode of sepsis
  • Blood cultures are being obtained or were obtained within the past 12 hours

Exclusion Criteria:

  • Inability to gain informed consent during the study period
  • Neonates
  • Death is likely imminent
  • Palliative care patient


To provide evidence of the feasibility of the real-time sequencing detection and other emerging rapid diagnostic approaches for sepsis pathogens.


The DIRECT study is funded by the Queensland Genomics Health Alliance Round 2 clinical implementation, innovation and incubation program. The DIRECT 2 sub-study evaluating the impact of software-guided antimicrobial dosing is funded by a Brisbane Diamantina Health Partners Health system Improvement Ideas Grant (MRFF Rapid Applied Research Translation Program).


Centre for Children’s Health Research
Queensland Children’s Hospital Precinct
Paediatric Critical Care Research Group
Level 6, 62 Graham Street
South Brisbane, Qld, 4101

Central team (CCHR – Brisbane)

For study queries contact

Study Coordinator

Natalie Sharp
e: Natalie.sharp2@health.qld.gov.au
t: 07 3069 7318

Chief Investigator

Dr Adam Irwin
e: a.irwin@uq.edu.au