GRADY: a computer tool to detect early sepsis and gram-negative bloodstream infection in ICU patients
Prospective Validation of the GRADY Bacteremia/Sepsis Prediction Model in Intensive Care Unit Patients: Clinical Performance and Feasibility as an Early Warning System
The GRADY tool will be tested to see if routine vital signs and lab results can detect early sepsis and gram-negative bloodstream infection in adult ICU patients.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 55 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Sisli Hamidiye Etfal Training and Research Hospital Academic / other |
| Locations | 1 site (Seyrantepe, Istanbul) |
| Trial ID | NCT07126106 on ClinicalTrials.gov |
What this trial studies
This is a prospective, single-center validation of the GRADY machine-learning models using routinely collected vital signs and laboratory data from adult ICU patients. The study will enroll patients who remain in the ICU at least 48 hours and have blood cultures taken, and will compare GRADY's predictions to standard scoring systems such as SOFA, SIRS, and NEWS2. Primary outcomes include detection accuracy for gram-negative bacteremia and sepsis versus blood culture-confirmed infection and clinical sepsis definitions. The protocol also explores how GRADY might be integrated as an early warning tool to support more rapid clinical intervention.
Who should consider this trial
Good fit: Adults aged 18 or older who are in the ICU for 48 hours or more, have had blood cultures obtained as part of routine care, and can provide informed consent are the intended participants.
Not a fit: Patients younger than 18, those with ICU stays under 48 hours, or patients who never had blood cultures collected are unlikely to benefit from this study's findings.
Why it matters
Potential benefit: If successful, GRADY could enable earlier identification of patients at risk for gram-negative bacteremia or sepsis, allowing faster targeted treatment and potentially reducing ICU morbidity and mortality.
How similar studies have performed: Previous retrospective studies of machine-learning sepsis prediction tools have shown promise, but prospective external validation of such models remains limited.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Patients aged 18 years or older * ICU stay of 48 hours or longer * Patients from whom blood cultures were obtained during routine monitoring * Signed informed consent form Exclusion Criteria: * Patients younger than 18 years * ICU stay shorter than 48 hours * Patients without blood cultures
Where this trial is running
Seyrantepe, Istanbul
- Sisli etfal research and training hospital — Seyrantepe, Istanbul, Turkey (Türkiye) (Recruiting)
Study contacts
- Study coordinator: okan derin
- Email: okanderin@gmail.com
- Phone: +905053580264
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.