Sepsis Trials In Critical Care (SepTIC)

Research questions:

1. Do rapid PCR-based microbiological diagnostics combined with procalcitonin improve outcomes and antibiotic stewardship compared to standard care in patients admitted to intensive care (ICU) with sepsis?

2. Does conservative fluid therapy with active removal of accumulated fluid (de-resuscitation) improve outcomes compared to standard care in patients admitted to ICU with sepsis? 3. Does GM-CSF compared to placebo improve outcomes in a high-risk subset of patients admitted to ICU with sepsis? 4. What is the relative cost-effectiveness of each of these interventions compared to current standard of care?

Design: A 2x2 factorial pragmatic open-label randomised controlled trial, with an embedded randomised double-blind parallel group trial Setting: 60 general ICUs in the UK Target population: Adult patients who are critically ill due to sepsis Inclusion criteria - Adult patients admitted to ICU due to suspected sepsis and expected to stay for at least 48 hours

- Receiving intravenous antibiotics for suspected sepsis - Initial early fluid resuscitation complete - Acute organ dysfunction due to infection Additional criteria for GM-CSF subset - Mechanically ventilated - Receiving vasopressors/inotropes - Absolute lymphocyte count <1.2 x109/L on two consecutive days Exclusion criteria - More than 12hours since ICU admission - Previously admitted to ICU due to sepsis on this hospital admission - Not expected to survive 90 days, due to pre-existing chronic disease - Not expected to survive initial resuscitation (24 hours) - Neutropenia due to chemotherapy/malignancy (but not due to sepsis) - A source of infection that will require a prolonged course of antibiotics, >21 days - Additional exclusion criteria for GM-CSF

Health Technologies being assessed 1. Rapid PCR-based microbiological diagnostics combined with procalcitonin compared with standard care 2. Conservative fluid therapy with active removal of accumulated fluid (de-resuscitation) compared to standard care 3. GM-CSF (sargramostim) 6µg/kg/day for 7 days compared to placebo

Measurement of costs & outcomes Primary: 90-day all-cause mortality (for all interventions) Secondary: - Duration of mechanical ventilation, shock, renal replacement therapy - Length of stay in ICU and hospital - Antibiotic use (defined daily doses per 1000 occupied bed days) - Infection relapse / recurrence or secondary infection requiring further antibiotic treatment

- Adverse events and drug reactions (including antibiotic related adverse events) - Health-related Quality of Life (EQ-5D-5L) and cognitive function (MoCA-Blind) at 6 months - 1-year mortality

Sample size: 3758 patients in total, allowing for 10% inflation due to the factorial design and 2% loss to follow-up. 1300 patients are required for the high-risk GM-CSF subset.

Timeline: After 9 months set-up time, 3-years are required for recruitment, based on 2 patients recruited per ICU per month. Allowing for 1-year follow up and 3 months final analysis and reporting, this totals 60 months for the project.

Expertise: Our team includes critical care, microbiology and emergency medicine clinicians, trialists, statisticians, health economists, a psychologist and a patient representative. Our team has expertise in sepsis management, antimicrobial resistance, diagnostics, ICU trial design/methodology/analysis, health economics, communication and strong links with patient groups.