2025 Critical Care Congress Skip to main content
2025 Critical Care Congress Main banner
Final Program


Back

Precision Medicine in Sepsis

Star Research Presentations: Pharmacology, Precision Medicine in Sepsis and Patient Safety

Optimizing Sepsis Time Zero Definitions Based on Associations of Time-to-Antibiotics With Mortality

Sunday, February 23, 2025
2:15pm – 3:45pm Eastern Time
Location: Connections Central Theater 1

  • Chanu Rhee, MD, MPH (he/him/his)

    Harvard Medical School

    Disclosure(s): No relevant financial relationship(s) to disclose.

  • TP

    Theodore Pak, MD, PhD

    Research Fellow
    Department of Population Medicine, Harvard Pilgrim Healthcare Institute / Division of Infectious Diseases, Department of Medicine, Massachusetts General Hospital

    Disclosure information not submitted.

  • JY

    Jessica Young, PhD

    Associate Professor
    Harvard Pilgrim Health Care Institute

    Disclosure information not submitted.

  • CM

    Caroline McKenna, N/A

    Analyst
    Harvard Pilgrim Health Care Institute

    Disclosure information not submitted.

  • AA

    Anna Agan, N/A

    Analyst
    Harvard Pilgrim Health Care Institute

    Disclosure information not submitted.

  • LD

    Laura DelloStritto, N/A

    Project Manager
    Harvard Pilgrim Health Care Institute

    Disclosure information not submitted.

  • Edward J. Septimus, MD

    Senior Lecturer
    Harvard Pilgrim Health Care Institute

    Disclosure information not submitted.

  • CS

    Claire Shappell, MD, MPH

    Instructor
    Division of Pulmonary and Critical Care Medicine, Department of Medicine, Brigham and Women's Hospital

    Disclosure information not submitted.

  • KS

    Kenneth Sands, MD, MPH

    Chief Epidemiologist
    HCA Healthcare

    Disclosure information not submitted.

  • JG

    Jeffrey Guy, MD, MSc

    Vice President
    HCA Healthcare

    Disclosure information not submitted.

  • RP

    Russell Poland, PhD

    AVP for Research
    HCA Healthcare

    Disclosure information not submitted.

  • MF

    Michael Filbin, MD, MSc

    Associate Professor
    n/a

    Disclosure information not submitted.

  • SD

    Sayon Dutta, MD, MPH

    Assistant Professor
    Massachusetts General Hospital / Harvard Medical School

    Disclosure information not submitted.

  • Sameer Kadri, MD, MS

    National Institutes of Health Clinical Center

    Disclosure information not submitted.

  • Henry Masur, MD

    Chief, Critical Care Medicine Department
    National Institutes of Health Clinical Center

    Disclosure information not submitted.

  • MK

    Michael Klompas, MD, MPH

    Professor
    Department of Population Medicine, Harvard Pilgrim Healthcare Institute / Division of Infectious Diseases, Department of Medicine, Brigham and Women's Hospital

    Disclosure information not submitted.

Disclosure(s):
Chanu Rhee, MD, MPH: No relevant financial relationship(s) to disclose.
Introduction: The CMS SEP-1 sepsis definition includes heterogeneous organ dysfunction criteria that likely differ in their need for urgent treatment. We evaluated which criteria identify patients that benefit from immediate antibiotics to inform more evidence-based time zero definitions and quality metrics.


Methods: We retrospectively identified all adults admitted to 145 hospitals with suspected infection (intravenous antibiotics and blood cultures within 24h of arrival) from Sept 2018-Dec 2023. We evaluated associations between time-to-antibiotics (1-3 vs 0-1h and 3-6 vs 0-3h) and in-hospital mortality relative to time zero definitions anchored on the onset of each SEP-1 organ dysfunction (hypotension [SBP < 90 or MAP < 65mmHg]; lactate ≥4.0 or 2-4mmol/L; respiratory failure [non-invasive or invasive ventilation], creatinine ≥2.0mg/dL, bilirubin ≥2.0mg/dL, platelets < 100,000/uL, or INR >1.5/PTT >60s), adjusting for baseline and time-varying confounders using inverse probability weighting.


Results: Among 2.3 million patients with suspected infection, 640,117 met ≥1 organ dysfunction criteria and received antibiotics between 0-6h of time zero. Organ dysfunction rates at time zero ranged from 6.6% (low platelets) to 40.2% (lactate 2-4); crude mortality ranged from 5.9% (lactate 2-4) to 21.4% (respiratory failure). Adjusted mortality risk was higher for antibiotic administration between 1-3 vs 0-1h for definitions anchored to hypotension (RR 1.07, 95% CI 1.03-1.10) and lactate ≥4 (RR 1.10, 95% CI 1.06-1.14) but not for other organ dysfunctions. When assessing antibiotic administration at 3-6 vs 0-3h, mortality risk was also elevated for definitions anchored by lactate 2-4, respiratory failure, and elevated INR/PTT.


Conclusions: In this cohort of over 640,000 patients with suspected sepsis, mortality risk was only increased with antibiotic delays >1h relative to the onset of hypotension or lactate ≥4mmol/L, and additionally increased with antibiotic delays >3h for respiratory failure, coagulopathy, or lactate >2. Modifying sepsis time zero definitions and quality metrics to emphasize these organ dysfunctions could help clinicians and regulators better target patients needing immediate antibiotics, simplify measurement, and reduce unnecessary pressure to rapidly treat patients with less severe presentations.