A Reproducible Swine Model of Endotoxemia-Induced Septic Shock for Critical Care Investigations

Sunday, March 22, 2026

2:30 PM - 3:30 PM Central Time

Location: Connections Central - RST 12

Rachel L. Choron, MD, FACS

Rutgers Robert Wood Johnson Medical School, New Jersey

Disclosure information not submitted.
DK

Divya Kewalramani, MD

Assistant Professor of Research
Rutgers Robert Wood Johnson Medical School, New Jersey

Disclosure information not submitted.
SS

Shareef Syed, BA

Medical Student
Rutgers Robert Wood Johnson Medical School

Disclosure information not submitted.
CL

Charles Levin, MD

AnesthesiaResident
Columbia University Irving Medical Center

Disclosure information not submitted.
JG

Julie Goswami, MD

Assistant Professor of Surgery
Rutgers Robert Wood Johnson Medical School

Disclosure information not submitted.
BP

Barbara Perry, MBA

Director SimLab & Ortho Research & Education Laboratory
Rutgers Robert Wood Johnson Medical School

Disclosure information not submitted.
AZ

Aleksandra Zawieska, PhD

Director SimLab & Ortho Research & Education Laboratory
Rutgers Robert Wood Johnson Medical School

Disclosure information not submitted.
MN

Mayur Narayan, MBA, MD, MPH, FCCM, FACS, MHPE, FICS

Professor of Surgery
New York-Presbyterian Hospital/Weill Cornell Medical Center

Disclosure information not submitted.

Introduction: Preclinical septic shock models are vital for translational evaluation of emerging therapies, yet current swine models vary in shock definitions, sampling, and hemodynamic thresholds—limiting reproducibility. We developed a rigorously controlled swine endotoxemia-induced model with four distinct phases reflecting septic shock: baseline, endotoxemia induction, shock with vasopressor initiation, and vasopressor steady state.

Methods: Four Yorkshire pigs (22±3 kg) were anesthetized, intubated, and monitored via arterial, central venous, and pulmonary artery catheters. Continuous mean arterial pressure (MAP), systolic/diastolic pressure (SBP/DBP), central venous (CVP), heart rate (HR), and ECG—were recorded at 30-second intervals. Baseline was defined as MAP ≥65 mmHg for 20 minutes. Endotoxemia was induced with peripheral intravenous lipopolysaccharide (E.coli O111:B4) infusion at 2.5 µg/kg/hr, titrated up every 6 minutes until SBP dropped ≥20% and MAP ≤65mmHg. A 30cc/kg crystalloid bolus was given and norepinephrine (0.01–0.30 µg/kg/min) initiated to restore MAP ≥65 mmHg. Steady state was defined after 15 minutes of stable vasopressor dosing and maintained for 2.5 hours. ABGs and lactate were sampled serially. Data were cleaned and aggregated into 10-minute means for analysis using linear mixed-effects models.

Results: Endotoxemia developed over 23.5±7.4 minutes. MAP fell from baseline 81.2±5.8 to 50±2.5 mmHg (Δ-38%) and SBP from 97.8±4.4 to 62.8±2.5 mmHg (Δ–36%) with shock onset. After norepinephrine, MAP and SBP improved to 61±4 and 82±6.5 mmHg at steady state. HR rose from baseline 107±9 bpm to 116±10 bpm during endotoxemia and peaked at 147±8.1 bpm after 2.5 hours. Lactate increased from baseline 2.9±1.2 to 3.8±2.5 mmol/L after endotoxemia and reached 4.5±1.4 mmol/L by 2.5 hours. Across all variables, inter-pig SEM remained ≤6% of the mean, confirming low variability supporting reproducibility of the model. Total endotoxin dosed averaged 7.4± 5.9 µg/kg with pigs receiving 54±18 µg/kg of norepinephrine in aggregate.

Conclusions: This novel four-phase swine model provides a reproducible, low-variance septic shock model using vasopressor therapy. It offers a robust control condition to test adjunctive interventions in septic shock and serves as a translational reference for preclinical research.