Abstract: TH-PO0733
Effect of Polymyxin B Hemoadsorption on One-Year Mortality in Patients with Endotoxic Septic Shock: The Tigris Trial
Session Information
- AKI: Critical Care, CKRT, Cardiorenal Syndromes, and Clinical Outcomes
October 22, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- Neyra, Javier A., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Legrand, Matthieu, University of California San Francisco, San Francisco, California, United States
- Kamaluddin, Esha, Spectral Medical Inc, Toronto, Ontario, Canada
- Tomlinson, George, University of Toronto, Toronto, Ontario, Canada
- Kellum, John A., University of Pittsburgh, Pittsburgh, Pennsylvania, United States
Background
Tigris was a Bayesian, randomized trial using an informative prior taken from the EUPHRATES trial testing the effect of polymyxin B hemoadsorption (PMX) on mortality at D28 and D90 in patients with endotoxic septic shock (ESS). A preplanned analysis was to examine mortality at 12 months.
Methods
Telephone/medical record follow-up was conducted by each of the 19 enrolling sites for all patients alive at hospital discharge. As per the prespecified primary analysis, treatment effect was estimated by the odds ratio (OR) comparing PMX to control, adjusting for the baseline APACHE II in a logistic regression model. The analysis used an informative normal prior distribution on the APACHE II-adjusted log-OR for PMX based on a 25% discounted estimate from logistic regression in a sample of 179 patients from the EUPHRATES trial. We also fit a Bayesian Cox model using an informative prior on the hazard ratio (HR).
Results
The final analysis included 157 patients from Tigris. Mortality at 1 year was 56/106 (52.8%) with PMX compared to 34/51 (66.7%) with standard of care alone (absolute risk reduction of 13.9%, number needed to treat of 7). In Tigris alone, we found a 95.9% posterior probability that PMX is beneficial at 1 year compared to standard care alone (OR with 95% credible interval: 0.54 (0.26, 1.08)). Using the 75%-weighted prior, we found an OR of 0.69 (0.31, 1.13), 93.0% probability of benefit, and a marginal average risk reduction of 8.8% (-3.0%, 20.3%) yielding a number needed to treat of 11. Using a 75%-weighted prior, survival analysis through 1 year (see Figure) yielded an adjusted HR of 0.66 (0.43, 1.02) in favor of PMX, with a posterior probability of benefit of 96.3%.
Conclusion
The benefit for mortality from PMX treatment compared to standard of care for patients with ESS persisted at 1 year after randomization.
Acknowledgment
On behalf of the TIGRIS investigators
Trial Registration: NCT03901807
Figure. One year survival for patients enrolled in the Tigris Trial
Funding
- Commercial Support – Spectral Medical