Abstract
Septic shock driven by endotoxemia is associated with high mortality despite advances in supportive care. Polymyxin B hemoperfusion (PMX-HP) selectively removes circulating endotoxin and has shown variable efficacy in randomized trials. While earlier studies, such as EUPHAS, suggested benefit, subsequent trials, including ABDO-MIX and EUPHRATES, yielded neutral results, partly due to heterogeneous patient selection. The recently completed TIGRIS trial addressed these limitations by enrolling septic shock patients with intermediate endotoxin activity (EAA: 0.60-0.89) and high multiple organ dysfunction syndrome (MODS > 9) using a Bayesian design. In 151 evaluable patients, PMX-HP achieved the primary endpoint, with a 95.3% posterior probability of 28-day survival benefit (adjusted odds ratio [OR]: 0.67; absolute risk reduction [ARR] 6.4%). At 90 days, mortality reduction was greater (ARR: 17.4%; adjusted OR: 0.54; > 99% posterior probability of benefit), corresponding to a number needed to treat of 8.1. These results support the targeted use of PMX-HP in a biomarker-defined subgroup and may facilitate broader regulatory approval.