Extracorporeal Cytokine Hemadsorption with oXiris® in Critically Ill Patients with Non-Septic Vasoplegic Shock: Hemodynamic Effects, Cytokine Kinetics, and Mortality Outcomes

Background: Vasoplegic shock (VS) in critically ill patients without microbiological evidence of infection poses a major clinical challenge in intensive care units (ICUs). Extracorporeal cytokine hemadsorption using the oXiris® membrane—a high-permeability polyacrylonitrile-based (AN69-ST) filter with adsorptive properties against inflammatory mediators—has emerged as a potential adjunct to restore haemodynamic stability. Evidence supporting its use remains limited, particularly regarding timing of initiation and downstream mortality biomarkers. Methods: We conducted a single-centre prospective observational study at the ICU of Istanbul Prof. Dr. Cemil Taşcıoğlu City Hospital between October 2022 and January 2023. Adults aged ≥18 years with VS (CRP ≥ 100 mg/L, procalcitonin [PCT] < 2 μg/L, no positive microbiological culture) requiring continuous renal replacement therapy (CRRT) with the oXiris® membrane were enrolled (n = 34), of whom 30 completed the study period without microbiological exclusion and comprised the final analysis cohort. Pre- and post-treatment (72 h) clinical and cytokine parameters were compared. The association of VS resolution and 7-day mortality with timing of oXiris® initiation, cytokine levels, and treatment duration was assessed. This study was conducted and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement. Results: Significant changes were observed across all principal haemodynamic and inflammatory parameters at 72 h of oXiris® treatment, including mean arterial pressure (MAP: 50.8 ± 6.3 to 69.3 ± 17 mmHg, p < 0.001), SOFA score (8.33 ± 2.29 to 4.9 ± 3.22, p < 0.001), IL-6 (767.3 ± 1205.7 to 294.4 ± 686.3 pg/mL, p < 0.001), IL-1β, TNF-α, CRP, lactate, and creatinine. VS resolved in 24/30 patients (80%). Younger age was associated with VS resolution (57.8 ± 19.7 vs. 76.6 ± 13.9 years; p = 0.021). Initiation of oXiris® within 8 h was associated with significantly shorter VS resolution time (52.5 ± 23.9 vs. 85.9 ± 48.2 h; p = 0.045). Seven-day mortality was 20% (n = 6) and hospital mortality was 50% (n = 15). Post-treatment IL-1β (856.7 ± 548.5 vs. 1086.9 ± 353.6 pg/mL; p = 0.044) and TNF-α (111.0 ± 70.0 vs. 145.4 ± 47.8 pg/mL; p = 0.011) at 72 h were significantly higher in hospital non-survivors, representing exploratory prognostic associations. Conclusions: Changes in haemodynamic and inflammatory parameters were observed during oXiris®-based CRRT treatment in critically ill patients with non-septic VS. Early initiation (≤8 h) was associated with shorter VS resolution time in this exploratory, uncontrolled analysis. Residual IL-1β and TNF-α at 72 h were associated with hospital mortality in exploratory analyses and may represent hypothesis-generating prognostic signals requiring prospective validation. Randomised controlled trials are warranted to confirm these findings and define optimal timing strategies.