Extracorporeal Membrane Oxygenation Modulates the Inflammatory Milieu and Organ Failure Trajectory in Severe COVID-19...

Background and Objectives: Coronavirus disease 2019 (COVID-19) triggers a dysregulated host response that may culminate in refractory hypoxaemic shock. Whether veno-venous ECMO modifies the inflammatory cascade more effectively in COVID-19 than in other septic states, and how it compares with conventional ventilatory support for COVID-19, remains uncertain. We compared three groups: COVID-19 patients supported with ECMO (COVID-ECMO, n = 25), non-COVID-19 septic shock patients on ECMO (SEPSIS-ECMO, n = 19) and critically ill COVID-19 patients managed without ECMO (COVID-CONV, n = 74). Methods: This retrospective study (January 2018-January 2025) extracted demographic, laboratory and clinical data at baseline, 48 h and 72 h. The primary end-point was the 72 h change in SOFA score (ΔSOFA). The secondary end-points included the evolution of interleukin-6 (IL-6), C-reactive protein (CRP), D-dimer and ferritin; haemodynamic variables; and 28 day mortality. A post hoc inverse-probability-of-treatment weighting (IPTW) sensitivity analysis adjusted for between-group severity imbalances. Results: Baseline APACHE II differed significantly (29.5 ± 5.8 COVID-ECMO, 27.4 ± 6.1 SEPSIS-ECMO, 18.2 ± 4.9 COVID-CONV; p < 0.001). At 48 h, IL-6 fell by 51.8% in COVID-ECMO (-1 116 ± 473 pg mL^-1) versus 32.4% in SEPSIS-ECMO and 18.7% in COVID-CONV (p < 0.001). The ΔSOFA values at 72 h were -4.6 ± 2.2, -3.1 ± 2.5 and -1.4 ± 1.9, respectively (p < 0.001). ECMO groups achieved larger mean arterial pressure rises (+16.8 and +14.2 mmHg) and greater norepinephrine reduction than COVID-CONV. The twenty-eight-day mortality was 36.0% (COVID-ECMO), 42.1% (SEPSIS-ECMO) and 39.2% (COVID-CONV) (p = 0.88). Across all patients, IL-6 clearance correlated with ΔSOFA (ρ = 0.48, p < 0.001) and with vasopressor-free days (ρ = 0.37, p = 0.002). Conclusions: ECMO, regardless of aetiology, accelerates inflammatory-marker decline and organ failure recovery compared with conventional COVID-19 management, but survival advantage remains elusive. COVID-19 appears to display a steeper cytokine-response curve to ECMO than bacterial sepsis, suggesting phenotype-specific benefits that merit confirmation in prospective trials.