Evaluation of Plasma Biomarkers to Understand the Biology and Heterogeneity of Treatment Effect in Lower Tidal Volume Ventilation Facilitated by Extracorporeal CO2 Removal in Acute Hypoxemic Respiratory Failure: A Secondary Analysis of the REST Trial.

OBJECTIVES: In patients with acute hypoxemic respiratory failure (AHRF), the use of lower tidal volume ventilation facilitated by veno-venous extracorporeal CO2 removal (vv-ECCO2R) does not improve clinical outcomes. The primary objective of this analysis was to evaluate for differences in indices of systemic inflammation and ventilator-induced lung injury between patients treated with lower tidal volume ventilation facilitated by vv-ECCO2R and standard care. Secondary objectives included an evaluation for heterogeneity of treatment effect. DESIGN: Substudy of a randomized clinical trial. SETTING: Nine U.K. ICUs. PATIENTS: Moderate-to-severe AHRF (Pao2: Fio2 < 150mmHg [20ka]). INTERVENTION: Plasma samples obtained at baseline and day 3. MEASUREMENTS AND MAIN RESULTS: The primary outcome was day 3 C-reactive protein (CRP). Clinical outcomes included 90-day mortality and ventilator-free days (VFD) until day 28. Exploratory analyses included an evaluation of plasma indices of lung injury, inflammation, and heterogeneity of treatment effect (HTE). Seventy-nine patients were enrolled, and 69 patients had paired plasma samples taken at baseline and day 3. There was no difference in day 3 plasma CRP (intervention 138.6 [70.4, 189.4] vs. standard care 113.0 [62.7, 233.8] mg/L; p = 0.72). Between baseline and day 3, there was a greater increase in plasma interleukin-18 in patients that received intervention compared with those that received standard care (Δ 337.7 [-128.9, 738.9] vs. 6.4 [-457.2, 6.4] pg/mL p = 0.05). In patients with high interleukin-18, allocation to intervention was associated with increased VFDs (p = 0.03). Similarly in patients with a hyperinflammatory phenotype, the intervention was independently associated with increased VFDs (p < 0.01) and decreased 90-day mortality (p = 0.01). CONCLUSIONS: In patients with moderate-to-severe AHRF, lower tidal volume ventilation, facilitated by vv-ECCO2R, was not associated with a difference in day 3 plasma CRP, but was associated with an increase in plasma interleukin-18 between baseline and day 3. Baseline plasma interleukin-18 and inflammatory phenotypes may identify subgroups of patients with moderate-to-severe AHRF that benefit from lower tidal volume ventilation facilitated by vv-ECCO2R. TRIAL REGISTRATION: NCT02654327.