Treatment of Acute Circulatory Failure Based on Carbon Dioxide-Oxygen (CO2-O2) Derived Indices
Pierre‐Grégoire Guinot, Corentin Evezard, Maxime Nguyen, Sebastien Pili-Floury, Vivien Berthoud, Guillaume Besch, Bélaïd Bouhemad, Valentin Kabbout, Mohamed Radhouani, Audrey Martin, Tiberiu Constandache, Bastien Durand, Pierre Voizeux, Sandrine GrosJean, Vincenza Caruzo, Pierre Alain Bahr, Juliette Bonneau, Justine Perrot, Elodie Chambade, Christophe Douguet, Eugenie Bernard
Abstract
Background Acute circulatory failure is critical in patients in the ICU. Indices derived from oxygen and CO 2 metabolism (CO 2 -O 2 -derived indices) including the central venous-to-arterial CO 2 difference and central venous-to-arterial CO 2 difference/arteriovenous oxygen content ratio are markers for global metabolic demand and tissue hypoxia. Research Question Does a resuscitation strategy using CO 2 -O 2 -derived indices improve tissular hypoperfusion compared with standard care? Study Design and Methods We conducted a randomized, prospective, multicenter, single-anonymized study in 3 ICUs. Patients aged ≥ 18 years with acute circulatory failure and arterial blood lactate levels ≥ 3 mM were included. Patients were randomized to receive either a CO 2 -O 2 -derived algorithm-based treatment or standard clinical practice. The primary outcome was lactate clearance > 10% within 2 hours. Secondary outcomes included Sepsis-Related Organ Failure Assessment score and mortality. Results Of the 179 patients enrolled (90 control patients and 89 treatment patients), there was no significant difference in achieving a lactate clearance > 10% at 2 hours between the control (50%) and interventional groups (43.8%) ( P = .497). At t2 hours, the median change in lactate levels in the control group was −10.53% (−29.27 to 5.68), whereas in the interventional group, it was −2.70% (−22.58 to 19.1; P = .096). Secondary outcomes did not differ between groups in Sepsis-Related Organ Failure Assessment scores (6 [3 to 9] vs 7 [4 to 10]; P = .719), ICU and hospital length of stay (4.5 days [2.0 to 10.8] vs 5.0 days [2.0 to10.0]; P = .963 and 11 days [3.0 to 27.0] vs 10 days [3.0 to 21.0]; P = .493), or 28-day mortality (44.9% vs 33.3%, P = .150). Interpretation Our results indicate that algorithm-based resuscitation using CO 2 -O 2 -derived indices did not improve lactate clearance or clinical outcomes compared with standard care. Further research is needed to identify specific patient subgroups who may benefit from this approach. Clinical Trial Registration ClinicalTrials.gov; No.: NCT05032521; URL: www.clinicaltrials.gov