270 - Oxygen Consumption in the Critically Ill: An Early Warning System
Thursday, May 14, 2020
1:24 PM – 1:32 PM
Location: Capitol: Terrace Level
Participants should be aware of the following financial/non-financial relationships:
Anne Kirstine Hoeyer-Nielsen: No financial relationships or conflicts of interest
Mathias J. Holmberg: No disclosure data submitted.
Anne V. Grossestreuer, PhD: No financial relationships or conflicts of interest
Michael W. Donnino, MD: No financial relationships or conflicts of interest
Kathrine M. Berg: No disclosure data submitted.
Anne Kirstein Hoyer-Nielsen: No disclosure data submitted.
Background and Objectives: Continuous measurement of oxygen consumption (VO2) and carbon dioxide production (VCO2) could provide real-time information about the degree of metabolic dysfunction a patient is experiencing. We hypothesized that the continuous measurement of VO2, VCO2, respiratory quotient (RQ), and end tidal CO2 (EtCO2) would provide a superior monitoring system for tissue perfusion and metabolism, and that changes in these parameters would provide an “early warning system” for detecting clinical deterioration.
Methods: In a prospective, observational, single-center study, we enrolled critically ill adults who required mechanical ventilation. Each subject was connected to the CARESCAPETM B650 monitor with respiratory module Es-COVX (General Electric Healthcare, Helsinki, Finland) for continuous measurements of VO2, VCO2, RQ, and EtCO2 for up to 48 hours. The primary outcome was changes in VO2 prior to clinical deterioration: 1) new hypotension, 2) decrease in hemoglobin by 20% between two consecutive measurements, or 3) increase in lactate by 20% between two consecutive measurements with a threshold of 3 mmol/L. Secondarily, we assessed changes in VCO2, RQ, or EtCO2 prior to a clinical deterioration. We used Wilcoxon Signed Rank and Wilcoxon Rank Sum test to analyze the changes in metabolic data prior to hypotension between intra-individual measurements and changes in lab values between events and non-events, respectively.
Results: A total of 141 subjects were included of which 27 patients required vasopressor during the observations period (median VO2:: 4 mL/kg/min [IQR: 3.0 – 5.4] from four to two hours versus 4 mL/kg/min [IQR: 3.1 – 5.1] two hours prior to start of vasopressor. VCO2, RQ, and EtCO2 did not change. There was no difference between the two groups based on 39 lactate events, but RQ showed significant changes: 2.4% [IQR: 0.2 – 7.3] in events versus -0.1% [IQR: -2.6 – 2.1] in non-events, p=<0.001. There was no difference in VO2, VCO2, RQ, or EtCO2 between events and non-events based on 12 hemoglobin events.
Conclusion: We found no change in VO2, VCO2, RQ, and EtCO2 prior to a clinical deterioration, however RQ, was significantly different between events and non-events prior to an increase in lactate.