Correlation and Agreement Between the CO2 Gap Obtained From Peripheral Venous Blood and From Mixed Venous Blood in Mechanically Ventilated Septic Patients

Abstract

Background Venous-arterial CO2 difference (Pv-aCO2) is a valuable marker that can identify a subset of patients in shock with inadequate cardiac output to meet tissue metabolic requirements. Some authors have found that Pv-aCO2 levels calculated from mixed vs central venous blood demonstrate a linear relationship. The purpose of this study is to determine whether there is a linear relationship between Pv-aCO2 obtained with peripheral venous blood (Pv-aCO2p) and with mixed venous blood, and the agreement between the 2 measures. Methods This was a prospective, single-center, observational clinical study enrolling mechanically ventilated patients in septic shock during the first 24 hours following admission to the intensive care unit. Results The Bravais–Pearson r-coefficient between Pv-aCO2 and Pv-aCO2p was .70 in 38 determinations (95%CI .48-.83; P-value = 1.25 x 10^-6). The Bland–Altman bias was 4.11 mmHg (95%CI 2.82–5.39), and the repeatability coefficient was 11.05. Using the Taffe approach, the differential and proportional biases were 2.81 (95%CI .52–5.11) and 1.29 (95%CI .86–1.72), respectively. Conclusion There was linear correlation between Pv-aCO2p and Pv-aCO2 in mechanically ventilated patients with septic shock. The bias showed a gradual increase in high Pv-aCO2 values in an upward trend.