Accuracy of transcutaneous carbon dioxide monitoring in hypotensive patients

Abstract

Objectives Continuous blood gas monitoring is frequently necessary in critically ill patients. Our aim was to assess the accuracy of transcutaneous CO2 tension (PtcCO(2)) monitoring in the emergency department (ED) assessment of hypotensive patients by comparing it with the gold standard of arterial blood gas analysis (ABGA). Methods All patients receiving PtcCO(2) monitoring in the ED were included. We excluded paediatric patients, patients with no ABGA results during a hypotensive event, patients whose ABGA was not performed simultaneously with PtcCO(2) monitoring, and patients who received sodium bicarbonate for resuscitation. The included patients were classified into hypotensive patients and normotensive patients. A hypotensive patient was defined as a patient showing a mean arterial pressure under 60 mm Hg. The agreement in measurement between PaCO2 tension (PaCO2) and PtcCO(2) were investigated in both groups. Results The mean difference between PaCO2 and PtcCO(2) was 2.1 mm Hg, and the Bland-Altman limits of agreement (bias +/- 1.96 SD) ranged from -15.6 to 19.7 mm Hg in the 28 normotensive patients. The mean difference between PaCO2 and PtcCO(2) was 1.1 mm Hg, and the Bland-Altman limits of agreement (bias +/- 1.96 SD) ranged from -19.5 to 21.7 mm Hg in the 26 hypotensive patients. The weighted kappa values were 0.64 in the normotensive patients and 0.60 in the hypotensive patients. Conclusions PtcCO(2) monitoring showed wider limits of agreement with PaCO2 in urgent situations in the ED environment. However, acutely developed hypotension does not affect the accuracy of PtcCO(2) monitoring.