Comparison of pH and PCO2 between arterial and venous blood gas samples in critically ill patients

Abstract BACKGROUND The reliability of peripheral venous blood sampling as a substitute for arterial blood sampling in determining the partial pressure of carbon dioxide (pCO2) in critically ill patients remains controversial. We aimed to determine the utility of venous blood in a medical intensive care unit population. METHODS Seventy-six critically ill patients requiring arterial blood sampling as part of their standard clinical management were prospectively enrolled in this study. A venous blood sample was drawn from the same upper extremity within 5 min of the arterial blood sampling. Venous pH, pCO2, and pO2 were measured in venous samples and compared with those in their arterial counterparts. RESULTS The venous pCO2 was, on average, higher by 5.8 mmHg (standard deviation = 4.6 mmHg) than the arterial pCO2 with Bland-Altman 95% limits of agreement of −3.4, 15.1 mmHg. CONCLUSIONS Obtaining venous blood from an intact upper extremity in critically ill individuals can provide a reliable estimate of arterial pCO2 and eliminate the need for invasive arterial blood sampling.