Comparison of central and mixed venous saturation during liver transplantation in cirrhotic patients: a pilot study

Liver transplantation is associated with important haemodynamic variations requiring cardiac output and oximetric data monitoring. The mixed venous saturation (SvO2) integrates parameters combining information about oxygen consumption, cardiac output and haemoglobin concentration. Central venous saturation (ScvO2) can be directly measured from blood drawn in the superior venous system via a central venous catheter. ScvO2 has been proposed as an alternative to SvO2 for intraoperative haemodynamic monitoring. The aim of the present study was to examine the level of agreement between SvO2 and ScvO2 during the preanhepatic and the neohepatic stage of liver transplantation in cirrhotic patients.

After agreement from the regulatory authorities for medical research and having obtained informed consent, 30 patients with cirrhosis undergoing liver transplantation were prospectively included. Blood gas samples were simultaneously drawn from the arterial line, the right atrium port and the pulmonary artery port of the catheter: during the preanhepatic stage (two times) and two times 30-40 min after graft revascularization. Arterial saturation (SaO2), haemoglobin concentration, cardiac index, SvO2, ScvO2 and oxygen consumption, delivery and extraction (VO2, DO2 and EO2, respectively) were measured. A Bland-Altman test was used to determine bias and limits of agreement between SvO2 and ScvO2. Both parameters were considered to be equivalent if limits of agreement were within +/-5%.

Bland-Altman analysis revealed a bias (limit of agreement) of -1.2% (-9.1 to 6.6%), -0.3% (-4.8 to 4%) and -2.1% (-12 to 7.8%) for the overall measurements and preanhepatic and postgraft reperfusion measurements, respectively. SvO2 decreased significantly between hepatectomy and reperfusion, whereas cardiac index, VO2, DO2 and EO2 showed significantly higher values after reperfusion. ScvO2 and SaO2 levels did not display different values between the two periods.

Measurements of SvO2 and ScvO2 showed a good level of agreement during the preanhepatic stage, whereas the level of agreement was low after liver graft reperfusion. The increase of VO2 associated with the decrease of SvO2 and the stability of ScvO2 between the two periods suggest an incomplete mixing of splanchnic venous blood into the right atrium. In addition, our samples were taken from the right atrium, which is not possible using a conventional central venous catheter, as the tip must lie in the superior vena cava and not in the right atrium. ScvO2 cannot be considered equivalent to SvO2 for the haemodynamic monitoring of patients with cirrhosis undergoing liver transplantation.