The Finometer measures haemodynamic parameters including cardiac output (CO) using non-invasive volume-clamp techniques. The aim of this study was to determine the accuracy of the Finometer in hyperdynamic cirrhotic patients using an invasive indicator dilution technique as control. CO was measured in twenty-three patients referred for invasive measurements of the hepatic venous pressure gradient on suspicion of cirrhosis. Invasive measurements of CO were performed using indicator dilution technique (CO(I)) and simultaneous measurements of CO were recorded with the Finometer (CO(F)). In six patients, measurements of CO were performed with invasive technique and the Finometer both before and after beta-blockade using 80 mg of propranolol and the changes in CO (DeltaCO(I) and DeltaCO(F) respectively) were calculated to evaluate the Finometers ability to detect relative changes in CO. Mean CO(I) was 6.1 +/- 1.6 [3.9;9.7] l min(-1) (mean +/- SD [range]) compared to mean CO(F) of 7.2 +/- 2.3 [3.1;11.9] l min(-1). There was a mean difference between CO(F) and CO(I) of 1.0 +/- 1.8 [-2.1;4.0] l min(-1) and 95% confidence interval of [0.2;1.8], P<0.001. In patients with measurements before and after beta-blockade, mean DeltaCO(I) was 1.6 +/- 1.4 [-0.1;3.3] l min(-1) compared to mean DeltaCO(F) of 1.9 +/- 1.3 [0.4;3.8] l min(-1). Mean difference between DeltaCO(F) and DeltaCO(I) was 0.3 +/- 0.3 [-0.2;0.7] l min(-1) with a 95% confidence interval of [-0.1;0.6], P = 0.11. Compared with invasive measurements, the Finometer can be used to measure changes in CO, whereas absolute measurements are associated with higher variation in patients with cirrhosis. The Finometer seems useful for repeated determinations such as in studies of effect of pharmacotherapy.