The variability of hydrostatic hepatic vein and ascitic fluid pressures and of plasma and ascitic fluid colloid osmotic (oncotic) pressures was assessed during hepatic venous catheterization by repeated measurements on different days and at different locations in patients with cirrhosis of the liver. Furthermore, calculation of oncotic pressure from protein determinations was compared to the directly measured value of plasma and ascitic fluid samples. Repeated measurements of hydrostatic pressure in the same hepatic vein within 15 min showed a standard deviation (SD) below 1 mmHg. The variation in hydrostatic hepatic vein pressures, pressure differences and ascitic fluid pressures (when measured at different locations within the liver and peritoneal space during a single examination) was 1.5, 1.0 and 1.0 mmHg (SD), respectively. When measured on different days, the variation of hydrostatic hepatic vein pressures, pressure differences and ascitic fluid pressures was 2.5, 1.6 and 1.3 mmHg (SD), respectively. Repeated measurements of oncotic pressure on the same plasma sample varied 0.3 and 0.6 mmHg (SD) when measured on the same day and a subsequent day, respectively. The variation in plasma and ascitic fluid oncotic pressure, when measured on samples obtained on different days, was 2.0 and 0.6 mmHg, respectively. The error of oncotic pressure calculated from protein determinations was 2.9 and 1.3 mmHg (SD) for plasma and ascitic fluid, respectively. It is concluded that measurements during catheterization give a good reproducibility in determination of the hydrostatic pressures in hepatic vein and ascitic fluid and of the colloid osmotic (oncotic) pressure in plasma and ascitic fluid in the resting supine patient with cirrhosis, which substantiates the use of measurements during a short period as representative for the patients long-time level.