Reliable plasma viscosity measurements are required for cardiovascular risk factor assessment. This study was designed to examine the influence of tourniquet pressure and duration on plasma viscosity measurement with a coaxial cylinder at 37 degrees C. Plasma viscosity was evaluated at low (40 mm Hg) and high (mean arterial blood pressure plus 10 mm Hg) tourniquet pressure and at two time intervals following (0-1 and 4-5 min) of tourniquet application in 38 healthy adults. Longer duration of tourniquet application was accompanied by a significant increase in plasma viscosity at both low (1.356 +/- 0.067 mPa s versus 1.393 +/- 0.067 mPa s (p = 0.0001)) and high (1.321 +/- 0.071 mPa s versus 1.430 +/- 0.086 mPa s (p = 0.0002)) tourniquet pressures. Plasma viscosity values were not influenced by the magnitude of tourniquet pressure. The accuracy of plasma viscosity measurements can be improved by sample acquisition within 1 min following tourniquet application.