On whole blood viscosity measurements in healthy individuals and in rheumatoid arthritis patients.

Different methods of measuring whole blood viscosity using a couette rotational viscometer were compared to establish its use in clinical rheumatological practice. The relationship between blood viscosity and hematocrit was approximately exponential and no significant differences in the slopes were found between healthy controls and rheumatoid arthritis patients. Correction of native blood viscosity to a standard hematocrit of 40% by extrapolation from a standard regression curve, established by concentration/dilution of samples from healthy persons to correct for hematocrit differences and at shear rate 92s-1, was the best method for differentiating between viscosities of patients and controls. It was also the least laborious method, requiring the smallest amounts of blood and having the lowest method error. Native blood viscosity, corrected blood viscosity, plasma viscosity and red cell aggregation were all significantly higher and hematocrit significantly lower in rheumatoid arthritis patients than in controls.