Viscosity measurements on very small capillary blood samples.

Viscosity measurements on very small capillary blood samples could be of considerable clinical interest. We have developed an oscillating viscometer for very small volumes, which consists of a glass capillary containing 7 mul of blood, which is part of an oscillating torsional resonator. The damping of the sinusoidal oscillations depends on the density and viscosity of the fluid, which allows blood viscosity measurements. The instrument was first evaluated in comparison with a standard blood viscometer (Contraves LS 30). Blood from healthy volunteers anticoagulated with EDTA was adjusted to hematocrit levels of 20, 30, 40, 50, and 60%, respectively. A strong correlation was found between hematocrit and oscillating viscosity (y=0.17x-2.05, r=0.969, p<0.0001) and between oscillating and conventional high shear viscosity (y=1.11x-0.62, r=0.971, p<0.0001). Blood viscosity measured in venous or capillary blood of normal subjects was similar (p=0.63). Bedside viscosity measurements on capillary blood drawn from a finger prick during routine blood glucose measurements in patients with diabetes mellitus showed lower blood viscosity than controls (3.62+/-0.87 vs 4.79+/-0.59 mPa.s, p=0.0007), which is in contrast to earlier publications, and may be explained by the lower hematocrit in our diabetic patients (34.7+/-6.0% vs. 43.1+/-1.9%, p<0.0001). Blood viscosity was independent of the actual glucose level (range 3-17 mmol/l). Capillary blood anticoagulated with EDTA was drawn by heel prick from 23 newborns. Blood viscosity was higher (5.66 +/-2.47 mPa.s) than in adult controls (see above), which could be explained by the dependence on the higher hematocrit (46.4 +/-8.6%). We conclude that viscosity measurements can be made on very small samples such as capillary blood from diabetic patients or newborn babies with this new oscillating viscometer. It remains to be determined if such new informations have clinical implications.