Background There is increasing evidence that blood viscosity and its major determinants (haematocrit and plasma viscosity) are associated with increased risks of cardiovascular disease (CVD) and premature mortality; however, their predictive value for CVD and mortality is not clear. Methods We prospectively assessed the added predictive value of plasma viscosity and whole blood viscosity and haematocrit in 3386 men and women aged 30-74 years participating in the Scottish Heart Health Extended Cohort study. Results Over a median follow-up of 17 years, 819 CVD events and 778 deaths were recorded. Hazard ratios (95% confidence intervals) for a 1 SD increase in plasma viscosity, adjusted for major CVD risk factors, were 1.12 (1.04-1.20) for CVD and 1.20 (1.12-1.29) for mortality. These remained significant after further adjustment for plasma fibrinogen: 1.09 (1.01-1.18) and 1.13 (1.04-1.22). The corresponding results for blood viscosity were 0.99 (0.90, 1.09) for CVD, and 1.11 (1.01, 1.22) for total mortality after adjustment for major CVD risk factors; and 0.97 (0.88, 1.08) and 1.06 (0.96, 1.18) after further adjustment for fibrinogen. Haematocrit showed similar associations to blood viscosity. When added to classical CVD risk factors, plasma viscosity improved the discrimination of CVD and mortality by 2.4% (0.7-4.4%) and 4.1% (2.0-6.5%). Conclusions Although plasma and blood viscosity may have a role in the pathogenesis of CVD and mortality, much of their association with CVD and mortality is due to the mutual effects of major CVD risk factors. However, plasma viscosity adds to the discrimination of CVD and mortality and might be considered for inclusion in multivariable risk scores.
Keywords: Viscosity; cardiovascular disease; fibrinogen; haematocrit; mortality.