(on behalf of the authors)

doi:10.1093/eurheartj/ehi153 Haemorheological variables as risk factors of ischaemic heart diseases: reply We thank Dr Késmárky and colleagues for their interest in our work. We agree in principle with their conclusions, but we note that there are two principal differ-ences between our respective scientific materials. First, our data are based on the general population which will include a proportion of subjects with epidemio-logical and clinical evidence of ischaemic heart disease (IHD) as well as ‘healthy subjects ’ without such evidence. These data permit calculation of risk of sub-sequent IHD in the general population rather than solely in patients who have visited a clinician. Secondly, we did not present data on whole blood viscosity; a recent meta-analysis showed that plasma viscosity was predictive for IHD but there were too few prospective studies of whole blood viscosity to warrant a full meta-analysis.1 Furthermore, it has been proposed that in epidemiological studies, plasma viscosity and haematocrit can be used to estimate whole blood viscosity, as these measures are far less technically demanding than whole blood viscosity.2 In our own paper, we have suggested that fibrinogen, plasma viscosity, and white cell count could be considered as robust and cheap additional risk predictors of IHD in the general population, but also that these results should be tested in other epidemiological studies. Reference