HEMATOCRIT - A PREDICTOR OF CARDIOVASCULAR MORTALITY

Objectives. The main purpose of the study was to assess a possible association between haematocrit (Hct) and long-term cardiovascular disease (CVD) mortality. Design. An extensive examination programme was carried out in 2014 men, defined as apparently healthy, during the period 1972 to 1975, including Hct measurements in a 25% random subsample. Sequential, cause-specific mortality was followed prospectively over a period of 16 years. Setting. The survey was conducted at Medical Department B, Rikshospitalet, Oslo, Norway. Subjects. The participants represented 86% of all eligible apparently healthy men working in five preselected companies in Oslo. Interventions. No intervention was given by the study group during follow-up. Main outcome measures. Complete, cause-specific mortality figures after 8-16 years were obtained from the Norwegian Central Bureau of Statistics. Results. Hct measures were obtained in 488 men (24.2%). Mean Hct was 47.2% (SD2.9%). After correcting for differences in age, plasma cholesterol, systolic blood pressure, erythrocyte sedimentation rate and smoking habits (Cox proportional hazards model), an increase in Hct by 2 SDS was associated with an increase in CVD mortality by a factor ranging between 2.9 at 10, and 1.9 at 16 years (P < 0.05). A similar increased risk was observed earlier during follow-up but the number of deaths was too small for meaningful statistical analysis. No association was found between Hct and non-CVD mortality. Conclusions. Our data show that increased Hct is associated with an increased risk of dying from CVD-independent of conventional CVD risk factors.