Arterial and venous thrombosis by high platelet count and high hematocrit: 108 521 individuals from the Copenhagen General Population Study

Background: It is unclear whether high platelet count or high hematocrit predict risk of thrombosis in individuals from the general population. Objectives: We tested the hypothesis that individuals from the general population with high platelet count or high hematocrit have high risk of arterial and venous thrombosis. Methods: We prospectively followed 108 521 individuals from The Copenhagen General Population Study for a median of 8 years. Platelet count and blood hematocrit were measured at study entry. Results and Conclusion: Multivariable adjusted hazard ratios for individuals with platelet counts in the top 5 percentiles (>398 × 10⁹/L) vs in the 25th-75th percentiles (231-316 × 10⁹/L) were 1.77 (95% confidence interval [CI], 1.38-2.24) for arterial thrombosis in the brain (38 and 26 events/10 000 person-years) and 0.82 (95%, 0.61-1.11) for arterial thrombosis in the heart (23 and 28 events/10 000 person-years). For individuals with hematocrit values in the top 5 percentiles (women/men: >45/>48%) vs the 25th-75th percentiles (women/men: 38.1-42/41.1-45%), hazard ratios were 1.27 (95% CI, 0.91-1.75) for arterial thrombosis in the brain (40 and 26 events/10 000 person-years) and 1.46 (95% CI, 1.06-2.00) for arterial thrombosis in the heart (43 and 25 events/10 000 person-years). Neither high platelet count nor high hematocrit was associated with risk of venous thromboembolism. When excluding individuals with myeloproliferative neoplasia from the main analyses, results on risk of thrombosis were similar. In this prospective study, high platelet counts were associated with 1.8-fold risk of arterial thrombosis in the brain, whereas high hematocrit was associated with 1.5-fold risk of arterial thrombosis in the heart.