Patients with essential thrombocythemia (ET) aged less than 60 years, who have not suffered a previous vascular event (low-risk patients), may develop thrombotic or hemorrhagic events. So far, it has not been possible to identify useful markers capable of predicting which of these patients are more likely to develop an event and therefore who needs to be treated. In the present study, we analysed the relationship between vascular complications and longitudinal blood counts of 136 low-risk ET patients taken over a sustained period of time (blood cells dynamism). After a median follow-up of 60 months, 45 out of 136 patients (33%) suffered 40 major thrombotic and 5 severe hemorrhagic complications. A total number of 5,781 blood counts were collected longitudinally. Thrombotic and hemorrhagic events were studied together (primary endpoint) but also separately (thrombotic alone = secondary endpoint; hemorrhagic alone = tertiary endpoint). The primary endpoint showed no significant association between platelet and WBC count at diagnosis and risk of any event (platelet, p = 0.797; WBC, p = 0.178), while Hb at baseline did show an association (p = 0.024). In the dynamic analysis with Cox regression model, where the blood count values were studied by time of follow-up, we observed that the risk for Hb was 1.49 (95% CI 1.13-1.97) for every increase of 1 g/dL, and that this risk then marginally decreased during follow-up. WBC was associated with an increased risk at baseline for every increase of 1 × 10(9)/L (hazard ratio (HR) 1.07, 95% CI 1.01-1.13, p = 0.034), the risk was stable during follow-up (HR 0.95, p = 0.187 at 60 months). Also, for each increment at baseline of 100 × 10(9) platelets/L, HR was increased by 1.08 (95% CI 0.97-1.22, p = 0.159) and decreases during follow-up. In conclusion, this study is the first to evaluate in ET low-risk patients, the risk of developing a thrombotic/hemorrhagic event considering blood counts over time. Overall our study shows that the risk changes over time. For example, the risk associated with WCC is not linear as previously reported. An interesting new finding is that PLT and even Hb contribute to the risk of developing vascular events. Future treatments should take into consideration these findings and aim to control all parameters over time. We believe this early study may help develop a dynamic analysis model to predict thrombosis in the single patient. Further studies are now warranted to further validate our findings.