Adults with primary immune thrombocytopenia (ITP) may be susceptible to thromboembolism (TE). The objective of this systematic review was to evaluate studies that reported the prevalence and risk of developing TE in the ITP population from ITP diagnosis and splenectomy. We searched several bibliographic databases and included 29 studies. Using meta-analytical techniques, the pooled prevalence of TE before ITP diagnosis was 7.84% (arterial 6.25%; venous 1.95%). The pooled 'annualised' cumulative incidence (without prior TE) and cumulative risk (irrespective of prior TE) were 1.29%/yr and 3.00%/yr, respectively. Splenectomised patients had pooled cumulative risk of arterial TE (ATE) and venous TE (VTE) of 0.19%/yr and 1.10%/yr, respectively. In cohorts, regardless of a history of TE, the pooled relative risk (RR) of any TE was 1.60 (1.34, 1.86) for ITP vs. ITP-free individuals [arterial: 1.52 (1.25, 1.80); venous: 1.70 (0.96, 2.43)]. Splenectomised patients were at higher risk of venous events, pooled RR 2.39 (1.61, 3.17). To conclude, we found an increased risk of TE (mainly ATE) among ITP individuals and a higher risk of VTEs after splenectomy. How intrinsic (ITP pathophysiology, age, gender) and extrinsic factors (treatment) contribute to this risk could not be investigated here but is a task for future studies. Primary immune thrombocytopenia (ITP) in adults is a rare heterogeneous blood disorder which is characterised by a reduced peripheral blood platelet count below 100 9 10 9 /L (1). Its underlying mechanisms are increasingly explained within an 'autoimmune' paradigm in which platelets are destroyed through antibody-mediated destruction and thrombocytopoiesis is reduced through immune-mediated megakaryocytic downregulation (2).Although bleeding is a prominent feature of the disease, patients often present with other comorbid conditions at ITP diagnosis or throughout the course of the disease (3, 4). Notably, there are studies that showed that the risk of thromboembolic events (TE) could be elevated among patients with ITP (4, 5) and also among patients who had ITP-related splenectomies (6). How such a paradoxical 'prothrombotic and thrombocytopenic' state coexists is not completely understood. One likely explanation for this heightened risk for TE is that the ITP pathophysiology promotes a prothrombotic state but the evidence is unclear and weak (7-9). On the other hand, the distribution and influence of typical prothrombotic risk factors, including obesity, smoking, family history of TE, and the impact of ITP-related treatments, such as long-term corticosteroid use or splenectomy, on the risk of TE in ITP have been not completely described and explained. Before engaging into investigative studies on the contributory factors of TE in ITP, it is important to review the published evidence to date to establish whether there is a heightened risk of TE in the ITP population.