LAUSR

Patients with immune thrombocytopenia (ITP) have low platelet counts, which may increase bleeding risk, but they are not protected from thrombosis. Multiple studies suggest that the risk of thrombosis is higher in patients with ITP than in the general population. Prior work has described an increased incidence of venous thrombotic events in patients who underwent splenectomy for ITP. However, the contribution of ITP-directed therapies to the increased thrombotic risk seen in patients with ITP is largely unknown. In this issue of the British Journal of Haemotology, Lafaurie et al. use a large French cohort of over 10 000 patients with ITP to assess the thrombotic risk associated with different treatments for ITP. The authors conclude that exposure to corticosteroids, thrombopoietin-receptor agonists (TPO-RA), IV immunoglobulin (IVIG) and splenectomy was associated with an increased risk of hospitalisation for venous thrombosis (VT). All listed agents were also associated with an increased risk of arterial thrombosis (AT), except for corticosteroids. Rituximab was not associated with an increased risk of either AT or VT. Fortunately, there now exists a growing list of therapies for ITP. First-line treatment for adult patients with ITP is relatively straightforward, using steroids +/ IVIG. However, second-line treatment varies based on individual patient factors. Selecting a therapy necessitates shared decision making between patients and their providers, taking into consideration the patient’s lifestyle, comorbidities and values. For example, a patient wanting to limit the need for chronic medications may choose to proceed with a splenectomy. In contrast, another patient may place a high value on decreasing infectious risk and favor a TPO-RA. The optimal strategy may change over time for an individual patient and, thus, needs to be continually re-assessed. Lafaurie et al. seeks to address whether different therapies for ITP may contribute to thrombotic risk. Indeed, this would be an important component of the risk–benefit discussion between clinicians and patients when selecting a treatment strategy. Distinguishing the thrombotic risk of ITP-directed therapies from disease-related factors is not an easy task. First, there is a great deal of heterogeneity in the disease course among individual patients, ranging from patients with platelet counts consistently above 50 9 10/l who never require treatment to patients with repeated weeklong hospitalisations for refractory platelet counts of <10 9 10/l. Clearly, the risk of VT differs in these patients, based on the known elevated VT risk from inpatient hospitalisation alone. Second, patients may have other co-existing autoimmune diseases known to increase thrombotic risk, such as antiphospholipid syndrome. Last, patients with ITP commonly have antiplatelet or anticoagulant therapy withheld during periods of moderate-severe thrombocytopenia or, in some instances, permanently discontinued. The lack of adequate primary or secondary prevention of VT/AT may play a role in the increased thrombotic events seen in ITP. Lafaurie et al. present a methodologically rigorous analysis of thrombotic risk that accounts for many confounding factors (e.g., age, comorbidities). However, because of the inherent limitations of retrospective work, other confounding variables may exist. In Lafaurie et al.’s current study, different ITP-directed therapies were not directly compared. Without comparing strategies, which is difficult for many of the reasons previously outlined, my approach to selecting ITP-directed therapies is not necessarily changed by the findings of this study. There are many additional considerations that must be weighed (e.g., the efficacy of the treatment and the rapidity of response needed) in treatment selection. However, the study importantly highlights that haematologists must remain vigilant of the fact that patients with ITP are not protected from thrombosis. The authors report in this cohort of patients with ITP the 5-year risk of hospitalisation for VT Correspondence: Allyson M. Pishko, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. E-mail: [email protected] commentary