The thalassemia syndromes are among the most frequent monogenic diseases, particularly but not exclusively in low-income countries. Regular red cell transfusion is the mainstay of treatment of anemia in beta thalassemia major, the most severe form of the syndromes. In the past 30 years, the adoption of transfusion regimens tailored to keep hemoglobin closer to normal levels has dramatically improved the life expectancy of these patients in countries where adequate therapeutic programs are available. However, iron overload, caused not only by multiple transfusions but also increased intestinal absorption, causes continuous oxidative stress and massive cardiac and hepatic deposition of this metal, which ultimately results in multiple organ failure and premature death. In the past few years the availability of efficacious chelation therapy by means of oral drugs such as deferiprone and deferasirox has been a major step forward in the reduction of the complications of iron overload. Owing to adequate transfusion regimens coupled with the use of these drugs, the life expectancy of newly diagnosed children with thalassemia is becoming increasingly similar to that of their unaffected age peers, at least in high-income countries.
Thalassemia intermedia is less severe than thalassemia major, with hemoglobin levels usually ranging from 8.0 to10.0 g dL−1, so that these patients are less transfusion dependent. Until recently, splenectomy was frequently performed in these patients, in order to deal with spleen enlargement, decrease transfusional need, avoid growth retardation and the hematological consequences of hypersplenism (leukopenia and thrombocytopenia). Splenectomy is not without risks, beside that of bacterial infectious complications. In the 1990s, Israeli scientists led by the late Amiram Eldor and Eliezer Rachmilewitz were the first to indicate that in patients with beta-thalassemia arterial and venous thrombotic episodes are more frequent than expected for their young age [1,2]. Subsequently large cohort studies carried out in Italy  and other regions of high endemicity such as Mediterranean countries, the Islamic Republic of Iran and Thailand [4–7] clearly established that thrombotic complications were mainly occurring in splenectomized patients with thalassemia intermedia (at least four times more frequently than in thalassemia major) . Thrombosis affects the venous more than the arterial vessels, although changes in the white matter suggestive of cerebral microthrombosis are relatively common in splenectomized young adults with thalassemia intermedia .
The mechanism of the increased risk of thrombosis in splenectomized patients with thalassemia intermedia (nearly 7-fold) is peculiar. Platelet hyperactivity is expressed by a decreased in vivo survival, increased adhesion and aggregation, membrane exposure of P-selectin and activation of the arachidonic acid pathway, with heightened urinary excretion of metabolites of platelet thromboxane [8–12] which decreases after the administration of low-dose aspirin . Another fundamental mechanism of platelet activation is oxidative stress, caused by iron overload that catalyzes hydroxyl radical generation from activated oxygen species . Beside platelet abnormalities, these patients have a hypercoagulable state expressed by laboratory signs of heightened enzymatic activity of thrombin . A global laboratory test that investigates overall coagulability such as the thrombin generation test gives normal results in plasma from splenectomized patients with thalassemia intermedia . It is only with the adoption of thromboelastometry that the presence of hypercoagulability was unequivocally documented in these patients . The discrepancy between the two global tests clearly directs towards the red blood cells and platelets (present in the whole blood used for thromboelastometry, but not in plasma used for the thrombin generation test) as the culprit for hypercoagulability . As early 1993 Israeli scientists had shown that the structurally abnormal red cells of patients with thalassemia intermedia expose, as a consequence of oxidative stress, the aminophospholipid phosphatidylserine in the outer leaflet of the membrane, so that they behave as activated platelets providing a procoagulant surface for heightened thrombin formation [1–3]. The red cells of patients with thalassemia intermedia are also characterized by enhanced adherence to vascular endothelial cells , also depending on the degree of exposure of phosphatidylserine in the cell membrane.
Why do these pathogenic mechanisms and thromboembolic complications occur more frequently in the milder thalassemia intermedia than in the more severe thalassemia major? Patients with thalassemia major are usually regularly transfused, so that the great majority of their circulating red cells are normal allogeneic cells from blood donors. Why then does venous thromboembolism occur more frequently in splenectomized patients than in those with an intact spleen? Because the presence of the spleen helps to remove the structurally abnormal and rigid red cells that cause hypercoagulability by playing as platelets, whereas after splenectomy a larger burden of abnormal red cells does circulate, particularly in non- or rarely transfused patients. On the whole, these data provide neat evidence that splenectomy in patients with thalassemia intermedia should be considered a risk factor for venous thromboembolism. Hence, the first message for the clinician is that splenectomy, a frequent occurrence in countries where the thalassemia intermedia trait is endemic, should be added to the established risk factors of venous thromboembolism.
What can be done to prevent this? The most obvious action is to avoid splenectomy, a key mechanism in the development of thrombosis (Table 1). Other clues for thrombo prophylaxis stem from the article of Taher et al.  in this issue. By comparing retrospectively splenectomized patients with thalassemia intermedia who had had thrombotic episodes (mainly venous) with age- and sex-matched splenectomized patients who were thrombosis free, and a third group of thalassemia intermedia patients who were thrombosis free and with an intact spleen, the intensity of blood transfusion was one of the main determinants for being thrombosis free . In the past, the tendency of limiting or avoiding transfusion in these patients by performing splenectomy was driven by the risk of transmission of blood-borne viral infections, and by the consequences of iron overload. Both these concerns are now much smaller, because blood supply is safer with appropriate donor testing, and oral chelating agents are more effective and easy to use (albeit more expensive) than the time-honored parenteral desferal. Hence, to have fewer splenectomized patients with thalassemia intermedia is the main thrust of prevention (Table 1), and the practice of recommending splenectomy should be much more limited than it was until recently.
|Avoid splenectomy and transfuse red blood cells as necessary according to hemoglobin values, with added iron chelation|
|Long-term aspirin (80–300 mg daily)|
|Hydroxyurea (daily dosage tailored to keep platelet count below 400 000 μL−1|
|Peri-operative prophylaxis with low-dose unfractionated or low-molecular-weight heparins in patients undergoing major surgery|
Taher et al.  provide other clues for the primary prevention of thrombosis in the many splenectomized patients with thalassemia intermedia who are currently at risk in endemic areas. Because thrombotic complications occur after an average interval of as many as 8 years after splenectomy , this long period of continued exposure to the risk makes it difficult to recommend long-term prophylaxis with such parenteral anticoagulants as unfractionated or low-molecular-weight heparins. Nor is it advisable to prescribe oral vitamin K antagonists, for the risk of bleeding entailed by a long period of treatment with these drugs and the need of laboratory control, at least until new oral anticoagulants are available and proven to be safer and equally effective [16,17]. Aspirin is perhaps the most reasonable option, owing to the evidence of platelet hyperactivity and of white matter changes suggestive of cerebral microthrombosis . Moreover, aspirin may also help to prevent venous thromboembolism, even although the efficacy of this antiplatelet agent is much smaller that that of anticoagulants [18–20]. Taher et al.  have also shown that in splenectomized patients with thalassemia intermedia who developed thrombosis, the platelet count was much higher than in those who did not in spite of being asplenic. Although it remains to be established whether or not the high platelet count is the cause of thrombosis, it is reasonable (but not evidence based) to recommend aspirin (100–300 mg daily) to patients at higher risk because of a high platelet count (Table 1). Another meaningful advice is to reduce the platelet count to at least below 400 000 μL−1 using hydroxyurea, that is frequently used in patients with thalassemia intermedia with the goal to increase the levels of fetal hemoglobin and thereby reduce transfusion requirement. Perhaps the dosage of hydroxyurea should be tailored with the goal of achieving not only a rise in fetal hemoglobin but also a reduction of platelet count (Table 1).
To sum up, thalassemia intermedia is to be added to the acquired risk factors of thrombosis, particularly of venous thromboembolism, less so of atherothrombosis. As the risk of thrombotic complications is particularly prominent in patients who underwent splenectomy, this operation should be avoided, even if this choice entails a greater need of red cell transfusion and iron chelation. Patients with thalassemia intermedia exposed to situations carrying an added risk of thrombosis such as major surgery (for instance, colecystectomy, frequently needed in these patients due to cholelithiasis) should definitely receive robust peri-operative antithrombotic prophylaxis with heparins , even if they are young. In patients at high risk because they are splenectomized, long-term antithrombotic prophylaxis may be considered, but the only approach that can be proposed right now is the inhibition of platelet aggregation with aspirin and control of high platelet counts with hydroxyurea. These approaches should be formally evaluated in the frame of controlled clinical trials.