Idiopathic thrombocytopenic purpura coexisting with polycythemia vera.

Sir,

Polycythemia vera (PV) coinciding with other hematological disorders[1-5] and idiopathic thrombocytopenic purpura (ITP) concomitantly associated with other blood-related neoplasm[67] have been sporadically described; however, ITP and PV coinciding in the same patient, as we have observed in three cases during the last 20 years, have not been described so far. All patients were men of 63, 68, and 77 years, respectively; they were referred to our attention because of the simultaneous finding of thrombocytopenia, with platelet counts ranging from 8 to 29 × 109/l. This finding was associated with abnormally high hematocrit levels (from 54 to 58%). Despite the relatively low platelet counts, they presented no petechiae or mucosal bleeding. At admission, two patients were receiving antihypertensive agents, whereas the third was affected by diabetes mellitus, which was well controlled by oral glucose-lowering agents. Two patients, out of three, no episodes of thromboses were present in their histories, whereas in the third patient, the most old, hematological abnormalities (a hematocrit level of 58% associated with a platelet count of 12 × 109/l) were discovered during a previous admission in another hospital where he had been referred because of a cerebral ischemic ictus. On physical examination, they were afebrile and not appeared acutely ill. Blood pressure, heart and respiratory rates were within normal values in all patients. A palpable spleen, 2-cm below the costal margin, was found in two patients. The gas analysis performed from the arterial blood samples revealed normal oxygen saturation. Moreover, respiratory functional tests and radiological examination of the chest ruled out any bronchopulmonary disorders. On ultrasonography and computed tomography of the abdomen, mild splenomegaly was present in two cases; normal liver and the absence of lymphadenopathy were also found. In addition, renal and liver functions and electrolytes tests were all normal. Autoimmunity panel was negative; in particular, antiphospholipid and antinuclear antibodies were not found. Again, antibodies against cytomegalovirus, EBV, HIV1/2, hepatitis A, B, C, and Helicobacter pylori were negative. Breath test for H. pylori infection was negative in all patients. Given that the causative role of this pathogenic bacteria in the setting of thrombocytopenia has been demonstrated only in recent years,[8-10] its presence was tested in two patients during the initial work-up and in the remaining case several years after the onset of the hematological disorders. The morphological examination of peripheral blood showed marked thrombocytopenia and platelet anisocytosis but not other significant abnormalities. Erythroid and megakaryocytic hyperplasia was revealed by the examination of the bone marrow (BM) specimens in all of the three cases; these findings were according to the erythrocytosis and to a peripheral platelet-consuming process. No myelofibrotic and/or myelodysplastic changes in the BM were found. Given the exclusion of all other causes of thrombocytopenia, an immune-mediated ITP was suggested with a concomitant diagnosis of PV; the latter diagnosis was also confirmed by the detection of the Janus Kinase 2 (JAK2 V617F) mutation on PB; given that this test was only recently available, in two cases, it was performed several years after the primary diagnoses with a confirmatory intent, whereas in the remaining patient, it was done at the disease onset as part of the initial work-up. PV was controlled by regular phlebotomies; prednisone 1 mg/kg was given to treat of ITP and a full response was achieved and maintained over the time; however, the reduction of steroid dosage was followed by the decreasing of the platelet count in all patients. Indeed, none of them achieved the cure of the supposed ITP, although a clear increase of platelets following full-dose steroids indirectly supported the putative immunological origin of the low platelet count in our patients. However, given the relative safe platelet counts (45 – 70 × 109/l) maintained by the patients, they are carefully followed and no salvage treatments have been offered so far. To date, with a follow-up of 18, 4, and 1 years, respectively, chronic but asymptomatic ITP persisted throughout the course of active PV, although the requirement for phlebotomy to maintain appropriate and safe levels of hematocrit is diminished over the time. In conclusion, we have reported three patients above 60 years of age with a definite diagnosis of PV (JAK2 V617F positive) and severe thrombocytopenia (platelet count <30 × 109/l) which was interpreted as concomitant ITP, given the exclusion of other diseases associated with thrombocytopenia and the rapid response to corticosteroids. The significance of our observation remained unexplained and no conclusions on the meaning of this coincidence, if it would be probably fortuitous or the manifestations of an underling unique disease, could be drawn.