Safety of anticoagulation in patients with atrial fibrillation and MDS/AML complicated by thrombocytopenia: An unresolved challenge: Can they be managed? A report of three cases and literature review

Atrial fibrillation has a fivefold increased risk of ischemic stroke and accounts for 15 percent of strokes nationwide.1 Treatment with anticoagulation reduces the risk of thromboembolic events in patients with atrial fibrillation; however, treatment in patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) poses a unique challenge. These patients have a high risk of thromboembolic events, both from atrial fibrillation and from hypercoagulability associated with malignancy. They also frequently have thrombocytopenia, which places them at increased risk for bleeding, a risk that only increases with anticoagulation. There is currently no literature to guide physicians as to how to manage this patient population. We report three cases of patients with MDS/AML and atrial fibrillation in whom anticoagulation was continued despite thrombocytopenia. We retrospectively reviewed cases of patients with known MDS/AML and thrombocytopenia treated with oral anticoagulation for thromboembolic prophylaxis for atrial fibrillation. Cases were selected if they met the following inclusion criteria: (1) thrombocytopenia defined as platelet count <50 × 109/L, (2) treatment with oral anticoagulation for thromboembolic prophylaxis for atrial fibrillation, (3) follow-up during treatment for MDS/AML. Platelet counts were calculated from the patient's initial presentation in the Division of Hematology/Oncology at our institution through October 1, 2017. Platelet transfusions were given to maintain a platelet count above 10 × 109/L, which is the standard operating procedure in our practice. Case 1: An 85-year-old man with anemia on routine laboratory tests was found to have MDS with 4 percent blasts on subsequent bone marrow biopsy. Initial labs at our institution showed thrombocytopenia with platelet count of 101 × 109/L. He completed five cycles of azacitidine for MDS. During a follow-up period of 321 days, he continued to have thrombocytopenia, with median platelet count of 50 × 109/L (9–114 × 109/L). During this time, he continued warfarin for atrial fibrillation, with an INR in the target range between 2 and 3. No major bleeding or thrombosis was reported, although patient reported self-limited epistaxis. Case 2: A 77-year-old man, diagnosed with AML at an outside institution, completed 7 + 3 induction chemotherapy, followed by two cycles of cytarabine consolidation therapy. He presented to our institution 11 months later with relapsed disease, where he was noted to have thrombocytopenia with platelet count of 54 × 109/L. He completed six cycles of azacitidine. He continued to have thrombocytopenia during a follow-up period of 191 days; median platelet count was 58 × 109/L (30–169 × 109/L). He continued rivaroxaban for atrial fibrillation and did not experience any bleeding or thrombotic complications. Case 3: A 76-year-old woman with progressive thrombocytopenia and anemia for several years was diagnosed with MDS on bone marrow biopsy. On presentation to our institution, she had thrombocytopenia with platelet count of 44 × 109/L. She received supportive care for MDS. She remained thrombocytopenic over a follow-up period of 98 days, with a median platelet count of 46 × 109/L (35–58 × 109/L). She continued warfarin for atrial fibrillation and did not experience any bleeding or thrombotic complications. Three patients with atrial fibrillation and MDS/AML complicated by thrombocytopenia are presented in this review. They all received oral anticoagulation for atrial fibrillation despite thrombocytopenia. The average follow-up period was 203 days (98–321 days). No bleeding or thrombotic events were reported. Table 1 summarizes patient characteristics, platelet counts, and bleeding/thrombotic outcomes. Atrial fibrillation is the most common arrhythmia in adults, particularly older adults as the prevalence increases significantly with age.2 A 2007 meta-analysis of 29 randomized clinical trials demonstrated that warfarin reduced the risk of stroke in patients with atrial fibrillation by 64%. By comparison, anti-platelet therapy reduced the risk of stroke by 22%.3 A scoring system, CHA2DS2-VASc, has been used to risk stratify patients for thromboembolic events from atrial fibrillation (2 points for previous embolic event and age ≥75 years; 1 point each for congestive heart failure, hypertension, diabetes mellitus, age 65–74 years, vascular disease, and female sex).4 Patients deemed to be “moderate to high risk” with scores ≥2 are recommended to start anticoagulation, either warfarin or a direct oral anticoagulant. In our case series, all three patients had CHA2DS2-VASc scores ranging from 2 to 5 and were prescribed anticoagulation. The major concern for all antithrombotic medications is the increased risk of bleeding, particularly intracranial bleeding. Multiple studies have shown an almost twofold increased risk of intracranial bleeding with warfarin use, although the incidence was still quite rare (0.46 per 100 person years).5 Factors shown to increase the risk of bleeding include advanced age, uncontrolled hypertension, history of myocardial infarction or ischemic heart disease, cerebrovascular disease, anemia, history of bleeding, and concomitant use of anticoagulant and antiplatelet agents.6 All three of our patients were of advanced age and two had history of cerebrovascular disease, placing them at increased risk for bleeding. MDS and AML occur most often in an age group fraught with comorbidities, such as atrial fibrillation, necessitating anticoagulation for thromboembolic prevention. Adding to this challenge is the increased risk of bleeding, which is only further elevated with use of anticoagulation. Our three patients with MDS/AML were safely treated with anticoagulation with no bleeding or thrombotic complications, which give us hope that this high-risk population can be safely managed. Nothing to report.