Clinical impact of atrial fibrillation on hospitalized patients with myeloproliferative neoplasms: A nationwide analysis

Abstract Background: Patients with myeloproliferative neoplasms (MPNs), including essential thrombocythemia (ET), polycythemia vera (PV), and primary myelofibrosis (PMF) are at increased risk of both thrombosis and bleeding due to disease-related coagulopathy and treatment-related factors. The coexistence of atrial fibrillation (AF) complicates risk stratification and management, yet real-world data remain limited. Methods: We conducted a retrospective analysis using the National Inpatient Sample (2018–2021) to evaluate the risk of thromboembolic and bleeding complications in hospitalized adults with myeloproliferative neoplasms (MPNs: polycythemia vera, essential thrombocythemia, myelofibrosis) and atrial fibrillation (AF). ICD-10 codes were used to identify MPNs, AF, thromboembolic events (e.g., stroke, pulmonary embolism, myocardial infarction), and major bleeding (e.g., intracranial or gastrointestinal hemorrhage). Primary outcomes included the occurrence of thromboembolism, major bleeding. Secondary outcomes included the length of stay, and total hospital charges. Chi-square and t-test were used to assess demographic information of patients, while multivariable survey-weighted logistic and linear regression models were used to assess the outcomes while adjusting for possible confounders. All analyses accounted for the complex sampling design of NIS using appropriate discharge weights. Results: Among 104,139 hospitalized patients with myeloproliferative neoplasms (MPNs), 17,729 had Atrial fibrillation (AFib) with mean age of 68.3 years. AFib was independently associated with an increased risk of thromboembolism (adjusted odds ratio [aOR] 1.10; 95% CI, 1.06–1.15; p<0.001), longer lengths of stay and increased hospital charges. Other significant predictors of thrombotic events included older age, obesity, and higher Charlson comorbidity index . Female sex was associated with lower thrombotic risk. Although AFib was not significantly associated with major bleeding after adjustment (aOR 0.98; 95% CI, 0.92–1.04; p=0.45), bleeding risk was significantly increased with older age, alcohol use, higher Charlson comorbidity index and anti-coagulant use- factors commonly observed in patients with AFib. Similar trends were observed when stratifying bleeding into subtypes of gastrointestinal and intracranial hemorrhages. Conclusion: Atrial fibrillation is associated with increased thromboembolic risk among hospitalized patients with myeloproliferative neoplasms. These findings emphasize the substantial burden of AFib and comorbidies on healthcare utilization and costs in MPN hospitalizations. This underscores the need for individualized anti-thrombotic strategies and highlight the importance of careful anticoagulation titration in high risk population, by balancing thrombosis and bleeding potential.