Validity of CHA2DS2-VASc score in patients with malignancies

Abstract

Abstract Introduction Anticoagulation for patients with atrial fibrillation/flutter (AF) after risk stratification based on CHA2DS2-VASc score (congestive heart failure, hypertension, age group, Diabetes Mellitus, prior stroke, transient ischemic attack or thromboembolism, prior heart attack, peripheral artery disease or aortic plaque and patient's gender) has been the gold standard practice to prevent cerebrovascular accidents (CVA). Purpose Since patients with malignancies are at increased risk of thrombophilia, we tried to study if cancer patients with atrial fibrillation are at increased risk for CVA beyond what should be expected based on their comorbidities. Methods We used the Nationwide Readmissions Database (NRD) database for the years 2016–2019. First, we extracted all cases with a diagnosis of atrial fibrillation/flutter. Exclusion criteria included patients younger than 18-year-old. We studied the comorbidities and calculated CHA2DS2-VASc score. Then, we compared the risk for admission with an acute CVA in patients with and without any diagnosis of malignancies (solid cancer, leukemia, lymphoma or metastatic disease). Patients with carcinoma in situ were considered non-cancer patients. Corrected logistic regression and Chi-square tests using survey procedures in SAS 8.1 were applied to accommodate for complex sampling (Rao-Scott design). Results We identified 1,748,619 weighted admissions with AF equally distributed across the years. 5% were patients with malignancies. The mean age was 74-year-old for cancer patients and 71-year-old for non-cancer patients with quite equal gender distribution (43–49% were female). Patients with a diagnosis of cancer has high inpatient mortality (13%). Every additional point in CHA2DS2-VASc score was associated with increased odds of admission with CVA. Having a diagnosis of cancer was not associated with increased odds of admission with CVA (odd ratio for 95% confidence interval 1.0 [0.95–1.1]). Patients with history of VTE (assuming they are on anticoagulation) or who are on chronic anticoagulation have lower risk of admission with acute CVA if they have AF. Conclusions Although malignancies is associated with hypercoagulable status including higher risk for stroke, deep venous thrombosis and pulmonary embolism, our cross-sectional study indicates that CHA2DS2-VASc score is still a reliable tool for risk stratification in patients with atrial fibrillation. More longitudinal studies are needed. Funding Acknowledgement Type of funding sources: None.