Impact of prior history of cancer in the outcomes of patients with atrial fibrillation

Abstract

Abstract Introduction Atrial fibrillation (AF) is common among cancer survivors, but anticoagulant therapy in is particularly challenging in this population. Aim We analyzed how a previous diagnosis of cancer influences the embolic and hemorrhagic risk in AF patients in order to guide anticoagulant therapies. Methods We used a large retrospective and observational cohort of AF patients from Vigo, Spain (2014–2018). Primary endpoint were embolic and haemorrhagic events during follow-up. Secondary endpoints were all cause mortality, CV mortality, acute myocardial infarction (AMI) and heart failure (HF) admission. Results From the 16,056 included patients, 7.1% had a current/past diagnosis of cancer. Median follow-up was 4.9 years. The rate of anticoagulation therapy was similar in cancer and non-cancer patients (74.8% vs 75.8) although there was a higher prescription of heparin and a lower prescription of VKA in cancer group. Cancer patients had a higher embolic (CHA2DS2-VASC 3.5±1.5 vs 3.2±1.5; p<.001) and hemorrhagic risk (HASBLED 3.0±1.2 vs 2.6±1.2; p<.001). Embolic and hemorrhagic risk stratification with CHA2DS2-VASC and HASBLED scores was only accurate in patients with past (non-active) cancer (sHR CHA2DS2-VASC for embolic events: 1.26, 95% CI 1.02–1.54, P=.028 and sHR HASBLED for bleeding events 1.11, 95% CI 1.01–1.23, P=.029) but not in active-cancer patients (CHADS2-VASC sHR 1.14, 95% CI 0.98–1.32; p=.076 and HASBLED sHR 1.08, 95% CI 0.99–1.17; p=.070). After adjusting by age, sex, CHA2DS2-VASC, HASBLED, and anticoagulation therapy, cancer was not associated with higher risk of embolic events (sHR 0.73, 95% CI 0.41–1.26; p=.256), neither in anticoagulated nor in non-anticoagulated patients. However, cancer patients presented a higher bleeding risk (sHR 1.18, 95% CI 1.07–1.30; p=0.001), both in anticoagulated and non-anticoagulated patients. Anticoagulation therapy was associated with lower rate of embolic events and with higher rate of bleeding events in both groups of patients with and without cancer, although the embolic-hemorrhagic balance associated with anticoagulation was more favorable in cancer patients (for each avoided embolism with anticoagulation, 5.7 bleeding and 11.7 bleeding episodes occurred in cancer and non-cancer group respectively). We found no association between cancer and AMI or CV mortality rates. All-cause mortality was higher in cancer patients. For HF, only chemotherapy-treated patients presented higher risk (sHR1.54, 95% CI 1.10–2.16; p=0.012). Conclusion Patients with prior history of cancer have higher bleeding risk but similar embolic risk compared to patients without cancer history. Anticoagulation significantly reduced embolic events, with an increase in bleeding outcomes. However, the embolic-hemorrhagic profile of anticoagulation was more favorable in patients with cancer than without it. Funding Acknowledgement Type of funding sources: None. Scores performances by cancer activityEmbolic/bleeding net clinical balance