Bleeding and thrombotic events in patients with atrial fibrillation and cancer: a systematic review and meta-regression analysis
European Heart Journal

Abstract
Atrial fibrillation (AF) and cancer are frequently coexisting in elderly patients. Pooled metanalytic data on the impact of cancer on outcomes in AF patients are lacking.
We want to investigate the impact of cancer in patients with AF, particularly in relation to the incidence of bleeding and ischemic events
Systematic review and meta-regression analysis of clinical studies retrieved from Medline (PubMed) and Cochrane (CENTRAL) databases according to PRISMA guidelines. Safety endpoints included any, major, gastrointestinal (GI) bleeding and intracranial haemorrhage (ICH). Efficacy endpoints included myocardial infarction (MI), ischemic stroke/systemic embolism (IS/SE), cardiovascular (CV) and all-cause death.
15 studies were included in the metanalysis: 4 prospective, 3 randomized clinical trials and 8 retrospective studies with 2,868,010 AF patients, of whom 479,571 (16.7%) had cancer. The pooled HR for cancer was 1.43 (95% CI 1.42–1.44) for any bleeding, 1.27 (95% CI 1.26–1.29) for major bleeding, 1.17 (95% CI 1.14–1.19) for GI bleeding, and 1.07 (95% CI 1.04–1.11) for ICH. The risk of major bleeding increased with the proportion of breast cancer. Cancer increased the risk of all-cause death (HR 2.00, 95% CI 1.99–2.02) whereas no association with MI (HR 0.97, 95% CI 0.94–1.01) and CV death (HR 1.01, 95% CI 0.99–1.03) was found. Patients with AF and cancer were less likely to suffer from IS/SE (HR 0.91, 95% CI 0.89–0.94).
The presence of cancer modifies the clinical history of AF patients, mainly increasing the risk of bleeding. Further analyses according to the type and stage of cancer is necessary to better stratify bleeding risk in these patients.
Type of funding sources: None.
Figure 1
Contributors

D Pastori
Author

D Menichelli
Author

A Di Rocco
Author

A Farcomeni
Author

A Sciacqua
Author

P Pignatelli
Author

L Fauchier
Author

G Y H Lip
Author


