CO046 | BLEEDING RISK FACTOR–BASED PHENOTYPES AND THEIR ASSOCIATION WITH MORTALITY AND HEMORRHAGIC OUTCOMES IN ATRIAL FIBRILLATION
Abstract
Background and aims. In patients with atrial fibrillation (AF), multiple bleeding risk factors are well established. However, whether these factors cluster into distinct clinical phenotypes and how these phenotypes relate to mortality and bleeding outcomes remains unclear. Methods. This prospective observational study included consecutive patients with AF who initiated non-vitamin K oral anticoagulant (NOAC) therapy between 2013 and 2018. Bleeding risk factors were defined a priori based on clinical relevance and literature evidence and included age >75 years, previous bleeding, chronic kidney disease (CKD), cirrhosis, systolic blood pressure >140 mmHg (uncontrolled hypertension), concomitant antiplatelet therapy, NSAID use, and alcohol abuse. These variables were collected at baseline, and patients were stratified into a low-risk group (0–1 risk factors) and a higher-risk group (≥2 risk factors). Unsupervised clustering using latent class analysis (LCA) was applied to the higher-risk group to identify distinct bleeding risk phenotypes. Associations with primary outcomes (all-cause mortality and composite bleeding) and secondary outcomes (major bleeding and clinically relevant non-major bleeding [CRNMB] according to the ISTH definition) were evaluated using Kaplan–Meier analysis and Cox proportional hazards models. Results. Overall, 1,039 patients were included (age 75.2±9.6 years, 56.8% female), of whom 316 (30.4%) were classified as low risk. Among the 723 (69.6%) patients in the higher-risk group, four distinct clusters were identified. Cluster 1 represented a frail phenotype, characterized by a high burden of bleeding risk factors, including advanced age, previous bleeding, CKD, and high fall risk. Cluster 2 was defined by alcohol abuse. Cluster 3 was characterized by a high prevalence of NSAID and antiplatelet use. Cluster 4 was characterized by uncontrolled hypertension. After a median follow-up of 1,539 days (IQR 1,124–1,863), 336 (32.3%) deaths and 277 (26.6%) composite bleeding events occurred. Compared with the low-risk group, all clusters showed an increased risk of all-cause mortality, with the highest risk observed in Cluster 1 (HR 2.99, 95% CI 2.17–4.12) (Figure 1). For composite bleeding, Cluster 1 (HR 1.87, 95% CI 1.30–2.69) and Cluster 4 (HR 1.49, 95% CI 1.06–2.10) were significantly associated with increased risk, whereas the association for Cluster 2 was borderline (HR 1.39, 95% CI 0.97–1.99) and not significant for Cluster 3 (Figure 1). Among secondary outcomes, Cluster 1 remained associated with an increased risk of both major bleeding (HR 2.03, 95% CI 1.04–3.95) and CRNMB (HR 1.79, 95% CI 1.18–2.72) (Figure 1). Clusters 2 and 4 showed an increased risk of CRNMB (HR 1.49, 95% CI 1.00-2.22 and HR 1.62, 95% CI 1.11-2.36, respectively), whereas associations for Cluster 3 were not significant (Figure 1). Conclusions. Bleeding risk factors cluster into distinct clinical phenotypes associated with different risks of mortality and hemorrhagic outcomes.
Related
- Association of ABO blood-group phenotype with postoperative bleeding and long-term mortality after coronary artery bypass grafting
- Association of Iron Homeostasis With One‐Year All‐Cause Mortality Among Intensive Care Unit Patients with Atrial Fibrillation: A Retrospective Study
- Acute Hemorrhagic Manifestations in a Pregnant Patient With Type 1 von Willebrand Disease: Challenges in Management of Bleeding Disorders During Gestation
- Association of D-dimer, fibrin degradation products, and coagulation parameters with progressive hemorrhagic injury following traumatic brain injury : A retrospective cohort study
- Stroke and Bleeding Risk in Long-Term Care Residents With Atrial Fibrillation Who Discontinue Anticoagulation
BibTeX
@article{bucci2026co046,
title = {{CO046 | BLEEDING RISK FACTOR–BASED PHENOTYPES AND THEIR ASSOCIATION WITH MORTALITY AND HEMORRHAGIC OUTCOMES IN ATRIAL FIBRILLATION}},
author = {Tommaso Bucci},
journal = {Bleeding, Thrombosis and Vascular Biology},
year = {2026},
month = oct,
url = {https://www.btvb.org/btvb/article/view/748},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - CO046 | BLEEDING RISK FACTOR–BASED PHENOTYPES AND THEIR ASSOCIATION WITH MORTALITY AND HEMORRHAGIC OUTCOMES IN ATRIAL FIBRILLATION
AU - Tommaso Bucci
PY - 2026
DA - 2026/10/26
JO - Bleeding, Thrombosis and Vascular Biology
UR - https://www.btvb.org/btvb/article/view/748
AB - Background and aims. In patients with atrial fibrillation (AF), multiple bleeding risk factors are well established. However, whether these factors cluster into distinct clinical phenotypes and how these phenotypes relate to mortality and bleeding outcomes remains unclear. Methods. This prospective observational study included consecutive patients with AF who initiated non-vitamin K oral anticoagulant (NOAC) therapy between 2013 and 2018. Bleeding risk factors were defined a priori based on clinical relevance and literature evidence and included age >75 years, previous bleeding, chronic kidney disease (CKD), cirrhosis, systolic blood pressure >140 mmHg (uncontrolled hypertension), concomitant antiplatelet therapy, NSAID use, and alcohol abuse. These variables were collected at baseline, and patients were stratified into a low-risk group (0–1 risk factors) and a higher-risk group (≥2 risk factors). Unsupervised clustering using latent class analysis (LCA) was applied to the higher-risk group to identify distinct bleeding risk phenotypes. Associations with primary outcomes (all-cause mortality and composite bleeding) and secondary outcomes (major bleeding and clinically relevant non-major bleeding [CRNMB] according to the ISTH definition) were evaluated using Kaplan–Meier analysis and Cox proportional hazards models. Results. Overall, 1,039 patients were included (age 75.2±9.6 years, 56.8% female), of whom 316 (30.4%) were classified as low risk. Among the 723 (69.6%) patients in the higher-risk group, four distinct clusters were identified. Cluster 1 represented a frail phenotype, characterized by a high burden of bleeding risk factors, including advanced age, previous bleeding, CKD, and high fall risk. Cluster 2 was defined by alcohol abuse. Cluster 3 was characterized by a high prevalence of NSAID and antiplatelet use. Cluster 4 was characterized by uncontrolled hypertension. After a median follow-up of 1,539 days (IQR 1,124–1,863), 336 (32.3%) deaths and 277 (26.6%) composite bleeding events occurred. Compared with the low-risk group, all clusters showed an increased risk of all-cause mortality, with the highest risk observed in Cluster 1 (HR 2.99, 95% CI 2.17–4.12) (Figure 1). For composite bleeding, Cluster 1 (HR 1.87, 95% CI 1.30–2.69) and Cluster 4 (HR 1.49, 95% CI 1.06–2.10) were significantly associated with increased risk, whereas the association for Cluster 2 was borderline (HR 1.39, 95% CI 0.97–1.99) and not significant for Cluster 3 (Figure 1). Among secondary outcomes, Cluster 1 remained associated with an increased risk of both major bleeding (HR 2.03, 95% CI 1.04–3.95) and CRNMB (HR 1.79, 95% CI 1.18–2.72) (Figure 1). Clusters 2 and 4 showed an increased risk of CRNMB (HR 1.49, 95% CI 1.00-2.22 and HR 1.62, 95% CI 1.11-2.36, respectively), whereas associations for Cluster 3 were not significant (Figure 1). Conclusions. Bleeding risk factors cluster into distinct clinical phenotypes associated with different risks of mortality and hemorrhagic outcomes.
SN - 2785-5309
ER -