OC 53.2 - External Validation of the AF-BLEED Score in Predicting Major Bleeding in Patients with Newly Diagnosed Atrial Fibrillation

Tuesday, June 25, 2024

09:45 – 10:00 ICT

Room: Ballroom B3

Presenting Author(s)

Sabine F.B van der Horst, MD

Leiden University Medical Center, department of Thrombosis and Hemostasis
Leiden, Zuid-Holland, Netherlands

Co-Author(s)

GC

Gordon Chu

MD, PhD-student
Department of Internal Medicine, Alrijne Hospital, Leiden, the Netherlands
Leiden, Zuid-Holland, Netherlands
ET

Emmy M. Trinks-Roerdink

MD, PhD
Department of General Practice & Nursing Science, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands
Utrecht, Utrecht, Netherlands
Td

Tim A.C. de Vries

MD, PhD-student
4 Department of Clinical and Experimental Cardiology and Cardiothoracic Surgery, Heart Center, Amsterdam UMC location University of Amsterdam, Amsterdam, The Netherlands; Heart Failure and Arrhythmias, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
Amsterdam, Noord-Holland, Netherlands
LV

Leonard Voorhout

MD, PhD-student
Department of Cardiology, Hospital Rijnstate, Arnhem, The Netherlands
Arnhem, Gelderland, Netherlands
JS

Jaap Seelig

MD, PhD
Department of Cardiology, Hospital Rijnstate, Arnhem, The Netherlands
Arnhem, Gelderland, Netherlands
Erik Klok, MD, PhD, FESC

Leiden University Medical Center
FR

Frans H. Rutten

MD, PhD, prof
Department of General Practice & Nursing Science, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands
Utrecht, Utrecht, Netherlands
MH

Martin E.W. Hemels

MD, PhD
Department of Cardiology, Hospital Rijnstate, Arnhem, The Netherlands
Arnhem, Utrecht, Netherlands
Geert-Jan Geersing

MD, PhD, associate prof
Department of General Practice & Nursing Science, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands
Utrecht, Utrecht, Netherlands
MH

Menno V. Huisman

Professor of Medicine
Department of Medicine – Division of Thrombosis and Hemostasis, Leiden University Medical Center, Leiden, the Netherlands; Dutch Thrombosis Network, The Netherlands.
Leiden, Zuid-Holland, Netherlands

Background: Accurate bleeding risk assessment is crucial in atrial fibrillation (AF) patients initiating oral anticoagulant (OAC) therapy. The AF-adapted VTE-BLEED (AF-BLEED) score has been developed to predict major bleeding (MB) within the first 180 days after OAC initiation, enabling tailored anticoagulant care in this population.

Aims: To evaluate the predictive performance of the AF-BLEED score in new-onset AF patients.

Methods: Utilizing DUTCH-AF registry data, a near-nationwide prospective study of patients (≥ 18 years) with recent-onset ( < 6 months) AF or atrial flutter in the Netherlands (July 2018 - July 2021), we included all patients starting OAC at diagnosis. The AF-BLEED score categorized patients as low-risk (score ≤3) or high-risk (score >3) for bleeding. The primary outcome was time to first major bleeding (MB), and the secondary outcome was time to first composite of MB or clinically relevant non-major bleeding (CRNMB) within 180 days. We accounted for death and OAC discontinuation as competing events. Subdistribution hazard ratios (SHR) were obtained and the predictive performance was evaluated in terms of discrimination (time-dependent area under the curve (AUC), i.e. C-statistic) and calibration.

Results: Among 4080 patients analyzed in the complete-case analysis (median age 72 years, 57.3% male, low-risk: 94.0%), the 180-day cumulative MB incidence was 0.50% (95%CI 0.27–0.72%) in low-risk patients and 1.64% (95%CI 0.04–3.24%) in high-risk patients (SHR 3.32, 95%CI 1.13–9.77). Cumulative MB/CRNMB incidences were 1.73% (95%CI 1.31–2.14%) and 4.10% (95%CI 1.61–6.59%) in low- and high-risk patients, respectively (SHR 2.41, 95%CI 1.24–4.70). C-statistics (95%CI) were 0.56 (0.48–0.64) for MB and 0.54 (0.50–0.57) for MB/CRNMB. Calibration plots showed overestimation of MB and slight underestimation of MB/CRNMB.

Conclusion(s): Although 180-day MB and MB/CRNMB incidences were higher among high-risk patients, the AF-BLEED score did not accurately predict relevant bleeding in this cohort of new-onset AF patients who initiated OAC.