PB0499 - Bleeding risk assessment in end-stage kidney disease: validation of existing risk scores and evaluation of a machine learning-based approach

Monday, July 11, 2022

6:30 PM – 7:30 PM

Location: Exhibition

Presentation Format: In-Person

Presenting Author(s)

Stephan Nopp, MD

PhD Student
Clinical Division of Haematology and Haemostaseology, Department of Medicine I, Medical University of Vienna, Vienna, Austria
Vienna, Wien, Austria

Co-Author(s)

CS

Clemens Spielvogel

PhD Student
Medical University of Vienna
Vienna, Wien, Austria
SS

Sabine Schmaldienst

Head of Department
Clinic Favoriten
Vienna, Wien, Austria
RK

Renate Klauser-Braun

Head of Department
Clinic Donaustadt
Vienna, Wien, Austria
ML

Matthias Lorenz

Medical Doctor
Vienna Dialysis Center, Vienna, Austria
Vienna, Wien, Austria
BB

Benedikt Bauer

Data Management Assistant
Department of Medicine I, Clinical Division of Haematology and Haemostaseology, Medical University of Vienna
Vienna, Wien, Austria
Ingrid Pabinger, MD

Co-Head of the Thrombosis and Haemostasis Research Group
Clinical Division of Hematology and Hemostaseology, Department of Medicine I, Medical University of Vienna, Vienna, Austria
Vienna, Wien, Austria
MS

Marcus Säemann

Head of Department
Clinic Ottakring
Vienna, Wien, Austria
Oliver Königsbrügge, MD PhD

Physician Scientist
Clinical Division of Haematology and Haemostaseology, Department of Medicine I, Medical University of Vienna, Vienna, Austria
Vienna, Wien, Austria
Cihan Ay, MD

Professor, Head of the Thrombosis and Haemostasis Research Group
Clinical Division of Haematology and Haemostaseology, Department of Medicine I, Medical University of Vienna, Vienna, Austria
Vienna, Wien, Austria

Background: Patients with end-stage kidney disease (ESKD) on hemodialysis (HD) are at increased risk for bleeding. However, despite relevant clinical implications regarding dialysis modalities or anticoagulation, no bleeding risk assessment strategy has been established in this challenging population.

Aims: This study sought to evaluate existing bleeding risk scores and develop a machine learning-based prediction model for major bleeding in patients on HD.

Methods: Analyses on bleeding risk assessment models were performed in the population-based Vienna InVestigation of Atrial fibrillation and thromboemboLism in patients on hemoDialysIs (VIVALDI) study including 625 patients. In this cohort study, patients were prospectively followed for a median observation period of 3.5 years for the occurrence of major bleeding. First, performances of existing bleeding risk scores (i.e., HAS-BLED, HEMORR2HAGES, ATRIA, and four others) were evaluated in terms of discrimination and calibration. Second, four machine learning-based prediction models that included clinical, dialysis-specific, and laboratory parameters were developed and tested using Monte-Carlo cross-validation.

Results: Of 625 patients (median age: 66 years, 38% women), 89 (14.2%) developed major bleeding, with a 1-year, 2-year, and 3-year cumulative incidence of 6.1% (95%CI 4.2-8.0), 10.3% (95%CI 8.0-12.8), and 13.5% (95%CI 10.8-16.2), respectively. C-statistics of seven contemporary bleeding risk scores ranged between 0.54 and 0.59 indicating poor discriminatory performance (Table 1). The HAS-BLED score showed the highest C-statistics of 0.59 (95% 0.53-0.56). Similarly, all four machine learning-based predictions models performed poorly in internal validation (C-statistics ranging from 0.49-0.55, Figure 1).

Conclusion(s): Existing bleeding risk scores and a machine learning approach including common clinical parameters fail to assist in bleeding risk prediction of patients on HD. Therefore, new approaches, including novel biomarkers, to improve bleeding risk prediction in patients on HD are needed.