Skip to main content
ISTH 2023 Main banner
Back

    Poster Networking

    Venous Thromboembolism

    Poster Networking Session

    PB0504 - Clinical usefulness of risk tools predicting recurrence of unprovoked venous thromboembolism: a systematic review and meta-analysis

    Sunday, June 25, 2023
    18:30 – 19:30 ET
    Room: Exhibition

      Co-Author(s)

    • SB

      Soerajja Bhoelan, MD

      MD, PhD student
      Department of Haematology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands
      Groningen, Groningen, Netherlands

      • Co-Author(s)

    • Tessa Elling, MD

      PhD student
      Department of Hematology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
      Groningen, Groningen, Netherlands

      • Presenting Author(s)

    • Karina Meijer, MD, PhD (she/her/hers)

      MD, PhD, Internist-haematologist, Professor
      Department of Hematology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
      Groningen, Groningen, Netherlands

      • Co-Author(s)

    • Jv

      Jasper van Miert

      MD, PhD
      Department of Haematology, University of Groningen, University Medical Centre Groningen
      Groningen, Groningen, Netherlands

    Background: The clinical usefulness of risk tools identifying patients with unprovoked venous thromboembolism (VTE) who could discontinue anticoagulation therapy is unclear.

    Aims: To determine calibration, discrimination and usefulness for clinical decisions of existing risk tools

    Methods: Pubmed and Embase were searched on December 1st, 2021, for risk tools in patients with unprovoked VTE.
    Calibration was evaluated by regression of observed vs expected recurrence from validation respectively derivation studies.
    C-statistics were calculated from contingency tables and pooled in a random-effects model.
    Recurrence risks for each score were pooled in a Bayesian model. We calculated post-test probabilities that patients were low-risk for recurrence, as defined by the ISTH (i.e. risk < 5% at 1 year and < 15% at 5 years, with the upper boundary of the 95% confidence interval < 8% and < 24%, respectively).

    Results: Eight original studies were included, 5 evaluating DASH and 4 HERDOO2. In both, a score ≤1 was considered low-risk. Reported recurrence risks are displayed in figure 1.
    Calibration of DASH showed overestimation of recurrence risks >10%, regardless of follow-up. HERDOO2 was similarly calibrated with 1-year follow-up, but not for 2-year follow-up; risks < 10% were underestimated.
    Pooled c-statistics for DASH were 0.63(95% CI 0.59-0.67), 0.67(0.64-0.71) and 0.69(0.64-0.74) for 1-, 2- and 5-year follow-up, respectively. Pooled c-statistics for HERDOO2 were 0.60(0.59-0.60) and 0.64(0.63-0.64) for 1- and 2-year follow-up, respectively.
    DASH score 1 provides only a 31% (1-year) and 15% (5-year follow-up) probability of being ISTH low-risk. For DASH 0, this was 51% and 76%, respectively. DASH 1 was more likely to extend over the upper risk border.
    Low HERDOO2-scores were truly low-risk at 1-year follow-up.

    Conclusion(s): At the designed cut-off ≤1, DASH provides insufficient certainty that a patient is truly low-risk; instead a cut-off ≤0 identifies patients who could safely discontinue anticoagulation therapy.
    A HERDOO2 ≤1 satisfactorily identifies low-risk patients.