What is the clinical evidence from this study?

Study design: Observational. Population: Venous thromboembolism (VTE) without active cancer (n=74398). Intervention: VTE-PREDICT risk score. Primary outcome: Recurrent VTE and clinically relevant bleeding (non-major and major) at 5 years (C-statistics 0.48-0.71 (recurrence) and 0.61-0.68 (bleeding)).

What does this research mean for the field?

The VTE-PREDICT risk score predicts 5-year risks of recurrent venous thromboembolism (C-statistics 0.48-0.71) and clinically relevant bleeding (C-statistics 0.61-0.68) to quantify individual benefits and harms of extended anticoagulation. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop and externally validate prediction models estimating 5-year risks of recurrent venous thromboembolism (VTE) and clinically relevant bleeding to inform extended anticoagulation decisions in patients without cancer.

August 14, 2026European Heart JournalOpen Access

Recurrent venous thromboembolism and bleeding with extended anticoagulation: the VTE-PREDICT risk score

View Full Paper

Ask AI

Bookmark

Key result

VTE-PREDICT score shows modest discrimination for 5-year recurrent VTE and bleeding in external validation.

C-statistics 0.48-0.71 (recurrence) and 0.61-0.68 (bleeding)
n=74,398

Why the study?

Deciding whether to stop or continue anticoagulation after initial treatment for VTE is challenging due to heterogeneous individual risks of recurrence and bleeding.

Key Points

To develop and externally validate prediction models estimating 5-year risks of recurrent venous thromboembolism (VTE) and clinically relevant bleeding to inform extended anticoagulation decisions in patients without cancer.
Derivation used individual patient data from five studies (n = 15,141; 220 recurrences, 189 bleeding events) using 14 clinical characteristics in competing risk-adjusted models.
External validation was conducted in five independent cohorts and trials (n = 59,257; 2,283 recurrences, 3,335 bleeding events) over up to 5 years of follow-up.
Absolute treatment effects were calculated by integrating model predictions with hazard ratios from trials and meta-analyses.
External validation demonstrated C-statistics ranging from 0.48 (95% CI: 0.45–0.52) to 0.71 (95% CI: 0.66–0.77) for recurrent VTE and 0.61 (95% CI: 0.54–0.67) to 0.68 (95% CI: 0.65–0.70) for bleeding.
In the Danish VTE cohort, predicted 5-year risks ranged from 4% to 19% for recurrent VTE and from 1% to 19% for bleeding.
Internal derivation achieved an overall C-statistic of 0.68 (95% CI: 0.65–0.72) for recurrence and 0.69 (95% CI: 0.67–0.72) for bleeding.

Study Design

Type

Observational (n=74,398)

Multicenter

Yes

Structured PICO

Population

74,398 patients with venous thromboembolism without active cancer who completed at least 3 months of initial anticoagulation, followed for up to 5 years.

Exposure

Extended anticoagulation (DOACs, VKAs, LMWH)

Comparator

No extended anticoagulation

Outcome

Time-to-first-recurrent VTE (objectively confirmed, fatal or non-fatal recurrent DVT or PE, or death to which PE contributed) and time-to-first-bleeding event (composite of major bleeding and clinically relevant non-major bleeding)composite

Main Result

Effect estimate: C-statistics 0.48-0.71 (recurrence) and 0.61-0.68 (bleeding)

The VTE-PREDICT risk score provides a validated tool to estimate individual 5-year risks of recurrent VTE and bleeding to guide shared decision-making on extended anticoagulation.

Limitations

Variation in definitions of predictors and outcomes among the individual studies used for model development
Predictive performance of the recurrent VTE model varied
Deaths in which PE could not be ruled out were included as recurrent VTE events
Assumed relative treatment effects to be constant over time, across studies and settings, and homogenous among all patients
Antiplatelet therapy status was registered only at the end of the initial treatment

Key expert perspectives

Captured external expert commentary on this paper, strongest first. Original sources are linked where available.

MN Mathilde NijkeuterInternist-vascular medicine specialist, UMC Utrecht

“A thrombosis is a blockage of a blood vessel by a blood clot. If a clot clogs a vein in a leg, we call it a thrombosis leg. If this clot breaks loose, it enters the lungs and is called a pulmonary embolism. Pulmonary embolism and thrombosis are two different manifestations of the same disease: venous thromboembolism (VTE). Treatment for thrombosis consists of blood thinners for at least three months. After that, doctor and patient together make a choice to stop blood thinners or continue treatment for life.”

UMC UtrechtNews Coverage

MD Maria de WinterResearcher, UMC Utrecht / Heidelberg University

“The risk of serious bleeding and a new thrombosis is different for everyone and depends on risk factors, such as age and gender. Guidelines recommend weighing these risks, but only average risks are known for the average patient. This leads to a dilemma: How should patient and physician jointly make a choice if individual risks are not known?”

UMC UtrechtNews Coverage

JD Jannick DorresteijnInternist-vascular physician, UMC Utrecht

“Through the tool, the physician calculates the risk of a new thrombosis and severe bleeding. This information helps the patient and physician to decide together about stopping or continuing blood thinners after a thrombosis. The tool is based on a computational algorithm and for this purpose aggregated large data sets, big data, from studies conducted in the past were used.”

UMC UtrechtNews Coverage

Implication

Limited discrimination in external cohorts cautions against routine clinical use; leaves open whether refined models can improve individualized anticoagulation decisions.

Abstract

AIMS: Deciding to stop or continue anticoagulation for venous thromboembolism (VTE) after initial treatment is challenging, as individual risks of recurrence and bleeding are heterogeneous. The present study aimed to develop and externally validate models for predicting 5-year risks of recurrence and bleeding in patients with VTE without cancer who completed at least 3 months of initial treatment, which can be used to estimate individual absolute benefits and harms of extended anticoagulation. METHODS AND RESULTS: Competing risk-adjusted models were derived to predict recurrent VTE and clinically relevant bleeding (non-major and major) using 14 readily available patient characteristics. The models were derived from combined individual patient data from the Bleeding Risk Study, Hokusai-VTE, PREFER-VTE, RE-MEDY, and RE-SONATE (n = 15,141, 220 recurrences, 189 bleeding events). External validity was assessed in the Danish VTE cohort, EINSTEIN-CHOICE, GARFIELD-VTE, MEGA, and Tromsø studies (n = 59 257, 2283 recurrences, 3335 bleeding events). Absolute treatment effects were estimated by combining the models with hazard ratios from trials and meta-analyses. External validation in different settings showed agreement between predicted and observed risks up to 5 years, with C-statistics ranging from 0.48-0.71 (recurrence) and 0.61-0.68 (bleeding). In the Danish VTE cohort, 5-year risks ranged from 4% to 19% for recurrent VTE and 1% -19% for bleeding. CONCLUSION: The VTE-PREDICT risk score can be applied to estimate the effect of extended anticoagulant treatment for individual patients with VTE and to support shared decision-making.

Discussion

No takes yet. Share an insight, caveat, or question.

Figures from the paper

Published by European Heart Journal · CC BY-NC

Cite This Study

Winter et al. (2023) conducted an observational in Venous thromboembolism (VTE) without active cancer (n=74,398). VTE-PREDICT risk score was evaluated on Recurrent VTE and clinically relevant bleeding (non-major and major) at 5 years (C-statistics 0.48-0.71 (recurrence) and 0.61-0.68 (bleeding)). The VTE-PREDICT risk score predicted 5-year risks of recurrent VTE and bleeding with C-statistics ranging from 0.48-0.71 and 0.61-0.68, respectively, in external validation cohorts.

synapsesocial.com/papers/6a7ea94fec5eed42c5a78cfe https://doi.org/10.1093/eurheartj/ehac776

Topics

Anticoagulation in AF

View Full Paper

Ask AI

Bookmark

Implication

Limited discrimination in external cohorts cautions against routine clinical use; leaves open whether refined models can improve individualized anticoagulation decisions.

Recurrent venous thromboembolism and bleeding with extended anticoagulation: the VTE-PREDICT risk score

Key Points

Study Design

Structured PICO

Main Result

Limitations

Key expert perspectives

Implication

Abstract

Discussion

Figures from the paper

Cite This Study

Topics

Recurrent venous thromboembolism and bleeding with extended anticoagulation: the VTE-PREDICT risk score

Key Points

Study Design

Structured PICO

Main Result

Limitations

Key expert perspectives

Implication

Abstract

Discussion

Figures from the paper

Cite This Study

Keep reading

Topics

Keep reading