Provoked clot recurrence varies by trigger

Recurrence after provoked venous thromboembolism was not uniformly low, especially after major transient triggers.
*Prospective cohort study; Level 2 (OCEBM).

Citation

Eichinger S, Eischer L, Kaider A, Heinze G, Quehenberger P, Kyrle PA. Risk of recurrence in patients with provoked venous thromboembolism: a prospective cohort study. Blood. Manuscript BLD-2026-034774R1.

Background

Provoked blood clots are often treated with short-term anticoagulation because recurrence is assumed to be low. This study tested whether recurrence differs by the strength and type of temporary trigger.

Patients

Adults with symptomatic leg deep vein thrombosis and/or pulmonary embolism who completed at least 3 months of anticoagulation. Exclusions included cancer, major inherited or immune clotting disorders, pregnancy-associated clots, muscle-vein thrombosis, or another reason for long-term anticoagulation.

Intervention

Stopping anticoagulation, with the initial clot classified as caused by major or minor temporary risk factors.

Control

Comparison between major and minor temporary provoking factors.

Outcome

Objectively confirmed symptomatic recurrent venous thromboembolism.

Follow-up Period

Median 23.9 months after stopping anticoagulation.

Results

Outcome Result
Overall recurrence at 24 months 9.5% (95% CI 6.0 to 14.0)
Major temporary trigger recurrence at 24 months 19.7% (95% CI 11.0 to 30.2)
Minor temporary trigger recurrence at 24 months 5.0% (95% CI 2.2 to 9.5)
Minor versus major trigger recurrence hazard HR 0.22 to 0.29 across adjusted models; all 95% CIs excluded 1.0
Hormone-associated clots in 81 women 0 recurrences

HR = hazard ratio.

The study was stopped early because recurrence after major provoking factors was unexpectedly high.

Limitations

Early termination left fewer participants and events than planned, limiting subgroup analyses. Single-center Austrian cohort, mostly White participants, may limit generalizability.

Funding

Funding not reported; authors declared no conflicts.

Clinical Application

Do not assume all provoked clots are low risk; consider longer anticoagulation after major triggers, while hormone-related clots appear very low risk.