The Nordic Aortic Valve Intervention Trial 4 — Prospective Study on the Impact of Different Anti-thrombotic Therapies on Subclinical Leaflet Thickening and Its Temporal Dynamics in Transcatheter Bioprosthetic Aortic Valves

doi:10.1016/j.jacc.2026.08.023

NOTION-4

Patient / Population Intervention / Exposition Comparaison Critère de jugement

In 352 patients after TAVI without an indication for anticoagulation, 3 months of a DOAC, then antiplatelet therapy did not reduce leaflet thrombosis on CT at 1 year compared with antiplatelet therapy alone.

N

352patients

Schéma

investigator-initiated, randomised, open-label, superiority trial at two Danish centres (Rigshospitalet Copenhagen, Aarhus University Hospital); early 3-month DOAC then single antiplatelet vs lifelong single antiplatelet therapy; 4D-CT and clinical assessment at 3 months, 1 year and 5 years

Critère

number of patients with at least one bioprosthetic aortic valve leaflet with HALT on 4-dimensional cardiac CT at 1 year after TAVI

Pertinence

2Important — l’un des piliers parmi d’autres.

RésultatHALT at 12 months (primary endpoint) 28.3% with 3-month DOAC vs 32.2% with SAPT (risk difference −3.9%, 95% CI −14.4 to 6.6; P = 0.54). At 3 months, on DOAC, HALT 12.1% vs 31.8%. All-cause death, stroke or major/life-threatening bleeding at 12 months 8.2% vs 2.3% (risk difference 5.9%, 95% CI 1.2 to 10.6).

Jørgensen TH, Larsen AF, Jensen JM, et al. Short-Term Anticoagulant Therapy and Subclinical Leaflet Thickening in Transcatheter Aortic Valves: The NOTION-4 Trial. J Am Coll Cardiol. 2026 Aug 30 (online first). doi:10.1016/j.jacc.2026.08.023 10.1016/j.jacc.2026.08.023

Discussion et critique

Three months of anticoagulation clears the leaflets while it is being taken and the effect is gone nine months after stopping — the thrombus comes back, so a short course buys nothing durable. Against that stands harm: death, stroke or major bleeding in 8.8% versus 2.3%, a difference the trial was not powered for but which points the same way as GALILEO. Two Danish centres and 352 patients, an imaging endpoint, and the honest reading is the presenters' own: this regimen does not belong in patients without another indication for anticoagulation.