Short-Course DOAC After TAVR Clears Leaflet Thrombosis Early — But the Benefit Disappears by 12 Months

Subclinical leaflet thrombosis — hypoattenuating leaflet thickening, or HALT, visible on cardiac CT but not causing symptoms — has become an intensely studied phenomenon in the TAVR literature. Since CT imaging after TAVR became routine, it has been clear that a substantial minority of patients develop HALT in the months after implantation. What remains contested is whether HALT matters clinically — does it predispose to stroke? Does it cause valve deterioration? — and whether anticoagulation prevents or reverses it. NOTION-4, presented at ESC 2026 from Danish centers, provides the cleanest randomized data yet on the anticoagulation question, and the answer is nuanced.[1]

Three hundred fifty-two patients at two high-volume Danish TAVR centers were randomized to one of two antithrombotic strategies after TAVI: either lifelong single antiplatelet therapy (SAPT) alone — the current standard for patients without an independent indication for anticoagulation — or a 3-month course of a DOAC (direct oral anticoagulant) followed by lifelong SAPT. Both groups underwent CT imaging at 3 months and 12 months to quantify HALT prevalence. The investigators were led by Jorgensen TH, De Backer O, and Sondergaard L, the Copenhagen structural heart group that has been at the forefront of TAVR antithrombotic research.[1]

At 3 months, the DOAC group had significantly lower HALT prevalence — consistent with what prior observational data and the GALILEO trial had suggested about anticoagulation clearing leaflet thrombus. But the critical finding is the 12-month assessment. Once the DOAC was stopped at 3 months and both groups were on SAPT alone, the HALT prevalence converged: there was no significant difference between groups at 12 months. The short-course DOAC cleared thrombus while it was being taken, but the effect did not persist.[1]

Clinical Context

The clinical importance of HALT remains genuinely uncertain. The GALILEO trial (2019) showed that a rivaroxaban-based strategy after TAVR reduced HALT prevalence but increased net adverse clinical events, driven by bleeding — and it was stopped early for harm.[2] The lesson many took from GALILEO was that for patients without an intrinsic indication for anticoagulation (such as AFib), DOACs after TAVR do more harm than good. NOTION-4 refines that picture: a 3-month course may clear HALT without exposing patients to 12 months of bleeding risk, but the HALT returns once the DOAC stops. This raises the uncomfortable possibility that unless HALT has durable consequences — valve deterioration years later — the clinical benefit of clearing it temporarily is unclear.

The significance of HALT for bioprosthetic valve durability is still being worked out. CT-detected HALT can be accompanied by reduced leaflet motion (RELM) on imaging, and some studies have linked RELM to impaired hemodynamics and potentially accelerated valve degeneration. Others have shown that many episodes of HALT resolve spontaneously on antiplatelet therapy. Until there is a randomized trial powered for valve durability or stroke outcomes rather than CT HALT prevalence, NOTION-4's findings are important but incomplete.

Why It Matters Clinically

For structuralists, NOTION-4 is the latest data point in a long debate about what to prescribe after TAVR. The current ESC and ACC/AHA positions converge on SAPT for patients without AFib, with some variability in the guidance about dual antiplatelet therapy in the first 3–6 months. NOTION-4 does not change that guidance substantially — it confirms that anticoagulation clears HALT but does not durably prevent it, and that the clinical implications of HALT remain to be definitively established.

Limitations

Single-country, two-center design limits generalizability. CT HALT is an imaging surrogate endpoint — clinical outcomes (stroke, valve deterioration) were not powered. The 3-month DOAC duration was chosen pragmatically, and longer or different regimens were not tested. HALT burden and its clinical significance may vary by valve type (intra-annular vs. supra-annular designs, different leaflet thicknesses). Patients with an independent indication for anticoagulation were excluded.