TAVI is moving into younger and healthier patients, which makes valve durability matter more than it used to. ACASA-TAVI randomised 360 patients aged 65 to 80 at the three Norwegian centres that perform most of the country's TAVI to 12 months of monotherapy with either a factor Xa inhibitor or aspirin. The design used co-primary endpoints: efficacy was hypoattenuated leaflet thickening on blinded core-lab four-dimensional cardiac CT at 12 months, and safety was a composite of VARC-3 bleeding, thromboembolic events and all-cause death.
Leaflet thrombosis was found in 16.2% on the NOAC against 28.6% on aspirin (risk ratio 0.55, 95% CI 0.37 to 0.82, p=0.004). The safety composite occurred in 7.5% against 10.6%, a risk difference of -3.3% (95% CI -9.5% to 2.8%), meeting non-inferiority comfortably. Mean age was 74.5 years and 37% were women; 336 of 360 completed.
The honest caveat is what the efficacy endpoint is. Hypoattenuated leaflet thickening is an imaging finding, not an event. Its link to valve failure, stroke or death is plausible and unproven, and a trial of 360 patients over a year cannot test it. What ACASA-TAVI establishes solidly is the safety half: a NOAC after TAVI in this age band did not cost bleeding, which is the objection that has kept anticoagulation off the table.
So this changes the conversation rather than the protocol. It is reasonable to consider a NOAC in a younger TAVI patient with no bleeding risk and a long expected valve life, and to say clearly that the reason is imaging rather than outcomes. It is not a reason to move a frail 85-year-old off aspirin.
- Consider NOAC monotherapy in younger TAVI patients where long valve durability is the priority.
- Be explicit with the patient that the benefit shown is on CT imaging, not on stroke or survival.
- Do not extrapolate to patients older or frailer than the 65 to 80 band studied.
- Bleeding was not increased, which removes the main historical objection to this strategy.
- Where a NOAC is already indicated for AF after TAVI, this supports monotherapy rather than adding aspirin.
The statistics, in plain English
Two things are being asked at once here, and they are judged differently. The efficacy comparison is a superiority test: a risk ratio of 0.55 with an interval of 0.37 to 0.82 excludes no difference. The safety comparison is a non-inferiority test, which asks only whether the new strategy is not unacceptably worse — the interval of -9.5% to 2.8% includes zero and stays within the pre-set margin. Passing a non-inferiority test is not the same as showing the NOAC was safer; it means a meaningful excess of harm has been ruled out at this sample size.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for cardiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free