After TAVR, a DOAC cleared the leaflets and cost more events

Design

randomised controlled trial, superiority design, 12-month CT endpoint

Population

347 analysed patients after successful TAVR without an indication for oral anticoagulation

Primary outcome

prevalence of hypoattenuated leaflet thickening on cardiac CT at 12 months

Effect

28.3% vs 32.2% at 12 months (risk difference −3.9%, 95% CI −14.4 to 6.6, P = 0.54); death/stroke/major bleeding 8.2% vs 2.3%

NOTION-4 randomised 352 patients without an indication for oral anticoagulation shortly after successful transcatheter aortic valve replacement to either lifelong single antiplatelet therapy or three months of a direct oral anticoagulant followed by lifelong single antiplatelet therapy. After screen failures and withdrawals, 176 and 171 patients respectively were analysed, with hypoattenuated leaflet thickening on cardiac CT at 12 months as the primary endpoint.

At three months, while the anticoagulant was still running, leaflet thickening was present in 12.1% of the DOAC group against 31.8% on antiplatelet therapy. By 12 months, nine months after the drug stopped, that difference had gone: 28.3% versus 32.2%, risk difference −3.9% (95% CI −14.4 to 6.6), P = 0.54. The trial was powered for superiority and did not find it.

The clinically important number is the other one. The combined risk of all-cause death, stroke or major or life-threatening bleeding at 12 months was 8.2% in the DOAC group against 2.3% with antiplatelet therapy alone — a risk difference of 5.9% (95% CI 1.2 to 10.6). A course of anticoagulation bought a temporary improvement in a CT appearance and was accompanied by more of the events that matter.

• Do not start a time-limited DOAC after TAVR to prevent leaflet thickening
• Continue single antiplatelet therapy where there is no independent anticoagulation indication
• Treat incidental leaflet thickening on surveillance CT as a finding to follow, not to anticoagulate
• Reassess any local protocol that routinely images leaflets after TAVR without a plan for what a positive result changes

Why it matters

The strategy improved the picture and worsened what the picture was supposed to stand in for.

Don't overread it

This was powered for an imaging endpoint; the excess of death, stroke and bleeding is a secondary observation in a modest trial.

The statistics, in plain English

The primary endpoint interval of −14.4% to 6.6% spans zero and both directions, so the trial cannot distinguish benefit from harm on leaflet thickening at one year. The safety difference is the more reliable signal because its interval, 1.2 to 10.6, excludes zero — though with 347 patients and few events, the size of that harm is imprecise. Leaflet thickening is an imaging surrogate, not an outcome patients experience.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.