First randomized PROBE trial testing continued vs interrupted oral anticoagulation during TAVI; non-inferiority with potential superiority, results pending.

Background

Approximately one-third of patients undergoing transcatheter aortic valve implantation (TAVI) are on oral anticoagulation (OAC), mainly for atrial fibrillation, and face competing periprocedural risks of thromboembolism and bleeding. While guidelines generally advise interrupting OAC for high-bleeding-risk procedures, observational studies suggest that continuing OAC during TAVI may be safe and could reduce thromboembolic events. High-quality randomized evidence is lacking.

Patients

• Inclusion: Adults on OAC (vitamin K antagonist [VKA] or direct oral anticoagulant [DOAC]) undergoing transfemoral or transsubclavian TAVI who provide written informed consent.
• Exclusion (true interruption not acceptable):
  • Mechanical heart valve prosthesis
  • Intracardiac thrombus
  • Venous thromboembolism within 3 months before TAVI
  • Transient ischemic attack or stroke within 6 months in patients with atrial fibrillation
• Setting: Pragmatic, multicenter; 21 sites in the Netherlands, Belgium, Luxembourg, Denmark, Ireland, and Italy.
• Enrollment status: Recruitment began November 2020; as of May 2023, 625 enrolled; target 858 patients.

Intervention

Periprocedural continuation of OAC (VKA or DOAC): patients continue their prescribed OAC through the day of TAVI (VKA dosed to usual INR target). OAC may be interrupted only for clear clinical contraindications at clinician discretion. Other procedural aspects (valve type, heparin/protamine, vascular closure, embolic protection) per local protocol.

Control

Periprocedural interruption of OAC without bridging:

• DOACs: Stop 48 hours pre-TAVI; dabigatran adjusted for renal function (eGFR 50–80: 72 hours; eGFR 30–50: 96 hours).
• VKAs: Acenocoumarol 72 hours; phenprocoumon or warfarin 120 hours pre-TAVI.
• No LMWH bridging (high-thromboembolic-risk patients excluded; bridging increases bleeding without antithrombotic benefit in this setting).
• Restart OAC after TAVI when deemed safe (evening or next morning).

Outcome

• Primary endpoint (30 days, VARC-3): Composite of cardiovascular mortality, all stroke, myocardial infarction, major vascular complications, and type 2–4 bleeding.
• Key secondary endpoints:
  • Individual components of the primary endpoint at discharge and 30 days
  • VARC-3 clinical safety and efficacy composites
  • Quality of life (SF-12, KCCQ, TASQ) at baseline, 30, and 90 days
  • NYHA class, hospital readmission, and cost-effectiveness
• Biochemical substudy: Serial measures of coagulation/platelet activation markers and vWF/ADAMTS13 (baseline to ~6 weeks) to explore mechanisms and early markers (e.g., paravalvular leak).

Study Design

• Type: Multicenter, randomized, open-label with blinded endpoint adjudication (PROBE), non-inferiority with hierarchical superiority testing.
• Randomization: 1:1 continuation vs interruption; variable block sizes; stratified by OAC type (VKA vs DOAC) and site (via REDCap).
• Analysis population: Modified intention-to-treat (events from 5 days pre- to 30 days post-TAVI).
• Sample size: 858 patients (90% power; 1-sided alpha 0.025; non-inferiority margin 4% absolute risk difference). Anticipated primary event rates: 17.5% (interruption) vs 13.5% (continuation).

Level of Evidence

Level I (randomized controlled trial; ongoing, results pending).

Follow up period

• Primary endpoint: 30 days post-TAVI
• Additional follow-up: Discharge, 30 days, and 90 days for clinical outcomes and quality of life; biochemical substudy to ~6 weeks.

Results

• Primary outcome (planned): 30-day composite of cardiovascular mortality, all stroke, myocardial infarction, major vascular complications, and VARC-3 type 2–4 bleeding.
• Secondary outcomes (planned): Individual and composite VARC-3 endpoints; quality of life (SF-12, KCCQ, TASQ); NYHA class; readmissions; cost-effectiveness; mechanistic biomarker analyses.
• Current status: Ongoing; no outcome data reported at time of design publication.
• NNT: Not calculable (no results available).

Limitations

• Open-label treatment may introduce reporting/ascertainment bias, mitigated by blinded clinical endpoint adjudication and standardized VARC-3 definitions.
• Powered for a composite endpoint; may be underpowered for individual outcomes (e.g., major bleeding, ischemic stroke).
• Excludes highest thromboembolic-risk patients; limits generalizability to those populations.
• Local procedural practices (e.g., heparin/protamine use, device/closure strategy) left to site discretion; potential heterogeneity.
• mITT window excludes events before 5 days pre-TAVI; appropriate for regimen-related risk but may omit very early events unrelated to strategy.

Funding

Investigator-initiated; funded by the St. Antonius Research Fund and the Netherlands Organization for Health Research and Development. No industry involvement. Independent Data Safety Monitoring Board oversight; conducted per GCP, Declaration of Helsinki, and national regulations.

Citation

van Ginkel DJ, Bor WL, Dubois CLF, et al. Periprocedural continuation versus interruption of oral anticoagulant drugs during transcatheter aortic valve implantation: rationale and design of the POPular PAUSE TAVI trial. EuroIntervention. 2023;19:766-771. Published online ahead of print August 2023. doi:10.4244/EIJ-D-23-00206.