PAUSE TAVI
Continuation versus Interruption of Oral Anticoagulation during TAVI
Year of Publication: 2025
Authors: D.J. van Ginkel, W.L. Bor, H.M. Aarts, ..., and J.M. ten Berg
Journal: The NEW ENGLAND JOURNAL of MEDICINE
Citation: N Engl J Med 2025:392:438-49
Link: https://doi.org/10.1056/NEJMoa2407794
PDF: https://dsm.units.it/sites/dsm.units.it/...arPauseTAVI.pdf
PDF: Clinical Question
In patients receiving oral anticoagulants and planning to undergo transcatheter aortic-valve implantation (TAVI), is a strategy of periprocedural continuation of oral anticoagulation non-inferior to interruption for a composite of death from cardiovascular causes, stroke, myocardial infarction, major vascular complications, or major bleeding within 30 days?
Bottom Line
In patients undergoing TAVI with a concomitant indication for oral anticoagulation, periprocedural continuation of anticoagulation was not non-inferior to interruption with respect to the primary composite outcome at 30 days. This was driven by a higher incidence of bleeding complications in the continuation group with no significant difference in thromboembolic events.
Major Points
- One-third of patients undergoing TAVI have an indication for oral anticoagulation owing to concomitant diseases.
- This was an international, open-label, randomized, noninferiority trial.
- The primary outcome occurred in 16.5% of the continuation group and 14.8% of the interruption group, failing to meet the noninferiority criterion.
- Any bleeding occurred in 31.1% of the continuation group versus 21.3% in the interruption group (risk difference, 9.8 percentage points; 95% CI, 3.9 to 15.6).
- Thromboembolic events were similar between the groups, occurring in 8.8% of the continuation group and 8.2% of the interruption group.
- The findings provide evidence supporting periprocedural interruption of oral anticoagulation in patients undergoing TAVI.
Design
Study Type: International, investigator-initiated, open-label, randomized clinical trial with blinded outcome assessment.
Randomization: 1
Blinding: Open-label with blinded outcome assessment by a clinical-events committee.
Enrollment Period: November 2020 through December 2023.
Follow-up Duration: 30 days.
Centers: 22
Countries: Netherlands, Belgium, Denmark, Italy, Ireland, Luxembourg
Sample Size: 858
Analysis: The primary analysis was performed in the modified intention-to-treat population.
Inclusion Criteria
- Patients planning to undergo transfemoral or transsubclavian TAVI.
- Patients receiving long-term oral anticoagulants.
- Patients who had provided written informed consent.
Exclusion Criteria
- Presence of a mechanical heart valve prosthesis.
- Intracardiac thrombus.
- Venous thromboembolism within 3 months before TAVI.
- Transient ischemic attack or stroke in patients with atrial fibrillation within 6 months before TAVI.
Baseline Characteristics
| Characteristic | Control | Active |
|---|---|---|
| Age-yr | 80.9±6.2 | 81.4±5.6 |
| Female sex-no. (%) | 138 (32.3) | 158 (36.7) |
| Median body-mass index (IQR) | 26.9 (24.3-30.8) | 26.5 (24.2-29.7) |
| Score on the EuroSCORE II-% | 3.9±4.3 | 3.8±3.9 |
| NYHA class I-no. (%) | 15 (3.5) | 11 (2.6) |
| NYHA class II-no. (%) | 146 (34.2) | 152 (35.3) |
| NYHA class III-no. (%) | 238 (55.7) | 241 (55.9) |
| NYHA class IV-no. (%) | 28 (6.6) | 27 (6.3) |
| Atrial fibrillation-no. (%) | 406 (95.1) | 414 (96.1) |
| Paroxysmal Atrial fibrillation-no./total no. (%) | 184/406 (45.3) | 192/414 (46.4) |
| CHA2DS2-VASc score | 4.4±1.4 | 4.5±1.4 |
| Hypertension-no. (%) | 322 (75.4) | 339 (78.7) |
| Diabetes (None)-no. (%) | 304 (71.2) | 303 (70.3) |
| Diabetes (Non-insulin-dependent)-no. (%) | 87 (20.4) | 90 (20.9) |
| Diabetes (Insulin-dependent)-no. (%) | 36 (8.4) | 38 (8.8) |
| Coronary artery disease-no. (%) | 206 (48.2) | 207 (48.0) |
| Previous CABG-no./total no. (%) | 72/206 (35.0) | 66/207 (31.9) |
| History of myocardial infarction-no. (%) | 75 (17.6) | 61 (14.2) |
| Previous cerebrovascular event (Transient ischemic attack)-no. (%) | 42 (9.7) | 42 (9.7) |
| Previous cerebrovascular event (Ischemic stroke)-no. (%) | 51 (11.9) | 39 (9.0) |
| Previous cerebrovascular event (Hemorrhagic stroke)-no. (%) | 4 (0.9) | 4 (0.9) |
| Previous cerebrovascular event (Undetermined stroke)-no. (%) | 3 (0.7) | 3 (0.7) |
| Peripheral artery disease-no. (%) | 85 (19.9) | 79 (18.3) |
| Chronic obstructive pulmonary disease-no. (%) | 49 (11.5) | 68 (15.8) |
| Chronic renal insufficiency-no. (%) | 221 (51.8) | 213 (49.4) |
| Previous aortic-valve surgery-no. (%) | 28 (6.6) | 36 (8.4) |
| Previous pacemaker implantation-no. (%) | 88 (20.6) | 75 (17.4) |
Arms
| Field | Control | Continuation Group |
|---|---|---|
| Intervention | Oral anticoagulation was interrupted before TAVI. Direct oral anticoagulants were stopped 48-96 hours prior, and vitamin K antagonists were stopped 72-120 hours prior. Bridging with heparin was not initiated. Oral anticoagulation was restarted after TAVI as soon as deemed safe by the physician. | Patients continued their oral anticoagulation regimen, including on the day of the TAVI procedure. |
| Duration | Periprocedural | Periprocedural |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| A composite of death from cardiovascular causes, stroke from any cause, myocardial infarction, major vascular complications, or major bleeding (VARC-3 type 2, 3, or 4) within 30 days after TAVI. | Primary | 14.8% (63/427) | 16.5% (71/431) | 0.18 for noninferiority | |
| Thromboembolic event at 30 days | Secondary | 8.2% (35/427) | 8.8% (38/431) | Risk Difference 0.6 (95% CI, -3.1 to 4.4) | |
| Cerebrovascular event (stroke or TIA) at 30 days | Secondary | 6.3% (27/427) | 6.5% (28/431) | Risk Difference 0.2 (95% CI, -3.1 to 3.5) | |
| Any bleeding | Adverse | 21.3% (91/427) | 31.1% (134/431) | Risk Difference 9.8 (95% CI, 3.9 to 15.6) | |
| Major bleeding (VARC-3 type 2, 3, or 4) | Adverse | 8.9% (38/427) | 11.1% (48/431) | Risk Difference 2.2 (95% CI, -1.8 to 6.3) |
Subgroup Analysis
The results for the primary outcome were generally consistent across prespecified subgroups. A potentially greater benefit from the interruption strategy was observed in patients with a history of coronary artery disease (Risk Difference 7.2, 95% CI 0.2 to 14.2) and those with high frailty (Edmonton Frail Scale score ≥5; Risk Difference 16.4, 95% CI 1.7 to 31.1). Confidence intervals were not adjusted for multiplicity.
Criticisms
- This was an open-label trial and was thereby potentially subject to reporting and ascertainment biases, although trial outcomes were adjudicated by a blinded clinical-events committee.
- The pragmatic nature of the trial did not include a neurologic examination or neuroimaging in all patients.
- The trial was powered for a composite primary outcome, so no clinical inferences should be drawn regarding the separate components.
- Almost all the patients enrolled were treated with the use of the transfemoral approach, so the results should not be generalized to other vascular-access approaches.
Funding
Netherlands Organization for Health Research and Development and the St. Antonius Research Fund.
Based on: PAUSE TAVI (The NEW ENGLAND JOURNAL of MEDICINE, 2025)
Authors: D.J. van Ginkel, W.L. Bor, H.M. Aarts, ..., and J.M. ten Berg
Citation: N Engl J Med 2025:392:438-49
Content summarized and formatted by NeuroTrials.ai.