For patients with valvular heart disease, when are non‑vitamin K antagonist oral anticoagulants (NOACs) appropriate and when should a vitamin K antagonist (warfarin) be used?

NOACs in Valvular Heart Disease: When to Use and When to Avoid

For patients with atrial fibrillation and valvular heart disease, NOACs should be used instead of warfarin in most cases, with two critical exceptions: patients with rheumatic mitral stenosis and those with mechanical prosthetic valves must receive warfarin. 1

Clear Indications for NOACs (Class I Recommendation)

NOACs are the preferred anticoagulant for patients with atrial fibrillation and the following valvular conditions, based on CHADS-VASc score 1, 2:

• Native valve disease (excluding rheumatic mitral stenosis) - including aortic stenosis, aortic regurgitation, mitral regurgitation, and tricuspid valve disease 1
• Bioprosthetic valves implanted more than 3 months ago (surgical or transcatheter) 1, 2
• Prior valve repair with prosthetic material 1

The evidence supporting NOACs in these populations comes from large randomized trials that included patients with valvular heart disease and demonstrated similar efficacy for stroke prevention with reduced intracranial bleeding compared to warfarin 3, 4. Meta-analyses confirm NOACs reduce stroke/systemic embolism (HR 0.70) and intracranial hemorrhage (HR 0.47) compared to warfarin in patients with native valvular disease 3.

Absolute Contraindications for NOACs (Class III: Harm)

Warfarin is mandatory in these two scenarios 1, 5:

1. Mechanical Prosthetic Heart Valves

• All mechanical valves (any position, any type) require warfarin indefinitely 1, 5
• NOACs are contraindicated due to demonstrated harm in the RE-ALIGN trial with dabigatran 1
• Target INR varies by valve type and position:
  • Bileaflet valve in aortic position: INR 2.5 (range 2.0-3.0) 5
  • Tilting disk or bileaflet in mitral position: INR 3.0 (range 2.5-3.5) 5
  • Caged ball/disk valves: INR 3.0 (range 2.5-3.5) plus aspirin 75-100 mg daily 5

2. Rheumatic Mitral Stenosis

• Long-term warfarin is required regardless of severity 1, 2, 5
• Target INR 2.5 (range 2.0-3.0) 5
• This applies even with mild stenosis if rheumatic in etiology 1

Special Consideration: Early Post-Bioprosthetic Valve Period

For the first 3 months after bioprosthetic valve implantation (surgical or transcatheter), warfarin is reasonable over NOACs (Class IIa recommendation) 1:

• Target INR 2.5 (range 2.0-3.0) 5
• After 3 months, transition to NOAC based on CHADS-VASc score if atrial fibrillation present 1
• Recent trials support NOAC safety in bioprosthetic valves, including transcatheter valves 6, 7

Critical Pitfalls to Avoid

Do not confuse "valvular atrial fibrillation" terminology - this outdated term has created confusion 6. The FDA label for warfarin mentions "non-valvular AF" but this does NOT mean all valvular disease requires warfarin 5. Only mechanical valves and rheumatic mitral stenosis are true contraindications to NOACs 1, 6.

Do not assume all mitral stenosis requires warfarin - only rheumatic mitral stenosis is a contraindication 1. Degenerative or calcific mitral stenosis can be treated with NOACs 1.

Do not extrapolate mechanical valve data to bioprosthetic valves - these are entirely different populations with different thrombogenicity 6, 7.

Algorithm for Decision-Making

1. Does the patient have a mechanical prosthetic valve?

   • Yes → Warfarin (INR based on valve type/position) 1, 5
   • No → Proceed to step 2

2. Does the patient have rheumatic mitral stenosis?

   • Yes → Warfarin (INR 2.0-3.0) 1, 5
   • No → Proceed to step 3

3. Does the patient have atrial fibrillation with other valvular disease?

   • Native valve disease → NOAC based on CHADS-VASc 1
   • Bioprosthetic valve <3 months → Consider warfarin 1
   • Bioprosthetic valve >3 months → NOAC based on CHADS-VASc 1

The evidence consistently demonstrates that NOACs provide equal or superior outcomes compared to warfarin in appropriate valvular heart disease populations, with the critical advantage of reduced intracranial bleeding risk 3, 4.