I am surprised how often I have to explain why a patient needs to stay on warfarin because he or she has a mechanical heart valve. My husband (as a cardiologist) has also expressed concerns that many clinicians often forget why we keep patients on warfarin. So here’s why.

That’s because so far, the evidence from DOACs shows that they cause more harm than good.

DOACs are contraindicated in patient with mechanical heart valves. It’s written all over the product monographs for dabigatran, rivaroxaban, apixaban and edoxaban.

But it started with the RE-ALIGN study which compared dabigatran with warfarin in patients with mechanical heart valves. It was published in 2013 in New England Journal of Medicine.

Here’s quick summary:

Study Design: a prospective, randomized, phase 2, open-label trial evaluating the effect of warfarin and dabigatran

Inclusion criteria:
– Patients between the age of 18 and 75 years
– Undergoing implantation of a mechanical bileaflet valve in the aortic or mitral position or both (Population A) OR
– If they had undergone implantation of a mechanical bileaflet mitral vavle (with or without mechanical bileaflet aortic-valve replacement) more than 3 months before randomization (Population B)

Exclusion criteria:
-Previous prosthetic heart valve replacement; aortic root replacement, replacement of ascending aorta, concomitant bioprosthetic valve replacement or mechanical tricuspid or pulmonary valve replacement at the time of the index valve replacement surgery (defined as the surgery that had been conducted within 7 days of randomization for population A or at least three months ago for population B);
-Clinically relevant paravalvular leaks, endocarditis, complex congenital heart abnormality, acute coronary syndrome within one month, uncontrolled hypertension, recent emergency surgery, planned surgery or intervention within one month after randomization,
-Any history of hemorrhagic stroke, at high risk for bleeding, active hepatitis or abnormal liver function with persistently elevated ALT, AST or alkaline phosphatase more than 3x the upper limit of normal, creatinine clearance <40 mL/min,
-Patients with a clear indication for long-term dual antiplatelet therapy or oral anticoagulant therapy for other indications for which dabigatran is not approved, recent malignancy or radiation therapy, treatment with selected drugs that may interact with dabigatran, at risk of pregnancy or with known allergy to dabigatran or warfarin.

Study-Drug Regimen

Dabigatran – The starting dose of dabigatran was based on renal function, with an initial dose of 150mg po BID in patients with a CrCL less than 70mL/min, 220mg po BID in those with a CrCL 70-109mL/min and 300mg po BID in those with CrCL > 110mL/min. This is to achieve a trough plasma level of 50 ng/mL of dabigatran. Dose is further adjusted based on trough level as well as renal function.

Warfarin – For patients in the warfarin group, the target range for the INR was 2-3 in those deemed to be at low thromboembolic risk (patients who had a mechanical aortic valve with no additional risk factor) and 2.5-3.5 in those deemd to be at intermediate or high risk (patients who had a mechanical aortic valve with additional risk factors or a mechanical mitral valve).

Primary Outcome

  • Trough plasma level of dabigatran (to validate a new dosing regimen for the prevention of thromboembolic complications in patient with mechanical heart valves)
  • Additional efficacy and safety outcomes included stroke, systemic embolism, transient ischemic attack, valve thrombosis, bleeding, venous thromboembolism, myocardial infarction and death

This study was stopped early because of an excess of thromboembolic and bleeding events in the dabigatran group, as compared with the warfarin group. Most of the events occurred in population A (where patients were started dabaigatran within 7 days after valve surgery), with fewer events in population B (patients who had undergone valve implantation more than 3 months before randomization).

The authors offered the following possible explanations:

  1. Inadequate plasma levels of dabigatran. The trough plasma levels of dabigatran in population A were lower during the first few weeks after surgery. This may have allowed early formation of blood clots . Thromboembolic events also occurred in patients with higher trough levels, so drug levels (alone) cannot explain for the increase rate in thromboembolic events.
  2. Higher dabigatran doses led to unacceptably high bleeding rates. The correlation between high doses and high bleeding rates are very clear from the results. However, the authors suggest that giving dabigatran more frequently without increasing the total daily dose may offer some benefits (but this has not been tested yet).
  3. Dabigatran offers a different mechanism of action than warfarin. In patients with mechanical heart valve, coagulation activation and thrombin generation are induced by the release of tissue factor from damaged tissue during surgery, in which warfarin may be more effective than dabigatran. This is because warfarin inhibits the activation of both tissue factor induced coagulation (Inhibition of factor IX) and contact pathway-induced coagulation (e.g. inhibition of factor X). However, dabigatran is a thrombin inhibitor for which it is not designed to target the various areas within the coagulation pathway. In addition if the contact activation is intense, it may result in excessive thrombin generation that overwhelms dabigatran.

So here are my thoughts:

  • While dabagatran is a thrombin inhibitor, rivaroxaban, apixaban and edoxaban are factor Xa inhibitors. They are all grouped together as DOAC (direct oral anticoagulants). But they behave differently and have different mechanism of actions. It may be unfair to group them together and to apply the same contraindications and warnings. However, none of the factor Xa inhibitors have been studied adequately in patients with mechanical heart valves, so there is still a lack of adequate evidence to support their use in this population.
  • In DAWA Pilot study, dabigatran was compared with warfarin for patients with atrial fibrillation and bioprosthetic mitral or aortic valve replacement. The study dose of dabigatran was 110mg po BID (and compared to warfarin). This study was terminated prematurely due to low enrollment but preliminary results did not suggest dabigatran caused more thromboembolic or bleeding events. So maybe there is hope?
  • At this time – while patients with mechanical heart valve cannot take DOAC because of clear evidence of harm from the results in RE-ALIGN, the study didn’t focus on stable patients who have mechanical heart valve replacement from a long time ago. For these stable patients where the source and risk of thromboembolism is less likely from local tissue injury, I wonder if there is a role for DOACs.

So the main reason that DOACs cannot be used in patients with mechanical heart valves isn’t the lack of evidence. In fact, it is because of evidence of significant harm that is derived from Dabigatran experience.

Until we have more evidence to prove otherwise, we have to avoid all DOACs in patients with mechanical heart valves. However, this may not be the case for other valvular heart disease. As per 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation, a DOAC may be used in patients with atrial fibrillation and native aortic valve disease, tricuspid valve disease, or mitral regurgitation when anticoagulation is required.  

Thanks for reading and I hope I have shed some light into a common question I get in clinical practice.