Catheter ablation can be performed safely while a patient is on uninterrupted warfarin


By Stephen Page and Richard Schilling

Electrophysiologists have been instinctively hesitant to perform invasive procedures in patients anticoagulated with warfarin. In fact, the risks of bleeding are overestimated and catheter ablation can be performed safely in many patients, provided that any risks of bleeding are balanced against the risks of thromboembolism in each case.

Catheter ablation for typical right atrial flutter is associated with a risk of thromboembolism and requires periprocedural anticoagulation. The risks of bleeding are largely related to vascular access as cardiac perforation is rare in this situation. Conventionally, warfarin was stopped with bridging low molecular weight heparin (LMWH) but we have previously shown that catheter ablation can be performed safely on uninterrupted warfarin, and that uninterrupted warfarin is associated with fewer bleeding complications related to vascular access, and is also cost-effective (LMWH is expensive; warfarin is cheap)1. This approach avoids a potentially hazardous period of sub-therapeutic anticoagulation while warfarin is being restarted and, although no data exist, may help to reduce the risk of periprocedural stroke.

Catheter ablation for atrial fibrillation (AF) is associated with a risk of periprocedural stroke of up to 1%2 and also requires periprocedural anticoagulation. Bleeding risks are also higher and associated with vascular access, transeptal puncture, and cardiac perforation of the atrial wall. There is now a large body of evidence showing that the risk of bleeding is lower in patients undergoing ablation on uninterrupted warfarin than on a bridging LWMH regime.

This is largely due to a reduction in bleeding related to vascular access3. There is also increasing experience of managing bleeding complications in patients with a therapeutic international normalised ratio. Prothrombin complex is expensive but provides rapid normalisation of the prothrombin time if required. In fact, vascular access bleeding can usually be managed with heparin reversal alone, and in our experience, significant pericardial effusions can be managed with insertion of a pericardial drain and heparin reversal without the need to reverse the warfarin with prothrombin complex. A recent meta-analysis has demonstrated the safety and efficacy of uninterrupted warfarin4 and the recent guidelines for catheter ablation for AF reflect this with 50% of Task Force members performing AF ablation without stopping warfarin5.

Ablation of the compact atrioventricular node, focal atrial tachycardias, slow pathway and accessory pathway ablation are not associated with a significant increased thromboembolic risk and periprocedural warfarin is not specifically required. There are no safety data for performing catheter ablation in these situations for individuals who happen to be on warfarin.

It seems reasonable that if the risk of thromboembolism is low, warfarin may be stopped and restarted after the ablation. If the risk of thromboembolism is felt to be high, the procedure can be performed on uninterrupted warfarin.

There are less data available for catheter ablation of ventricular tachycardia. Many patients, especially those with structural heart disease, will require arterial access for monitoring, percutaneous ventricular assist or to reach the left ventricle via the aorta. Furthermore, some patients may require pericardial access. Efforts to avoid perform ventricular tachycardia on warfarin should usually be made particularly when the substrate is likely to be in the more fragile right ventricle, although there will be occasions when an emergency ablation procedure is required and the risks of warfarin reversal/cessation will need to be assessed depending on planned access, and likely need for pericardial access etc.

In summary, catheter ablation can be performed safely in most patients with uninterrupted warfarin. The data are now compelling for performing catheter ablation for both right atrial flutter and AF on uninterrupted warfarin. Periprocedural anticoagulation for catheter ablation for other arrhythmias should be determined on a case-by-case basis but, in general, the risks of bleeding are low on uninterrupted warfarin and the decision to stop warfarin should be considered carefully.

Stephen Page and Richard Schilling are both at the Department of Electrophysiology, St Bartholomew’s Hospital, UK.


1. Finlay et al. J Cardiovasc Electrophysiol. 2010; 21:150–54

2. Di Biase et al. Circulation. 2010; 121: 2550–56

3. Page et al. J Cardiovasc Electrophysiol. 2011; 22:265–70

4. Santangeli et al. Circ Arrhythm electrophysiol. 2012; 5:302–11

5. Calkins et al. Heart Rhythm. 2012; 9:632–96