Patient Education Blog

New Clots in Spite of Being on Warfarin – “Warfarin Failure”

Stephan Moll, MD writes…

Occasionally, new blood clots (DVT or PE) develop in spite of a patient being on warfarin (Coumadin, Jantoven). This happens particularly in patients with (a) fluctuating INRs (i.e. when the INR is low at times), (b) the clotting disorder lupus anticoagulant or antiphospholipid antibody (APLA) syndrome or (c) cancer. But it also happens in others. When a new DVT or PE occurs (referred to together as venous thromboembolism or VTE) in spite of adequate warfarin therapy, health care professionals talk about a “warfarin failure”.

A.  How common is it?

Warfarin failure occurs in approximately:

    • 2-3 % of patients with previous unprovoked VTE over 1-4 years of treatment with warfarin [ref 1].
    • 17 % of patients who had a VTE due to cancer over 6 months of treatment with warfarin [ref 2].
    • Up to 11 % of patients with VTE due to lupus anticoagulant and antiphospholipid antibody syndrome over 2 ½ years of treatment with warfarin [ref 3].

B.  Figuring out why warfarin failed a patient

a)      Getting a radiological imaging study

If there is fairly strong suspicion for a new DVT or PE, it is appropriate to get a Doppler ultrasound of the extremity or CT scan of the chest, respectively, to verify that the symptoms are truly due to a new clot [ref 4]. One cannot simply assume that new symptoms are due to a new clot. For example: increased leg swelling and pain after a DVT may be due to post-thrombotic syndrome and more than usual physical activity rather than due to a new clot.

b)      Did the Doppler of CT or VQ truly show a new clot?

Diagnosing “recurrent DVT or PE” can be a major challenge [ref 4], and the consequences can be profound. Doppler ultrasounds can be difficult to perform and read, particularly when (a) examining the calf veins, as these are relatively small, (b) performing the study on a patient who is quite overweight, and (c) if there has been previous DVT in the leg. Previous DVT often leads to some chronic changes of the veins, and it can be difficult to differentiate between old clot and new clot. It is always helpful for the health care professional to have an old Doppler ultrasound for comparison, so that one can hopefully tell whether clot seen on the ultrasound is new since the previous study, or was already present. But even when having a comparison study it can still be difficult. Your health care professional clearly wants to make sure that a good and experienced Doppler technologist performed the study and a good technologist and radiologist interpreted the study.

Obtaining a blood D-dimer test can be another helpful piece of information: if the D-dimer is positive it is one additional reason to believe that the patient does have a new clot [ref 4]; however, the D-dimer results should not be overvalued and should not be the only reason to conclude that a patient does or does not have a new clot.

c)      What were the INRs in the preceding weeks?

The health care professional will need to review the quality of the warfarin management of the preceding weeks by looking at the INRs. May be the INRs were low (below the typical 2-3 range) and a clot formed at that time.

d)      Has the patient been non-compliant?

It is important to find out whether the patient may have interrupted the warfarin therapy for some time prior to the clot or stopped taking it altogether.

e)      Does the patient have a lupus anticoagulant?

This is a clotting disorder, which can be detected by doing a blood test. While it can be found in patients with lupus (a rheumatologic disease), it can also be present in other patients with blood clots. The presence of a lupus anticoagulant can lead to inaccurate INR values: An INR may look like it is satisfactory (for example a value of 2.8), but in the patient with a lupus anticoagulant that INR value may actually not indicate adequate blood thinning.

f)       Does the patient have cancer?

It is well known that warfarin is not optimally effective in patients with cancer, i.e. that clots can form in spite of the patient being on warfarin with an INR of 2-3.  Therefore, in the patient with a new clot in spite of adequate warfarin therapy, the suspicion that the patient may have a cancer is raised, and a thorough investigation (additional scans or blood work) for cancer may be appropriate.

C.  Treatment options

Once the reasons listed above have been considered as potential causes for warfarin failure, a new treatment plan can be made. If the patient truly had a “warfarin failure”, then the treatment options are:

    1. Continue warfarin, but increase the target INR (to 3.0-4.0, for example)
    2. Switch to a different blood thinner
      • Low molecular weight heparin (LMWH) injections (Lovenox = enoxaparin; Fragmin = dalteparin)
      • Arixtra (fondaparinux) injections – once daily
      • Xarelto (rivaroxaban), the new oral blood thinners FDA approved for DVT and PE treatment
      • Pradaxa (Dabigatran) or Eliquis (apixaban), not FDA approved at this point for DVT and PE treatment
    3. Add aspirin to the blood thinner

Only in patients with cancer is LMWH known to be better (i.e. more effective) than warfarin in preventing further clots. However, in non-cancer patients, no data exist that any of these 3 choices would be better (more effective) than warfarin. It is also not known whether any one of the 3 choices listed above would be better than the other. Thus, the best approach is not known.

Placing an IVC (inferior vena cava) filter into the big vein in the abdomen is not a solution, as the filter does not prevent new blood clots from forming; it just prevents clots that break of from a DVT to get to the lung to cause a PE.

D.  Personal Approach

  1. Warfarin failure: My typical approach to patients with “warfarin failure”, once I have considered the causes of warfarin failure discussed above, is to consider twice daily low molecular weight heparin; after 3 months, once the acute clotting event has been appropriately treated, I may discuss a switch in management, such as (1) fondaparinux injections into the skin because of the convenience of once daily dosing, (2) a switch to warfarin with a higher target INR if the patient has a history of fairly stable INRs, or (3) may be a switch  to one of the new oral anticoagulants. However, at this point, while we are just learning how Xarelto performs in the real world, i.e. outside of clinical trials, I would be very hesitant to put a patient with true warfarin failure that was not due to fluctuating INRs on one of the new oral blood thinners.
  2. New oral anticoagulant failure: In the studies that compared Xarelto (rivaroxaban) to warfarin in the treatment of DVT and PE, new clots occurred in both treatment groups equally often: 2-3 % of patients had a new clot over a 1 year treatment period [ref 5,6]. If a patient who is being treated with one of the new oral anticoagulants has clearly documented new clot, then a treatment change seems appropriate. I would choose and discuss with the patient a switch to warfarin, long-term. However, it is not known how effective this strategy is in preventing further clots.

E.  Summary

  • Some (few) people develop new clots despite being on warfarin.
  • Determining why this happens requires investigation and possibly lab tests.
  • Alternative blood thinning medications are available.

F.  References

  1. Kearon C et al. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin for long-term prevention of recurrent venous thromboembolism. N Engl J Med. 2003 Aug 14;349(7):631-9.
  2. Crowther MA et al. A comparison of two intensities of warfarin for the prevention of recurrent thrombosis in patients with the antiphospholipid antibody syndrome. N Engl J Med. 2003;Sep 18;349(12):1133-8.
  3. Lee AY et al. Low-molecular weight heparin versus coumarin for the prevention of recurrent venous thromboembolism in patients with cancer.  N Engl J Med. 2003 Jul 10;349(2):146-53.
  4. Bates SM et al. Diagnosis of DVT: Antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. CHEST 2012; 141(2)(Suppl):e351S–e418S
  5. Buller HR et al. Oral rivaroxaban for the symptomatic venous thromboembolism. N Engl Med 2010;363:2499-510.
  6. Buller HR et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl Med 2012;366:1287-97.

F.  Support Forum: Ask questions, share information and connect with other patients on Clot Connect’s support forum.


Disclosure: I have consulted for Janssen, Boehringer-Ingelheim, and Daiichi.

Last Updated: 3/4/2013

Tags: ,

This entry was posted by clotconnect on at and is filed under Blood thinners (anticoagulants), Therapy. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

Comments are closed.