This article is more than 5 years old. Information may no longer be current.
Several factors determine duration of anticoagulation for VTE
Click Here to Manage Email Alerts
As a coagulationist with a special interest in thrombosis, thrombophilia and anticoagulation, I occasionally receive email inquiries like the following:
“Hi, Dr. Moll. I am seeking your valued opinion: 56-year-old man; 2013 R leg DVT. Heterozygous FV Leiden. Long-term warfarin? Thanks! Kind regards, Dr. ...”
I am always stunned that some health care professionals think a decision about length of anticoagulation could be made on such sparse data. The question above to the hematologist is similar to asking an oncologist: “Seeing a 56-year-old man with lung cancer. What treatment do you recommend? Chemotherapy?”
Stephan Moll
In both cases, details are needed.
Decision-making about length of anticoagulation in patients with venous thromboembolism requires having several pieces of information, as the decision depends on three factors (Figure 1): the risk for recurrent VTE; the risk for a major or life-threatening bleed; and a patient’s personal values and preferences with regard to treatment.
Location of VTE, risk factors
I encourage health care professionals to clearly define each venous thrombotic episode that a patient has had in respect to anatomy (ie, Was it a superficial thrombophlebitis or a deep vein thrombosis? Was it a distal or proximal DVT?), as well as the VTE risk factors associated with each thrombotic event in an “a, b, c” fashion as shown in Figure 1, as both play a role in the assessment of the risk for recurrent VTE.
Anatomical location of thrombosis matters.
Source: Images courtesy of Stephan Moll, MD.
Examples: Patients with asymptomatic distal leg DVT may not require any anticoagulation. Patients with symptomatic distal leg DVT typically only require 3 months. Patients with unprovoked proximal DVT may require long-term anticoagulation.
Also, the circumstances of the VTE episode matter. A patient with a VTE associated with a major transient risk factor — such as major surgery — is treated with 3 months of anticoagulation because of a low risk for recurrence if off blood thinners (3% during 5 years), whereas the patient with the same type of DVT — if unprovoked (idiopathic) — may be considered for long-term anticoagulation because of a high risk for recurrence (up to 30% during 5 years).
Bleeding risk
As the primary risk associated with long-term anticoagulation is bleeding, an assessment of bleeding risk with continued anticoagulation is the second pillar of the decision process about how long to treat.
Existing formal bleeding risk scores only modestly predict major bleeding in patients with VTE, partially because these scores were developed in the atrial fibrillation population, not VTE patients. Thus, common-sense bleeding risk assessment by the clinician is typically needed and seems appropriate: Has the patient had significant bleeding on anticoagulation? Have the international normalized ratios fluctuated a lot? Does the patient have extreme INR values at times? Has the patient had serious falls, or is he/she at risk for such? Is the patient elderly or does he/she have comorbid diseases?
I always encourage health care professionals to list all bleeding risk factors in an “a, b, c” fashion as shown in Figure 1.
‘Warfarin Hate Factor’
Finally, a patient’s personal preference plays an important role in the decision about how long to treat with anticoagulation for an episode of VTE.
I use a “Warfarin Hate Factor” (Figure 1) — or, now that the non-vitamin K oral anticoagulants are in clinical use, a “Blood Thinner Dislike Factor” scale — for that assessment.
I ask the patient to give me a number on a scale from 0 to 10 to indicate how much he or she “hates” being on warfarin, taking into consideration INR fluctuations, the need for monitoring and the inconvenience of physician office visits for INR testing, the dietary restrictions, the effect of the anticoagulant on the patient’s lifestyle (with respect to profession and hobbies), and the cost of the drug and INR testing.
A hate factor of 0 of 10 means: “I don’t mind at all being on warfarin. It’s just a pill I have to take.” A value of 10 means: “I hate warfarin so much, I absolutely want to come off. I am not worried about another clot.”
This Warfarin Hate Factor question often opens up a floodgate in the discussion with the patient, bringing out what he or she really thinks about being on warfarin or one of the novel anticoagulants. If the patient’s hate factor is high because of dietary restrictions or cumbersomeness of monitoring, then one can discuss switching from warfarin to a novel anticoagulant if long-term anticoagulation is indicated. However, if the hate factor is high because of the fear from bleeding or impact on professional or recreational activities while on an anticoagulant, then such a switch would, obviously, not remedy that issue.
‘Risk of Recurrence Triangle’
For my discussion with a patient, I use a “Risk of Recurrence Triangle” (Figure 2) to come to a decision about whether we should treat with long-term anticoagulation.
I use the American College of Chest Physicians’ 2012 guidelines as the major evidence-based building block for this triangle.
A patient who had a VTE associated with a major transient risk factor has a low risk for recurrence once anticoagulation is stopped after 3 months; thus, the patient is at the top — or in the tip — of the triangle. Patients with unprovoked VTE have a higher risk for recurrence and are in the lower part, or the broad base of the triangle.
Men with unprovoked VTE have the highest risk for recurrence. Women with unprovoked nonhormone-associated VTE are still in the lower part of the triangle, but above where men fall.
Finally, patients with VTE associated with minor risk factors (eg, contraceptives, estrogen-replacement therapy or pregnancy, minor surgery, or minor immobility such as long-distance travel) appear to be at intermediate risk for recurrence (ie, the middle part of the triangle).
It is this intermediate-risk patient population in whom I obtain a thrombophilia workup and a D-dimer while on anticoagulation. If the thrombophilia workup does not show a strong thrombophilia (Table 1) and a D-dimer test is negative, then we stop anticoagulation and recheck a D-dimer (and test for protein C and S activities, which should not be tested while on warfarin) 4 weeks later. If the D-dimer test again is negative, then we continue without anticoagulation. However, if the D-dimer on anticoagulation is clearly positive or turns positive 4 weeks after having come off the anticoagulant, we continue/restart anticoagulation.
The horizontal dashed line (Figure 2) indicates the approximate level above which the risk for dying of a recurrent VTE is lower than the risk for dying of bleeding on anticoagulation (ie, a balance of the case–fatality in favor of stopping anticoagulation). Thus, patients above the line are the ones in whom one would stop anticoagulation after the initial 3 months of treatment.
However, case–fatality from bleeding for patients on long-term warfarin for secondary VTE prevention, as well as for patients on the novel anticoagulants, is not accurately known. The position of the horizontal dashed line is not the same in every patient and is not necessarily static over time; it depends on an individual’s bleeding risk. It is moved up for the patient at low risk for bleeding and moved down for the patient at high risk for bleeding, as the case–fatality rate from bleeding changes and, thus, the acceptance to treat with long-term anticoagulation.
Similarly, although not because of a change in case–fatality rate: In the patient with a strong preference not to be on anticoagulation (ie, a high Warfarin Hate Factor), the line would move down (ie, the patient would find a risk for recurrence off anticoagulation acceptable, that a person with a low Warfarin Hate Factor would not find acceptable). In a patient with a low Warfarin Hate Factor, the line would move up.
Definition of ‘long term’
“Long-term” anticoagulation also is referred to as “extended” or — in everyday clinical practice terminology — as “indefinite” or “life-long” (ie, for years to come).
However, periodic re-evaluation of the patient is indicated, with (a) new risk–benefit assessment of anticoagulation; (b) discussion of new trial data; (c) updates on the status of available anticoagulants; and (d) follow-up discussion about a patient’s treatment preference (ie, a reassessment of where in the recurrence triangle the patient is and where the red line should be drawn).
Caveats about using a D-Dimer
Although it has been shown that a positive D-dimer on warfarin, as well as off warfarin, indicates a higher risk for recurrence, the D-dimer should only be considered for routine clinical use if the clinician knows (a) what type of D-dimer assay his/her clinical laboratory uses (is it a highly or moderately sensitive test, or an insensitive test?); and (b) what the cutoff level of D-dimer result is above which the test predicts a higher risk for recurrent VTE. If the clinician does not know this information, the D-dimer cannot be used safely for decision-making.
References:
Douketis J. BMJ. 2011;342:d813.
Kearon C. Chest. 2012;141:e419S-494-S.
Kearon C. Blood. 2014;123:1794-1801.
MacLean S. Chest. 2012;141:e1S-e23S.
Moll S. Hematology. 2014;297-305.
Riva N. Thromb Haemost. 2014;112:511-521.
For more information:
Stephan Moll, MD, is professor of medicine in the department of medicine and division of hematology-oncology at the University of North Carolina School of Medicine in Chapel Hill, N.C. He also is medical director of the Clot Connect patient education program (www.clotconnect.org), an initiative of the University of North Carolina Hemophilia and Thrombosis Center. He can be reached at UNC School of Medicine, Campus Box 7036, Chapel Hill, NC 27599; email: smoll@med.unc.edu.
Disclosure: Moll has been a consultant for CSL Behring, Daiichi and Janssen.