Samuel Z. Goldhaber, MD


August 13, 2015

This feature requires the newest version of Flash. You can download it here.

D-dimer Disappoints

Hello. This is Dr Sam Goldhaber for the Clot Blog at on Medscape, recording from the American College of Cardiology Scientific Sessions in San Diego. Today our topic is "D-dimer Disappoints."

As clinicians who take care of patients with DVT [deep venous thromboembolism] and pulmonary embolism, we know that there is a high rate of recurrence after anticoagulation is discontinued. Generally, we will treat our patients with 3-6 months of anticoagulation and then face the question and dilemma: Should we continue extended-duration anticoagulation or should we stop and take our chances?

If we stop anticoagulation, the rate of recurrence within 10 years of cessation of anticoagulation is actually 50%. One out of 2 patients will have a recurrence if the original DVT or pulmonary embolism was idiopathic and unprovoked. If the original DVT or pulmonary embolism was provoked, let's say by surgery or by pregnancy, the risk for recurrence within 10 years of cessation of anticoagulation is about 20%, or 1 in 5. Either way, it's a high rate of recurrence. We think that's probably because venous thromboembolism (VTE) is a chronic illness like coronary disease or diabetes.

DODS Study: VTE Recurrence Rate With Negative D-dimer

There is a Canadian study,[1] which was recently published, which had the idea of testing what has been a very common practice: namely, come to the conclusion of anticoagulation (usually 3-6 months of anticoagulation); stop anticoagulation for a month and check the D-dimer; if the D-dimer is negative—which is thought to mean that the anticoagulation system is very calm and quiet and that a recurring VTE is unlikely—then call up the patient and say, "You know, you don't need to go back on anticoagulation because your risk for recurrence is very low."

This practice was tested formally in patients with VTE. Men and women got their 3-6 months of anticoagulation, the anticoagulation was discontinued, and the D-dimer was obtained.

Now, in men with a negative D-dimer rate, the recurrence was surprisingly high: 8% per year in the first year. That continued to be a total of 16% recurrence in 2 years—clearly unacceptable. That is in the patients who had a negative D-dimer. For the men who had a positive D-dimer and who did not restart their anticoagulation, the recurrence rate was actually double.

What about women? In women in the first year after stopping anticoagulation with a negative D-dimer, the lack of anticoagulation was problematic because their recurrence rate was 5% in the first year and 10% within 2 years, with a negative D-dimer. If they had an elevated D-dimer, the recurrence rate was doubled to 10% the first year and then 20% in the first 2 years.

What we find is that having a positive D-dimer doubles the risk for a recurrent VTE if anticoagulation is stopped, but having a negative D-dimer really does not confer immunity against a recurrent VTE after anticoagulation is discontinued. We really can't reliably use a negative D-dimer to reassure our patients that they're free of the chance of a recurrence. I think that an 8%/year recurrence rate in men and a 5%/year recurrence rate in women are unacceptably high.

Idiopathic or Provoked?

What I think is the most important predictor of recurrence is whether the initial DVT or pulmonary embolism is idiopathic or provoked. If it's idiopathic with an overall likelihood of recurrence of 50% within 10 years, I generally encourage my patients to consider indefinite duration anticoagulation. If the initial DVT or pulmonary embolism is provoked, let's say due to surgery, I often recommend to my patients that at least they take low-dose aspirin to help reduce, to some extent, the risk for recurrence. It won't do as big a job as an anticoagulant, but it's better than taking nothing at all.

In summary, I think we should discard the D-dimer as a way to make a decision about how long to treat a patient with anticoagulation after an initial DVT or pulmonary embolism. We need to consider clinical factors, whether the DVT or pulmonary embolism was provoked or unprovoked, and, of course, take patient preference into consideration.

This is Dr Sam Goldhaber, signing off for the Clot Blog.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: