What are the data to guide predicting risk of recurrence and hence need for possibly indefinite duration of anticoagulation?  The Figure shows the factors to be considered, from a paper by Dr. John Heit from the Mayo Clinic,[13] listing male gender, increasing age, obesity, a neurologic disease with leg paresis, active cancer certainly, lupus anticoagulant or antiphospholipid antibody, and antithrombin, protein C, or protein S deficiency.  If the patient is homozygous for factor V Leiden or has a combination of genetic factors, for example, heterozygous for factor V Leiden coupled with the prothrombin gene mutation, then there is also increased risk.  There is still some question about the utility of persistently elevated D-dimer levels and residual venous thrombosis after a full course of treatment.  

However, even after taking account of all these risk factors, the bottom line is that clinicians still do not have a full appreciation as to what is happening at the level of the endothelium and what is it that keeps driving these clots to recur.  As a result, it is becoming more widespread in clinical practice to treat unprovoked DVT with really extended durations of therapy.   Hylek E. Am J Med 2013; published on-line at http://education.amjmed.com/00000.