stopping anticoagulation after first DVT

one pretty common primary care issue is the safety of stopping oral anticoagulant therapy (OAC) in patients who have a first, unprovoked venous thromboembolism (VTE). the risk of recurrent VTEs is variable in the studies (5-27% in first year, then 2-4%/yr). A Canadian study was just released looking at 410 adults with first unprovoked VTE who completed 3-7 months of OAC (see dvt stop anticoag AIM 2015 in dropbox, or Ann Intern Med. 2015;162:27-34​).  OAC was discontinued if the D-dimer were negative on therapy, and not restarted if D-dimer were still negative after 1 month. they assessed recurrent VTE over 2.2 years.

baseline: mean age 51, BMI 37, 45% with DVT and 55% with PE, and not much difference in baseline characteristics between men and women. women were divided into 2 groups: those with VTE who were on estrogen therapy (who then stopped the estrogens), and those who were not on estrogen therapy.

results:

    --78% had 2 negative D-dimers and did not restart OAC. of those with an initially normal D-dimer, 15% became positive 1 month after stopping OAC, including 2 with recurrent VTE within that one month period

    --overall rate of recurrent VTE was 6.7% per patient-year: 9.7% in men and 5.4% in women who had VTE not associated with estrogen therapy and 0% in women with VTE associated with estrogens (of interest, the % of men and women who had 2 negative D-dimers was the same -- it was not that men had higher D-dimer levels than women; also the fact that no women who had initial VTE while on estrogens had recurrence, suggesting that those on estrogens should not be considered to have an unprovoked VTE)

    ​--adverse events: in those put back on OAC during follow-up, the rate of major bleeding was 2.3% per patient-year. 9 patients were diagnosed with cancer during follow-up.

    

A few issues. clearly, there are significant risks and benefits at stake. continuing OAC for life after an unprovoked VTE is a major medicalization of the patient, committing them perhaps to decades of OAC and their very real likelihood of adverse effects. stopping OAC could lead to life-threatening consequences. a few other studies shed some light on this issue. 

another Canadian study found that in 646 people with a first unprovoked VTE treated with OAC for 6 months and then followed 4 years, there were 91 confirmed recurrences, with a 9.3% recurrence rate. Men had a 13.7% annual risk, women a 5.5% annual risk. women with 0-1 risk factors (edema/redness/venous stasis changes of leg, D-dimer >250 while on warfarin, BMI>30, or age >65) had an annual risk of 1.6%,  and those with >1 risk factor had annual risk of 14.1% -- see dvt length of anticoag in dropbox, or CMAJ 2008;179(5):417-26.

an Italian study looked at serial D-dimers to see if OAC could be discontinued, where 1010 patients <70 years old with unprovoked first VTE, given at least 3 months of OAC, were followed for 2 years with D-dimer measurements -- if the D-dimer was negative while on therapy, OAC was stopped and D-dimer was rechecked at 15, 30, 60, and 90 days, with suggestion to resume OAC at the time any D-dimer was positive. in 528 (52.3%) the D-dimer was persistently negative, and there were 25 recurrences (3.0% per patient-year). of the 482 with a positive D-dimer, 373 resumed OAC (where there were 4 subsequent VTEs, or 1.1%) and 109 refused. in the latter group 15 patients had a recurrence (8.8% per patient-year). major bleeding occurred in 2.3% per patient-year in those who resumed anticoagulation. there was no difference in recurrences between men and women (see dvt serial to stop anticoag blood 2014 in dropbox, or Blood. 2014;124(2):196-203).

so, what is one to do???

--my sense is that we should speak with the patient to see how they value the different outcomes, including the medicalization of continued OAC but decreased likelihood of recurrent VTE. the stakes are higher in those with a PE, since they are more likely to have a PE as the recurrent VTE.

--i'm not sure how to synthesize the conflicting data on male/female differences. i think the Italian study is pretty informative (and did not have a gender difference), given the large numbers of people involved and the impressive results. for the past several years i have been checking D-dimer levels prior to stopping OAC, then a month later (as in the first study). at this point, i will suggest to patients to follow the Italian study, with D-dimer checks prior to stopping OAC, then at 2 weeks , 1 month, 2 months, and 3 months. it was interesting in this Italian study that D-dimer positivity was highest at 15 days (20.8%), and lower thereafter (13.1% at 30 days, 6.6% at 60 days, and 3.6% at 90 days) -- as a side note one of my female patients with a negative D-dimer on OAC did have a recurrent PE 3 weeks off the OAC and before i checked the 1 month D-dimer, so I suspect I would have picked up a high D-dimer at 15 days and restarted the OAC).