unprovoked VTE: high risk of bleeding with longterm anticoag

 Though the general approach to people with unprovoked venous thromboembolism has been for extended anticoagulation, a recent systematic review and meta-analysis suggested that long-term anticoagulation may make the risk of major bleeding outweigh the benefit (see dvt bleeding risk anticoag after unprov VTE AIM2021 in dropbox, or doi:10.7326/M21-1094) 

 

Details: 

-- 14 RCTs and 13 prospective cohort studies of patients with unprovoked venous thromboembolism (VTE) were evaluated for major bleeding with extended anticoagulation. Overall, low risk of bias in these chosen studies 

-- 70,202 patients: 9982 patients were receiving a vitamin K antagonist (VKA, such as warfarin/coumadin and its derivatives) and 7220 received a direct oral anticoagulant (DOAC, though most of the studies used rivaroxaban) 

    -- all patients had had an unprovoked or minimally provoked VTE (the latter in those with mild transient VTE risk factors), were on a VKA with an INR goal of 2-3 for a minimum of six additional months beyond completion of at least three months of initial anticoagulation, and these studies reported major bleeding during extended anticoagulation (the researchers obtained additional information from the investigators beyond what was reported in the studies themselves) 

--a major bleeding event was as defined by the individual studies or by International Society on Thrombosis and Hemostasis (ISTH) criteria: overt bleeding associated with a decrease of hemoglobin of at least 20 g/L, requirement for transfusion or at least of at least two units blood, occurring in a critical site (intracranial, intraspinal, intraocular, pericardial, intra-articular, intramuscular with compartment syndrome, or retroperitoneal) or is fatal or contributes to death 

-- primary outcome: the incidence of major bleeding during extended anticoagulation for up to 5 years 

-- also reported was longer term data from studies with 10-year evaluation

 

Results: 

-- the incidence of major bleeding per 100 person-years: 

    -- VKA: 1.74 events (1.34-2.20) 

    -- DOAC: 1.12 events (0.72-1.62) 

 

-- 5-year cumulative incidence of major bleeding: 

    -- VKA: 6.3% (3.6%-10.0%) 

    -- DOAC: insufficient data to evaluate the incidence beyond 1 year after discontinuing the med 

 

-- subgroups with significantly higher incidence of major bleeding: 

    

    -- of note in the subgroups:

        -- essentially all of the groups (VKA and DOAC) had increased bleeding rate if not on an anticoagulant

        -- comparing the relative risk of bleeding, there was a statistically increased risk using DOACs vs VKAs in those with a history of bleeding and concomitant antiplatelet therapy, 

-- case fatality rate of major bleeding: 

    -- VKA: 8.3% (5.1%-12.2%) 

    -- DOAC: 9.7% (3.2%-19.2%) 

 

-- no significant different in the incidence of major bleeding when analyzing the RCTs vs prospective cohort studies 

-- no difference in patients with initial isolated proximal DVT vs those with isolated PE; between those with initial isolated proximal DVT and those with concomitant PE and DVT; or those with isolated PE vs those with concomitant PE and DVT 

-- no difference in outcomes if the initial anticoagulant treatment was only 3 months 

-- no difference if using the definition of major bleed by the studies or by the ISTH criteria 

 

Commentary: 

-- the 2016 guidelines from the American College of Chest Physicians and the 2020 guidelines from the American Society of Hematology do suggest continuing anticoagulation indefinitely after a first unprovoked proximal DVT or PE, except in those people considered to have a high risk of major bleeding. Some specialists have suggested not considering indefinite anticoagulation regardless of their risk of recurrent VTE if there is a major risk of bleeding (at least 3% per year)

-- the population incidence of VTE is 1 to 2 cases per 1000 persons/yr, mostly unprovoked or associated with minor transient risk factors (i.e. weakly provoked) 

-- general agreement is of at least 3 to 6 months of initial anticoagulant therapy, where the risk of a fatal recurrent VTE exceeds the potential of a fatal major bleed from anticoagulation 

-- a recent meta-analysis of 7515 patients, by the same authors as in the bleeding meta-analysis above, found that for a first unprovoked or weakly provoked VTE with only the initial 3+ months of anticoagulation (ie no further anticoagulation), there was a 10% recurrent VTE risk at 1 year; with a risk of 25% at 5 years, and 36% at 10 years. there was also a 4% risk of death from a recurrent VTE event (see VTE risk after disc anticoag BMJ2019 in dropbox, or doi:10.1136/bmj.l4363). the anticoagulant used in all of these studies was apparently warfarin 

    --overall extended anticoagulation is associated with > 80% decreased risk of recurrent VTE if anticoagulation is continued 

 

-- the above review on bleeding found in their overall analysis that there was no significant difference between VKA or DOAC as anticoagulants, though there was a nonsignificant trend favoring the DOACs. However, in their subgroup analyses, there were statistically significant differences: those with a history of bleeding did better with VKAs, those with concomitant use of antiplatelet therapy did better with VKAs; and there were pretty strong favorable trends for VKAs in those with hemoglobin levels <100 g/L and in those >65 years old, and for DOACs for women

 

--the researchers offered a helpful case study based on the above numbers: 

    -- a patient with first unprovoked VTE who had completed at least 3 months of anticoagulation, and either has a history of bleeding or receiving concomitant antiplatelet therapy, may not be a candidate for anticoagulation with the VKA (not enough long-term data for DOACs): discontinuing the anticoagulation would put the 5-year risk of death from a recurrent VTE at about 1.0% (25% risk for recurrent VTE at 5 years multiplied by the 4% case fatality rate of recurrent VTE); if anticoagulation is continued, risk of death or major bleed at 5 years will be >1.2% (>15% risk for bleeding at 5 years multiplied by 8% case fatality rate of major bleeding). And over a 10-year horizon, the risk of death from recurrent VTE if anticoagulation was discontinued would be 1.4% (36% risk for recurrent VTE at 10 years multiplied of the 4% case fatality rate of recurrent VTE) and the risk of death from major bleed if anticoagulated would be >2.4% (>30% risk at 10 years multiplied by the 8% case fatality rate from a major bleed) [this longer-term data for up to 10 years is from the researchers' earlier paper noted above on risk of recurrent VTE after discontinuing anticoagulation, which decreases over time:        

     -- but, from their previous study cited above on VTE risks if stopping anticoagulation, there was significantly more harm in the first year (10.3%), vs at 5 years (25.2%, an average of 5%/year), vs at 10 years (36.1%, an average of 3.6%/yr). they did not calculate a specific number of years when major bleeding risks of stopping anticoagulation starts to outweigh the benefits (and we would need data of benefits and risks for each individual year of continued anticoagulant therapy vs stopping it, since I assume that there is an annual and nonlinear change in the risks of bleeding as was found in the potential benefits). And would be great to have more granular modeling: which type of anticoagulant to use for those at higher risk, as identified above.

 

Limitations: 

-- for VKA meds, there were no data about frequency of testing INRs or the consistency of how often those results were within the goal range, which might affect the VKA bleeding risk 

-- not enough data on DOACs (only 1 year) to assess long-term effects in this study, all limiting the interpretation of long-term risks/benefits with these drugs 

    --and, since many clinicians now are using DOACs, this is really important, since the benefit with VKAs decreases over time. If that were not true for DOACs, that would change the risk calculation 

    --also, do these results apply to other DOACs besides rivaroxaban? A recent study comparing warfarin with DOACs suggested that in those with atrial fibrillation on DOACs had less bleeding, except for rivaroxaban (see https://www.ahajournals.org/doi/10.1161/CIRCOUTCOMES.120.007230 ). and rivaroxaban was the predominant one used in the above analysis

        -- though, despite some studies to the contrary, I am still hesitant to use DOACs in people who travel a lot to their native countries where reversal agents are not available. They can bring vitamin K with them pretty easily…  and this might also apply to some areas of the US… 

-- the researchers did have access to more information on these studies than was printed in the studies themselves, but they did not do an individual patient data meta-analysis. So, as with meta-analyses in general, they are combining studies with different inclusion and exclusion criteria, different demographics of patients, different adherence to anticoagulants, and it is hard to know whether their conclusions in any way reflect the individual patient sitting in front of us 

 

So, this analysis is helpful in primary care, since it strongly suggests that the benefit/risk ratios of continuing anticoagulation for life may change dramatically over time with these meds. there clearly need to be long-term high-quality studies assessing DOACs (breaking down different ones, since the outcomes might be different), as well as more information on long-term VKAs (assessing the risks associated per time in the therapeutic range, which might differentiate the benefit/risk assessment for different patients on VKAs). Unprovoked or weekly provoked VTEs are really common and the benefits of early anticoagulation are clear, but the lack of strong data on the risks in terms of major bleeding make it difficult to have an informed discussion with patients about this complex treatment decision. so the real benefit of this study is that it (hopefully) spurs these important further studies since there are such profound consequences of jumping to life-long anticoagulation as in the current guidelines.... and, this issue may apply to other conditions requiring long-term anticoagulation (atrial fibrillation, for example)

 geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list