Pros, cons of prolonged anticoagulation after a clot

Although the discussion was friendly, there was a lot of provocation involved when two experts debated extending anticoagulation after venous thromboembolism (VTE) at the American Heart Association's 2019 Scientific Sessions, held in Philadelphia in November.

Specifically, they disagreed about whether VTEs provoked by any factor other than major surgery or trauma should be treated in the same way as unprovoked cases. Samuel Z. Goldhaber, MD, director of the thrombosis research group at Brigham and Women's Hospital in Boston, argued for yes.

“VTE is mostly a chronic illness like diabetes or coronary artery disease,” said Dr. Goldhaber, who is also a professor of medicine at Harvard Medical School. “With the exception of postop patients who develop VTE, we should not tell our patients that we can cure your VTE with time-limited anticoagulation.”

To support his case, he turned to data from the Framingham Heart Study, published in Thrombosis Research in 2016, which followed almost 10,000 patients after a VTE. “Over the course of 15 years, their event-free survival—no PE [pulmonary embolism] and no VTE—was exactly the same among those with unprovoked VTE and those with provoked VTE,” he said. Cancer patients made up a separate category with lower survival, he noted.

For evidence that treatment effects, as well as risk, are similar regardless of provocation, he turned to the EINSTEIN-CHOICE trial results, published in the New England Journal of Medicine in 2017. Patients in that industry-funded study, 60% of whom had provoked VTE, were randomized to six to 12 months of either of two doses of rivaroxaban or aspirin.

Both doses of the newer drug were superior to aspirin in reducing recurrent VTEs. “But most importantly, they caused a 70% reduction in provoked VTE patients, as well as in unprovoked VTE patients,” said Dr. Goldhaber.

A prespecified analysis of the trial classified patients by the factors that may have provoked their original clots. The minor persistent risk factors included the common problems of heart failure, obesity, chronic kidney disease, family history of VTE, and hereditary thrombophilia. Immobility was considered a minor transient risk factor.

For these patients, treatment with rivaroxaban was associated with a threefold reduction in recurrent VTE, according to results published April 9, 2018, in Blood Advances. “Among patients not receiving extended-duration anticoagulation, the annual risk of recurrent VTE was at least 7% in VTE provoked by minor persistent or transient risk factors,” said Dr. Goldhaber.

Based on findings like these, the European Society of Cardiology (ESC) recommended ending the dichotomy in their latest PE guidelines, published by the European Heart Journal on Aug. 31, 2019. “Terminology such as ‘provoked’ vs. ‘unprovoked’ PE/VTE is no longer supported by the guidelines, as it is potentially misleading and not helpful for decision making regarding the duration of anticoagulation,” quoted Dr. Goldhaber.

Instead, the guidelines recommend classifying patients by their risk for VTE recurrence if they don't continue anticoagulation: low, intermediate, or high. “The only patients in the low group were those who had VTE after major surgery or major trauma,” he said.

The intermediate group (thought to have a recurrence risk between 3% and 8%) includes patients whose VTE was provoked by minor surgery, hospitalization for medical illness, pregnancy, estrogen use, a long flight, and inflammatory bowel or autoimmune disease.

This is also the category for those patients who used to be described as having had an unprovoked VTE. “We now call it no-identifiable-risk-factor VTE,” said Dr. Goldhaber. He and two colleagues argued for these changes even before the ESC guidelines came out, writing an editorial that was published by Circulation on Dec. 3, 2018.

“We suggested placing the unprovoked VTE and the provoked VTE into the same funnel and having a duration of anticoagulation that was proportional to the risk of recurrence,” said Dr. Goldhaber. “Provoked VTE and persistent VTE risk factors do not receive the respect they deserve.”

However, his debate opponent, Teresa L. Carman, MD, offered her respect for these risks. “When these patients are in my office, I will tell them you are never no-risk. Once you've identified yourself as having a DVT [deep venous thrombosis] or pulmonary embolism, I can never take your elevated risk away,” said Dr. Carman, who is an assistant professor at Case Western Reserve University and director of vascular medicine at University Hospitals Cleveland Medical Center in Ohio.

But she also expressed her concern about the risks of long-term anticoagulation. She noted that 2012 CHEST guidelines on antithrombotic therapy for VTE recommend risk stratification based on bleeding risk when deciding whether to treat patients long term.

“In patients with low bleeding risk, maybe extending secondary prophylaxis is warranted. If there's high bleeding risk, those are not the patients that you're going to want to subject to that risk,” she cited.

The guideline's focus is echoed in many algorithms for anticoagulation after VTE, such as the one published by Blood Advances on Nov. 13, 2018. “The first thing in most algorithms is consider the risk of bleeding,” said Dr. Carman. “We have to pay attention to that in all our patients.”

Attention should also be given to individual patients' circumstances and preferences, she argued. “How is this impacting their lifestyle? How's it impacting their job? How's it impacting the things they want to do like ride a bike, go skiing … all the other things that may not be safe, if you will, on anticoagulation?”

Patient characteristics should also factor into anticoagulation decisions, Dr. Carman added, offering the example case of a 27-year-old woman who had a PE after a long flight. “To tell her you're going to put her on anticoagulation for life—that's a long, long, long time,” she said.

That span of anticoagulation carries significant financial cost, said Dr. Carman, who offered the estimate of 60 years. “Using today's values, that's enough to put her through medical school,” she said. For some patients, cost may justify consideration of aspirin therapy, which is “not antiquated” despite having been proven less effective than rivaroxaban, according to Dr. Carman.

Another important consideration is that recurrence risk varies with age, sex, and type of clot, she noted. Five-year recurrence rates differed among patients who had PE (22%), proximal DVT (26.4%), or calf DVT (7.6%), according to a meta-analysis published in the Journal of Thrombosis and Haemostasis in 2010.

There are also tools for estimating patients' recurrence risk. “I find D-dimer to be one of those useful biomarker risk stratification tools that we need to employ,” said Dr. Carman. “We know that if patients have a negative D-dimer 30 days after coming off anticoagulation, their risk is lower.”

If the D-dimer stays low, then VTE risk does too, but if it goes up, so does the patient's risk, which might lead to reconsideration of anticoagulation. “We have opportunities for intervention. We have opportunities for discussion,” she said.

In discussions with patients who have had a VTE, physicians should note that it “may be a chronic condition,” Dr. Carman acknowledged in her conclusion. “But I think we have to know the magnitude of that elevated risk and be able to manage these patients appropriately,” she said. “We can use patient characteristics, biomarkers, and VTE characteristics to help us with that.”

Her closing advice foretold closer accord between the debaters in the future. At the end of his talk, Dr. Goldhaber had noted that decision making about anticoagulation after VTE should eventually become more precise, thanks to advances in genetic screening, mathematical modeling, and biomarkers.