Validating HERDOO2 – When is it Safe to Stop Therapy After an Unprovoked VTE?

Have you ever been asked about when to stop anticoagulation therapy after an unprovoked venous thromboembolism (VTE)?  If so, then you know that this decision is controversial.  According to the 2016 CHEST VTE Guidelines, at least 3 months of therapy is recommended for an unprovoked DVT or PE (Grade 1B).¹  Thereafter, the clinician is expected to weigh the risks and benefits to determine if extended therapy is appropriate.  For patients who are at low-to-moderate bleeding risk, indefinite therapy is suggested — but this is a weak Grade 2B recommendation.¹ The International Society on Thrombosis and Haemostasis (ISTH) suggests it is safe to stop anticoagulation if the risk of recurrent VTE is < 5% one year after discontinuing treatment.² Balancing the risk of mortality from recurrent VTE versus major bleeding has been challenging.  A validated clinical decision tool is sorely needed!

A few tools such as DASH, Vienna, and HERDOO2 have been developed over the years.  These tools purportedly identify patients at low-risk of recurrent VTE and thus can safely stop anticoagulation therapy.³ Until recently, none of these tools have been validated and therefore they have not been widely adopted into practice. Each tool considers several criteria and use a scoring system to predict VTE recurrence.  An elevated D-dimer serum concentration is a risk factor used in all three models.  DASH and Vienna predict the risk of recurrent VTE 1 month after anticoagulation has been stopped.  Given that many recurrent VTE events will occur within 30 days of treatment discontinuation, these models clearly have limited utility in practice!  In contrast, the HERDOO2 tool can be used to calculate risk in patients currently on therapy. The original HERDOO2 clinical decision rule was developed to identify women at low-risk of recurrence.⁴ Researchers were unable to create a rule that identified low-risk men and concluded that all men are at a high risk of recurrence.

The HERDOO2 validation study was the largest and most comprehensive study to evaluate risk factors for recurrent VTE in patients with unprovoked VTE.  The authors concluded it may be safe for women who have been treated with anticoagulants for 5-7 months after an unprovoked VTE to discontinue therapy if they have HERDOO2 score of 0 or 1.  See Table 1.

To validate HERDOO2, the investigators conducted a multinational, prospective cohort study called REVERSE II.⁵ REVERSE II included 44 secondary and tertiary care centers in 7 countries and targeted patients who experienced a first major, unprovoked VTE and who had received at least 5 months but no more than 12 months of anticoagulation.  The VTE was considered unprovoked in the absence of: leg fracture or lower extremity plaster cast, immobilization for more than three days, major surgery in the three months before the VTE event, and diagnosis of malignancy in the past five years.  Participants must be at least 18 years old.  Patients who already discontinued anticoagulation therapy or those requiring ongoing treatment were excluded.

Low-risk patients were classified as women with a HERDOO2 score of 0 or 1. High-risk patients included women with 2 or more HERDOO2 criteria and all men. There were two arms in this study.  The intervention arm included low risk patients who discontinued anticoagulants.  The observation arm included high-risk patients where the anticoagulation treatment decision was left to clinician discretion.

Baseline assessments occurred when patients were still taking anticoagulation therapy. Patients received follow up at 6 months and 1 year.  Participants were instructed to contact study staff if they developed any symptoms of recurrent VTE during the follow-up period.

The primary outcome was adjudicated symptomatic major VTE defined as a new popliteal vein or proximal deep vein thrombosis or new pulmonary embolism. Secondary outcomes were major bleeding and mortality.  Based on the ISTH recommendation that a recurrent VTE risk threshold of < 5% would allow patients to safely discontinue anticoagulants, a sample size of N = 600 was required to meet power.

There were 2747 patients included in the analysis with a mean age of 54 years old, 44% were female, and 83% were white.  Of the 1213 women, about half were classified as low-risk using the HERDOO2 tool.  Of the 591 low-risk women who discontinued anticoagulation, 17 developed recurrent major VTE during 564 patient-years of follow-up (3.0% recurrent VTE per patient-year, 95% CI 1.8 to 4.8%). In 323 high-risk patients who discontinued anticoagulation, 25 recurrent VTE events occurred during 309 patient-years of follow-up (8.1% recurrent VTE per patient-year, 95% CI 5.2 to 11.9%). There were 22 major bleeding events during 1788 person-years follow-up (1.2% major bleeds per patient-year, 95% CI 0.8% to 1.9%) in those continuing anticoagulants. 

Table 2: Major Results of REVERSE II⁵

Primary and secondary outcome analysis by risk category and whether anticoagulants were continued or discontinued.  Events (%) per 100 patient years (95% confidence intervals)

^{OAC = oral anticoagulants}

This study prospectively validated HERDOO2 and suggests the tool is able to identify low-risk patients who can safely discontinue anticoagulants after completing at least 5 months of treatment following a first unprovoked VTE.  Using the HERDOO2 tool, more than half of women with unprovoked VTE can discontinue anticoagulants and be spared the risks, costs, and burden of chronic anticoagulant therapy.  The risk of recurrent major VTE of 3% found in the study is less than the 5% threshold recommended by the ISTH.

Are the results from the REVERSE II study likely to impact clinical practice?  Strengths of the study include a large sample size, increasing the precision of the estimates of recurrent VTE.  Furthermore, the sample was drawn from a diverse patient population from multiple centers around the world, increasing generalizability of results.

A major limitation of HERDOO2 – the results are restricted to women. The investigators were unable to identify any combination of characteristics in men that would deem them as low risk for recurrent VTE. Researchers chose one year as the follow-up period because the highest risk of recurrence is in the first year after discontinuing anticoagulants; however, a longer period of follow-up would be necessary to see if the rate of recurrence substantially increases.  The PADIS-PE trial compared patients with first unprovoked VTE who received 6 months of anticoagulants and then were randomized to 18 months of anticoagulation or placebo.⁶ During the 18 months follow-up, patients receiving anticoagulation had a lower composite of recurrent VTE and major bleeding.  However after 42 months of follow-up, the two study arms did not differ significantly in the composite endpoint.⁶ Examining outcomes from the use of HERDOO2 tool beyond 1 year would provide further insight in its potential utility.

With the increased use of DOACs for both acute and extended treatment of VTE, the risk:benefit ratio may be altered.  Low-dose apixaban 2.5mg BID was highly effective in the AMPLIFY-EXT when given after 6 months of anticoagulation therapy.⁷ More recently, rivaroxaban 10 mg daily was highly effective for extended VTE treatment in EINSTEIN-CHOICE.⁷ These lower dose regimens are more convenient and safer options (although admittedly not cheaper) and patients (and clinicians) may be more reluctant to stop anticoagulation therapy.

Despite these limitations, the HERDOO2 tool has the potential to save many women from a lifetime of anticoagulation therapy. Routine use of HERDOO2 to inform treatment decisions can reduce pill burden, the bleeding risk from anticoagulants, and the burdens of INR monitoring (for patients prescribed warfarin). It’s a quick and easy way to assess patients and can help guide the decision to stop anticoagulation. What do you think? Should we routinely use HERDOO2 to guide treatment decisions? Or should be advocate the use of DOACs for extended treatment after an unprovoked VTE?