atrial fib in rheumatic dz: warfarin better than rivaroxaban

 Another article on atrial fibrillation (AF) was just published, finding that warfarin was superior to rivaroxaban in those with rheumatic heart disease (see afib warf better than rivarox in RHD NEJM2022 in dropbox or DOI: 10.1056/NEJMoa2209051)

 

Details:

-- 4531 patients with atrial fibrillation or atrial flutter and echocardiographically documented rheumatic heart disease who had any of:

    -- CHA₂ DS₂VASc score of at least 2

    -- or echocardiographic mitral-valve area of no more than 2 cm², left atrial spontaneous echo contrast, or left atrial thrombus

-- mean age of the patients was 51 years, 72% were women, BMI 24.5

-- eGFR 81, heart failure 39%, hypertension 23% (mean SBP 116 mmHg), stroke 11%

-- CHA₂ DS₂VASc score 1.9

-- patients were from Africa, Asia, and Latin Amercia

-- those who were included in the study were because of: mitral stenosis in 82%, CHA₂ DS₂VASc score in 56%,  and left atrial spontaneous echo contrast in 12%

-- meds prior to enrollment: b-blockers 72%, ACE/ARB 28%, digoxin 42%, diuretic 84%, vitamin K antagonist 53%

-- patients were randomly assigned to receive standard doses of rivaroxaban 15 or 20mg daily depending on renal function or dose-adjusted vitamin K antagonist (goal INR 2-3, with INR checks at least monthly)

     -- vitamin K antagonist (VKA) used: 79-85% were on warfarin and the rest acenocoumarol

-- primary efficacy outcome: composite of stroke, systemic embolism, myocardial infarction, or death from vascular (cardiac or noncardiac) or unknown causes

-- primary safety outcome was major bleeding, as defined by the International Society of Thrombosis and Hemostasis.

-- mean followup 3.1 years

 

Results:

-- INR in target range (2.0-3.0) for those on VKA: rate of 56% at 6 months, 59% at 1 year, 65% at 2 years, 65% at 3 years, 64% at 4 years

-- comparing % of patients continuing to receive VKA vs rivaroxaban: 98% vs 88% at 1 year, 98% vs 84% at 2 years, 97% vs 81% at 3 years, 97% vs 79% at 4 years

    -- permanent discontinuation of trial medication was more common with rivaroxaban than with VKA therapy at all visits, some related to patient preference, some to subsequent valve replacement (31%) necessitating a VKA 

-- In the intention-to-treat analysis, 560 patients in the rivaroxaban group and 446 in the VKA group had a primary-outcome event, proportional-hazards risk 1.25 (1.10-1.41), a 25% increase

 

Note that the curves start to diverge around 24-30 months and are on a trajectory of increased divergence at 54 months

-- by components of primary outcome (only the statistically significant ones), rivaroxaban vs VKA, in intention-to-treat analysis:

    -- stroke: 90 vs 65 patients, 37% increased risk, proportional HR 1.37 (1.00-1.89)

    -- ischemic stroke: 74 vs 48 patients, 53% increased risk, prop HR 1.53 (1.06-2.20)

    -- death: 552 vs 442 patients, 23% increased risk, prop HR 1.23 (1.09-1.40), with death in 8.0% vs 6.4% per year

        -- death due to vascular cause: 439 vs 337 patients, 29% increased risk, prop HR 1.29 (1.12-1.49)

        -- sudden cardiac death: 141 vs 94 patients, 51% increased risk, prop HR 1.51 (1.16-1.96)

        -- death due to mechanical or pump failure: 237 vs 174 patients, 35% increased risk, prop HR 1.35 (1.11-1.64)

-- no difference in on-treatment analysis: ie, assessing results in those actually taking the different meds

-- no difference in rate of valve replacement surgery or mitral valvuloplasty 

-- no difference in rate of major bleeding (none statistically significant, but major bleeds in 40 on rivaroxaban and 56 on VKAs, with 4 vs 15 fatal bleeds, 8 vs 14 intracranial bleeds)

Commentary:

-- though rheumatic heart disease is not so common in those born in the US, it is a major cause of valvular heart disease, resultant atrial enlargement (especially if mitral stenosis), and atrial fibrillation in the world. But it is pretty common in people coming here from high prevalence countries who had rheumatic fever, typically in their youth and more often in females

-- trials of DOACs (direct oral anticoagulants, which includes rivaroxaban) in patients with AF are typically excluded those with rheumatic heart disease

-- patients with mitral stenosis have a very high risk of stroke (pre-anticoagulation, in the 30-40% range), and this was the most common reason for inclusion in this study

 

-- This study confirmed that VKAs are the preferred anticoagulation in those with rheumatic heart disease, with impressive differences in major clinical outcomes (stroke and death). These benefits appeared to dwarf the much smaller (and not statistically significant) increase in major bleeding with VKAs vs rivaroxaban

-- of note, rivaroxaban probably should not be considered the best choice of the DOACs (see http://gmodestmedblogs.blogspot.com/2022/01/atrial-fib-apixaban-outperforms.html )
-- it is unclear in this study why there was such a profound benefit of VKAs on deaths vs strokes, since mortality is a less expected outcome than strokes, and stroke risk is so high if not adequately anticoagulated (of course, some with strokes do die, but the numbers in this study of deaths are high: the absolute number or strokes prevented was 25, yet the number of deaths prevented was 110). this all suggests that the VKA effect was largely independent of the expected benefit for the usual AF culprits: stroke and systemic embolism

-- it did take 24-30 months on these therapies to start to see a difference. not sure why.

-- the percent of people in appropriate INR range on VKA varies pretty dramatically between individuals (in my experience): some are about 100% in-range, and some need pretty regular VKA adjustments (perhaps related to changes in diet, meds including acetaminophen, VKA medication adherence). this study found about 60% in INR range at their visits. not sure how they approached those out-of-range (not documented): did they test these patients more frequently and adjust their dose? what was their algorithm and patient adherence to it? there is not much information that i can find in the medical literature, but one expert opinion was that the goal should be at least 70% of the time in-range. would the benefits of VKA be much higher if 70-80% of patients were in-range? what are the benefits and risks for my patients in the 100% range? we could potentially tweak our anticoagulation approach some if there were a significant benefit to achieving a certain target of % of INR's in-range.

    -- another issue is what should we consider the correct range to be, given the target range of 2.0-3.0?? there is pretty strong argument that patients with an INR between 1.7-2.0 or between 3.0-3.3 have a strong likelihood of not needing a VKA dose change to actually be in the 2.0-3.0 target range, that subsequent INR values without change are often in the 2.0-3.0 range (see https://pubmed.ncbi.nlm.nih.gov/18983486/ ). my practice is to repeat the INR a week later if in the 1.7-2.0 or 3.0-3.3 range to make sure it is okay, prior to changing VKA dosing, and i have typically found this to be true

        -- but if the 60% of those in target range as found in the above study actually meant that 70-80% actually had INRs between 1.7 and 3.3, that seems likely to be okay and they are attaining a pretty consistent and appropriate VKA effect

Limitations:

-- there is a fundamental difference in this type of study, whereby one group (VKAs) are seen regularly by their care providers (at least monthly for INR testing) vs every several months with the other group (rivaroxaban). And this difference may well reflect differences in access to care as well as the beneficial effect of an increased therapeutic relationship in those getting more care. this may explain part of the issue of medication adherence

-- those in the rivaroxaban group who needed valve surgery were then put on VKAs, which could distort the intention-to-treat analysis, but the on-treatment analysis did not have much difference in results. but, on-treatment analyses are less statistically rigorous, by potentially inducing a bias.

so, this study does support the guidelines of using VKAs for patients with AF related to rheumatic heart disease (eg see https://reader.elsevier.com/reader/sd/pii/S0735109714017409?token=297F8D0D95101C3E18E082D01FE75CC9305AA388D5FD30A1F81133C0627A08273924D7E794AC4D62A201BD2763232510&originRegion=us-east-1&originCreation=20220921141442 ).  But it does bring up the question of how we should quantify patients in "target INR range", both in these studies but also in our clinical practice, perhaps considering the 1.7-3.3 as the goal range, as found in observational studies to very likely reflect the 2.0-3.0 range (though should probably be rechecked within a week or so)

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