Potential Warning Signs Seen for NOAC Use After TAVR
Bleedings risks were similar with NOACs and VKAs in a registry study, but ischemic risks were higher with the newer agents.
Patients who are taking oral anticoagulation for atrial fibrillation and then undergo TAVR may be better off with a vitamin K antagonist (VKA) than with a non-VKA oral anticoagulant (NOAC), a registry study suggests.
Though bleeding risks did not differ between the two types of drugs in the year after TAVR, NOACs carried a higher risk of a composite of all-cause mortality, MI, or any cerebrovascular event (21.2% vs 15.0%), a difference that was borderline significant (HR 1.44; 95% CI 1.00-2.07).
“We are suggesting that we have a risk of ischemic events if we use NOACs instead of vitamin K antagonists in TAVI patients,” senior author Julinda Mehilli, MD (Munich University Clinic, Ludwig-Maximilians University, Germany), told TCTMD, noting that most patients who require an oral anticoagulant for A-fib—even after TAVR—receive NOACs because of their greater ease of use compared with VKAs.
Asked whether there was enough evidence from this study, as well as smaller investigations hinting at the same differences between NOACs and VKAs after TAVR, to advise more judicious use of NOACs in this setting, Mehilli pointed to the need for randomized trials before recommending changes to clinical practice. She called the current findings hypothesis-generating but not confirmatory.
“For the clinical routine, if you are able to identify patients who have the highest risk of thromboembolic events in the future and tailor therapy for these people—maybe by selecting vitamin K antagonists instead of NOACs or using the higher dose of NOACs among these people—[that] is a possible reaction to our data,” she said, adding, however, that results from ongoing randomized trials (ATLANTIS and ENVISAGE-TAVI AF, for example) will provide more information than can be gleaned from observational studies.
It’s possible, too, that routine monitoring might be required for NOACs, though assessing the anticoagulant activity of NOACs is more difficult than for VKAs, Mehilli said.
Many patients undergoing TAVR have an indication for chronic oral anticoagulation, typically A-fib, but there is limited evidence to guide the choice of agent in this setting. Guideline writers have responded to that gap differently. European guidelines give preference to VKAs over NOACs in the first 3 months after implantation of an aortic bioprosthesis but say NOACs should be considered as an alternative after that point. US guidelines, in contrast, make no specific mention of NOACs in the setting of TAVR, saying use of a VKA for 3 to 6 months after the procedure is reasonable for all patients with a low risk of bleeding.
NOACs are contraindicated in patients with mechanical valves in both sets of guidelines, based mainly on the findings of the RE-ALIGN trial.
The current study, published online June 12, 2019, ahead of print in JACC: Cardiovascular Interventions with lead author David Jochheim, MD (Munich University Clinic, Ludwig-Maximilians University), aimed to provide some insights into the choice of anticoagulant in the post-TAVR setting. The investigators examined registry data on 962 patients who underwent TAVR at four European centers between June 2007 and February 2017. Mean age of the population was 81.3 years, and the median STS score was 4.5%. Slightly more than half of the patients (52.5%) were women, and most (62.7%) received a balloon-expandable valve.
The indication for oral anticoagulation was A-fib in nearly all cases. Two-thirds of patients were discharged on a VKA and the rest on a NOAC. Of the latter group, 53.7% received rivaroxaban (Xarelto; Bayer/Janssen), 39.2% apixaban (Eliquis; Bristol-Myers Squibb), and 7.1% dabigatran (Pradaxa; Boehringer Ingelheim).
To account for differences between the NOAC and VKA groups, the investigators used propensity scores for inverse probability of treatment weighting.
After such adjustment, the rate of the composite ischemic outcome was borderline significantly higher in the NOAC versus VKA group at 1 year. There were no differences for any of the other outcomes, including any BARC bleed (33.9% vs 34.1%; HR 0.97; 95% CI 0.74-1.26), major or life-threatening bleeds (20.7% vs 22.5%; HR 0.90; 95% CI 0.64-1.26), or all-cause mortality (16.5% vs 12.2%; HR 1.36; 95% CI 0.90-2.06).
What’s Going on Here?
The higher rate of ischemic events with NOACs versus VKAs seen here conflicts with extensive data in the broader A-fib population showing efficacy that is at least as good—if not better—and safety that is superior with the NOACs.
Mehilli said the discrepancy could have something to do with the high level of risk seen in patients undergoing TAVR—who typically have been older with higher rates of comorbidities and prior stroke. Previous studies have shown, she noted, that NOAC-treated patients with A-fib and high CHA2DS2-VASc scores frequently have subtherapeutic plasma drug levels, and these types of patients might be more frequent in a post-TAVR setting relative to the wider A-fib population. Prior research has also demonstrated, Mehilli said, that patients with low plasma concentrations of NOACs at steady state are more likely to have thromboembolic events.
That is the most probable explanation, Mehilli agreed, but she speculated that the findings might also be related to differences in the effects of NOACs and VKAs. VKAs, she explained, more broadly interfere with the coagulation cascade—suppressing both tissue factor-induced and contact pathway-induced coagulation—and that additional suppression might be necessary for patients with stented bioprostheses.
In an accompanying editorial, however, Dominick Angiolillo, MD, PhD, and Andres Pineda, MD (University of Florida College of Medicine – Jacksonville), say “the findings from this investigation need to be interpreted with caution as in the context of a nonrandomized study with inherent limitations, including unbalanced baseline characteristics and unmeasurable confounders.
“In fact,” they continue, “while the authors attribute the increased risk of thrombotic complications with a NOAC to its lesser degree of anticoagulation effect compared with a VKA, this did not result in an increase in bleeding. It is also important to note that the increased risk of ischemic events was of borderline statistical significance.”
On the other hand, there is some evidence from other studies that bolster the current results, Angiolillo and Pineda say, pointing to a small study hinting at greater risks of all-cause mortality and stroke at 1 year with a NOAC versus warfarin and the prematurely halted GALILEO trial, which—though it excluded A-fib patients—showed worse outcomes with rivaroxaban versus antiplatelet therapy after TAVR.
“Collectively, with the findings from this investigation, these observations must be seriously considered,” they say, noting, as Mehilli did, that ongoing trials will help clarify the ideal antithrombotic regimen after TAVR for patients with A-fib.
“Despite the supportive data associated with the use of NOACs in patients with atrial fibrillation, including among those undergoing other interventional procedures such as coronary stenting, to date there is [a] lack of evidence that these should be preferred over a VKA after TAVR,” the editorialists conclude. “Until then, NOACs should be used with caution and the choice of antithrombotic therapy should be individualized, taking into account patients’ bleeding and thrombotic risk profile.”
Jochheim D, Barbanti M, Capretti G, et al. Oral anticoagulant type and outcomes after transcatheter aortic valve replacement. J Am Coll Cardiol Intv. 2019;Epub ahead of print.
Angiolillo DJ, Pineda AM. Oral anticoagulation after TAVR in patients with atrial fibrillation: the certainty of uncertainty. J Am Coll Cardiol Intv. 2019;Epub ahead of print.
- The registry was partially funded by the Munich Heart Alliance partner site of DZHK.
- Mehilli reports receiving lecture fees from Abbott Vascular, Boston Scientific, Biotronik, Edwards Lifesciences, BMS, and Terumo and institutional research grants from Edwards Lifesciences, Boston Scientific, and Abbott Vascular.
- Jochheim reports no relevant conflicts of interest.
- Angiolillo reports having received consulting fees or honoraria from Amgen, Aralez, AstraZeneca, Bayer, Biosensors, Boehringer Ingelheim, Bristol-Myers Squibb, Chiesi, Daiichi Sankyo, Eli Lilly, Haemonetics, Janssen, Merck, PhaseBio, PLx Pharma, Pfizer, Sanofi, and The Medicines Company and having received payments for participation in review activities from CeloNova and St. Jude Medical. He also reports that his institution has received research grants from Amgen, AstraZeneca, Bayer, Biosensors, CeloNova, CSL Behring, Daiichi Sankyo, Eisai, Eli Lilly, Gilead, Idorsia, Janssen, Matsutani Chemical Industry Co., Merck, Novartis, Osprey Medical, and Renal Guard Solutions.
- Pineda reports having received consulting fees from Pfizer and TZ Medical.