Anticoagulant Type May Influence Fracture Risk in A-fib Patients

The overall risk of osteoporotic fractures in patients on chronic oral anticoagulation for A-fib is low, but the choice of agent may make a difference, according to nationwide data from Denmark.

The adjusted 2-year risk of having any fracture or starting medications for osteoporosis was significantly lower among patients taking a direct oral anticoagulant (DOAC) versus a vitamin K antagonist (5.21% vs 6.43%; HR 0.84; 95% CI 0.76-0.93), Casper Binding, BMedSc (Copenhagen University Hospital Herlev and Gentofte, Hellerup, Denmark), and colleagues report.

The researchers calculated that for every 1,000 patients treated with a DOAC instead of a vitamin K antagonist (VKA) for 2 years, 6.8 osteoporotic fractures and 7.1 osteoporosis treatment initiations could be avoided.

“This study shows that skeletal health profile should be taken into account when assessing OAC [oral anticoagulant] treatment among patients with atrial fibrillation, and I think this is the key message for practicing clinicians,” Binding, a medical student, told TCTMD in an email.

“I don’t think DOAC treatment should categorically be the choice of drug in all cases, but I do think clinicians need to assess every patient according to comorbidities when choosing OAC treatment, and in doing so our findings should be taken into account,” he continued. “In other words, patients with strong risk factors for osteoporotic fractures should be considered for DOAC treatment rather than VKA treatment.”

But that conclusion might be a bit premature, according to Brian Gage, MD (Washington University School of Medicine, St. Louis, MO), who noted that prior studies looking into the link between VKA therapy and fractures have provided mixed results.

“There are a lot of factors we take into account when we’re choosing antithrombotic therapy either for atrial fibrillation or for other disorders, and the ones that are always going to be most important are risk of stroke and risk of bleeding,” Gage commented to TCTMD.

He pointed out that a typical patient with A-fib who is not taking anticoagulation has a stroke risk of 4% to 5% per year. Anticoagulation cuts that by 60% to 70%. “That’s what’s going to drive our decision because stroke is so catastrophic for these patients. These other factors are also going to be important, but not as important,” he noted.

That said, Gage added, if the difference in fracture risk between VKAs and DOACs is corroborated with additional studies, it might be one factor to consider when choosing an oral anticoagulant, particularly in patients with A-fib and a high baseline risk of fracture.

Binding and colleagues published their findings online ahead of the October 29, 2019, issue of the Journal of the American College of Cardiology.

Inconsistent Prior Evidence

As noted by Gage, previous studies exploring the link between oral anticoagulation and fracture risk have yielded inconsistent findings, with some showing an increased risk with VKAs. Putative mechanisms involve dietary restrictions required for warfarin use and a direct effect of vitamin K antagonism on bone health.

To see whether there might be a difference between VKAs and DOACs, which do not have an effect on vitamin K, the investigators looked at data from Danish national registries on 37,350 patients who had received at least 180 days of oral anticoagulation for A-fib and had no prior use of medications for osteoporosis. Overall, 32.6% of patients were treated with a VKA and 67.4% received a DOAC.

DOAC-treated patients were slightly older and more likely to be female, be on hormone replacement therapy, have a history of prior stroke or fracture, and have a diagnosis of alcohol abuse. After adjustment for potential confounders, DOAC versus VKA therapy was associated with lower risks of a variety of fracture-related outcomes, with the exception of hip fracture.

Standardized 2-Year Risks

^aAny fracture or initiation of osteoporosis medications.

The researchers point out two potential mechanisms to explain the observed higher risk of fractures with VKA therapy. “Warfarin is a VKA, and by regulating vitamin K, warfarin inhibits the g-carboxylation of several proteins, including coagulation factors II, VII, IX, and X,” they explain. “Studies suggest a link between warfarin and undercarboxylated osteocalcin, which is associated with low bone mineral density.”

In addition, Binding et al say, VKA-treated patients are required to restrict various types of food, particularly certain vegetables containing folic acid, that may interact with the drug effects. That can increase the risk of hyperhomocysteinemia. “Hyperhomocysteinemia is associated with a decrease in osteoblast activity and an increase in osteoclast activity, thus reducing bone strength,” the authors write. “Hyperhomocysteinemia is also found to increase the amount of matrix metalloproteinases, which is associated with degradation of extracellular bone matrix.”

In contrast, DOACs do not influence the synthesis of osteocalcin and are not subject to dietary restrictions.

Impact of Fractures on Quality of Life

The results of the study are important, Binding said, because “many patients with atrial fibrillation are elderly, and studies have shown that osteoporotic fractures are associated with high mortality and reduced life quality among elderly patients. This is why it is important to investigate the association between OACs and osteoporotic fractures, and why practicing clinicians should take these findings into account.”

Asked whether the results would be similar when looking at individual DOACs—rather than the class as a whole—Binding indicated that they probably would be.

“The hypothesis of the study, and the reason why VKAs are associated with a higher risk of osteoporotic fractures, is based on the biochemical function of VKAs and the dietary restrictions accompanying the treatment,” he said. “On this basis, the different DOACs should all be associated with a lower risk of osteoporotic fractures compared to VKAs, and it would be interesting to evaluate the DOACs separately to test the hypothesis.”

There are a lot of factors we take into account when we’re choosing antithrombotic therapy either for atrial fibrillation or for other disorders, and the ones that are always going to be most important are risk of stroke and risk of bleeding. Brian Gage

Gage agreed that there shouldn’t be a difference between the DOACs in this regard. But he said these findings shouldn’t affect clinical decision-making at this point and require corroboration in other data sets.

“This study in and of itself is unlikely to change clinical practice but studies like this in aggregate at some point may influence our prescribing decision in a secondary way,” he said.

As Gage concludes in an accompanying editorial, the researchers “had a strong biochemical basis to search for a link between VKAs and osteoporotic fracture. Their well-done observational study provides additional evidence of this link, but residual confounding may have exaggerated the association. Thus, at least for patients who have atrial fibrillation and no prior osteoporotic fracture (the population studied), the decision to prescribe a VKA or a DOAC should depend on the risks of ischemic stroke, hemorrhage, need for monitoring, and affordability rather than on the risk of osteoporotic fracture.”