Warfarin Is Bad to the Bone, Fracture Data Affirms

In U.S. claims analysis, fracture risk lower with DOACs

This article was first published on Monday, November 25, 2019 in MedPage Today.

By Nicole Lou, Staff Writer, MedPage Today.

Warfarin (Coumadin) may not be the best oral anticoagulant in atrial fibrillation (Afib) in terms of bone health, a retrospective study suggested.

On multivariable-adjusted, propensity score-matched analyses of insurance claims, new users of direct oral anticoagulants (DOACs) were at lower fracture risk over an average 16.9 months of follow-up compared with new warfarin users:

• Fractures requiring hospitalization: HR 0.87 (95% CI 0.79-0.96)
   
• All clinical fractures: HR 0.93 (95% CI 0.88-0.98)
   
• Hip fractures: HR 0.91 (95% CI 0.78-1.07)

The strongest effect estimates were observed when comparing apixaban (Eliquis) and warfarin, whereas head-to-head comparisons among DOACs showed similar fracture risk, Pamela Lutsey, PhD, of University of Minnesota School of Public Health in Minneapolis, and colleagues reported in JAMA Internal Medicine.

"Collectively, our findings support the notion that warfarin may be harmful to bone health," they concluded. They argued that the lack of statistical significance for the outcome of hip fracture may have to do with lower precision for this outcome.

Warfarin had been associated with higher fracture risk than dabigatran (Pradaxa) in a 2017 study from Hong Kong.

Recently, other groups have also reported fewer cases of poor osteoporotic outcomes when Afib patients took DOACs in lieu of the vitamin K antagonist.

"Vitamin K is important in posttranslational glutamination of osteocalcin, the major noncollagenous bone matrix protein. Warfarin interferes with this process and consequently inhibits the activation of bone matrix proteins," the investigators noted.

They recommended caution when prescribing warfarin to patients with Afib at high risk of fracture.

The study used MarketScan administrative claims databases for information on a commercially-insured population of people with non-valvular Afib who were prescribed oral anticoagulants in 2010-2015. These were new users without known prior exposure to oral anticoagulation.

Groups were categorized according to the first oral anticoagulant they were prescribed: dabigatran (18.9%), rivaroxaban (Xarelto, 21.1%), apixaban (10.6%), warfarin (49.4%).

In the end, 167,275 individuals were matched for comparison. Their average age was 68.9 years; 38.0% were women. Each DOAC user was matched to up to three warfarin recipients.

The subgroup of people diagnosed with osteoporosis especially benefited from DOACs over warfarin, according to Lutsey's team.

Potential confounding is an inherent limitation to such a retrospective analysis, the authors acknowledged. Even after matching, the warfarin and apixaban groups in this study were older and had higher CHA₂DS₂-VASC scores compared to the dabigatran and rivaroxaban groups.

Furthermore, the databases used for the analysis excluded people without insurance, and edoxaban (Savaysa) was excluded from the study because there were too few users of this DOAC in the dataset.

The study was supported by grants from the National Heart, Lung, and Blood Institute; the National Institute on Aging; and the American Heart Association.

Lutsey disclosed receiving NIH grants.