Oral blood thinners and bone loss

July 15, 20212 min readSAVE

Oral anticoagulants may lower bone density, trabecular bone score


Adults prescribed any oral anticoagulation therapy have lower bone mineral density and trabecular bone score compared with controls, with a stronger negative effect observed among those prescribed warfarin, data published in Bone show.

“Patients taking oral anticoagulants are at risk of low bone mass and bone quality, and those taking warfarin had an even higher risk,” Victoria Z.C. Borba, MD, PhDassociate professor of endocrinology at the Federal University of Paraná in Brazil, told Healio. “Patients undertaking oral anticoagulants, no matter the type, should have their bone mineral density and quality by trabecular bone score investigated.”

Borba is an associate professor of endocrinology at the Federal University of Paraná in Brazil.

Direct-acting oral anticoagulants are therapeutic alternatives to warfarin that act independently of vitamin K, thus not affecting bone matrix formation, Borba and colleagues wrote in the study background. The researchers aimed to compare BMD and microarchitecture in patients treated with direct-acting oral anticoagulants compared with warfarin.

In a cross-sectional, observational study, Borba and colleagues analyzed data from 100 patients prescribed oral anticoagulants for at least 1 year following up at a cardiology service, and a control group of 50 patients matched by age, sex and race (mean age, 61 years; 64% men). Researchers then stratified participants based on the type of anticoagulant used, either direct-acting oral anticoagulants (n = 50) or warfarin groups (n = 50). Participants completed questionnaires and underwent BMD measurements via DXA as well as trabecular bone score assessment.

In the direct-acting oral anticoagulant group, 38% of participants used rivaroxaban (Xarelto, Janssen/Bayer), 34% used apixaban (Eliquis, Bristol Myers Squibb/Pfizer), 22% used dabigatran (Pradaxa, Boehringer Ingelheim) and 6% used edoxaban (Savaysa, Daiichi Sankyo). Eleven patients reported previous use of warfarin that was switched to a direct-acting oral anticoagulant more than 1 year before the study. The mean duration of anticoagulation was shorter in the direct-acting group compared with the warfarin group, at 30 months vs. 81 months, respectively (P < .001).

Within the cohort, researchers observed low bone mass in 42%, 50% and 66% of participants in the control group, direct-acting oral anticoagulants group and warfarin group, respectively (P = .012).

In logistic regression analysis, BMD was associated with BMI (OR = 0.846; 95% CI, 0.763-0.926), creatinine level (OR = 0.024; 95% CI, 0.001-0.434) and trabecular bone score (OR = 17.777; 95% CI, 4.526-96.903).

Mean trabecular bone score decreased progressively across the three groups — 1.328 (control group), 1.264 (direct-acting oral anticoagulants) and 0.138 (warfarin group; P < .001).

“All three groups differed in terms of bone microarchitecture assessed by trabecular bone score, which was more degraded in the warfarin group compared with the direct-acting oral anticoagulant group, and more degraded in the direct-acting oral anticoagulant group compared with the control group,” the researchers wrote.

Researchers found that negative predictors of trabecular bone score included warfarin use (–0.06; 95% CI, –0.11 to –0.02), BMI (–0.01; 95% CI, –0.01 to 0) and hyperglycemia (–0.07; 95% CI, –0.11 to –0.03).

Positive predictors of trabecular bone score were an active International Physical Activity Questionnaire classification (0.06; 95% CI, 0.01-0.11) and family history of hip fracture (0.07; 95% CI, 0.01-0.14).

“A study investigating fractures in patients taking oral anticoagulants should clarify the real impact of these medications on bone,” Borba told Healio. “This type of study could also direct us to which are the bone-friendly anticoagulants.”

For more information:

Victoria Z.C. Borba, MD, PhD, can be reached at vzcborba@gmail.com.

Leave a Comment

Your email address will not be published. Required fields are marked *