Factor Xa inhibitor use linked to risk for ILD in patients with atrial fibrillation

In patients with nonvalvular atrial fibrillation receiving oral anticoagulants, factor Xa inhibitors appeared associated with increased risk for interstitial lung disease, according to a study published in JAMA Network Open.

“This is an observational study which describes associations, not causality,” Gregory Y.H. Lip, MD, of the Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart and Chest Hospital, told Healio. “Thus, while initially surprising, the results illustrate that the direct oral anticoagulants are not all the same.”

Various non-vitamin K antagonist oral anticoagulants (NOACs) have demonstrated lower bleeding risk profiles than warfarin, but there has been some evidence of possible risk for interstitial lung disease (ILD) associated with their use, according to Lip and colleagues.

In a nationwide retrospective cohort study, Lip and colleagues analyzed 106,044 patients (mean age, 73.4 years; 56.6% men) with nonvalvular atrial fibrillation receiving oral anticoagulants from the Taiwan National Health Insurance Research Database between June 1, 2012, and Dec. 31, 2017, to determine the risk for incident ILD in this patient population.

Of the total cohort, 60.9% of patients received factor Xa (FXa) inhibitors — which included rivaroxaban (56.9%; Xarelto, Janssen), apixaban (23.8%; Eliquis; Bristol Myers Squibb, Pfizer) and edoxaban (19.2%; Savaysa, Daiichi Sankyo) — 21.2% received dabigatran and 17.9% received warfarin, which served as the reference group. In order to control covariates across these three groups, researchers used propensity score stabilized weighting.

Follow-up continued until Dec. 31, 2019, or until patients were diagnosed with ILD or died.

After balancing covariates for the groups, researchers found that patients receiving FXa inhibitors were at a higher risk for incident ILD than those receiving warfarin, with a hazard ratio of 1.54 (95% CI, 1.22-1.94; 0.29 vs. 0.17 per 100 patient-years).

Similar to the overall FXa inhibitor group, apixaban (0.35 vs. 0.17 per 100 patient-years; HR = 1.72; 95% CI, 1.27-2.31), edoxaban (0.37 vs. 0.17 per 100 patient-years; HR = 1.6; 95% CI, 1.12-2.27) and rivaroxaban (0.27 vs. 0.17 per 100 patient-years; HR = 1.48; 95% CI, 1.16-1.88) each carried significantly higher risk for incident ILD.

Researchers found that the association between FXa use and ILD remained regardless of patient age; sex; CHA₂DS₂-VASc score; HAS-BLED score; use of angiotensin system inhibitors, amiodarone, statin or beta-blocker; or risk for stroke or major bleeding.

Notably, treatment consisting of FXa inhibitors plus amiodarone, a frequent treatment for atrial fibrillation, showed the highest risk for incident ILD compared with treatment with warfarin alone (0.38 vs. 0.13 per 100 patient-years; HR = 2.69; 95% CI, 1.88-3.85) among the groups.

Additionally, when comparing FXa inhibitor to warfarin, there was an absolute risk decrease per 100 patient-years of 0.78 (95% CI, 0.63-0.94) for ischemic stroke or systemic embolism and 0.78 (95% CI, 0.63-0.93) for major bleeding in these patients. Dabigatran also showed an absolute risk decrease per 100 patient-years of 0.64 (95% CI, 0.46-0.82) for ischemic stroke or systemic embolism and 1.01 (95% CI, 0.84-1.17) for major bleeding according to researchers. Researchers noted that the increased risk for ILD with FXa inhibitors was much smaller than these decreases in risk.

“We have described associations, not causality,” Lip told Healio. “Thus, vigilance in monitoring for any potential adverse lung outcomes associated with the use of these drugs is recommended as part of the holistic approach to [atrial fibrillation] care and management. This requires further studies in different populations and additional prospective work or clinical trials. We would initially look to see if the observations could be observed in other large cohorts, including non-Asian studies.”

This study by Lip and colleagues adds to the literature indicating that more studies need to be done on the association between NOACs and ILD to prove causation, according to an accompanying editorial by Emanuel Raschi, MD, PhD, associate professor in pharmacology in the department of medical and surgical sciences at the University of Bologna, Italy.

“The welcome contribution by Chan et al should not be viewed as an alarm but rather as an alert for clinicians, including general practitioners, hospitals, and specialized physicians,” Raschi wrote. “Recommending the close monitoring of lung function in patients who were treated with NOACs is not justified, and any regulatory measure cannot be envisioned other than an update of the summaries of product characteristics. However, patients should be instructed to timely communicate early respiratory signs and symptoms to their clinicians, who should remain vigilant for adverse lung effects, especially in patients receiving anti–FXa agents and concomitant amiodarone.”

For more information:

Gregory Y.H. Lip, MD, can be reached at Gregory.Lip@liverpool.ac.uk.

Reference:

Raschi E. JAMA Netw Open. 2022; doi:10.1001/jamanetworkopen.2022.43316.