COX-2-Selective NSAIDs a Better Choice in DOAC-Treated AF Patients

COX-2-selective nonsteroidal anti-inflammatory drugs (NSAIDs) appear to be a safer option than their nonselective counterparts among patients with atrial fibrillation (AF) who are taking a direct oral anticoagulant (DOAC), according to observational data.

Concomitant use of a COX-2-selective NSAID instead of a nonselective NSAID was associated with lower risks of both gastrointestinal and non-GI bleeds, researchers led by Fabian Maximilian Meinert, MD (Charité-Universitätsmedizin Berlin, Germany), report in a study published online recently in JAMA Network Open.

The difference in GI bleeding based on NSAID type appeared to be greater in women than in men.

“Our findings suggest that the beneficial effects of COX-2-selective NSAIDs versus nonselective NSAIDs regarding the risk of GI bleeding that were first shown in randomized clinical trials among patients with indications for analgesic treatment are retained among patients with nonvalvular AF who are concomitantly treated with DOACs,” the researchers conclude, adding that “this finding is important given the high baseline risk of GI bleeding during treatment with DOACs.”

The investigators examined electronic medical records from the UK Clinical Practice Research Datalink Aurum and claims data from the Régie de l’assurance maladie du Québec, which includes information from all residents of the Canadian province ages 65 years and older along with individuals who lack private insurance plans and those receiving financial assistance. The analysis included 30,240 patients with AF who had at least one episode of concomitant DOAC and NSAID use (defined by overlapping prescriptions for at least 1 day) between January 1, 2011, and December 12, 2020. Mean patient age was 72.1 years, and 43.3% were women.

Most episodes of concomitant use (54.8%) involved nonselective NSAIDs, including ibuprofen, naproxen, and numerous others, with the rest of the episodes involving selective agents like celecoxib, diclofenac, etodolac, mefenamic acid, and meloxicam. The most commonly used COX-2-selective NSAIDs were diclofenac in the UK and celecoxib in Quebec; naproxen was the most frequently used nonselective agent in both cohorts.

The primary outcome was hospitalization for GI bleeding, which was less frequent with selective versus nonselective NSAIDs over mean follow-ups of 44 days in the UK (unweighted incidence rate [IR] 20.47 vs 34.77 per 1,000 person-years) and 30 days in Quebec (unweighted IR 27.04 vs 52.52 per 1,000 person-years). Similar differences were seen between NSAID types for non-GI bleeding in the UK (unweighted IR 17.92 vs 34.82 per 1,000 person-years) and Quebec (unweighted IR 31.31 vs 66.92 per 1,000 person-years).

In random-effects models, use of a COX-2-selective NSAID was associated with lower risks of GI bleeding (pooled weighted HR 0.63; 95% CI 0.46-0.87) and non-GI bleeding (pooled weighted HR 0.54; 95% CI 0.40-0.74).

The findings were consistent across subgroups defined by age, the order of DOAC-NSAID initiation, baseline bleeding risk, and type of DOAC, although women appeared to derive a greater benefit from using the selective agents compared with men.

“While this observation requires replication and should be considered as hypothesis-generating, it may be associated with potential differences in the risk of DOAC-induced GI bleeding between sexes,” Meinert et al write.

Tailored Care and Antithrombotic Stewardship

Commenting for TCTMD, Craig Beavers, PharmD (Baptist Health Paducah, KY), said “it’s great to see that there may still be some ability to provide some relief in those patients that need some form of pain management therapy or therapy related to, for example, osteoarthritis or something to that effect and not feel that we can’t give them an option because they are on a [DOAC].”

The study has limitations stemming from its retrospective design, but it “really speaks to the construct of proper selection of agents, trying to do some form of mitigating strategy to reduce harm, and working through a developed individualized plan,” he said. This might also include a choice of one DOAC over another to lessen bleeding risk.

Beavers added that he has a focus on antithrombotic stewardship programs, an issue taken up in an editorial accompanying the study by Kelly Rudd, PharmD (Oklahoma State University-Tulsa), and Geoffrey D. Barnes, MD (University of Michigan, Ann Arbor).

The importance of the study, which aligns with prior evidence, “lies in the reminder that coprescribing of NSAIDs with DOAC therapy, regardless of NSAID selectivity, could increase bleeding risk up to twofold, prompting a clinical imperative by the healthcare community to give thoughtful attention to mitigating this risk,” they write.

The editorialists note that the data suggest that using COX-2-selective versus nonselective NSAIDs could lead to about 16 fewer hospitalizations for GI bleeding and 19 for non-GI bleeding per 1,000 patient-years.

“Considering these data in the context of approximately 1 million patients concurrently taking DOACs and NSAIDs in the US (and larger numbers globally), the safety and financial impacts could be staggering,” Rudd and Barnes say.

Antithrombotic stewardship programs provide an opportunity to incorporate considerations around concomitant NSAID use into clinical decision-making, they indicate. For example, members of the team can mull whether NSAID therapy is necessary in the first place, preferentially use COX-2-selective agents when they’re an option, choose a nonselective NSAID with lower bleeding risks, and take other measures to lessen the overall hazard of bleeding.

These new data “reaffirm that thoughtful NSAID selection could meaningfully reduce patient harm, but true progress requires more than individual prescribing decisions,” Rudd and Barnes say. “By adopting the principles of antithrombotic stewardship, the healthcare community should transform these insights into lasting, system-wide action. In a time when preventable medication harm remains alarmingly common, implementing antithrombotic stewardship is not just an option but a responsibility.”