Concomitant use of NSAIDs and oral anticoagulants (DOACs) – including DOACs – may increase the risk for major bleeding, according to a new post hoc analysis of the ARISTOTLE trial.

Study investigators analysed data from 17,423 individuals who took part in the trial comparing apixaban with warfarin in patients with AF.  NSAID use was common: at baseline 5% patients were taking one and an additional 13% used NSAIDs at least once during the trial.

Patients taking NSAIDs with an oral anticoagulant had significantly higher rates of major bleeding and more clinically relevant non-major (CRNM) bleeding than those who did not use NSAIDs (24.5% baseline users vs. 21.0% incident vs. 15.6% never users).

Researchers said the association of major bleeding with incident NSAID use during the trial was significant (HR 1.61, 95% CI 1.11–2.33) when they excluded NSAID use at baseline.

And the same pattern was found for NSAID use in relation to CRNM bleeding (HR 1.70, 95% CI 1.16–2.48) but not for gastrointestinal bleeding (HR 1.26, 95% CI 0.59–2.73), the researchers said in Circulation.

Major bleeding in patients with baseline and incident NSAID use were primarily upper gastrointestinal (19.3%), ‘other’ (19.3%), and intracranial bleeding (14.8%).

For CRNM bleeding the primary locations were ‘other’ (23.5%), epistaxis (22.2%), and haematuria (14.8%).

The study investigators noted that the safety and efficacy of apixaban versus warfarin appeared not to be significantly altered by NSAID. In patients using an NSAID at baseline, the incidence rate per 100-person years of major bleeding was 2.7 in the apixaban arm and 4.2 in the warfarin arm.

In patients not using an NSAID, the incidence rate per 100-person years of major bleeding was lower in both groups (2.1 on apixaban and 3.0 on warfarin) while the incidence rate per 100-person years of gastrointestinal bleeding was low in those with NSAID use (0.8 on apixaban and 1.2 on warfarin) and never users (0.7 in both treatment arms).

Speaking to the limbic about the implications of the study, cardiologist Associate Professor John Amerena, director of the Geelong Cardiology Research Unit and director of the Heart Failure Programme at Barwon Health in Victoria, described the findings as ‘intuitive’ and supported recommendations to avoid the combination of anticoagulants and NSAIDs wherever possible.

However he said that in practice it’s not always easy to avoid using an NSAID, so he suggested using a COX-2 inhibitor such as celecoxib, which is associated with less bleeding and does not decrease platelet function but still has anti-inflammatory properties.

“What I say to patients, is if you really can’t function without [an NSAID], take low dose [celecoxib] on a day when they are going to do something active that they know is going to stir their joint pain. That one-off use every now and then is not going to hurt people,” he said.

And in the case of an acute injury with a lot of discomfort he suggests a week of NSAID use at the most.

“We’d say, use the lowest dose possible to control the discomfort for the shortest duration possible and then you’ve got to play it individually on a case by case basis,” he added.

In the study, the most commonly used NSAID at baseline was diclofenac (15.0%), followed by COX-2 inhibitors (14.4%), and ibuprofen (14.3%). As for patients with incident NSAID use, the most commonly used drug was also diclofenac (24.6%), followed by ibuprofen (18.4%) and very few initiated Cox II inhibitor (0.2%).