Additional Perspective on NSAIDs and Patients with CVD Who May be Using Aspirin

A recent study suggests that NSAIDs may not interfere with the cardiac benefits of aspirin. Those are the widely reported top-line results of the latest post hoc analysis of the PRECISION (Prospective Randomized Evaluation of Celecoxib Integrated Safety Versus Ibuprofen or Naproxen) trial, originally published in 2016 and reported at the American Heart Association Scientific Sessions that year.

And while we now have additional data on interactions between aspirin and NSAIDs, these new findings provide no additional clarity that might inform clinical practice. The AHA continues to recommend that clinicians minimize NSAID use in general and use the lowest effective dose when these agents are needed as outlined in the 2007 Scientific Statement from the American Heart Association.

PRECISION was a double-blinded post marketing study of patients who had osteoarthritis or rheumatoid arthritis and needed NSAID treatment. All study patients either had established cardiovascular disease (CVD) or were at high risk for CVD. Patients were randomized to celecoxib, ibuprofen, or naproxen. The trial compared the effects of celecoxib (up to 200 mg twice daily), naproxen (up to 500 mg twice daily), and ibuprofen (up to 800 mg three times daily) in terms of cardiovascular safety. A majority of PRECISION patients (69%) discontinued the study drug at some point during the course of the trial, there was an unknown number of crossovers to a non-study NSAID, and 27% of the original cohort discontinued follow up. About 45% of enrollees reported aspirin use at baseline in the PRECISION trial; this was used to assign them to the aspirin or non-aspirin group for the aspirin analysis. Patients were allowed up to 325 mg aspirin daily, but neither the timing of aspirin use, nor the actual dosage taken were reported.

The initial PRECISION data, published in 2016 in the New England Journal of Medicine, were followed by subanalyses of ambulatory blood pressure that was reported at the European Society of Cardiology Conference in 2017 and by the Journal of the American College of Cardiology report on co-administration of aspirin in 2018. Like the earlier two PRECISION reports, the JACC report comes with major gaps, caveats, statistical flaws, and limitations, noted Elliott Antman, MD, senior physician at Brigham and Women’s Hospital, associate dean for Clinical and Translational Research at Harvard medical School, and a past president of the American Heart Association.

Although the JACC report focuses on the concomitant use of aspirin and three NSAIDS, it contains no information regarding the actual dosing of aspirin during the study period. And while the JACC article reveals that 304 patients discontinued aspirin after randomization and 964 patients started aspirin during the study, there is no information regarding the timing of aspirin crossovers relative to cardiovascular events.

The aspirin analysis reported in JACC was pre-specified in the trial protocol, but the authors note that the pre-specified analysis was not designed to detect any interaction between the study NSAIDs and aspirin. The JACC analysis was not conducted on a strict intention-to-treat basis but on a post hoc “on-treatment” cohort that investigators reported to be taking study drug during follow-up. Rather than using the primary MACE endpoint of the original trial, investigators created a new post hoc endpoint that included cardiovascular, gastrointestinal, and renal components.

Like the original PRECISION trial, the JACC report is based on asymmetric dosing across the three arms of the trial. The celecoxib dose was low compared to the naproxen and ibuprofen doses, making it difficult to interpret the results because the doses of the drugs used across the different arms of the trial are not pharmacologically comparable. In addition, poor adherence to the PRECISION protocol and low retention negatively impacted the trial and subsequent analyses. More than a quarter of the original 24,000 patients enrolled in PRECISION dropped out or were otherwise lost to follow up, making it impossible to evaluate their outcomes. Even within the on-treatment cohort created for the JACC analysis, about 1.5 percent of patients had missing baseline data, which was filled in by imputation, adding a recognized statistical vulnerability to an already questionable analysis. Additional manipulations attempted to convert observational data regarding the use of aspirin into the appearance of a randomized dataset, but the lack of true randomization means there cannot be any correction for unmeasured confounders.

“These factors, the absence of correction for multiple analyses, insufficient details on the actual use of aspirin during the trial period, and the limitations of the main PRECISION trial make it difficult to provide any definitive recommendations to direct clinical practice,” Antman wrote in a JACC editorial that accompanied the PRECISION report. “It seems best to minimize the use of NSAIDs in general and especially in patients with CVD. If they (NSAIDs) must be used, the drug with the safest profile in the lowest dose for the shortest period of time remains the best advice for practice.”