Risk of stroke high when anticoagulant drugs are stopped

Fran Lowry

May 02, 2012

Durham, NC - A retrospective analysis of the data from the ROCKET AF trial, an international trial that compared rivaroxaban (Xarelto, Bayer) with warfarin for the prevention of stroke in patients with nonvalvular atrial fibrillation, shows that stopping these anticoagulant medications puts patients at high risk for stroke or systemic embolism [1].

Dr Jonathan P Piccini

The analysis also shows that the stroke rate after the end of the study was higher among patients in the rivaroxaban arm of the trial who had to come off rivaroxaban and transition back to warfarin. However, there did not appear to be a difference between rivaroxaban- and warfarin-treated patients following a temporary discontinuation or early permanent discontinuation for clinical indications.

The results of the analysis were presented at the American Heart Association's Emerging Science Series webinar by Dr Jonathan P Piccini (Duke University School of Medicine, Durham, NC).

"No matter what drug they are on, patients who need anticoagulation revert back to their intrinsic risk of stroke and embolism after discontinuation, so it shouldn't be done lightly," lead author Dr Manesh Patel (Duke University School of Medicine) said in a statement. "Unfortunately, it's unclear how to provide optimal anticoagulation coverage during periods of transition."

Rivaroxaban vs warfarin

ROCKET AF included 14 264 moderate- to high-risk AF patients and found that the newer anticoagulant rivaroxaban was as effective as warfarin in preventing stroke and systemic embolism. An advantage of rivaroxaban is that it can be taken once daily and does not require the continual monitoring of warfarin, which requires frequent dose adjustment.

In this new analysis, the researchers sought to explore some of the concerns that arose from the trial, including the increased rates of stroke and blood clots that were observed after discontinuing rivaroxaban at the end of the study, an observation that led to a warning in rivaroxaban's prescribing information.

They retrospectively evaluated patients who had a temporary discontinuation, early permanent study-drug discontinuation, as well as all patients who completed the trial and transitioned to open-label therapy for stroke and non-central nervous system (CNS) embolism and other thrombotic events, including MI and death up to 30 days' postdiscontinuation.

Their analysis showed that stroke and non-CNS embolism occurred at similar rates after both temporary and early permanent discontinuation with both anticoagulants.

After temporary discontinuation, there were nine such events with rivaroxaban and eight with warfarin (6.20 vs 5.05/100 patient-years; hazard ratio [HR] 1.28; 95% CI 0.49-3.31; p=0.62). After early permanent discontinuation, there were 42 events with rivaroxaban and 36 with warfarin (25.60 vs 23.28 per 100 patient-years; HR 1.10; 95% CI 0.71-1.72; p=0.66).

At the end of the trial, the risk for stroke, but no other events, was higher for the patients who transitioned from rivaroxaban to open-label warfarin than for patients already randomized to warfarin who transitioned to open-label warfarin.

In the rivaroxaban-treated patients, there were 22 strokes, compared with six in the warfarin-treated patients (6.42 vs 1.73 per 100 patient-years; HR 3.72; 95% CI 1.51-9.16; p<0.0044).

"When clinicians have patients with atrial fibrillation and have to stop their medication or transition their patients to another drug, that transition needs to be done very carefully, with the risk of stroke kept in mind," Piccini said in an interview.

The excess stroke risk that was seen with rivaroxaban in the mandated transition at the end of the study was probably due to the drug's shorter half-life, he said.

"Patients who were already taking warfarin in a blinded fashion continued taking warfarin in an open-label fashion, whereas the rivaroxaban patients had to start their warfarin as a new medication, and it took them time to build up a therapeutic INR and be protected," he said.

Theoretically, such a situation could happen in clinical practice, but it is likely to be unusual, Piccini continued. "Usually, patients stop their medication for a clinical indication, either because they have to undergo a procedure or some event occurs that mandates that they stop."

For now, doctors can tell their patients who are taking rivaroxaban that they can be confident that if their rivaroxaban has to be stopped, whether due to a clinical indication or to a bleeding event, there does not appear to be any difference in their risk of stroke relative to warfarin, he said. "But in the rare situations where someone has to transition from rivaroxaban to a different medication, something that would not be common in clinical practice but could certainly occur, say if someone's drug prescription coverage changed, it is important for them to understand that if they transition from a short-acting anticoagulant to a long-acting anticoagulant in the absence of a clinical event, there is going to be a gap in anticoagulation coverage that entails risk, and that risk needs to be addressed," he said.

Confirmed risk

Asked for comment on these findings, Dr Joseph P Broderick (University of Cincinnati, OH) pointed out that these new findings confirm that coming off these medications has risks. "There was definitely a risk, whether you're coming off the rivaroxaban or warfarin, for having an ischemic event—an MI or stroke—while you were coming off for whatever procedure."

Event rates increased about fivefold when patients were permanently discontinued, he noted. "They had to come off it because they were at risk for bleeding, but the point is there was a risk."

However, what is also clear is that transitioning from rivaroxaban to warfarin entailed a higher risk than from warfarin to open-label warfarin, Broderick said. At the time, rivaroxaban was not approved, so patients coming off anticoagulation in the trial all transitioned to warfarin, which would be understandably easier for those patients previously on warfarin.

"What that tells me is that even within a controlled trial, there's risk, and we probably need to better understand how best to make that transition," he said. "I think it's an important observation and underscores the importance of anticoagulation coverage during transition to get to a therapeutic state. That's important, and it's important whether you're on rivaroxaban or warfarin."

Last year, Broderick and colleagues published an article in Stroke on the contribution of withdrawal of antithrombotic medications to overall stroke risk. "We found that in the population, about 5% of all the ischemic strokes occur within the two months after someone gets stopped on warfarin, an antithrombotic, or even an antiplatelet," he said. "About half of those were related to an antithrombotic, which means we've got to pay more attention to the transitions, and we need to make sure we can do that better."

Strategies include using heparin to bridge the period before the oral medication gets to therapeutic levels. Rivaroxaban has a shorter half-life, which would have to be taken into account during transitions, he noted.

Patel reports financial relationships with Johnson & Johnson Pharmaceutical Research & Development, Daiichi Sankyo, and Genzyme Corporation. Piccini reports financial relationships with Bayer HealthCare Pharmaceuticals, Boston Scientific, and Johnson & Johnson Pharmaceutical Research & Development.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.