After acute myocardial infarction (MI), bleeding risk increases proportionally with the number of antithrombotic medications used, such that patients on triple therapy consisting of aspirin, clopidogrel, and a vitamin K antagonist are approximately 4 times more likely to experience bleeding events than those who receive aspirin alone. Moreover, nonfatal bleeding raises the ongoing risk of recurrent MI and death, reports a Danish study published in the December 12, 2009, issue of the Lancet.

Using nationwide registers, researchers led by Rikke Sørensen, MD, of Copenhagen University Hospital Gentofte (Hellerup, Denmark), identified 40,812 patients aged ≥ 30 years who had been admitted to the hospital with first-time MI between 2000 and 2005. They calculated antithrombotic drug use based on  prescriptions for aspirin, clopidogrel, and vitamin K antagonists (typically warfarin) from the time of hospital discharge onward. Bleeding was defined as admission to a Danish hospital with a diagnosis of nonfatal bleeding or a diagnosis of bleeding as the cause of death.

During a mean follow-up of 476.5 days, 4.6% of patients experienced nonfatal bleeding that led to hospital admission. The frequency of these events increased over the course of the study. Most cases were gastrointestinal. Fatal bleeding, on the other hand, occurred in 0.3% of patients and remained at a steady level throughout the study. These events were mainly associated with treatment for femoral pseudoaneurysm. 

Bleeding incidence (both fatal and nonfatal) was lowest among patients taking aspirin alone and highest in patients either on clopidogrel plus a vitamin K antagonist or on triple therapy. Cox models showed that the hazard of bleeding increased proportionally with the number of drugs patients received (table 1).

Table 1. Fatal/Nonfatal Bleeding Associated with Antithrombotic Drug Use

Patient characteristics associated with higher bleeding risk were male sex (HR 1.33; 95% CI 1.21-1.47), age ≥ 80 years (HR 2.81; 95% CI 2.37-3.35), presence of malignant disease (HR 1.52; 95% CI 1.22-1.90), previous bleeding (HR 2.46; 95% CI 2.15-2.81), and concomitant use of several medications including loop diuretics, glucose-lowering drugs, NSAIDs, and PPIs. Curiously, patients who had undergone PCI had a slightly lower risk of bleeding (HR 0.88; 95% CI 0.78-0.99).

Nonfatal bleeding appeared to have long-term consequences; among patients who experienced such events, 37.9% had recurrent MI or died during the study period, compared with 18.4% of those without nonfatal bleeding (HR 3.00; 95% CI 2.75-3.27; P < 0.0001).

Surprising Risk with Clopidogrel Plus Vitamin K Antagonist

An unexpected finding was that “dual therapy with vitamin K antagonists and clopidogrel was nearly as hazardous as triple therapy,” Dr. Sørensen told TCTMD in an e-mail communication, adding that previous studies have found that the 2-drug combination has only moderate effects. “Until now, it has been assumed that treatment with a vitamin K antagonist plus clopidogrel was a good alternative to triple therapy in patients at high risk of bleeding,” she wrote.

In an editorial accompanying the paper, Erik L. Grove, MD, of Aarhus University Hospital (Skejby, Denmark), and Robert F. Storey, MD, of the University of Sheffield (Sheffield, United Kingdom), write that dissecting what part of this combination is responsible for bleeding can be tricky.

“The pharmacodynamic response to the platelet P2Y₁₂ inhibitor clopidogrel is variable and, by inference, patients with high levels of platelet inhibition will be at increased risk of bleeding, especially when combined with swings above the target international normalized ratio [INR] in patients treated with vitamin K antagonists,” they note, adding that new drugs with more predictable actions might improve the situation.

Stephen G. Ellis, MD, of the Cleveland Clinic Foundation (Cleveland, OH), noted in a telephone interview with TCTMD that a major weakness of the study is its lack of details about warfarin therapy. Without INR measurements, “[y]ou don’t know if [patients] were in range or out of range, or what the expected range was,” he said.

Deepak L. Bhatt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), said that the finding about elevated bleeding with clopidogrel and vitamin K antagonists should be interpreted cautiously. “That doesn’t really make biological sense,” he commented to TCTMD in a telephone interview, explaining that based on his understanding the combination should have approximately the same risk as aspirin and clopidogrel.

The study is observational and thus could suffer from confounding, Dr. Bhatt stressed, adding that another questionable finding is the lower risk of bleeding in PCI patients.

Dr. Sørensen agreed that the explanation for the reduced risk associated with PCI could be residual confounding. Data published in the paper’s appendix show that PCI patients were generally younger, had fewer comorbidities, and were more often treated with concomitant medications such as statins, she explained, adding that the effects might have remained even after multivariable adjustment.

Findings Relevant for Real World

The study’s strengths are its large size and the fact that it represents the “real world,” said Dr. Ellis. The actual bleeding risk in this population may be “well underestimated by randomized clinical trials,” he said, because subjects who enter those studies tend to be less sick than average patients. “Now for the first time, we have estimates of what some of these risks are, and they’re substantial.”

Dr. Bhatt pointed out that vitamin K antagonists have many cardiovascular indications and “when you lump all those different patient types together, that’s a fair number of cardiology patients.” When these patients receive stents, they are candidates for triple therapy.

In the Danish study, only 0.77% of patients initially received triple therapy, but that number rose to 3.6% when additional patients crossed over from other regimens, Dr. Sørensen reported.

Managing Risks a Challenge

Dr. Ellis said balancing all these risks in patients is challenging.

“The first step is recognizing that the drugs we normally use with stenting don’t mesh too well with [warfarin], and that we need to be very careful about using drug-eluting stents in patients that have to be on [warfarin],” he said. “The next step is to try and sort out which of these patients are really at high risk. There are a number of scoring algorithms that could potentially be applied here to help triage patients.”

Avoiding triple therapy should be a priority, said Dr. Ellis, “but if you’re stuck, one would like to think if you ran [warfarin] at a lower INR than you normally would, say 1.8 to 2 or something like that, if you can get it in that range, then you would minimize risk. But we really don’t know.”

For his part, Dr. Bhatt advised that when it is necessary to stop antithrombotic therapy in a patient who has a bleeding complication but also is at high risk for ischemic events, “it’s important to be vigilant” and to take symptoms such as recurrent chest pain seriously. “Something that is biologically very true but sometimes forgotten by clinicians is that if someone’s bleeding, the natural response is [for the body] to try to stop that bleeding,” he said. Thus, halting antithrombotics can set up a hypercoagulable state and increase the likelihood of ischemic events.

Sources:

1. Sørensen R, Hansen ML, Abildstrom SZ, et al. Risk of bleeding in patients with acute myocardial infarction treated with different combinations of aspirin, clopidogrel, and vitamin K antagonists in Denmark: A retrospective analysis of nationwide registry data. Lancet. 2009;374;1967-1974.
2. Grove EL, Storey RF. The right oral antithrombotics in acute coronary syndromes. Lancet. 2009;374:1947-1948.

Disclosures:

• The study was funded by the Danish Heart Foundation and Danish Medical Research Council.
• The study authors report no relevant conflicts of interest.
• Dr. Grove reports receiving lecture fees from Pfizer and travel support from Pfizer and MSD.
• Dr. Storey reports receiving honoraria, consultancy fees, institutional research grants, and/or travel support from AstraZeneca, Eli Lilly, Daiichi Sankyo, Schering Plough, The Medicines Company, and Teva.
• Dr. Ellis reports consulting for many of the major drug-eluting stent companies, particularly Boston Scientific, Cordis, and Abbott.
• Dr. Bhatt reports that his institution receives research funding from a number of companies involved with development of antithrombotic drugs.

Related Stories:

• Managing Antiplatelet Meds Tricky When Patients Need GI Endoscopy
• Short-Term Break from Thienopyridines OK, but Not from Aspirin
• Bleeding Risk Low Yet Persistent with Prolonged Dual Antiplatelet Therapy