Anticoagulate in Heart Failure -- Do We Have an Answer?

Ileana L. Piña, MD, MPH; Shunichi Homma, MD


April 19, 2012

This feature requires the newest version of Flash. You can download it here.

Introduction: Heart Failure and Anticoagulation

Ileana L. Piña, MD, MPH: Hi. I'm Ileana Piña and this is my blog. I'm the Associate Chair of Academic Affairs at Montefiore Medical Center and Albert Einstein College of Medicine in the Bronx. Being in New York gives me a terrific opportunity to meet some very important people that we are going to talk to.

Today we have Dr. Shunichi Homma, Associate Chief of Cardiology at Columbia University College of Physicians and Surgeons. He has been working for many years on a very important trial, which we are going to talk about today. It is called WARCEF.[1]

This trial is very important because for years, we in the heart failure world have had many discussions and arguments among ourselves about what to do about anticoagulation. We look at echocardiograms and see very big ventricles where the blood is literally just swirling around, and we wonder what is going to happen to those patients. What if they have a stroke? In the 1990s, when I was at Temple University, we used to anticoagulate everyone. Then we started looking at the literature and saw that it wasn't that clear. There are studies that have shown that if we tried aspirin, it may decrease or negate the effects of angiotensin-converting enzyme (ACE) inhibitors, and we lived by ACE inhibitors in the heart failure world. We needed a study to give the right answer, and I think this is what we have.

I would like to introduce Dr. Homma from Columbia University. Welcome. It's really good to have you here. You have been working very hard these past few years.

The WARCEF and WATCH Studies

Dr. Piña: Tell us about the WARCEF trial. What does WARCEF stand for and what type of trial was it?

Shunichi Homma, MD: Thank you for having me. WARCEF stands for Warfarin vs Aspirin in Reduced Cardiac Ejection Fraction. As you have just mentioned, the issue about how to treat patients with reduced left ventricular ejection fraction and sinus rhythm in heart failure has been a question for many years. Out of the approximately 5 to 6 million patients each year with heart failure in the United States, some of them have atrial fibrillation, which means that they are not the target of our study.

Dr. Piña: But we have to anticoagulate them anyway.

Dr. Homma: Exactly. And the question remains for the rest of the patients, which is a very large number of patients. As you have mentioned, whether we should use warfarin or aspirin has not been clarified. The latest study was the Warfarin and Antiplatelet Therapy Study in Chronic Heart Failure (WATCH) study, which enrolled approximately 1500 patients. The study had 3 arms, so the number of patient-years in the aspirin and warfarin arms was about 2000. It was a rather small study that was stopped early.

Dr. Piña: Did we get any answers from WATCH?

Dr. Homma: Some answers. What was interesting about WATCH was that it showed that heart failure hospitalization was increased by aspirin, possibly because it had counteracted the effect of ACE inhibitors. With that as the background, we have our study, the WARCEF study, which enrolled 2305 patients with a mean follow-up of 3.5 years.

Dr. Piña: How long did it take to get here?

Dr. Homma: Many years.

Dr. Piña: So it was a very long study.

Dr. Homma: Yes. Initially, the study started in 2002 with the first enrollment, so it was a long time. But we finished, which is the good news. The randomization was to warfarin 325 mg.

Dr. Piña: It was one-to-one randomization?

Dr. Homma: One-to-one randomization or to warfarin, with a target international normalized ratio (INR) of 2.75. We achieved an INR of 2.5 in the warfarin arm.

Dr. Piña: That's still pretty good for a clinical trial.

Dr. Homma: Yes. After the mean follow-up of 3.5 years with a composite primary endpoint consisting of death, ischemic stroke, or intracerebral hemorrhage, we compared both arms.

WARCEF Results: Why the Small Event Rate?

Dr. Homma: We found that there was no difference between the 2 arms. The annual event rate was approximately 7.5% in the warfarin arm and 7.9% in the aspirin arm.

Dr. Piña: That's not a huge number of events. Would you agree?

Dr. Homma: Yes, and it has been brought up by heart failure experts that, given the heart failure population, death rate was rather low -- about 6.5% or so.

Dr. Piña: Which puts them somewhere around a class II [in the New York Heart Association functional classification system]. Maybe even some class I patients were in there and a smattering of class III patients, because we're seeing that class II or III's annual mortality is about 8%. It's good because it fits the literature. It fits what we know.

Dr. Homma: Yes. Approximately 55% of patients were class II, 30% were class III, and about 12% were class I. There were very few class IV patients. These patients are very difficult to enroll. In terms of the characteristics of the patients, the mean age was 61, about 80% of patients were men, and about 13% of patients had previous ischemic stroke or transient ischemic attack.

Why WARCEF Was a Good Clinical Trial

Dr. Piña: It wasn't a fully enriched population in stroke, but my experience has been that heart failure patients in general don't have many strokes. If you look at the cause of death or events on event committees, the number of strokes is small. Most of what they have is hospitalizations for heart failure, which I know you tracked as well.

Dr. Homma: Yes, we tracked that also. An important point about the study that I should mention is that it was double-blinded; therefore, this is the first warfarin-vs-aspirin comparison in heart failure using a double-blind design.

Dr. Piña: That's the value of it.

Dr. Homma: It is valuable because it is very important to avoid the placebo effect. Also, the adjudication, obviously, was independently performed. And importantly, for the entry criteria, we made sure that all of the patients who failed the left ventricular ejection fraction criteria were using a centralized echocardiogram.

Dr. Piña: You had a core lab and you had an independent adjudication committee. These, to me, are the pieces of a good clinical trial in which people are truly blinded and the blinding isn't broken. I think that's very important.

What Was the Bleeding Rate From Warfarin?

bleed, they're older, and they may have other risks for bleeding.

Dr. Homma: With warfarin, one of the major complications is bleeding. In terms of the bleeding rate per year, it was approximately 1.8% in the warfarin arm for major hemorrhage, as opposed to about 0.8% or 0.9% with aspirin.

Dr. Piña: Almost double.

Dr. Homma: Basically double the rate, which is very consistent with previous studies looking at bleeding rate in these patients.

Dr. Piña: What kind of bleeding was the most common?

Dr. Homma: The most common were gastrointestinal and genitourinary hemorrhage.

Dr. Piña: Were there any deaths because of bleeding?

Dr. Homma: A few, but not too many -- much fewer than we expected with this many patients.

Would a Subgroup Analysis Find Any Differences?

Dr. Piña: Was there any subgroup that gave you a hypothesis that should be tested?

Dr. Homma: Yes. You bring up a very interesting point. We analyzed a large group of patients, over 2300, but these patients were tied together by a common factor in that they were in sinus rhythm and happened to have a low ejection fraction. However, there are a variety of groups of patients; some are old, some are young, some are women, some are men, and of course there are other variables, such as cause of heart failure.

Dr. Piña: Ischemic or nonischemic? What was your percentage of ischemic patients?

Dr. Homma: About 60% were ischemic.

Dr. Piña: So you had a large number of ischemic patients. Was there any interaction between etiology of heart failure (ischemic vs nonischemic) and the endpoints?

Dr. Homma: That's a very interesting question because that's exactly what we're trying to tease out at this point. Overall, there was no significant difference, but perhaps within certain groups there is a difference in the effectiveness of medication. Because it was not a primary endpoint of the study, we are not publishing it in the manuscript that will be coming out.

What Happened With Heart Failure Hospitalizations?

Dr. Piña: Tell me about the heart failure hospitalizations. I know that it's something your adjudication committee looked at very carefully because of previous data, especially from WATCH, about heart failure hospitalizations. Was there anything different?

Dr. Homma: As you mentioned, the WATCH study showed increasing heart failure hospitalizations with aspirin, as did another smaller study comparing warfarin with aspirin. However, in our study, there was no difference. In fact, there was a trend towards increased heart failure hospitalization in the warfarin group. It is very interesting and quite different from what we have seen, but I'm very confident about the adjudication process that took place, particularly because the adjudication chairperson happened to be a heart failure specialist.

Dr. Piña: A good one, too, and experienced. It makes a lot of difference when you have a good adjudication committee that is experienced and can really look at the fine details.

I think an interesting point to be made is about post-hoc analysis and subgroup analysis, which very often lead us to think, "This must be real, this must be true." Then when we test it, it's not, and there's the hospitalization. That's the perfect example, in WATCH: You didn't see the heart failure hospitalizations with the bigger trial -- probably better trial -- without a placebo group. It just wasn't there. For our audience, it's learning that subgroup analyses can have its problems and that, very often, at least in heart failure, we have been fooled by subgroup analyses. Then when we truly test prospectively, it's not there.

A Global Initiative: Any Differences?

Dr. Piña: I know that this was a huge international trial. How many countries did you have?

Dr. Homma: We had 11 countries involved, including countries in Europe and South America in addition to the United States and Canada.

Dr. Piña: Who enrolled more patients?

Dr. Homma: It's actually a 50% American product and 50% foreign.

Dr. Piña: Did you see any difference? Because, again, some trials have seen a benefit in the US population but not abroad, and vice versa. Was there any difference?

Dr. Homma: You bring up a very interesting point. We analyzed by country and continents. There were some differences, but we need to understand the reason for the difference. For example, some of the subgroup composition may be different, heart failure classes may be different, etiology may be different, and another issue is the INR. The effectiveness of warfarin or time in therapy may be different. All of these have to be adjusted and figured out.

Dr. Piña: These are all the challenges that we go through with international trials.

What Are the Clinical Implications of WARCEF?

Dr. Piña: If you had to give advice to us, the working cardiologists, what would you tell us? What should we do with our patients?

Dr. Homma: The overall result of this study was that there was no difference between the effectiveness of warfarin vs aspirin. This applies to patients with low ejection fraction and sinus rhythm, all combined.

Dr. Piña: And well medicated.

Dr. Homma: Yes, and well medicated. It's very important to mention that almost 100% of patients were using ACE inhibitors, angiotensin II receptor blockers, or beta-blockers.

Dr. Piña: That's a credit to the steering committee.

Dr. Homma: Yes, it was a very well-treated group of patients.

Dr. Piña: So, in a well-treated group of patients, there is no difference.

Dr. Homma: There was no difference overall.

Dr. Piña: How about the ischemic patients? We use aspirin in ischemic patients. Should physicians keep using aspirin even when the ejection fraction is low?

Dr. Homma: One can infer that, overall, there shouldn't be a difference, but again, I would like to hold some of the recommendations until we look further into the study.

Dr. Piña: But at least for now, the decision can be made on purely clinical grounds. I shouldn't think that if I start aspirin I'm going to be hurting this patient in any way, or if I put them on warfarin. So, I think the decision can be made clinically.

Dr. Homma: At this point, yes.


Dr. Piña: I want to thank you for coming here. I wish you luck. Is your manuscript going to be seen soon?

Dr. Homma: Yes, it's in press at this point.

Dr. Piña: Wonderful. You presented this at the stroke meetings, right?

Dr. Homma: Yes, at the American Stroke Association's International Stroke Conference.

Dr. Piña: Are you going to publish any other papers?

Dr. Homma: There will be a presentation at the annual meeting of the American Academy of Neurology that is coming up in April, as well as the European Society of Cardiology Heart Failure Congress. We are in the process of preparing a manuscript to look further into the data to see if there might be certain groups that have differential effects. We are also interested in the potential use of new types of antithrombotic agents. Because warfarin is being replaced for many indications, perhaps new antithrombotics may be used for certain patients.

Dr. Piña: Maybe that's your next study.

Dr. Homma: WARCEF 2 -- and I'm sure we'll be talking again.

Dr. Piña: Let's not tell that to the National Institutes of Health, with all the money that they spent. Anyway, congratulations. I'm very proud of all of you. I think you have done an outstanding job. This was a difficult trial to do, managing all those countries, patients not following therapy, coming off therapy, and so many events that these patients have, particularly the hospitalizations. I wish you luck with the manuscript. We look forward to seeing it. I want to thank Dr. Homma for being with us today and for doing such excellent work. Hopefully I will see you very soon. This is Ileana Piña and I'm signing off. Have a great day.


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.