Anthony N. DeMaria, MD

There has been a debate about the value of aspirin as a prophylactic agent since I was an intern, which was a long time ago. Although aspirin seems to work very well in patients who have disease, its prophylactic value has been jumbled.

ASCEND is a well-designed randomized prospective trial. The aspirin was purely preventative, but the population was at higher risk because they had diabetes. How I interpret the findings is that you trade a lower incidence of ischemic events for a higher incidence of bleeding. This raises all kinds of questions. If I say to one of my patients, would you rather have an MI or a GI bleed, they would probably take the GI bleed. But people tend to equate these two, despite infarctions being more likely to be fatal. I don’t think they’re exactly equal. In the authors’ analysis, the number needed to treat to receive benefit was smaller than the number needed to treat to have harm. It would be an easier decision if this were all GI bleeding. But I worry about intracranial bleeding. Those were very small, but not zero.

The bottom line is that there is no role for aspirin as a prophylactic agent that has been documented at this point in time. But given the fact that most people would trade bleeding, which is usually treatable, for a reduction in major ischemic events, I tend to risk-stratify patients. If they are diabetic with no other risk factors, I probably go easy. If they are diabetic with risk for bleeding, then there is no way I would prescribe aspirin. But if they are diabetic and without risk for bleeding but have other risk factors such as hypertension, smoking or family history, I will tell them I will give them aspirin and it will reduce their risk for MI, and if it causes bleeding we will deal with it.

The omega-3 fatty acid portion was also well-done. The primary endpoint was negative. Frankly, I stopped recommending omega-3 fatty acids for my patients a number of years ago based on other negative trials. This is yet another negative trial in a higher-risk group of patients than usually studied. The 1-g dose may have been too small. People who take these supplements usually take 2 to 4 g per day. These are not very expensive, but they’re not cheap, especially for those using pharmaceutical-grade supplements. And then you get expensive urine.

Prakash Deedwania, MD, FACC, FAHA, FCCP, FACP, FASH

Even though the trial was done with very good intentions, and it studied diabetic patients supposedly at high risk, the overall event rate was low, which was surprising. I don’t know of any diabetes prospective trial that has had that kind of event rate. That tells me there may have been some selection bias. Most patients had a low or moderate vascular risk score. You get what you select for.

Omega-3 fatty acids have never been convincingly proven to reduce risk on top of statins. In this study, 70% of patients were on statins. So it is not entirely surprising that no benefit was found on top of statins.

Another problem with the trial is that the dose of omega-3 fatty acids was low at 1 g per day. Most of the trials being conducted now are with a dose of 3 to 4 g per day. It’s purely speculation, but we may have seen different results if the dose were higher. These days, most clinicians who prescribe omega-3 fatty acids prescribe 2 to 4 g per day. Of course, this study was started a long time ago, and hindsight is 20/20. We have learned so much more since the study began.

The aspirin portion of the trial provides some exciting data. I was pleased to see that aspirin does reduce the vascular event rate, but it is counterbalanced by the bleeding risk. We have many ways of balancing the risks against the benefits. Also of note, there are data from Germany showing that enteric-coated aspirin does not provide as good antiplatelet efficacy in diabetic patients. Extended-release aspirin may be a different story.

I don’t think these results will change guidelines. We have never recommended omega-3 fatty acids for diabetic patients, and that will continue until there are convincing data otherwise. This does provide further data that aspirin provides benefit in diabetic patients, which we have been saying all along.

Ileana L. Piña, MD, MPH

ASCEND was a well-done study with 12,000 patients. The investigators did a terrific job of conducting a pragmatic trial. We are not as good at these in the U.S. They were able to have pretty complete datasets using electronic records that talk to each other, which we in the U.S. do not have.

I have a lot of patients who tell me that they want to take omega-3 fatty acid supplements, and I tell them to eat fish; it tastes better. When I see diabetic women with HF, they do not do well, and I tell them not to spend the money on supplements, and to eat fish at least three times per week. It’s part of the Mediterranean diet. If they eat fish, then maybe they are not eating steak or fatty foods. The total substitution in their diet may help.

I will be talking to the American Heart Association guideline committee about looking into this issue. This is a well-done randomized trial that meets our criteria for inclusion in guidelines.