Medical Examiner

The Power of the Placebo

A recent study revealed that stents are likely ineffective at relieving chest pain. Why didn’t we know that?

Heart Stent
Dmitry Stolyariv, a heart surgeon, holds a stent as he performs a surgery in the Siberian city of Krasnoyarsk, Russia, on Sept. 28, 2016.

Ilya Naymushin/Reuters

Every so often, a new study comes along that challenges conventional wisdom in medicine or science. When the conditions are right, these studies can generate a lot of attention in both the popular press and the medical community. In early November, one of these such studies, called the ORBITA study, was published in the Lancet by a group of cardiologists.

The authors had set out to ask and answer a simple question: Does placement of a small wire mesh (called a stent) inside the artery that feeds blood to the heart (the coronary artery) relieve chest pain? One might ask what was novel about this question. The truth is that there was and is nothing novel about the question. The novelty was in the methods the authors used to answer the question: They conducted a prospective randomized controlled clinical trial, or RCT, the gold standard of research. The best RCTs compare the effect of the active intervention to a placebo and the best of the best keep both the subjects and the investigators blind to the intervention. The authors managed to do this for stents and chest pain, something that had never been done before, and in doing so, they had the best chance of preventing the placebo effect from skewing the results.

To understand how stents work, it helps to understand what causes chest pain. Our hearts have a system to feed oxygen rich blood to the heart muscle, through a network of coronary arteries. Arteries can become clogged with a mix of inflammatory cells and fats called plaques that accumulate over a lifetime. When plaques are severe enough, they can cause a restriction of flow causing a mismatch in oxygen supply and demand, and this can lead to what we cardiologists call “chest pain,” or angina. The presentation of angina typically happens in someone who is exercising and is reduced when the person rests, but over time, things can progress to the point where symptoms can leave people unable to do even basic daily activities, or the symptoms of angina can even occur at rest. Sometimes, the plaques can burst like a pimple and that can set off a reaction resulting in the formation of a blood clot cutting off the blood flow completely—resulting in a heart attack if left untreated.

The role of ruptured plaques leading to heart attacks was defined in the early 1980s by pathologists who studied the arteries of patients who had recently died of heart attacks and by a group of cardiologists who fed contrast dye into the coronary arteries of patients who showed up in the emergency room with heart attacks. This was groundbreaking and transformational research demonstrating that heart attacks were caused by ruptured plaques with clots overlying.

In the 1980s a group of doctors developed a technique to compress the plaque by inflating a small balloon advanced over a wire and placed directly in the coronary artery. When inflated at high pressure, this balloon effectively opens the blockage, a technique called angioplasty. Stents were developed to act as a scaffold to keep the artery open. It makes sense to think they would work to relieve pain. They have become the standard of care. Stenting is now used extensively to treat patients with what we call chronic stable angina, or chest pain, and it’s also used to open a completely blocked artery during a heart attack. While the technique is the same, there is one important difference: Angioplasties during a heart attack have been proven to saves lives. Angioplasties to relieve chronic angina have never been proven to actually reduce chest pain.

In 2007, a study much like ORBITA called COURAGE asked whether people with stable chest pain (not heart attacks) did better when treated with optimized medicines alone or medicines plus a stent. The takeaway was that there was no difference—medicines appeared just as good as stents, which generated a ton of attention and controversy. The resulting interpretation seemed to be that medicines were as good as stents when it came to preventing bad things like heart attacks and dying, but that stents were still better at relieving chest pain.

This was always an assumption. There had never been a gold-standard placebo-controlled trial asking whether stents actually improved symptoms better than just medicines alone. The reasons are probably many, but the biggest is that doing a placebo-controlled trial for stents is harder—you can’t just use a sugar pill as the placebo, you actually have to do a fake operation to mimic stent implementation even though half of participants don’t get a stent at all.

ORBITA has several limitations. The study was small (200 patients) as cardiology trials go, and it was limited to people with blockages in just one artery (many patients have multiple blockages). The patients were not very symptomatic, so some argue that this made it hard to detect an improvement with stents, among other smaller complaints, but in my opinion, these criticisms do not take away from the core results (though they may change how we interpret and apply them in the real world of clinical medicine). Ultimately, what ORBITA found was that there was effectively no difference between the patients who got stents and those who got dummy stents. Importantly, there was no difference in the rates of chest pain. None.

The big question, of course, is for patients who have stents. What does this mean for them? Well for one, the study showed no real difference, not that people with stents did worse. They just did the same whether they got the stent or not. In fact, both groups improved. So for patients with stents, there is no reason to panic. There is nothing to do. Again, many patients get stents during a heart attack, and nobody, even the most ardent skeptics, argues with that. Others get stents for chest pain without a heart attack. Those patients can and should know that the stent might not have had a big impact on their chest pain, but if they are feeling better, there is nothing more to do.

The big question for us in cardiology is how we will approach treating patients with chest pain going forward. This is just one study, and though it will likely inform recommendations that make their way down to practicing physicians, I can’t predict how it will all end up. It is very possible that given the results of ORBITA, we will work harder to treat chest pain with medicines before trying stents. I also expect there will be other well-done placebo-controlled trials looking at the effect of stents and other devices.

What ORBITA suggests is that the placebo effect is real and in play in these cases. It suggests that knowing you are getting a procedure can make you feel better. This is very interesting and will hopefully be the basis of future research questions. It’s also a reminder that the researchers’ effort in figuring out how to do a placebo-controlled study for a device like a heart stent was well worth it, and should be used again, to continue to help us understand how much of medicine is what we do, and how much is what people think we are doing.