On the meaning of a coronary blockage. Chapter 2 of the mini-series

What follows is by far the most interesting effect in all of medical science

Last week I wrote about the COURAGE trial. At the time, in 2007, the cardiology world was stunned to learn that opening partially blocked coronary arteries—in stable patients—did not lead to fewer heart attacks or less cardiac death compared to tablets alone.

I chose the verb stunned because nearly the entire focus of cardiology at the time was the search for ischemia due to these blockages—so that they could be “fixed.”

Od habits are hard to break, so, post-COURAGE, cardiologists didn’t stop doing stress tests, then angiograms, and then stents. They just changed their reasoning: the idea went from we persist on this path to open clogged arteries not to extend life but to relieve angina. How could anyone be against the relief of suffering?

Recall that graph from COURAGE which showed the higher rates of patients who were angina-free for the first three years in the invasive arm.

So the engine of cardiology, stress tests→angiograms→ stents continued to roar.

Ten years passed. Then came the next shocker.

A group in London, led by a strong woman named Rasha Al-Lamee, had this idea: what if the relief of angina seen in the COURAGE trial was due to placebo effect?

The history of medicine is replete with examples of placebo effect. I feel no need to give examples. All you need to consider is that many (if not most) new drugs gain regulatory approval by passing muster against a placebo.

Recall also that in COURAGE, one group got a procedure and were likely told—perhaps shown—that they were fixed. The other group knew they weren’t fixed.

The idea behind the ORBITA trial was to find patients with a blockage in a single vessel, and then test whether opening that blockage led to relief of angina—against a placebo. They (primarily) measured angina using exercise time on a treadmill.

ORBITA may be one of the most brilliant trials of this generation.

They found 200 patients who had a single-vessel partial blockage who were willing to delay their stent procedure for a few weeks to participate in this experiment. Half would be randomized to have the stent; half would undergo a placebo (sham) procedure in which a flow wire was placed across the lesion, but it was left unfixed. Yes, that is right, some patients had their widow-maker left alone.

No one in this trial would know the treatment arm. The patients wore headphones during the angiogram procedure. When they left the cath lab, none of the doctors or nurses would know. The elaborate and elegant blinding procedures were necessary.

Before the procedure all patients were stabilized on proper meds. Their exercise time and degree of symptoms of angina recorded on a data sheet.

You might wonder how they came up with 200 patients? This, too, was brilliant thinking. There had never been a placebo controlled trial of opening an artery. But there had been two placebo-controlled drug trials that documented a 45-55 second increase in exercise time.

The ORBITA team powered their trial to detect a 30-second increase in exercise time after the stent. This was a conservative estimate because if a tablet increased exercise time by 50 seconds, surely a total relief of blood flow would do better.

Now to the Results:

A repeat treadmill stress was done at six weeks after either the real stent or the sham procedure.

The increase in exercise time did NOT reach statistical significance. It was only 16 seconds. A later re-analysis increased the time to 21 seconds, which was still well below statistical significance.

They also found no difference in the proportion of patients with improvement in one or two classes of angina. There were no differences in physical limitation scores.

One interesting finding came from looking at how the heart’s walls moved after the two procedures. Using an echocardiogram (ultrasound), they found that muscle downstream from the stented vessel looked a lot better in the treatment arm vs placebo. This means the stent actually worked to relieve the blockage and improve heart function—BUT IT MADE NO DIFFERENCE IN SYMPTOMS.

The Attacks:

Because ORBITA turned upside down the idea—from COURAGE—that stents relieve angina, they faced blistering attacks from the believers. Eg: The trial was too short, too small, recruited special patients, etc.

In one of the most candid and entertaining pieces of academic writing that I have read, two ORBITA authors swatted these concerns away—often with good doses of British ridicule. Please do read this “Authors reply.”

Here is one delicious response:

We suggest that Rahul Muthalaly and colleagues might have misunderstood ORBITA and possibly the purpose of randomisation and blinding. Patients in both arms had the same medical therapy initially and were then randomised to the PCI and placebo groups. Therefore, ORBITA really is a comparison between PCI and placebo.

---

The take-home messages from ORBITA are numerous:

First is that the way forward to knowing things in medicine requires randomization. When clinicians—not randomization—choose a therapy, and you look at results after the fact, you can never know if the patients were similar.

Second is that nearly everything we do in medicine should pass muster against a proper control arm. That might be a placebo, a sham procedure, or the standard of care. Blinding must be maintained in trials.

Third is that gaining real knowledge requires courage. The ORBITA investigators actually published every angiogram in the supplement. They did this to show doctors that they included patients with terrible looking blockages.

In my field of electrophysiology, there is great fear in doing a sham-control for ablation of AF. Many say it would not be ethical. Or that we know ablation works. I encourage doubters to open up the supplement and look at those angiograms in ORBITA. Electrophysiologists lack the courage of Rasha Al-Lamee and colleagues.

---

Fortunately, the ORBITA investigators are not resting.

They are in the process of studying the placebo effect of stenting patients with more serious angina and possibly multiple partial blockages. They will call this ORBITA-2.

---

Thanks for your enthusiastic support of Sensible Medicine. We are excited to bring you medical thinking free of advertisements.