The Stent Trial That Compared It to a Fake Procedure

Until 2017, no one had ever run a placebo-controlled trial on coronary stents. Doctors had implanted millions of them in patients with stable angina — the squeezing chest pain that comes on with exertion. The reasoning seemed self-evident: a narrow artery hurts, you prop it open, the patient feels better.

The ORBITA trial, led by Rasha Al-Lamee at Imperial College London and run across five UK centres, did the test that surgical specialties almost never do. Two hundred patients with severe single-vessel stenosis were sedated and wheeled into the catheterization lab. Half got a real percutaneous coronary intervention. Half got a sham: catheters were threaded in, the patient was kept on the table for at least fifteen minutes, and the catheters came out with nothing deployed. Neither the patients nor the cardiologists who later assessed them knew which arm they had been in.

Six weeks later, both groups walked on a treadmill. The stent patients lasted, on average, 16.6 seconds longer than the sham patients. The 95 percent confidence interval ran from minus 8.9 to plus 42 seconds. The p-value was 0.20. By the standards of any other specialty, the gap between a real stent and an imaginary one had vanished into noise. The result, published in The Lancet in November 2017, detonated.

Scope matters here. ORBITA was about stable angina — patients not in the middle of a heart attack. Stenting an actively occluding artery during a heart attack still saves lives, and nothing in the trial disturbs that. What it disturbed was the much larger elective population getting stents on the assumption that the metal, not the ritual around it, was doing the work.