RCTs in the Age of SoMe: ISCHEMIA Trial Publishes Baseline Data

Debra L Beck

March 13, 2019

Anticipating presentation and publication of the full results sometime later this year, investigators with the International Study of Comparative Health Effectiveness With Medical and Invasive Approaches (ISCHEMIA) trial have published their baseline data and once again stirred discussion about whether the trial will really answer the question it was originally designed to answer.

"This is a very high visibility trial," study chair Judith S. Hochman, MD, NYU Langone Health, New York City, told theheart.org | Medscape Cardiology. "In this era of social media there have been commentaries out there that motivated us to report the baseline criteria for the trial, both so that the community can understand the population that was randomized in this trial and to show that we achieved our goals in terms of randomizing the target population that we sought to randomize."

ISCHEMIA is designed to better define the role of revascularization in stable ischemic heart disease (SIHD), which was called into question when the COURAGE and BARI 2D trials showed no reduction in death or myocardial infarction (MI) for percutaneous coronary revascularization (PCI) added to optimal medical therapy (OMT) over OMT alone in patients with SIHD.

In brief, of the 5179 patents randomly assigned to either OMT or OMT plus early revascularization, median age was 64 years, 22.6% were women, 33.7% were nonwhite, 41% had diabetes, 19% had had a myocardial infarction, and 90% had previous angina.

Stress imaging was the qualifying test for 75% of patients. To boost enrollment in countries where cardiac imaging is not commonly performed, the trial allowed entry on the basis of an exercise electrocardiogram (ECG). One-quarter of the participants were enrolled on the basis of nonimaging exercise tolerance testing (ETT).

"ETT is the most common stress test used worldwide, and the [American College of Cardiology/American Heart Association] guidelines recommend treadmill testing when you want to do a stress test for those who are able to exercise and have an interpretable baseline ECG," said Hochman.

Among participants randomized after stress imaging, core laboratory assessment of ischemia severity was severe in 44.8%, moderate in 41.0%, mild in 8.1%, and nonexistent or uninterpretable in 6.1%.

"In regular clinical practice there is no core lab, and you can see that trying to explain the trial, consent a patient, have them wait for the core lab review, and then have a CCTA [coronary CT angiography] and wait for core lab review is cumbersome," said Hochman.

"We debated this point upfront and wanted to balance trial applicability to clinical practice and rigor. We also wanted to avoid unnecessarily impeding the work flow and, therefore, from the outset allowed some centers that had high concordance with the reads of the core lab to randomize their patients without waiting for the core lab review. We subsequently increased the number of sites with this permission," she explained.

Among the 25% of participants randomized after nonimaging ETT, core laboratory determination of ischemia severity was severe (an eligibility criterion) in 83.0%, moderate in 8.0%, mild in 2.7%, and nonexistence or uninterpretable in 6.3%.

Of those who also underwent CCTA, 79% had multivessel disease, 87% had left anterior descending coronary artery disease (LAD), and 47% had proximal LAD disease.

Participants undergoing ETT had a greater frequency of three-vessel CAD, LAD, and proximal LAD stenosis than participants undergoing stress imaging.

"I think it should be reassuring to the community that we randomized the right patients. These patients had more extensive coronary disease than COURAGE or BARI 2D, more triple-vessel disease, more LAD disease, and more proximal LAD disease, by far," said Hochman.

But it is this use of ETT and the non-negligible proportion of patients who, on core lab reading, were found to have less than qualifying levels of ischemia that worries editorial writer Raymond Gibbons, MD, from the Mayo Clinic, Rochester, Minnesota.

"This step certainly looks like it has a cost in terms of the patients who got enrolled who didn't then meet the core lab criteria, and that cost is decreased power of the study, said Gibbons. However, he stressed to theheart.org | Medscape Cardiology that this is a "glass half empty and half full" story.

"Their baseline data suggest that they did a wonderful job of truly delivering contemporary optimal medical therapy and that's not been done before," said Gibbons, adding that if the results go against revascularization, they should be viewed as a success for OMT in stable ischemic heart disease.

Antiplatelet medications were being taken by 94% of patients in the trial, antiplatelet or anticoagulation by 96%, statins by 95%, and antihypertensives or antianginal medications by 97%. Notable exceptions were angiotensin-converting-enzyme inhibitors, which were taken by only 41.4%, and high-intensity statins, which were taken by 37%.

"But these patients had a baseline median low-density lipoprotein level of 83, which is pretty darn good," said Gibbons.

The ISCHEMIA trial has come under some scrutiny in recent months. Last summer, an editorial critiquing four aspects of the trial's design was posted and then removed from the American Heart Journal website, presumably to correct some errors. The paper was ultimately reposted unchanged.

A few months before that, the Twittersphere erupted with criticism when it was revealed that the trial had changed its primary end point from the composite of death and MI to include also resuscitated cardiac arrest, hospitalization for unstable angina, and hospitalization for heart failure.

These last two end points, specifically, are "vulnerable to bias in an unblinded trial," even if the end points themselves are adjudicated, an editorial pointed out.

In truth, the trial was originally approved with the five-component end point as the primary end point, as the study leadership explained in an editorial published in Circulation in August 2018.

"After funding, we changed to the two-component end point hoping we would be powered, but then changed back to original," said Hochman.

Hochman is study chair and clinical coordinating center principal investigator for the International Study of Comparative Health Effectiveness With Medical and Invasive Approaches (ISCHEMIA) trial for which, in addition to support by a National Heart, Lung, and Blood Institute grant, devices and medications were provided by Abbott Vascular, Medtronic, Inc, St. Jude Medical Inc, Volcano Corporation, Arbor Pharmaceuticals LLC, AstraZeneca, Merck Sharp and Dohme Corp, Omron Healthcare Inc, and financial donations came from Arbor Pharmaceuticals LLC and AstraZeneca. Gibbons reported personal fees from Medtronic.

JAMA Cardiol. Published online February 27, 2019. Abstract, Editorial


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.
Post as: