Coronary angiography is being overused

Originally published in MedPage Today

by Kristina Fiore, MedPage Today Staff Writer

For patients without known heart disease, the diagnostic yield of coronary angiography appears to be low, researchers found.

Only about 38% of those referred to the cath lab for the elective procedure had obstructive coronary artery disease, Manesh R. Patel, MD, of Duke University, and colleagues reported online in the New England Journal of Medicine.

“In patients without known heart disease, we should carefully re-evaluate our process for determining how we send people to the cath lab,” Patel told MedPage Today. “It is still a very important place for patients with acute MI, unstable symptoms, and certainly for patients for whom we’re concerned about artery blockages.”

But, Patel continued, physicians need to “maximize the workup — what’s the most efficient way to do all the tests in order to determine if someone has obstructive coronary artery disease.”

Current guidelines for triaging patients for cardiac catheterization recommend a risk assessment and noninvasive testing.

Yet the methods for determining how invasive tests are ordered “are far from perfect,” said Cam Patterson, MD, PhD, of the University of North Carolina at Chapel Hill, who was not involved in the study. Patterson also noted that “having an angiogram in 2010 doesn’t mean that a patient is buying himself a stent or a bypass surgery,” he wrote in an e-mail to MedPage Today.

To investigate patterns of noninvasive testing and the diagnostic yield of catheterization among patients with suspected coronary artery disease, Patel and colleagues assessed 398,978 patients seen at 663 hospitals between January 2004 and April 2008 who were logged in the American College of Cardiology National Cardiovascular Data Registry.

Patel noted that this population accounts for roughly 20% of the patients going into a cath lab. The vast majority are seen for other indications including heart attack and heart failure. The researchers also excluded patients who’ve had a prior heart procedure such as a valve replacement or stent.

They defined obstructive coronary artery disease as stenosis of 50% or more of the diameter of the left main coronary artery, or 70% or more of a major epicardial vessel.

Median patient age was 61; 52.7% of the sample population was male.

At catheterization, the researchers found, 37.6% of patients had obstructive coronary artery disease.

The percentage was similar — 41% — when the definition of obstructive disease was expanded to include stenosis of 50% or more of any coronary vessel.

There was also a similar percentage of patients with no coronary artery disease: 39.2%.

“That [only] 38% had an occlusion … was surprising to us,” Patel said. Yet he noted that when data were included on patients with known disease or a previous condition or procedure, the rate of detection increased to 60.3%.

And being able to rule out a heart attack is reassuring for both patients and doctors, said Ralph Brindis, MD, of the University of California San Francisco and an author of the study.

“A normal cath is incredibly reassuring and may actually save money in the long run as the patient and the physician don’t repeatedly do noninvasive tests and repeatedly admit the patient to the hospital based on their anxiety or uncertainty,” Brindis said.

“The real question here is what is the ‘correct’ percentage of normal caths in this patient population that should be the ideal benchmark,” he added. “We do not yet know this answer.”

The researchers also assessed noninvasive testing prior to angiography, and found that 84% of patients had some form of noninvasive test — whether it was electrocardiography, echocardiography, CT angiography, or a stress test. The database did not provide information on specific tests.

Those with a positive result on a noninvasive test were moderately more likely to have obstructive coronary artery disease than those who didn’t have any testing (41% versus 35%, P<0.001; OR 1.28, 95% CI 1.19 to 1.37).

The researchers also saw risk factors for disease that were congruent with previously known risk factors:

* Male sex (OR 2.70, 95% CI 2.64 to 2.76)
* Older age (OR 1.29, 95% CI 1.29 to 1.30)
* Presence of insulin-dependent diabetes (OR 2.14, 95% CI 2.07 to 2.21)
* Presence of dyslipidemia (OR 1.62, 95% CI 1.57 to 1.67)

Based on those findings, Patel said that “every step along the way in determining who gets to the cardiac cath lab needs to be improved.”

Step one, he said, is to determine more efficiently a patient’s risk for significant blockages: “Can we improve our metrics in ways of identifying patients with chest pain who might have obstructive coronary artery disease.”

The second is to determine which noninvasive test might be most appropriate for this population instead, and several trials investigating this are under way.

“We need those types of studies to tell us who should be getting noninvasive tests and how well they work,” Patel said.

In an accompanying editorial, David J. Brenner, PhD, of Columbia University Medical Center in New York City, wrote that the study is correct in suggesting the need for optimizing “the application of gatekeeper tests such as myocardial perfusion scintigraphy in order to decrease the disturbingly large proportion of invasive coronary angiographic procedures that yield negative results.”

With so many high-tech imaging tools available, he wrote, “it is essential to optimize their use.”

The researchers acknowledged that the study was limited because they couldn’t distinguish between types of noninvasive tests, and they had no information on the “undoubtedly large population” of patients who were evaluated but didn’t undergo catheterization.

Visit for more cardiology news.

Leave a Comment

Most Popular

✓ Join 150,000+ subscribers
✓ Get KevinMD's most popular stories