Noninvasive CT scans are nearly as accurate at imaging coronary artery blockages as conventional angiography and are much safer for many patients, according to researchers who published a study today in the New England Journal of Medicine.
Angiograms are considered the gold standard for detecting blockages in the arteries. But the procedure involves inserting a guide wire and catheter into the groin, threading them through the blood vessels to the heart and injecting a dye that allows blockages to be seen in an X-ray.
Using a CT machine instead to make a three-dimensional image of the heart could eliminate the risks involved with traditional angiograms, which include heavy bleeding, damage to blood vessels and even death, said Dr. Julie Miller, an interventional cardiologist at Johns Hopkins University School of Medicine in Baltimore and lead author of the study.
More than 1.2 million patients in the U.S. undergo cardiac angiograms each year, and 1% to 2% of those cases result in complications, according to the American Heart Assn.
The National Center for Health Statistics at the Centers for Disease Control and Prevention estimates that 25 people die every year as a result of the procedure.
About 20% to 30% of those tests give patients a clean bill of health -- and that means hundreds of thousands of people are exposed to needless risk, Miller said.
Many cardiologists see CT scans as a safer alternative because the scans are powerful enough to create a high-resolution image even though the contrast dye is administered by a simple intravenous line and thus more dilute.
Miller and her colleagues at nine hospitals in the U.S., Canada, Germany, Japan, Brazil, Singapore and the Netherlands identified 291 patients with symptoms of coronary artery disease who were candidates for traditional angiograms. Their median age was 59, and 74% of them were men.
Before the patients underwent conventional angiograms, their hearts were imaged in 8.5 seconds with a 64-slice CT scanner made by Toshiba Medical Systems, which funded the study along with the National Institutes of Health and private foundations.
Two physicians examined each image and graded the degree of narrowing in 19 places in the main coronary arteries. Then the researchers compared the results from both procedures.
In 163 patients with the highest degree of coronary artery disease -- a narrowing of at least 50% in at least one artery -- the CT angiograms were 93% as good as the traditional angiograms, according to the study.
Overall, the CT scans accurately identified 85% of the patients who had the biggest blockages and 90% of the patients who did not.
The researchers also found that 91% of patients who were identified by the CT scans as having the most severe disease were correctly diagnosed, as were 83% of patients whose scans did not reveal large blockages.
Two patients in the study had a reaction to the contrast dye used to perform the CT angiogram, and one patient died as a result of the conventional angiogram.
Dr. Matthew Budoff, director of cardiac CT at Harbor-UCLA Medical Center, said the study confirmed results from his own research using a similar scanner made by General Electric Co.
His study, which was funded by GE, was published this month in the Journal of the American College of Cardiology.
"It's not quite but almost as good as an invasive angiogram," said Budoff, who also receives speaking fees from GE.
The CT test is faster and costs thousands of dollars less, and patients leave "with a Band-Aid and a bottle of water," he said.
"The benefits for many patients outweigh the risk of missing 1% of disease," he said.
But other doctors say more data are needed to prove that coronary CT angiograms are worthwhile, especially as a screening tool.
"What we really need is a study that compares cardiac CT to traditional ways of working up chest pain, like stress testing, and look at patient outcomes in both groups," said Dr. Rita Redberg, director of women's cardiovascular services at UC San Francisco Medical Center, who co-wrote a perspective article accompanying the study.
"Without actual outcome data, we don't know that this is going to help patients at all," Redberg said.
Until their benefits are proven, Medicare should not be paying for the tests, which typically cost $700 to $1,000, Redberg wrote with her UCSF colleague, Dr. Judith Walsh. An attempt by the Centers for Medicare and Medicaid Services to cut off funding for such procedures was met with resistance by physicians and ultimately scuttled in March, they wrote.
An official from the federal agency responded that "additional studies are needed to clearly delineate the appropriate use of this technology in the Medicare population."