Study shows inaccuracies in cardiac PET-CT imaging

Jul 4 2007

Current PET-CT scanners with standard commercial software designed to provide images of the heart are falsely indicating coronary artery disease in as many as 40 percent of patients, according to a study published today on the cover page of the Journal of Nuclear Medicine.

Lead author K. Lance Gould, M.D., professor in the Division of Cardiology at The University of Texas Medical School at Houston, said he discovered the abnormalities upon his initial use of cardiac PET-CT scanners in the Weatherhead P.E.T. Center For Preventing and Reversing Atherosclerosis at Memorial Hermann-Texas Medical Center.

Positron emission tomography, or PET, is nuclear cardiac imaging for optimally assessing blood flow in the heart. When paired with CT, or computed tomography, the PET-CT scanners should be an accurate, noninvasive tool for detecting or assessing severity of heart disease, said Gould, executive director of the Weatherhead P.E.T. Center.

However, he said, an erroneous basic concept in the software functions make the PET-CT scanners prone to generating false-positive results.

The technology showed false-positive results in 40 percent of the 259 patients who participated in Gould's study. In 23 percent of those patients, it showed severe false abnormalities suggesting heart disease, which Gould said could have resulted in unwarranted, emergency heart procedures if he had not run additional tests to confirm his suspicions of the inaccurate results.

"We found errors that are inherent in the technology as now used. Failure of physicians to recognize and correct those errors can have a profound impact on patients lives," said Gould, the Martin Bucksbaum Distinguished University Chair. "These falsely positive results could lead to unnecessary procedures for nonexistent heart problems or for problems that could be treated without surgery."

The commercial software does not account for movement in the lungs and heart while the patient is breathing during the test, Gould said, and as a result, the PET and CT images don't always match or co-register. "Consequently, the images have big holes or gaps where the processing is incorrect in areas where the PET and CT data fail to match," Gould said. "It looks like someone has severe disease."

On discovering the problems, Gould developed a solution with Tinsu Pan, Ph.D., associate professor in the Department of Imaging Physics at The University of Texas M. D. Anderson Cancer Center, by rewriting the software to account for breathing motions and correctly aligning PET and CT images.

"When done properly, PET-CT produces absolutely perfect images of blood flow in the heart muscle," he said. "It's the best way to assess and direct management of heart disease." Unfortunately, according to Gould, the redesigned software is not generally commercially available at this time. Gould also says, "This medical product is out there on the market, and physicians and their patients trust it, yet there is a big problem with it. Since PET-CT proved ideal for cancer, body and brain imaging, everyone assumed its application to the heart would work with the same protocols and software a big mistake."

Gould, a preventive cardiologist, has been using traditional PET imaging throughout his medical career to detect heart disease, to determine the need for arteriograms, stents or bypass surgery and to monitor the reversal of the disease in his patients who follow his program of diet, exercise and medication.

Adding the CT component to PET is a faster way to produce cardiac images and obtain additional information about coronary calcium deposits due to atherosclerosis.

"Properly co-registered PET/CT data provide definitive perfusion images demonstrated in this study and, as reported earlier, suitable for assessing severity of coronary artery disease, for following its regression or progression, and for identifying mild early non-obstructive coronary atherosclerosis as the basis for intense, lifelong pharmacologic and lifestyle management," the publication states.

PET-CT is a valuable tool in the detection and management of coronary artery disease, Gould said, and he hopes other cardiologists and their patients will insist that scanner manufacturers resolve problems with the software to produce the most accurate results.

For patients who have recently had a PET-CT and been diagnosed with coronary artery disease, Gould recommends they seek a second opinion and request review of their PET-CT scans for these errors in matching the PET and CT images.

Alternatively, traditional or "Pure PET" without the CT component is less subject to this problem.

However, Gould cautioned: "Alternative non-invasive technologies such standard SPECT nuclear cardiac imaging and CT angiography have been proven to have comparable false positive results for different reasons. Pure PET or PET-CT done properly are the best non-invasive tests for early coronary heart disease or for assessing its severity but have to be done correctly with attention to technical details."