Tailoring MR to patients improves cardiac imaging

November 14, 2006

The possibility that there is no one right answer for everyone should come as no surprise, yet cardiac MR protocols have persisted in acquiring data at diastole for all patients. Mid-diastole, the fraction of a second when the heart rests between beats, intuitively would seem the right time to collect data, especially for coronary MR angiography, as cardiac vessels are the most susceptible to motion artifact. Diastole is, in fact, a good starting point -- but not necessarily the end point, at least not in all cases.

The possibility that there is no one right answer for everyone should come as no surprise, yet cardiac MR protocols have persisted in acquiring data at diastole for all patients. Mid-diastole, the fraction of a second when the heart rests between beats, intuitively would seem the right time to collect data, especially for coronary MR angiography, as cardiac vessels are the most susceptible to motion artifact. Diastole is, in fact, a good starting point - but not necessarily the end point, at least not in all cases.

Research conducted in Japan indicates that tailoring the acquisition strategy to the subject may be the best way to reduce motion artifact.

"We found the optimal timing of data acquisition varies in each patient," said Dr. Hajime Sakuma, professor and vice chairman of Mie University Hospital.

Sakuma and colleagues at Mie University Hospital and Matsusaka Central Hospital evaluated patients to determine whether the right time for some might be during systole, when their hearts are actively contracting to pump blood. The decision might be based on heart rates and other factors, they hypothesized.

In a study published in the Nov. 21 issue of the Journal of the American College of Cardiology, the researchers used whole-heart MRIs to evaluate 131 patients suspected of having coronary artery stenosis. In 48, especially those with faster heart rates, the optimal time to collect data occurred while the heart was contracting rather than resting. Overall, useable images were collected in 113 patients or 86% of the study participants.

"This customized approach substantially reduced motion blurring of the coronary artery and improved the detection of coronary arterial stenoses compared with previous studies," Sakuma said.

The research team compared the accuracy of diagnoses based on tailored coronary MRA to those obtained using traditional x-ray coronary angiography. Coronary MR produced accurate diagnoses of "significant narrowing" of the lumen in 87% of coronary arteries at least 2 mm in diameter. (Significant narrowing was defined as narrowing of at least half the normal diameter of the lumen.) When researchers studied isolated segments of arteries, their ability to accurately diagnose significant disease using MRA rose to 94%.

Further statistical analyses led the researchers to conclude that using tailored MRA protocols detected significant narrowing of the arteries with "moderate sensitivity and high specificity," 82% and 90%, respectively. This means the tests produced only a small number of false-positive results, but a more moderate level of false-negative findings.

"While more research and refinement are needed before coronary MR angiography is ready for widespread clinical use, the technology is ideally suited for screening coronary artery disease and is worth the effort for further development," Sakuma said.

The research comes at a time when 64-slice CT appears on the brink of revolutionizing cardiac evaluations, creating the opportunity to follow patients after interventions or even screen high-risk but asymptomatic patients. It is much faster than MRA, 10 to 15 seconds versus 15 minutes or more, but presents a substantial cumulative risk from radiation exposure.

"It exposes patients to up to twice as much as a traditional cardiac catheterization and coronary angiography and it requires patients to be injected with a dye solution," said

Dr. Warren J. Manning, section chief of noninvasive cardiac imaging at Beth Israel Deaconess Medical Center in Boston and a professor of medicine and of radiology at Harvard.

Manning noted, however, that both techniques have been studied only in patients very likely to have coronary artery disease, which is much different than the typical patient.

"The use of both technologies in patients who have atypical chest pain or simply have multiple risk factors for coronary artery disease is unproven," he said. "In the end, it's not just whether we can get pictures. We have to understand what they mean and their impact on our patients' care."