MRI has gained increasing acceptance for identifying patients with even subtle coronary artery disease (CAD). In a new study from Brazil, researchers found that first-pass MRI can distinguish segments with different degrees of obstructive CAD.
The results suggest a new diagnostic option for patients with symptoms of cardiovascular disease but in whom angiographic results may be negative, a problem highlighted in a recent series of studies on obstructive CAD in women.
Overall mortality rates for patients with coronary artery disease have been declining for a generation, but ischemic heart disease remains a major public health problem, wrote Dr. Luiz Rodriguez de Ávila, Ph.D.; Dr. Juliano Lara Fernandez; and colleagues from the University of Sao Paolo in Brazil in the February issue of Radiology.
Coronary angiography, the diagnostic gold standard for CAD, is invasive and has low accuracy in estimating the physiologic importance of coronary stenosis in the population at risk. These limitations, combined with the modality's low accuracy in estimating the physiologic importance of stenosis and functional importance of an obstruction, have led to the development of alternative diagnostic imaging methods, including thin-section CT, Tc-99m sestamibi SPECT, and stress echocardiography for disease detection. The Brazilian team sees MRI as an important modality for detecting different degrees of obstructive CAD.
"Compared with other noninvasive methods, (MRI) has been increasingly accepted as an imaging modality that could improve the identification of patients with CAD," the authors wrote. "By permitting an integrated examination of cardiac function, myocardial perfusion, and viability, MR imaging has become important in the management of both acute and chronic coronary syndromes" (Radiology, February 2006, Vol. 238:2, pp. 464-472).
The traditional method for these exams, first-pass MR with peripherally injected contrast material, offers quantitative assessment of first-pass contrast-enhanced perfusion using signal-intensity time curves, with sensitivities and specificities ranging from 90% to 91% and 62% to 85%, respectively, compared to radionuclide scanning or coronary angiography.
In the study, 37 patients (29 men, eight women; mean age 57.2 years ± 10.5 standard deviation) with CAD risk factors and clinical indications for coronary angiography, such as positive stress test results, underwent first-pass contrast-enhanced MRI on a 1.5-tesla MR scanner (Signa Horizon, CVi; GE Healthcare, Chalfont St. Giles, U.K.). Patients with a history of myocardial infusion were excluded.
Images were acquired after infusion of the stress agent dipyridamole (Persantin, Farmacia USP, Sao Paolo, Brazil) at 0.56 mg/kg of body weight and six minutes after the start of infusion of 0.05 mmol/kg (0.1 mL/kg) gadopentetate dimeglumine (Magnevist, Schering, Berlin) at 5 mL/sec, followed by 20 mL of saline.
Patients maintained their breath-hold and expiration as long as possible throughout the acquisition of 50-70 images. Cine short-axis images were obtained first to plan the perfusion imaging. Six to eight sections were then acquired to cover the left ventricle (ECG-triggered enhanced fast gradient-echo-train pulse sequence with repetition time of 6.9 msec, echo time of 1.8 msec, flip angle of 10º, 160 x 160 matrix, field-of-view of 320-360 mm, 8-mm section thickness, and echo-train length of 4).
Aminophylline (Farmacia USP) was administered to reverse the effects of the stress agent, then rest perfusion images were obtained using the same MR imaging parameters. Patients were examined supine with monitoring of heart rate, ECG results, respiratory rate, and arterial pressure. All underwent coronary angiography within 15 days of MR.
For interpretation, myocardial segments were divided into five groups (A-E) based on the degree of obstruction in the supplying artery, with A signifying negative results and E meaning significant obstruction greater than 70%. Time-to-peak signal intensity and hyperemia-to-rest (HR) ratios for each of these three variables were analyzed for each segment using a generalized linear model.
The results showed key differences in signal-intensity upslope, with significantly higher values in patients with normal coronary arteries compared to those with CAD (p < 0.001), both at rest and after administration of dipyridamole for all groups (p < 0.001).
There were significant differences in signal-intensity slope (p < 0.001) between segments in group A patients (no obstruction seen at angiography) and normal-appearing results in group B patients (having segments containing lesions that corresponded to any degree of stenosis): 26% at hyperemia and 17% at rest. In the other groups, a significant difference at rest was not observed after stress agent infusion, the authors wrote.
Both signal-intensity upslope (p < 0.05) and peak signal intensity (p < 0.01) enabled the differentiation of segments with more than 70% reduction in luminal diameter from those in all other groups. Unlike the findings for signal-intensity upslope, however, no significant differences were found at rest between segments in group A and those in group E with considerable lumen obstruction.
Findings for HR ratios were similar to those seen using each signal-intensity variable. The HR ratio for signal-intensity upslope was found to be significantly different for all groups (p < 0.001); however, no differences were found in the HR ratio for either peak signal intensity or time-to-peak signal intensity.
"Findings from this study demonstrate that first-pass myocardial perfusion MR can, in a stepwise fashion, demonstrate myocardial ischemia in segments with different degrees of coronary obstruction," the authors wrote. "The detection of myocardial ischemia was more pronounced with analysis of signal-intensity upslope curve than with peak signal intensity, time-to-peak intensity, and HR ratios. To the best of our knowledge, these results are the first to show that this technique can demonstrate impaired perfusion in segments that are supplied by normal-appearing epicardial arteries in patients with obstructive lesions in other coronary territories."
Microcirculatory impairment may be the cause of reduced perfusion in normal-appearing segments, they wrote. Vasodialation by dipyridamole increases perfusion by increasing intracellular levels of adenosine and provoking vasodilation.
"The variable that was found to be the most useful in distinguishing groups of segments was signal-intensity upslope," a finding that was in accordance with studies by other groups, de Ávila and colleagues wrote. "Signal-intensity upslope allowed for the discrimination of patients without CAD or who have alterations in endothelial function. Signal-intensity upslope could also be used to differentiate segments with clinically important lesions (more than 70% reduction in luminal diameter) from all other groups both at rest or hyperemia. The use of signal-intensity upslope might therefore permit the identification of patients who have a low risk for CAD or who have alterations in endothelial function."
The method could also be useful for separating segments with significant luminal narrowing from other segments at hyperemia by demonstrating areas of the myocardium that could be suitable for revascularization, they wrote.
"The ability to detect segments in the myocardium that have impaired perfusion, but contain normal coronary arteries may demonstrate the potential clinical importance of identifying patients with increased risk of cardiovascular disease who would otherwise be considered at lower risk," the authors wrote.
Limitations of the study included the authors' inability to compare findings directly with those of other studies, especially with PET, though previous results suggest that the findings are highly concordant, they wrote. Sections thinner than 8 mm may also have served to better distinguish perfusion deficits.
"The quantitative assessment of signal intensity time curve ... can be used to differentiate segments with significant obstructions from all other segments," the authors concluded. "Most importantly, first-pass perfusion MR imaging can be used to distinguish segments in normal individuals from those in normal-appearing areas of the myocardium in patients with diffuse CAD. The identification of these regions at MR imaging might have an effect on risk stratification and treatment strategies for these patients and could be assessed in larger clinical studies."
By Eric Barnes
AuntMinnie.com staff writer
February 21, 2006
Related Reading
MRI differentiates AMI from myocarditis, November 21, 2006
Cardiovascular MRI accurately diagnoses acute myocarditis, June 16, 2005
Study quantifies prevalence of noncardiac findings on cardiac MR, January 25, 2005
Myocardial perfusion measured by MRI useful in detecting heart disease, August 4, 2003
Copyright © 2006 AuntMinnie.com
