Cardiovascular Magnetic Resonance in Ischemic Heart Disease
Left Ventricular Structure and Function at Rest
The most fundamental quantitative measurements of the heart include those of ventricular size and function. Due to its high spatial and temporal resolution and the freedom to image in any plane, cardiovascular magnetic resonance (CMR) imaging is especially suited for this task. Traditionally, black-blood spin echo techniques and bright-blood gradient echo cine techniques have been utilized for this purpose, but these have been replaced by the newer steady-state free precession-based approach, which results in substantially improved blood pool-to-myocardium contrast (see Figure 1a and 1b).
Contiguous short axis slices of the left ventricle are obtained from base to apex, endocardial and epicardial borders are planimetered, volumes for each slice are calculated by multiplying the area by slice thickness, and finally, total ventricular volume is calculated by adding the volumes of individual slices. Myocardial mass is calculated by multiplying myocardial volume by its density. Normal values in volunteers have been published for both the left and the right ventricle. Since measurements of ventricular size and function are very reproducible with CMR, this allows for smaller sample sizes for clinical studies using these parameters as end-points.
Regional function, measured as myocardial wall thickening, is usually calculated by the centerline method, where a line is created in the center of the myocardium, equidistant from the endocardial and epicardial borders. Multiple chords are placed perpendicular to the centerline between the epicardial and endocardial borders and thickening is calculated as the ratio of the length of the chord at end-systole and end-diastole.
Functional Stress Testing
Detection of regional dysfunction during pharmacologic stress can be used to identify patients with underlying coronary artery disease. Investigators have shown improved sensitivity and specificity for dobutamine CMR over stress echocardiography and that dobutamine magnetic resonance imaging (MRI) stress testing has excellent accuracy in patients who had inadequate acoustic windows for stress echocardiographic studies (sensitivity 83%, specificity 83%). Performing dobutamine CMR stress testing requires a carefully assembled team with adequate nursing support, online monitoring of ventricular function and rhythm during the study, and standardized protocols.
Myocardial Perfusion Imaging
For the purposes of CMR perfusion imaging, gadolinium-diethylenetriamine penta-acetic acid (Gd-DTPA) is injected into a peripheral vein at the time of maximal vasodilation and its first pass is imaged through the circulation and the myocardium with relatively high spatial and temporal resolution, covering representative slices of the heart. Gd-DTPA shortens T1-relaxation time and therefore brightens tissues in the compartment where the agents are distributed. The typical dose of Gd is 0.025-0.075mmol/kg, injected as a tight bolus by a power injector. Perfusion defects appear as areas of hypointense myocardium, usually in the subendocardium (see Figure 2).
For clinical purposes, images can be analyzed qualitatively by visual inspection for the presence of perfusion defects, although the accuracy of visual analysis has not been well-validated. More quantitative techniques have been validated against positron emission tomography (PET) measures of blood flow, with a sensitivity and specificity of 87% and 85%, respectively, compared with quantitative angiography. The accuracy of CMR perfusion imaging is comparable to radionuclide modalities and in a recent study had sensitivity of 88% and specificity 90% versus angiography.
Imaging of Myocardial Viability
With the widespread use of percutaneous and surgical revascularization techniques, determination of myocardial viability has become an important clinical goal. Delayed contrast-enhanced imaging for the detection of myocardial viability is a novel application of CMR with a recent increase in use due to an improved imaging sequence using inversion recovery that significantly improved image quality and contrast between normal tissue and infarcted myocardium (signal intensity ~500% of normal tissue). This technique is based on delayed wash-in and wash-out kinetics of Gd-DTPA into and out of infarcted tissue, compared with normal myocardium. Infarct detection has been carefully validated with this technique in animal models against histology, both in acute reperfused and non-reperfused infarcts, as well as in the chronic setting.
Currently, CMR is the only technique able to resolve the transmural extent of myocardial infarction and is more sensitive than single positron emission computed tomography (SPECT) techniques for identifying subendocardial infarction. This is important, since it has been shown both in the acute and chronic setting that the transmural extent of hyper-enhancement is inversely proportional to the likelihood of functional recovery after myocardial infarction or revascularization in chronic ischemic disease. In general, when there is no hyper-enhancement, the likelihood of recovery is close to 80%.
On the other hand, when the transmural extent of hyper-enhancement is more than 75%, the likelihood of recovery is less than 5%. It has also been shown that when the transmural extent of hyper-enhancement is between 1% and 50%, functional recovery may not be accurately predicted based on the hyper-enhancement pattern alone; contractile reserve in response to low-dose dobutamine may be able to distinguish viable from non-viable myocardium in these cases.
Coronary Artery Imaging
A most appealing role for CMR in ischemic heart disease would be the non-invasive assessment of the coronary arteries with high temporal and spatial resolution, during relatively short acquisition times. Given the small size, tortuous course, and motion of the coronary arteries, several technical challenges must be overcome in order to obtain images of diagnostic quality. Best in-plane resolution for CMR angiography (CMRA) is about 700 to 900µm, which is still about twice the pixel size compared with conventional X-ray angiography. Compensation for cardiac and coronary arterial motion is achieved by using short acquisition times and obtaining images during isovolumic relaxation. The most commonly used techniques for respiratory motion correction rely on diaphragmatic navigators, in which the lung-diaphragm interface is tracked and is used to predict the motion and position of the coronary arteries.
A prospective multi-center study of native coronary arteries was recently conducted using a single hardware platform with a standardized imaging protocol using a navigator technique. CMRA detected a total of 78 of 94 coronary stenoses found on the conventional angiogram (83%). Sensitivity, specificity, and accuracy for the detection of left main coronary artery (LMCA) disease or three-vessel disease were quite good at 100%, 85%, and 87%, respectively, but sensitivity and specificity were suboptimal in individual vessels. Further refinement of these techniques is clearly needed.
Conclusion
CMR is evolving rapidly for use in the evaluation of patients with ischemic heart disease. Additional applications include evaluation of aortic disease, complex congenital heart disease, cardiac masses, cardiomyopathies, pericardial disease, and atherosclerotic plaque in the aorta, carotid, and peripheral vasculature. CMR offers the potential for a comprehensive evaluation of cardiovascular disease and its promise is being increasingly recognized and put to use at many centers around the world.