Champion of the Myocardial Imaging Debate: MDCT (For Now)
James P. Earls, MD
The scientific session "CT II: Myocardial Imaging" focused on the rapidly advancing arena of noninvasive cardiac imaging that received so much attention at RSNA 2002.[1] The development of the 8- and now 16-detector-row CT scanner has recently facilitated cardiac imaging protocols that offer both high-temporal and spatial resolution. Combined with recent innovations in cardiac MRI technology, the entire field of noninvasive cardiac imaging has undergone explosive growth in the past few years.
Cardiac CT Outstrips MRI
The session was moderated by 2 well-known and highly respected cardiac imaging researchers, Andre Duerinckx, MD, Ph.D., University of Texas Southwestern Medical Center, Dallas, Texas, and Zahi Fayad, Ph.D., Mount Sinai Medical Center, New York, NY. It is interesting to note that while both of the moderators are well published in the field of cardiovascular MRI, 9 of the 10 scientific papers presented in this session used multi-detector-row CT (MDCT) and not MRI as an imaging tool. This is likely due to the brisk pace of cardiac CT development; this area of imaging is so new and changing so rapidly that well-recognized or widely published researchers who are experts in cardiac CT are not yet known or available to chair scientific sessions at RSNA.
Coronary artery disease is the #1 cause of death in the United States. The new cardiac CT and MR applications represent a potential vast new area for the practice of radiology. However, these new noninvasive cardiac imaging techniques can appear intimidating for the radiologist or clinician who has not recently been involved with diagnostic heart imaging or who is not familiar with the array of treatments for heart disease. Cardiac imaging not only involves morphologic (or anatomic) imaging, but also includes the evaluation of cardiac function, coronary angiography, myocardial perfusion, infarct imaging, and myocardial viability.
Because of the many techniques available and the relative complexity of the new cardiac CT and MRI systems, there may be some inertia getting these new applications accepted into routine clinical practice. Although cardiac MRI techniques for the assessment of myocardial perfusion and viability are well established and are widely available in the community, they are not performed at many sites other than specialized centers, which are usually the large academic hospitals.[4] MDCT techniques are somewhat easier to perform and interpret than many of the recently launched cardiac MRI applications. For this reason, MDCT may eventually gain wider clinical acceptance and achieve widespread use for noninvasive cardiac imaging than the apparently more complex MRI applications.
It was clear at this session that the development of both CT and MRI cardiac imaging technology is proceeding at a rapid pace. One of the most promising and potentially useful applications of cardiac CT is in coronary angiography. Although this was extensively covered in a separate scientific session (see RSNA report "Cardiac MDCT and MRI Vie for Supremacy"), 1 paper highlighting the utility of this technique was presented during this session. M, Kim, MD, Seoul, South Korea, presented work spotlighting the accuracy of MDCT in evaluating the patency of coronary artery bypass grafts.[1] This group studied 31 patients who underwent MDCT and conventional graft angiography and reported a sensitivity, specificity, and accuracy of 100%, 97%, and 98% in defining graft stenoses. The negative predictive value was reported as 100%, which is a very important value if this technology is ultimately to be used for screening populations so that no true-positives are erroneously reported as false-negatives.
Characterizing LV Function
In the remaining presentations during this session, several general areas of research were highlighted. These included 3 presentations outlining the evaluation of ventricular function by MDCT scanning, 3 studies looking at MDCT of myocardial infarct, and 1 presentation each focusing on coronary artery calcium scoring studies, mapping for atrial fibrillation ablation, and first-pass perfusion on MRI.
The evaluation of left ventricular (LV) function using MDCT is important as an adjunct to other MDCT applications in the effective diagnosis and management of patients with cardiovascular disease.[1] It is helpful clinically to determine LV function noninvasively with a high degree of accuracy, at the same time that a coronary angiogram is performed. The 3 papers reporting on this topic in this session all used a contrast-enhanced approach in which LV function data and coronary angiography data were acquired simultaneously.
Walter Hundt, MD, Stanford, California, postulated that MDCT with retrospective cardiac electrocardiogram (ECG) gating facilitates the calculation and quantification of LV volumes and function. However, this group of investigators found significant overestimation of the end systolic volume, which generated an underestimation of the stroke volume and ejection fraction.[1] Another study, reported by Satoshi Hosoi, MD, Onsen, Japan, focused on the feasibility of MDCT in the determination of LV function and wall motion. Although this group did not report the same underestimation of the end systolic volume and achieved good results, they did have some difficulty evaluating wall motion in the cardiac apex (AHA segment 1).
Finally, Sandra Halliburton, Ph.D., Cleveland, Ohio, reported on the novel approach her group was taking to increase the temporal resolution of cardiac MDCT by using interpolated data from 4 consecutive cardiac cycles. This technique reduced temporal resolution from 500 ms to approximately 90-250 ms (depending on heart rate) but increased the effective slice thickness from 1 mm to 6-7 mm. Using cine TrueFISP (fast imaging with steady-state precession) MRI as a gold standard, the group found that the accurate measurement of LV volumes could be significantly improved using high-temporal-resolution image reconstruction.
Breakthroughs in Infarct Detection
The detection and quantification of myocardial infarction is another critical new application for both cardiac MRI and MDCT. In clinical practice, the ability to rapidly and noninvasively delineate and quantify a myocardial infarction with a high degree of accuracy would be of enormous utility, especially in the work-up of patients presenting with acute chest pain. The identification and differentiation of viable from nonviable myocardium plays a pivotal role in determining treatment strategies for patients with heart disease. In the past few years, MRI has made a quantum leap in this arena with the introduction and rapid acceptance of the "delayed-enhancement" technique (DE-MRI).[5] In the myocardial imaging session at RSNA 2002, several researchers also reported success with the use of contrast-enhanced MDCT in pinpointing an area of infarcted myocardium. In general, myocardial infarcts were hypoperfused on early images and then became hyperenhanced on delayed images after contrast infusion. These findings were similar to those on MRI.
Myocardial Imaging Potpourri
Udo Hoffman, MD, Boston, Massachusetts, reported results using a porcine model and correlated imaging findings with pathologic analysis of heart specimens stained with triphenyl tetrazolium chloride (TTC).[1] This group of investigators found that MDCT can detect and quantify acute myocardial infarction with a high degree of accuracy, although they also reported that end-systolic datasets were more precise than end-diastolic datasets. Another group had similar success in this regard using a rabbit model. J. Paul, MD, Le Plessis-Robinson, France, reported data from a human group of 25 patients.[1] This group of investigators found subepicardial hyperenhancement on delayed images in all patients. The shape and extent of the delayed hyperenhancing area correlated well with ECG and echocardiographic findings and with the peak serum creatine kinase level.
Denis Foley, MD, Milwaukee, Wisconsin, discussed the role of cardiac MDCT in planning and optimizing atrial fibrillation ablation. A 3D model of the left atrium was exported to an interventional system and then registered and used during radiofrequency ablation procedures to isolate the pulmonary veins. P. Hunold, MD, Essen, Germany, and colleagues found that a saturation-recovery steady-state free precession (TrueFISP) sequence offers better signal characteristics than the previously used TurboFLASH techniques for the evaluation of first-pass myocardial perfusion.
In addition to the ability to noninvasively image the coronary arteries, it appears that cardiac MDCT may have additional utility in the area of assessing coronary function and viability. Researchers in this session at RSNA 2002 demonstrated that cardiac function could also be derived from contrast-enhanced MDCT exams. Other reports determined that infarct imaging with delayed CT is also a possibility, although this work is clearly at an earlier stage of evolution than the coronary angiography or functional analysis applications.
The future of noninvasive cardiac imaging using MDCT and MRI continues to offer the promise that this new and rapidly advancing field will likely substantially change clinical practice and referral patterns for the evaluation of cardiac disease within the next several years.
|
| |
| |