Cardiac CT and MR may replace many of the millions of echocardiograms, nuclear medicine studies, and coronary angiograms done annually. Several useful clinical applications are already available for both cardiac CT and MR, but, unfortunately, their adoption has been slow, in part because the technology has not been widely available. Now that new clinical CT and MR systems offer excellent cardiac imaging capability, many centers can finally establish cutting-edge cardiac imaging programs.
We have built a successful cardiac CT and MR program using a multipronged approach. The clinical practice made a group-level commitment to developing and supporting the program, and several radiologists were trained in performing and interpreting the studies. We actively marketed the services to referring physicians and established productive relationships with cardiologists, surgeons, and internists. The approach our practice used in developing our cardiac CT and MR program can guide other clinical practices in a similar situation.
Cardiac applications for CT are relatively new. Coronary artery calcium scoring was initially performed solely on electron-beam CT scanners, but it has recently migrated to multislice CT. Coronary CT angiography (CTA) is now best performed on MSCT systems with eight, 16, or more detector rows.
- Calcium scoring studies. Calcium scoring determines the total load of calcified plaque in the coronary arteries by calculating a score based on the volume and density of calcified plaque, using a standardized formula. Calcium scoring does not replace any prior diagnostic test, and its clinical use remains somewhat controversial. We have performed coronary calcium scoring for about four years and find these studies easy to perform and interpret. Many clinicians now order the study for patients with intermediate coronary risk factors. Often the results are used to determine whether patients should use lipid-lowering medications or follow other types of risk-factor reduction.
- Coronary CTA. Coronary CTA is an accurate and reliable method for performing noninvasive coronary angiography.1-7 The development of 16-detector row scanners has improved the performance of coronary CTA over prior CT techniques, increasing the average number of assessable coronary segments to approximately 90% (Figure 1). CTA studies have shown a 92% to 95% sensitivity for detecting stenoses of 50% or greater. Coronary CTA can accurately predict the presence and extent of coronary atherosclerosis long before it advances to the point where a stress nuclear study may be positive. It holds great promise for both early diagnosis and preventive treatment. Treatment in the future may be based not on luminal narrowing, but on actual plaque histology (Figure 2).
We view coronary CTA as both a screening tool and a problem solver. It effectively lowers the threshold for obtaining a "screening" coronary angiogram, and it allows for an additional diagnostic depiction of the coronaries in patients who are not diagnosable using other means. Our experience has shown that many cardiologists will embrace the technique when it is put forward in this context. If it is presented more aggressively, they may reject it outright as an imperfect replacement of diagnostic coronary angiography.
We typically perform two to three coronary CT angiograms daily, and cardiologists order more than 50% of these studies. The most common indication is chest pain, usually atypical or chronic. In the proper setting, coronary CTA may be an excellent tool in the evaluation of acute chest pain, but additional research is needed. Another common indication is for clarification of inconclusive noninvasive studies. Typically, this will be a suspected false-positive nuclear stress test. Other reasons include screening in high-risk patients, patients who refuse conventional coronary angiography, incomplete coronary angiography, known or suspected anomalous coronary arteries, suspected graft failure following coronary artery bypass grafting, congenital heart disease, and the evaluation of coronary stents.
We use a relatively standard approach. Beta-blocked patients receive a small bolus of contrast to determine circulation time. A diagnostic dose of 80 to 100 cc of nonionic contrast is infused at 4 to 5 cc/sec followed immediately by 50 cc of normal saline using a dual-head injector. We obtain enough 0.625-mm-thick slices to cover the entire heart and portions of the ascending aorta. Scan time is typically 18 to 26 seconds. Postprocessing is time-consuming and requires experience using a third workstation and specialized software. Volume-rendered 3D images depict the heart and coronary arteries. Reconstructions of the left and right coronary circulations include curved multiplanar reformations and subvolume maximum intensity projections.
Few MR applications have held greater promise and encountered bigger challenges than cardiac imaging.8,9 MR accurately depicts cardiac structure, function, perfusion, and myocardial viability with a capacity unmatched by any other imaging modality. While it is an accepted and widely used tool for cardiac research, however, it is used less commonly in clinical practice than many researchers had predicted.
- Morphology. Numerous methods used to depict cardiac structure are referred to as black blood and bright blood techniques. T2-weighted inversion recovery imaging is the frontline black blood sequence. Bright blood imaging yields both morphologic and functional data. Blood generates bright signal intensity, and multiple consecutive images are acquired that can be viewed dynamically to depict cardiac motion. A new approach to improve cine imaging involves a technique known as steady-state free precession.
We commonly use these techniques for evaluation of the pericardium and for structural evaluation of the heart. Depiction of a thickened pericardium is relatively simple by MR but is difficult with echocardiography (Figure 3). The occasional pericardial mass or abnormal fluid collection can be evaluated using these sequences. Other indications include evaluation of arrhythmogenic right ventricular dysplasia and differentiation of constrictive versus restrictive pericarditis.
- Cardiac function. MR evaluation of ventricular and valvular function is well established. Quantification of ventricular volumes with MR has been shown to be accurate and more reproducible than echocardiography.10 In clinical practice, MR imaging is used much less frequently than echocardiography for the evaluation of cardiac function because of reduced availability and its higher cost and longer exam times. Useful, though possibly less accurate, functional information is already available with echocardiography, which is generally performed within cardiologists' practices. Thus, no strong need to refer for MRI exists. While MR imaging is well established for assessment of cardiac function, in our practice we have had limited success in using it as a clinical tool.
- Myocardial perfusion. Myocardial regional blood flow is assessed using dynamic MR imaging during the first pass of a contrast agent. Under pharmacologic stress, a stenotic artery is unable to respond like a healthy vessel, and a perfusion deficit appears in the myocardium. Myocardial perfusion studies can depict even small, subendocardial regions of myocardial ischemia.11 The need to infuse a stress agent during the study requires a physician comfortable with its administration to be present. Radiologists sometimes perform this function, but most centers use a cardiologist or internist for this critical aspect of the study. While myocardial perfusion studies offer some improvement over currently available nuclear cardiac studies, we have not been able to develop a routine clinical service. Referrals are limited because many centers, as with echocardiography, have a competing standard technology (nuclear SPECT stress imaging) already available.
- Myocardial viability. Identification and differentiation of viable from nonviable myocardium plays a critical role in prognosis for patients with coronary artery disease. MR imaging has made dramatic progress with the introduction and rapid acceptance of the delayed-enhancement technique (DE-MRI).12 After an appropriate delay (typically 10 to 20 minutes), breath-hold inversion recovery-prepared, T1-weighted gradient-echo images are acquired. Viable tissue is dark, while nonviable, fibrotic, or scarred tissue enhances. The amount of enhancement is inversely correlated with recovery of function following revascularization (Figure 4). DE-MRI can visualize small, subendocardial areas of infarction that may be missed by nuclear techniques, including PET.
Use of DE-MRI has grown rapidly. The studies are easy to do and involve no stress agents or monitoring. DE-MRI can be promoted to both cardiologists and cardiac surgeons, who often have the difficult task of determining whether to revascularize a patient with known coronary stenoses.
- Coronary MRA. The coronary arteries, very small structures residing on a beating heart in a respiring chest, are the most difficult arterial circulations to image using MR. While many techniques have been investigated, none has received universal acceptance. Recent studies have shown increased accuracy, especially for detection of patients with left main stem or three-vessel disease.13 But exam time, reliability, and ease of use continue to be major practical impediments to clinical acceptance. Clinically, we perform coronary MR primarily for suspected anomalous coronary arteries. Because coronary CTA has proven to be such a robust and useful technique, there is little call for coronary MR in our practice.
- Congenital heart disease. Clinical management of patients with congenital heart disease depends on characterization of cardiac morphology and evaluation of hemodynamics. Traditionally, cardiac catheterization and echocardiography are used to assess these patients. Cardiac MR imaging can in many cases completely evaluate cardiac and vascular morphology, venoatrial connections, visceral situs, and extracardiac abnormalities.9 We have teamed with pediatric and adult cardiologists specializing in congenital heart disease, and have used MR in many cases to replace or reduce the need for cardiac catheterization.
DEVELOPING A CARDIAC CT OR MR PROGRAM
Once potential referring physicians recognize the value of these new studies, practices should take a deliberate approach to starting a program. Directing resources and personnel to a noninvasive cardiac imaging program is no small endeavor. The department must prepare a carefully thought out, deliberate plan prior to opening the doors for new patients.
If executed correctly, numerous rewards follow, including increased utilization of CT and MR systems and increased revenue. A successful program leads to greater job satisfaction for physicians, nurses, and technologists and to enhanced visibility and respect within the local medical community. A well-run program improves relationships with cardiology and other referring physicians and places the radiology group in an advantageous position in a competitive environment.
- Program director(s). Appointment of an overall director, or individual CT and MR directors, provides a go-to person who can develop protocols, attend meetings, make recommendations on equipment, ensure that studies are performed and interpreted correctly, and oversee quality assurance. While many radiologists may interpret exams, having a single director greatly simplifies the process of getting the program up and running smoothly.
Our practice appointed a physician as the director of combined cardiac CT and MR early in the program development. The director saw that numerous critical elements were put into place, making sure that the CT and MR technologists received appropriate cardiac applications training, that the business office was able to schedule and bill for the studies, and that other radiologists received necessary training. The director also gave talks and attended meetings with referring cardiologists, internists, and surgeons.
- CT and MR technology. New cardiac programs do not always require expensive new CT or MR systems. Most MR systems are capable of performing at least some cardiac studies. Sequences required to evaluate function, morphology, and viability are often standard or may have been purchased as options in the past. The sequences available in advanced cardiac software packages are often already included as slightly slower standard sequences on many MR systems. We have six operating MR systems: Two were purchased with advanced cardiac software packages, but all are used for cardiac studies.
Many centers already have one or more MSCT systems. Often the required software or hardware for performing either calcium scoring or coronary CTA has already been purchased or can be obtained as an add-on option. We purchased a two-slice CT system four years ago that had calcium-scoring capability and added a 16-slice system for coronary angiography 18 months ago. Currently, all cardiac CT studies are performed on our newest 16-slice MSCT system.
- Radiologist training. Interpreting radiologists require training in applying protocols, processing, and reading cardiac studies. Until formal cardiovascular CT and MR fellowships are developed, the number of radiologists with useful experience will be limited. Few radiologists have received in-depth training in these advanced applications, but most practicing radiologists with a working familiarity with state-of-the-art CT and MR can learn to read cardiac CT and MR studies by using available resources.
Many university hospitals offer minifellowships in cardiovascular CT and/or MR that last from a week to several months and give hands-on experience. Other options include intensive courses sponsored by CT and MR vendors. The annual meetings of several organizations-including the RSNA, American Roentgen Ray Society, Society for Computed Body Tomography/Magnetic Resonance, and International Society for Magnetic Resonance in Medicine-offer lectures and other programs taught by recognized leaders in the field. Many CME courses focus on these new applications as well. Finally, in-depth review articles with practical advice for clinical applications have appeared in the radiology journals.
- Marketing of services. Services must be marketed to referring physicians who may not be aware that they are available or knowledgeable about their use. We took numerous approaches to marketing: word of mouth, didactic lectures, a regular presence in clinical meetings, and printed materials distributed to physicians.
We gave a formal presentation to our own radiology group when we started the coronary CTA program, as many of our physicians initially understood little about this new technique. Once our radiologists were informed, they were more likely to recommend the procedures when discussing clinical cases and could talk intelligently about them if approached by other physicians. This type of physician-to-physician discussion and promotion eventually accounts for a large portion of new referrals.
Short PowerPoint talks, structured around clinical cases and the supporting literature, are another useful tool. We have given lectures to cardiology, internal medicine, and family practice groups. Almost without exception, referrals from these groups increase in the weeks following one of these meetings. These groups enjoy the chance to have an open discussion about the studies and to meet the physicians who will be interpreting them. We also attend a weekly cardiology/cardiac surgery conference.
When we started performing coronary CTA, we mailed an information packet to about 2000 potential referring physicians in our area.
It included detailed information and clinical examples of the exams we offered, a phone number for a dedicated scheduler who handles all outpatient cardiac CT studies, information about billing and reimbursement, and a reprint of an important article from the literature. We established a section on our practice Web site (www.fairfaxradiology.com) containing this and other pertinent material and also produced a patient information brochure that physicians could give to patients for whom they were ordering the exams.
- Developing new relationships. Establishing and maintaining open communications with referring cardiologists, surgeons, and internists is critical to the long-term success of the program. Good communication goes both ways and is helpful to the radiologist as well as the referring physician. Before performing the exams, we obtain reports of all prior cardiac studies when possible. We may contact physicians directly if any questions remain. If adequate history and prior diagnostic information are obtained, the odds are good that the correct study will be performed, the actual clinical question answered, and the referring physician satisfied.
Radiologists need to speak the language of cardiologists and other cardiac care providers. We use standard report templates for our cardiac CT exams to ensure that all cardiac CT readers include the same information with every study and that the terminology used is correct and familiar to the clinicians. This report is a modified version of that recommended for coronary angiography by the American Heart Association. We also provide selected color images with the written report, highlighting and labeling important findings.
We include a CD containing images and one or more movies depicting a 3D volume-rendered view of the heart and coronary arteries for coronary CTA studies. This allows the referring physician to see the angiograms for patients about whom they are making critical decisions and gives them materials that they can show to their colleagues or patients. This makes for good patient care and is a secondary form of marketing that has led to new referrals.
We encourage feedback from the clinicians to obtain follow-up information regarding previously imaged patients. Review of this data is helpful in refining imaging protocols and influencing how subsequent studies are interpreted. We have a database established for each of the 500 or so coronary CTAs we have performed to date.
Advanced cardiac CT and MR techniques and capabilities have migrated from research and academic centers to community and general hospitals and outpatient imaging centers. Radiologists now have the tools needed to perform sophisticated noninvasive cardiac imaging. With a careful and deliberate approach, radiology departments and practices can build successful cardiac imaging programs. By doing high-quality cardiac CT and MR and building effective relationships with other physicians, we have gradually been accepted as a new member of the clinical team caring for cardiac patients.
Dr. Earls is director of cardiovascular CT and MRI at INOVA Fairfax Hospital and a partner of Fairfax Radiological Consultants in Fairfax, VA.
Dr. Earls has received grants/research support from GE Healthcare, is a consultant with Mallinckrodt/Tyco, and is a member of the speakers bureau for Berlex Laboratories.
1. Achenbach S, Ulzheimer S, Baum U, et al. Noninvasive coronary angiography by retrospectively ECG-gated multislice spiral CT. Circulation 2000;102:2823-2828.
2. Knez A, Becker C, Ohnesorge B, et al. Noninvasive detection of coronary artery stenosis by multislice helical computed tomography. Circulation 2000;101:E221-E222.
3. Nieman K, Oudkerk M, Rensing BJ, et al. Coronary angiography with multi-slice computed tomogaphy. Lancet 2001;357:599-603.
4. Achenbach S, Giesler T, Ropers D, et al. Detection of coronary artery stenoses by contrast-enhanced, retrospectively electrocardiographically-gated, multislice spiral computed tomography. Circulation 2001;103(21):2535-2538.
5. Nieman K, Cademartiri F, Lemos PA, et al. Reliable noninvasive coronary angiography with fast submillimeter multislice spiral computed tomography. Circulation.2002;106(16):2036-2038.
6. Ropers D, Baum U, Pohle K, et al. Detection of coronary artery stenoses with thin-slice multi-detector row spiral computed tomography and multiplanar reconstruction. Circulation 2003;107(5):664-666.
7. Mollet NR, Cademartiri F, Nieman K, et al. Multislice spiral computed tomography coronary angiography in patients with stable angina pectoris. J Am Coll Cardiol 2004;43(12):2265-2270.
8. Earls JP, Ho VB, Foo TK, et al. Cardiac MRI: recent progress and continued challenges. J Magn Reson Imaging 2002;16(2):111-127.
9. Constantine G, Shan K, Flamm SD, Sivananthan MU. Role of MRI in clinical cardiology. Lancet 2004;363(9427):2162-2171.
10. Bellenger NG, Burgess MI, Ray SG, et al. Comparison of left ventricular ejection fraction and volumes in heart failure by echocardiography, radionuclide ventriculography and cardiovascular magnetic resonance: are they interchangeable. Eur Heart J 2000;21:1387-1396.
11. Al-Saadi N, Nagel E, Gross M, et al. Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance. Circulation 2002;101:1379-1383.
12. Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. NEJM 2000;343:1445-1453.
13. Kim WY, Danias PG, Stuber M,et al. Coronary magnetic resonance angiography for the detection of coronary stenoses. NEJM 2001;345:1863-1869.