- Email: [email protected]
Referrals to a Dedicated Cardiac MRI Service: Who Sends What?
Referrals to a Dedicated Cardiac MRI Service: Who Sends What? Michael K. Atalay, MD, PhD, Ethan A. Prince, MD, John J. Cronan, MD
The purpose of this study was to retrospectively evaluate the referral patterns to the authors’ cardiac magnetic resonance imaging service in its first 28 months of operation. In late November 2003, the authors’ radiology practice established a cardiac magnetic resonance imaging service for a 719-bed teaching hospital and a 247-bed community hospital. Data relevant to referrals were reviewed. Between December 1, 2003, and April 1, 2006, 780 patients were imaged, 556 (71%) at the teaching hospital. Referrals came from 157 physicians in 17 different medical specialties, including adult cardiology (64%), cardiothoracic surgery (15%), pediatric cardiology (8%), internal medicine (6%), and others (7%). Overall, primary indications were function and viability evaluation (29%), aorta and valve assessment (24%), congenital heart disease (17%), arrhythmogenic right ventricular dysplasia exclusion (13%), cardiac masses (9%), pericardial disease (4%), and others (4%). Referrals for function and viability represented a much greater percentage of cases at the community hospital (53%) than at the teaching hospital (19%). The reverse was true for congenital heart disease (7% vs 21%). This study demonstrates that cardiac specialists generate the vast majority of referrals (87%) to the authors’ cardiac magnetic resonance imaging service and that there are substantial differences in the referral patterns between a large teaching hospital and a smaller community hospital. Key Words: Cardiac imaging, cardiac MRI, imaging service, referral patterns J Am Coll Radiol 2008;5:638-643. Copyright © 2008 American College of Radiology
INTRODUCTION Magnetic resonance imaging (MRI) has been used to evaluate the heart and great vessels for more than 2 decades. However, only in the past 5 or 6 years has cardiac MRI (CMR) emerged as a robust tool for routine clinical application [1-4]. This is due largely to technologic advances in both hardware and software that contribute to general improvements in image quality and acquisition speed. High-quality, detailed, clinically relevant MRI examinations can be performed in reasonable time frames. In addition to routine anatomic assessments of cardiac and aortic anatomy, novel pulse sequences permit detailed evaluations of myocardial function, blood flow physiology, and tissue characterization, most notably myocardial scar delineation. Because MRI has become a powerful tool for evaluating cardioaortic pathology, particularly in a problem-solving capacity, and because examination times are generally reasonable, it is possible to establish a dedicated CMR service within a busy, multipurpose MRI practice [5-7]. Department of Diagnostic Imaging, Rhode Island Hospital, Providence, Rhode Island. Corresponding author and reprints: Michael K. Atalay, MD, PhD, Rhode Island Hospital, Department of Diagnostic Imaging, 593 Eddy Street, Providence, RI 02903; e-mail: [email protected].
638
Although trends related to overall CMR utilization have been documented, the nature of CMR referrals has not [8,9]. Information pertaining to referrals may help design a new service or improve and tailor an existing service. It may provide a better understanding of the perceived role of CMR among our clinical colleagues, including relative strengths and weaknesses, and identify target audiences for education. The purpose of the present study was to retrospectively evaluate the referral patterns to our imaging department’s CMR service in its first 28 months of operation. MATERIALS AND METHODS Study Design and Inclusion Criteria Our institutional review board approved the design of this retrospective study, and all patient data were handled in a manner compliant with the Health Insurance Portability and Accountability Act. Informed consent was waived. In late November 2003, our imaging department established a dedicated CMR service for a 719-bed teaching hospital (TH) and a 247-bed community hospital (CH). A total of 4 half days per week of both radiologist and scanner time was allocated to this service. Both institutions have identical 1.5-T cardiac-capable MRI © 2008 American College of Radiology 0091-2182/08/$34.00 ● DOI 10.1016/j.jacr.2008.01.012
Atalay et al/Referrals to Cardiac MRI 639
scanners. Each hospital has an established cardiothoracic surgery division and a well-recognized adult cardiology division with fully equipped interventional suites. The TH also has both pediatric and adult cardiologists specializing in congenital heart disease. A CMR-specific referral form was generated and distributed widely by our marketing department throughout the hospitals’ referral regions. For this study, data relevant to our referrals and the respective magnetic resonance examinations were reviewed. This included review of all referral forms, patient demographics, and study reports. Referral forms that were generated by residents, fellows, nurses, or physician assistants were assigned to appropriate attending physicians. Attending physicians were categorized by area of specialty. Cardiac MRI Referral Indications Indications for referrals included: (1) function and viability evaluation, (2) congenital heart disease, (3) arrhythmogenic right ventricular dysplasia exclusion, (4) intracardiac or extracardiac mass, (5) pericardial disease, (6) anomalous coronary artery assessment, (7) bypass graft evaluation, (8) cardiac valve assessment, (9) thoracic aorta with or without aortic valve assessment, and (10) pulmonary vein mapping before radiofrequency ablation. Function and viability evaluation has a largely standardized protocol and represents a broad category covering referrals typically related to the evaluation of a suspected or known cardiomyopathy, ischemic or otherwise (eg, restrictive cardiomyopathy, hypertrophic cardiomyopathy, dilated cardio-
myopathy). When multiple indications were cited, only the indication relevant to the chief clinical concern was noted. As in most imaging departments, magnetic resonance angiography of the aorta and peripheral vasculature is routinely performed in our MRI practice outside of the CMR service. However, only those aortic cases specifically referred to the CMR service were included in this study. All of these cases included cine imaging in one or more planes, and many included phase contrast imaging of the aorta and aortic valve with the intent of identifying and quantifying various valve lesions or coarctation. Data Analysis Data were analyzed for the frequency of referral on a physician subspecialty basis and on an indication basis for both hospitals independently and combined. The number of cases referred by each physician was tallied, as was the total number of cases per month to assess the growth of the service. Patients’ ages and sex were also noted. An effort was made to determine how many referrals were based on the recommendations of radiologists. RESULTS Between December 1, 2003, and April 1, 2006, 780 patients were imaged, 556 (71%) at the TH. Most patients were men (59%), and most were over 18 years old (91%). Referrals came from 157 physicians in 17 different medical specialties, including adult cardiology (64%), cardiothoracic surgery (15%), pediatric cardiol-
Table 1. Referral patterns for cardiac magnetic resonance imaging by physician specialty Combined Data Teaching Hospital Community Hospital Specialty No. of Cases % No. of Cases % No. of Cases % Adult cardiology 497 64 319 57 178 79 Cardiothoracic surgery 120 15 104 19 16 7 Pediatric cardiology 61 8 59 11 2 1 Internal medicine 47 6 32 6 15 7 Nephrology 15 2 5 1 10 4 Pulmonary 12 2 12 2 0 0 Pediatrics 7 1 7 1 0 0 Oncology 6 1 6 1 0 0 General surgery 5 1 5 1 0 0 Rheumatology 2 0 2 0 0 0 Urology 2 0 1 0 1 0 Endocrine 1 0 1 0 0 0 Family practice 1 0 0 0 1 0 Infectious diseases 1 0 0 0 1 0 Neurology 1 0 1 0 0 0 Neurosurgery 1 0 1 0 0 0 Obstetrics-gynecology 1 0 1 0 0 0 Total 780 100% 556 100% 224 100%
640 Journal of the American College of Radiology/ Vol. 5 No. 5 May 2008
Table 2. Physician referral patterns for CMR No. of Physicians Who Referred That Many No. of CMR Cases Patients per Referring Physician Combined TH CH 1-4 125 82 43 5-9 16 12 4 10-14 6 1 5 15-19 2 2 0 20-29 2 1 1 ⱖ30 6 5 1
Total No. of Studies Referred by Those Physicians Combined 203 (26%) 102 (13%) 75 (10%) 30 (4%) 47 (6%) 323 (41%)
TH 134 (24%) 79 (14%) 13 (2%) 30 (5%) 26 (5%) 274 (49%)
CH 69 (31%) 23 (10%) 62 (28%) 0 (0%) 21 (9%) 49 (22%)
Note: CH ⫽ community hospital; CMR ⫽ cardiac magnetic resonance imaging; TH ⫽ teaching hospital.
ogy (8%), internal medicine (6%), and others (7%) (Table 1). Overall, 32 physicians referred 5 or more patients and accounted for 74% of the total referrals, 16 physicians referred 10 or more patients for 61% of the total, and 8 physicians referred 20 or more patients for 47% of the total (Table 2). Primary indications in order of frequency were function and viability evaluation (29%), aorta and valve assessment (24%), congenital heart disease (17%), arrhythmogenic right ventricular dysplasia exclusion (13%), cardiac masses (9%), pericardial disease (4%), and others (4%) (Table 3). Considerable differences were seen in referrals between the TH and the CH (Tables 1 to 3). For example, referrals for function and viability evaluation represented a greater percentage of cases at the CH (53%) than at the TH (19%). The reverse was true for congenital disease (6% at CH and 20% at the TH). Also, referrals from adult cardiologists constituted a greater percentage of the total at
the CH (79%) than at the TH (57%), whereas the percentage of referrals from cardiothoracic surgeons was higher at the TH (19%) than at the CH (7%). Growth for the first 18 months was steady, at a rate of approximately 2 additional studies per month (Figure 1). For the last 5 months of the period examined, 180 patients were imaged (36 per month), 128 at the TH (25.6 per month) and 52 at the CH (10.4 per month). Of the 780 referrals, 8 were confidently linked to the recommendations of radiologists. DISCUSSION Cardiac MRI has emerged as a powerful tool for imaging the heart and vascular system. As the technology continues to advance, the cardiovascular applications grow, image quality improves, and scan durations decrease. These factors contribute to an ever increasing role of MRI in the
Table 3. Indications for cardiac magnetic resonance imaging with relative frequencies Combined Data Teaching Hospital Community Hospital Indication Function and viability evaluation Aorta with or without aortic valve Congenital heart disease Arrhythmogenic right ventricular dysplasia Cardiac mass Pericardial disease Anomalous coronary Cardiac valve Pulmonary vein mapping Magnetic resonance pulmonary angiography Bypass graft patency Total
No. of Cases 225 184
% 29 24
No. of Cases 106 156
% 19 28
No. of Cases 119 28
% 53 13
130 102
17 13
114 83
21 15
16 19
7 8
74 35 13 11 3 2
9 4 2 1 0 0
46 29 8 9 3 2
8 5 1 2 1 0
28 6 5 2 0 0
13 3 2 1 0 0
1 780
0 100%
0 556
0 100%
1 224
0 100%
Atalay et al/Referrals to Cardiac MRI 641
Fig. 1. Number of cardiac magnetic resonance imaging (CMR) studies performed each month from the end of November 2003, when the CMR service become operational, to the end of March 2006.
diagnosis and management of cardiovascular disease. Consequently, more and more facilities are conducting CMR studies, and some are creating dedicated CMR services. Our experience developing a dedicated service in a diagnostic imaging department has been described previously [5]. In our case, a cardiac-specific referral form has been generated and widely distributed, and regular scanner time is allocated each week for CMR studies. In this study, we observed that 87% of referrals to our CMR service come from cardiac specialists. It is likely that the actual percentage in which cardiac specialists have a role is higher. Many referrals from noncardiac specialists, on the basis of the physicians’ names on the referral forms, were likely prompted by consultation with cardiac specialists. The implications of our study are that the success of a CMR service on the basis of study volume rests heavily on the relationships of the imaging service with physicians who specialize in cardiovascular disease, particularly adult cardiologists. These relationships presumably rely in turn on examination quality, interpretation accuracy, report turnaround time, additive examination value, and other typical determinants of referring physicians’ confidence and satisfaction. In essence, a favorable working relationship with cardiologists and cardiothoracic surgeons is critical for CMR success. There are a few potential explanations for the strong preponderance of referrals from cardiac specialists. First, although CMR is clearly versatile and very powerful, it is currently viewed largely as a problem-solving tool rather than a first-line diagnostic measure. Physicians are in-
clined to invoke MRI when other more conventional cardiac diagnostic methods are spurious or inconclusive. Often, by this time, a cardiac specialist is involved. Second, the education of potential referring physicians on the strengths and applications of CMR is lacking. Many physicians, including both cardiologists and radiologists, struggle to understand the role of MRI in the evaluation of the heart, among the myriad other more traditional techniques in their arsenal. The education of potential referring physicians is an ongoing endeavor, particularly given the evolving indications and technologic advances. Recently, appropriateness criteria for both CMR and cardiac CT have been established [10]. These criteria will undoubtedly help guide utilization. One notably absent referral indication in our study was pharmacologic stress perfusion using MRI. This is a somewhat more involved application and one that is gaining recognition (and popularity) as a reasonable alternative to other methods for detecting flow-limiting coronary artery stenoses [11-14]. A comprehensive examination, including first-pass stress imaging, viability, and function, may indeed provide equal or greater information than other techniques. As such, it may become a routine, or “primary,” indication for CMR from noncardiac specialists, just as is a stress-rest nuclear examination. Our service currently offers stress perfusion CMR, but to date, we have had little interest from potential referring physicians. This is again likely due to limited education, to physician confidence in the technique, and to referral inertia, from both cardiac and noncardiac specialists. In
642 Journal of the American College of Radiology/ Vol. 5 No. 5 May 2008
time, we may see this emerge as a more substantial referral indication, particularly from primary care physicians and family practitioners. Of all the referrals to our service, only 8 (1%) can be confidently linked to the recommendations of radiologists. Radiologists may represent a large, relatively untapped resource of referrals to a CMR service. The detection of cardiac pathology on routine chest or abdominal computed tomographic scan may lead directly to appropriate CMR referrals. For example, right-heart pathology, such as dilated chambers or right ventricular hypertrophy, can be better evaluated with MRI than with echocardiography. Congenital lesions detected on computed tomography (CT), such as anomalous pulmonary veins, can be further evaluated and accurately quantified using MRI. Intracardiac and extracardiac masses detected on CT, including suspected cavitary thrombus, are typically better characterized with MRI [15-17]. The pericardium is better visualized with MRI than with echocardiography, and the functional significance of pericardial effusions or calcifications seen on CT can be elaborated with MRI. Incidentally detected left ventricular myocardial fat may be due to fatty metaplasia of unsuspected myocardial infarction [18,19]. If a patient’s heart disease is unknown, MRI may be appropriate to evaluate the extent of infarction, its effect on function, and the likelihood of recovery with revascularization. As these examples illustrate, awareness of cardiac pathology on CT may benefit a CMR service. Differences observed between referral patterns at our 2 hospitals are likely reflective of their relative strengths and subspecialization, and the patterns may as a consequence not be generalizable. For example, pediatric and adult congenital heart sections are present at our TH but not at our CH. Conversely, the CH has a strong general cardiology tradition. Knowledge of the particular strengths of the cardiology and cardiac surgery divisions of a hospital may guide the development and “marketing” of the CMR service. Of interest, it is noted in our experience that the percentage of cases referred at the TH (71%) is slightly lower than the relative sizes of the hospitals on the basis of bed census (75%). The overall growth of the service in terms of the number of referrals per month was steady for the first 18 months of the period evaluated but seemed to plateau during the last 5 months (Figure 1). The plateau is likely due to a number of factors. One chief factor is the emergence of competition; a local cardiology group, which previously referred CMR cases to our service, opened its own outpatient CMR service. Other factors leading to a plateau of referrals may include limited referring physician education, nonsustained marketing efforts, high examination costs, and perceived or real limitations of scanner and reader availability. For many reasons outlined
above, it is unlikely that we have reached a saturation of referrals at which any and all patients with appropriate indications for CMR are being scanned. In summary, the vast majority of referrals to the CMR service of our imaging department come from physicians specializing in the cardiovascular system. As such, the success of our service depends largely on a strong working relationship with these specialists, most notably adult cardiologists. One potentially appealing option in developing and growing a CMR service is to include cardiologists in magnetic resonance study performance and interpretation. A host of financial and political hurdles would need to be overcome, but by combining the expertise of the different specialties, the benefits of such a collaboration could be vast. In general, efforts to broaden the referral basis may benefit from educational initiatives, including the education of fellow radiologists, who through their reports, recommendations, and discussions may be able to propagate understanding of CMR and influence referral patterns. REFERENCES 1. Constantine G, Shan K, Flamm SD, Sivananthan MU. Role of MRI in clinical cardiology. Lancet 2004;363:2162-71. 2. Earls JP, Ho VB, Foo TK, Castillo E, Flamm SD. Cardiac MRI: recent progress and continued challenges. J Magn Reson Imaging 2002;16: 111-27. 3. Lima JA, Desai MY. Cardiovascular magnetic resonance imaging: current and emerging applications. J Am Coll Cardiol 2004;44:1164-71. 4. Poon M, Fuster V, Fayad Z. Cardiac magnetic resonance imaging: a “one-stop-shop” evaluation of myocardial dysfunction. Curr Opin Cardiol 2002;17:663-70. 5. Atalay MK. Establishing a cardiac MRI program: problems, pitfalls, and expectations. J Am Coll Radiol 2005;2:740-8. 6. Earls JP. Planning and education build cardiac CT and MR service. Diagn Imaging 2004;11(suppl):2-6. 7. Wolff SD, Comeau CR. Setting up a clinical cardiac MR imaging program: practical issues and economics. Magn Reson Imaging Clin North Am 2003;11:19-26. 8. Stillman AE. What has happened to cardiac MRI? J Am Coll Radiol 2007;4:224-6. 9. Wittram C, Meehan MJ, Halpern EF, Shepard JA, McLoud TC, Thrall JH. Trends in thoracic radiology over a decade at a large academic medical center. J Thorac Imaging 2004;19:164-70. 10. Hendel RC, Patel MR, Kramer CM, et al. ACCF/ACR/SCCT/SCMR/ ASNC/NASCI/SCAI/SIR 2006 appropriateness criteria for cardiac computed tomography and cardiac magnetic resonance imaging: a report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group, American College of Radiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, American Society of Nuclear Cardiology, North American Society for Cardiac Imaging, Society for Cardiovascular Angiography and Interventions, and Society of Interventional Radiology. J Am Coll Cardiol 2006;48:1475-97.
Atalay et al/Referrals to Cardiac MRI 643 11. Ingkanisorn WP, Kwong RY, Bohme NS, et al. Prognosis of negative adenosine stress magnetic resonance in patients presenting to an emergency department with chest pain. J Am Coll Cardiol 2006;47:1427-32. 12. Pennell DJ. Cardiovascular magnetic resonance and the role of adenosine pharmacologic stress. Am J Cardiol 2004;94(suppl):26D-31D. 13. Pilz G, Bernhardt P, Klos M, Ali E, Wild M, Hofling B. Clinical implication of adenosine-stress cardiac magnetic resonance imaging as potential gatekeeper prior to invasive examination in patients with AHA/ACC class II indication for coronary angiography. Clin Res Cardiol 2006;95: 531-8. 14. Plein S, Radjenovic A, Ridgway JP, et al. Coronary artery disease: myocardial perfusion MR imaging with sensitivity encoding versus conventional angiography. Radiology 2005;235:423-30.
15. Gilkeson RC, Chiles C. MR evaluation of cardiac and pericardial malignancy. Magn Reson Imaging Clin North Am 2003;11:173-86. 16. Schvartzman PR, White RD. Imaging of cardiac and paracardiac masses. J Thorac Imaging 2000;15:265-73. 17. White CS. MR evaluation of the pericardium and cardiac malignancies. Magn Reson Imaging Clin North Am 1996;4:237-51. 18. Baroldi G, Silver MD, De Maria R, Parodi O, Pellegrini A. Lipomatous metaplasia in left ventricular scar. Can J Cardiol 1997;13:65-71. 19. Winer-Muram HT, Tann M, Aisen AM, Ford L, Jennings SG, Bretz R. Computed tomography demonstration of lipomatous metaplasia of the left ventricle following myocardial infarction. J Comput Assist Tomogr 2004;28:455-8.
The purpose of this study was to retrospectively evaluate the referral patterns to the authors’ cardiac magnetic resonance imaging service in its first 28 months of operation. In late November 2003, the authors’ radiology practice established a cardiac magnetic resonance imaging service for a 719-bed teaching hospital and a 247-bed community hospital. Data relevant to referrals were reviewed. Between December 1, 2003, and April 1, 2006, 780 patients were imaged, 556 (71%) at the teaching hospital. Referrals came from 157 physicians in 17 different medical specialties, including adult cardiology (64%), cardiothoracic surgery (15%), pediatric cardiology (8%), internal medicine (6%), and others (7%). Overall, primary indications were function and viability evaluation (29%), aorta and valve assessment (24%), congenital heart disease (17%), arrhythmogenic right ventricular dysplasia exclusion (13%), cardiac masses (9%), pericardial disease (4%), and others (4%). Referrals for function and viability represented a much greater percentage of cases at the community hospital (53%) than at the teaching hospital (19%). The reverse was true for congenital heart disease (7% vs 21%). This study demonstrates that cardiac specialists generate the vast majority of referrals (87%) to the authors’ cardiac magnetic resonance imaging service and that there are substantial differences in the referral patterns between a large teaching hospital and a smaller community hospital. Key Words: Cardiac imaging, cardiac MRI, imaging service, referral patterns J Am Coll Radiol 2008;5:638-643. Copyright © 2008 American College of Radiology
INTRODUCTION Magnetic resonance imaging (MRI) has been used to evaluate the heart and great vessels for more than 2 decades. However, only in the past 5 or 6 years has cardiac MRI (CMR) emerged as a robust tool for routine clinical application [1-4]. This is due largely to technologic advances in both hardware and software that contribute to general improvements in image quality and acquisition speed. High-quality, detailed, clinically relevant MRI examinations can be performed in reasonable time frames. In addition to routine anatomic assessments of cardiac and aortic anatomy, novel pulse sequences permit detailed evaluations of myocardial function, blood flow physiology, and tissue characterization, most notably myocardial scar delineation. Because MRI has become a powerful tool for evaluating cardioaortic pathology, particularly in a problem-solving capacity, and because examination times are generally reasonable, it is possible to establish a dedicated CMR service within a busy, multipurpose MRI practice [5-7]. Department of Diagnostic Imaging, Rhode Island Hospital, Providence, Rhode Island. Corresponding author and reprints: Michael K. Atalay, MD, PhD, Rhode Island Hospital, Department of Diagnostic Imaging, 593 Eddy Street, Providence, RI 02903; e-mail: [email protected].
638
Although trends related to overall CMR utilization have been documented, the nature of CMR referrals has not [8,9]. Information pertaining to referrals may help design a new service or improve and tailor an existing service. It may provide a better understanding of the perceived role of CMR among our clinical colleagues, including relative strengths and weaknesses, and identify target audiences for education. The purpose of the present study was to retrospectively evaluate the referral patterns to our imaging department’s CMR service in its first 28 months of operation. MATERIALS AND METHODS Study Design and Inclusion Criteria Our institutional review board approved the design of this retrospective study, and all patient data were handled in a manner compliant with the Health Insurance Portability and Accountability Act. Informed consent was waived. In late November 2003, our imaging department established a dedicated CMR service for a 719-bed teaching hospital (TH) and a 247-bed community hospital (CH). A total of 4 half days per week of both radiologist and scanner time was allocated to this service. Both institutions have identical 1.5-T cardiac-capable MRI © 2008 American College of Radiology 0091-2182/08/$34.00 ● DOI 10.1016/j.jacr.2008.01.012
Atalay et al/Referrals to Cardiac MRI 639
scanners. Each hospital has an established cardiothoracic surgery division and a well-recognized adult cardiology division with fully equipped interventional suites. The TH also has both pediatric and adult cardiologists specializing in congenital heart disease. A CMR-specific referral form was generated and distributed widely by our marketing department throughout the hospitals’ referral regions. For this study, data relevant to our referrals and the respective magnetic resonance examinations were reviewed. This included review of all referral forms, patient demographics, and study reports. Referral forms that were generated by residents, fellows, nurses, or physician assistants were assigned to appropriate attending physicians. Attending physicians were categorized by area of specialty. Cardiac MRI Referral Indications Indications for referrals included: (1) function and viability evaluation, (2) congenital heart disease, (3) arrhythmogenic right ventricular dysplasia exclusion, (4) intracardiac or extracardiac mass, (5) pericardial disease, (6) anomalous coronary artery assessment, (7) bypass graft evaluation, (8) cardiac valve assessment, (9) thoracic aorta with or without aortic valve assessment, and (10) pulmonary vein mapping before radiofrequency ablation. Function and viability evaluation has a largely standardized protocol and represents a broad category covering referrals typically related to the evaluation of a suspected or known cardiomyopathy, ischemic or otherwise (eg, restrictive cardiomyopathy, hypertrophic cardiomyopathy, dilated cardio-
myopathy). When multiple indications were cited, only the indication relevant to the chief clinical concern was noted. As in most imaging departments, magnetic resonance angiography of the aorta and peripheral vasculature is routinely performed in our MRI practice outside of the CMR service. However, only those aortic cases specifically referred to the CMR service were included in this study. All of these cases included cine imaging in one or more planes, and many included phase contrast imaging of the aorta and aortic valve with the intent of identifying and quantifying various valve lesions or coarctation. Data Analysis Data were analyzed for the frequency of referral on a physician subspecialty basis and on an indication basis for both hospitals independently and combined. The number of cases referred by each physician was tallied, as was the total number of cases per month to assess the growth of the service. Patients’ ages and sex were also noted. An effort was made to determine how many referrals were based on the recommendations of radiologists. RESULTS Between December 1, 2003, and April 1, 2006, 780 patients were imaged, 556 (71%) at the TH. Most patients were men (59%), and most were over 18 years old (91%). Referrals came from 157 physicians in 17 different medical specialties, including adult cardiology (64%), cardiothoracic surgery (15%), pediatric cardiol-
Table 1. Referral patterns for cardiac magnetic resonance imaging by physician specialty Combined Data Teaching Hospital Community Hospital Specialty No. of Cases % No. of Cases % No. of Cases % Adult cardiology 497 64 319 57 178 79 Cardiothoracic surgery 120 15 104 19 16 7 Pediatric cardiology 61 8 59 11 2 1 Internal medicine 47 6 32 6 15 7 Nephrology 15 2 5 1 10 4 Pulmonary 12 2 12 2 0 0 Pediatrics 7 1 7 1 0 0 Oncology 6 1 6 1 0 0 General surgery 5 1 5 1 0 0 Rheumatology 2 0 2 0 0 0 Urology 2 0 1 0 1 0 Endocrine 1 0 1 0 0 0 Family practice 1 0 0 0 1 0 Infectious diseases 1 0 0 0 1 0 Neurology 1 0 1 0 0 0 Neurosurgery 1 0 1 0 0 0 Obstetrics-gynecology 1 0 1 0 0 0 Total 780 100% 556 100% 224 100%
640 Journal of the American College of Radiology/ Vol. 5 No. 5 May 2008
Table 2. Physician referral patterns for CMR No. of Physicians Who Referred That Many No. of CMR Cases Patients per Referring Physician Combined TH CH 1-4 125 82 43 5-9 16 12 4 10-14 6 1 5 15-19 2 2 0 20-29 2 1 1 ⱖ30 6 5 1
Total No. of Studies Referred by Those Physicians Combined 203 (26%) 102 (13%) 75 (10%) 30 (4%) 47 (6%) 323 (41%)
TH 134 (24%) 79 (14%) 13 (2%) 30 (5%) 26 (5%) 274 (49%)
CH 69 (31%) 23 (10%) 62 (28%) 0 (0%) 21 (9%) 49 (22%)
Note: CH ⫽ community hospital; CMR ⫽ cardiac magnetic resonance imaging; TH ⫽ teaching hospital.
ogy (8%), internal medicine (6%), and others (7%) (Table 1). Overall, 32 physicians referred 5 or more patients and accounted for 74% of the total referrals, 16 physicians referred 10 or more patients for 61% of the total, and 8 physicians referred 20 or more patients for 47% of the total (Table 2). Primary indications in order of frequency were function and viability evaluation (29%), aorta and valve assessment (24%), congenital heart disease (17%), arrhythmogenic right ventricular dysplasia exclusion (13%), cardiac masses (9%), pericardial disease (4%), and others (4%) (Table 3). Considerable differences were seen in referrals between the TH and the CH (Tables 1 to 3). For example, referrals for function and viability evaluation represented a greater percentage of cases at the CH (53%) than at the TH (19%). The reverse was true for congenital disease (6% at CH and 20% at the TH). Also, referrals from adult cardiologists constituted a greater percentage of the total at
the CH (79%) than at the TH (57%), whereas the percentage of referrals from cardiothoracic surgeons was higher at the TH (19%) than at the CH (7%). Growth for the first 18 months was steady, at a rate of approximately 2 additional studies per month (Figure 1). For the last 5 months of the period examined, 180 patients were imaged (36 per month), 128 at the TH (25.6 per month) and 52 at the CH (10.4 per month). Of the 780 referrals, 8 were confidently linked to the recommendations of radiologists. DISCUSSION Cardiac MRI has emerged as a powerful tool for imaging the heart and vascular system. As the technology continues to advance, the cardiovascular applications grow, image quality improves, and scan durations decrease. These factors contribute to an ever increasing role of MRI in the
Table 3. Indications for cardiac magnetic resonance imaging with relative frequencies Combined Data Teaching Hospital Community Hospital Indication Function and viability evaluation Aorta with or without aortic valve Congenital heart disease Arrhythmogenic right ventricular dysplasia Cardiac mass Pericardial disease Anomalous coronary Cardiac valve Pulmonary vein mapping Magnetic resonance pulmonary angiography Bypass graft patency Total
No. of Cases 225 184
% 29 24
No. of Cases 106 156
% 19 28
No. of Cases 119 28
% 53 13
130 102
17 13
114 83
21 15
16 19
7 8
74 35 13 11 3 2
9 4 2 1 0 0
46 29 8 9 3 2
8 5 1 2 1 0
28 6 5 2 0 0
13 3 2 1 0 0
1 780
0 100%
0 556
0 100%
1 224
0 100%
Atalay et al/Referrals to Cardiac MRI 641
Fig. 1. Number of cardiac magnetic resonance imaging (CMR) studies performed each month from the end of November 2003, when the CMR service become operational, to the end of March 2006.
diagnosis and management of cardiovascular disease. Consequently, more and more facilities are conducting CMR studies, and some are creating dedicated CMR services. Our experience developing a dedicated service in a diagnostic imaging department has been described previously [5]. In our case, a cardiac-specific referral form has been generated and widely distributed, and regular scanner time is allocated each week for CMR studies. In this study, we observed that 87% of referrals to our CMR service come from cardiac specialists. It is likely that the actual percentage in which cardiac specialists have a role is higher. Many referrals from noncardiac specialists, on the basis of the physicians’ names on the referral forms, were likely prompted by consultation with cardiac specialists. The implications of our study are that the success of a CMR service on the basis of study volume rests heavily on the relationships of the imaging service with physicians who specialize in cardiovascular disease, particularly adult cardiologists. These relationships presumably rely in turn on examination quality, interpretation accuracy, report turnaround time, additive examination value, and other typical determinants of referring physicians’ confidence and satisfaction. In essence, a favorable working relationship with cardiologists and cardiothoracic surgeons is critical for CMR success. There are a few potential explanations for the strong preponderance of referrals from cardiac specialists. First, although CMR is clearly versatile and very powerful, it is currently viewed largely as a problem-solving tool rather than a first-line diagnostic measure. Physicians are in-
clined to invoke MRI when other more conventional cardiac diagnostic methods are spurious or inconclusive. Often, by this time, a cardiac specialist is involved. Second, the education of potential referring physicians on the strengths and applications of CMR is lacking. Many physicians, including both cardiologists and radiologists, struggle to understand the role of MRI in the evaluation of the heart, among the myriad other more traditional techniques in their arsenal. The education of potential referring physicians is an ongoing endeavor, particularly given the evolving indications and technologic advances. Recently, appropriateness criteria for both CMR and cardiac CT have been established [10]. These criteria will undoubtedly help guide utilization. One notably absent referral indication in our study was pharmacologic stress perfusion using MRI. This is a somewhat more involved application and one that is gaining recognition (and popularity) as a reasonable alternative to other methods for detecting flow-limiting coronary artery stenoses [11-14]. A comprehensive examination, including first-pass stress imaging, viability, and function, may indeed provide equal or greater information than other techniques. As such, it may become a routine, or “primary,” indication for CMR from noncardiac specialists, just as is a stress-rest nuclear examination. Our service currently offers stress perfusion CMR, but to date, we have had little interest from potential referring physicians. This is again likely due to limited education, to physician confidence in the technique, and to referral inertia, from both cardiac and noncardiac specialists. In
642 Journal of the American College of Radiology/ Vol. 5 No. 5 May 2008
time, we may see this emerge as a more substantial referral indication, particularly from primary care physicians and family practitioners. Of all the referrals to our service, only 8 (1%) can be confidently linked to the recommendations of radiologists. Radiologists may represent a large, relatively untapped resource of referrals to a CMR service. The detection of cardiac pathology on routine chest or abdominal computed tomographic scan may lead directly to appropriate CMR referrals. For example, right-heart pathology, such as dilated chambers or right ventricular hypertrophy, can be better evaluated with MRI than with echocardiography. Congenital lesions detected on computed tomography (CT), such as anomalous pulmonary veins, can be further evaluated and accurately quantified using MRI. Intracardiac and extracardiac masses detected on CT, including suspected cavitary thrombus, are typically better characterized with MRI [15-17]. The pericardium is better visualized with MRI than with echocardiography, and the functional significance of pericardial effusions or calcifications seen on CT can be elaborated with MRI. Incidentally detected left ventricular myocardial fat may be due to fatty metaplasia of unsuspected myocardial infarction [18,19]. If a patient’s heart disease is unknown, MRI may be appropriate to evaluate the extent of infarction, its effect on function, and the likelihood of recovery with revascularization. As these examples illustrate, awareness of cardiac pathology on CT may benefit a CMR service. Differences observed between referral patterns at our 2 hospitals are likely reflective of their relative strengths and subspecialization, and the patterns may as a consequence not be generalizable. For example, pediatric and adult congenital heart sections are present at our TH but not at our CH. Conversely, the CH has a strong general cardiology tradition. Knowledge of the particular strengths of the cardiology and cardiac surgery divisions of a hospital may guide the development and “marketing” of the CMR service. Of interest, it is noted in our experience that the percentage of cases referred at the TH (71%) is slightly lower than the relative sizes of the hospitals on the basis of bed census (75%). The overall growth of the service in terms of the number of referrals per month was steady for the first 18 months of the period evaluated but seemed to plateau during the last 5 months (Figure 1). The plateau is likely due to a number of factors. One chief factor is the emergence of competition; a local cardiology group, which previously referred CMR cases to our service, opened its own outpatient CMR service. Other factors leading to a plateau of referrals may include limited referring physician education, nonsustained marketing efforts, high examination costs, and perceived or real limitations of scanner and reader availability. For many reasons outlined
above, it is unlikely that we have reached a saturation of referrals at which any and all patients with appropriate indications for CMR are being scanned. In summary, the vast majority of referrals to the CMR service of our imaging department come from physicians specializing in the cardiovascular system. As such, the success of our service depends largely on a strong working relationship with these specialists, most notably adult cardiologists. One potentially appealing option in developing and growing a CMR service is to include cardiologists in magnetic resonance study performance and interpretation. A host of financial and political hurdles would need to be overcome, but by combining the expertise of the different specialties, the benefits of such a collaboration could be vast. In general, efforts to broaden the referral basis may benefit from educational initiatives, including the education of fellow radiologists, who through their reports, recommendations, and discussions may be able to propagate understanding of CMR and influence referral patterns. REFERENCES 1. Constantine G, Shan K, Flamm SD, Sivananthan MU. Role of MRI in clinical cardiology. Lancet 2004;363:2162-71. 2. Earls JP, Ho VB, Foo TK, Castillo E, Flamm SD. Cardiac MRI: recent progress and continued challenges. J Magn Reson Imaging 2002;16: 111-27. 3. Lima JA, Desai MY. Cardiovascular magnetic resonance imaging: current and emerging applications. J Am Coll Cardiol 2004;44:1164-71. 4. Poon M, Fuster V, Fayad Z. Cardiac magnetic resonance imaging: a “one-stop-shop” evaluation of myocardial dysfunction. Curr Opin Cardiol 2002;17:663-70. 5. Atalay MK. Establishing a cardiac MRI program: problems, pitfalls, and expectations. J Am Coll Radiol 2005;2:740-8. 6. Earls JP. Planning and education build cardiac CT and MR service. Diagn Imaging 2004;11(suppl):2-6. 7. Wolff SD, Comeau CR. Setting up a clinical cardiac MR imaging program: practical issues and economics. Magn Reson Imaging Clin North Am 2003;11:19-26. 8. Stillman AE. What has happened to cardiac MRI? J Am Coll Radiol 2007;4:224-6. 9. Wittram C, Meehan MJ, Halpern EF, Shepard JA, McLoud TC, Thrall JH. Trends in thoracic radiology over a decade at a large academic medical center. J Thorac Imaging 2004;19:164-70. 10. Hendel RC, Patel MR, Kramer CM, et al. ACCF/ACR/SCCT/SCMR/ ASNC/NASCI/SCAI/SIR 2006 appropriateness criteria for cardiac computed tomography and cardiac magnetic resonance imaging: a report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group, American College of Radiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, American Society of Nuclear Cardiology, North American Society for Cardiac Imaging, Society for Cardiovascular Angiography and Interventions, and Society of Interventional Radiology. J Am Coll Cardiol 2006;48:1475-97.
Atalay et al/Referrals to Cardiac MRI 643 11. Ingkanisorn WP, Kwong RY, Bohme NS, et al. Prognosis of negative adenosine stress magnetic resonance in patients presenting to an emergency department with chest pain. J Am Coll Cardiol 2006;47:1427-32. 12. Pennell DJ. Cardiovascular magnetic resonance and the role of adenosine pharmacologic stress. Am J Cardiol 2004;94(suppl):26D-31D. 13. Pilz G, Bernhardt P, Klos M, Ali E, Wild M, Hofling B. Clinical implication of adenosine-stress cardiac magnetic resonance imaging as potential gatekeeper prior to invasive examination in patients with AHA/ACC class II indication for coronary angiography. Clin Res Cardiol 2006;95: 531-8. 14. Plein S, Radjenovic A, Ridgway JP, et al. Coronary artery disease: myocardial perfusion MR imaging with sensitivity encoding versus conventional angiography. Radiology 2005;235:423-30.
15. Gilkeson RC, Chiles C. MR evaluation of cardiac and pericardial malignancy. Magn Reson Imaging Clin North Am 2003;11:173-86. 16. Schvartzman PR, White RD. Imaging of cardiac and paracardiac masses. J Thorac Imaging 2000;15:265-73. 17. White CS. MR evaluation of the pericardium and cardiac malignancies. Magn Reson Imaging Clin North Am 1996;4:237-51. 18. Baroldi G, Silver MD, De Maria R, Parodi O, Pellegrini A. Lipomatous metaplasia in left ventricular scar. Can J Cardiol 1997;13:65-71. 19. Winer-Muram HT, Tann M, Aisen AM, Ford L, Jennings SG, Bretz R. Computed tomography demonstration of lipomatous metaplasia of the left ventricle following myocardial infarction. J Comput Assist Tomogr 2004;28:455-8.