Medical imaging is undergoing exponential growth. A recent report from the US Government Accountability Office noted a doubling of Medicare spending on imaging services from 2000 to 2006.1 Cardiovascular imaging represents the largest portion of these totals, and there has been universal acknowledgment of the need to increase quality and reduce inappropriate testing in an effort to control costs.2 In parallel with this trend, thought leaders within academic cardiology are calling for the training of a new generation of cardiologists with expertise in multimodality imaging (MMI)—namely echocardiography, nuclear cardiology, cardiovascular computed tomography (CCT), and cardiovascular magnetic resonance (CMR)—instead of the traditional one or two.3–6 The rationale is to provide future cardiovascular imaging subspecialists with a broader base of skills in both mature and evolving technologies, resulting in more efficient, cost-effective use with improved diagnostic accuracy and patient outcomes. The release of the American College of Cardiology (ACC) Training Statement on Multimodality Noninvasive Cardiovascular Imaging in 2008 provides a basic structure for fellowship training (see Table 1) that follows this new paradigm, recommending a ‘patient-centered’ rather than ‘technology-centered’ approach; consolidating the technical curriculum of each modality into unified, ‘parallel’ training; assessing imaging competency instead of the traditional ‘apprentice model;’ and proposing a combined reading room where these integrated imaging practices can occur.7
In addition to training concerns, cardiovascular imaging subspecialists face passage of a ‘non-traditional’ set of board examinations in three of the four modalities (echocardiography, nuclear cardiology, and computed tomography—there is currently no exam for magnetic resonance) that have been established outside the realm of the American Boards of Medical Subspecialities (ABMS) as passage of the American Board of Internal Medicine (ABIM) examination in cardiovascular disease is seen as an insufficient measure of detailed imaging knowledge.8 Eligibility and maintenance of competence for these exams extend beyond simple board passage, requiring continuing education hours and an implication that practitioners will be lifelong learners and consumers of publications pertinent to cardiovascular imaging.
While there is sufficient momentum behind the development of a future expert in MMI, the idea itself has far outpaced any data to prove its superiority. Establishing this model of practice will require a specific cohort of leaders in academic cardiovascular imaging who can demonstrate proficiency in all major imaging technologies while training cardiology fellows as they develop a research career in an area largely devoid of similarly trained mentors. The objective of this study was to look beyond the MMI training statement and identify the practical logistics of training, maintaining technical proficiency, obtaining credentialing, and achieving board certification in all four major imaging modalities in order to develop a career as an academic cardiovascular imaging specialist involved in the training of future practitioners.
Imaging training data were compiled from a list of academic medical centers with Accreditation Council of Graduate Medical Education (ACGME)-accredited fellowships in cardiovascular medicine via the Electronic Residency Application Service’s (ERAS) online database. These programs were cross-referenced with available data detailing advanced training in CMR and CCT listed on the educational portion of the websites for their associated societies (the Society for Cardiovascular Magnetic Resonance [SCMR: www.scmr.org] and the Society of Cardiovascular Computed Tomography [SCCT: www.scct.org]). Because additional training in imaging is not an ACGME-accredited fellowship, no official roster of available programs exists.
The monthly volume of publications was aggregated from the tables of contents for all commonly read general cardiology and subspecialty cardiology journals surveyed in summer 2009 using our institution’s library website. Each journal’s monthly page total (including editorials, original research, review articles, and letters to the editor) was averaged using a three-month period (June to August 2009). If journals were published on a weekly basis, four consecutive issues were combined to calculate page totals. If they were published every six or eight weeks, an average four-week, or one month, total was calculated.
The training and competency statements issued jointly by the ACC and the various subspecialty societies (available online on the ACC’s website: www.acc.org) were reviewed to compile the full amount of continuing medical education (CME) requirements for each cardiovascular subspecialty during a standard three-year period. All data delineating the cost of subspecialty societal membership, board certification, and fellowship application were tabulated from the websites of each representative organization:
- general cardiology: ACC, American Heart Association (AHA);
- electrophysiology: Heart Rhythm Society (HRS);
- heart failure/transplant: Heart Failure Society of America (HFSA), International Society for Heart and Lung Transplantation (ISHLT);
- interventional cardiology: Society for Cardiovascular Angiography and Interventions (SCAI); and
- cardiovascular imaging: American Society of Echocardiography (ASE), American Society of Nuclear Cardiology (ASNC), SCMR, and SCCT.
Training in Multimodality Imaging. To merely be eligible for competency in MMI requires some degree of serendipity. A candidate’s fellowship training program must have faculty and facilities that allow adequate exposure to all types of cardiovascular imaging. Of the 162 fellowships in cardiology that participate in the ERAS, only 21 of them (13%) have active three- to 12-month training in CMR documented by the SCMR (www.scmr.org/education/fellowships.cfm). Many of these programs also offer CCT exposure, but there are no available statistics describing penetrance into fellowship training programs in the US. It is also unknown which programs may have arrangements with other training sites to provide CMR or CCT exposure for their fellows if it is not offered at their center. There are ample educational opportunities for physicians in practice to pay for level 2 (in most cases) or even level 3 (less frequently) training in CT. Fewer courses offer level 2 CMR training, but level 3 requires at least one full year of formal fellowship. Fellows enrolled in ACGME-accredited fellowships are typically limited in their ability to participate in private courses or rotations away from their hospitals for either financial or medical–legal reasons.
Technical Proficiency. Compared with other specialized areas within cardiology, namely electrophysiology, heart failure/transplantation, and interventional cardiology, MMI poses a variety of unique challenges. Most importantly, it requires technical proficiency across four specific areas that are each demanding in isolation. In addition to the everyday technical knowledge needed to practice clinical multimodality cardiovascular imaging, a physician active in all four areas must stay current with the primary literature.
Again, this differs in comparison with the other specialized areas of cardiology. Of the highest-impact journals considered necessary reading in ‘general’ cardiology (Circulation, Journal of the American College of Cardiology, American Heart Journal, and American Journal of Cardiology), the amount of cumulative monthly material devoted to the main components of the table of contents equaled 1,575 pages (see Table 2). By narrowing each subspecialty area to the journal published by its representative society, imaging stands alone with four separate journals and an additional average of 276 pages per month—21% more than the next highest total (in heart failure/transplant). Granted, all of these societal journals have a lower average impact score (2.52) than the core group of general cardiology journals (6.79),9 but the need to be conversant in the latest technical publications makes these required reading for any physician active in all four types of imaging. Additionally, each subspecialty has noted an increase in spin-off journals from both the ACC and the AHA. As of 2009, there are six additional journals—specifically, in electrophysiology (Circ Electrophysiol), heart failure/transplant (Circ Heart Fail), interventional cardiology (Circ Cardiovasc Intervent and J Am Coll Cardiol Intv), and cardiovascular imaging (Circ Cardiovasc Imaging and J Am Coll Cardiol Img)—which add to the above totals but do not alter the overall ratios.
Credentialing and Board Certification. Hospital credentialing for the performance and interpretation of cardiovascular imaging procedures commonly requires adequate training defined by the ACC series of Competence and Training Statements (either Level 2— ‘Additional training in one or more specialized areas that enables the cardiologist to perform or interpret (or both) specific procedures at an intermediate skill level’— or Level 3—’Advanced training in a specialized area that enables a cardiologist to perform, interpret, and train others to perform and interpret specific procedures at a high skill level’), a minimum number of procedures per year to maintain technical skills, and a current medical license. The CME hours for all four areas of subspeciality cardiology are listed in Table 3. Strict adherence to the totals dictated by these documents in echocardiography, nuclear stress testing, and CMR/CCT would require a specialist in MMI to obtain a minimum of 90 (if level 2 in all areas) and a maximum of 140 hours (level 2 in echocardiography and nuclear cardiology; level 3 in CMR and CCT) of total CME in three years. Compared with the other three subspecialty areas, these totals are 300% and 467% higher, respectively, than each corresponding total of 30 required hours in the same time period.
In similar fashion, board certification in imaging is fragmented, expensive and laborious. It commonly starts with membership in each respective society (see Table 3), again in contrast to other subspecialty areas of cardiology, each with membership in only one or two societies. There is no single board exam in cardiovascular imaging. Instead there are three (National Board of Echocardiography, Certification Board of Nuclear Cardiology, and Certification Board of Cardiovascular Computed Tomography). Each of these exams requires significant investments of time and money, and each requires recertification within 10 years. Additionally, physicians have the option of pursuing a societal fellowship within specific areas of imaging. Currently these are offered only by the ASE (FASE), ASNC (FASNC), and, to a limited extent, the SCCT (FSCCT), and each costs additional fees to apply. Were an academic cardiovascular imager to maintain annual societal memberships, pursue comprehensive board certification, and obtain fellowship in multiple areas, the total cost over the three-year period would be relatively prohibitive.
These data demonstrate the onerous logistical barriers to developing a career as a technically conversant, multiple-board- certified, multimodality academic cardiovascular imaging specialist. While it is a worthy and achievable goal, the current model is in evolution. The need for consolidation of board certification has been voiced2,3,6–8 and is currently a topic of discussion among members of the imaging societies and their associated boards. The anticipated goal is to have an ABIM-administered exam with optional modules that correspond to each of the four modalities. While there is certainly an argument that taking the current group of tests is unnecessary, they are clearly becoming the standard measure for physicians to obtain credentialing and receive appropriate reimbursement from insurance carriers. Most importantly, it is also considered appropriate for academic practice, where one is held to higher standards due to frequent educational contact with fellows-in-training.
One might believe, based on reading the material referenced above, that this expert’s existence is imminent, and that all future imaging specialists will have no choice but to have interchangeable skills across the board. If that is indeed the future, who will remain to be valid authorities in each individual modality? Will the amount of parallel technical knowledge and clinical time necessary to be active in all types of imaging result in subspecialists whose skills are fair in all areas and great in none? Or should the heterogeneity of imaging graduates remain, with some pursuing focused expertise in one specific area and others electing to become proficient in all?
In an informal survey of clinical members within the division of cardiovascular medicine at The Ohio State University (n=35 responders), 56% said they would prefer clinical imaging studies on their patients to be interpreted by faculty who practice in a single modality for three to four days per week rather than faculty with credentialing/ certification in multiple modalities, reading each one to two days per week. Even in our medical center, where all four modalities are performed and interpreted by cardiologists, there are clearly varying opinions.
In our single-center experience, Ohio State has had a year-long advanced imaging training program for four years, focusing on CMR and CCT and offering two positions per year. With the 2009–2010 entering class, the program has extended to include two years of training in all four modalities. Our training emphasis is balanced between patient care, research, and education—with moderate- to high-volume labs; tenured, funded, nationally recognized medical directors, and a relationship with the Department of Biomedical Engineering. Thus far, three of seven graduates are practicing true MMI in at least three modalities, and all are doing so at our institution. Within this small cohort, anecdotal evidence has shown nationwide availability of four-modality positions to be highly limited, primarily due to inconsistent penetrance of CMR in both academic and community practices. However, positions working in two (echocardiography and nuclear cardiology) and sometimes three (previous plus CCT) are relatively common.
Future Directions in Academic Multimodality Imaging (MMI)—Development of MMI Research Models
Perhaps most imperative amid the technological explosion and proliferation of cardiovascular imaging is the current lack of data demonstrating that diagnostic studies chosen and interpreted by experts in MMI lead to more cost-effective care and improved patient outcomes compared with the practice of single-modality experts. Technical advances and added applications are published monthly, but they have far outpaced our knowledge of how best to employ these new tools. The Appropriateness Criteria series of documents published by the ACC beginning in 2005 detail clinical instances when imaging is appropriate to use, but they themselves are primarily based on consensus opinion. Data such as those describing the potential risk of radiation exposure during CCT10 only heighten the need for erudite trials that define consistent quality parameters and patient benefit when utilizing imaging modalities to diagnose disease and make treatment decisions. These investigations are the ideal research niche for the academic MMI specialist, who differs from traditional faculty with expertise in one or sometimes two types of imaging.
Traditional academic imagers have a very sophisticated understanding of their particular modality and devote the majority of their clinical and research time to their specific area of focus. The academic comprehensive imaging specialist would likely be unable to master all four technical areas and would have to shift focus away from technical innovation and towards investigating innovative utilization of integrated strategies.
Approaches to validate the current MMI paradigm require organization around several specific metrics: ensuring quality and consistency among faculty with varying levels of expertise; tracking cost-effectiveness and/or patient benefit versus risk (radiation and contrast exposure, sedation, etc.) when comparing an ‘imaging triage’ approach utilizing an MMI specialist with non-MMI practitioners; demonstrating that integrated information accrued from multiple imaging studies leads to more informed clinical decisions; and, perhaps most importantly, demonstrating that utilization of cardiovascular imaging leads to improved short- and long-term patient outcomes. These investigations necessitate a comprehensive, multi-center effort, with infrastructure allowing data collection and long-term clinical follow-up, similar to the recent models published by the National Heart, Lung, and Blood Institute workshop in July 2008 that focused on outcomes research in cardiovascular imaging.11
Clearly, this is a mission requiring involvement from the entire academic cardiology community, but MMI specialists will play a crucial role. Thus, the importance of career development for young academic multimodality imagers requires an equally daunting, parallel effort. Barriers to their success must be addressed by thought leaders within the imaging community. The new multimodality Core Cardiology Training Symposium (COCATS) provides some guidelines, but there remains a lack of similarly trained mid- and senior-level faculty mentors to lead the way.