Radiology and the Heart
Cardiology and radiology: Are they two specialties working in tandem for optimal patient care or two opposing armies in a turf battle? The answer, of course, is complicated, and can’t be approached without an acknowledgement of the ground already ceded to cardiology. In today’s hospitals, cardiac ultrasound, cardiac catheterization, and (in many cases) nuclear cardiology are all the purview of the cardiologist. Newer, noninvasive techniques such as CT angiography (CTA) and MR angiography provide a fresh proving ground for each of the two specialties to demonstrate their unique worth—while they are under pressure to care for patients more efficiently and cost effectively than ever. As cardiologist Guy Weigold, MD, director of cardiac CT at Washington Hospital Center, Washington, DC, points out, “You take an image of the heart, and put it in front of a cardiologist and a radiologist, and those two people are going to see very different things. The cardiologist is going to see the disease and innately understand the clinical implications, but would have a hard time appreciating the subtleties of the artifacts. The radiologist, on the other hand, might have a better understanding of the artifacts and other issues related to image quality, but doesn’t innately understand the clinical implications.” Models for collaboration between cardiology and radiology vary widely. At Washington Hospital Center, for instance, the bulk of cardiac imaging is handled by cardiologists, although cardiac CT and MRI are still shared. Cardiologists do an initial interpretation for disease, and radiologists perform a second interpretation with the intention of identifying any incidental findings and issuing necessary follow-up recommendations. By contrast, at the University of Maryland Medical Center (UMMC) in Baltimore, radiologists still perform most cardiac imaging, according to Charles White, MD, chief of thoracic radiology. “We’re all the way on the side of the spectrum where radiology really does the bulk of the CT and MRI,” he says. “We do it all. We’re happy to have the cardiologists’ participation, but the issue has been that they are involved in so many other things that there hasn’t been anyone with the time or bandwidth to be heavily involved from their side. We value their clinical expertise, but the protocol, patient work-up, and decision making of the imaging piece are all performed within radiology.” Natural Evolution This model evolved, White says, from the fact that the radiology department at UMMC was an early developer of a strong cardiac MRI program. When cardiac CT became increasingly prevalent, it made sense for radiology to continue superintending its use. “It was not much of a stretch for us to expand into cardiac CT,” White recalls. “A turf battle is more likely to occur when you have a vacuum—when there wasn’t a cardiac CT or MRI program before, it was new ground, and so it became a new negotiation. A lot depends on referral patterns and on whether the services and relationships that already exist are good.” Ricardo Cury, MD, director of cardiac imaging at Baptist Hospital of Miami in Florida and of Baptist Cardiac and Vascular Institute, concurs. “Of our cardiac CT, 40% to 50% comes from the emergency department, and radiology already provides emergency-department work for CT imaging, so it makes sense for radiology to do cardiac CT for the emergency department as well,” he says. He adds, “We also have a history, here at Baptist Hospital of Miami, of over 15 years of providing very good service for nuclear cardiology due to the efforts of Jack Ziffer, MD. That helped, over time, to maintain cardiac CT and cardiac MRI. Obviously, building that confidence from other clinicians is a process that takes time.” James Earls, MD, a radiologist with Fairfax Radiological Consultants in Virginia, notes that in the case of his group, its primary hospital client tasked it with developing a program that would be shared between cardiology and radiology. “When we went out and purchased the cardiac CT scanner for the hospital, it said we would share it equally,” he says. “Because we have an established outpatient cardiac CT program (we’ve been doing it since 2002), the cardiologists had a lot of respect for the radiologists’ experience doing it even before we opened the joint program at the hospital.” He adds, however, that Virginia happens to be a very restrictive state when it comes to issuing certificates of need; otherwise, most of the cardiology practices probably would have tried to get their own scanners in the past. What of scenarios wherein the natural evolution of roles has favored cardiology, not radiology? “Cardiac CT and MRI are relatively new fields,” Weigold notes. “The bulk of the work being done on any given CT or MRI system in the hospital is, far and away, not cardiac in nature. Here, I think the radiologists feel they have tons to do already; they’ve never felt the need to try to wrestle it away from the cardiologists. At the end of the day, a lot comes down to the economics of things—the practicalities.” The Efficiency Factor In many cases, in fact, it would appear that the turf battles of a happier, wealthier time in health care have all but evaporated. In today’s hospitals, efficiency is the watchword when it comes to caring for cardiac patients. “Everyone’s busy, and there’s a lot of work to be done,” Weigold says. “It’s not efficient to have two physicians working on the same task together. That’s the major hurdle: How do you make it efficient?” In fact, Washington Hospital Center’s collaborative cardiac program began with Weigold and a representative from the radiology department literally reading each case shoulder to shoulder—an arrangement that enabled each to learn from the other, but that proved to be cumbersome over time. “Up until recently, the radiologists were occasionally doing the initial reading of the study, but the radiologists have become so busy with their own CT work that they don’t anymore,” he says. “Right now, we have three cardiologists reading scans each day, and the radiologists look at the scans separately for incidental findings. The final report is a collaboration—they’re still with us, but in a virtual way.” In a similar way, White observes, the cardiologists at UMMC have plenty on their collective plate without dipping a toe into cardiac CT and MRI. “I don’t feel that the cardiologists want to be doing more,” he notes. “They’re busy in many other aspects of cardiac imaging, including echocardiography, catheterization, and transesophageal echocardiography, so they have a nice amount to do. We’ve had a very amicable relationship with them over the years.” At Fairfax Radiological Consultants’ hospital, where cardiac imaging is shared as evenly as possible between the two specialties, a reading panel is split between cardiologists and radiologists who alternate half days. “We obtain the study and interpret it, and we dictate the results into our PACS, where they’re accessible for the cardiologists,” Earls says. “Both sides read their studies in a timely fashion—the morning shift, by 1 pm, and the afternoon shift, by 5 pm.” He adds, however, that it would be much more difficult for a general radiologist to maintain this pace. “Several of our radiologists are cardiac specific, and that’s very helpful,” he says. “The fact that we’re subspecialized makes a big difference.” At Baptist Hospital of Miami, the imperative for increased efficiency has been a boon to the radiology department. “Radiology brings a lot to the table because you have one physician reading the whole image, as opposed to a model wherein two physicians read one study,” he says. “That would be cumbersome—and even more cumbersome if you’re providing the service beyond regular hours.” This, of course, raises the issue of the detection of incidental findings outside the heart. While Weigold and Cury both take it as an assumption that incidental findings by radiologists should be a consideration in parceling out cardiac imaging, Earls is operating in a hospital that no longer pays for the second interpretation. “My group said we wouldn’t do it anymore,” he says. At the 2010 annual meeting of the RSNA in Chicago, Illinois, Earls presented “Cardiac Imaging: Should You Evaluate or Ignore the Extracardiac Structures?” on December 1, as part of a controversy session on this topic. The approach he advocates for those facing similar financial sanctions is one of informal, collegial collaboration. “In theory, as radiologists, we’re responsible for everything on the film, and a small (but significant) number of people will have clinically important extracardiac findings,” he says, “but that doesn’t mean cardiologists can’t be trained to find them. Often, they’ll call us and say, ‘There’s something in the lung here,’ and we’ll take a look. We don’t get paid for it, but we’ll informally tell them, and we’ll recommend an additional study as necessary, depending on the finding.” Business Models Money, as Earls’ experience indicates, is a critical factor in how collaboration between cardiology and radiology develops. “There are places where the cardiologist reads the heart and the radiologist reads the rest, which to me is less than optimal,” White says. “I’d like radiology to have a piece of the heart. You can always have one of each on a study, but the problem is that Medicare requires one primary name on a study.” At UMMC, where radiology has what White characterizes as a huge component of cardiac imaging under its purview, “Our business model is radiology predominant,” he says. “The business piece of it, the reimbursement, all goes to radiology. Cardiology does not share in that.” Fairfax Radiological Consultants’ cardiac program at its hospital solves this problem by splitting interpretations as evenly as possible between cardiology and radiology, with each side acting as an informal consultant to the other as needed. Even this simple arrangement is the result of some negotiation on the part of both specialties, Earls notes. “There was a loophole when we set up the nuclear-cardiology program,” he says. “They were alternating a cardiology reader with a radiology reader, but if a patient was referred from a cardiologist’s practice, he or she could still read those cases, allowing the majority of studies to be read by someone who wasn’t the assigned reader.” When it was time to establish the hospital’s cardiac CT program, Earls was alert to the possibility of this loophole recurring: “I put some things in place to make sure we’d really be doing it 50–50,” he says. At Baptist Hospital of Miami, which specialty handles a patient’s interpretation is dependent on a number of factors, including the clinical indication and the source of the referral. “Here, cardiology is mainly doing echocardiography. Nuclear cardiology is shared (nuclear radiologists perform the interpretations and cardiologists perform the stress tests), and radiologists trained in cardiac imaging are doing cardiac MRI and CT,” Cury says. He attributes radiology’s success in these two modalities to its provision of CT interpretation for the emergency department (and to high-quality service). “We worked out a very good model on patients presenting with chest pain to the emergency department; if their risk score on the scale for thrombolysis in myocardial infarction, or TIMI, is 0, 1, or 2, we do coronary CTA (CCTA), and if it’s more than 2, we do myocardial-perfusion imaging with SPECT,” he says. “We have our own data showing that CTA really decreases the length of stay in the hospital for patients with chest pain, and it is very accurate. There’s a unique opportunity for radiology in getting involved with this type of imaging in the emergency department, and a lot of data supporting the use of CCTA in that setting.” At Washington Hospital Center, where cardiology controls most cardiac imaging (with the exception of overreading by radiologists), Weigold notes, turf battles have been avoided largely as a result of a strong, respectful relationship between the two specialties. “It was naturally fostered because we were under the same roof, which is a little unique,” he says. “It may be the reason the paradigm is not replicated much in the community.” Serving Each Other Both sides agree that providing the optimal level of service to each other enhances collaboration—and, of course, improves patient care. “Level of service is fundamental,” Cury says. “With our emergency-department interpretations, every case is read in less than 60 minutes. This is not easy to achieve—by the time the scan is done, you have a lot of images to process and reconstruct, but it improves our interaction with cardiology and emergency-department physicians by providing that good service. We call and discuss the results with them, and we’re also available to discuss which tests would be best for the specific clinical situation.” White adds that providing good service is one way that radiologists can maintain their place at the cardiac table. “If you’re not giving anyone else a reason to home in on the business, it lessens the chance that will happen,” he says. “A good service model is key, and you have to be willing to talk. If there’s a desire for cardiologists to get involved in imaging, it’s not realistic or wise to shut the door. Cardiologists have the patient—that’s a big advantage, and it would be silly to alienate them. My approach is to provide good service and keep an open door.” Earls agrees. “I try to attend a lot of the multidisciplinary meetings where surgeons and cardiologists get together—being there, being able to speak the lingo, ensures you still have a place in the imaging,” he says. “A lot of people just sit up and demand it, whereas we’ve been demonstrating that we do a good job. The cardiologists respect us—they’ll give us a study and ask our opinion. We’ve had some turf issues over the years, but we treat each other respectfully.” Cury also notes that education is critical. “It’s important for radiology to have expertise in cardiac imaging and sufficient training to provide the service at a high level, as well as to work with the cardiologists and emergency-department physicians to develop protocols and try to select the best test for the best patient,” he says. “I’ve also seen many cardiologists who have a lot of experience with CCTA, and over time, they can identify major pathologies outside the heart. This shouldn’t be an argument for radiology to have a place in cardiac imaging: The most important argument is for a high level of training and service to each other.” Weigold concludes that the success of collaborations between cardiology and radiology will largely be dependent on the attitudes of the individual clinicians involved. “It has a lot do to with personalities and their tolerance, willingness, and interest in getting this collaboration going,” he says. “Here, we quickly found out that we could learn a lot from each other.” Cat Vasko is associate editor of Radiology Business Journal. Additional Reading — Technical Advances: Radiation-sparing Techniques and Automated AV Tools