Cardiac Imaging: Radiologists move to protect MR and CT turf

Radiology's cardiac turf losses are legion: cardiac catheterization, cardiac angioplasty, echocardiography, and nuclear cardiology. Today, another turf battle is taking shape: control of cardiac CT and MR.

The outcome is by no means assured, and many radiologists have signaled by their words and actions that they intend to put up a vigorous fight. But so have cardiologists. MR and CT vendors, meanwhile, are ready to sell into a larger market that includes both groups of specialists.

In fact, clinicians other than cardiologists have been using CT and MR scanners for years, with and without radiologists reporting the final read. As the technology becomes even easier to use, it's reasonable to expect more nonradiologist involvement in CT and MR. In most cases, nonradiologists, including cardiologists, are drawn to CT and MR because of declining revenues due to either a change in the reimbursement scheme or the wholesale replacement of invasive procedures by CT and MR (see "Urologists eye CT scanners as reimbursement drops," p. 15).

Radiology has responded-with some success-to these encroachments by aggressively lobbying insurance carriers to enforce credentialing criteria and Congress to fix loopholes in the Stark II self-referral laws. But cardiology is a different animal, one to which radiology has traditionally lost turf battles. While many radiologists play down the impending showdown with cardiologists, others warn of a major conflagration.

"This will be the mother of all turf battles," said Dr. Jeff Jones, an interventional radiologist and director of cardiovascular imaging at Wilson Medical Center in North Carolina. "We thought iliac and renal stents was a big turf issue, but this is bigger."

Jones was part of a triumvirate of interventional radiologists who spoke at the Society of Interventional Radiology meeting in March. The lecturers explained why interventional radiologists are best qualified to perform cardiac MR and CT. After one presentation, an audience member quipped that perhaps the SIR should reinsert "cardiovascular" into its name. (Two years ago, the society shortened its name from the Society of Cardiovascular and Interventional Radiology in an effort to distance itself from cardiologists and become more aligned with diagnostic radiology.)

Dr. David Levin, emeritus radiology chair at Thomas Jefferson University and an expert on turf issues, sees the battle as a dire threat, arguing that cardiology control of CT and MR might be the linchpin that, if loosened or lost, could lead many other subspecialties to assume in-house control of CT and MR scanning and interpretation.

"The field is in real jeopardy if we lose cardiac CT and MR," Levin said. "If radiologists lose the heart, they could feasibly begin to lose MR and CT imaging to other subspecialties, including orthopedic surgeons, oncologists, neurologists, and urologists."

In the view of Dr. Stephan Koch, a radiologist and medical director for Imaging Heart in the New York metropolitan area, the battle may already be over.

"The wave of the future will be for cardiologists to have this technology in their offices," Koch said.

In Koch's world, cardiologists already have CT scanners in their offices because he brokers for them. He seeks out large multispecialty cardiology groups who practice within the legal scope of the Stark II self-referral laws and persuades them to buy CT scanners. Koch's group supplies the technologists and ensures that the scanning is held to the highest standards. In return, Koch interprets all extracardiac findings on CT angiography scans. All revenues from CTA are split 50/50, which Koch calls a win-win situation.

"I know I am not going to make a lot of friends in the radiology community, but because the cardiologists own all the patients, I would rather work with them and provide a high-quality service to them," he said.

Levin won't surrender so easily. He is currently setting up a cardiac CTA program at Jefferson with colleague Dr. Ethan Halpern. The 40 or so referrals to date have come from cardiology, cardiac surgery, and primary-care physicians. The future of radiologists' control of CTA depends on their working with internists and primary-care physicians, Levin said. These clinicians would rather not refer to cardiology and risk losing the patient. Radiologists must educate referring physicians about the benefits of CTA and of having radiologists do the imaging.

Dr. Elliot Fishman, director of diagnostic radiology and body CT at Johns Hopkins University, has witnessed the acquisition of CT and MR scanners by cardiologists in many centers and institutions. In his travels, he has asked cardiologists-in-training to name their field of interest. Noninvasive cardiac imaging is often the answer.

"They're not talking about echocardiography or ECG. They mean MR and CT," Fishman said.

At the premier meetings for cardiologists-the American Heart Association and the American College of Cardiology-the number of papers on cardiac CT and MR has burgeoned, according to Dr. Gerald Pohost, a cardiologist and director of cardiovascular research at the University of Southern California. The Society for Cardiovascular Magnetic Resonance, of which Pohost was a founding member, boasts a 60/40 membership ratio of cardiologists to radiologists. The SCMR meeting has become a showcase for cutting-edge cardiac MR research. The North American Society for Cardiac Imaging, a staunch advocate for radiologists, attracts at least 20% cardiologists at its annual meeting.

"There's no doubt about cardiology's enormous interest in CT and MR," Fishman said. "It's a matter of fact, not paranoia. And it's happening all over the country, not just in one or two areas."

RADIOLOGY'S RESPONSE

Radiologists seem to understand the gravity of the situation. The July issue of Radiology contained two special reviews on CT evaluation of the coronary arteries. Journal editor Dr. Anthony V. Proto prefaced each article with a note to the effect that this topic is so important it warrants two special reviews. Dr. Melvin Clouse, a professor of radiology at Beth Israel Deaconess Medical Center in Boston, said he can't ever remember the journal running back-to-back special reviews on the same topic.

Three years ago, the RSNA created a separate cardiac imaging section to promote cardiac CT and MR research. Abstracts in those fields have steadily increased, as has participation by radiologists at the sessions, said Clouse, who has chaired the RSNA cardiac imaging subcommittee for the last three years.

In 2003, there was an explosion of papers documenting the role of 16-slice CT in cardiac imaging, but a disproportionate number of them came from European researchers. U.S. radiologists have to do more publishing, Clouse said, because whoever controls the research inevitably controls the modality.

The American College of Radiology earlier this year reinvigorated its CT committee with a charge to determine where the college could best focus its energy. The answer: cardiac CT, according to Dr. Geoffrey Rubin, chief of cardiovascular imaging at Stanford University and a member of the newly formed ACR CT subcommittee. A preliminary plan by the subcommittee calls for ensuring the availability of appropriate cardiac CT educational venues for radiologists, making certain that new trainees and programs are integrating cardiac CT techniques appropriately, enticing radiologists and residents to enlist in cardiac imaging fellowships, and encouraging radiologists to work with cardiologists in this area.

CT and MR present substantial challenges to cardiologists, Rubin said. He describes echocardiography and catheter angiography as "flashlight" imaging, meaning that the image is narrowly focused. CT and MR imaging, on the other hand, are like turning on a fluorescent ceiling light where everything is immediately visible. Because CT and MR can visualize disease throughout the body, cardiologists will be challenged to maintain a certain standard of care and expertise.

In January 2003, the NASCI sent a position paper to the American Board of Radiology in support of integrating a cardiac CT and MR imaging section into the existing oral boards. If radiology did not act swiftly, there would be serious consequences to the entire field. The ABR listened and for the first time included a "virtual" cardiac exam in the June boards.

"This step was seen as essential in order to increase the radiology residents' interest in learning about the heart and residency programs' motivation in providing them with an appropriate education on the subject," said Dr. Kay Vydareny, head of the cardiopulmonary section of the ABR.

At last year's RSNA meeting, a small group of radiologists led by Dr. Dieter Enzmann, radiology chair at the University of California, Los Angeles, met to find a solution to the challenges of the impending turf battle. The Manhattan Project emerged. Essentially, it is a program dedicated to educating radiologists, through didactic and hands-on methods, about cardiac anatomy and advanced imaging. Its name, adopted from the crash U.S. program initiated in 1942 to build the first atomic bomb, is meant to signify the urgency of radiologists' involvement in cardiac CT and MR imaging. But it also signifies to some cardiologists an arrogant intent on the part of radiology to control the imaging.

"The name suggests an offensive strike, and I don't think it needs to be that way," Pohost said.

Like many in academia, Pohost has assembled clinical and research teams comprising both cardiologists and radiologists. The key, he said, is to recruit competent personnel who can work together. Enzmann agrees but counters that the number of competent cardiac radiologists is few, and interest in entering the field is low.

"The Manhattan Project is simply a means for radiology as a subspecialty to train itself in this new area. It has nothing to do with turf," he said.

The Manhattan Project may indeed boost cardiac CT and MR turf, but more so for cardiologists than for radiologists. Twelve institutions are listed by the Manhattan Project as offering hands-on cardiac CT and MR fellowships (www.scardonline.org). But rather than being bastions of radiological strength, some of these programs are being overrun by cardiologists.

The number of inquiries from cardiologists for the one-year fellowships at the University of Pennsylvania outnumbers those from radiologists 10:1 (only radiologists are accepted for one-year fellowships). For three-month fellowships at Penn, cardiologists again outnumber radiologists 2:1. The one-week courses are split more evenly. At Northwestern University, the ratio of cardiologists to radiologists has been 50:50. Both groups are studying both modalities, although cardiology favors CT because of the potential for coronary CTA, said program director Dr. James Carr. At the Cleveland Clinic, the ratio of cardiologists to radiologists has been 50:50, while at the University of Florida in Jacksonville, it is 2.5:1.

ACOUSTIC IMAGING

In a Sept. 15 editorial titled "Who Owns Images of the Heart?" Dr. Alfred A. Bove, editor-in-chief of CardioSource, an online publication of the ACC, argued that CT and MR are just two more imaging devices for cardiologists, who began centuries ago with acoustic images generated by the stethoscope. Anyone who has the proper training can essentially interpret CT and MR images, he said.

"Cardiologists have been interpreting images for years," Bove said during a phone interview. "For someone to say we can't do that with MR and CT doesn't make sense. If we gain the skills and are qualified and credentialed, we should be able to do so."

Bove said that he is not advocating sole ownership of cardiac MR and CT; he just does not want cardiologists to be excluded. But statistics tell a different story about exclusion. In a study reported at the 2003 RSNA meeting, Levin and colleagues at Jefferson found that although overall payments for noninvasive diagnostic imaging rose 71% from 1993 to 2001, cardiologists' share of those Medicare dollars increased almost twice as rapidly as that of radiologists. More important, payments to cardiologists in offices and private imaging centers grew at a rate of 240%.

"This trend calls for closer scrutiny of practices in private imaging centers and offices, especially when given the potential for self-referred studies by cardiologists," said Dr. Vijay Rao, radiology chair at Jefferson.

In the editorial, Bove referenced a recent opinion piece in The New York Times by Levin, who blamed the high cost of imaging on physicians' self-referral patterns, particularly those of cardiologists. Bove wrote that radiologists' concern about self-referral is likely to be irrelevant as the ACC develops appropriateness criteria for utilization of imaging procedures. Levin said it's ridiculous to believe that self-referral problems will disappear if cardiologists set guidelines for scanning.

"That's a joke. They order scans for everyone who walks in the door," he said.

While some radiologists may be more vocal than others about cardiologists' voracious appetite for patients, the majority of leaders in radiology-and cardiology-espouse collaboration. Two facts fuel the desire to collaborate: Radiologists don't own the patients, and cardiologists, generally, don't own the scanners. Radiologists' control of imaging modalities gives them an excellent opportunity to establish a strong foothold in cardiac imaging, particularly if it is bolstered by research, education, and accreditation. Many cardiology groups cannot afford to spend upward of $2 million for a scanner that will sit idle much of the time. In some states, certificate of need (CON) conditions further restrict cardiologists from owning scanners.

But the above advantages for radiologists have caveats. As cardiologists see their salaries negatively affected by advanced CT and MR imaging, groups are beginning to pool their resources to buy scanners. They then fill up the scanner's time with peripheral imaging. In some cases, cardiologists can bypass CON restrictions by purchasing a scanner for research and then quietly slipping into clinical scanning. The 16-, 32-, and 64-slice CT scanners are rapidly changing the way cardiologists think. Compared with MR scanners, these machines are relatively inexpensive, involve minimal setup and operating costs, and can challenge MR for some applications. Philips Medical Systems recently introduced the Brilliance CT Private Practice CV system, a 16-slice scanner restricted to the analysis of cardiac and peripheral vasculature that cannot be used to perform radiological exams. More important, it is priced under $700,000. Philips expects most buyers to be cardiologists.

The Brilliance CT will not necessarily sound the death knell for radiologists' involvement in cardiac CT, however. Radiologists will still have to be involved in interpreting extracardiac findings because the images inevitably will have some piece of the lung, aorta, or pulmonary arteries in it, Koch said. Hearts come in many different configurations, and these machines are not smart enough to localize only the heart.

ACC guidelines will eventually require that noncardiologists have to read the outside portion of the study, Fishman said. A cheap 16-slice scanner isn't as worrisome to Fishman as an inexpensive 64-slice machine. Without even trying, a 64 slice scanner blows away 16 slices.

"If 64 slices ever become affordable to cardiologists, it will kill us," he said.

The effort to keep cardiology at bay may ultimately rely on how effective the ACR is at persuading Congress to plug loopholes in the Stark II self-referral laws and how effective radiology is at enforcing a standard of care based on accreditation and credentialing. Leaders in the field are also reorganizing their radiology departments to strengthen and maintain the hold on cardiac CT and MR imaging (see accompanying article). Hope is being pinned on individual radiologists becoming involved in cardiac imaging. Many radiologists express optimism about the future, while others are already waving white flags.

"Even when radiologists are aggressive, it doesn't mean they will win," Fishman said. "If I were in cardiologists' shoes, I'd be doing the same thing."

On the flip side, Levin has found surprising support and enthusiasm for cardiac imaging.

"I thought radiologists were just going to roll over and play dead like we did with catheter angiography," he said. "But that's not what I'm finding. They want to do this."

The short supply of radiologists gives them extra bargaining power within their hospitals to set the standard for cardiac imaging, Levin said. Collaboration is the ideal, but if that isn't possible, radiology will compete, and compete very nicely.

"Bring 'em on," he said.

--

Top 10 indications for cardiac CT*

1. Cardiac calcification

2. Coronary calcification

3. Anomalous coronary arteries

4. Ruling out coronary stenosis

5. Bypass graft patency/location

6. Congenital heart disease

7. Cardiac masses

8. Pulmonary veins

9. Cardiac veins

10. Plaque imaging

*According to Dr. Stephan Achenbach, assistant professor of medicine at the University of Erlangen, Germany

Top 10 indications for cardiac MR*

1. LV function/mass/volumes/stress MR

2. Viability

3. Valvular heart disease

4. Right ventricular anatomy, function

5. Congenital heart disease, anomalous coronaries

6. Coronary artery integrity

7. Characterization of cardiomyopathies

8. Subclinical atherosclerosis

9. Pericardial disease

10. Pulmonary vein anatomy postablation

*According to Dr. Warren J. Manning, section chief of noninvasive cardiac imaging at Beth Israel Deaconess Medical Center, Boston