Not long ago, invasive diagnostic catheterizations to search for arterial blockages and disease were the interventional radiologist’s stock-in-trade. In some locales, that may still be the case. The gold standard for the diagnosis of vascular stenosis and other ills, however, has shifted to the noninvasive side. CT angiography and MR angiography have become so capable of defining vascular structures that external scans can put the diseases on view without having to snake a catheter through the patient’s body. Clearly, patients have benefited. So have IRs—maybe.

Coronal MIP images from four-station whole-body contrast-enhanced MR angiography at 3T acquired using a Trio from Siemens Medical Solutions with TimTM. Images courtesy of Kambiz Nael, MD, University of California, Los Angeles. (Click the image for a larger version.)

“I don’t know the number, but I think back in my fellowship days in the 1980s, the proportion of angiography among vascular radiologists for diagnosis-only was probably 80%. Now it has just about flip-flopped. Now we do angiograms to intervene 80% of the time, out of all our vascular work,” says J. Bayne Selby, Jr, MD, FSIR, a professor of radiology in the division of vascular and interventional radiology at the Medical University of South Carolina (MUSC) in Charleston.

“The carotid angiogram has pretty much ceased to exist,” he adds, “and the rest of the body is starting to go that way too. In our practice, if you need a run-off arteriogram, we still do the conventional angiography on the legs, but other practices do that all by CTA or MRA, so the lower leg run-off is going the way of the dinosaurs too.”

The deployment of the latest-generation CT and MR scanners has allowed IRs to do what many say they like best, to focus their practices on interventional procedures such as stenting while the vascular diagnostic work is handled through the noninvasive interpretation of images.

The new scanners have made “a huge, huge impact,” says Barry Stein, MD, FSIR, division of vascular and interventional radiology and chief of cardiovascular MRI and CT at Hartford Hospital, the largest tertiary care center in Connecticut, and Jefferson Radiology.

Whole body CTA, courtesy of Osman Ratib, MD. (Click the image for a larger version.)

“I’m no longer just a technician in the interventional suite,” Stein says. “I’m a clinician, and I can offer a full range of services at the highest level. I’m very passionate about it. This is a massive opportunity for radiologists to distinguish themselves from other practitioners.”

Robert A. Lookstein, MD, is assistant professor of radiology in the division of interventional radiology at Mt. Sinai Medical Center in New York City. He calls CTA and MRA “state of the art in the diagnosis of peripheral arterial disease.”

IRs who do not embrace the imaging will be “pushed out of the diagnosis of peripheral vascular disease,” he adds. “It really is a dichotomy. Most people would argue that diagnostic invasive angiography is an obsolete technology.”

To indicate just how obsolete, Lookstein cites this example: “I was just at a major endovascular meeting,” he says. “They performed over four days 14 or 15 live teaching cases. Every single case had imaging that had made the diagnosis before the patient came in to be treated.”

Lookstein is upbeat about the impact of CTA and MRA on IR practices.

“We’re much busier now than in the invasive days,” he says. “On average, one third to one fourth of our time back then was dedicated to [invasive] diagnostic studies. Now that entire time during the day is open to treat more patients. We can treat more patients using invasive techniques and diagnose more patients using noninvasive techniques.The patient gets the therapy clearly mapped out ahead of time, and that reduces the time for the [treatment] procedure and lowers the risk.”

Lookstein is also upbeat about the IR practice revenue that can be generated using CTA and MRA.

“There are studies that are reimbursed that can be performed very rapidly, and there’s clearly a need for them,” he says. “If physicians are very motivated and want to build a practice based on cardiovascular imaging, they can grow the revenue for their practice.”


Detailing just which invasive diagnostic procedures have been eclipsed by CTA and MRA is a task that most IRs interviewed for this story balked at, preferring to note that they included just about any invasive diagnostic catheterization. Hartford Hospital’s Stein did e-mail a list that he had prepared previously, which in reality includes diagnostic interrogation of all the vascular territories for nearly all common pathologies. It included extracranial carotid artery scans; renal artery scans for stenosis and the evaluation of bypass grafts, as well as renal donor workups; mesenteric evaluations; aorta and peripheral run-off tests, including abdominal aortic aneurysm and tests for limb-threatening ischemia; and a number of other aortic and vascular conditions.

The point not to be missed is that the catheterization procedures themselves are not falling by the wayside. But they are becoming exploratory rather than diagnostic, and there are fewer of them. As Selby and others explain, the typical catheterization now would be done subsequent to the CTA or MRA, and it would occur as part of stenting or other treatment. “The intervention is an extension of the arteriogram,” Selby says. “We would do the arteriogram to make sure we’re in the right place.”

Patients whose scans turned out clean on CTA or MRA would never be catheterized, as they would have been previously. Patients whose scans showed the need for surgery would avoid catheterization too and would go straight to the operating room.

Norbert Wilke, MD, FACC, is an associate professor of radiology and chief of cardiovascular MR and CT at the University of Florida Health Science Center in Jacksonville. Wilke makes the point that CTA and MRA scans not only help IRs plan interventions, they help surgeons plan surgeries. “We are in the midst of changing the paradigm to cross-sectional imaging from conventional x-ray,” he says.

He also notes that many patients are still undergoing diagnostic catheterizations. “There are so many patients that some are still getting x-rays to be timely,” he says. “If we can’t get them to the MR fast enough when there is an overload—especially if you don’t operate dedicated cardiovascular scanners—many patients still end up in the cath lab.”


Wilke also notes that the use of CTA possibly imposes a radiation risk to the patient that ought to give IRs pause about selecting CTA over MRA. Wilke and others also say that CTA contrast agent can pose a risk to kidney patients, whereas the gadolinium agent used with an MRA does not.

“Elective patients should consider what the radiation dose is,” Wilke says, “especially if you have to do later x-rays during stenting or mammographic x-ray screening studies in women. The MR study would be much more favorable to save unnecessary radiation and iodine contrast load.”

Wilke says more study is needed to pin down the exact radiation risk of CT using different protocols, but he adds that “of patients receiving 10 millisieverts [of radiation exposure] per year, one out of 1,000 may develop cancer, based on recent Japanese data presented.” He adds that the radiation dose from a single CT scan is hard to determine “but it’s in the range of 10 to 20 millisieverts, depending on the different vendors offering the latest 64 detector CT scanners currently.” Wilke says that women under 50 should not undergo CTA if MRA can be used, and he adds that for any younger patients, MRA is the preferred method because of the potential radiation risk caused by extensive CT imaging. The future generations of so-called “dual-source” CT scanners as recently introduced at the RSNA 2005 will substantially reduce the CT radiation dose.

But Wilke also notes that the newer MRA machines capable of the most detailed vascular imaging are less widely deployed than CTA scanners, and he says that using them compared with CTA has more difficulty and needs more investment and staff training in order to establish a viable service line.

Jeffrey E. Jones, MD, is managing senior partner at Carolina Radiology Consultants (CRC), a private practice group whose clients include a 350-bed hospital in Wilson, NC. Jones says CRC has made MRA the “gold standard” for its vascular imaging because renal patients can handle the gadolinium and because CTA is often “challenging” to interpret if calcified plaque is involved. But Jones says CTA or MRA scans are used to diagnose peripheral vascular disease (PVD) in virtually all patients. “Nobody gets a diagnostic catheterization for peripheral or renal,” he says, “except in rare circumstances where a surgeon may request a diagnostic catheterization arteriogram.”

The point should be made that CTA is not used at the expense of MRA or vice versa. The scans are not mutually exclusive. As Brian K. Herman, MD, medical director at the Eisenhower Imaging Center in Rancho Mirage, Calif, stresses, “CTA and MRA are complementary. CTA is good in the absence of calcium; in MRA calcium plays no role.”


According to Jones, Stein, and others, IR specialists are following a well-accepted protocol for PVD, which usually begins with duplex ultrasound imaging of carotids and/or peripheral run-off vessels, in conjunction with an ankle-brachial index test. If no evidence of disease shows up, then the patient is sent home without further testing. If the patient tests positive, then CTA or MRA is done to elucidate the exact location and extent of the disease to facilitate appropriate triage and planning of any therapeutic intervention.

According to Texas interventional radiologist Gregory Karnaze, MD, that is the rough protocol followed at Austin Radiological Association (ARA), where Karnaze is president. “The patients come in and we set up a consult,” he says. “All our information is on PACS (picture archiving and communications system), so we can bring it up and show it to the patient in the interview room, and let them know what the options and treatments are.”


With streamlined vascular screening possible through CTA and MRA, the business thrust at many IR clinics has been to promote the screens to referring physicians and then try to capture any ensuing interventional business when a referred patient screens positive for disease.

This sounds like a rosy picture for IR—quicker, safer diagnostic screening; more patients screened, therefore more positive patients; more positive patients, therefore more interventional work. For some clinics—as Lookstein suggests at the outset of this story—this pattern may lead to prosperity. For others, however, one step does not necessarily lead to the next.

The Medical University of South Carolina’s Selby agrees that IRs logically should be seeing more patients because of CTA and MRA.

“Diagnostic angiography always had a small but real risk, which meant that the primary care guy might not send the [asymptomatic] patient,” Selby says, “but now he shouldn’t hesitate to send the patient for CTA or MRA.”

But so far, Selby says, the increase in patient referrals has not happened at his practice. “Even with rampant diabetes and renal failure, we’re not seeing enough patients. We need to make sure the primary care people understand this new game in town.”

Karnaze at ARA says the impact of CTA and MRA as a revenue generator is still unclear. “IR is such a changing field, I’m not sure that can be answered,” he says. “I don’t think it’s been a negative revenue issue. In the past the catheterized arteriogram would have been done in the hospital, so they got that technical component, so overall I would say the impact has probably been fairly neutral.”


Another factor that determines how much IRs will profit from the new diagnostic procedures is the extent to which they face competition from other specialists, like cardiologists and vascular surgeons. Cardiologists especially have been bringing CTA into their practices, doing the diagnostics that would previously have gone to IRs. This competition is occurring not just between clinics but between specialists at the same hospital.

At the Eisenhower Medical Center in Rancho Mirage, says Brian Herman, the IR practice has lost out to in-house cardiologists and vascular surgeons who have usurped CTA and MRA diagnostics for themselves.

“If we didn’t have competition on the IR side, the MRA and CTA would be excellent complements to what we do. But instead, the patients are being imaged elsewhere and treated elsewhere,” he says.

Herman says it is pointless for radiologists to compete head to head with cardiologists in a hospital setting because “cardiologists have too much power.” They bring too much revenue into hospitals to be challengeable in competition, he explains.

Instead of directly competing, Herman advises radiologists to focus on neurological and stroke interventions to showcase their own skills and bring themselves power that way.

Although Herman says his department at Eisenhower is “as strong in terms of status and political power as radiology has been in three decades,” he is still concerned enough about encroachment from other specialists to voice this caveat:

“When you talk about hospitals supporting the department of radiology, or any department of the hospital, if you don’t have a long-sighted vision of supporting those attributes against the inevitable erosion to cardiologists or other entities, it will be an inevitable deterioration of the quality of those services. If you simply roll over—and over and over—to the political push of those entities that are just looking for more dollars, you eventually have eroded the department of your hospital, in this case radiology, but also the field of radiology in general across the country.”

Even within radiology, competition between body imagers and IRs to interpret CTA and MRA diagnostic scans can have disruptive effects for IRs. If body imagers are reporting diagnostic findings to referring physicians without giving a heads-up to their IR clinicians, treatments that might have gone to IRs may be lost.

Selby at MUSC says that in the old days of diagnostic catheterization it was “pretty hard” for a referring physician to say no to an IR seeking to do an intervention when the patient was “right there on the table.” But CTA and MRA diagnostic reports now going to referrers prior to intervention have empowered referring doctors to make other choices about who the patient is sent to, Selby adds. The IR doctors may lose out to competitors. But the pendulum swings both ways. Referring physicians could send work to IRs that competing specialists want. “This is something that has the cardiologists worried with coronary artery CTA. They still want to be the gatekeepers,” Selby says.


Despite the risks from competing physicians, most IRs interviewed for this story say CTA and MRA have strengthened radiology’s position.

“This is a natural for radiologists,” says Curtis W. Bakal, MD, MPH, professor of radiology and vice chairman of radiology at the University of Medicine and Dentistry of New Jersey in Newark. “This is a huge opportunity for radiology.”

Bakal argues that in the days of invasive diagnostic catheterization, primary care doctors might have hesitated to send patients they were unsure about because of the risks associated with invasion. With CTA and MRA, there isn’t that hesitation. “This is a terrific way to screen patients and keep them in radiology,” he says. “You can pick the patients up early and maintain the practice. When we screened renal artery stenosis, it went up in incidence, and we were able to retain much of that business. This is a great way to screen.”

IRs argue that as trained radiologists they are in a much better position to interpret images than nonradiologist subspecialists. A key element of this may be the post-processing of complex CT scans.

“The technology isn’t something you can learn in a weekend,” says Hartford Hospital’s Barry Stein. “I can do all the post-processing and bring it up right there next to me in the interventional radiology suite. I can figure out what the best orientation is to see a blood vessel more advantageously. It’s important for IRs to become versatile [in post-processing] and not run away from it. That gives us the advantage.

“You always want to be moving the goal posts as the competition tries to catch you,” Stein adds. “Our radiology background is very strong, and we should embrace the new technologies much more than the competition. My only plea is that most radiology groups allow interventional radiologists to become involved in this so they become the diagnostic and therapeutic vascular specialists of choice. This is a great opportunity for radiology and for interventional radiology to make a significant impact on the management of vascular disease with this powerful technology. If the right people with the right support in radiology embrace this, for radiology this will be major.”

But just how successful interventional radiologists will be at defining the vascular imaging territory as their own remains to be seen. While some clinics report glowing financial results, others have not seen the inflow of patients they expected. The overall picture is mixed.

Internecine Turf Issue: Whose Scan Is This Anyway?

When a patient completes a CTA or MRA diagnostic scan, who should interpret it? Should it be interventional radiologists who hope to do follow-up interventional treatments, or should it be body imagers who may feel they are better prepared to interpret CTA and MRA since they are the ones to whom other CT and MR scans are routinely routed?

In some places this in-house struggle between radiologists over who interprets CTA and MRA diagnostic scans has become a real sore point.

“In my own experience, body imagers want to do this and learn about it, and IRs have been too busy to object-and that’s a mistake,” says Curtis W. Bakal, MD, MPH, an interventional radiologist and vice chair of the radiology department at the University of Medicine and Dentistry of New Jersey. “It’s a mistake to let the body imagers do this. CTA and MRA are the interventional radiologist’s tools.”

CTA and MRA are a key technology “that IRs should be taking control of,” he adds, “because it’s a way to get into patient care early in the care chain.”

Bakal says he plans to “reallocate responsibilities” in his department “very shortly” to reflect this view, although he admits that he does not know yet exactly how he is going to do it.

At the Medical University of South Carolina, J. Bayne Selby, MD, calls the conflict between body imagers and IRs “the elephant in the room that nobody talks about.”

The body imagers want to learn vascular interpretation, he says, but the interventionalists are afraid they will be cut out of the treatment loop if the body imagers simply report back their findings to referring physicians.

“Basically, I think it is going to be the old ‘all politics is local,'” Selby says, meaning that different clinics and institutions will come up with their own solutions. “But I do think 95% of IRs would say they’re more qualified to interpret blood vessels on a CTA. Body imagers would say they are in a better position to issue the report.”

Who reads, and for what reasons, can get complicated simply because of the way technology is deployed too.

Steven J. Citron, MD, is chief of interventional radiology at Piedmont Medical Center in Atlanta, where there are, he says, about six IRs and about 13 body imagers. It is the body imagers who are interpreting the CTA and MRA scans.

“In our particular practice, the body imagers have asked to own these scans. In the interest of group harmony, the IR section has acquiesced,” Citron says. “We initially said we wanted CT and MR angiography as a joint venture between body imaging and interventional radiology. We said please call us and we’ll go over the salient features-but the fact of the matter is, those calls never came.”

Citron says the workstations where the reconstructions are done are in the body imaging section and not the interventional radiology section. But the decision for body imagers to read the scans was more than that, he adds. “It was also a sense of ownership with them,” he says. “They thought they knew the pulse sequences better, but we thought we understood the anatomy better. We thought there needed to be a good joint effort.”

Citron says now the IR doctors do not see a list of the CTA and MRA reports completed by the body imagers and therefore do not know what interventions they may be losing out on. “We are dependent on referring physicians for our cases and not on the body imagers. It’s a big issue,” Citron says. “Certainly, there is a profit motive, but despite e-mails and discussions in group meetings, it seems to have fallen on deaf ears. To be fair, the body imagers may feel overburdened with lots of work and not have the time or take the time to let us know…. We could set up a service line and follow up, but we haven’t done that heretofore.”

In the meantime, Citron adds, “the outside arterial business has fallen off because a lot of it is being done by the cardiologists.” He says Piedmont Medical is looking into the development of a vascular treatment center that would involve all vascular specialists and “hopefully raise the tide for all boats.”

“I’m putting my foot on the accelerator because I want to do more vascular business,” Citron concludes.

It is a goal shared by lots of IRs who have the diagnostic tools of CTA and MRA but may not be the ones asked to interpret the studies.

G. Wiley

George Wiley is a contributing writer for Decisions in Axis Imaging News.