My practice has changed dramatically in the last 15 years,” says Dennis Kay, MD, associate chairman, Department of Radiology, at the Ochsner Clinic in New Orleans. “Back then, an average day would be three to five vascular cases, usually aortic runoff and carotid studies with an occasional venogram. We are now approaching 20 vascular cases per day. Many of our runoff studies are done by MRA [magnetic resonance angiography], nearly all carotid studies are done by ultrasound, and we seldom do venograms.

“If you look at the older radiology textbooks, you see a large number of angiograms of lesions such as hepatic hemangiomas or focal nodular hyperplasia,” Kay remarks. “Today, those angiograms have been replaced by CT, MR, nuclear medicine, and ultrasonography.”

Kay’s observation encapsulates what is happening in vascular imaging. Increasingly, the classic contrast angiogram is being displaced. For many indications, there are imaging options, and part of the radiologist’s job may be educating the referring physician on the merits of the various studies for particular situations.

Arrival of Ultrasonography

The first modality to make a significant inroad in vascular imaging was ultrasonography. For the extremities, duplex ultrasonography and color Doppler imaging are now the first choices in most situations. In the leg, ultrasonography has a sensitivity of greater than 80% and a specificity of more than 90% in predicting the severity of an arterial obstruction.1

The carotid vessels are another site in which ultrasonography has displaced contrast angiography. When properly acquired, the images are so accurate that many surgeons will not ask for an additional imaging study before performing endarterectomy. Because of this success, duplex screening of asymptomatic persons for significant carotid stenoses is gaining adherents. The best technique is not clear, however.

Edward I. Bluth, MD, chairman of the Department of Radiology at the Ochsner Clinic, and colleagues at other medical centers recently conducted a blinded comparison of duplex and power Doppler ultrasonography in 120 asymptomatic patients referred for carotid artery screening.2 The power Doppler study had a sensitivity of 70% and a specificity of 91%. These investigators calculated the cost-effectiveness of carotid ultrasonography with endarterectomy as indicated to be $47,000 per quality-adjusted life year and noted that the area under the receiver operating characteristic curve, 0.87, is similar to that of screening mammography.

“We are in the same situation with vascular imaging that we were with mammography 20 years ago,” according to Bluth. “Then, mammography was done strictly for diagnosis; screening was not considered an indication, and insurance would not pay for mammography done for that purpose. Today, of course, we know that screening mammography saves lives. Screening carotid ultrasonography can identify and quantify disease accurately, but insurance does not cover it, and no national organization is recommending it. To change this situation, we need long-term studies that will show that mass screening improves the quality of life.”

Endorsement of screening carotid ultrasonography by a national organization may not be far away. At the 2000 Scientific Sessions of the American Heart Association (AHA) in November, there was a session entitled “Carotid Imaging in the Year 2000: Which Modality for Which Patient?” that looked at screening as well as diagnostic studies. The AHA, through its American Stroke Association, has already adopted the cause of stroke prevention and treatment, a program into which advocacy of screening carotid ultrasonography would fit well.

Some radiology and vascular practices have established screening clinics, sometimes mobile operations, that provide low-cost carotid screening studies. Bluth has some concerns about these efforts.

“If the studies are to be of value, the technicians must be properly trained and certified in vascular sonography. Also, the laboratory must be certified by an organization such as the Intersocietal Commission on Accreditation of Vascular Laboratories, the American Institute of Ultrasound in Medicine, or the American College of Radiology. Otherwise, we could see a repeat of the situation in which substandard mammography, performed before the Mammography Quality Standards Act (MQSA) regulations, overlooked significant disease.”

A situation in which vascular screening by ultrasonography is of undisputed benefit is diabetes. Two presentations at the European Congress of Radiology Meeting in March 2000 demonstrated the value of color Doppler ultrasonography in diabetic patients, even those with no symptoms of vascular disease. Matteo Salcuni, MD, and associates of the University d’Annunzio in Imbaro, Italy, screened the carotid and leg arteries and the abdominal aorta of 217 patients, with all those having evidence of significant stenosis undergoing angiography. In 73% of the patients, stenoses of 50% or greater were discovered. Importantly, almost half of the patients with such lesions were asymptomatic. In the other series, P. Hadjilira, MD, and colleagues from Athens, who examined 40 patients with Type II diabetes being seen at the clinic for the first time, concluded that vascular color Doppler ultrasonography is essential for the initial evaluation of diabetic patients.

Not Just for Research

In 1986, Valentin Fuster, MD, of Mt Sinai Hospital in New York City found that 50% of patients who had undergone angiography shortly before suffering an acute coronary event had stenoses of no greater than 50%. Clearly, knowledge of more than the size of plaques was needed. As Fuster explained in a presentation at the AHA meeting last November, MRI can provide what he referred to as a biopsy of the vascular wall. He and his associates have used MRI to confirm the regression and stabilization of plaques during dietary or drug treatment of animal and human atherosclerosis and to monitor vascular disease over time.

In the minds of many, MRA and CT angiography (CTA) are still research tools. Geoffrey D. Rubin, MD, chief of cardiovascular imaging at Stanford University School of Medicine, Stanford, Calif, disagrees.

“Most CTA and MRA are well beyond the research stage,” he stresses. “Although there are some variations in the relative indications for these techniques, at institutions where they have been adopted, they have virtually replaced diagnostic contrast angiography.” He reports that his department trains three to five private practice radiologists every week in the techniques of CTA, MRA, or both.

Both CTA and MRA have the advantage of noninvasiveness or at least minimal invasiveness. Moreover, there is less patient discomfort and morbidity, and the study takes less time than is required for contrast angiography. The result is increased department throughput and less time away from work or other activities for the patient. The noninvasive studies also are often less expensive, according to Rubin.

The ability of CTA and MRA to depict the vessel wall as well as the lumen is of particular value in aneurysmal disease, where it is important to determine the true extent of the lesion when planning therapy. Rubin and his associates recently described a comparison of multidetector array and standard CTA in 48 patients with aortic aneurysms or dissections.3 The studies on the multidetector array scanners were 2.6 times as fast, with a fourfold increase in scanning efficiency, a 40% reduction in section thickness (mean 3.2 vs 5.3 mm), and a smaller dose of contrast medium: an average of 97 mL vs 232 mL. In one patient with diabetes-associated renal impairment, gadolinium was used successfully as the contrast agent.

Other uses of MRA and CTA are being pioneered. Hatsukami and associates of the VA Puget Sound Health Center in Seattle recently demonstrated that high-resolution MRA can depict the thickness of the fibrous cap, and thus the risk of rupture, of plaque in the carotid artery.4 At the 49th Annual Scientific Sessions of the American College of Cardiology, Clifford R. Weiss, MD, of the National Heart, Lung, and Blood Institute described the use of gadolinium-enhanced MRA with a double inversion recovery fast spin echo sequence to study the carotid and vertebral arteries of patients with suspected cerebrovascular disease. Of particular interest was the evidence of vascular inflammation such as increased T2 values or contrast enhancement of the wall, increased wall thickness, or both, in the carotid artery or intrarenal aorta. All 15 subjects with one of these MRI findings had higher serum concentrations of the inflammation markers VCAM-1, ICAM-1, and interleukin-6 than did the other patients. The results suggest that MRA can identify atherosclerosis at an earlier stage than can be detected with the usual morphologic imaging methods.

The relative roles of CTA and MRA are still being defined. The former is more widely available, particularly for patients in emergency departments or intensive care units, and the examination protocols are largely standardized. Moreover, scans can be obtained on patients with defibrillators or pacemakers, which is not true of MRI. At present, MR scans require more physician input and customization, and it can be more difficult to obtain a high-quality scan. On the other hand, MR does not require iodinated contrast, an advantage in a patient who is allergic to iodine or who has borderline renal function. Finally, three-dimensional renderings are easier to create with MRI, in part because the bones are not seen.

Role for Contrast Angiography

Contrast angiography still has a place. For example, neither CTA nor MRA provides good images of the smallest arteries and veins (1 mm). Angiography is preferred for studying arteriovenous fistulas or malformations, particularly when the lesions are being mapped for treatment. Intracranially, berry aneurysms have traditionally been studied by angiography, although CTA and MRA are getting better at depicting these lesions.

The method of contrast angiography is changing too, as the traditional cut film study is being phased out. At Ochsner, one of the angiography suites is still equipped to obtain cut films, “and I think we have not used that machine for years,” Kay observes. “We have gone strictly to digital subtraction imaging. The earliest DSA equipment was much more difficult to use, and the quality was not ideal. In the early days, we also did some intravenous subtraction studies. Today, we do not even think about using cut film, and intravenous subtraction studies are rare.”

There is one situation in which the subtraction images alone are not entirely satisfactory. “In the aortograms and runoff studies, our surgeons like to see the anatomic background. Also, if there is a stenosis near the carotid bifurcation, they want to know how high up in the neck it is.” In these cases, the angiographer prints some nonsubtracted images because surgeons like to see bones.

The principal use of contrast angiography today is for guidance of interventional procedures. Coding can be a challenge.

“There is a code for nonselective placement of an aortic catheter,” Kay reports. “There is another for the abdominal and bilateral lower extremity study. If we do iliac angioplasty, there is a code for that, and if we place a stent, there is another.”

Even in interventional procedures, however, angiography is giving way to CT and MR.

“Endovascular techniques are becoming more popular for the treatment of aortic aneurysms, and CTA is important for these procedures,” Rubin explains. “There is a growing awareness that CT is more sensitive than angiography in identifying endoleaks-areas where flow around the stent graft fills the aneurysmal sac. This is a controversial endpoint, but it is considered important now.

“Contrast angiography should be fading except for planning or carrying out an interventional procedure,” Rubin believes. “If an institution is diligently working to direct patients toward the most appropriate imaging modality, they will find their use of contrast angiography to be maybe 5% to 10% of what it was a decade ago.”

So why are so many diagnostic contrast studies still being performed? Part of the reason is lack of education, not only of radiologists, but also of referring physicians. Sometimes, the path of least resistance is for the radiologist simply to do what is ordered, not explain to the referring physician why CTA or MRA might be better than contrast angiography.

And, of course, there is the question of money. “There may still be some economic advantages to doing the invasive procedure,” Rubin points out. “The reimbursement is higher. If there are economic motivators in selecting a modality, you are going to pay a price for choosing the noninvasive vascular imaging method.”

Lighting Up the Lesions

Nuclear medicine also has a small role in the diagnosis of vascular disease. At the May meeting of the American Roentgen Ray Society, Ahmed

Outif, MBBS, from the University of Bristol in England, described a method of measuring the arterial circulation in the leg. The leg is placed in the field of view of a gamma camera, and after the patient has rested for 10 minutes, Technetium Tc 99m is injected. Three minutes later, a blood pressure cuff is inflated above the knee for 4 minutes. Beginning 15 seconds before the cuff is deflated, images are acquired, and a time-activity curve is generated. These curves identify the site of maximum flow obstruction.

In 1998, the Food and Drug Administration granted marketing approval to a technetium-labeled synthetic peptide that binds to activated platelets as a tool for diagnosing deep venous thrombosis. Other radiopharmaceuticals are under development for this indication. At the Society of Nuclear Medicine meeting in June, Bryson Borg, MD, and associates from Pennsylvania State College described the use of a technetium-labeled agent to identify deep venous thrombosis after total knee arthroplasty. The nuclear scan had a sensitivity of 91% vs 27% for ultrasonography. However, the latter study was more specific: 100% vs 75% for the technetium-labeled agent.

Reimbursement Issues for CTA

In the summer of 2000, the American College of Radiology created eight codes for CTA. However, the valuation is currently in flux. Data have been collected for all eight codes, which will be presented to the Relative Valuation Utility Commission in February. It is hoped that the valuations will be firmly established by the end of 2001.

Even this list of vascular imaging methods does not exhaust the possibilities. Endovascular ultrasonography has been used extensively in some medical centers, particularly for cardiac studies. Angiography also is seeking a place. It is likely to be some time before it is possible to automatically select an imaging modality for vascular disease.

Judith Gunn Bronson, MS, is a contributing writer for Decisions in Axis Imaging News.


  1. Spittell PC. Peripheral vascular disease. In: Pohost GM, O’Rourke RA, Berman DS, Shah PM, eds. Imaging in Cardiovascular Disease. Philadelphia: Lippincott Williams & Wilkins; 2000:845.
  2. Bluth EI, Sunshine JH, Lyons JB, et al. Power Doppler imaging: initial evaluation as a screening examination for carotid artery stenosis. Radiology. 2000;215:791-8000.
  3. Rubin GD, Shiau MS, Leung AN, Kee ST, Logan LJ, Sofilos MS. Aorta and iliac arteries: single vs multiple detector-row helical CT angiography. Radiology. 2000;215:670-676.
  4. Hatsukami TS, Ross R, Polissar NL, Yuan C. Visualization of fibrous cap thickness and rupture in human atherosclerotic carotid plaque in vivo with high resolution magnetic resonance imaging. Circulation. 2000;102:959-964.