February 25, 2008

Adjusting Contrast — One Size Does Not Fit All
By Beth W. Orenstein
Radiology Today
Vol. 9 No. 4 P. 20

Clinicians increasingly recognize that some patients need more contrast—or less—than standardized protocols dictate.

As scanner technology has advanced, CT angiography (CTA) has become a more commonly used tool to help detect coronary artery disease in patients. Traditionally, imaging has used a one-size-fits-all approach to delivering the contrast that is needed when scanning cardiac patients. However, it is widely recognized that some patients need more or less contrast agent than others depending on their body size, the scan duration, and other factors.

“A little old lady whose heart is not pumping so vigorously anymore will need less contrast than a younger person with a fast cardiac output,” says U. Joseph Schoepf, MD, associate professor of radiology in diagnostic radiology at the Medical University of South Carolina in Charleston.

To solve this issue, Medrad Inc. of Indianola, Pa., maker of the Stellant CT Injection System, has developed personalized patient protocol technology (P3T) for the accurate, personalized delivery of contrast during CT exams.

Medrad showcased CardiacFlow, the first application of its P3T technology, at RSNA 2007 in November. CardiacFlow, which is specifically designed for cardiac CT imaging procedures, received FDA clearance in December.

Coming Soon
Anthony Emerick, RT(R)(CT), a clinical marketing specialist for Medrad, says CardiacFlow is in use in Europe and should be commercially available in the United States later this year. Harald Seifarth, MD, assistant professor of radiology at the University of Muenster in Germany, evaluated CardiacFlow and continues to use it for cardiac CTA cases.

In development since 2005, CardiacFlow works with any scanner from any vendor. With the system, a patient is given an optional test bolus of contrast to determine how long it takes to reach the region of interest and its peak. Factors such as the patient’s weight, contrast concentration, scan duration, and maximum tolerable flow rate are then entered into the injector. Future software revisions may also consider kidney function, body surface area, heart rate, other scanner configurations such as tube voltage, and whether the cardiac procedure is prospectively triggered or retrospectively gated.

Using algorithms developed by the research scientists at Medrad and informed by radiology literature, the software calculates the circulating blood volume and thus the proper amount of contrast needed to complete the exam. It determines not only the amount of contrast but also the rate of delivery and whether a mix of contrast and saline is needed. The software can be run from a PC next to the scanner or be attached to the injector.

The work of Kyongtae Bae, MD, PhD, of the University of Pittsburgh Medical Center, and Dominik Fleischmann, MD, of Stanford University Medical Center, was heavily referenced in the design of CardiacFlow’s algorithm, according to John Kalafut, MSEE, principal research scientist at Medrad. “Bae’s full-bodied, contrast medium pharmacokinetic model aided during validation of the concepts,” he explains.

Individualized, patient-based contrast dosing, especially for a procedure as technical and time critical as CTA, solves many issues, say the researchers who have worked on the patent-pending CardiacFlow.

Better Images
Personalizing contrast delivery not only makes the procedure safer for patients, but it also produces higher quality images that reduce the need for repeat exams. More accurate dosing and fewer repeat exams also reduce costs and improve efficiency.

Emerick says personalized dosing does not necessarily mean less contrast will be used. He says in some patients, particularly those who are heavier, the opposite may be true. Because the algorithms are based on factors such as the patient’s weight, “If you have a patient who is 300 pounds, you could actually use a greater amount of contrast than you normally would on your preprogrammed protocol,” he explains.

What P3T technology guarantees is that the proper amount of contrast is used every time, and in some patients, that may be less than the facility’s protocols normally require, Emerick says.

Emerick says CT technologists traditionally have relied on their experience to adjust contrast delivery depending on a patient’s body size and circulation. An experienced technologist is more likely to make the appropriate adjustments and deliver the proper amount of contrast to a larger patient, Emerick says. But a novice or someone who does few CTAs could have more difficulty appropriately adjusting the contrast amount and the timing of the delivery.

P3T technology “takes the guesswork out of the practice,” which could help reduce overdosing with contrast or potentially having to rescan the patient, according to Emerick.

Technologists, especially those who are less experienced, tend to err on the side of giving too much rather than too little contrast, Emerick says. “They’ll say, ‘I’ll give them 150 [cubic centimeters],’ when in actuality they may have achieved high-quality diagnostic imaging with 80 [cubic centimeters].” In those cases, scientifically calculating the dosage would reduce the amount of contrast given.

More precise contrast delivery dosing also reduces the risk of contrast nephropathy in certain patients, says Pal Suranyi, MD, PhD, of the diagnostic radiology department at the Medical University of South Carolina. “Technically, contrast media is not good for sick kidneys,” he says. “If a patient has poor renal function, you want to minimize the amount of contrast injected while also ensuring that sufficient enhancement will be present in the coronaries.”

Carefully selecting the proper dosage and timing for each patient also improves image quality. Suranyi and colleagues conducted a study of 120 patients undergoing cardiac CT. Seventy patients were given individualized contrast volume and injection parameters of a triphasic (contrast, contrast/saline, and saline phase) injection protocol. The personalized algorithm was based on scan duration, patient-specific variables, and the time-attenuation response in the aorta and pulmonary artery following a test bolus. A control group of 50 patients was injected with routine contrast protocols. The images from the two groups were analyzed to compare the level and homogeneity of enhancement with the aorta and the coronary arteries.

“Our results showed the computerized, patient-based protocols achieved very homogeneous and high attenuation throughout the entire coronary tree,” Suranyi says. The patient-specific contrast protocols clearly produced higher and more uniform coronary enhancement, and the diagnostic display of the heart was improved, he adds.

When standard protocols are used, it is not uncommon for contrast material to be brighter at the beginning of the vessel under investigation and fade out toward the end of the vessel, according to Suranyi. Individualized dosing provides greater homogeneity, resulting in better quality images for the interpreting physician, he says.

Because the image quality is greater with the individualized protocols, it reduces the likelihood of a repeat exam, which would not only be costly but also time-consuming for the technologists and physicians, as well as unnerving for the patients, the researchers say.

It also improves workflow. With better images, the radiologists aren’t slowed by having to call the technologist down the hall and reviewing the protocols he or she used. They can read in an efficient fashion. Also, Emerick says the physicians need quality images to be able to generate quality 3D renderings.

Suranyi also says that P3T technology is a good tool for researchers like himself. “It’s a gold mine being able to retrospectively analyze the injection parameters in every patient,” he says. “You can go back and look at the contrast media protocol and see what effect it has had.”

Other improvements also are in the works. In March 2007, Medrad received FDA clearance for its XDS Extravasation Detector, which is designed to prevent moderate to severe contrast leaks into surrounding tissue that may seriously injure patients. Because XDS technology directly senses fluid pooling under the skin in order to stop an injection during a CT procedure, it minimizes false positives to maintain efficient workflow.

Medrad is also working to improve workflow by integrating the P3T injector system with the scanner. Medrad sells scanner interfaces for Siemens, Toshiba, and Philips CT scanners. However, the current functionality only allows for the scanner to trigger the injector; advanced functionality, such as exchanging parameters to generate P3T protocols, does not exist.

“Currently if you do a CT scan, what the technologist has to do is press two buttons—one to start the injector and another to start the CT,” Schoepf says. “With integration, the entire injector interface can be brought up on the actual CT scanner interface, and you’re basically dealing with one device instead of two so you can determine your injection protocol from your CT scanner console.”

When the technologist is well trained, “They almost always get it right,” Schoepf says. “There is very little of what we see in terms of technologist error. However, if the technologist is in a rush or you have a novice who hasn’t done it a whole lot, it is a whole lot easier to have your contrast media injection protocol integrated into your CT scan protocol and to have to press only one button instead of doing double action on the CT scanner and the injector.”

A former technologist, Emerick says the integration and individualized dosing technology also can serve as a recruiting tool when radiologic technologists are in short supply. Because of the shortage, many facilities are cross-training their technologists. “With cross-training, you’re working in CT a few days and MRI a couple days. That happened to me years ago when it was a really popular concept,” Emerick says. Unfortunately, he adds that this can mean technologists are not well trained in one procedure or modality. It doesn’t mean the technologist can’t perform CTA, but if the contrast dosing were programmed in and all he or she had to do was push a button, it would make it that much easier, he says.

“There was so much interest [in CardiacFlow], I lost my voice by Tuesday,” Emerick says of discussing the product at RSNA 2007. Many physicians he spoke with at the conference were complaining that they had a problem with high turnover among technologists, and they would often train someone only for him or her to move on. Personalized protocols for contrast delivery could help because it would make it easier for new technologists to step in, Emerick says.

Integrating the dosing protocol software with the facility’s RIS and PACS is also promising, says Kalafut. The exchange of contrast injection data with those systems is also under development in response to clinical need.

Information about the amount of contrast that was used can be sent from the scanner to the billing staff, which can then ascertain that the proper amount is entered for reimbursement. Also, having the scanner transfer information about the protocols that were used eliminates errors that can occur when the information is entered by hand, Kalafut says.

Medrad is planning other applications for its P3T technology including neuroradiology scans and body angiography. Meanwhile, reimbursement—the future of CTA of the heart—is somewhat uncertain because the Centers for Medicare & Medicaid Services has proposed limiting Medicare coverage for diagnostic CTA to just two indications and only in cases where the patient is enrolled in a clinical trial. However, many physicians favor CTA as a noninvasive procedure for diagnosing coronary artery disease. With personalized contrast dosing making the procedure even better and safer.

— Beth W. Orenstein is a freelance medical writer based in Northampton, Pa and a frequent contributor to Radiology Today.