Interventional radiology sets agenda for 21st century

Cardiovascular" may no longer be part of the Society of Interventional Radiology's name, but vascular access remains a crucial element of the latest research presented at the society's annual meeting in Baltimore in April. Stent design, embolization therapies, and gene delivery topped the clinical agenda even as the society launched ambitious efforts to reinvent its image.

Politics and public relations played a major role at the meeting: After research confirmed that most patients lack a clear understanding of what interventional radiologists do, members of the Society of Cardiovascular and Interventional Radiology voted at the meeting to shorten their society's name. They also launched efforts to educate both patients and referring physicians about the skills and therapies interventionalists offer.

"It's less about changing our name and more about educating the public to the benefits of interventional radiology," said SIR executive director Paul Pomerantz. "Interventional radiology has set the agenda for medicine in the 21st century. But if we don't tell people who we are and what we do, we'll lose to other specialties."

ROBOT BIOPSIES

Some research presented at the meeting seemed designed to advance the specialty while streamlining the work of interventionalists. Researchers at Johns Hopkins University have begun using a robotic arm to perform successful core needle biopsies. In the future, the technology may be applied to drainages and ablations as well.

The robotic arm attached to a gantry uses a radio-lucent finger to hold a core biopsy needle that can advance or retreat. The remotely controlled robot can move around a fixed point from any angle. The needle tip in the skin and the target lesion are imaged with CT, and those data are used to plot the needle's trajectory. Dr. Stephen B. Solomon, an assistant professor of radiology and urology at Hopkins, revealed that 10 patients with lesions ranging from 1 to 3 cm had been successfully biopsied, with CT confirmation of the needle tip within the lesions.

An operator can point to the needle tip or the target on the computer screen or use a joystick for real-time guidance. The robotic arm then rotates to the appropriate trajectory to drive the needle to the target. CT fluoroscopic imaging was used to visualize needle placement.

Occasionally, respiratory motion changed the coordinates. In those instances, the robot missed the target, Solomon said. Future research will focus on a respiratory tracking method. Solomon and his team also intend to try the robot's hand in radio-frequency ablation and drainages.

"The best application may be for difficult trajectories and to reduce radiation exposure to physicians," Solomon said.

A study comparing standard biopsy with robotic biopsy is planned for 2003.

COLD VIRUS FIGHTS CANCER

A new approach to cancer therapy, meanwhile, involves fighting one of the deadliest diseases with one of the most common.

Phase I of a multicenter trial testing dose limits of a synthetic adenovirus-the common cold-showed tumor suppression in patients who received higher doses, said lead investigator Dr. Daniel Y. Sze, an assistant professor of radiology at Stanford University Medical Center. The therapeutic virus, when injected intra-arterially to the liver, appeared to kill tumors without harming healthy tissue.

"You can think of this virus as a new generation of chemotherapy that is much more selective about what it attacks," Sze said.

Unlike most viruses used in gene therapy, this one retains the ability to replicate. When an infected cancer cell dies, it breaks open, releasing the virus and all its copies, which can then infect other cancerous cells and start the process again.

All 35 patients in the study had gastrointestinal cancer-most originating in the colon-that had spread to the liver. Each had received unsuccessful chemotherapy, and none were surgical candidates. The life expectancy of each was about six months. The median survival time of the 28 patients who received the highest doses was about one year. More research is needed to prove the efficacy, Sze said.

The virus, which is genetically engineered to be weaker than normal, is designed to infect only cells with an abnormality in the tumor suppressor gene p53. Up to two-thirds of cancers have abnormal p53 function.

"The serendipitous finding was that these cells cannot recognize when they're infected by a virus. It makes the cell particularly susceptible to infection by this engineered virus," Sze said.

CT imaging showed that the tumors grew slightly larger immediately after treatment, in a suspected inflammatory response to the viral infection. The tumors then slowly got smaller. More impressive than tumor shrinkage were blood tests that showed a significant decrease in abnormal proteins being secreted by the tumors, Sze said.

"That suggests the tumors, although still visible on the CT scan, are dying or dead," he said.

Sze won the Young Investigator Award for the paper. Phase II study, due to start this year, will pair the virus with chemotherapy, which seems to increase its potency. Sze added that it could be years before this treatment is ready for routine use.

STENT CONSTRUCTION FLAWED

Most stent research has focused on examining the prosthetics after they are extracted from the body to determine the nature of devolution. But Dr. Julio C. Palmaz, a pioneer in stent research, has taken a different approach by examining intra-arterial stents straight from the package.

Using x-ray photoelectron spectroscopy, he and colleagues from the University of Texas Health Science Center at San Antonio found significant surface defects in various stent brands currently available and widely used in the U.S.

"This is the first observation of the surface of stents with a metallurgic microscope. It shows that we should address the composition of stents the same as we do with drugs," said Palmaz, chief of cardiovascular and interventional radiology at UTHSC.

Palmaz presented the research for lead investigator Dr. Cristina S. Fuss. The perfect stent has yet to be developed, he said, but interventional radiologists would be happy with a 5% restenosis rate, compared with the current 50% rate.

The researchers examined nine coronary and three peripheral stent brands. The bulk stent material was stainless steel in eight, nitinol in three, and gold-coated stainless steel in one. As expected, the predominant surface elements were carbon and oxygen. The next most abundant element should have been iron, but it was silicon from surface lubricants, Palmaz said.

Researchers also found a relationship between chromium and silicon. Stents with the least chromium, the most important surface element of expandable stents, had the most silicon. Many stents had sodium residue, an alkali component that is not biocompatible, Palmaz said. The results varied not only among stent brands but within batches of the same product.

"What does this mean clinically?" Palmaz said. "We're using radiation and elution for a product that's imperfect."

Another study, from Germany, examined the thrombogenicity of various endovascular stent types and found a wide range of platelet activation following stent implantation.

"What matters with stents are placement and long-term results," said lead investigator Dr. Gunnar Tepe from the University of Tuebingen.

Tepe and colleagues placed 104 total stents in a modified Chandler-Loop. The platelet count, beta-thromboglobulin, and the TAT (thrombin-antithrombin-III complex) were assessed at start time and after two hours.

After two hours, significant differences were seen. TAT (mg/L) ranged between 31 (control, no stent), 328 (Bard peripheral), and 5897 (Jomed SelfX, without electropolishing). While some stents did well with polishing, others did not.

Researchers concluded that electropolishing reduces the thrombogenicity of the stents. Altering the stent surface for local delivery with radiation did not increase platelet activation.

Stents were ranked overall from best to worse:

• Saxx

• Palmaz Corinthian (with electropolishing)

• Jomed peripheral (with electro-polishing)

• Palmaz Schatz, Schatz Re188, and Schatz Re186

• Bridge peripheral

• Palmaz Corinthian (without electropolishing)

• Jomed peripheral (without electropolishing)

• Jomed SelfX (with and without electropolishing)

MR FILTERS UFE CANDIDATES

Chicago researchers caution that interventional radiologists should routinely use pelvic MRI to assess women prior to uterine fibroid embolization. In their study, 98% of women were thought to have fibroids prior to MRI examination. After MR, the number dropped to 83%. In addition, interventional radiologists had a 22% increase in diagnostic confidence after MRI.

Five interventional radiology attending physicians at Northwestern University Medical School were asked to prospectively complete questionnaires before and after MRI was obtained in their evaluation of women presenting for potential UFE, said. Dr. Reed A. Omary, a professor of radiology at Northwestern.

The questionnaires posed three queries to the physicians:

• What is your single most likely diagnosis prior to MR?

• What is your diagnostic confidence in this single diagnosis (in terms of 0 to 100% certainty)?

• What was your anticipated treatment plan prior to MR and what is your final treatment post-MR?

Diagnostic choices included fibroids, adenomyosis, endometrial diseases, adnexal masses, and multifactorial. Treatment choices were clinical management, biopsy, UFE, and surgery. Pelvic MR scans included axial/sagittal T2-weighted and pre/post-gadolinium-enhanced T1-weighted images.

The IRs completed 60 consecutive questionnaires. Before MRI, mean diagnostic confidence was 76%, while after MRI, it was 98%. MRI thus caused a mean gain in diagnostic confidence of 22%, Omary said.

MRI changed the initial diagnoses of IRs in 11 patients (18%). Initial treatment plans changed in 12 patients (20%). Before MRI, 57/60 (95%) women were expected to receive UFE, but after MRI, UFE was not recommended in 10 of the 57 women (18%). In these 10 women, IRs recommended surgery for eight, clinical management for one, and biopsy for one.

"Interventional radiologists increa-singly use pelvic MRI for this population, but the effect MR has on altering the diagnosis and treatment decisions had not been quantified until now," Omary said.

Some IRs who consider MRI too expensive asked Omary how high-quality ultrasound would compare. The researcher said that, anecdotally, the attending IRs had many patients who were diagnosed with diffuse fibroids based on ultrasound scans.