BBI Contributing Editor

SAN DIEGO, California – The annual meeting of the Society of Cardiovascular and Interventional Radiology (SCVIR; Fairfax, Virginia), held here this spring, provided an opportunity to capture a comprehensive overview of the fast-growing field of interventional radiology. As shown in Table 1 on page 122, the total estimated U.S. market for interventional radiology products in 2000 is expected to exceed $600 million, a figure that includes a wide variety of product segments such as PTA catheters and stents, endovascular grafts, gastrostomy and nephrology products, interventional neuroradiology products, thrombectomy and embolectomy devices, vena cava filters, and drainage catheters. Devices for vascular therapy including balloon catheters, stents, and ancillary devices, account for over 60% of the market, or approximately $370 million. The market is expected to grow rapidly, at over 15% per year, over the next five years due to a number of factors, including the adoption of new, high-cost devices such as stent-grafts, increasing use of interventional procedures in the treatment of neurovascular disease, and growth in the use of devices for the treatment of peripheral artery, venous thrombosis, and thromboembolism.

Another rapidly growing application for interventional radiologists is percutaneous vertebroplasty, a less-invasive approach for treating vertebral compression fractures caused by osteoporosis, malignancy or conditions such as vertebral hemangioma. The technique involves the injection of acrylic cements into the damaged vertebra under fluoroscopic guidance, and provides instant relief of pain in many cases. Although the technique was first introduced in the U.S. in 1993, vertebroplasty has yet to become a standard of care in this country, while in Europe the procedure has been widely adopted. But as more U.S. interventional radiologists have become trained in performing the procedure, use has begun to expand rapidly.

Another growth segment of the market is interventional neuroradiology products, including microcatheters, occlusion coils and particles, and, in the future, neurovascular stents. Key suppliers in the interventional neuroradiology segment continue to achieve sales growth of 30% or more. Sales for Target/Boston Scientific (Fremont, California) increased about 30% in the most recent quarter, and sales for the neurovascular products unit of Cordis/ Johnson & Johnson (Miami Lakes, Florida) rose 38% in 1999. Emerging new product segments such as carotid stents and protection filters promise to allow further rapid expansion of sales. Additional growth areas of the interventional radiology market include hemodialysis access devices; new materials and devices for embolization, including uterine artery embolization; new stents for applications in treatment of renal artery disease and biliary disease; new technologies for prevention of restenosis or for revascularization such as brachytherapy and gene therapy; products for radiofrequency (RF) ablation of tumors and chemoembolization; and new products for removing thrombus from blood vessels.

Although in the past, the interventional radiology market has been limited somewhat by the small number of physicians specializing in the field, the number of interventional radiologists is continuing to grow. As of March 2000, SCVIR had a total of about 3,500 members, vs. less than 2,000 as recently as 1996, and registration for this year's SCVIR conference exceeded 5,000. The market remains limited to a degree by the scarcity of specialists, particularly in the interventional neurology segment where there are only about 250 specialists worldwide. However, other specialists, including cardiologists and vascular surgeons, are also moving aggressively to adopt interventional radiology techniques as part of their practice, expanding the available market. Demand for less-invasive procedures is expected to continue to grow as new techniques and devices allow an expanded range of diseases to be treated, and as patients become increasingly aware of interventional radiology as an alternative to surgical therapy.

Market opens for endovascular grafts

An important segment of the IR market that is just now beginning to develop is endovascular stent-grafts. As shown in Table 2 on page 123, a number of new stent grafts are under development in addition to the two FDA-approved devices from Medtronic and Guidant. Pricing for the new devices is considerably higher than for conventional stents used for vascular treatment. Medtronic's AneuRx stent-graft is priced at approximately $11,000 in the U.S., and the Guidant Ancure device is priced at $9,950. Suppliers believe that, even at prices of around $10,000 per graft, the total cost of treatment will be less than for surgical repair of an abdominal aortic aneurysm. Hospital stay is reduced considerably, based on data from clinical studies, where patients treated with stent-grafts had stays of two to three days, vs. six to 10 or more days for surgical treatment. A 50% reduction in morbidity was achieved using the AneuRx device as compared to surgery, and there was a 56% reduction in the need for blood transfusions. In the studies performed with the Guidant Ancure stent-graft, complications were also reduced considerably, from about 40% in the surgically-treated group to 19.6% for tube grafts and 28.7% for bifurcated grafts.

Over time, prices for endovascular grafts are expected to fall dramatically as new devices are approved for marketing and competition in the market increases. In Europe, for example, there are now five devices available, and prices have already dropped to around $6,000 per device. Some vendors expect that prices in the U.S. will drop to the $2,000 range within two to three years, assuming additional devices are approved during that time. However, because of patient demand for endovascular graft therapy, and because of the large potential market (about 200,000 patients are diagnosed with AAA each year, and 15,000 patients die each year from the condition), unit volume is expected to grow very rapidly. Sales may exceed $200 million by 2005 in the U.S. alone.

Endoleaks remain an important concern for clinicians evaluating the safety of stent-graft therapy. A variety of factors can result in an endoleak, including attachment failures, branch flow of blood from collateral vessels, graft defects such as tears in the material, and excessive porosity of the graft fabric. Most endoleaks are due to branch flow from collaterals. However, such leaks rarely lead to rupture of the aneurysm, the adverse event that therapy is intended to prevent. In studies with the Ancure device, for example, there were no ruptures in 597 implants in spite of an endoleak rate approaching 30%. In a study with the AneurRx device described by Christopher Zarins, MD, of Stanford University Medical Center (Palo Alto, California) at the SCVIR conference, the endoleak rate was 13% at one month post-discharge, but only 0.7% of patients suffered a rupture (7 out of 1,067 patients). In addition, it is believed that several of the observed ruptures could have been prevented if the most recent version of the device had been employed. There were no deaths due to aneurysm rupture in the patients treated with the AneuRx device.

New devices under development are expected to have some additional advantages that could allow other suppliers to capture a share of the growing endovascular graft market. For example, Cook (Bloomington, Indiana) is developing modified versions of its tracheo-bronchial covered stent that can be custom manufactured to match a specific patient's anatomy. The devices have been used to treat aortic dissection with good results as compared to surgery. The Medtronic Talent device and the W.L. Gore (Flagstaff, Arizona) Excluder also have proven useful in that application. In the case of the Talent stent-graft, hospital stay was reduced from 40 days to seven days in a study of 24 patients, and mortality was reduced to zero vs. 33%. According to Lindsay Machan, MD, of the University of British Columbia (Vancouver, British Columbia, Canada), who discussed stent-grafts at SCVIR, the Gore Hemobahn and Excluder devices are easiest to deliver, while the Talent provides the greatest variety of shapes to match patient anatomy. The development-stage Cordis device offers the smallest delivery system available, while the Ancure and AneuRx devices have the obvious advantage of FDA-approved status.

However, based on the most recent data, stent-graft applications may prove to be somewhat more limited than developers had initially hoped. While the devices perform well in terms of patency when implanted in large-diameter vessels such as the abdominal aorta, studies using endovascular grafts in small arteries have been discouraging. For example, the Wallgraft from Boston Scientific (Natick, Massachusetts) has been evaluated for the treatment of lesions in the carotid, iliac and subclavian arteries, but patency at 33 months was low at 48%. Similar results were obtained with the Gore Hemobahn for the treatment of femoro-popliteal lesions, with patency of about 72% at six months. While such applications are clearly challenging, the early results are not as promising as had been expected. As more is understood regarding the mechanisms of restenosis and tissue in-growth following stenting in smaller vessels, it is becoming apparent that there is a considerable amount of research needed in order to adequately understand the effects of vascular biology.

Vertebroplasty emerges as major procedure

Percutaneous vertebroplasty has been in use in the U.S. since 1993, and has been widely used in Europe for an even longer period. However, the procedure is only now beginning to grow rapidly in volume, as more patients become aware of the benefits as compared to alternative treatments. In addition, more radiologists are learning the technique, driving growth in procedure volume. Nevertheless, the procedure has not yet become the standard of care in the U.S., in part because there are no FDA-approved materials for use in vertebroplasty. The most commonly used material is Simplex bone cement, manufactured by Stryker's (Kalamazoo, Michigan) Howmedica division (Allendale, New Jersey). However, Simplex is not radiopaque, and consequently interventional radiologists must add materials such as barium sulfate or tantalum powder in order to visualize the distribution of the injected material. Another material that is commonly used for vertebroplasty procedures is Cranioplastic Type I PMMA cement from Codman/Johnson & Johnson (Berkshire, United Kingdom), and new hydroxyapatite materials are under development by Howmedica, Norian (Cupertino, California), and Orthovita (Malvern, Pennsylvania). In addition, a bis-gamyl methacrylate material is being studied by Orthocomp, and a bioactive glass component is being studied by Orthovita. Hydroxyapatite materials have the advantage of not requiring the addition of opacification agents.

To perform the vertebroplasty procedure, the diseased vertebra is first identified radiographically, and an 11-13 gauge needle is inserted under fluoroscopic or CT guidance. Suppliers of commonly used needles include Cook, Parallax Medical (Mountain View, California), and MD Tech (Gainesville, Florida). Usually, a test injection is performed using contrast material to assess probable flow patterns, followed by injection of the PMMA under direct fluoroscopic control. Proper control of the placement of the PMMA or other material is critical, since serious adverse consequences, including death, can occur if the material is allowed to migrate out of the target region. Although standard bone biopsy needles are currently used for the procedure by many physicians, dedicated vertebroplasty needles are under development by various manufacturers. One novel device that has proven useful in vertebroplasty is the EZflow bone cement delivery system manufactured by Parallax Medical, which is targeting vertebroplasty as a major area of market focus. The EZflow system allows physicians to use real-time fluoroscopy throughout the cement delivery procedure while minimizing radiation exposure. The system includes specially designed access needles and a volumetrically controlled injector. Prices for the disposable EZflow kit range from $295 to $495 in single quantity, and the prep kit is priced at $90.

Another new device with applications in the treatment of vertebral compression fractures is being introduced by Kyphon (Santa Clara, California). Kyphon's Kyphoplasty system, the KyphX Inflatable Bone Tamp, includes a balloon catheter that can be inserted into the vertebral space and inflated, creating a void that allows bone cement to be injected in a highly viscous state. As a result, migration of the cement outside of the target zone is minimized. In addition, and perhaps more importantly, the balloon inflation provides height restoration of the vertebra, an added benefit not provided by standard vertebroplasty procedures. As with all vertebroplasty procedures, a certain portion of the vertebral wall – the posterior wall in this case – must be intact in order for the patient to be a candidate for the kyphoplasty procedure. The Kyphon device has been cleared by the FDA, and will be introduced later this year. The product will be priced at about $2,000 for treatment of an initial level, and each additional level treated will cost about $1,000. Over 400 fractures have been treated to date using the kyphoplasty procedure. Kyphon is a privately held company.

There is a clear need for improved devices for vertebroplasty that can help to ensure correct placement of cement, as well as for new cements that have improved biomechanical properties and that are radiopaque. Proper control of the procedure is critical, since there have already been some patient deaths reported due to leakage of cement outside of the target zone. Nevertheless, complication rates have generally been in the range of 1% for vertebroplasty. Another potential advance is to apply robotics to control needle placement and cement injection. In addition, there is the potential to incorporate chemotherapy agents in the injected material, which could benefit patients who have suffered vertebral damage due to tumor in-growth.

As shown in Table 3, a major market opportunity exists for vertebroplasty products, estimated at $400 million. According to Jacques Dion, MD, of Emory University Hospital (Atlanta, Georgia), who discussed the procedure at SCVIR, only about 15 physicians were actively performing vertebroplasty two years ago. Now, the number has increased to more than 400, and thousands of patients have been treated. Medicare now provides reimbursement for the procedure in at least 26 states, although specific reimbursement practices vary from state to state. The procedure is clearly cost-effective, since it can be performed on an outpatient basis, saving the cost of an eight-day hospital stay.

Interventional neuroradiology grows

Interventional neuroradiology is a well-established segment of the peripheral intervention market, one that emerged in the mid-1980's with the introduction of the first variable stiffness microcatheters by Target Therapeutics, now a unit of Boston Scientific. Growth of the market has continued to be limited to a degree by the small number of physicians specializing in this field. However, new application segments may emerge that will not only contribute to continued strong market growth, but that also will attract participation by physicians from other specialties including cardiology and vascular surgery. Suppliers foresee a major opportunity for carotid angioplasty and stents, along with vascular protection devices to help prevent procedure-related embolization. In addition, there are major needs for improved approaches to stroke treatment, and interventional neuroradiology has much to offer in that area. Forecast U.S. sales of interventional neuroradiology products, including microcatheters, guidewires, ancillary devices, and embolization materials and coils, are shown in Table 4 on page 126.

The primary stents being used in carotid procedures at present are the Bard Memotherm, the Cordis SMART stent, and the Wallstent from Boston Scientific. Roubin is planning to begin using a new Cordis stent, the 18 SmartStent, for carotid procedures. In addition, EndoTex Interventional Systems (Cupertino, California) has received an investigational device exemption to begin trials with its NexStent carotid stent, a nitinol device that can be delivered via a 5F catheter. Guidant is a major investor in EndoTex. Embolic protection devices that may prove valuable for carotid stenting include the GuardWire device from PercuSurge (Sunnyvale, California); the AngioGuard device from Angioguard/Cordis; devices from Embolic Protection and MedNova (Galway, Ireland), and the TRAP from Microvena (White Bear Lake, Minnesota). Scion Cardiovascular (Miami, Florida) has described a new carotid stent, and the company also is developing the Angio-Pro intravascular protection filter.

An emerging area of applications in interventional neurology that may create additional opportunities for device companies is treatment of intracranial stenosis, a condition that can lead to stroke, and that also can cause significant degradation in quality of life. As discussed by J.J. Connors, MD, of Inova Fairfax Hospital (Falls Church, Virginia) at SCVIR, intracranial stenosis often leads to transient ischemic attacks (TIAs) lasting under 24 hours. However, according to Connors, if such attacks last for more than one hour, over 80% can be classified as an actual stroke. Between 20% and 30% of patients who suffer a TIA due to intracranial stenosis will have a stroke within two years, and 50% will have a stroke in three to five years. About 40,000 strokes per year are believed to be attributable to intracranial atherosclerosis. Often, patients who present with a TIA are treated with drug therapy, but according to Connors, a soon-to-be-published study will show that 45% of patients with intracranial stenosis fail medical therapy within 36 days, and many die or have a major stroke when they fail. Many of these patients are African-American and Asian, and have not been studied since major stroke therapy studies such as the NASCET trial mainly included caucasians.

Outcomes for patients with intracranial stenosis can potentially be improved substantially by the use of intracranial stents. Although some trials of intracranial angioplasty and stenting were performed in the early 1990s, techniques and devices were poor, according to Conners, and as a result outcomes were discouraging. However, with newer devices and better techniques, particularly the use of less traumatic balloon inflation techniques, it appears possible to achieve satisfactory results that provide significant symptomatic improvement. Long-term follow-up will demonstrate if the technique helps to prevent stroke. Connors is using the Medtronic S670 stents and the Equinox balloon catheter. However, the new INX family of stents is under development by Medtronic for neuro applications that should be useful in treating intracranial stenosis. The INX stents will be available in diameters of 2.5 mm, 3.0 mm, and 3.5 mm, with lengths ranging from 4 mm to 18 mm. In addition, Target/Boston Scientific will launch a new neuro balloon catheter shortly. According to the American Heart Association (Dallas, Texas), the prevalence of TIA in men is 2.7% for ages 65-69 and 3.6% for ages 75-79. For women, TIA prevalence is 1.6% for ages 65-69 and 4.1% for ages 75-79. Based on U.S. Census Bureau estimates, there are about 14.4 million men and 20.5 million women age 65 and older in the U.S.

Devices may transform dialysis access

Another area of innovation highlighted at the SCVIR conference is new devices for dialysis access. Two companies exhibited new subcutaneous access devices, similar to drug delivery ports, that are intended to improve hemodialysis access. Neither device is yet FDA-approved. Biolink (Middleboro, Massachusetts) is developing the Dialock Hemodialysis Access System, a titanium port connected to twin nickel/titanium-reinforced silicone catheters. The port contains two needle-accessible channels, each incorporating a valve assembly. The valves are opened by insertion of specially designed 15-gauge needle cannulas, and automatically close when the needles are withdrawn. The needles lock together when both are properly in place, helping to prevent accidental dislodgment. Complication rates compare favorably to currently available access systems, and patency is about 60% at one year vs. about 40% for PTFE grafts or less than 40% for A-V fistulas or catheters. Biolink plans to launch the Dialink later this year at a target price of $2,500 to $3,500.

A second device, the LifeSite from Vasca (Tewksbury, Massachusetts), is implanted subcutaneously in the upper chest, and accepts a 14-gauge needle for interconnection to the dialysis system. Two LifeSite systems are implanted for dialysis access. The LifeSite closely resembles a drug infusion port, is fabricated of titanium alloy and stainless steel, and is currently available in Europe for hemodialysis and peritoneal dialysis access. Vasca, a venture-funded firm, hopes to introduce the LifeSite in the U.S. by mid-year, and pricing is expected to be approximately $3,500 for two devices.

Dialysis access is an area where considerable improvement is needed. Worldwide, it is estimated that about 1 million patients are now undergoing dialysis. Access site complications are frequent, and generally must be treated with time-consuming thrombolytic therapy or catheter intervention. New access devices could well prove cost-effective if they can help to avoid thrombotic occlusions. Furthermore, such devices could facilitate the movement of dialysis treatment into the home, improving convenience and possibly improving patient outcome by allowing more frequent dialysis.

Peripheral embolization growth continues

Yet another attractive segment of opportunity in interventional procedures is peripheral embolization. Key application areas include uterine artery embolization for the treatment of uterine fibroids; chemoembolization for cancer and, perhaps, sterilization. New materials and embolization devices are under development that promise to improve the accuracy and efficiency of embolization of uterine fibroids that cause excessive bleeding and pain, and to provide more permanent occlusion than existing devices or materials. Dr. S. Hori of Rinku General Medical Center in Japan described new sodium acrylate-vinyl alcohol polymer particles that increase in volume by about fourfold when exposed to human serum. The particles can be delivered via catheter while still unexpanded, and can then be positioned as desired before being allowed to expand and set up in place. The particles have been used for uterine fibroid embolization with no accompanying pain or fever. Furthermore, it is possible to predict the degree of occlusion prior to allowing the particles to set up in the vessel.

Microtherapeutics (San Clemente, California) is developing new occlusion coils, comprised of a nitinol core covered with stainless steel, that provide significantly improved retention force and more reliable self-anchoring. Biosphere Medical (Marlborough, Massachusetts) is developing the Embosphere microspheres for embolization, and plans to launch the product in the U.S. during 2000. The Embosphere microspheres are precisely calibrated, and are made of an advanced copolymer that enables precisely targeted and controlled embolization. The initial target application is uterine fibroid embolization, but versions for neuro embolization also are under development.

Chemoembolization is another growing segment of IR applications. As shown in Table 5, typical revenues for an oncological IR practice can exceed those for a vascular surgery practice, and are considerably greater than for gynecological interventional practices. Chemoembolization is being applied primarily for the treatment of liver cancer, where the technique can allow treatment of some patients who cannot be treated surgically. Most patients have developed liver cancer due to metastasis from a primary colon cancer. There are about 500,000 patients in the U.S. under surveillance for colon cancer, and 130,000 to 150,000 new cases each year. Survival is increased about two-fold using chemoembolization as compared to systemic chemotherapy, since an approximately 35% increase in chemotherapy dose can be achieved. However, a negative factor is that the Health Care Financing Administration has recently retroactively denied reimbursement for hospital stays for chemoembolization procedures, and is now reimbursing the procedure as an outpatient chemotherapy procedure. The change has prompted efforts by physicians to develop techniques allowing chemoembolization to be performed on an outpatient basis.

According to experts who commented on the Genyx study, similar approaches have been demonstrated in the past, but concerns about DMSO toxicity and possible teratogenic effects of DMSO prevented widespread adoption. Although such issues may have now been resolved, there will likely be continued reluctance to adopt novel new methods for sterilization, particularly in light of the Dalkon Shield debacle of some years ago. However, the potential procedure volume and associated market opportunity is quite large, making this an attractive area for further development.