Applied Radiology

Sanjeev Ravipudi, MD is currently an associate of Missouri Cardiovascular Specialists as a staff Interventional Cardiologist at Boone Hospital Center in Columbia, MO. After receiving his MD in 1995 at the University of Missouri in Kansas City, he completed his Internal Medicine Residency at Baylor University Medical Center in Dallas, TX, and his General and Interventional Cardiovascular Fellowship at the University of Southern California in Los Angeles, CA.

With more than 1 million percutaneous interventions performed each year in the United States alone, the scope of the vexing problem of restenosis is staggering. To date, the development and clinical application of intracoronary stents has had the single greatest impact on the prevention of restenosis. Bare metal stents effectively prevent remodeling but are unable to counteract neointimal proliferation. Elucidation of downstream molecular events responsible for restenosis has made it easier to rationally select from potential therapeutic options. Recent attention has turned toward the development of novel systems for local drug delivery; drug-coated stents represent the foremost innovation in this area. The marriage of stenting and antiproliferative agents has the potential to eradicate restenosis. Although it does not seem that drug-eluting stents are the cure for restenosis, they do represent the next major advancement.

Restenosis has long been recognized as the "Achilles heel" of coronary angioplasty. ¹ It is a limiting factor that occurs in 12% to 48% of patients who undergo successful percutaneous coronary interventions. ²

Development of restenosis subjects patients to additional risks as well as the complications associated with repeat interventions or bypass surgery. Retrospective analysis of Emory University data showed that restenotic patients, when compared with patients without restenosis, had a higher incidence of angina (71% versus 39%) and target vessel reintervention (56% versus 4% at 6 months). Patients with restenosis were also found to have a lower freedom from myocardial infarction (88% versus 97% at 6 months, and 85% versus 93% at 6 years) and coronary artery bypass surgery (94% versus 99% at 6 months, and 78% versus 91% at 6 years), as well as a poorer survival rate (93% versus 95% at 6 years). ³ With over 1 million percutaneous interventions performed each year in the United States alone, the scope of this vexing problem is staggering. Increased awareness of unfavorable outcomes associated with restenosis has led to improvements in disease recognition and treatment.

Defining restenosis

Numerous definitions of restenosis have been reported. Gruntzig originally proposed the most commonly accepted definition of angiographic restenosis, a >50% luminal diameter stenosis at follow-up angiography. ⁴ Additional classification of in-stent restenosis into focal and diffuse is based on angiographic appearance. Focal restenosis is considered <10 mm in length and can occur at the articulation, margin, in the body, or be multifocal. Diffuse restenosis is considered >10 mm in length and can be intrastent, intrasegment/proliferative, or totally occlusive. Diffuse restenosis has been associated with a greater incidence of target vessel revascularization following repeat intervention. ⁵

Indices have been developed to measure and describe the mechanics of angiographic restenosis. Acute gain is the change in minimal luminal diameter (MLD) from baseline to immediately after percutaneous coronary angioplasty (PTCA). Late loss is the difference between the immediate postprocedure MLD and that measured at follow-up angiography. Net gain is the difference between follow-up MLD and baseline (acute gain ­ late loss). Late loss index refers to the percentage of early gain that is ultimately lost to restenosis (late loss/acute gain). ⁶ These indices vary depending on which devices are used for intervention. The acute gain and the net gain are greatest with stenting. In aggressive directional atherectomy, the net gain may approach that of stenting. Balloon angioplasty, laser atherectomy, and rotational atherectomy usually have only moderate net gain. The major angiographic predictors of restenosis are long lesions, multivessel lesions, chronic total occlusions, ostial lesions, angulated lesions (>45š), saphenous vein graft proximal and body lesions, and residual stenosis >30%. ^7-9 Angiographic predictors of in-stent restenosis are prior stenosis, long lesions, multiple stents, small vessels, and high-pressure inflation. ¹⁰

Clinical restenosis refers to the incidence of clinical events that ultimately lead to repeat revascularization of the index lesion. In general, patients who develop typical symptom recurrence gradually 1 to 6 months after angioplasty are more likely to have restenosis of the index lesion. These patients warrant further studies to define the presence of significant restenosis. Development of anginal symptoms within 1 month is more commonly associated with incomplete revascularization or abrupt vessel closure. Symptom occurrence beyond 6 months is likely a result of disease progression involving a different stenosis. ¹¹ Although patients who present with symptoms atypical of those prior to angioplasty infrequently have restenosis, ¹² the presence of angiographic restenosis is commonly associated with a return of preangioplasty symptoms. Approximately 2% to 50% of these patients remain asymptomatic. ^3,4,13-15 The long-term significance of silent restenosis is yet to be determined. However, data suggest that the clinical outcome of these patients is favorable. ¹⁶ The major clinical risk factors associated with an increased incidence of restenosis are diabetes and unstable angina. ^7-9

Mechanisms

The restenotic process has been described as a "healing" response to vascular injury with lesion development occurring gradually over a period of weeks to months (Figure 1). ^17,18 Elastic recoil, the difference in vessel cross-sectional area during and shortly after balloon inflation, occurs within minutes to hours of an intervention. ¹⁹ Approximately 15% of successful PTCA cases develop a >0.5 mm luminal loss at 24 hours due to recoil and thrombus, ¹⁸ with one-half of these progressing to significant restenosis. ²⁰

The vast majority of restenosis cases occur between 1 and 3 months due to neointimal hyperplasia and unfavorable remodeling (Figure 2). Occurrence of clinical or angiographic restenosis after 6 months is rare. ^2,4,18 Intracoronary stenting effectively limits elastic recoil and unfavorable remodeling. When compared with balloon angioplasty, stenting reduces the incidence of restenosis. ²¹ However, added stent-induced mechanical arterial injury along with a foreign-body response further promotes an acute and chronic inflammatory process within the vessel wall. ²² Subsequent elaboration of cytokines and growth factors (ie, platelet derived growth factor [PDGF], endothelial cell growth factor, transforming growth factor-beta [TGF-ß], and fibroblast growth factor [FGF]) induce multiple signaling pathways. This, in turn, activates vascular smooth muscle cell (VSMC) migration and proliferation and excess deposition of extracellular matrix proteins and fibrosis. ^23-25 In the context of restenosis, these factors induce an excessive neointimal fibroproliferative response composed mostly of smooth muscle cells (Figure 3). Progressing over a period of weeks to months, the fibromuscular overgrowth histologically comprises the majority of the restenotic lesion. ^2,17 In-stent restenosis is primarily due to neointimal hyperplasia.

Targeting therapy

Thrombus formation

The process of thrombus generation begins immediately at the site of vessel injury and persists for up to 4 weeks following intracoronary stent placement. ^23,26 Subsequent platelet activation and aggregation lead to the release of potent vasoactive agents and mitogens including PDGF, thrombin, and thromboxane A2 (TXA2). In addition to promoting intimal hyperplasia by propagating the secretion of PDGF and FGF, thrombin also induces VSMC migration and production of VSMC growth factors, such as serotonin. ²⁷ Thus, it is suggested that platelet activation and aggregation as a result of balloon or stent injury is one of the earliest causative events in restenosis. ²⁸

As a result, considerable effort has been directed toward the study of antithrombotic agents in the prevention of restenosis. Hirudin, an anti-thrombin factor, inhibits the mitogenic effect of thrombin and its synergistic interaction with serotonin. ²⁷ Along with low molecular weight heparins and glycoprotein IIb/IIIa inhibitors, hirudin theoretically offers a viable avenue for therapeutic intervention. However, clinical studies utilizing these agents have thus far failed to show a definite reduction in coronary restenosis. ²⁹

Inflammation

After balloon or stent injury, acute and chronic inflammatory cells participate in the ensuing process of arterial repair. Neutrophils infiltrate the vessel wall within 24 hours, followed by monocyte adhesion and infiltration. ³⁰ Observations have shown that the extent of acute inflammatory cell infiltration depends on the arterial substrate and the degree of vascular injury. For example, presence of a lipid core and deep-vessel wall injury is associated with greater inflammatory cell infiltration. ²³ It is surmised that inflammatory cells play an integral role in the exaggerated healing response seen in restenosis. ^23,24

Chronic inflammatory cells also contribute to the process with the release of cytokines (ie, interleukin-1 [IL-1] and tumor necrosis factor [TNF]) that stimulate VSMC migration and proliferation. ³¹ Additional observations of in-stent restenosis have shown lymphocyte, histiocyte, and giant cell infiltrate gathered around stent struts, which also further the activation of VSMC growth. The presence of these cells exhibits the importance of the body's immune response to the foreign stent struts in promoting restenosis. ^23,24

Systemic anti-inflammatory and immunosuppressive therapy, in theory, appear promising for treating restenosis. Small trials utilizing nonsteroidal anti-inflammatory drugs and tranilast have reported positive results ²⁹ ; however, no trial has proven any definitive benefit. Currently, exciting possibilities for preventing restenosis involve the use of anti-inflammatory agents with local drug delivery systems.

Vascular smooth muscle cell proliferation

Vascular smooth muscle cell growth is a key component of the pathophysiology of restenosis. Mechanical injury of the vascular wall induces migration of VSMCs from the media to the intima. ⁵ After migration, they proliferate and synthesize an extracellular matrix, the major component of the restenotic lesion. Vascular smooth muscle cells also release growth factors (ie, PDGF and FGF) and chemotactic agents that potentiate the proliferative process. ³² As a major participant in the development of restenosis, the VSMC is a prime target in the treatment of restenosis. Brachytherapy and drug-eluting stents are successful examples of antiproliferative therapies that target the VSMC.

Growth factors and cytokines

Vascular injury incites the release of mitogenic and chemotactic signals. These signals are mediated by growth factors and cytokines (ie, IL-1, tumor necrosis factor alpha [TNF-*], and interferon alpha [INF-*]) that modulate VSMC response. ³³ Signaling in response to these factors (Figure 4) is mediated through their specific cell-surface receptor and transduced by intracellular signaling pathways such as molecular target of rapamycin (sirolimus) (mTOR). ³⁴

A key regulatory kinase, mTOR is part of a signal transduction pathway in smooth muscle cells and lymphocytes. This enzyme is necessary for growth-factor and cytokine-signaling responses that initiate cell proliferation, which is essential in the development of neointimal hyperplasia. Inhibition of mTOR causes disruption of the cell cycle at the GAP 1-synthesis phase (Figure 5) via p27, promoting a cytostatic state. The protein p27 is an endogenous inhibitor of cyclinE/cdk2, which controls cell cycle progression to S phase. Additionally, mTOR inhibits protein synthesis and cell migration. ³⁴

Many intracellular signaling pathways have been elucidated and are targets of various treatment strategies. Rapamycin is produced by Streptomyces hygroscopicus and was found originally on Easter Island (Rapa Nui). ³² When bound to FKBP ₁₂ , it inhibits the activity of mTOR. Sirolimus also produces antimigratory and anti-inflammatory effects (Figure 6). ³⁴

Paclitaxel is extracted from the bark of the Pacific yew. It binds to microtubules and inhibits their depolymerization (molecular disassembly) into tubulin. Thus, paclitaxel blocks a cell's ability to break down the mitotic spindle during mitosis (cell division). With the spindle still in place, the cell cannot divide into daughter cells (Figure 7).

Coupled with the coated-stent drug-delivery system, these therapeutic agents have generated tremendous enthusiasm for the prevention of restenosis. Elucidation of downstream molecular events responsible for growth-factor or cytokine-mediated smooth muscle cell proliferation has made it easier to rationally select potential therapeutic options for the treatment of restenosis. The first generation of antiproliferative stents will likely be followed by further generations of coated stents with improved delivery systems, tailored dosing, and utilization of multiple agents providing a more effective blockade of the final common pathway that leads to neointimal proliferation.

Extracellular matrix and integrins

In addition to exaggerated VSMC proliferation, major components of a restenotic lesion also include excess deposition of extracellular matrix (ECM) proteins and fibrosis. Initially it was thought that the main function of the ECM was to provide a stabilizing structural lattice for the vascular wall. However, accumulating evidence indicates that matrix proteins also contribute to various biochemical and mechanical stimuli. ^35,36 Many extracellular components transmit their signals through specialized cell surface receptors known as integrins, and support a variety of cellular functions, such as migration and proliferation in response to mitogens. Other matrix components and growth factors have been found to work in concert with the ECM to regulate VSMC activity responsible for vascular development and remodeling. ³² Matrix metalloproteinases (MMPs) are a group of proteins essential for ECM dissolution. This family of enzymes has been implicated in VSMC and leukocyte migration during the restenotic process. ^35,36 Thus, the ECM represents another possible target for therapeutic intervention.

Primary prevention of restenosis

Bare metal stenting

To date, the development and clinical application of intracoronary stents has had the single greatest impact on interventional cardiology. In comparison with balloon angioplasty, stenting maximizes acute gain allowing for larger MLD and subsequently improving procedural success rates. Despite suffering a greater late loss, stenting has maintained its superiority over balloon angioplasty with reduced incidence of restenosis and target vessel revascularization (TVR) in most lesion subtypes ²¹ :

* In cases of acute myocardial infarction, data currently available supports stenting.

* In the setting of chronic total occlusion, trials demonstrate superior angiographic and clinical outcomes in patients randomized to stenting after successful recanalization.

* It is known that diabetic patients suffer higher rates of restenosis when compared with nondiabetic patients. Observational data suggest that stenting in diabetics is associated with lower restenosis rates when compared with angioplasty alone ^37,38 ; however, no randomized study to date has formally evaluated its efficacy.

* One subset that has failed to show benefit with stenting has been saphenous vein grafts.

Overall, based on the results of major clinical trials comparing angioplasty with stenting in native coronary vessels ^39,40 (Table 1), a strategy of stenting to prevent restenosis, reduce overall costs, and improve procedural success is considered appropriate in many patients. Coronary stenting is currently performed in 80% to 90% of patients undergoing percutaneous coronary intervention (PCI).

Brachytherapy

Used in other clinical situations to successfully stunt excess growth, vascular brachytherapy, the intraluminal delivery of radiation following angioplasty, was viewed as a viable solution to inhibit restenosis. The principal of radiation biology for prevention of restenosis is to induce apoptosis (programmed cell death), especially in those cells undergoing mitosis after vascular injury (Figure 8). This is accomplished by direct ionization or indirect interaction with other molecules to produce free radicals, which will subsequently damage critical targets (ie, DNA). The overall effect of radiation therapy is strongly dependent on the cumulative dose, dose rate, and cell cycle. ⁴¹

The main platforms developed for delivery of intravascular ionizing radiation use catheter-based systems with line-source wires, radioactive beads, gas/liquid-filled balloons, or stents that utilize beta or gamma ray emitters. Gamma rays are photons originating from the center of the nucleus; they differ from X-rays, which originate from the orbital outside of the nucleus. Gamma rays offer deep penetration but require excess shielding and prolonged dwell times when compared with beta emitters. Iridium-192 (192Ir) is currently the only gamma ray isotope in use. ⁴¹ Beta rays are high-energy electrons emitted by nuclei and contain too many or too few neutrons. These negatively charged particles have a wide variety of energies and a diverse range of half-lives. Beta emitters are associated with high gradients to the near wall, as they rapidly lose their energy to surrounding tissue. This quality limits the depth that brachytherapy can be delivered effectively. ⁴¹

Early trials evaluating safety, efficacy, and dosing of brachytherapy delivered after angioplasty or stenting in de novo coronary lesions showed promising results. ^42-44 The landmark SCRIPPS trial of 55 patients demonstrated an improvement in target vessel revascularization at 6 months with 45% TVR in the placebo group versus 12% TVR in the group treated by catheter-based gamma radiation. ⁴² The difference between the two groups persisted at follow-up even years after treatment. ⁴³ The beta energy restenosis trial (BERT) was a feasibility study of 23 patients that evaluated catheter-based beta radiation delivered after conventional PTCA. At 6-month follow-up, the rate of restenosis for the entire cohort was 17%. ⁴⁴

BETA-CATH, in contrast to the smaller preliminary trials, was a large prospective, randomized, blinded, placebo-controlled study that tested brachytherapy in de novo lesions after PTCA alone or with combined stent placement. ⁴⁵ In spite of a 38% reduction in in-lesion restenosis, there was no significant reduction in target vessel failure. This was explained by an increase in in-segment (injury and radiation segment + 5 mm at each end) stenosis and edge stenosis. Additionally, there was a higher rate of late thrombosis, especially in the stent plus radiation group. ⁴⁶

Similar disappointing findings were seen with radioactive stents. At dosing levels that inhibited in-stent restenosis, a 36% to 44% rate of edge stenosis was observed. This unique angiographic pattern is known as the "candy wrapper" effect. ⁴⁷

These observations identified two major problems associated with the use of this technology for primary prevention of restenosis: late thrombosis and edge stenosis. Late thrombosis is due to a delay in healing associated with radiation and is avoided easily with extended antiplatelet therapy following intervention. ⁴⁸ The main explanation for the incidence of edge effect is a combination of low/dropped-off dose delivery at the edges of the radiation source and an injury created by the device for intervention that is not covered by the radiation source (a geographic miss). ⁴⁵

Before brachytherapy can be considered for general use after de novo PCI, further evaluation is necessary to address the identified problems. Alternatively, brachytherapy has demonstrated definitive advantages in the treatment of in-stent restenosis and is FDA-approved (beta and gamma) for secondary prevention.

Drug-eluting stents

Bare metal stents effectively prevent remodeling but are unable to counteract neoinimal proliferation. The marriage of stenting and antiproliferative agents has the potential to eradicate restenosis. Several systemic approaches have been tested, but promising experimental results have not translated into clinical effectiveness. A proposed explanation for the repeated failure of clinical drug studies has been the inability of agents given systemically to reach sufficient levels in the injured arteries to effectively inhibit restenosis. Local drug administration would allow active drug delivery to the site of injury at the time of injury and the ability to achieve higher tissue concentrations of the drug. Recent attention has turned toward the development of novel systems for local drug delivery; drug-coated stents represent the foremost innovation. Current ubiquitous use of bare metal stents along with permanent scaffolding properties make bare metal stents an attractive reservoir for medication delivery.

In the development of drug-eluting stents (DES), a number of issues had to be addressed ^49,50 :

* Stents cover <10% of the target coronary segment and only a limited amount of drug can be loaded onto the stent (varies with stent design).

* The delivery system has to be tissue compatible.

* Coatings and antiproliferative agents should not induce inflammation or other detrimental responses.

* Stent expansion and sterilization should not affect coating or drug properties.

* Types of coating or drug binding should not interfere with pharmacokinetics or biological activities of the compound. The release of the drug must be predictable and controllable in terms of length of time and concentration.

* The coating/drug should not affect the mechanical properties of the stent.

Polymers are long-chain molecules consisting of small repeating units. Their ability to attach medications to metallic surfaces and facilitate prolonged drug delivery make polymers an appealing material to serve as coating matrices. Other coating considerations include naturally occurring substances, such as phosphorylcholine (PC), fibrin, cellulose, or albumin. ^49,50

In addition to resolving all of the preceding issues, optimal drug selection from a long list of potential agents has to be determined (Table 2). A major requirement of the selected drug would include the ability to inhibit key components involved in the restenosis process. As yet, trials employing stents coated with actinomycin-D, batimastat, heparin, ⁵¹ low-molecular-weight heparins, and dexamethasone have not shown a favorable influence on neointimal proliferation. To date, stents with or without polymer coatings utilizing antiproliferative medications (anti-neoplastic or immunosuppressive) have demonstrated the greatest success.

Sirolimus blended with a mixture of nonerodable polymers has been mounted on the BX Velocity stent (Cypher, Cordis, Johnson & Johnson; Warren, NJ). Early studies using these coated stents in animal models exhibited a significant reduction in neointimal proliferation. ²² Sousa et al, ⁵² in Brazil, then implanted two different formulations of sirolimus-coated stents (slow and fast release) in 30 patients with angina pectoris. On follow-up of 4 months, 8 months, and 1 year, no restenosis or adverse outcomes were observed. ^52,53

This led to the RAVEL trial, which prospectively compared the BX Velocity bare metal stent (Cordis) with the BX Velocity sirolimus-coated stent (Cypher, Cordis) for treatment of de novo coronary lesions among 230 patients in a randomized, double-blinded manner (Table 3). ⁵⁴ Even though treatment was limited to favorable lesion subtypes, the restenosis rate of zero and the event-free survival of 97% are impressive. ⁵⁴ To date, the sirolimus group remains patent and no long-term adverse events have occurred.

Advancing to the next step, SIRIUS is a multicenter, prospective, randomized, double-blinded trial that has enrolled 1101 patients in the United States with focal de novo coronary artery lesions. Patients were randomized to receive either a bare metal stent or a sirolimus-coated stent. Preliminary results of 400 patients were recently presented in Paris at Euro PCR. ⁵⁵ Results are still impressive; in-stent restenosis was 2% versus 31% in the control group with no significant differences in other major adverse events. Slightly dulling the brilliance of a significant reduction in in-stent restenosis, investigators reported an observed increase in in-segment restenosis of 9% versus 32%. Thoughts accounting for this finding included possible geographic miss, inadequate stent coverage, and possible need for higher medication delivery at the edges. Again, this is preliminary data and no definite conclusions can be made from these early observations. As anticipated, the restenosis rates are not zero, but remain extremely low and represent a hopeful advance in the treatment of restenosis. Also important to note is the lack of thrombosis, aneurysms, or late malaposition, which are some of the possible delayed complications.

TAXUS is a family of trials on paclitaxel polymeric-coated NIR Conformer and Express stents (both stents from Boston Scientific; Natick, MA). TAXUS I randomized 61 patients to a NIR Conformer slow-release drug-coated or non-drug-coated stent. Zero percent restenosis rates were observed in this early safety study. ⁵⁶ TAXUS II evaluated efficacy and enrolled 532 patients to receive paclitaxel-coated NIR Conformer stents with slow- and moderate-release formulations. ⁵⁶ Preliminary 30-day observational data shows 0% restenosis in both groups. The rate of target lesion revascularization (TLR) was 0.4% and 0% in the slow- and moderate-release groups, respectively, and was due to subacute thrombosis in both cases. Results should be available later this year. ⁵⁶ TAXUS III is a feasibility trial that enrolled 30 patients with in-stent restenosis, a less favorable treatment group. Patients were treated with paclitaxel-coated slow-release NIR Conformer stents. Cumulative 10-month major adverse cardiac event (MACE) rates for 28 of the 30 TAXUS III patients included 6 TLRs, 1 coronary artery bypass graft, 1 periprocedural non-Q-wave MI, 0 stent thromboses, and 0 deaths. Two of the TLRs targeted restenosis in gaps between two stented segments, while two others were performed to improve stent apposition to the artery wall. A fifth TLR was warranted after a bare stent adjacent to two paclitaxel-eluting stents became occluded, while the sixth TLR was symptom-driven. ⁵⁶ TAXUS IV and V are pivotal United States studies that are currently underway. They will be enrolling 1600 patients with de novo or in-stent restenosis lesions, respectively. Patients with de novo lesions will be randomized to moderate-release paclitaxel-coated or non-coated Express stents. Patients with in-stent restenosis will be randomized to PCI with a moderate-release paclitaxel-coated Express stent or brachytherapy. ⁵⁶ Enrollment is complete and results will be anticipated eagerly.

Cook has combined its paclitaxel coating with their Supra G and V-Flex plus stents (Cook Inc.; Bloomington, IN) in the ASPECT ⁵⁷ and ELUTES ⁵⁸ trials. ASPECT randomized 177 patients in Asia to high-dose, low-dose, and no paclitaxel coating. Restenosis rates were 4%, 12%, and 27%, respectively. ⁵⁷ ELUTES assessed the safety and efficacy of four doses of paclitaxel in 192 patients. The rate of restenosis for the bare stent group was 34%, 33% for the coated stents with the lowest dose, then 26%, 23%, and 14% for respective ascending doses. ⁵⁸

Many other stents, drug coatings, and indications are being tested actively. The DELIVER trial is investigating the ACHIEVE drug-coating (Cook Inc.) on a multilink stent platform (Guidant Corp.; Indianapolis, IN). ⁵⁹ Numerous potential future indications are also being evaluated, including use in patients with diabetes, saphenous vein grafts, chronic total occlusions, left main artery disease, complex/bifurcation lesions, multivessel disease, in-stent restenosis, small vessel disease, and peripheral arterial disease.

Although it does not seem that drug- eluting stents are the cure for restenosis, they do represent the next major advancement. Future advancements in understanding restenosis, drug delivery systems, and drug therapies, including combination therapy, will likely further the success of the technology. At present, the next major hurdles are the repercussions following widespread use and the possible complications that may be revealed on long-term follow-up.

Secondary prevention of restenosis

Repeat intervention for the treatment of in-stent restenosis with or without rotational atherectomy and cutting balloon is effective in acutely restoring luminal patency and dimensions. However, high rates of recurrent restenosis continue to plague standard treatment options.

At present, brachytherapy is the only treatment available demonstrating significant reduction in recurrent in-stent restenosis rates. Multiple trials have supported the use of gamma radiation. Key trials, including SCRIPPS, ^42,43 WRIST, ⁶⁰ GAMMA 1, ⁶¹ and ARTISTIC, ⁶² all showed definitive evidence that adjunctive brachytherapy after PCI for in-stent restenosis significantly reduces recurrent stenosis and recurrent revascularization rates. ^41,63 Similar support exists for beta radiation. BETA WRIST ⁶⁴ demonstrated a reduction in restenosis rates that approached that of gamma. ^41,63

The current debate regarding brachytherapy involves the choice between gamma and beta radiation. Gamma radiation has been demonstrated conclusively to be effective in the treatment of in-stent restenosis. Its drawbacks remain those associated with increased radiation exposure, the need for special shielding, and slightly longer dwell times. Beta radiation is unquestionably easier to use, which is its greatest advantage. Currently both are FDA approved for use in in-stent restenosis.

Future trial results of drug-eluting stents for the treatment of in-stent restenosis are greatly anticipated. It is possible that favorable results for drug-eluting stents will greatly limit the need for brachytherapy. However, as we have seen in early trial results, the restenosis rate associated with drug-eluting stents is not yet zero. Until that occurs, brachytherapy will continue to have this niche role.

Conclusion

At the start of the millennium, the problem of restenosis remains unresolved. ⁴⁹ As we gain a better understanding of the complex process involved, more effective targeted therapies can be developed. Although it appears that the current generation of drug-eluting stents have not eliminated the problem, there is little doubt that future generations have the potential to do so. Advancements in delivery platforms and drug therapies will help refine this technology. Future directions, generating just as much excitement, include genetic therapy. Although we have not yet discovered a panacea, we have never been so close. ⁴⁹