Cost-Effectiveness of Gamma Radiation for Treatment of In-Stent Restenosis
Results From the Gamma-1 Trial
Background— Recently, several randomized trials have demonstrated that intracoronary brachytherapy can reduce the rates of both angiographic and clinical restenosis in patients undergoing percutaneous coronary intervention (PCI) for in-stent restenosis. Whether this practice is cost-effective is unknown.
Methods and Results— Between December 1997 and July 1998, 252 patients with in-stent restenosis were randomized to receive brachytherapy or placebo after successful PCI as part of the Gamma-1 trial. We collected detailed resource utilization and cost data for each patient’s initial hospitalization and for 1 year after randomization. Compared with conventional treatment, intracoronary brachytherapy increased procedure duration, physician services, and equipment costs. As a result, initial costs were increased by nearly $4100 per patient ($15 724 versus $11 675, P<0.001). Over the 1-year follow-up period, brachytherapy reduced the need for repeat revascularization by 21% and reduced the need for bypass surgery by 44%. Although follow-up medical care costs were $2200/patient lower with brachytherapy, total costs remained higher at 1 year ($28 543 versus $26 737, P=0.46). In a sensitivity analysis that incorporated recent technical modifications and the use of prolonged antiplatelet therapy to prevent late thrombotic occlusion, follow-up cost savings increased to $3600/patient, and 1-year costs were slightly lower with brachytherapy ($26 352 versus $26 729, P=0.87). Subgroup analysis demonstrated significant cost savings in patients with diabetes and patients who did not undergo repeat stenting.
Conclusions— As performed in the Gamma-1 trial, coronary brachytherapy for in-stent restenosis improved clinical outcomes but increased 1-year costs compared with standard therapy. If late thrombosis can be eliminated, however, this technology has the potential to reduce overall medical care costs.
Received April 8, 2002; revision received May 9, 2002; accepted May 9, 2002.
Over the last decade, stenting has emerged as the dominant form of percutaneous coronary intervention (PCI) and is currently performed in ≈80% of all coronary interventions in the United States.1,2⇓ As a result, treatment of in-stent restenosis has become an increasingly frequent challenge for the interventional cardiologist. Recently, several randomized clinical trials have demonstrated that intracoronary brachytherapy can substantially reduce the rates of both angiographic and clinical restenosis in patients undergoing PCI for in-stent restenosis,3–5⇓⇓ which has led to Food and Drug Administration approval of several brachytherapy systems for treatment of in-stent restenosis.6
See p 646
Although these trials have clearly demonstrated the efficacy of brachytherapy for this challenging condition, the optimal role of this new procedure is unresolved, in part because of concerns about its cost.7 Compared with standard treatments for in-stent restenosis, coronary brachytherapy involves considerable additional expense related to specialized equipment, additional procedural time, and specialized physician services required for performance of radiation therapy. Whether these initial treatment costs are offset by subsequent savings due to reductions in the need for further revascularization procedures and related medical care is currently unknown. To evaluate the overall cost-effectiveness of coronary brachytherapy for treatment of in-stent restenosis, we therefore performed a prospective economic evaluation alongside the Gamma-1 trial, a randomized, double-blind, placebo-controlled trial of brachytherapy for patients undergoing PCI for in-stent restenosis.
Patient Population and Treatment Protocol
Between December 1997 and July 1998, 252 patients undergoing PCI for in-stent restenosis were enrolled in the Gamma-1 trial. Details of the study design have been described previously.3 Patients were eligible if they were undergoing repeat PCI for in-stent restenosis in a native coronary artery with a reference vessel diameter 2.75 to 4.0 mm and a lesion length ≤45 mm. The institutional review boards at each site approved the study protocol, and each patient provided informed consent before enrollment.
The target lesion was treated by means of conventional techniques, including high-pressure balloon dilation, rotational atherectomy or excimer laser, and additional stent placement, if necessary. Once an adequate angiographic result had been obtained, intravascular ultrasound was performed to permit radiation dose calculations as described previously.3 The radiation-delivery catheter was then positioned within the treated lesion, and the patient was randomized to receive either active or placebo brachytherapy. Brachytherapy was delivered by an 192Ir source that was manually advanced to the end of the radiation catheter by a radiation oncologist and left in place for the prescribed duration (generally 15 to 20 minutes). Patients assigned to placebo were treated identically, except that the “radiation” source used was an identical-appearing, nonradioactive ribbon. All patients were pretreated with aspirin and a thienopyridine and received intravenous heparin during the procedure to maintain an activated clotting time >300 seconds. After the procedure, aspirin was continued indefinitely, and ticlopidine (250 mg BID) or clopidogrel (75 mg QD) was prescribed for 8 weeks.
Assessment of In-Hospital Outcomes and Clinical Follow-Up
Case report forms concerning baseline demographic and clinical data, procedural details, and clinical outcomes during the initial hospitalization and 1-year follow-up period were completed by a research coordinator at each site and submitted to the data coordinating center (Cardiovascular Data Analysis Center, Boston, Mass). All end points (death, myocardial infarction, late stent thrombosis, and repeat revascularization) were reviewed by an independent clinical events committee who were blinded to treatment assignment. To limit contamination of clinical outcomes by the performance of routine angiographic follow-up (ie, the “oculostenotic reflex”), any repeat revascularization procedures (and their associated costs) judged not to have been clinically indicated were excluded from our economic analysis.
Summary of Principal Clinical Outcomes
As reported previously, the primary outcome of the Gamma-1 trial was the composite of death, myocardial infarction, or repeat revascularization of the target lesion at 9-month follow-up, which was reduced from 43.8% in the placebo group to 28.2% in the brachytherapy group (36% relative risk reduction; P=0.02).3 There were no significant differences in late mortality or myocardial infarction between the 2 groups. Thus, the benefit of brachytherapy was primarily due to a 42% relative reduction in the need for clinically driven target-lesion revascularization (24.4% versus 42.1%, P<0.01). Angiographic follow-up demonstrated parallel reductions in in-stent restenosis (21.6% versus 50.5%, P<0.005) and in-lesion restenosis (32.4% versus 55.3%, P=0.01).
Determination of Medical Care Costs
Medical care costs for the initial hospitalization and for the 1-year follow-up period were assessed by a combination of “bottom-up” and “top-down” methods as described previously.8
Cardiac Catheterization Laboratory Costs
Detailed resource utilization was recorded for each procedure, and the cost of each item was estimated on the basis of the mean hospital acquisition cost for the item in 2000. Costs of additional disposable equipment, overhead, and depreciation for the cardiac catheterization laboratory and for nonphysician personnel were estimated on the basis of the average cost per procedure at Beth Israel Deaconess Medical Center in 2000 and adjusted for actual procedure duration. For patients in the placebo group, procedure duration was corrected so as to exclude the time required for radiation planning, performance of placebo brachytherapy, and any associated waiting time. The cost of the equipment for the brachytherapy procedure was set at $2500 per procedure based on the manufacturer’s sales price for the delivery catheter and radioactive sources. We also assumed that the cost to set up the catheterization laboratory for brachytherapy ($25 000) would be amortized over a 5-year period, during which we estimated that a typical hospital would perform 50 brachytherapy procedures per year.
Other Hospital Costs
All other hospital costs were determined by top-down accounting methods based on each hospital’s annual Medicare Cost Report.9 Itemized bills were obtained for each patient’s initial hospitalization and any subsequent cardiovascular hospitalizations during the 1-year follow-up period. Hospital costs were determined by multiplying itemized hospital charges by the cost-center-specific cost-to-charge ratio obtained from the hospital’s Medicare Cost Report. Hospital admissions that were purely for the purpose of protocol-mandated cardiac catheterization were excluded from the economic analysis unless ischemia-driven coronary revascularization (as determined by the independent clinical events committee) was performed at that time. All costs were converted to 2000 dollars based on the medical care component of the Consumer Price Index.
For those admissions with missing billing information (n=49, 11%), nonprocedural hospital costs were imputed based on a linear regression model developed with the hospital admissions for which complete billing data were available (n=403). Independent variables for this model included length of stay, intensive care unit length of stay, vascular complications, and revascularization procedures (model R2=0.72).
Outpatient services related to management of cardiovascular disease were estimated by patient self-report at 3-month intervals during the follow-up period. Costs for these services were estimated based on 2000 Medicare reimbursement rates. Outpatient medications were not tracked, with the exception of thienopyridine therapy. Although all patients received such therapy after their index procedure (to maintain blinding), thienopyridine costs for the full duration of treatment were only assigned to patients in the active brachytherapy group.
Physician’s professional fees for inpatient services (admission and daily care), major cardiac procedures (coronary revascularization, stent placement, intravascular ultrasound, and diagnostic catheterization), surgical procedures, and outpatient services (office visits, echocardiography, and exercise testing) were estimated based on the Medicare Fee Schedule for Massachusetts. In addition, professional fees for radiation oncology and radiation physics services for planning and delivery of coronary brachytherapy were estimated based on review of applicable Current Procedural Terminology codes with 2 experienced radiation oncologists.
Discrete data are reported as frequencies, whereas continuous data are reported as mean±1 SD. Cost data are reported as both mean±SD and median values. Discrete variables were compared by Fisher’s exact test. Continuous variables (including cost data) were compared by Student’s t test. Although cost data are typically not normally distributed, the question addressed by the t test (“Is the cost difference between the 2 groups equal to $0?”) more closely approximates the point of view of a third-party payer or health plan administrator in evaluating a new technology than would a nonparametric comparison. We used 2-way ANOVA to test for an interaction between treatment effect and several prespecified patient- and treatment-related characteristics including diabetes, number of previous episodes of in-stent restenosis, and type of PCI. All analyses were performed according to the intention-to-treat principle.
Because the major benefit of brachytherapy was a reduction in the need for repeat revascularization, the primary end point for the cost-effectiveness analysis was the incremental cost per repeat revascularization avoided by brachytherapy compared with placebo. This cost-effectiveness ratio was calculated by dividing the difference in mean 1-year medical care costs for the 2 treatment groups by the difference in repeat revascularization rates (including both target-vessel and non-target-vessel procedures). Because the direct economic benefits of reduced repeat revascularization are already captured in the numerator of the cost-effectiveness ratio, it is important to note that the denominator of the ratio refers only to the health benefits of avoiding recurrent angina and the need for repeat revascularization. This approach is reasonable, because previous studies from our group and others have demonstrated that avoidance of coronary restenosis is associated with improved health-related quality of life10,11⇓ and improved quality-adjusted life expectancy,12 at least in the short run. Sensitivity analyses were performed to explore the impact of recent changes in brachytherapy techniques, adjunct pharmacology, and outcomes on the overall cost and cost-effectiveness of coronary brachytherapy for the study population. Bias-corrected confidence intervals for cost-effectiveness ratios were estimated by the bootstrap method, with 1000 resamplings of the study population.
Baseline clinical and angiographic characteristics were well matched for the 2 treatment groups (Table 1). The mean age of the study population was 60±11 years. Approximately 75% of the patients were men, 31% had diabetes mellitus, and 41% had multivessel coronary disease. Diffuse in-stent restenosis (lesion length >15 mm) was present in 60% of the study population, and 48% had had at least 1 previous treatment for in-stent restenosis.
Initial Treatment Costs
Table 2 summarizes selected resource utilization measures and costs for the initial revascularization procedures. After the time associated with planning and performance of placebo brachytherapy was excluded, procedure duration was ≈33 minutes longer for patients in the active brachytherapy group (145±64 versus 112±72 minutes, P<0.001). This difference was related to the required use of intravascular ultrasound for dose calculation in the brachytherapy group and the radiation dwell time itself (mean 21±5 minutes). Given the longer procedure duration, brachytherapy increased catheterization laboratory room and overhead costs by $239 (P=0.002), drug and disposable supply costs by $42 (P=0.10), and nonphysician personnel costs by $37 (P=0.002) per procedure compared with conventional therapy. Brachytherapy also increased procedure-related device costs (including the radiation-delivery catheter) by $2986 (P<0.001). Overall, initial procedure costs were approximately $3300 higher for patients in the brachytherapy arm than for standard care (P<0.001).
Other than the differences during the index procedures, there were no significant differences in initial hospital resource utilization between the 2 groups (Table 3). Professional fees were significantly higher for the brachytherapy group, however, which reflects the need for radiation physics and radiation oncology services to store and deliver the brachytherapy dose. Thus, initial hospital costs (including professional fees) were approximately $4100 higher for the brachytherapy group than for standard care ($15 724±4203 [median $14 805] versus $11 675±4920 [median $10 009]; P<0.001).
Follow-Up Medical Resource Utilization and Costs
Over the 1-year follow-up period, medical resource utilization tended to be lower in patients who received brachytherapy than in those who had conventional treatment (Table 4). Coronary brachytherapy reduced the need for any repeat revascularization by 21% (37.4% versus 47.1%, P=0.12), including a 45% reduction in the need for bypass surgery (11.5% versus 20.7%, P=0.06). There were less striking but consistent reductions in the need for rehospitalization as well. As a result, mean follow-up medical care costs were approximately $2200 per patient lower in the brachytherapy group than with conventional treatment ($12 818 versus $15 062, P=0.32). These savings were insufficient to fully offset the higher initial cost of brachytherapy, such that overall 1-year costs remained approximately $1800/patient higher for the brachytherapy group ($28 543±18 847 [median $22 677] versus $26 737±19 432 [median $20 670]; P=0.46).
Since the completion of the Gamma-1 trial, there have been several major changes in the technique and outcomes of γ-brachytherapy. In Gamma-1, intravascular ultrasound was required for radiation dosimetry. The Gamma-2 registry subsequently found that a simplified dosimetry regimen based on angiographic measurements alone produced clinical and angiographic outcomes identical to those seen in Gamma-1 (M. Leon, MD, written communication, September 2, 2001). Moreover, in the Gamma-1 trial, the effectiveness of brachytherapy was compromised by the occurrence of late stent thrombosis (beyond the usual 30-day window) in 5.3% of the brachytherapy group compared with 0.8% of the placebo group (P=0.07). Recently, however, several prospective registries involving >500 patients have demonstrated that avoidance of new stent placement combined with a 6-month course of aspirin plus clopidogrel antiplatelet therapy after γ-brachytherapy have reduced the incidence of late thrombosis to <1%.6,13⇓
We therefore performed a sensitivity analysis using updated assumptions as to both the costs and outcome of γ-brachytherapy. We assumed that intravascular ultrasound would not be required for dosimetry and that after successful radiation treatment, all patients would receive 6 months of combined antiplatelet therapy. Finally, we assumed that the use of prolonged antiplatelet therapy would reduce the incidence of late thrombosis to the rate observed in the placebo group. To perform this last sensitivity analysis, we randomly selected 6 of 7 episodes of late stent thrombosis from the brachytherapy group and eliminated their associated resource utilization and costs from our analysis. This effectively reduced the late stent thrombosis rate to 0.8% in both the placebo and brachytherapy groups.
Under these updated assumptions, brachytherapy resulted in substantial reductions in follow-up medical resource utilization compared with conventional treatment (Table 5). The need for rehospitalization was reduced by 23% (48.1% versus 61.1%, P=0.04), whereas the need for repeat revascularization was reduced by 30% (32.8% versus 47.1%, P=0.03), including a 60% reduction in the need for bypass surgery (8.4% versus 20.7%, P=0.007). As a result, follow-up medical care costs were approximately $3600 per patient lower in the brachytherapy group than for standard care ($11 446 versus $15 055, P=0.06), and there was no significant difference in aggregate 1-year medical care costs between the brachytherapy and placebo groups ($26 352±16 079 [median $21 247] versus $26 729±$19 392 [median $20 670]; P=0.87).
In our base case analysis, the incremental cost-effectiveness ratio for brachytherapy was $17 690 per repeat revascularization procedure avoided. This value is somewhat higher than the cost-effectiveness ratio for stenting compared with balloon angioplasty in both the Benestent II and Stent-PAMI trials ($10 000 to $13 000 per repeat revascularization avoided).12,14⇓ When reanalyzed according to our updated treatment and outcome assumptions, however, the cost-effectiveness of coronary brachytherapy for the Gamma-1 population was substantially improved. Under these conditions, coronary brachytherapy was an economically dominant strategy compared with conventional care that resulted in improved clinical outcomes at 1-year follow-up with comparable or slightly lower aggregate costs. Bootstrap simulation demonstrated that under these updated assumptions, the cost-effectiveness ratio for brachytherapy remained <$10 000 per repeat revascularization procedure avoided in 82.4% of samples.
Stratified analyses based on the primary Gamma-1 results demonstrated significant interactions between treatment assignment and overall medical care costs for several prespecified subgroups (Figure, A). For diabetic patients (n=79), brachytherapy reduced overall 1-year costs by $7998 (95% CI, $271 higher to $16 267 lower), whereas there was a trend toward increased overall costs in the nondiabetic subgroup (probability value for interaction, 0.005). Similarly, brachytherapy reduced overall costs by $11 226 (95% CI, $935 higher to $23 388 lower) for the subgroup of patients who did not undergo restenting at the time of treatment for in-stent restenosis (n=38), whereas 1-year costs tended to be somewhat higher in the group who did undergo restenting (probability value for interaction, 0.023). There were weaker trends toward reduced long-term costs with brachytherapy in patients with >1 episode of previous in-stent restenosis and for lesion lengths >15 mm. Of note, for each of the less favorable subgroups, the incremental cost-effectiveness ratio for brachytherapy compared with conventional care exceeded $100 000 per repeat revascularization procedure avoided in our base case analysis.
Under our updated treatment and outcome assumptions, stratified analyses demonstrated qualitatively similar treatment-subgroup interactions, with statistically significant net cost savings with brachytherapy in diabetic patients and patients who did not undergo restenting (Figure, B). Even under these more optimistic assumptions, however, the cost-effectiveness ratio for brachytherapy remained >$19 000 per repeat revascularization avoided in each of the less favorable treatment groups.
As the costs of health care continue to rise, both physicians and policy makers have become increasingly interested in understanding the cost-effectiveness of emerging medical technologies. This study represents the first formal economic evaluation of one such technology, intracoronary brachytherapy for treatment of in-stent restenosis. In the randomized Gamma-1 trial, we found that intracoronary brachytherapy increased initial treatment costs by approximately $4100 per patient compared with conventional treatment. These additional costs were primarily attributable to the brachytherapy device itself but were also related to the longer procedure duration, need for radiation oncology and radiation physics services, and specific equipment (eg, intravascular ultrasound) and adjunctive medical therapy (eg, extended thienopyridine antiplatelet coverage) required for the brachytherapy procedure. Over the 1-year follow-up period, brachytherapy improved clinical outcomes (mainly target-vessel revascularization) and reduced subsequent medical care costs by $2200 per patient. Nonetheless, these cost savings were insufficient to fully offset the higher initial cost of brachytherapy. At 1 year, aggregate medical care costs remained $1800 per patient higher with brachytherapy than with placebo.
Because brachytherapy was associated with higher 1-year costs in our primary analysis, we performed a formal cost-effectiveness analysis to assess the relationship between the clinical benefits of treatment and its overall cost. Because brachytherapy did not reduce long-term mortality and we did not directly assess quality of life in Gamma-1, we evaluated the cost-effectiveness of brachytherapy in terms of its main clinical benefit, a reduction in the need for repeat revascularization over a 1-year follow-up period. As performed in the Gamma-1 trial, the incremental cost-effectiveness ratio for brachytherapy compared with conventional treatment for in-stent restenosis was $17 690 per repeat revascularization procedure avoided over the 1-year follow-up period.
Although this cost-effectiveness ratio cannot be compared directly with standard “league tables” that are generally expressed in terms of cost per quality-adjusted year of life gained, some insight into the relative attractiveness of coronary brachytherapy may be derived from other economic studies in interventional cardiology. For example, the cost-effectiveness of elective coronary stenting (compared with balloon angioplasty) as performed in the Benestent II trial was approximately $11 000 to $13 000 per repeat revascularization avoided.14 Recently, we have reported that the cost-effectiveness of primary stenting (versus balloon angioplasty) for patients with acute myocardial infarction was approximately $10 000 per repeat revascularization avoided.12 Thus, as performed in Gamma-1, coronary brachytherapy was a relatively unfavorable use of scarce healthcare resources compared with these generally accepted applications.
Impact of Late Stent Thrombosis on Cost-Effectiveness
Since the performance of the Gamma-1 trial, late thrombosis has been recognized as an important complication of coronary brachytherapy. These events, which are rare after conventional treatment of in-stent restenosis, are presumably related to delayed or dysfunctional endothelialization of newly placed stents.3,15⇓ In the present study, we found that the rate of late stent thrombosis was a critical determinant of the cost-effectiveness of coronary brachytherapy. When we excluded those hospitalizations related specifically to the treatment of late thrombosis and its associated complications, follow-up cost savings with brachytherapy increased from $2200/patient to $3600/patient. Under these conditions, coronary brachytherapy was projected to reduce long-term medical care costs by approximately $300/patient, and its cost-effectiveness ratio was <$10 000 per repeat revascularization avoided in >80% of simulations.
The critical impact of late stent thrombosis on both the clinical and economic outcomes of coronary brachytherapy illustrates one of the main challenges in evaluating the cost-effectiveness of medical devices. Unlike medications, medical devices are typically in continuous evolution, and their results may depend on both patient characteristics and operator skill. As a result, although randomized trials remain the optimal method to eliminate selection bias in the evaluation of medical devices, they may not capture the full clinical or economic benefit of an evolving technology or procedure. In this case, we used sensitivity analysis based on updated assumptions regarding both the cost and outcomes of coronary brachytherapy in an effort to provide the most contemporary cost-effectiveness estimates. It must be recognized, however, that these results are somewhat speculative and require validation in a separate clinical trial. Unfortunately, once a device has achieved initial approval, such secondary trials are rarely performed in the same patient population. Alternatively, one could attempt to estimate cost-effectiveness based on observational registries or simulation models based on pooled clinical trial results. These methods are subject to numerous forms of bias, however, and may not necessarily produce results that are as valid as those based on primary clinical trial data.16
Clinical and Policy Implications
If future studies continue to confirm that late stent thrombosis can be eliminated either by avoidance of restenting or by extended antiplatelet therapy, our study suggests that use of coronary brachytherapy for treatment of patients similar to the Gamma-1 population will be economically attractive from a societal perspective and should not be denied on the basis of cost. The present study also suggests that certain patient characteristics, such as the presence of diabetes and the number of previous treatments, can be used to identify those patients who will derive the greatest clinical and economic benefits from brachytherapy. In fact, for many subgroups at high risk for recurrent restenosis, our study suggests that use of brachytherapy is associated with substantial long-term cost savings. For other patients (eg, initial presentation of focal in-stent restenosis), our findings suggest that initial use of brachytherapy is not particularly attractive (even under favorable assumptions) and might best be delayed until a second recurrence.
This study has several important limitations. As a placebo-controlled trial, the Gamma-1 trial may have underestimated the economic benefits of brachytherapy. In practice, it is possible that use of open-label brachytherapy will reduce the need for aggressive debulking or restenting at the time of treatment for in-stent restenosis, thus partially offsetting the treatment costs. In addition, the time horizon of our economic analysis was relatively brief, given the constraints of the clinical trial design and the available clinical follow-up. Thus, we cannot exclude the possibility that some of the economic benefits we observed might erode with further follow-up. Currently available data, however, suggest that the benefits of γ-brachytherapy are largely sustained through at least 3-year follow-up.17
This study was supported by a grant from Cordis, Inc (Warren, NJ).
Dr Leon serves as a consultant to Cordis, Inc. Dr Teirstein serves as a consultant to several brachytherapy companies and receives royalties from the sale of brachytherapy devices.
Williams DO, Holubkov R, Yeh W, et al. Percutaneous coronary intervention in the current era compared with 1985–1986: the National Heart, Lung, and Blood Institute Registries. Circulation. 2000; 102: 2945–2951.
Waksman R, Bhargava B, White L, et al. Intracoronary beta-radiation therapy inhibits recurrence of in-stent restenosis. Circulation. 2000; 101: 1895–1898.
Waksman R, White RL, Chan RC, et al. Intracoronary gamma-radiation therapy after angioplasty inhibits recurrence in patients with in-stent restenosis. Circulation. 2000; 101: 2165–2171.
Cohen DJ, Krumholz HM, Sukin CA, et al. In-hospital and one-year economic outcomes after coronary stenting or balloon angioplasty: results from a randomized clinical trial. Circulation. 1995; 92: 2480–2487.
Mark DB. Medical economics and health policy issues for interventional cardiology. In: Topol EJ, ed. Textbook of Interventional Cardiology. Philadelphia, Pa: WB Saunders; 1994: 1323–1353.
Cohen DJ, Taira DA, Berezin RH, et al. Cost-effectiveness of coronary stenting in acute myocardial infarction: results from the Stent-PAMI trial. Circulation. 2001; 104: 3039–3045.
Waksman R, Ajani AE, White RL, et al. Prolonged antiplatelet therapy to prevent late thrombosis after intracoronary gamma-radiation in patients with in-stent restenosis: Washington Radiation for In-Stent Restenosis Trial plus 6 months of clopidogrel (WRIST PLUS). Circulation. 2001; 103: 2332–2335.
Costa MA, Sabat M, van der Giessen WJ, et al. Late coronary occlusion after intracoronary brachytherapy. Circulation. 1999; 100: 789–792.
Teirstein PS, Massullo V, Jani S, et al. Three-year clinical and angiographic follow-up after intracoronary radiation: results of a randomized clinical trial. Circulation. 2000; 101: 360–365.