Author + information
- Received May 24, 2001
- Revision received January 7, 2002
- Accepted January 18, 2002
- Published online April 17, 2002.
- Andrew E Ajani, MBBS*,
- Ron Waksman, MD, FACC*,* (, )
- Dong-Hun Cha, MD*,
- Luis Gruberg, MD*,
- Lowell F Satler, MD, FACC*,
- Augusto D Pichard, MD, FACC* and
- Kenneth M Kent, MD, FACC*
- ↵*Reprint requests and correspondence:
Dr. Ron Waksman, 110 Irving Street, NW, Suite 4B-1, Washington, DC 20010, USA
Objectives The study assessed the influence of lesion length and reference vessel diameter (RVD) on recurrent restenosis after gamma intracoronary radiation therapy (ICRT) for in-stent restenosis (IRS).
Background Intracoronary radiation therapy reduces angiographic and clinical restenosis in patients with ISR. The impact of ICRT on challenging subgroups, such as long lesions and small vessels, has not been established.
Methods Six-month quantitative coronary angiography and clinical follow-up were conducted to evaluate the influence of lesion length and RVD in patients with ISR treated with ICRT who were enrolled in gamma radiation trials. Angiographic binary restenosis (>50% diameter stenosis) and clinical events were assessed in 311 patients treated with gamma ICRT and 105 patients who received placebo.
Results Baseline demographic, angiographic and procedural details were similar in the two treatment groups. The ICRT group had reduced binary restenosis in vessels of all sizes (30% vs. 66%, p < 0.001), with the most benefit seen in small vessels. A trend toward reduced restenosis with ICRT was found across all lesion lengths. At six months, major adverse cardiac events (MACE) were reduced in the ICRT group compared to placebo (34% vs. 71%, p < 0.0001), driven by reduced target vessel revascularization (27% vs. 71%, p < 0.0001). The independent predictors of angiographic restenosis include ICRT (OR [odds ratio] 0.16; CI [confidence interval] 0.10 to 0.28, p < 0.001), lesion length (OR 1.03; CI 1.01 to 1.05, p = 0.004) and RVD (OR 0.40; CI 0.23 to 0.67, p < 0.001).
Conclusions Intracoronary radiation therapy, compared to placebo, results in a significant reduction of angiographic restenosis across all vessel sizes, with a trend toward reduction of angiographic restenosis across all lesion lengths; this effect is seen predominantly in small vessels and diffuse lesions.
Postangioplasty restenosis is a major limitation of coronary interventions, and it has been significantly reduced with the widespread application of coronary stents (1,2). While coronary stenting has progressively evolved with improvement in stent design and deployment strategies, the treatment of in-stent restenosis (ISR) remains a challenge. Lesion length is recognized as a predictor of recurrent restenosis, with the highest rates found in diffuse lesions (>10 mm) (3–5). The incidence of restenosis is higher after stenting in smaller compared to larger vessels (6,7). Intracoronary radiation therapy (ICRT) reduces angiographic and clinical restenosis in patients with ISR (8,9). The impact of lesion length and reference vessel diameter (RVD) on the rate of recurrent restenosis in patients treated with ICRT has yet to be established. The purpose of this study was to assess the influence of lesion length and RVD on six-month angiographic and clinical outcomes in irradiated and placebo-treated patients with ISR.
Four hundred and sixteen consecutive patients enrolled in the Washington Radiation for In-Stent restenosis Trial (WRIST) series of gamma radiation studies (192Iridium), conducted at the Washington Hospital Center between February 1997 and June 2000, were selected for this study. These trials were designed to test the effectiveness of radiation as adjunctive treatment to conventional intervention of ISR. The selected cohort included patients treated with either radiation or placebo therapy who had completed six-month angiographic follow-up. Patients were included from the randomized, double-blinded WRIST and Long WRIST (long ISR lesions 36 to 80 mm) trials, and from gamma radiation registries including: Long WRIST High Dose (long ISR lesions, 36 to 80 mm in length using 18 Gy at 2 mm from the source); Plavix WRIST (six-month clopidogrel therapy postcoronary intervention and radiation); Compassionate WRIST (intracoronary localized radiation compassionate protocol for prevention of recurrence of restenosis); and WRIST X-over (patients who initially failed placebo therapy and were subsequently treated with radiation).
All studies involved an Investigational Device Exemption granted by the Food and Drug Administration, and they were approved by the Institutional Review Board and the Radiation Safety Committee at the Washington Hospital Center. Informed consent was obtained for all patients.
The inclusion criteria included ISR, lesion diameter stenosis >50%, RVD 2.5 to 5.0 mm, lesion length <80 mm and successful primary coronary intervention. Exclusion criteria included acute myocardial infarction (MI) <72 h, left ventricular ejection fraction (LVEF) <20%, angiographic visible thrombus, multiple coronary lesions and prior radiation therapy.
Radiation delivery system and dosimetry
In the WRIST series examined in this study, radiation was delivered via a closed end-lumen 5F noncentered catheter (Medtronic Vascular Interventional, Minneapolis, Minnesota), with patients receiving a nylon ribbon (0.003-in. diameter) containing different seed trains of either 192Iridium or placebo, positioned with at least a 4-mm overlap of the normal segments on each end. The radiation oncologist hand-loaded the ribbon from a lead container into the closed end-lumen catheter. Depending on the study protocol, the prescribed dose was 15 to 18 Gy at 2.0 mm from the surface of the source for vessels between 2.5 and 4.0 mm, and 15 to 18 Gy at 2.4 mm for vessels >4.0 mm in diameter. The mean specific activity of the seeds was 25.9 ± 4.1 mCi, with varying seed trains required to cover different lesion lengths.
Coronary intervention details
Interventional procedures included balloon angioplasty, excimer laser coronary angioplasty, rotational atherectomy and/or additional stent implantation. Device selection for the coronary intervention was based on lesion morphology and was at the discretion of the operator. A final angiogram was performed following radiation to ensure <30% final diameter stenosis. All patients received routine postangioplasty care and were discharged on aspirin (325 mg daily) and additional antiplatelet therapy, including ticlopidine 500 mg oral bolus and 250 mg twice daily continued for one month, or clopidogrel 300 mg oral bolus followed by 75 mg daily continued for either 30 days or six months (depending on the study protocol).
Data were prospectively recorded and forwarded to the Data Coordinating Center (Cardiology Research Institute Data Analysis Center at the Washington Hospital Center, Washington, D.C.). Clinical, angiographic and procedural details were recorded at baseline, in-hospital and at six-month follow-up (taken as within 220 days of the index procedure to ensure the entire cohort was included in the analysis). An adjudication committee independently reviewed all clinical events.
Quantitative coronary angiography (QCA)
All patients received six-month follow-up coronary angiography, with QCA performed at the Washington Hospital Center laboratory using the CMS–GFT system (Medis, The Netherlands). Angiographic measurements were performed on the entire artery segment covered by the radiation therapy. Late loss (mm) was defined as the change in lesion minimal luminal diameter (MLD) from the final to the follow-up angiogram. The arithmetic late loss index was defined as late loss/acute gain (increase in MLD from pre- to postintervention angiogram). Angiographic binary restenosis at six-month follow-up was defined as ≥50% diameter narrowing. This was determined both for the stented segment and the segment including the stent and its edges (within 5 mm of stent).
Death was defined as all-cause mortality. Myocardial infarction was defined as a total creatinine kinase elevation ≥3 times the baseline value and/or elevated creatinine kinase-MB fraction ≥20 ng/ml, with or without new pathological Q waves (>0.4 ms), in two or more contiguous leads. Target lesion revascularization (TLR) refers to the need for repeat percutaneous intervention or cardiac surgery due to restenosis at the site of the treated lesion. Target vessel revascularization (TVR) is defined as the need for repeat percutaneous intervention or cardiac surgery due to restenosis at the site of the treated lesion or a diseased segment in the target vessel distinct from the target lesion. Both TLR and TVR were driven by clinical signs of ischemia in the presence of angiographic restenosis. Major adverse cardiac events (MACE) were defined as death, Q-wave MI and/or TLR/TVR. Late total occlusion (LTO) was defined as angiographically documented total occlusion at the lesion site >30 days and within six months of the index procedure.
Continuous variables were expressed as means ± SD, and categorical data were expressed as percentages. To analyze the efficacy of ICRT, the Student ttest was used to compare continuous variables; chi-square test or the Fisher exact test was used to assess discrete variables. Stepwise logistic regression was used to determine independent predictors of angiographic binary restenosis and MACE (TVR). The variables used in this analysis included ICRT, age, gender, hypertension, diabetes, history of smoking, hypercholesterolemia, history of MI, history of coronary artery bypass grafting, history of percutaneous transluminal coronary angioplasty (PTCA), unstable angina, multivessel disease, LVEF, location of treated vessel, lesion length, RVD and the pre- and postintervention MLD. Independent variables are expressed as an odds ratio (OR) with 95% confidence intervals (CI). A value of p < 0.05 was considered statistically significant. The association of angiographic restenosis with lesion length and RVD was expressed by probability curves based on logistic regression.
Of the 416 patients enrolled in this cohort from the WRIST series, 311 were treated with active radiation and 105 with placebo. The baseline clinical, lesion and procedural characteristics were similar between the two groups (Table 1). Overall, 38% of patients were diabetic, 50% had multivessel disease and 53% had previous treatment of ISR, representing a high-risk cohort. Atheroablative devices were predominantly used for ISR, with 56% of patients treated with rotational atherectomy. Additional stenting was employed in 44% of patients. Radiation was delivered and tolerated in all ICRT patients (mean dose 14.6 ± 0.9 Gy, dwell time 21.9 ± 6.5 min).
The ISR lesions in both treated groups were predominantly diffuse (ICRT 22.9 ± 11.8 mm vs. placebo 25.0 ± 12.6 mm, p = NS), as outlined by QCA analysis (Table 2). There were no coronary perforations or aneurysms. In patients treated with adjunctive ICRT, follow-up MLD (1.55 ± 0.79 mm vs. 1.09 ± 0.68 mm, p < 0.001) and diameter stenosis (48 ± 26% vs. 61 ± 21%, p < 0.001) were favorable as compared to placebo. Patients receiving ICRT had reduced late loss (0.37 ± 0.78 mm vs. 0.84 ± 0.62 mm, p < 0.001), with an associated improved late loss index (0.48 ± 0.67 vs. 0.70 ± 0.65, p < 0.001), compared to placebo, demonstrating a powerful ability to reduce neointimal proliferation after coronary intervention. This translated into reduced angiographic binary restenosis with ICRT, both within the stent (24% vs. 61%, p < 0.001) and in the coronary segment involving the stent and its edges (30% vs. 66%, p < 0.001).
The ICRT showed a trend toward a reduction in six-month angiographic restenosis across all lesion lengths (p = 0.09) compared to placebo, especially in diffuse lesions (Fig. 1). In a similar analysis, ICRT resulted in reduced angiographic restenosis across vessels of all sizes (p = 0.03), with the effect seen predominantly in small vessels (<2.5 mm) (Fig. 2). Angiographic restenosis was also assessed in relation to both lesion length and RVD, demonstrating a higher rate in long lesions and small vessels (Fig. 3). It is this patient population that potentially derives the maximal benefit from ICRT. Analysis of Figure 3suggests that RVD is a more influential factor (than lesion length) for angiographic restenosis, which is supported by the statistical difference of the probability curves (Figs. 1 and 2).
Clinical events at six months among the entire cohort are outlined (Table 3). Event-free survival (freedom from death, MI and repeat revascularization) was greater for patients assigned to ICRT compared to placebo (68% vs. 29%, p < 0.001). This difference was driven by reduced rates of TLR (22% vs. 69%, p < 0.001) and TVR (27% vs. 71%, p < 0.001). There were no differences in rates of death or MI between each group. Late total occlusion was a phenomenon related to ICRT (5% vs. 0%, p = 0.015). The independent predictors of angiographic restenosis and MACE (TVR) at six months are outlined (Table 4). Radiation therapy, shorter lesion lengths and increased RVD (larger vessels) significantly reduce these end points.
The principal findings of this study are as follows: 1) ICRT using 192Iridium can be effectively and safely delivered to a high-risk population of patients with ISR; 2) ICRT, compared to placebo, shows a trend toward reduced angiographic restenosis across all lesion lengths, especially with diffuse lesions; 3) ICRT reduces angiographic restenosis in all vessel sizes, with the effect seen predominantly in small vessels (<2.5 mm); 4) a higher rate of angiographic restenosis was evident with a combination of long lesions and small vessels, and it is this patient population that appears to derive the maximal benefit from ICRT; 5) vessel size (RVD) is a more influential predictor of angiographic restenosis than lesion length; and 6) ICRT reduces late loss, angiographic binary restenosis within the stent (including stent edges), and clinical MACE driven by reduced TVR.
Mechanism of restenosis
Restenosis is a problem of exaggerated healing in the coronary artery after balloon injury, with smooth muscle migration and proliferation causing luminal compromise, in association with a lack of compensatory vessel wall dilation (10). Following PTCA, geometric remodeling is a critical response, with up to 73% of late loss due to chronic vessel constriction (11). Stents provide a scaffold for the vessel wall, and they largely eliminate this pathologic remodeling (12). Serial intravascular ultrasound (IVUS) studies in stented lesions show that most late loss is due to neointimal proliferation distributed over the length of the stent (13). Longer lesions with greater plaque burden provide an increased source of smooth muscle cells that will proliferate to form neointima (3,14). Stenting elicits a relatively higher proliferative response in small vessels. One explanation is that lumen gain in a small vessel requires a relatively higher degree of vessel wall stretch and thus creates more injury (7).
Efficacy of radiation
Use of ICRT has been shown to reduce lesion formation following arterial injury in a variety of animal models (15,16). Targeting the adventitia with an effective dose of radiation to inhibit the proliferation of adventitial myofibroblasts is one proposed key mechanism of reducing restenosis (17). Adjunctive ICRT after PTCA is efficacious, with the Scripps Coronary Radiation to Inhibit Proliferation Post Stenting (SCRIPPS) study (9)showing a 48% reduction in angiographic restenosis and a 58% reduction in composite MACE at three-year follow-up. Consistent with the clinical efficacy, IVUS analysis from the Gamma-1 study has shown that intimal hyperplasia is reduced with 192Iridium compared to placebo controls (0.8 ± 1.0 mm2vs. 1.6 ± 1.2 mm2, p = 0.007), when averaged over the length of the stent. In the original WRIST and SCRIPPS studies, ICRT with 192Iridium was the only predictor of freedom from angiographic or clinical restenosis (8,18). Our study is the first to show that both lesion length and RVD (in addition to ICRT) are important predictors in assessing risk of recurrent ISR.
Long lesions and small vessels
The mean lesion length in our cohort predominately represents diffuse ISR, which has been shown to be an independent predictor of TLR (OR 1.7, p = 0.038) following conventional therapy for ISR, predating ICRT (19). The IVUS analysis of radiated ISR lesions (mean length 56 ± 14 mm) from Long WRIST, when compared to WRIST, revealed an increased intimal hyperplasia cross-sectional area and a greater variability in neointimal response along the length of the stent (20). This supports the notion that long lesions are more challenging and may require higher dose prescriptions to eliminate the increasing probability of restenosis (Fig. 1). Long WRIST high dose was associated with a similar increase in mean intimal hyperplasia cross-sectional area (0.6 ± 1.5 mm2vs. 0.6 ± 1.1 mm2) compared to the Long WRIST group, despite longer lesions (mean 66 ± 17 mm) in the high-dose patients. Long ISR lesions also predict LTO and late thrombosis (OR 1.15, p = 0.04) especially in conjunction with radiation and additional stenting (21).
Endeavors to achieve the largest final MLD appear justified as larger vessels have less restenosis (known as the “bigger is better” hypothesis). The risk of a bigger lumen is more injury and exaggerated intimal hyperplasia (22). The limitation of ISR analysis in small vessels (RVD <2.5 mm) is that this subset has largely been excluded from radiation trials to date. In the current study, smaller vessels derive the maximum benefit from ICRT. The concept of freezing the final angiographic result with radiation and, in some instances, melting the residual stenosis (as seen in an IVUS subset of WRIST patients), is a potential explanation for our findings that smaller vessels have the greatest relative benefit.
These gamma radiation studies were not specifically designed to test the impact of lesion length and RVD on angiographic and clinical outcomes. Although this is a retrospective single-center experience, it is based on data acquired prospectively from the WRIST series that have similar entry criteria and clinical characteristics. Patients were subjected to an independent adjudication committee providing reliability to the observations.
Intracoronary radiation remains the foremost therapeutic strategy in treating patients with ISR. This therapy using 192Iridium improves six-month outcomes compared to placebo and results in a reduced rate of angiographic restenosis in all vessel sizes, predominantly small vessels (<2.5 mm), with a trend toward reduced restenosis across all lesion lengths, especially long lesions (>10 mm). The incidence of angiographic restenosis is reduced most notably in small vessels involving long lesions, and it is this patient population that potentially derives the maximal benefit from ICRT.
- confidence interval
- intracoronary radiation therapy
- in-stent restenosis
- intravascular ultrasound
- late total occlusion
- left ventricular ejection fraction
- major adverse cardiac events
- myocardial infarction
- minimal luminal diameter
- odds ratio
- percutaneous transluminal coronary angioplasty
- quantitative coronary angiography
- reference vessel diameter
- Scripps Coronary Radiation to Inhibit Proliferation Post Stenting
- target lesion revascularization
- target vessel revascularization
- Washington Radiation for In-Stent restenosis Trial
- Received May 24, 2001.
- Revision received January 7, 2002.
- Accepted January 18, 2002.
- American College of Cardiology Foundation
- Hoffmann R,
- Mintz G.S,
- Mehran R,
- et al.
- Bauters C,
- Banos J.L,
- Van Belle E,
- et al.
- Serruys P.W,
- Kay I.P,
- Disco C,
- Deshpande N.V,
- de Feyter P.J
- Akiyama T,
- Moussa I,
- Reimers B,
- et al.
- Waksman R,
- White R.L,
- Chan R.C,
- et al.
- Teirstein P.S,
- Massullo V,
- Jani S,
- et al.
- Post M.J,
- Borst C,
- Kuntz R.E
- Mintz G.S,
- Popma J.J,
- Pichard A.D,
- et al.
- Hoffmann R,
- Mintz G.S,
- Popma J.J,
- et al.
- Hoffmann R,
- Mintz G.S,
- Dussaillant G.R,
- et al.
- Kobayashi Y,
- De Gregorio J,
- Kobayashi N,
- et al.
- Waksman R,
- Robinson K.A,
- Crocker I.R,
- Gravanis M.B,
- Cipolla G.D,
- King S.B 3rd.
- Wiedermann J.G,
- Marboe C,
- Amols H,
- Schwartz A,
- Weinberger J
- Mehran R,
- Dangas G,
- Abizaid A.S,
- et al.
- Mintz G
- Waksman R,
- Bhargava B,
- Mintz G.S,
- et al.
- Faxon D