Skip main navigation

Angiographical Follow-Up After Radioactive “Cold Ends” Stent Implantation

A Multicenter Trial
Originally published 2002;105:550–553


    Background Radioactive stents with an activity of 0.75 to 12 μCi have shown >40% edge restenosis due to neointimal hyperplasia and negative remodeling. This trial evaluated whether radioactive Cold Ends stents might resolve edge restenosis by preventing remodeling at the injured extremities.

    Methods and Results The 25-mm long (15-mm radioactive center and 5-mm nonradioactive ends) Cold Ends stents had an activity of 3 to 12 μCi at implantation. Forty-three stents were implanted in 43 patients with de novo native coronary artery disease. Two procedural, 1 subacute, and 1 late stent thrombosis occurred. A restenosis rate of 22% was observed with a shift of the restenosis, usually occurring at the stent edges of radioactive stents, into the Cold Ends stents. The most severe restenosis occurred at the transition zone from radioactive to nonradioactive segments, a region located in dose fall-off.

    Conclusion Cold Ends stents did not resolve edge restenosis.

    Radioactive stents demonstrated a dose-related reduction of in-stent restenosis in animal trials.1,2 However, a dose-dependent delay in endothelialization of the stent was observed, thereby increasing the risk of thrombotic occlusions.3

    Radioactive stents with activities ranging from 0.75 to 12 μCi implanted in patients have shown a restenosis rate of >40%. Restenosis was primarily located at the stent edges and caused by neointimal hyperplasia and negative remodeling.4,5 The edges represent an area where tissue is subjected both to balloon-induced trauma and low-dose radiation.6,7 The aim of this trial was to evaluate whether extending the radioactive stent with nonradioactive “Cold Ends” might resolve edge restenosis by preventing remodeling at the injured extremities.


    Patient Population

    The 32P radioactive Cold Ends stent trial was a nonrandomized multicenter trial. Inclusion and exclusion criteria for enrollment in this trial were previously reported.6,7 The Medical Ethical Committees of all centers approved the study. All patients provided written informed consent. The trial was conducted according to GCP guidelines.

    Radioactive Stent

    The BXI Cold Ends Betastent (Isostent Inc) was 25-mm long with diameters of 3.0 and 3.5 mm. The 15-mm center segment was made radioactive by Phosphorus-32, whereas both 5-mm edges (the Cold Ends) were kept nonradioactive.2 The initial stent activity was measured and the date at which the radioactivity had decreased to 3 to 12 μCi (suitable for implantation) was calculated.

    Quantitative Coronary Angiography

    Quantitative coronary angiography (QCA) was performed before procedure, after procedure, and at 6-month follow-up by the Thoraxcenter Rotterdam angiographic corelab.5,6 Procedural success was defined as diameter stenosis (DS) <30%. Three regions of interest were analyzed: (1) target vessel, (2) target lesion, and (3) radioactive area. The radioactive area was defined as the segment, including only the radioactive segment of the stent. The target lesion was defined as the total stent area (radioactive segment and both Cold Ends). The target vessel was defined as the target lesion and the remaining segments of the treated vessel.

    QCA Methodology

    First, the minimum lumen diameter (MLD), reference diameter (RD), and DS were measured from side-branch to side-branch for the target vessel. Subsequently, the MLD, mean diameter (MD), and DS were measured for the total stent. Finally, the MLD, MD, and DS were measured for the radioactive segment. The total stent (approximately 25 mm in length) was delineated as the region of interest, by means of a proximal and distal cursor. Thereafter, by moving the cursors 1/5 of the total stent length (approximately 5 mm) away from the proximal and distal stent edges toward the center of the stent, the radioactive segment (approximately 15 mm) was identified. Restenosis was defined as >50% DS at follow up.

    Procedure and Follow-Up

    Patients received 250 mg aspirin and 10 000 international units heparin at the initiation of the procedure. The radioactive stent was implanted at a nominal deployment pressure of 8 to 18 atmospheres. The radioactive segment of the stent had to completely cover the original lesion length. Trauma at both edges was minimized by preferably performing direct stenting, avoiding postdilatation (see Table 1), and taking all previously published precautions.5,6 Patients received ticlopidine 250 mg BID or clopidogrel 75 mg daily for 6 to 7 months and aspirin ≥80 mg daily indefinitely after stent implantation.

    Balloon Inflation and Stent Deployment Data

    PredilatationStent DeploymentPostdilatation
    Nom. size balloon, mm2.68±0.513.21±0.373.48±0.56
    Balloon length, mm18±420±5
    Stentlength, mm25±3
    Balloon artery ratio0.90:11.16:1
    Stent artery ratio1.07:1
    Direct stenting16 (37%)
    No postdilatation10 (23%)

    Patients returned for 1- and 6-month clinical and 6-month angiographical follow-up. Death, Myocardial Infarction (MI), target lesion revascularization (TLR), target vessel revascularization (TVR), non–target vessel revascularization (non-TVR), subacute stent thrombosis (SAT), and late stent thrombosis (LT) were defined as previously reported.5,6

    Revascularization was performed on the basis of clinical symptoms and/or objective evidence of ischemia.

    Statistical Analysis

    Data are presented as mean±standard deviation. Continuous data were compared by means of the 2-tailed Student’s t test or linear regression when appropriate.


    Baseline Characteristics

    Baseline characteristics are shown in Tables 2 and 3.

    Baseline Patient Demographics and Anginal Status

        Male34 (79%)
        Female9 (21%)
    Risk factors
        Previous MI20 (47%)
        Diabetes mellitus3 (7%)
        Hyperlipidemia32 (74%)
        Hypertension16 (37%)
        Smoking16 (37%)
        Family history19 (44%)
    Angina pectoris
        Stable32 (74%)
        Unstable11 (26%)

    Baseline Lesion Characteristics

    LAD indicates left anterior descending coronary artery; LCx, left circumflex coronary artery, and RCA, right coronary artery.
    Treated vessel
        LAD20 (46.5%)
        LCx8 (19%)
        RCA15 (34.5%)
    Lesion type
        A15 (35%)
        B26 (60%)
        C2 (5%)
    Total occlusions2 (5%)
    Lesion length10±4 mm

    Procedural Data

    All 43 stents were successfully implanted in 43 patients. All patients were treated with a single Cold Ends stent. Five patients received additional nonradioactive stents due to procedural dissections. There were 3 transient occlusions: 2 procedural stent thrombosis and 1 SAT. The SAT occurred during an intravascular ultrasound evaluation of the stent performed 1 day after procedure for logistic reasons. None of these occlusions lead to an MI. All other procedures were uncomplicated. Procedural success was achieved in all but 2 patients. These patients had a DS of 33% and 30%.


    All patients were asymptomatic at hospital discharge. At 30-day follow-up, 93% remained asymptomatic and no additional SAT occurred. One LT, without total occlusion, occurred at 3.5 months after discontinuation of both clopidogrel and aspirin by the patient himself, leading to a non–Q-wave MI and a primary TLR. All patients returned for 6-month clinical follow-up (see Table 4). Angiographical follow-up was performed in 37 (86%) patients, and 6 patients refused. Of these 6 patients, 5 were asymptomatic and 1 had angina pectoris, as defined by the Canadian Cardiovascular Society Classification System, Class 2. Nine (21%) patients underwent a TLR: 7 due to restenosis and 2 due to recurrent angina without restenosis. Of these 2 patients, 1 had recurrent angina with an MLD of 1.71, and 1 had the LT, leading to an MI (see above) with an MLD of 1.69 mm. One restenotic patient was treated medically because he was both asymptomatic and had a negative stress test. Additionally, 1 TVR and 1 non-TVR were performed due to progression of atherosclerosis.

    Six-Month Clinical Follow-Up

    MACE indicates major cardiac events.
    Death0 (0%)
    Myocardial infarction1 (2%)
    Late occlusion1 (2%)
    Target lesion revascularization (TLR)
        PTCA9 (21%)
        CABG0 (0%)
    Target vessel revascularization (incl. TLR)10 (23%)
    Non–target vessel revascularization1 (2%)
    Total MACE11 (26%)
    Angina pectoris
        Stable8 (18%)
        Unstable (including MI)2 (5%)
        Angina free33 (77%)

    QCA Measurements

    QCA results and location of restenosis are presented in Tables 5 and 6. See Figure 1 for a patient example.

    QCA Analysis (n=37)

    RadioactiveTotal StentVesselRadioactiveTotal StentVessel
    %DS indicates % diameter stenosis, FU, follow-up; MLD, minimum lumen diameter; MD, mean diameter; Pre, pre-procedure; Post, post-procedure; Radioactive, radioactive segment (15 mm in length); RD, reference diameter; Total Stent, radioactive segment and both Cold Ends (25 mm in length); and Vessel, Total vessel (sidebranch to sidebranch). MLD, MD, RD, Acute gain, and Late loss measurements are in mm.
    RD or MD2.99±0.603.00±0.453.03±0.453.00±0.522.66±0.512.56±0.492.87±0.51
    Acute gain1.66±0.571.55±0.551.11±0.54
    Late loss0.65±0.570.79±0.580.49±0.49
    Late loss index0.42±0.410.60±0.640.55±0.83
    Restenosis, n (%)5 (14%)8 (22%)8 (22%)

    Location of Restenosis (n=8)

    *Both radioactive segment and proximal Cold End.
    Stent segment
        Diffuse*1 (3%)
        Proximal4 (11%)
        Distal1 (3%)
        Proximal and Distal2 (5%)

    Figure 1. Patient example at 6-month angiographic follow-up. The Wallstent (4.5 long), implanted in the ramus circumflexus, shows diffuse in-stent restenosis (A). The Cold Ends stent, implanted in the right coronary artery of the same patient, shows no restenosis in the radioactive segment but does show restenosis in both Cold Ends (B).

    Radiation Doses

    Stent activity level was 7.58±2.78 μCi at implantation, resulting in a calculated cumulative dose given over 100 days delivered to 1-mm depth from the stent surface of 53±19 Gy (Figure 2). There was no significant correlation between stent activity or cumulative dose and late loss or late loss index at follow-up.

    Figure 2. Two-dimensional dose representation for 12 μCi Cold Ends stent. Cumulative dose given over 100 days is shown (source, Isostent, Inc). Note that the total stent length is 2.5 cm. There is 1% activity on both (0.5 cm) Cold Ends.


    Edge restenosis occurs in >40% after radioactive stent implantation due to neointimal hyperplasia and negative remodeling and is caused by a combination of balloon injury and low-dose radiation at the stent edges.4–8 Therefore, it was hypothesized that edge restenosis could be resolved if the remodeling component of the edge restenosis phenomenon could be attenuated by the scaffolding properties of Cold Ends stents. The Cold Ends stents were manufactured as 15-mm isodose radioactive stents, extended with 5-mm nonradioactive Cold Ends segments. In order to control the extent of the barotrauma caused by balloon predilatation, direct stenting was preferably performed. Also, postdilatation of the stent was only performed if strictly necessary.

    What was observed was a shift of the restenosis, usually occurring at the edges of the radioactive stent,5,9 into the Cold Ends segment. More specifically, the most severe restenosis occurred at the transition zone from radioactive to nonradioactive segments, a region located in dose fall-off. Intravascular ultrasound analysis of a subgroup demonstrated significant neointimal in-growth distally and proximally from 2 to 3 mm within the radioactive segment extending on average into the Cold Ends segments of the stent.9 This phenomenon has also been recently demonstrated by the Thoraxcenter group in an experimental setup in which a half radioactive/half nonradioactive stent was implanted in coronary arteries of pigs. The peak level of neointimal hyperplasia was observed specifically in front of the last radioactive strut, thus in the region of dose fall-off. This strongly suggested that the combination of low-dose radiation in association with chronic injury by the stent is the main determinant of neointimal hyperplasia.10 Despite the prevention of the deleterious effects of negative remodeling seen at the stent edges in previous radioactive stent trials by the Cold Ends segments, a restenosis rate of 22% was observed. Therefore, Cold Ends stents did not resolve edge restenosis. One of the other options currently under investigation to reduce edge injuries is the use of square shouldered balloons, in which the entire balloon remains within the stent during stent deployment, thereby minimizing barotrauma at the proximal and distal edges. If edge restenosis is not reduced in that trial, it will give support to the hypothesis that low-dose radiation, or dose fall-off, is the main contributor of edge restenosis.


    Cold Ends stents did not resolve edge restenosis.

    The authors appreciate the efforts of the catheterization laboratory staff, the radiation staff, and the department of clinical epidemiology.


    Correspondence to Prof Patrick W. Serruys, MD, PhD, Dept of Interventional Cardiology, Thoraxcenter Bd 406, University Hospital Dijkzigt, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands. E-mail


    • 1 Carter AJ, Laird JR, Bailey LR, et al. Effects of endovascular radiation from a β-particle–emitting stent in a porcine coronary restenosis model: a dose-response study. Circulation. 1996; 94: 2364–2368.CrossrefMedlineGoogle Scholar
    • 2 Hehrlein C, Stintz M, Kinscherf R, et al. Pure β-particle–emitting stents inhibit neointima formation in rabbits. Circulation. 1996; 93: 641–645.CrossrefMedlineGoogle Scholar
    • 3 Carter AJ, Laird JR. Experimental results with endovascular irradiation via a radioactive stent. Int J Radiat Oncol Biol Phys. 1996; 36: 797–803.CrossrefMedlineGoogle Scholar
    • 4 Albiero R, Nishida T, Adamian M, et al. Edge restenosis after implantation of high activity 32P radioactive beta-emitting stents. Circulation. 2000; 101: 2454–2457.CrossrefMedlineGoogle Scholar
    • 5 Wardeh AJ, Knook AH, Kay IP, et al. Clinical and angiographical follow-up after implantation of a 6–12 microCi radioactive stent in patients with coronary artery disease. Eur Heart J. 2001; 22: 669–675.CrossrefMedlineGoogle Scholar
    • 6 Wardeh AJ, Kay IP, Sabate M, et al. β-Particle–emitting radioactive stent implantation: a safety and feasibility study. Circulation. 1999; 100: 1684–1689.CrossrefMedlineGoogle Scholar
    • 7 Albiero R, Adamian M, Kobayashi N, et al. Short- and intermediate-term results of 32P radioactive β-emitting stent implantation in patients with coronary artery disease: the Milan Dose-Response Study. Circulation. 2000; 101: 18–26.CrossrefMedlineGoogle Scholar
    • 8 Kay IP, Wardeh AJ, Kozuma K, et al. Radioactive stents delay but do not prevent in-stent neointimal hyperplasia. Circulation. 2001; 103: 14–17.CrossrefMedlineGoogle Scholar
    • 9 Kay IP, Wardeh AJ, Kozuma K, et al. The pattern of restenosis and vascular remodelling after cold-end radioactive stent implantation. Eur Heart J. 2001; 22: 1311–1317.CrossrefMedlineGoogle Scholar
    • 10 van der Giessen WJ, Regar E, Harteveld MS, et al. The edge-effect of coronary radioactive stents is caused by a combination of (chronic) injury and radioactive dose fall-off. Circulation. 2000; 102: II-667.Google Scholar