- Email: [email protected]
Design and dosimetry of a novel 90y beta source to prevent restenosis after angioplasty1
Int. J. Radiation Oncology Biol. Phys., Vol. 46, No. 1, pp. 249 –255, 2000 Copyright © 2000 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/00/$–see front matter
PII S0360-3016(99)00350-8
PHYSICS CONTRIBUTION
DESIGN AND DOSIMETRY OF A NOVEL 90Y BETA SOURCE TO PREVENT RESTENOSIS AFTER ANGIOPLASTY KONRAD MU¨ CK, PH.D.,* WERNER SCHMIDT, PH.D.,† PAUL WEXBERG, M.D.,‡ WALTER GO¨ RZ,* GERALD MAURER, M.D.,‡ AND MICHAEL GOTTSAUNER-WOLF, M.D.‡ *Austrian Research Center Seibersdorf, Seibersdorf, Austria; †Donauspital Vienna, Institute for Radiooncology, Vienna, Austria; and ‡ University of Vienna, Department of Cardiology, Vienna, Austria Purpose: Post-dilatation irradiation of the vessel wall is currently under investigation for prevention of restenosis after balloon dilatation. For the irradiation, special sources were designed for animal experiments which would give equivalent irradiation conditions and doses to the vessel wall that would later be employed for human application. Methods and Materials: For the planned irradiations, a specially designed yttrium-wire of 0.45-mm diameter coated with a thin shrink tube to prevent contamination was deployed. Several leakage tests applied before and after application proved that the irradiation source was leakproof. Dosimetry was performed by using 0.1-mmthick thermoluminescent dosimeters (TLD-100) calibrated against a primary standard. A shielding transport and application container was designed to facilitate the handling of the source during use, while reducing exposure of the medical personnel. Results: The designed source proves to be flexible for the insertion into proximal coronary vessels, and positioning at the site of stenosis. It provides an optimum protection of the animal and requires little radiation protection efforts on behalf of the medical staff. Dosimetric calculations and measurements showed that a centering of the source inside the vessel could be achieved with a maximum deviation of 50% between maximum and average dose levels. Conclusion: A yttrium-90 beta brachytherapy source was designed which provides high flexibility within proximal coronary arteries, ensures an adequate centering inside the artery, and provides irradiation conditions to the vessel wall of the experimental animal comparable to the application inside a human artery. © 2000 Elsevier Science Inc. Brachytherapy, Yttrium-90, Coronary artery disease, Angioplasty, Restenosis.
INTRODUCTION Atherosclerotic cardiovascular diseases are the main causes of death in the western world. Since the late 1970s, percutaneous transluminal coronary revascularization (PTCR) has become a potent alternative to surgical revascularization in patients with coronary artery disease (1). Balloon dilatation of atherosclerotic stenoses is initially successful in over 90%, with a low mortality rate of less than 1%. However, renarrowing (restenosis) occurs in 30 – 50% of patients, as a response to the PTCR 3– 6 months after the intervention (2). One of the main mechanisms of this phenomenon is vascular smooth muscle cell proliferation (3). Several devices (directional atherectomy, laser, rotational atherectomy, and intravascular stents) and pharmacological approaches (aspirin, anticoagulants, Ca-channel blockers, beta-blockers, angiotensin-converting enzyme inhibitors, fish-oil, IIb/IIIa inhibitors, etc.) have been tested in various clinical trials to reduce restenosis rate (4). Only
stents and blockage of the IIb/IIIa receptor showed favorable effects. However, the restenosis rate after sucessful PTCR remains at 20 –30% (5, 6). Recently, the antiproliferative effect of ionizing radiation has been proven to reduce neointimal growth up to 70% after experimental PTCR in animals and in clinical trials (7). An optimal intravascular brachytherapy device must be: (a) low-profile, flexible, and steerable, to pass tortuous vessel segments; (b) low in radiation exposure to the patient and operator; and (c) effective in reducing smooth muscle cell growth following the intervention in order to reduce restenosis. Beta sources of yttrium-90 (90Y) were selected to be applied for the human application (8) because they possess several advantages as compared to gamma radiation (9). They will be applied both in the proposed animal experiments and in human treatment. With beta sources, the determination of their absolute activity is not as easily achievable as with gamma sources; however, it will be shown that this is not detrimental to the
Reprint requests to: Konrad Mu¨ck, Ph.D., Austrian Research Center Seibersdorf, A-2444 Seibersdorf, Austria. Tel: ⫹⫹43 2254 780 3200; Fax: ⫹⫹43 2254 780 3206; E-mail: [email protected]
Supported by a grant of the Ludwig Boltzmann Institute for Cardiovascular Research. Accepted for publication 6 August 1999. 249
250
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 46, Number 1, 2000
Fig. 1. Design of beta treatment course.
application. A method to precisely ensure the required exposure values is described in this paper. The aim of the present study was to design and evaluate a novel intravascular beta-emitting brachytherapy device fulfilling the above-mentioned requirements. For the animal experiments ongoing at the Department of Cardiology of the University of Vienna, using New Zealand white rabbits, special radiation sources were designed and constructed to obtain irradiation inside the animal artery equivalent to that to be applied in the human treatment. This paper describes the design of the source to ensure its flexibility and to limit the irradiation time to less than a few minutes inside the artery, the centering inside the vessel, and the dose and its depth distribution to be comparable to later human applications. METHODS AND MATERIALS Design of brachytherapy device A source was designed to assure high flexibility, as well as optimum centering, inside the vascular vessel to obtain a high-dose homogeneity of irradiation. It consists of 5-mm long segments of an 90Y wire of 0.45-mm inner diameter positioned inside a shrink tube of 0.9-mm outer diameter (Fig. 1). This source is inserted through a 5-French catheter (1.65-mm diameter). To obtain sufficiently short irradiation times (⬃ 2– 4 min) inside the vessel, a nominal source activity of approximately 4 GBq is required. This activity is obtained by activating the source at the reactor of the Austrian Research Center Seibersdorf (ARCS) at a flux of 6.1013 cm-2 s-1 for 60 h. After activation, the source is stored for 24 h until the 3.19-h half-life 90Y has decayed to a sufficiently low level. The source is then placed inside the shrink tube according to a special patented technique (10) developed at ARCS for 192 Ir wires used in cancer brachytherapy. The source was attached to the anterior end of a standard 0.014-inch guide wire (standard length 175 cm, Fig. 1) by shrinking the shrink tube, and additionally secured with glue (LOCTITE 407). Flexibility of source The flexibility of the source was tested according to standard tests applied to catheters. The devices to be tested were fixed in a positioning tube in such a way that half of the balloon or the source, respectively, was outside the distal end of the guiding catheter (Fig. 2). The devices were
warmed to 37°C in a water bath for 2 min. The force to bend the device by 30° was measured with a digital scale (Satorius R160P). For comparison purposes, a standard guidewire (0.014 inch; Scimed, Naticle, MA) and two standard angioplasty balloons (brand y and brand x) were tested by the same device. Because of asymmetry, each device was tested in the pre-bent direction and the 180° reverse direction. Each experiment was repeated 10 times. Dosimetry Absolute dosimetry was performed by EXT-RAD-TLDs (1 ⫻ 1 mm2, 40 mg/cm2, thermoluminescent dosimeters (TLDs) fixed on a CAPTON-foil; Bicron). A special annealing (on a PC-driven PTW THELDO oven) and evaluation cycle (on a Bicron 3500 system) was established (11). Maximum temperature in both cases was 300°C. Annealing took about 20 min, with temperatures falling from 300°C to 80°C in 1-min increments; during readout the TLDs were heated to 300°C in 1 min, with a constant temperature rise of 5°C/s. The reproducibility of the area of the glowcurve was ⬎ 97%, the rising sensitivity due to incomplete annealing (maximum allowed temperature 300°C; in particular, after application of doses ⬎ 3 Gy) was compensated by individual calibration of the TLDs. The TLDs were calibrated both against 60Co sources
Fig. 2. Setup for testing the flexibility of the source.
Design and dosimetry of
90
Y beta sources
● K. MU¨ CK et al.
251
Fig. 3. Beta source and shielding container at the application site.
calibrated by the Austrian National Bureau of Standards and a secondary standard 90Sr/90Y reference source of the Physikalisch-Technische-Bundesanstalt, FRG (Beta Secondary Standard BeSS2 (12)). The ratio of the areas under the glowcurves for 1 Gy 60Co to 1 Gy 90Sr/90Y was ⬍ 1.05, which demonstrates the qualification of thin TLDs for beta dose measurements and that, as with gamma measurements, a reliable dose at fixed distances from the source center may be measured by placing thin TLDs in given positions from the axis and determining the thermoluminescence response, thus establishing a fixed relation between reference dose and beta emission. Depth-dose distributions perpendicular to the source axis were determined according to AAPM specifications (13) at specified distances up to 10 mm from the source. Also, absolute dose measurements were performed with TLDs. For depth-dose measurements, PMMA-sheets or polycarbonate foils (Makrofol G; Bayer) of different thicknesses were positioned between the TLDs. A special beta detector system was designed, consisting of a beta-plastic scintillator of 10-mm diameter and 5-mm height at a fixed distance from the source. The detection system is routinely checked with a standard 90Sr-source. The reproducibility of the system was extensively tested and is ⱖ 95%. The beta detector system is calibrated for dose measurements by use of the above-mentioned TLD calibration measurements. By this calibrated system, the actual beta emission of the activated source is determined before application, and the irradiation time is set accordingly.
To permit easy handling of the source and ensure a low radiation exposure of the medical staff during the application, as well as to facilitate the transport from the reactor to the application room, a special transport and application container was designed and built at ARCS. It consists of a shield of 10-mm plexiglass and 15-mm Pb coated by a thin Fe cover (Fig. 3). Despite its low weight of only 1.6 kg, the container gives excellent shielding.
Radiation protection issues After attaching the guidewire and heat shrinking the tube, a leakproof source is obtained. The tightness of the source was tested according to ISO-standard 9978 (14) by boiling the source in a water-alcohol solution for 1 h. Then the source is placed in the shielding/transport container and transferred to the hospital.
Dosimetry Results of depth-dose distribution measurements perpendicular to the source axis determined according to AAPM specifications are shown in Fig. 4. Resulting depth profiles show a decrease in dose by about a factor of three per 1-mm depth into the vascular tissue, both for an ideally centered source and one positioned close
RESULTS Flexibility The flexibility of the irradiation source, which is enabled by the special construction of the source in cylindrical segments, was tested by determining the deviation force required to bend the device to an angle of 30° in an experimental setup as shown in Fig. 2. Forces required for bending the brachytherapy device, as well as the two types of catheters and the guidewire, by 30° are summarized in Table 1. As expected, the lowest force is needed to bend the guidewire (1.02 ⫾ 0.02 mN in one direction and 1.28 ⫾ 0.02 mN in the opposite). The deflated balloons required a substantially greater force (4.5 to 6.8 mN, respectively), while in the opposite side, the required force in one case amounted to 16.9 ⫾ 0.5 mN. The bending force required for the inflated balloons was around 22– 40 mN (5–7 times stiffer than noninflated). The new brachytherapy device showed a comparable bending force of 35.1 ⫾ 0.9 mN and 37.6 ⫾ 0.8 mN in the opposite direction.
252
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 46, Number 1, 2000
Table 1. Results of flexibility test Bending force required for 30° angle (mN) Device
Test condition
Orientation (°)
Mean ⫾ SD
Minimum
Maximum
—
0 180 0 180 0 180 0 180 0 180 0 180
1.02 ⫾ 0.02 1.28 ⫾ 0.02 4.5 ⫾ 0.2 4.8 ⫾ 0.4 22.3 ⫾ 0.7 33.5 ⫾ 1.2 6.8 ⫾ 0.1 16.9 ⫾ 0.5 38.4 ⫾ 1.4 36.3 ⫾ 1.5 35.1 ⫾ 0.9 37.6 ⫾ 0.8
0.99 1.23 4.2 4.5 21.6 31.4 6.6 16.3 36.3 34.3 34.3 36.3
1.05 1.30 4.7 5.8 23.5 35.3 6.9 17.7 40.2 39.2 36.3 39.2
Guide wire 0.36 mm Balloon brand x
Not expanded Expanded (300 kPa)
Balloon brand y
Not expanded Expanded (300 kPa)
New brachytherapy device
Shrink tube shrunk by heating
to the vessel wall. It is expected that this depth profile is suitable to ensure a sufficient dose at lower layers of the vessel wall to inhibit restenosis. With the source design, a nominal activity of approximately 4 GBq can be readily achieved. This high activity ensures a short irradiation time of a few minutes to deliver 15–30 Gy to the adventitial border of the vessel wall. The dose rate was 17 Gy/min at 1.5-mm distance from the center and 10 Gy/min at 2.0 mm from the surface of the source at the time of irradiation (Fig. 4). Homogeneity of irradiation of vessel walls The required homogeneous irradiation of the stenosis zone depends significantly on the radial centering of the source inside the vessel. Optimally, the source should be positioned in the center of the vessel (Fig. 5, left side). However, any radial source inside the vessel may deviate from this position to the side of the artery, which results in
Fig. 4. Dose-depth distribution in radial direction of source.
an inhomogeneous dose to the artery wall. For this source design with a radioactive source of 0.45-mm diameter and a surrounding shrink tube diameter of 0.9 mm, the most adverse position inside a blood vessel with an inner diameter of 1.7 mm (average diameter of the investigated rabbit vessel) is displayed in Fig. 5, right side. The dose values derived for different sections of the arterial wall with the source in this most adverse position possible are given in columns 3 and 4 of Table 2. Thus, if the nominal activity and irradiation time of the source is chosen to give a planned dose of 30 Gy to the vessel wall for the source being in the well-centered position, the maximum dose in the closest position (position a in Fig. 5) would not exceed 46 Gy, still below lethal radiation levels for the cell, while in the most distant position, a dose of 19 Gy may result (position c). If the dose is set to 30 Gy at a depth of 0.5 mm in the vessel wall (to obtain a dose in the animal artery wall comparable to that of the human exposure as later described), the dose may vary, as given in columns 5 and 6 of Table 2. The minimum dose at 0.5-mm depth would not be less than 20 Gy (position c, 5-mm depth), while the maximum dose would not exceed 46 Gy (position a). Thus, dose levels are well within the desired dose range on all sides of the arterial wall, that is, above dose levels expected to be
Fig. 5. Maximum deviation of dose by improper centering of source.
Design and dosimetry of
90
Y beta sources
● K. MU¨ CK et al.
253
Table 2. Dose variations due to eccentric positioning of the irradiation source inside the artery Source activity and irradiation time set to 30 Gy at inner surface of vessel wall Source position Centered position Eccentric position Position a Position b Position c
Source activity and irradiation time set to 30 Gy at 0.5-mm depth of vessel wall
Distance from wire axis to wall surface (mm)
Dose at surface (Gy)
Dose at 0.5-mm depth (Gy)
Dose at surface (Gy)
Dose at 0.5-mm depth (Gy)
0.85
30
16
55
30
0.50 0.94 1.25
46 27 19
25 15 11
84 49 34
46 28 20
ineffective with regard to restenosis prevention, but below any critical radiation levels causing negative radiation effects. The dose in the closest position to the wall surface (position a) in any case would be not more than about 50% above the planned treatment dose and not less than 40% below that value in the furthest point (position c). Radiation protection To test any possible leakage of the sources, the following measurements of radioactive material outside the source were performed: (a) measurement of the activity content of the boiling water-alcohol solution (ISO-9978), (b) measurement of smear tests on the source before delivery to the hospital, (c) measurement of surface contamination of the surgery table and equipment after application, and (d) measurement of contamination of tissues and plastic covers after application. None of these tests showed any sigificant amount of activity released from the source. The designed sources, therefore, provide an adequate protection against any internal contamination of the test animal or any contamination of the application room, equipment, or personnel. Due to the good shielding of the transport container, the exposure of personnel in the treatment room is low. The beta dose rate is practically zero and the gamma dose rate at the outside of the container amounts to only 600 Sv/h at the surface and 5 Sv/h at a 50-cm distance from the surface. For the very short time when the source is not shielded when moving from the container into the body (2–5 s), an exposure of only 1–3 Sv to the medical personnel is expected per treatment. The low dose was demonstrated during preliminary tests and is considerably less than the x-ray dose from the angiographic surveillance to personnel during application. DISCUSSION The new brachytherapy device allows intravascular irradiation in peripheral arteries, and also in proximal coronary arteries. The high activity of the source permits short irradiation times to the application of therapeutic doses of 15–30 Gy at a depth of 1–2 mm. Due to the segmentation
into 90Y seeds of 5-mm length, the source is not only flexible, but by assembling different numbers of seeds, the production of different lengths of sources is easily achieved. Thereby, a perfect adaptation of source length to the length of the stenosis zone may be obtained which is advantageous over other devices. Because the wire segments are not in contact with the blood, and are freshly enclosed inside the tight, inactive shrink tube upon each application, they may be reactivated in the reactor and reused several times. Several other beta-emitting devices have been previously tested. Popowski et al. (15) developed a 29-mm-long 90Y coil coated with titanium and fixed to the end of a thrust wire. The advantage of this device is that it can be used with a centering balloon. However, the disadvantage is that the device-artery distance is bigger; therefore, longer irradiation times are required for a given source activity. A recently published clinical trial using this device failed to reduce restenosis (16), the most probable reason being a too low dose to the wall (target dose of 18 Gy set at the inner vessel wall, resulting in a dose of 7.2 Gy at a depth of 1 mm or of 2.7 Gy at 2 mm). Similarly, Xu et al. (17) reported on a 32P source that is 3-mm long and 0.3 mm in diameter, encapsulated in a Ni-Ti wire with a total diameter of 0.4 mm. This device is shorter than the standard balloon, and therefore, has to be retracted stepwise to obtain uniform irradiation over lesion length, a disadvantage not present in our design. Our study presents the first comparison of flexibility between radioactive source and standard interventional devices. The difference between a standard noninflated balloon and the radioactive source in force required for centrally bending by 30° was approximately 29 N. This is comparable to an inflated balloon, which needs 32 N more. Therefore, this source may not be suitable for distal coronary vessels, where tortuous artery segments have to be passed. However, peripheral coronary branches usually have a mean diameter of less than 2 mm, and are therefore not ideal for any intraluminal device, because of the risk of myocardial ischemia. An unfavorable device/ lumen ratio may cause ischemia by reduced coronary blood flow and/or vascular spasms, which lead to reversible occlusion of the vessel. Extensive tests by boiling in water/alcohol solution and by smear tests according to the ISO-standard 9978 proved that the radioactive source in the device was leakproof. Therefore, the
254
I. J. Radiation Oncology
●
Biology
●
Physics
source will not cause contamination within the treated body, or to the application room or personnel therein. In gamma brachytherapy, quality assurance of the dose to be applied is achieved by routine activity measurements before application. Because of self-shielding of betas inside the source, a determination of the true 90Y activity of the source is not achievable without destruction of the radioactive source. However, a determination of the true activity of the source is not really required; rather the rate of betas leaving the source boundary (“beta emissivity”) is relevant for the brachytherapy treatment. This emissivity is then calibrated against the dose. By designing a special betaplastic scintillator measuring device, a calibration of source emissivity to the dose to be applied to the stenosis zone could be achieved. Despite the problem that a determination of the source activity is not easily achievable for beta sources, the proper dose to the stenosis zone can be assured by this approach. The absolute dose is derived by thin TLDs measurements, which were standardized against available secondary standards. According to AAPM task group recommendations (13), the reference dose should be presented to a distance of 2.0 mm from the source axis. For a human coronary artery, the inner diameter amounts to approximately 3 mm, resulting in a reference dose according to AAPM at a depth of 0.5 mm from the inner surface of the blood vessel. Because the diameter of the artery of the planned experimental animal is 1.7 mm (a value which was established in the preparation of the animal experiment), the same depth of 0.5 mm from the inner vessel surface would correspond to a reference point at a distance of 1.4 mm from the center of the source for the animal experiment (see Fig. 6). For that reason, in the experimental setup and observed results, the reference dose always refers to a distance of 1.4 mm from source axis in the animal vessel. The designed source provides an optimum protection of the patient and requires little radiation protection efforts on behalf of the medical staff. Dosimetric calculations and measurements showed that a high homogeneity of the dose inside the vessel could be achieved due to a good centering of the source. The maximum deviation between highest dose and average dose under most adverse self-positioning of the brachytherapy device in the artery would not exceed 50%. This parameter,
Volume 46, Number 1, 2000
Fig. 6. Best source geometry for animal arteries as compared to human arteries.
probably most important for the success of the method of post-dilatation irradiation to prevent restenosis, was not determined thus far, and may be the cause for the negative results in some of the previous investigations. CONCLUSION This new device is safe and seems to have adequate flexibility to reach peripheral arteries and proximal coronary vessels. The special design of the source is leakproof, with no contamination detected by standard tests or smear tests in the application room. Due to the half-life of 64 h, and a special design for cleaning the source after irradiation, the source may be reused several times for repetitive experiments. Therefore, due to the half-life of the source and the treatment time, it is expected that up to 30 animal irradiations can be performed with one source. By activating the source to an activity of approximately 4 GBq nominal activity, irradiation times of a few minutes are obtained. Such short irradiation times should ensure no occlusions in the artery during treatment. The device ensures a sufficiently radially homogeneous beta irradiation of the vessel walls, whereby the deviations in the dose to the vascular wall at a depth of 0.5 mm inside the vessel wall would not exceed a factor of more than 2.5 between the highest and lowest dose. Generally, much lower deviations are to be expected. It is believed that this ensures a sufficiently homogeneous irradiation of the relevant tissue. Further studies are needed to prove efficacy to reduce neointimal proliferation after balloon dilatation.
REFERENCES 1. CABRI. First-year results of CABRI (coronary angioplasty versus bypass revascularisation investigation). Lancet 1995; 346:1179 –1184. 2. Gottsauner-Wolf M, Moliterno DJ, Lincoff AM, et al. Restenosis - an open file. Clin Cardiol 1996;19:347–356. 3. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury. I. Smooth muscle growth in the absence of endothelium. Lab Invest 1983;49:327–333. 4. Landau C, Lange RA, Hillis LD. Percutaneous transluminal coronary angioplasty. N Engl J Med 1994;330:981–993. 5. Serruys P, De Jaegere P, Kiemeniej F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331:489 – 495.
6. The Epilog Investigators. Platelet glycoprotein IIb/IIIa receptor blockade and low-dose heparin during percutaneous coronary revascularization. N Engl J Med 1997;336:1689 –1696. 7. Teirstein PS, Massullo V, Jani S, et al. Catheter-based radiotherapy to inhibit restenosis after coronary stenting. N Engl J Med 1997;336:1697–1703. 8. Schmidt W, Mu¨ck K, Lehmann D, et al. Tubular beta sources for endovascular irradiations: Design and dosimetry. AAPM Annual Meeting San Antonio, TX, 1998. Med Phys 1998;25;175. 9. Mu¨ck K, Schmidt W, Go¨rz W, et al. Betaquellen fu¨r endovaskula¨re, intracoronare Brachytherapie (in German). In: Schmidt R, editor. Medizinische Physik ’97. 28th Congress German Soc. for Medical Physics, 24 –27 Sept 1997, Hamburg. p. 127. 10. Go¨rz W, Nedelik A, Miksche W. Source manipulator as well
Design and dosimetry of
90
as design and device for producing the same. European patent 0 466 681 A1. 1991. 11. Schmidt W, Mu¨ck K, Steger F, et al. Dosimetrie an Betaquellen fu¨r endovaskula¨re, intracoronare Brachytherapie (in German). In: Schmidt R, editor. Medizinische physik ’97. 28th Congress German Soc. for Medical Physics, 24 –27 Sept 1997, Hamburg. p. 125. 12. Drake KH, Bo¨hm J. Automatisierter Mestand fu¨r die Dosimetrie von Betastrahlung (in German) PTB-Dos-19, Braunschweig;1990, ISBN 3-89429-012-9. 13. Nath R, Amols H, Coffey C, et al. Intravascular brachytherapy physics: Report of the Radiation Therapy Committee Task Group No. 6. Med Phys 1999;26:119 –152.
Y beta sources
● K. MU¨ CK et al.
255
14. International Standard Organization: Radiation protection— sealed radioactive sources—leakage test methods, ISO-standard 9978. 1992. 15. Popowski Y, Verin V, Schwager M, et al. A novel system for intracoronary beta-irradiation: Description and dosimetric results. Int J Radiat Oncol Biol Phys 1996;36:923–931. 16. Verin V, Urban P, Popowski Y, et al. Feasibility of angiocoronary beta-irradiation to reduce restenosis after balloon angioplasty. A clinical pilot study. Circulation 1997;95:1138 – 1144. 17. Xu Z, Almond PR, Deasy JO. The dose distribution produced by a 32P source for endovascular irradiation. Int J Radiat Oncol Biol Phys 1996;36:933–939.
PII S0360-3016(99)00350-8
PHYSICS CONTRIBUTION
DESIGN AND DOSIMETRY OF A NOVEL 90Y BETA SOURCE TO PREVENT RESTENOSIS AFTER ANGIOPLASTY KONRAD MU¨ CK, PH.D.,* WERNER SCHMIDT, PH.D.,† PAUL WEXBERG, M.D.,‡ WALTER GO¨ RZ,* GERALD MAURER, M.D.,‡ AND MICHAEL GOTTSAUNER-WOLF, M.D.‡ *Austrian Research Center Seibersdorf, Seibersdorf, Austria; †Donauspital Vienna, Institute for Radiooncology, Vienna, Austria; and ‡ University of Vienna, Department of Cardiology, Vienna, Austria Purpose: Post-dilatation irradiation of the vessel wall is currently under investigation for prevention of restenosis after balloon dilatation. For the irradiation, special sources were designed for animal experiments which would give equivalent irradiation conditions and doses to the vessel wall that would later be employed for human application. Methods and Materials: For the planned irradiations, a specially designed yttrium-wire of 0.45-mm diameter coated with a thin shrink tube to prevent contamination was deployed. Several leakage tests applied before and after application proved that the irradiation source was leakproof. Dosimetry was performed by using 0.1-mmthick thermoluminescent dosimeters (TLD-100) calibrated against a primary standard. A shielding transport and application container was designed to facilitate the handling of the source during use, while reducing exposure of the medical personnel. Results: The designed source proves to be flexible for the insertion into proximal coronary vessels, and positioning at the site of stenosis. It provides an optimum protection of the animal and requires little radiation protection efforts on behalf of the medical staff. Dosimetric calculations and measurements showed that a centering of the source inside the vessel could be achieved with a maximum deviation of 50% between maximum and average dose levels. Conclusion: A yttrium-90 beta brachytherapy source was designed which provides high flexibility within proximal coronary arteries, ensures an adequate centering inside the artery, and provides irradiation conditions to the vessel wall of the experimental animal comparable to the application inside a human artery. © 2000 Elsevier Science Inc. Brachytherapy, Yttrium-90, Coronary artery disease, Angioplasty, Restenosis.
INTRODUCTION Atherosclerotic cardiovascular diseases are the main causes of death in the western world. Since the late 1970s, percutaneous transluminal coronary revascularization (PTCR) has become a potent alternative to surgical revascularization in patients with coronary artery disease (1). Balloon dilatation of atherosclerotic stenoses is initially successful in over 90%, with a low mortality rate of less than 1%. However, renarrowing (restenosis) occurs in 30 – 50% of patients, as a response to the PTCR 3– 6 months after the intervention (2). One of the main mechanisms of this phenomenon is vascular smooth muscle cell proliferation (3). Several devices (directional atherectomy, laser, rotational atherectomy, and intravascular stents) and pharmacological approaches (aspirin, anticoagulants, Ca-channel blockers, beta-blockers, angiotensin-converting enzyme inhibitors, fish-oil, IIb/IIIa inhibitors, etc.) have been tested in various clinical trials to reduce restenosis rate (4). Only
stents and blockage of the IIb/IIIa receptor showed favorable effects. However, the restenosis rate after sucessful PTCR remains at 20 –30% (5, 6). Recently, the antiproliferative effect of ionizing radiation has been proven to reduce neointimal growth up to 70% after experimental PTCR in animals and in clinical trials (7). An optimal intravascular brachytherapy device must be: (a) low-profile, flexible, and steerable, to pass tortuous vessel segments; (b) low in radiation exposure to the patient and operator; and (c) effective in reducing smooth muscle cell growth following the intervention in order to reduce restenosis. Beta sources of yttrium-90 (90Y) were selected to be applied for the human application (8) because they possess several advantages as compared to gamma radiation (9). They will be applied both in the proposed animal experiments and in human treatment. With beta sources, the determination of their absolute activity is not as easily achievable as with gamma sources; however, it will be shown that this is not detrimental to the
Reprint requests to: Konrad Mu¨ck, Ph.D., Austrian Research Center Seibersdorf, A-2444 Seibersdorf, Austria. Tel: ⫹⫹43 2254 780 3200; Fax: ⫹⫹43 2254 780 3206; E-mail: [email protected]
Supported by a grant of the Ludwig Boltzmann Institute for Cardiovascular Research. Accepted for publication 6 August 1999. 249
250
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 46, Number 1, 2000
Fig. 1. Design of beta treatment course.
application. A method to precisely ensure the required exposure values is described in this paper. The aim of the present study was to design and evaluate a novel intravascular beta-emitting brachytherapy device fulfilling the above-mentioned requirements. For the animal experiments ongoing at the Department of Cardiology of the University of Vienna, using New Zealand white rabbits, special radiation sources were designed and constructed to obtain irradiation inside the animal artery equivalent to that to be applied in the human treatment. This paper describes the design of the source to ensure its flexibility and to limit the irradiation time to less than a few minutes inside the artery, the centering inside the vessel, and the dose and its depth distribution to be comparable to later human applications. METHODS AND MATERIALS Design of brachytherapy device A source was designed to assure high flexibility, as well as optimum centering, inside the vascular vessel to obtain a high-dose homogeneity of irradiation. It consists of 5-mm long segments of an 90Y wire of 0.45-mm inner diameter positioned inside a shrink tube of 0.9-mm outer diameter (Fig. 1). This source is inserted through a 5-French catheter (1.65-mm diameter). To obtain sufficiently short irradiation times (⬃ 2– 4 min) inside the vessel, a nominal source activity of approximately 4 GBq is required. This activity is obtained by activating the source at the reactor of the Austrian Research Center Seibersdorf (ARCS) at a flux of 6.1013 cm-2 s-1 for 60 h. After activation, the source is stored for 24 h until the 3.19-h half-life 90Y has decayed to a sufficiently low level. The source is then placed inside the shrink tube according to a special patented technique (10) developed at ARCS for 192 Ir wires used in cancer brachytherapy. The source was attached to the anterior end of a standard 0.014-inch guide wire (standard length 175 cm, Fig. 1) by shrinking the shrink tube, and additionally secured with glue (LOCTITE 407). Flexibility of source The flexibility of the source was tested according to standard tests applied to catheters. The devices to be tested were fixed in a positioning tube in such a way that half of the balloon or the source, respectively, was outside the distal end of the guiding catheter (Fig. 2). The devices were
warmed to 37°C in a water bath for 2 min. The force to bend the device by 30° was measured with a digital scale (Satorius R160P). For comparison purposes, a standard guidewire (0.014 inch; Scimed, Naticle, MA) and two standard angioplasty balloons (brand y and brand x) were tested by the same device. Because of asymmetry, each device was tested in the pre-bent direction and the 180° reverse direction. Each experiment was repeated 10 times. Dosimetry Absolute dosimetry was performed by EXT-RAD-TLDs (1 ⫻ 1 mm2, 40 mg/cm2, thermoluminescent dosimeters (TLDs) fixed on a CAPTON-foil; Bicron). A special annealing (on a PC-driven PTW THELDO oven) and evaluation cycle (on a Bicron 3500 system) was established (11). Maximum temperature in both cases was 300°C. Annealing took about 20 min, with temperatures falling from 300°C to 80°C in 1-min increments; during readout the TLDs were heated to 300°C in 1 min, with a constant temperature rise of 5°C/s. The reproducibility of the area of the glowcurve was ⬎ 97%, the rising sensitivity due to incomplete annealing (maximum allowed temperature 300°C; in particular, after application of doses ⬎ 3 Gy) was compensated by individual calibration of the TLDs. The TLDs were calibrated both against 60Co sources
Fig. 2. Setup for testing the flexibility of the source.
Design and dosimetry of
90
Y beta sources
● K. MU¨ CK et al.
251
Fig. 3. Beta source and shielding container at the application site.
calibrated by the Austrian National Bureau of Standards and a secondary standard 90Sr/90Y reference source of the Physikalisch-Technische-Bundesanstalt, FRG (Beta Secondary Standard BeSS2 (12)). The ratio of the areas under the glowcurves for 1 Gy 60Co to 1 Gy 90Sr/90Y was ⬍ 1.05, which demonstrates the qualification of thin TLDs for beta dose measurements and that, as with gamma measurements, a reliable dose at fixed distances from the source center may be measured by placing thin TLDs in given positions from the axis and determining the thermoluminescence response, thus establishing a fixed relation between reference dose and beta emission. Depth-dose distributions perpendicular to the source axis were determined according to AAPM specifications (13) at specified distances up to 10 mm from the source. Also, absolute dose measurements were performed with TLDs. For depth-dose measurements, PMMA-sheets or polycarbonate foils (Makrofol G; Bayer) of different thicknesses were positioned between the TLDs. A special beta detector system was designed, consisting of a beta-plastic scintillator of 10-mm diameter and 5-mm height at a fixed distance from the source. The detection system is routinely checked with a standard 90Sr-source. The reproducibility of the system was extensively tested and is ⱖ 95%. The beta detector system is calibrated for dose measurements by use of the above-mentioned TLD calibration measurements. By this calibrated system, the actual beta emission of the activated source is determined before application, and the irradiation time is set accordingly.
To permit easy handling of the source and ensure a low radiation exposure of the medical staff during the application, as well as to facilitate the transport from the reactor to the application room, a special transport and application container was designed and built at ARCS. It consists of a shield of 10-mm plexiglass and 15-mm Pb coated by a thin Fe cover (Fig. 3). Despite its low weight of only 1.6 kg, the container gives excellent shielding.
Radiation protection issues After attaching the guidewire and heat shrinking the tube, a leakproof source is obtained. The tightness of the source was tested according to ISO-standard 9978 (14) by boiling the source in a water-alcohol solution for 1 h. Then the source is placed in the shielding/transport container and transferred to the hospital.
Dosimetry Results of depth-dose distribution measurements perpendicular to the source axis determined according to AAPM specifications are shown in Fig. 4. Resulting depth profiles show a decrease in dose by about a factor of three per 1-mm depth into the vascular tissue, both for an ideally centered source and one positioned close
RESULTS Flexibility The flexibility of the irradiation source, which is enabled by the special construction of the source in cylindrical segments, was tested by determining the deviation force required to bend the device to an angle of 30° in an experimental setup as shown in Fig. 2. Forces required for bending the brachytherapy device, as well as the two types of catheters and the guidewire, by 30° are summarized in Table 1. As expected, the lowest force is needed to bend the guidewire (1.02 ⫾ 0.02 mN in one direction and 1.28 ⫾ 0.02 mN in the opposite). The deflated balloons required a substantially greater force (4.5 to 6.8 mN, respectively), while in the opposite side, the required force in one case amounted to 16.9 ⫾ 0.5 mN. The bending force required for the inflated balloons was around 22– 40 mN (5–7 times stiffer than noninflated). The new brachytherapy device showed a comparable bending force of 35.1 ⫾ 0.9 mN and 37.6 ⫾ 0.8 mN in the opposite direction.
252
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 46, Number 1, 2000
Table 1. Results of flexibility test Bending force required for 30° angle (mN) Device
Test condition
Orientation (°)
Mean ⫾ SD
Minimum
Maximum
—
0 180 0 180 0 180 0 180 0 180 0 180
1.02 ⫾ 0.02 1.28 ⫾ 0.02 4.5 ⫾ 0.2 4.8 ⫾ 0.4 22.3 ⫾ 0.7 33.5 ⫾ 1.2 6.8 ⫾ 0.1 16.9 ⫾ 0.5 38.4 ⫾ 1.4 36.3 ⫾ 1.5 35.1 ⫾ 0.9 37.6 ⫾ 0.8
0.99 1.23 4.2 4.5 21.6 31.4 6.6 16.3 36.3 34.3 34.3 36.3
1.05 1.30 4.7 5.8 23.5 35.3 6.9 17.7 40.2 39.2 36.3 39.2
Guide wire 0.36 mm Balloon brand x
Not expanded Expanded (300 kPa)
Balloon brand y
Not expanded Expanded (300 kPa)
New brachytherapy device
Shrink tube shrunk by heating
to the vessel wall. It is expected that this depth profile is suitable to ensure a sufficient dose at lower layers of the vessel wall to inhibit restenosis. With the source design, a nominal activity of approximately 4 GBq can be readily achieved. This high activity ensures a short irradiation time of a few minutes to deliver 15–30 Gy to the adventitial border of the vessel wall. The dose rate was 17 Gy/min at 1.5-mm distance from the center and 10 Gy/min at 2.0 mm from the surface of the source at the time of irradiation (Fig. 4). Homogeneity of irradiation of vessel walls The required homogeneous irradiation of the stenosis zone depends significantly on the radial centering of the source inside the vessel. Optimally, the source should be positioned in the center of the vessel (Fig. 5, left side). However, any radial source inside the vessel may deviate from this position to the side of the artery, which results in
Fig. 4. Dose-depth distribution in radial direction of source.
an inhomogeneous dose to the artery wall. For this source design with a radioactive source of 0.45-mm diameter and a surrounding shrink tube diameter of 0.9 mm, the most adverse position inside a blood vessel with an inner diameter of 1.7 mm (average diameter of the investigated rabbit vessel) is displayed in Fig. 5, right side. The dose values derived for different sections of the arterial wall with the source in this most adverse position possible are given in columns 3 and 4 of Table 2. Thus, if the nominal activity and irradiation time of the source is chosen to give a planned dose of 30 Gy to the vessel wall for the source being in the well-centered position, the maximum dose in the closest position (position a in Fig. 5) would not exceed 46 Gy, still below lethal radiation levels for the cell, while in the most distant position, a dose of 19 Gy may result (position c). If the dose is set to 30 Gy at a depth of 0.5 mm in the vessel wall (to obtain a dose in the animal artery wall comparable to that of the human exposure as later described), the dose may vary, as given in columns 5 and 6 of Table 2. The minimum dose at 0.5-mm depth would not be less than 20 Gy (position c, 5-mm depth), while the maximum dose would not exceed 46 Gy (position a). Thus, dose levels are well within the desired dose range on all sides of the arterial wall, that is, above dose levels expected to be
Fig. 5. Maximum deviation of dose by improper centering of source.
Design and dosimetry of
90
Y beta sources
● K. MU¨ CK et al.
253
Table 2. Dose variations due to eccentric positioning of the irradiation source inside the artery Source activity and irradiation time set to 30 Gy at inner surface of vessel wall Source position Centered position Eccentric position Position a Position b Position c
Source activity and irradiation time set to 30 Gy at 0.5-mm depth of vessel wall
Distance from wire axis to wall surface (mm)
Dose at surface (Gy)
Dose at 0.5-mm depth (Gy)
Dose at surface (Gy)
Dose at 0.5-mm depth (Gy)
0.85
30
16
55
30
0.50 0.94 1.25
46 27 19
25 15 11
84 49 34
46 28 20
ineffective with regard to restenosis prevention, but below any critical radiation levels causing negative radiation effects. The dose in the closest position to the wall surface (position a) in any case would be not more than about 50% above the planned treatment dose and not less than 40% below that value in the furthest point (position c). Radiation protection To test any possible leakage of the sources, the following measurements of radioactive material outside the source were performed: (a) measurement of the activity content of the boiling water-alcohol solution (ISO-9978), (b) measurement of smear tests on the source before delivery to the hospital, (c) measurement of surface contamination of the surgery table and equipment after application, and (d) measurement of contamination of tissues and plastic covers after application. None of these tests showed any sigificant amount of activity released from the source. The designed sources, therefore, provide an adequate protection against any internal contamination of the test animal or any contamination of the application room, equipment, or personnel. Due to the good shielding of the transport container, the exposure of personnel in the treatment room is low. The beta dose rate is practically zero and the gamma dose rate at the outside of the container amounts to only 600 Sv/h at the surface and 5 Sv/h at a 50-cm distance from the surface. For the very short time when the source is not shielded when moving from the container into the body (2–5 s), an exposure of only 1–3 Sv to the medical personnel is expected per treatment. The low dose was demonstrated during preliminary tests and is considerably less than the x-ray dose from the angiographic surveillance to personnel during application. DISCUSSION The new brachytherapy device allows intravascular irradiation in peripheral arteries, and also in proximal coronary arteries. The high activity of the source permits short irradiation times to the application of therapeutic doses of 15–30 Gy at a depth of 1–2 mm. Due to the segmentation
into 90Y seeds of 5-mm length, the source is not only flexible, but by assembling different numbers of seeds, the production of different lengths of sources is easily achieved. Thereby, a perfect adaptation of source length to the length of the stenosis zone may be obtained which is advantageous over other devices. Because the wire segments are not in contact with the blood, and are freshly enclosed inside the tight, inactive shrink tube upon each application, they may be reactivated in the reactor and reused several times. Several other beta-emitting devices have been previously tested. Popowski et al. (15) developed a 29-mm-long 90Y coil coated with titanium and fixed to the end of a thrust wire. The advantage of this device is that it can be used with a centering balloon. However, the disadvantage is that the device-artery distance is bigger; therefore, longer irradiation times are required for a given source activity. A recently published clinical trial using this device failed to reduce restenosis (16), the most probable reason being a too low dose to the wall (target dose of 18 Gy set at the inner vessel wall, resulting in a dose of 7.2 Gy at a depth of 1 mm or of 2.7 Gy at 2 mm). Similarly, Xu et al. (17) reported on a 32P source that is 3-mm long and 0.3 mm in diameter, encapsulated in a Ni-Ti wire with a total diameter of 0.4 mm. This device is shorter than the standard balloon, and therefore, has to be retracted stepwise to obtain uniform irradiation over lesion length, a disadvantage not present in our design. Our study presents the first comparison of flexibility between radioactive source and standard interventional devices. The difference between a standard noninflated balloon and the radioactive source in force required for centrally bending by 30° was approximately 29 N. This is comparable to an inflated balloon, which needs 32 N more. Therefore, this source may not be suitable for distal coronary vessels, where tortuous artery segments have to be passed. However, peripheral coronary branches usually have a mean diameter of less than 2 mm, and are therefore not ideal for any intraluminal device, because of the risk of myocardial ischemia. An unfavorable device/ lumen ratio may cause ischemia by reduced coronary blood flow and/or vascular spasms, which lead to reversible occlusion of the vessel. Extensive tests by boiling in water/alcohol solution and by smear tests according to the ISO-standard 9978 proved that the radioactive source in the device was leakproof. Therefore, the
254
I. J. Radiation Oncology
●
Biology
●
Physics
source will not cause contamination within the treated body, or to the application room or personnel therein. In gamma brachytherapy, quality assurance of the dose to be applied is achieved by routine activity measurements before application. Because of self-shielding of betas inside the source, a determination of the true 90Y activity of the source is not achievable without destruction of the radioactive source. However, a determination of the true activity of the source is not really required; rather the rate of betas leaving the source boundary (“beta emissivity”) is relevant for the brachytherapy treatment. This emissivity is then calibrated against the dose. By designing a special betaplastic scintillator measuring device, a calibration of source emissivity to the dose to be applied to the stenosis zone could be achieved. Despite the problem that a determination of the source activity is not easily achievable for beta sources, the proper dose to the stenosis zone can be assured by this approach. The absolute dose is derived by thin TLDs measurements, which were standardized against available secondary standards. According to AAPM task group recommendations (13), the reference dose should be presented to a distance of 2.0 mm from the source axis. For a human coronary artery, the inner diameter amounts to approximately 3 mm, resulting in a reference dose according to AAPM at a depth of 0.5 mm from the inner surface of the blood vessel. Because the diameter of the artery of the planned experimental animal is 1.7 mm (a value which was established in the preparation of the animal experiment), the same depth of 0.5 mm from the inner vessel surface would correspond to a reference point at a distance of 1.4 mm from the center of the source for the animal experiment (see Fig. 6). For that reason, in the experimental setup and observed results, the reference dose always refers to a distance of 1.4 mm from source axis in the animal vessel. The designed source provides an optimum protection of the patient and requires little radiation protection efforts on behalf of the medical staff. Dosimetric calculations and measurements showed that a high homogeneity of the dose inside the vessel could be achieved due to a good centering of the source. The maximum deviation between highest dose and average dose under most adverse self-positioning of the brachytherapy device in the artery would not exceed 50%. This parameter,
Volume 46, Number 1, 2000
Fig. 6. Best source geometry for animal arteries as compared to human arteries.
probably most important for the success of the method of post-dilatation irradiation to prevent restenosis, was not determined thus far, and may be the cause for the negative results in some of the previous investigations. CONCLUSION This new device is safe and seems to have adequate flexibility to reach peripheral arteries and proximal coronary vessels. The special design of the source is leakproof, with no contamination detected by standard tests or smear tests in the application room. Due to the half-life of 64 h, and a special design for cleaning the source after irradiation, the source may be reused several times for repetitive experiments. Therefore, due to the half-life of the source and the treatment time, it is expected that up to 30 animal irradiations can be performed with one source. By activating the source to an activity of approximately 4 GBq nominal activity, irradiation times of a few minutes are obtained. Such short irradiation times should ensure no occlusions in the artery during treatment. The device ensures a sufficiently radially homogeneous beta irradiation of the vessel walls, whereby the deviations in the dose to the vascular wall at a depth of 0.5 mm inside the vessel wall would not exceed a factor of more than 2.5 between the highest and lowest dose. Generally, much lower deviations are to be expected. It is believed that this ensures a sufficiently homogeneous irradiation of the relevant tissue. Further studies are needed to prove efficacy to reduce neointimal proliferation after balloon dilatation.
REFERENCES 1. CABRI. First-year results of CABRI (coronary angioplasty versus bypass revascularisation investigation). Lancet 1995; 346:1179 –1184. 2. Gottsauner-Wolf M, Moliterno DJ, Lincoff AM, et al. Restenosis - an open file. Clin Cardiol 1996;19:347–356. 3. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury. I. Smooth muscle growth in the absence of endothelium. Lab Invest 1983;49:327–333. 4. Landau C, Lange RA, Hillis LD. Percutaneous transluminal coronary angioplasty. N Engl J Med 1994;330:981–993. 5. Serruys P, De Jaegere P, Kiemeniej F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331:489 – 495.
6. The Epilog Investigators. Platelet glycoprotein IIb/IIIa receptor blockade and low-dose heparin during percutaneous coronary revascularization. N Engl J Med 1997;336:1689 –1696. 7. Teirstein PS, Massullo V, Jani S, et al. Catheter-based radiotherapy to inhibit restenosis after coronary stenting. N Engl J Med 1997;336:1697–1703. 8. Schmidt W, Mu¨ck K, Lehmann D, et al. Tubular beta sources for endovascular irradiations: Design and dosimetry. AAPM Annual Meeting San Antonio, TX, 1998. Med Phys 1998;25;175. 9. Mu¨ck K, Schmidt W, Go¨rz W, et al. Betaquellen fu¨r endovaskula¨re, intracoronare Brachytherapie (in German). In: Schmidt R, editor. Medizinische Physik ’97. 28th Congress German Soc. for Medical Physics, 24 –27 Sept 1997, Hamburg. p. 127. 10. Go¨rz W, Nedelik A, Miksche W. Source manipulator as well
Design and dosimetry of
90
as design and device for producing the same. European patent 0 466 681 A1. 1991. 11. Schmidt W, Mu¨ck K, Steger F, et al. Dosimetrie an Betaquellen fu¨r endovaskula¨re, intracoronare Brachytherapie (in German). In: Schmidt R, editor. Medizinische physik ’97. 28th Congress German Soc. for Medical Physics, 24 –27 Sept 1997, Hamburg. p. 125. 12. Drake KH, Bo¨hm J. Automatisierter Mestand fu¨r die Dosimetrie von Betastrahlung (in German) PTB-Dos-19, Braunschweig;1990, ISBN 3-89429-012-9. 13. Nath R, Amols H, Coffey C, et al. Intravascular brachytherapy physics: Report of the Radiation Therapy Committee Task Group No. 6. Med Phys 1999;26:119 –152.
Y beta sources
● K. MU¨ CK et al.
255
14. International Standard Organization: Radiation protection— sealed radioactive sources—leakage test methods, ISO-standard 9978. 1992. 15. Popowski Y, Verin V, Schwager M, et al. A novel system for intracoronary beta-irradiation: Description and dosimetric results. Int J Radiat Oncol Biol Phys 1996;36:923–931. 16. Verin V, Urban P, Popowski Y, et al. Feasibility of angiocoronary beta-irradiation to reduce restenosis after balloon angioplasty. A clinical pilot study. Circulation 1997;95:1138 – 1144. 17. Xu Z, Almond PR, Deasy JO. The dose distribution produced by a 32P source for endovascular irradiation. Int J Radiat Oncol Biol Phys 1996;36:933–939.