Intracoronary β-Irradiation With a Liquid 188Re-Filled Balloon
Six-Month Results From a Clinical Safety and Feasibility Study
Background—Coronary irradiation is a new concept to reduce restenosis. We evaluated the feasibility and safety of intracoronary irradiation with a balloon catheter filled with 188Re, a liquid, high-energy β-emitter.
Methods and Results—Irradiation with 15 Gy at 0.5-mm tissue depth was performed in 28 lesions after balloon dilation (n=9) or stenting (n=19). Lesions included 19 de novo stenoses, 4 occlusions, and 5 restenoses. Irradiation time was 515±199 seconds in 1 to 4 fractions. There were no procedural complications. One patient died of noncardiac causes at day 23. One asymptomatic patient refused 6-month angiography. Quantitative angiography after intervention showed a reference diameter of 2.77±0.35 mm and a minimal lumen diameter of 2.36±0.43 mm. At 6-month follow-up, minimal lumen diameter was 1.45±0.88 mm (late loss index 0.57). Target lesion restenosis rate (>50% in diameter) was low (12%; 3 of 26). In addition, we observed 9 stenoses at the proximal or distal end of the irradiation zone, potentially caused by the short irradiation segment and the decreasing irradiation dose at its borders (“edge” stenoses). The total restenosis rate was 46% and was significantly lower (29% vs 70%, P=0.042) when the length of the irradiated segment was more than twice the lesion length.
Conclusions—Coronary irradiation with a 188Re-filled balloon is technically feasible and safe, requiring only standard percutaneous transluminal coronary angioplasty techniques. The target lesion restenosis rate was low. The observed edge stenoses appear to be avoidable by increasing the length of the irradiated segment.
Restenosis is a major limitation of coronary angioplasty. Its biology includes inflammation, proliferation, and migration of smooth muscle cells and remodeling of the vascular wall.1 Local irradiation has been shown to inhibit neointima formation in animal experiments.2 3 4 Initial trials using γ-sources indicated the clinical effectiveness of coronary brachytherapy.5 6 In a first randomized trial, Teirstein et al6 showed a significant reduction of the restenosis rate after intracoronary γ-irradiation of dilated restenotic lesions. Recently, they demonstrated the persistence of the initial effect and a significantly lower event rate after 2 years.7 Promising angiographic results have also been reported with 90Sr/90Y β-irradiation.8 9 A procedural advantage of β-radiation is its low penetration depth, minimizing the ionizing exposure of both the patient and the operator. However, the fast decrease of β-radiation energy within 2 to 5 mm has raised the question of an inhomogeneous dose delivery to the vascular tissue and the need for centering the radiation device.
188Re is a high-energy β-emitter available in liquid form. The purpose of this study was to evaluate the feasibility and safety of intracoronary β-irradiation with a liquid 188Re-filled balloon.
This study was approved by the local ethics committee and the German radiation authorities. Inclusion criteria were written informed consent; age 18 to 80 years; ischemia by symptoms or exercise testing; intended angioplasty of a native coronary artery or a bypass graft; and a reference vessel diameter suitable for a 3.0-mm irradiation balloon. Exclusion criteria were acute myocardial infarction, unprotected main stem disease, strong vessel tortuosity, pregnancy, contraindication to aspirin, ticlopidine or heparin, and illness limiting the survival within 6 months.
Radiation Source and Dosimetry
Liquid 188Re is a high-energy β-emitter that is available daily from the 188W/188Re generator (Oak Ridge National Laboratory, Oak Ridge, Tenn) and has a half-life of 17 hours. The β-particles have a maximal energy of 2.12 MeV and a mean energy of 764 keV. A principle γ-ray component of 155 keV accounts for 15% of the radiation intensity and allows excellent control of contamination.
Before this clinical trial, we performed detailed in vitro studies to evaluate the dosimetry of a liquid 188Re-filled angioplasty balloon catheter.10 The dose decrease with increasing distance from the balloon surface was measured and compared with expected values derived from the point kernel function of 188Re in water.11 Very good correlation was demonstrated for the energy dose deposited and the fast dose decrease to 50% at 0.5 mm distance and to 10% at 2.5 mm distance from the balloon surface. From these data, the irradiation times required for a targeted dose can be calculated, dependent on the size of the irradiation balloon and the actual specific volume of 188Re.
Irradiation times were calculated to deliver 15 Gy in a tissue depth of 0.5 mm. This corresponds to a dose of 30 Gy at the surface of the balloon. Potential variations in dose delivery caused by minor differences between the actual and the nominal diameter of the irradiation balloon are <10%.10 12
Carrier-free 188Re (perrhenate) was obtained from a 188W/188Re generator by elution with saline and was concentrated by anion exchange columns to a specific volume of 1.7±0.6 GBq/mL (46.5±16.6 mCi/mL).13 A volume of 1.3 mL was filled into a 3.0-mL syringe shielded with 5-mm plastic. Because of the γ-radiation component, a lead container was used for transportation.
Previously, we have demonstrated that in the exceptional event of balloon rupture, perchlorate can be administered orally to block the thyroid and gastric mucosa and to reduce the radiation burden to an effective dose of 0.16 mSv/MBq 188Re.14 Perchlorate was not used prophylactically.
All patients were pretreated with 100 mg/d aspirin for ≥5 days. They received a 10 000-U bolus of heparin before angioplasty, adjusted to the activated clotting time (>280 seconds). Aspirin (100 mg/d) was continued throughout the study. Patients with stents additionally received 250 mg ticlopidine BID for 6 weeks. Irradiation was carried out after successful angioplasty with or without stenting. Patient inclusion was based on on-line quantitative coronary angiography. Only patients with a vessel size suitable for a 3.0-mm balloon and an ischemic tolerance of ≥1 minute during the preceding angioplasty received irradiation.
For irradiation, the angioplasty balloon was replaced by a noncompliant balloon of the same diameter and the same length equipped with a proximal and distal radio-opaque marker (Tacker, Cordis Europe). During intermittent contrast injections from the guiding catheter, indicating the position of the lesion and of landmark side branches, the deflated balloon was positioned, covering the target lesion and matching the position of the previous angioplasty balloon. The 188Re-filled syringe and an empty 50-mL syringe were connected to the balloon by a 3-way valve (Figure 1⇓). With the empty syringe, a negative pressure was applied to the balloon. Afterward, the valve was closed and the empty syringe was replaced by a stopper. The 3-way valve was operated with a 10-cm plastic stick. For irradiation, the balloon was manually inflated with an approximate inflation pressure of 3 to 4 atm. Because of the low viscosity of the liquid 188Re, this procedure is sufficient for complete filling of the balloon without bubbles, as we know from our in vitro trials. During inflation, the ECG and anginal symptoms were attentively observed. Inflation was stopped and dose delivery was fractionated in the case of severe anginal pain, marked ST-segment changes, decrease of blood pressure, or frequent ventricular arrhythmias. When the irradiation had to be fractionated, the balloon was deflated for 3 to 5 minutes but left in place across the target stenosis to avoid shifts of the balloon and to minimize radiation exposure to other areas. After irradiation, the balloon was removed with forceps, and the whole system including the shielded syringe was brought back to the department of nuclear medicine for decay. Finally, the catheterization laboratory was checked for radiation contamination.
Patients scheduled for coronary angioplasty were informed and gave their written consent for intracoronary irradiation and angiographic follow-up examination. In this safety and feasibility trial, no restrictions were made according to restenosis, bypass grafts, or total occlusions. The inclusion of patients into the study was done after successful angioplasty. Clinical follow-up was performed after 3 months and angiographic follow-up after 6 months.
Quantitative Coronary Angiography
Coronary angiography before and after angioplasty and at follow-up was performed in the same projections of the treated lesion after intracoronary glycerol-trinitrate. Angiographic measurements were done with the Pie Medical software 2.1 (Pie Medical Imaging). Each projection calibration was done from the unfilled guiding catheter.
Continuous variables are presented as mean±SD and compared with the unpaired Mann-Whitney U test or the paired Wilcoxon test. Discrete variables were expressed as counts and percentages and compared by means of χ2 analysis. Statistical significance was set at the 5% α-error level (P<0.05).
Between December 1997 and August 1998, 28 patients received irradiation after successful coronary angioplasty. Lesions were located in the left anterior descending coronary artery (n=9), the circumflex artery (Cx, n=8), the right coronary artery (n=10), and a bypass graft (n=1). In 17 patients, stenting was done before irradiation and in another 2 patients, stenting was performed afterward. Abciximab was given in 7 of 28 patients.
Irradiation was done with 3.0-mm (n=16) and 3.5-mm (n=12) balloons, with a length of 20 mm (n=24), 30 mm (n=3), and 40 mm (n=1), sufficient to cover the whole lesion. The irradiation balloon was selected with the same diameter and length as the angioplasty balloon. In all patients, the prescribed dose of 15 Gy was delivered. The mean irradiation time was 515±199 seconds (range 220 to 990). Irradiation was performed in 1 fraction (n=12), 2 fractions (n=7), 3 fractions (n=7), and 4 fractions (n=2) to limit ischemia. There were no adverse effects of the irradiation procedure except anginal pain and ST-segment changes. There was no radiation leakage within a patient. One minimal leak of the 3-way valve contaminated only the underlying water-resistant drape.
Clinical and Procedural Data
The study collective consisted of 18 men and 10 women with a mean age of 63.4±12.7 years. Four patients had diabetes mellitus, 16 had hyperlipidemia, 18 had hypertension, and 14 had a history of smoking. Single-vessel disease was present in 9 patients, 2-vessel disease in 14 patients, and 3-vessel disease in 5 patients. Left ventricular ejection fraction was 68±15%. Ten patients had a recent myocardial infarction; 11 patients had unstable angina. There were 19 de novo stenoses, 4 occlusions, 3 restenoses, and 2 in-stent restenoses. Lesion morphology included 13 type A/B1 and 15 type B2 lesions. Nonoccluded lesions were predominantly eccentric (n=19). Angioplasty before irradiation was performed with 1 balloon inflation in 4 lesions, with 2 inflations in 8 lesions, and with ≥3 inflations in 16 lesions (mean 3.0±1.8 inflations).
The mean lesion length was 11.8±4.3 mm (range 6.0 to 25.0). Six patients had minor vessel calcifications; no vessel was heavily calcified. Angioplasty resulted in an acute gain of 1.74±0.59 mm. The mean ratio of the irradiation balloon diameter to the reference diameter after angioplasty was 1.17±0.13 mm (range 0.93 to 1.49) and to the largest angioplasty balloon used was 1.07±0.09 mm (range 1.0 to 1.2). The mean difference of the length of the irradiation balloon and the length of initial lesion was 11.1±3.6 mm (range 3.7 to 18.6). The mean ratio of the length of the irradiation balloon to the lesion length was 2.21±0.63 (range 1.23 to 3.36).
There were no in-hospital complications related to the coronary intervention. One 76-year-old male patient died 23 days after stenting and irradiation of the right coronary artery as the result of kidney failure after unsuccessful percutaneous transluminal angioplasty of a renal artery. Autopsy was declined by his relatives. The remaining 27 patients were followed up for 6 months. Clinical follow-up after 3 months revealed no further death, no myocardial infarctions, and no revascularizations. After 6 months, there were still no further deaths but 1 myocardial infarction caused by a late stent thrombosis 109 days after stenting and irradiation. This patient also had received ticlopidine for the protocol specified for 6 weeks, as specified by the protocol. At 6 months, 13 patients were asymptomatic and 13 had stable angina. One asymptomatic female patient with a negative stress test declined follow-up angiography 6 months after angioplasty and irradiation of the Cx.
Follow-up angiography was performed in 26 patients after 177±27 days. Figure 2⇓ shows an example of a good 6-month result after angioplasty and irradiation of the Cx. The dichotomous restenosis rate (>50%) restricted to the target lesion was only 12% (3 of 26). In terms of the lesion type, target lesion restenosis occurred in 1 de novo lesion, in 1 occlusion, and in 1 restenosis.
Additionally, we observed 9 newly developed stenoses (average 78±16%) outside of the target lesion located at the proximal or distal end of the irradiation zone (Figure 3⇓). Thus, the overall stenosis rate at follow-up was 46% (12 of 26). There was no significant change of the reference diameter. Including both target lesion restenoses and newly developed “edge” stenoses at the end of the irradiation zone, the mean late lumen loss was 0.91±0.81 mm and mean late loss index was 0.57±0.57 (Table⇓). Repeat PTCA was performed in a total of 6 (23%) patients: 1 patient with target lesion restenosis and in 5 patients with edge stenoses.
A low restenosis rate was observed in proximal lesions (11%, 1 of 9) and in stented de novo stenoses (21%, 3 of 14). High restenosis rates were seen in the small groups of previously occluded (75%, 3 of 4) or restenosed (50%, 2 of 4) vessels.
To gain further insights into the mechanisms of edge stenoses, we analyzed the ratio of the length of the irradiation balloon to the length of the lesion, that is, the extent to which the irradiation balloon exceeded the stenosed area. The restenosis rate was significantly lower in patients in whom this ratio exceeded 2 compared with those with a ratio below 2 (Figure 4⇓).
Local irradiation from different sources has been shown to inhibit neointima formation in animal experiments.2 3 4 Thus far, 4 human studies of intracoronary brachytherapy have been reported.5 6 7 8 15 Condado et al5 and Teirstein et al6 7 showed a reduction of late lumen loss and a lower restenosis rate at 6 months and during 2 to 3 years of follow-up with γ-irradiation from a 192Ir source. Verin et al15 demonstrated the technical feasibility of intracoronary β-irradiation with an 18-Gy inner arterial surface dose applied from a centered 90Y wire but did not find an obviously reduced restenosis rate at the treated site (6 of 15, 40%). Recently, King et al8 showed a low restenosis rate of 15% (3 of 20) of the target lesion after intracoronary β-irradiation with 12 to 16 Gy at 2-mm distance from the center of a 90Y source. In addition, they found some positive remodeling of the treated segments in 9 of the 20 patients but no aneurysms.
This study is the first report the use of a liquid 188Re-filled balloon catheter for intracoronary irradiation in humans. We demonstrated the technical feasibility (100%) and the safety of this new technique. With the use of standard PTCA technology, coronary 188Re brachytherapy could be performed with low additional technical efforts. Like other high-energy β-emitters, 188Re allows us to apply a high focal radiation dose over a short period of time. Because of its low penetration depth, the use of β-radiation reduces the ionizing exposure of the patient and the operator and requires less effort for radioprotection in the catheterization laboratory compared with γ-sources. Features of 188Re, facilitating intracoronary application, are the high β-energy, its liquid form, and the short half-life of 17 hours with daily availability from the 188W/188Re generator. Filling a conventional balloon catheter with liquid 188Re results in a self-centering irradiation source independent from bending of the artery, cardiac motion, and stenosis morphology. When the size of the radiation balloon is matched to the diameter of the artery, there is direct contact between the radiation source and the inner vessel wall. This results in equal and predefined radiation doses at equidistant points from the luminal surface of the vessel independent from the vessel size. The 188Re-filled balloon therefore avoids the problem of centering a thin radioactive wire within the vessel. However, it does not affect the characteristic fast radial dose decrease of β-radiation in contrast with γ-radiation, resulting in a significantly lower dose at the deeper adventitial layers compared with the luminal surface.
The dose prescribed in this initial feasibility trial of liquid 188Re was chosen from the following considerations: Teirstein et al6 showed a significant effect with 26.5±3.5 Gy at a distance of 1.02±0.16 mm from the ribbon source to the target lesion. This corresponds to our dose of 30 Gy at the balloon surface in contact with the luminal vessel side. Compared with the pilot study of Verin et al15 that found no impressive effect with 18 Gy at the inner vessel side, our dose was chosen ≥50% higher. In this first feasibility trial with 188Re, we took care to avoid radiation induced pseudoaneurysms that previously had been described by Condado et al5 with a recalculated dose of 38.5 Gy delivered at a distance of 1.5 mm from the radioactive wire. Although the risk for developing pseudoaneurysms might be much less with low-penetrating β-radiation and the fact that Condado et al considered it likely that the treated site received up to 92 Gy, we decided to stay below the limit of 38 Gy at the balloon surface. With regard to the fast dose decrease of β-radiation, we accepted a low, potentially inadequate dose in the deeper, adventitial layers. Prior animal studies from Waksman et al16 showed that with a dosage of 14 Gy at 2 mm from the radiation source, the effect of β- and γ-radiation on the suppression of neointima formation was comparable. However, when delivering radiation from a centered wire source, the artery is not expanded to 3.0 mm and therefore the tissue dose at a 2-mm radius from the source center is probably higher at least in some portions of the arterial wall compared with the 15 Gy dose in 0.5-mm tissue depth of this study.
Our 6-month angiographic results after 188Re irradiation were characterized by 2 aspects: The target lesion restenosis rate was only 12%, which is comparable to the results from the BERT feasibility trial,8 showing a 15% restenosis rate at the target lesion. However, a disappointing finding of this feasibility study was the high incidence (35%) of new edge stenoses, increasing the overall restenosis rate to 46%. Including these edge stenoses, late lumen loss was 0.57 in this trial and comparable to the results from Verin et al15 (≈0.50). Such edge stenoses could be due to a low irradiation dose within a traumatized vessel segment. A proliferative tissue response has been described with low dose 192Ir irradiation of 10 Gy applied before balloon overdilation in a pig model4 and after external x-ray application.17 However, edge stenoses have not been described from the BERT trial,8 9 18 although the authors applied very similar β-irradiation doses. Recently Sabaté et al19 analyzed the volumetric changes after irradiation according to the BERT trial with the use of 3D ultrasound. They also found a reduced effect at the edges with a decrease in luminal volume due to an increase in plaque volume and no net change in external elastic membrane volume, but only 1 of 21 lesions showed an angiographic edge stenosis.
Our data indicate that the occurrence of edge stenoses is significantly influenced by the extent to which the irradiated segment is longer than the dilated lesion. In this study, the 188Re balloon had the same length as the dilation balloon. With a mean balloon length of 21.8±4.8 mm (range 20 to 40 mm), the irradiated segment was shorter than in the BERT trial, in which a 30-mm delivery system was used. Mean lesion length was slightly longer (11 mm) in this study compared with BERT (9 mm). Possibly, a decreasing dose at the ends of the balloon applied in an already traumatized tissue segment triggers a proliferative response. In this study, a mean of 3.0±1.9 balloon inflations were performed during angioplasty before irradiation. Because of slight shifts of the angioplasty balloon, the traumatized segment can be expected to be even longer than the simple length of the dilation balloon.
There are some other factors that potentially reduce the effective tissue dose. Two thirds of the patients received coronary stents before irradiation. Experimental studies indicate that stents reduce the radiation dose by 4% to 14%, depending on the stent type.20 We did not correct the irradiation dose with and without stenting. However, assuming a maximum dose reduction of 14%, the remaining dose of 12.9 Gy would still have been in the dose range applied in the BERT trial. Furthermore, β-radiation is shielded by calcium. Although there were no heavily calcified lesions in this study, it should be mentioned that angiography has a limited sensitivity for inhomogeneous, focal calcific deposits.21 As a safety and a feasibility study, this trial was not restricted to certain lesion types. Although the case numbers were small, restenoses were frequent (5 of 8) in previously restenosed and occluded lesions. Because of the fast dose decrease of β-radiation within the vessel wall, it cannot be excluded that in lesions with a high plaque burden the effective dose to the proliferating tissue elements might have been too low.
The number of patients in this feasibility trial was limited. The study did not include a control group and thus was not randomized. Because of the lack of the control group, we could not prove the efficacy of 188Re irradiation, and it remains open whether the low target lesion restenosis rate is indeed due to the irradiation procedure. Concerning the safety of this new method, our study is limited to the procedural results and 6-month follow-up. Because of the geometry of the 188Re-filled balloon, the dose was homogenous at the inner vessel surface. However, we did not use intravascular ultrasound to evaluate plaque burden and did not adapt the prescribed β-irradiation dose to vessel wall thickness.
Hypoxia reduces the biological effect of radiation in malignant tissue.22 This raises the question whether radiation delivered from an inflated balloon is less effective than irradiation from a nonoccluding device caused by compression of the vessel wall and compromise of the vasa vasorum. However, our previous animal studies with the 188Re-filled balloon showed a significant reduction of neointimal hyperplasia with 7.5 Gy in 0.5-mm tissue depths, which is comparable to the data from nonoccluding β-emitting catheters.2 Furthermore, it is unclear which degree of hypoxia is reached in the vessel wall and whether oxygen concentration in the target cells is low enough (0.02%) to diminish radiation-induced formation of DNA strand breaks,22 taking into account the short duration of balloon inflations (mean 262±140 seconds), the low inflation pressure (3 to 4 atm), and the anatomy of vasa vasorum in the adventitia.23
With the use of a liquid- filled, radioactive balloon, there is always a decreasing irradiation dose at the edges of the balloon. From our data, delivery of such a decreasing dose in a diseased segment traumatized by angioplasty predisposes for the development of edge stenosis. Usage of longer radioactive balloons exceeding the traumatized segment should limit the occurrence of edges stenosis, shifting the dose decrease at the edge of the balloon into nontraumatized segments. Low-pressure inflation of the radioactive balloon alone will not cause new vessel trauma, as we know from vessel occlusion during angioscopy.24
Coronary irradiation with a 188Re-filled balloon is technically feasible and safe. The target lesion restenosis rate was low. The observed edge stenoses appear to be avoidable by increasing the length of the irradiated segment. A randomized trial is warranted to evaluate the efficacy of 188Re brachytherapy.
This study was supported by the Deutsche Forschungsgemeinschaft (SFB 451-B6).
- Received September 29, 1999.
- Revision received December 1, 1999.
- Accepted December 22, 1999.
- Copyright © 2000 by American Heart Association
Waksman R, Robinson KA, Crocker IR, et al. Intracoronary low-dose β-irradiation inhibits neointima formation after coronary artery balloon injury in the swine restenosis model. Circulation. 1995;92:3025–3031.
Condado JA, Waksman R, Gurdiel O, et al. Long-term angiographic and clinical outcome after percutaneous transluminal coronary angioplasty and intracoronary radiation therapy in humans. Circulation. 1997;96:727–732.
Teirstein PS, Massullo V, Jani S, et al. Two-year follow-up after catheter-based radiotherapy to inhibit coronary restenosis. Circulation. 1999;99:243–247.
King SB, Williams DO, Chougule P, et al. Endovascular beta-radiation to reduce restenosis after coronary balloon angioplasty: Results of the Beta Energy Restenosis Trial (BERT). Circulation. 1998;97:2025–2030.
Meerkin D, Tardif JC, Crocker IR, et al. Effects of intracoronary β-radiation therapy after coronary angioplasty: an intravascular ultrasound study. Circulation. 1999;99:1660–1665.
Verin V, Urban P, Popowski Y, et al. Feasibility of intracoronary β-irradiation to reduce restenosis after balloon angioplasty: a clinical pilot study. Circulation. 1997;95:1138–1144.
Waksman R, Robinson KA, Crocker IR, et al. Endovascular low-dose irradiation inhibits neointima formation after coronary artery balloon injury in swine: a possible role for radiation therapy in restenosis prevention. Circulation. 1995;91:1533–1539.
Sabate M, Serruys PW, van der Giessen WJ, et al. Geometric vascular remodeling after balloon angioplasty and beta- radiation therapy: a 3-dimensional intravascular ultrasound study. Circulation. 1999;100:1182–1188.
Amols HI, Trichter F, Weinberger J. Intracoronary radiation for prevention of restenosis: dose perturbations caused by stents. Circulation. 1998;98:2024–2029.