Abstract
β-radiation is a novel potential energy source for the creation of myocardial lesions. While the feasibility of delivering β-radiation via a deflectable transvenous catheter has been described, dose effects and the time course of histopathological changes have not been previously assessed. The purpose of this study was to characterize pathological aspects of cardiac lesions induced by β-radiation in an animal model at various stages of evolution and in response to different dose exposures.
Nine dogs and one pig were studied. The cavotricuspid isthmus, antrum of pulmonary veins (PVs), and mitral isthmus were irradiated (25, 50, 75, or 100 Gy) with strontium–yttrium-90, delivered via a deflectable catheter (cavotricuspid isthmus and mitral isthmus) or a double-loop catheter (antrum of PVs). Eighteen lesions were created. Animals were sacrificed at 2 weeks, 1 month, 3 months, or 6 months. Lesions were processed for morphometric histopathological analyses. Over the first month, lesions were characterized by inflammation, haemorrhage, and myocyte necrosis. Thereafter, fibrotic replacement was predominant. Transmurality of lesions was observed in 50% of cases, with no dose–response effect (P = 0.976). Surface fibrin thrombus was present in 50% of cases and was essentially limited to lesions assessed within the first month. No neighbouring injury or pulmonary venous stenosis was observed.
Atrial lesions created by β-radiation are characterized by an inflammatory phase with surface fibrin thrombosis during the first month and replacement fibrosis thereafter. No appreciable dose–response effect was noted within the 25–100 Gy range tested.
Introduction
Radiofrequency is the most commonly utilized energy source for creating intracardiac lesions for the treatment of various arrhythmias. Ablation procedures have become increasingly complex, as a growing number of targeted arrhythmia substrates necessitate the creation of long, linear lesions. These procedures can be time consuming, and often require extensive catheter manipulation in difficult-to-access areas. Of particular concern are the challenges sometimes encountered when attempting to create contiguous ablation lines in the left atrium, including to isolate the pulmonary veins (PVs).1,2 In the case of discontinuity within a linear lesion, proarrhythmic effects may be observed.2 Moreover, radiofrequency ablation is painful, can be responsible for thrombus formation or charring, and can damage neighbouring structures such as the oesophagus or phrenic nerve.3,4
Alternative energy sources such as cryotherapy and high-intensity-focused ultrasound have, therefore, been developed.5,6 In creating linear or circumferential lesions, each has its limitations, particularly with regard to damaging adjacent structures. In this light, radiation-based ablative sources were explored. On the basis of experience with β-radiation to treat in-stent restenosis, it has been shown that low tissue penetration permits the delivery of a high-radiation dose to adjacent myocardium, producing cell death with minimal neighbouring tissue exposure.7,8 When applied to the cavotricuspid isthmus, β-radiation can create bidirectional block and contiguous linear lesions as assessed by histopathological analyses in a canine model. Additionally, intravascular β-radiation can induce transmural necrosis and fibrosis of PV myocardial sleeves.9,10 These studies demonstrated the feasibility of using a non-contact-dependent radiation source to create ablation lesions in the myocardium with the desired electrophysiological effect, i.e. conduction block.
The interest in radiation as an ablative energy source has been heightened by the recent development of an innovative non-invasive radiation delivery approach using stereotactic robotic radio surgery to produce atrial lesions.11 Although this nascent energy source shows promise, the nature of radiation lesions remain to be characterized. More specifically, studies regarding the time course for injury, histopathological delineation of lesions, and dosing effects are lacking. Such information is critical in refining the technology in order to optimize parameters to maximize safety and efficacy. The purpose of the present study was, therefore, to systematically assess the histopathological evolution of cardiac lesions created by β-radiation over time and to explore a potential dose–response effect.
Methods
β-Radiation catheters design
The β-radiation delivery system developed by Novoste (Beta-Cath Delivery Catheter, Novoste Corp., Norcross, GA, USA) is similar to that used for coronary arteries.12 It consists of a hydraulic delivery system, source train, and catheter. The hydraulic delivery system houses the source train [strontium/yttrium ([Sr/Y)-90] and deploys it to the distal catheter. Two prototype catheters were specifically tailored for the purpose of this experiment (Figure 1). The first catheter used for linear lesions (i.e. cavotricuspid isthmus and mitral isthmus) was a 7F deflectable catheter with two radiopaque markers identifying the distal and proximal limits of the radiation source train (60 mm in length) (Figure 1A). A 7F double loop over-the-wire catheter was used for PV applications (Figure 1B–D).
Catheter designs. (A) Fluoroscopic lateral view of the 7F steerable Beta-Cath Delivery Catheter used for linear lesions. Note the two radiopaque markers identifying distal and proximal limits of the radiation source train (40 mm in length). (B and C) Lateral (B) and frontal (C) views of the 7F double loop over-the-wire Beta-Cath Delivery Catheter used for pulmonary vein applications. Note the two radiopaque markers (B) identifying the distal and proximal limits of the radiation source train. (D) Fluoroscopal anterioposterior view of the 7F double loop over-the-wire Beta-Cath Delivery Catheter.
Animal preparation and β-radiation application
Nine healthy mongrel dogs were induced with Pentothal 25 mg/kg, intubated, and maintained on a respirator with halothane. In addition, a single pig was studied to confirm similar effects of β-radiation on larger calibre PVs. Through two femoral venous punctures, two sheaths (9F and 10F) were introduced.
Cavotricuspid isthmus applications
Cavotricuspid isthmus ablation using the Beta-Cath has been previously described.10 In short, the steerable catheter was advanced through the 9F sheath. The distal marker of the catheter was first placed in the right ventricle past the tricuspid valve, with the catheter deflected across the valve to ensure apposition against the cavotricuspid isthmus. The radiation source train was advanced into position by use of the hydraulic transfer device, with the distal marker indicating the position of the distal end of the 40-mm source train. A first irradiation was performed in this position. The catheter was then withdrawn until the proximal marker was in the inferior vena cava while the same catheter apposition against the cavotricuspid isthmus was maintained. With the catheter in this more proximal isthmus position, the radiation source train was once again advanced to the catheter tip, with the proximal marker indicating that the proximal end of the source train was in the inferior vena cava. Irradiation in this second position was performed.
Pulmonary vein and mitral isthmus applications
A 8F transseptal sheath (Fast-Cath SL-1, Daig, St Jude Medical, Minnetonka, MN, USA) was advanced through femoral access. Once transseptal puncture was successfully achieved, an intravenous non-fractionated heparin bolus was administrated to maintain an activated clotting time between 250 and 350 s. Once the transseptal sheath was in the left atrium, a selective PV angiography was performed to identify the targeted PV ostium and proximal trunk. The Beta-Cath double-loop delivery catheter was advanced through the transseptal sheath, and the guidewire was inserted into the right or left superior PV. The catheter was then advanced over the wire in order to position the source train (the double loop) at the PV antrum. The radiation source train was advanced into position by use of the hydraulic transfer device, and the application duration was maintained in order to deliver the prescribed dose. The double-loop catheter was then removed and replaced with the deflectable catheter. It was positioned in the area of the mitral isthmus under fluoroscopic guidance. When a stable position was achieved, the radiation source train was advanced into position by the use of the hydraulic transfer device, and the application duration was maintained in order to deliver the prescribed dose.
By design, the prescribed dose was delivered at a point 2 mm from the centre of the source axis. The total delivered dose was determined by the duration of exposure to the radiation source. Therefore, the radiation source train was maintained for either 4 min, resulting in a delivered dose of 25 Gy, 8 min for 50 Gy, 12 min for 75 Gy, and 16 min for 100 Gy in accordance with the pre-defined protocol. The catheter was not displaced during these applications. After each irradiation, the source train was drawn back into the transfer device. Upon completion of the intervention, catheters were removed and the animals extubated and housed for the pre-determined follow-up period (2 weeks, 1, 3, or 6 months). At the end of the follow-up, animals were euthanized with a lethal injection of pentobarbital. The study was approved by the Montreal Heart Institute Animal Care and Use Committee. All animals were cared for in accordance with the ‘Principles of Laboratory Animal Care’ formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences.
Pathological preparation
After euthanasia, hearts and lungs were explanted and fixed in formaldehyde, then sent to Dr Renu Virmani at the Armed Force Institute of Pathology (Washington, DC, USA) for histological evaluation. Digital photographs (macro and micro) were obtained prior to tissue sampling. For histological evaluation, 5–10 serial sections (3–4 mm thick) from each of the treatment sites were obtained (cavotricuspid isthmus, left superior PV, right superior PV, right inferior PV, mitral isthmus). In addition, representative sections were obtained from the lungs and labelled; right upper lobe, right middle lobe, right lower lobe, left upper lobe, and left lower lobe and submitted for histology. All tissue samples and sections were dehydrated in a graded series of ethanol and embedded in paraffin. The sections were then cut at 4–6 μm, using a rotary microtome, mounted on a charged slide, and stained with haematoxylin–eosin, Movat's pentachrome and Masson's trichrome. All sections were examined by light microscopy for the presence of fibrosis, inflammation, haemorrhage, thrombus, and myocyte necrosis.
Statistical analyses
Given that several ablation lesions were created in each animal, the effect of dose on lesion transmurality considered the non-independent data structure. A generalized estimating equations model was used for cluster sampling data by specifying link (i.e. logit) and distribution (i.e. binomial) functions, with an exchangeable correlation structure. A P value <0.05 was considered statistically significant. Statistical testing was performed using SAS software version 9.2 (SAS Institute, Cary, NC, USA).
Results
Macroscopic analysis
An initial macroscopic inspection was performed on all cardiac chambers that were irradiated, as well as all adjacent structures. No macroscopic changes were seen in the lungs or oesophageal regions, and no pericardial inflammation or effusions were noted on gross inspection. In terms of actual lesion sites, 100 Gy applications with sacrifice at 2 weeks induced easily visualized dark-tan lesions (Figure 2A). All other lesions (25, 50, or 75 Gy with sacrifice at 1, 3, or 6 months) resulted in either a pale discolouration of the endocardial surface (Figure 2B) or no clear evidence of tissue injury.
Macroscopic lesions in the left atrium (A) and cavotricuspid isthmus (B). (A) Macroscopic view of the left atrium (animal 3; 100 Gy; 2 weeks) showing a 1.5cm × 0.4 cm dark-tan lesion (1) adjacent to the right inferior pulmonary vein (RIPV) ostium and above the mitral valve annulus. In addition, there is a small 1.5 cm × 1.0 cm, hemispherical, haemorrhagic lesion of the posterior leaflet (atrial side) of the mitral valve (2). (B) Macroscopic view of the right atrium (animal 10; 25 Gy; 6 months) depicting the cavotricuspid isthmus and interatrial septum. Gross inspection shows a focal (non-discrete) area of pale discolouration in the cavotricuspid isthmus. FO, foramen ovale; IVC, inferior vena cava.
Histological analysis
Lesions were discrete, with varying degrees of haemorrhage, inflammation, necrosis, and fibrosis (Figure 3). Main histological findings according to time of sacrifice are summarized in Table 1. Acute lesions (2 weeks) were characterized by damage to the endothelial cell layer, with surface erosion, organizing fibrin thrombi on the atrial surface, myocyte necrosis, and collagen deposition in a proteoglycan rich matrix. Acute inflammation involved at least 50% of the atrial wall thickness with 100 Gy radiation. At 1 month, haemorrhage and myocyte necrosis persisted, but inflammatory changes were no longer seen. Fibrotic replacement was observed at this stage. Thrombus was still present on the endocardial surface, but was absent thereafter. At later time points (3 and 6 months), changes consisted mostly of localized endocardial thickening with underlying fibrosis. No dose–response effect was observed with regard to transmurality of lesions [odds ratio for a 25 Gy increase in dose 1.0, 95% confidence interval (0.7, 1.4), P= 0.976].
Summary of histological findings
| Animal | Race | Site | Dose | Sacrifice | Thrombus | Haemorrhage | Transmural | Inflammation | Necrosis | Fibrosis | Metaplasia |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Dog | RSPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | − | − |
| 1 | Dog | MI | 100 Gy | 2 weeks | + | + | + | + | ++ | + | − |
| 2 | Dog | RSPV | 100 Gy | 2 weeks | + | + | − | ++ | + | + | − |
| 2 | Dog | MI | 100 Gy | 2 weeks | + | + | − | + | + | + | − |
| 3 | Pig | RIPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 3 | Pig | CTI | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 4 | Dog | LSPV | 50 Gy | 1 month | + | + | − | − | + | + | − |
| 5 | Dog | RSPV | 50 Gy | 1 month | + | + | + | − | − | + | − |
| 6 | Dog | RSPV | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 6 | Dog | CTI | 75 Gy | 3 months | − | − | − | + | + | + | − |
| 7 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 7 | Dog | CTI | 25 Gy | 3 months | − | − | − | − | − | + | − |
| 8 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 8 | Dog | CTI | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 9 | Dog | LSPV | 50 Gy | 6 months | − | − | + | − | − | ++ | + |
| 9 | Dog | CTI | 75 Gy | 6 months | − | + | − | − | − | ++ | − |
| 10 | Dog | RSPV | 50 Gy | 6 months | + | − | − | − | − | ++ | + |
| 10 | Dog | CTI | 25 Gy | 6 months | − | − | + | − | − | ++ | − |
| Animal | Race | Site | Dose | Sacrifice | Thrombus | Haemorrhage | Transmural | Inflammation | Necrosis | Fibrosis | Metaplasia |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Dog | RSPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | − | − |
| 1 | Dog | MI | 100 Gy | 2 weeks | + | + | + | + | ++ | + | − |
| 2 | Dog | RSPV | 100 Gy | 2 weeks | + | + | − | ++ | + | + | − |
| 2 | Dog | MI | 100 Gy | 2 weeks | + | + | − | + | + | + | − |
| 3 | Pig | RIPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 3 | Pig | CTI | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 4 | Dog | LSPV | 50 Gy | 1 month | + | + | − | − | + | + | − |
| 5 | Dog | RSPV | 50 Gy | 1 month | + | + | + | − | − | + | − |
| 6 | Dog | RSPV | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 6 | Dog | CTI | 75 Gy | 3 months | − | − | − | + | + | + | − |
| 7 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 7 | Dog | CTI | 25 Gy | 3 months | − | − | − | − | − | + | − |
| 8 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 8 | Dog | CTI | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 9 | Dog | LSPV | 50 Gy | 6 months | − | − | + | − | − | ++ | + |
| 9 | Dog | CTI | 75 Gy | 6 months | − | + | − | − | − | ++ | − |
| 10 | Dog | RSPV | 50 Gy | 6 months | + | − | − | − | − | ++ | + |
| 10 | Dog | CTI | 25 Gy | 6 months | − | − | + | − | − | ++ | − |
RSPV, right superior pulmonary vein; MI, mitral isthmus; RIPV, right inferior pulmonary vein; CTI, cavotricuspid isthmus; LSPV, left superior pulmonary vein; Metaplasia refers to cartilaginous endocardial metaplasia; Transmural, transmural lesion;.− , absent; +, mild to moderate; ++, extensive.
Summary of histological findings
| Animal | Race | Site | Dose | Sacrifice | Thrombus | Haemorrhage | Transmural | Inflammation | Necrosis | Fibrosis | Metaplasia |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Dog | RSPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | − | − |
| 1 | Dog | MI | 100 Gy | 2 weeks | + | + | + | + | ++ | + | − |
| 2 | Dog | RSPV | 100 Gy | 2 weeks | + | + | − | ++ | + | + | − |
| 2 | Dog | MI | 100 Gy | 2 weeks | + | + | − | + | + | + | − |
| 3 | Pig | RIPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 3 | Pig | CTI | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 4 | Dog | LSPV | 50 Gy | 1 month | + | + | − | − | + | + | − |
| 5 | Dog | RSPV | 50 Gy | 1 month | + | + | + | − | − | + | − |
| 6 | Dog | RSPV | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 6 | Dog | CTI | 75 Gy | 3 months | − | − | − | + | + | + | − |
| 7 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 7 | Dog | CTI | 25 Gy | 3 months | − | − | − | − | − | + | − |
| 8 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 8 | Dog | CTI | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 9 | Dog | LSPV | 50 Gy | 6 months | − | − | + | − | − | ++ | + |
| 9 | Dog | CTI | 75 Gy | 6 months | − | + | − | − | − | ++ | − |
| 10 | Dog | RSPV | 50 Gy | 6 months | + | − | − | − | − | ++ | + |
| 10 | Dog | CTI | 25 Gy | 6 months | − | − | + | − | − | ++ | − |
| Animal | Race | Site | Dose | Sacrifice | Thrombus | Haemorrhage | Transmural | Inflammation | Necrosis | Fibrosis | Metaplasia |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Dog | RSPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | − | − |
| 1 | Dog | MI | 100 Gy | 2 weeks | + | + | + | + | ++ | + | − |
| 2 | Dog | RSPV | 100 Gy | 2 weeks | + | + | − | ++ | + | + | − |
| 2 | Dog | MI | 100 Gy | 2 weeks | + | + | − | + | + | + | − |
| 3 | Pig | RIPV | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 3 | Pig | CTI | 100 Gy | 2 weeks | ++ | + | + | ++ | + | + | − |
| 4 | Dog | LSPV | 50 Gy | 1 month | + | + | − | − | + | + | − |
| 5 | Dog | RSPV | 50 Gy | 1 month | + | + | + | − | − | + | − |
| 6 | Dog | RSPV | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 6 | Dog | CTI | 75 Gy | 3 months | − | − | − | + | + | + | − |
| 7 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 7 | Dog | CTI | 25 Gy | 3 months | − | − | − | − | − | + | − |
| 8 | Dog | RSPV | 50 Gy | 3 months | − | − | + | − | − | ++ | − |
| 8 | Dog | CTI | 50 Gy | 3 months | − | − | − | − | − | + | − |
| 9 | Dog | LSPV | 50 Gy | 6 months | − | − | + | − | − | ++ | + |
| 9 | Dog | CTI | 75 Gy | 6 months | − | + | − | − | − | ++ | − |
| 10 | Dog | RSPV | 50 Gy | 6 months | + | − | − | − | − | ++ | + |
| 10 | Dog | CTI | 25 Gy | 6 months | − | − | + | − | − | ++ | − |
RSPV, right superior pulmonary vein; MI, mitral isthmus; RIPV, right inferior pulmonary vein; CTI, cavotricuspid isthmus; LSPV, left superior pulmonary vein; Metaplasia refers to cartilaginous endocardial metaplasia; Transmural, transmural lesion;.− , absent; +, mild to moderate; ++, extensive.
Microscopic views at different magnifications in right and left atria. (A) (animal 7; 25 Gy/3 months): low-power (Masson-stained) representative section from the right atrium in the area of the isthmus, showing marked transmural fibrosis without inflammation, thrombus, haemorrhage, or necrosis. (B) (animal 3; 100 Gy/2 weeks): low-power (Masson-stained) representative section of the right atrium lesion in the area of the isthmus just above the mitral valve annulus showing extensive transmural fibrosis consisting of surface fibrin thrombus with underlying marked chronic inflammation (lymphocytes, macrophages, and giant cells), haemorrhage, and myocyte necrosis. (C) (animal 1; 100 Gy/2 weeks): higher magnification compared with A and B, haematoxylin and eosin (H&E) stained. This representative section from the left atrium adjacent to the right superior pulmonary vein shows an extensive transmural lesion consisting of surface fibrin thrombus (T) with transmural chronic inflammation (stars) in the underlying subendocardial tissue, and moderate myocyte necrosis (N) and haemorrhage. (D) (animal 1; 100 Gy/2 weeks): higher magnification compared with C (H&E stained section). This representative histological view of the left posterior atrial wall shows acute and chronic inflammation with myocyte necrosis.
Surrounding structures
Most lesions were localized, discrete, and continuous/contiguous in multiple sections. There was no microscopic evidence of pericardial extension at the site of irradiation. Two hearts (animals 9 and 10) showed localized pericardial thickening remote from the radiation site, involving right and left atrial appendages. These changes were deemed most likely not to be related to radiation. The radiation effect (100 Gy dose) on the mitral valve at the 2-week time point (animals 1 and 2) consisted of focal haemorrhage and minimal surface thrombosis. Furthermore, sections of the coronary arteries showed no radiation effects and focal sampling of the conduction system was within normal limits (animal 9 had minimal fibrosis in the atrioventricular node considered to be normal). Overall, radiation lesions in the left atrium were localized to the orifices of the right superior PV, left superior PV, or right inferior PV. No animal had PV stenosis. The histopathological changes in the lungs consisted of occasional inflammatory infiltrates. A single pulmonary artery in two animals showed the presence of a clot. These clots may have arisen as in situ thrombi or as embolic events from the use of catheters and/or radiation lesions on the right side of the heart.
Discussion
This study describes pathological effects of 18 β-radiation applications performed with transvenous access in canine right and left atria, using different catheter shapes, application sites, and dose exposures. Additionally, sacrifice times varied from 2 weeks to 6 months in order to evaluate the time course of changes. Importantly, no neighbouring injury was observed in pericardium, lungs, or coronary vessels. No PV stenosis occurred. To our knowledge, it is the first study to describe the progressive cardiac histological changes seen during the evolution of β-radiation induced lesions. Our results suggest that the histological process can be divided into two phases. Over the course of the first month, inflammation, haemorrhage, and myocyte necrosis were observed. After the first month, fibrotic replacement was predominant. This time course appears similar to radiofrequency lesions, as acute lesions are characterized by coagulation necrosis with surrounding haemorrhage and acute inflammation. The lesions mature within 2 months, with fibrosis, granulation tissue, chronic inflammatory infiltrates, and significant volume contraction.13
β-Radiation for the creation of cardiac lesions
Various radiation effects on the heart have been described. For example, radiation-associated cardiac disease has been seen in patients receiving high-dose external beam radiation for various types of cancer, as well as in atomic bomb survivors. Acute injury is often manifested by pericarditis. Injury effects months to years post-radiotherapy include ventricular dysfunction, valvular disease, coronary artery disease, and myocardial infarction.14 Conduction disturbances have also been noted.
Guerra et al.10 first showed that catheters specifically designed to deliver β-radiation to the cavotricuspid isthmus could effectively create bidirectional conduction block with contiguous linear lesions. Interestingly, bidirectional block was not seen acutely but developed 1 week post-application. Perez-Castellan et al. demonstrated the feasibility of creating lesions in the PV ostia of minipigs using an intravascular catheter delivering β-radiation.14 Consistent with our findings, no PV stenosis was observed.
The concept of creating specific areas of conduction block in the myocardium was further explored using a novel stereotactic robotic approach designed to deliver external radiation beams in a non-invasive fashion. In this albeit limited report, focused external beam radiation was shown to produce cavotricuspid isthmus block, atrioventricular nodal block, and a reduction in voltages at the PV–left atrial junction in 16 miniswine. Lesions were created in a manner akin to that used to deliver radiotherapy to neoplasms. The authors postulated that this form of therapy could be of use in patients for whom a traditional approach (invasive, involving sedation, or anaesthesia) is prohibitive or for patients in whom the arrhythmic substrate is not easily accessible. Should these novel methods develop further, information about the time course and histopathological effects of radiation on the myocardium, such as described in this study, will prove useful.
Lesion transmurality
In exploring a potential dose–response effect, delivering more than a 25 Gy application was not associated with a higher likelihood of creating transmural lesions. There were, however, several factors that complicate the assessment of lesion transmurality. First, sacrifices were carried out after a minimum of 2 weeks and a maximum of 6 months. Chronic lesions may be more difficult to locate on macroscopic histopathology, making it equally difficult to select the ideal site for serial sections. This difficulty is compounded by the fact that radiation lesions are created by gradual fibrotic replacement as opposed to a thermal burn (e.g. electrophysiological effects typically occur only 1 week post-ablation).10 Finally, radiation lesions, contrary to energy sources such as cryoenergy, are fairly irregular in shape and distribution. Thus, serial sections can underestimate the degree of transmurality in certain instances.
In a prior study of seven animals, most lesions were confirmed to be transmural when a 50 Gy dose was used in all but one (25 Gy) dog.10 In contrast, transmurality was observed in 6 of 10 lesions (60 Gy) performed at the PV ostium by Perez-Castellano et al.9 In the stereotactic robotic radiosurgical study,11 it appeared that when the targeted ablation area encompassed the entire myocardial wall, histological changes were transmural with doses ranging from 32 to 80 Gy. Taken together, these results suggest no additional gains with regard to transmurality of lesions beyond 25–50 Gy.
Thrombus formation
In our study, surface fibrin thrombus was present on all ablation lesions at 2 weeks and 1 month. Only one thrombus was observed thereafter (i.e. left-sided at 6 months). This pattern is similar to radiofrequency energy that results in acute disruption of the integrity of the endothelial cell layer with thrombus formation. By design, 50 or 100 Gy was delivered for all lesions within the first month, precluding extrapolations to 25 Gy doses. Similarly, Perez-Castellano et al.9 found endothelial disruption with fibrin thrombus at the surface of acute lesions. Consistently, studies reporting more chronic lesions have likewise noted a low to non-existent incidence of surface thrombosis.10,11
The remote possibility that a late thrombus may occur remains of concern, particularly for left-sided lesions. While underlying mechanisms are unknown, it may be speculated that delayed radiation effects are implicated. Delayed coronary stent thrombosis is occasionally observed with brachytherapy.15 At the luminal surface, brachytherapy is associated with delayed re-endothelization, medial thinning, areas of fibrinoid necrosis, and loss of smooth muscle cell nuclei, potentially explaining the increased risk of late coronary thrombosis.15,16 It is, therefore, possible that delayed re-endothelialization may contribute to late surface thrombosis.
Clinical perspectives
Our study suggests that β-radiation could effectively create ablation lesions in myocardial tissue. Since radiation is delivered from the centre of the source axis, direct contact with myocardial tissue is not required. However, proximity remains important since delivered radiation decreases with increasing distance from the source. Potential advantages may include the ability to treat difficult to access regions and the development of non-invasive therapies.11 In this respect, 25–50 Gy seems adequate to induce transmural lesions. An important drawback, however, is the prolonged time course before electrophysiological effects become manifest. This precludes the possibility of monitoring the effectiveness of any given lesion during ablation. In terms of thrombogenicity, this energy source appears comparable with radiofrequency, with similar implications with respect to antithrombotic or anticoagulant therapy post-ablation.
Limitations
The study contains a limited number of ablation lesions and parameters tested. The ‘prescribed dose’, by definition, is the dose received by the tissue at 2 mm from the centre of the source axis. This dose may be considered an approximation, since in vivo variations in the dose delivered to various parts of the endocardium may be substantial, reflecting in part the actual distance between the catheter and myocardial tissue. Dose calculations were performed in idealized situations and with anatomical assumptions that may not always reflect the clinical situation. Finally, caution must be exerted in the extrapolation of animal studies to humans. While such research may provide useful information and define priorities for clinical studies, reliable quantitative estimates of human risk cannot be inferred.
Conclusion
β-radiation can produce transmural lesions in right and left atria with different catheter configurations and radiation doses. During the first month post-ablation, lesions are inflammatory and exudative, with surface thrombosis. Fibrotic replacement occurs thereafter, with scar formation. Well-formed lesions are produced with applications as low as 25 Gy. Higher doses do not result in an appreciable dose–response effect. However, even doses as high as 100 Gy are not associated with injury to neighbouring structures. The absence of an immediate electrophysiological effect and lack of a dose–response effect, both of which are traditionally used to guide ablation procedures, are identified limitations of this energy source.
Conflict of interest: At the time of the study, R.B. was VP of Medical Affairs at Novoste Corp and P.G.G. was a consultant for Novoste. The authors have no other potential conflicts of interest to disclose.
