|
|
||||||||
Ann Thorac Surg 2005;80:1417-1423
© 2005 The Society of Thoracic Surgeons
a Canadian Surgical Technologies and Advance Robotics, Lawson Health Research Institute, London, Ontario, Canada
b Imaging Group, Robarts Research Institute, London, Ontario, Canada
c Department of Physiology & Pharmacology, London, Ontario, Canada
d Department of Medicine, London, Ontario, Canada
e Department of Anaesthesia, London, Ontario, Canada
f Department of Pathology, London, Ontario, Canada
g Department of Medical Biophysics, University of Western Ontario, and London Health Science Center, London, Ontario, Canada
Accepted for publication March 3, 2005.
* Address reprint requests to Dr Guiraudon, Cardiac Surgery Department, CSTAR Building, LHSC-University Campus, 339 Windermere Road, London, Ontario N6A 5A5, Canada (Email: gguiraud{at}uwo.ca).
| Abstract |
|---|
|
|
|---|
METHODS: We report an acute study of en bloc, cryo-exclusion of the pulmonary vein region in 7 pigs. The left atrial appendage (LAA) was approached via a left thoracotomy. Electrodes were attached to the posterior wall of the left atrium (LPA) and right atrial appendage (RAA) for pacing and electrophysiological monitoring. A modified Surgifrost probe was introduced via the LAA and positioned using transesophageal (TEE) and intracardiac (ICE) echocardiographic guidance to generate encircling cryolesions (3 minutes, 105°C) of the pulmonary vein region.
RESULTS: A complete two-way block was achieved in 6 pigs and an incomplete block in 1. The excluded segment had very slow idiosyncratic rhythm or was electrically silent. In all pigs before isolation, sustained AF was inducible with the most rapid rhythms and fractionated electrograms recorded from the LA. While sustained AF was induced before exclusion, it was not after isolation in either the exclude or non-excluded segments. All tissue samples taken along the encircling cryolesions had transmural cryolesions on pathological examination.
CONCLUSIONS: We conclude that off-pump, closed heart, beating, intracardiac AF surgery is feasible, reliable and can duplicate the accuracy and precision of the open-heart approach. However, further developments are needed to make this novel approach an alternative to current approaches for catheter ablation.
| Introduction |
|---|
|
|
|---|
Minimally-invasive surgery for lone atrial fibrillation (AF) has an opportunity to become the intervention of choice if the very high efficacy of the conventional open heart approaches [912] could be achieved with minimal surgical invasiveness [13, 14]. This goal requires an off-pump, closed, beating heart intervention with minimal surgical trauma in terms of access.
Left atrial surgery for AF is currently performed under those conditions using an epicardial approach [15, 16]. However, the epicardial approach has inherent limitations in terms of access due to the complex anatomy of the posterior left atrium and pericardial sac, and in terms of energy delivery with limited effectiveness of some energy sources to attain transmural lesions [1719]. Despite recent advances in technology, the epicardial approach does not duplicate the effectiveness of open heart surgical techniques [15, 16].
The intracardiac approach is not associated with any limitations in term of access of the entire left atrium but has challenges in terms of safe intracardiac access, absence of direct vision, navigation of tools and effective energy delivery.
We report a feasibly study of en bloc exclusion of the pulmonary vein region in the pig using closed heart, off-pump, beating, intracardiac surgery with cryoablation.
| Material and Methods |
|---|
|
|
|---|
Surgical Preparation
The 7 animals were tranquillized with Telazol and Rompun before transportation to the laboratory. After intubation they were ventilated mechanically and anaesthesia was maintained with nitrous oxide and 1% to 2% Isoflurane during the intervention. They received Buprinorphine analgesic and Lidocaine was given subcutaneously on the site of the incision line followed by neuromuscular blockade with Pancuronium bromide. The heart was exposed via a small left thoracotomy in the fifth intracostal space, and the left lung was deflected and covered by a cotton sponge. The pericardium was opened longitudinally, posterior to the phrenic nerve. The two edges of the pericardial sac were attached to the chest wall with silk sutures to expose the heart, which was supported in the pericardial sling. Right and left atrial quadripolar electrode plaques were attached with monofilament sutures.
The left atrial appendage was excluded using a large Satinski vascular clamp. The atrial appendage was opened longitudinally and its edges were then re-enforced using Dacron® fabric. Before introducing the probe, full heparinization was established.
The porcine left atrial anatomy differs from human anatomy. There is a single ostium with two to four pulmonary vein orifices that is situated on the "posterior inferior" (dorso-caudal) segment of the left atrium. The right and left pulmonary vein orifices are very close to each other and only separated by a crest. The posterior wall is therefore above (cephalad) the pulmonary vein ostium (Fig 1). There is a sizable cuff of left atrium between the pulmonary vein ostium and the mitral valve annulus over the inferior wall that comprises the "left atrial isthmus" [17]. Beneath runs the coronary sinus that is the terminal segment of a patent left superior vena cava.
|
The Bachmann bundle [18] is a thick inter-atrial bundle, whose width and thickness can preclude transmural cryoablation. Therefore the cryoablation line was located dorsal to the bundle over the posterior wall, while the Bachmann bundle was left attached to the nonexcluded left atrial cuff.
The encircling line of block ran from the upper segment of the left atrial appendage, then along the left atrial posterior wall dorsal to the Bachmann bundle, circumscribing the right pulmonary vein orifice. It then traveled along the left atrial inferior wall at a distance from the mitral valve annulus and posterior to the coronary sinus, to end at the inferior segment of the left atrial appendage (Fig 1).
Manipulation of the cryoprobe was guided using echocardiography and the effect was assessed by monitoring of interatrial conduction delay. In particular, the latter was of value in positioning the probe dorsal to the Bachmann bundle. The correct application of the cryoprobe to attain uniform contact and optimal freezing was assessed using a mark on the tool, the pattern of temperature drop and surgeon's experience. The overlapping applications were conducted in a clockwise fashion from the left superior quadrant contiguous to the superior segment the left atrial appendage, to the left inferior quadrant contiguous to the inferior segment of the left atrial appendage (Fig 1).
Image Guidance
The probe for intracardiac echocardiography was introduced after a cut down via the left internal jugular vein into the right atrium and complemented by transesophageal echocardiography.
Cryoablation
Cryoablation was carried using a modified SurgiFrost (CryoCath Inc, Kirkland, Quebec, Canada) cryoprobe introduced intra-atrially via the opening in the left atrial appendage. The cryoprobe was cooled at maximum capacity between 95°C to 135°C for 3 minutes for each application.
Electrophysiological Studies
A custom designed quadripolar electrode plaque was attached to the right atrial appendage (RAA) and another was positioned over the posterior wall of the left atrium (LPA). In the pig the posterior wall of the left atrium is in direct contact with the pulmonary artery trunk. A small pouch was constructed between the two, the plaque electrode was inserted and the pouch closed by a 4-0 monofilament stitch. This maneuver allowed insertion of the plaque at the center of the excluded segment at a distance from the cryoablation lines.
A surface electrocardiogram (ECG) signal, two bipolar electrograms from each of the RAA and LPA electrodes and the arterial blood pressure were continuously monitored by a Quinton EPAmp recording system (Quinton Electrophysiology Corp, Seattle, WA), while selected segments were recorded on optical disk or paper. Right and left atrial pacing thresholds were determined, and the atria were then paced at twice threshold to determine interatrial conduction delays.
Before ablation, burst pacing (2 seconds at 60 Hz) of the RAA and LPA electrodes was used to induce atrial arrhythmia: fibrillation or flutter. During cryoablation, RAA and LPA electrograms were monitored to assess pulmonary vein region isolation. Occurrence of right to left atrial delay, dissociation of the right and left electrograms, or disappearance of the left electrograms was of particular interest. Pacing of the posterior left atrium was used to verify the electrical isolation of the pulmonary vein region.
Following cryoablation, a repeat electrophysiological study was performed 15 min or more after completion of isolation. Stimulation of either the excluded pulmonary vein region or the non-excluded atrial segment (RAA), while recording surface and electrical activity of the other segment, was employed to assess a two-way isolation of the pulmonary vein region. Burst pacing (60 Hz) was also delivered to the RAA and LPA, to assess the impact of isolation on the inducibility of atrial arrhythmias (atrial vulnerability).
Pathological Examination
After completion of isolation and the final electrophysiological study, the heart was excised en bloc, with care being taken to keep the left atrium with its posterior wall and pulmonary veins intact. Macroscopic visual and photographic examination of the left atrium was conducted via the left atrial appendage, the mitral valve orifice after partial resection of the left ventricle and/or via a small window constructed through the posterior wall. The heart was subsequently fixed in 10% buffered formalin solution, prior to being sent to the pathology department, where, after fixation, 5 to 7 tissue samples were obtained along the circular cryoablation line (Fig 2). Alternate tissue slides were stained with Hematoxylin-Eosin and Movat's for examination by the pathologist (CG).
|
| Results |
|---|
|
|
|---|
Image Guidance
Echocardiographic imaging provided critical guidance, with a good correlation to the surgeon's estimated position of the cryoprobe. No iceball was observed around the intra-atrial probe during freezing. Change associated with frozen atrial myocardium was not obvious by echocardiographic imaging during or after ablation. Various modes of ultrasonic imaging were employed to help visualize the probe within the left atrium and guide catheter manipulation. Initially two-dimensional (2D) transesophageal echocardiographic method was used for guidance, whereas four-dimensional (4D) imaging reconstructed from off line serial 2D images was used to create a three-dimensional (3D) image set for catheter visualization with respect to the pulmonary veins and mitral valve. However, 4D imaging proved difficult and therefore live 3D imaging with an epicardial probe was employed with good result.
Pulmonary Vein Isolation by Electrophysiological Assessment
A good marker for the adequate positioning of the probe dorsal to the Bachmann bundle was the recording of a significant right to left atrial delay (Fig 3A) when applying the second or third cryoapplication involving the right upper quadrant of the left atrium (Fig 1). This delay suggested that the nonexcluded segment remained in direct connection with the Bachmann bundle while the activation now took a "longer or slower" conducting route to reach the posterior wall (Table 1).
|
|
Atrial Arrhythmias Before Isolation of Pulmonary Vein Region
Sustained AF with a pattern consistent with left atrial origin was inducible using burst pacing of the LPA and often several seconds after atrial flutter induced with burst pacing of the RAA (Table 2). Two animals with sustained atrial fibrillation had to be cardioverted because of low blood pressure. The shortest cycle length varied from 87 to 135 ms when measurable. Reproducible fractionated electrograms were recorded from the LPA, and rarely from the RAA electrodes. RAA burst pacing induced non-sustained atrial flutter in 3 animals with a cycle length that varied from 92 ms to 137 ms.
|
In the nonexcluded segment, short runs of atrial flutter (
3 seconds) were induced by RAA pacing in 5 animals (Fig 3B), but no atrial arrhythmia was inducible in 3. The animal with intermittent isolation exhibited a response to stimulation similar to that observed in pigs with the fully isolated LPA.
Pathological Findings
Macroscopic examination
Shortly after excision and opening of the left atrium, most of the lesions were not readily visible, although a subtle reddish discoloration could be detected in some locations around the pulmonary vein region. There was neither erosion of the endocardium nor thrombi. After fixation, the lesions were readily visible, with a much darker band of endocardium circumscribing the pulmonary vein region (Fig 2). The left atrial wall thickness varied significantly from 7 mm over the superior wall in the prolongation of the Bachmann bundle, to 1.5 mm at the inferior wall posterior to the coronary sinus. The line of cryoablation was positioned posterior to the coronary sinus as planned (Fig 2).
Histological examination documented typical acute cryolesions (Fig 4). Transmural lesions were obtained with myocytes necrosis, vascular congestion and interstitial haemorrhage. Lesions were completely transmural in the thinner segments, while the epicardial fat was spared in thicker segments with some hemorrhage at the surface. In the endocardium, hemorrhage and edema were a constant feature.
|
| Comment |
|---|
|
|
|---|
We have demonstrated that an open-heart surgical procedure could be duplicated in the off-pump, closed, beating heart without complications or limitations associated with the challenges of this novel approach. This study is part of a multidisciplinary project which includes bioengineering imaging [2426], haptics and other artificial intelligence tools and manipulators to offer beating, intracardiac, off-pump surgery as a preferred alternative to open-heart surgery for most of its current targets. At this stage of the evolution of this technique, the most important priority is the development of a reliable 4D image guidance environment, to compensate for the absence of direct vision, along with the integration of robotic and haptic tools.
Image Guidance
Echocardiographic imaging provided good guidance to ascertain the location of the probe in relation with the pulmonary vein orifices, the Bachmann bundle, the mitral valve orifice and the coronary sinus. However, the quality of the contact of the probe was left to the surgeon's touch. In support of this procedure and other intracardiac interventions, we have developed an extensive virtual cardiac environment to permit preoperative dynamic compute tomographic (CT) or magnetic resonance (MR) patient images to be registered to and synchronized with the patient. Preliminary work towards this goal has been reported [2428]. We are currently addressing the issues of image-to-patient registration, as well as the mapping of electrophysiological data, acquired via a tracked (CARTO System; BioSense Webster, Diamond Bar, CA) probe onto the model. In addition we plan to integrate tracked intracardiac ultrasound with the virtual model in order to give the intraoperatively acquired images their appropriate context with respect to the model. We believe that the realization of this environment, when integrated with intra-cardiac procedures such as that described in this paper, will result in enhanced planning, more precise targeting, and increased efficacy of the intracardiac, off-pump procedures.
Cryoablation
Transmural lesions were reliably obtained using modified SurgiFrost probes. Failure to achieve complete isolation was more likely associated with misguided application [21], and the difficulty of precisely deploying the full length of the 6-cm probe to assure contiguous contact with the atrial wall, particularly in the convoluted regions of the septum to pulmonary vein junction.
Criteria for Isolation
Electrical isolation can only be documented using electrophysiological testing to confirm a bidirectional bock. Because cryoablation is associated with three zones: a central zone of irreversible myocardial damage, a zone of reversible cellular change and a distal zone of intact tissue, observed electrical isolation may be only temporary. Based on our clinical experience of reversible cryogenic damage [21], we allowed 15 minutes before reassessment. Macroscopic and/or microscopic examination can only suggest or document incomplete isolation when borderline transmurality or a gap of viable myocardium is observed. Because the entire circumference of the ablation zone cannot realistically be examined using serial histological slides, the histological examination is used only in support of electrophysiology and to document unwanted effects on tissue or vasculature.
Conclusion
These acute studies produced electrophysiological features very similar to those observed in surgical patients with chronic or persistent lone AF [12]. Before isolation, all animals exhibited evidence consistent with sustained left sided atrial fibrillation and right-sided atrial flutter, a condition similar to patients with lone AF. Acute effects of isolation of the pulmonary vein region was associated with inability to induce sustained AF in the excluded segment, while only short lived atrial flutter could be induced in the non-excluded segment. These findings are very similar to those of patients with persistent or paroxysmal AF and suggests that the mechanism of acutely induce AF in "normal" porcine heart and that of patients may be similar and that the pig is a valid experimental model for AF.
Further developments need to address the study of the pathophysiology of AF in a chronic isolation model, while at the same time developing sophisticated technology to transfer this approach to a "user-friendly" clinical setting. Additional developments will be needed in ultrasound guidance, as well as improvements in the design of the lesioning probe to optimize the efficacy of this procedure, while developing the 4D virtual reality platform [2428].
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. G. Katritsis and A. J. Camm Catheter ablation of atrial fibrillation: do we know what we are doing? Europace, November 1, 2007; 9(11): 1002 - 1005. [Full Text] [PDF] |
||||
![]() |
G. M. Guiraudon, D. L. Jones, D. Bainbridge, and T. M. Peters Mitral valve implantation using off-pump closed beating intracardiac surgery: a feasibility study Interactive CardioVascular and Thoracic Surgery, October 1, 2007; 6(5): 603 - 607. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |