| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2003;126:1357-1365
© 2003 The American Association for Thoracic Surgery
Surgery for acquired cardiovascular disease |
a Department of Cardiac Surgery, Austin and Repatriation Medical Centre, Heidelberg, Victoria, Australia
b Baker Institute of Medical Research, Prahran, Victoria, Australia
Read at the Eighty-second Annual Meeting of The American Association for Thoracic Surgery, Washington, DC, May 5-8, 2002.
Received for publication April 22, 2002; revisions received July 8, 2002; revisions received September 24, 2002; accepted for publication November 18, 2002.
* Address for reprints: Jai Raman, MBBS, MMed, FRACS, PhD, Section of Cardiac and Thoracic Surgery, University of Chicago Medical Center, 5841 S Maryland Ave, MC 5040, Chicago, IL 60637 United States
jraman{at}surgery.bsd.uchicago.edu
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionDiscussionReferences |
|---|
METHODS: One hundred thirty-two patients in 20 centers underwent radiofrequency ablation as an adjunct to conventional cardiac surgery, with a standardized lesion set created with a flexible, 7-electrode, temperature-controlled probe (Cobra; EPTechnologies, San Jose, Calif). All data were entered into a central registry, with regular follow-up prompted by the registry cocoordinator. Each radiofrequency scar was made with standard parameters requiring 2 minutes of tissue coagulation at 80°C to 85°C. Patients undergoing mitral procedures had radiofrequency ablation performed in the left atrium endocardially. Patients undergoing aortic valve replacement or coronary artery bypass surgery underwent epicardial radiofrequency ablation of the left atrium. Epicardial radiofrequency ablation lesions on the right atrium were common to both groups of patients. Preoperatively, 75% of the patients had chronic atrial fibrillation, 21% had paroxysmal atrial fibrillation, and 4% had flutter. Surgical procedures performed included mitral valve procedure in 60%, coronary artery bypass grafting in 14%, aortic valve replacement in 7%, and coronary artery bypass grafting plus aortic valve replacement in 4%.
RESULTS: There were no major complications related to the use of radiofrequency ablation. There were no soft tissue or cardiac perforations. Ten patients were defibrillated into sinus rhythm within 3 months postoperatively. The freedom from atrial fibrillation was 84% at 3 months, 90% at 6 months, and 100% at 12 months. All patients at 12 and 18 months' follow-up were in sinus rhythm. There were no thromboembolic complications.
CONCLUSIONS: Surgical radiofrequency ablation can be performed safely as an adjunct to conventional cardiac surgery. A standardized lesion set created by using similar temperature settings can be adopted in multiple centers and might be effective in treating atrial fibrillation. Data collection through a central registry has helped in monitoring the effectiveness of this new technique in a scattered population.
We chose to replace the cut-and-sew approach with radiofrequency ablation (RFA) to create transmural lines of electrical block on the basis of animal work performed at our laboratory.14 Our early clinical experience15 helped pave the way for the adoption of RFA as an adjunct to conventional cardiac surgery in many centers across Australia and New Zealand. This is a preliminary report on this multicenter experience.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionDiscussionReferences |
|---|
Figure 1 shows the schematic diagram of the lesion set chosen by our group (4a being the endocardial lesions used in the mitral procedures, 1b being the epicardial lesions on the left atrium used in aortic valve/coronary artery bypass grafting [CABG] procedures, and 1c being epicardial right atrial lesions used in both). The numbered lines are a suggested sequence of lesions meant to help with training surgeons unfamiliar with RFA of the atria, the maze procedure, or both. The steps of the procedure are described in the diagram. The left atrial appendage was oversewn in a linear fashion from within, and this plicating suture helped gather up redundant atrial tissue in this region going up to the atrial roof. Pulling up on the uncut ends of the suture also helped display the left-sided pulmonary vein orifices.
|
In the first half of our experience, all patients had AF documented on a preoperative 24-hour Holter electrocardiographic monitor, before being offered RFA. Initially, the procedure was performed only in association with mitral valve surgery. The RFA procedure was also used in such a way that there was no significant alteration of the surgeon's current approach and technique.
Safety precautions were prescribed and consisted of the following:
Epicardial applications
As confidence in the technique grew and evidence of the transmural nature of the epicardial lesions was established, surgical RFA was extended to patients undergoing aortic valve procedures and coronary artery surgery. Epicardial lesions were created in these patients. All procedures were performed with aid of cardiopulmonary bypass. In the epicardial setting, all lesions were created during cardiopulmonary bypass. In the early part of the experience, the posterior left atrial lesions were created on an arrested heart to facilitate access. Later in the experience, the trend was to perform these lesions on a decompressed heart during CPB, either with the aid of a left ventricular vent or bicaval cannulation. The fat pad around the pulmonary veins was dissected until the underlying muscle was seen. Unfortunately, this meant that the epicardial lesions in the posterior left atrium did not reach the mitral valve annulus. The lesion to the tricuspid valve annulus was performed low down on the anterior aspect of the right atrioventricular groove. If required, the right coronary artery was dissected free and pushed anteriorly to allow the probe to sit at the atrioventricular groove.
There were no stand-alone procedures performed for AF alone.
When each new surgeon and center was trained in the use of the probe, care was taken to ensure uniformity of the lesion sets and parameters used to create them.
Details of all patients who had the procedure performed with the Cobra probe were entered into a centrally maintained registry. The purchase price of each probe also included a fee to help with the set-up and running costs of the registry. A protocol of clinical, echocardiographic, and electrocardiographic follow-up was suggested to each of the participating surgeons.
Study population
All patients underwent a definite cardiac surgical procedure, which was the primary indication for surgical intervention. Patients were referred for surgical intervention by their cardiologists, and the decision to proceed was taken in consultation with the cardiac surgeon. Although the addition of RFA to the procedure did increase operative time, it was perceived by most participating surgeons as relatively innocuous. If there was a history of AF, random 24-hour Holter monitoring was performed, and the patient was offered RFA. All surgical RFA procedures were performed as an adjunct to conventional cardiac surgical procedures with the aid of cardiopulmonary bypass. All patients who underwent RFA were included in this study, irrespective of the risk and urgency of the procedure.
One hundred thirty-two patients underwent surgical RFA procedures between March 2000 and March 2002 at 20 hospitals across Australia and New Zealand. Table 1 outlines the patient demographics of this study. The mean age of these patients was 66 years (range, 21-86 years). Preoperatively, 75% of the patients had chronic AF, 21% had paroxysmal AF, and 4% had a combination of flutter and fibrillation. The mean duration of preoperative rhythm disturbance was 36 months, ranging from 1 to 120 months. Although the recommendation was to apply this procedure in patients with long-standing AF (ie, duration >6 months), the multicenter nature of this experience placed the onus of patient enrollment on the surgeon's discretion and judgment. Almost all of the patients had clinically documented AF for more than 6 months. This report includes all the patients recruited, regardless of indication, and is a true warts and all account.
|
The primary surgical procedures were mitral valve procedures in 62 patients. Mitral valve procedures and CABG were performed in 10 patients. Combined aortic and mitral procedures were performed in 7 patients. CABG was performed in 19 patients. Aortic valve procedures were performed in 9 patients. In total, endocardial RFA on the left atrium was performed in 92 patients, and epicardial RFA on the left atrium was applied in 40 patients.
Pharmacologic protocol
Postoperatively, patients were not started on any specific regimen of anticoagulation. However, as the multicenter experience began, patients were encouraged to take low-dose amiodarone (200 mg/d) for 6 months unless specifically contraindicated.
Statistical analysis
Statistical analysis was conducted with SAS version 5.0 software (SAS Institute, Inc, Cary, NC). Continuous data were expressed as means ± SD. The Student t test and 
2 test were used for statistical analysis.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionDiscussionReferences |
|---|
There were 9 perioperative deaths. None were related to the use of RFA. One was due to remote aortic dissection in a patient who had replacement of the ascending aorta. Another was due to severe liver dysfunction caused by coagulopathy and tamponade in an elderly woman with a very fragile and calcified aortic root that leaked after valve replacement. The third was a decompensated patient on a balloon pump with uncontrolled AF, who had a very poor ventricle caused by large infarcts, severe aortic regurgitation, and uncontrolled tachycardia. He arrested soon after anesthetic induction and was salvaged. RFA was used in this patient to achieve good rate control and in the hope of avoiding amiodarone. Despite a relatively stable postoperative course, there was no recovery of ventricular function on day 5, and prolonged mechanical support was not an option. In retrospect, this patient should have had another means of rate control. However, the radiofrequency lesions were created with the heart beating on CPB, with no addition to the crossclamp time. A fourth patient died on the eighth postoperative day at home from a hemorrhaging duodenal ulcer. There was one death caused by cardiogenic shock as a result of a large infarct complicating a delayed low-output syndrome on day 6 after the operation. In this instance the infarct was septal and related to a very diseased left anterior descending artery. It is unlikely that radiofrequency energy was applied close to the left anterior descending artery. There was one death caused by perioperative gut ischemia in a patient who underwent successful mitral valve repair and CABG. One patient with a very poor ventricle who had mitral valve repair for severe regurgitation died of a postoperative low systemic vascular resistance syndrome resistant to norepinephrine and vasopressin. Details on 2 other deaths were sketchy but related to postoperative low-output states in elderly patients.
We attempted to study a control group of patients with preexisting AF who had cardiac surgery but did not have any antiarrhythmic procedures. However, this proved extremely difficult in a multicenter setting and was abandoned.
Although the recommendation was that patients continue to receive amiodarone for 6 months, there was variable compliance by treating surgeons and patients
Follow-up
From a total of 132 patients undergoing the procedure, follow-up was attempted in the 123 survivors. Twelve patients had undergone operations within 3 months of this publication and hence were not available for out-of-hospital follow-up. Mean follow-up was 6.4 months (range, 3-24 months), with complete clinical review in 110 patients. Eighty-seven patients underwent the follow-up evaluation for 3 months, 50 for more than 6 months, and 15 for more than 12 months. Follow-up data at 6 months were not yet available in another 12 patients. Figure 2
is a representation of rhythm preoperatively and postoperatively in the patients enrolled in the study. The sinus recovery rate was 84% at 3 months (72 of 87), 90% at 6 months (45 of 50), and 100% at 12 months. Ten patients required cardioversion postoperatively. Five of them had cardioversion within 2 weeks, and 5 had cardioversion within 3 months. Recovery of sinus rhythm occurred in 9 (90%) of these patients. In the follow-up period, return to sinus rhythm was documented by means of electrocardiography and Holter monitoring, which was usually performed between 3 and 6 months postoperatively. No stress tests were performed to examine sinus node function.
|
|
|
|
There were 6 readmissions with recurrent AF, sinus bradycardia, or both in the first 3 months postoperatively. There was one late death in a patient who had infective endocarditis complicated by liver failure 8 months postoperatively.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionDiscussionReferences |
|---|
In the race to create scars of electrical block, a range of different energy sources is being tested. RFA, which uses a form of alternating current, causing tissue coagulation by means of resistive heating, has been adapted from the electrophysiologist's repertoire.22 Cardiac electrophysiologists work from the endocardial aspect, causing focal ablation of ectopic foci of atrial tissue, especially in the pulmonary vein orifices.23 These techniques are limited by swirling blood that reduces energy delivery and poor geographic localization, relying on intracardiac echocardiography. To reduce the risk of thromboembolism and to allow for maximal focal delivery of radiofrequency energy, most electrophysiologists rely on temperature settings at a maximum of 70°C for less than 60 seconds. Their goal also is to cause focal ablation endocardially. This approach cannot be used for the treatment of AF.24
Surgical goals in the treatment of AF are different. If the principles of the maze procedure are to be emulated or replicated without cutting and sewing, transmural scars that act as lines of electrical block are required. Surgeons performing a combined surgical procedure have little time to measure electrograms or study maps intraoperatively to ensure transmurality.25 Patwardhan and associates26 used radiofrequency microbipolar coagulation in an early attempt to perform a maze modification. While pioneering RFA for AF, Melo and coworkers27 used parameters set by the cardiac electrophysiologists, causing lesions that might not have been quite transmural. That might explain the mediocre results obtained in his early reports. Numerous authors have described the creation of transmural lesions with surgical probes used for RFA.28,29 The Cobra probe specifically has been shown to create transmural lesions when used appropriately in animals.30
Our experience with the Cobra probe and its predecessor was in the animal laboratory. Quite by accident, we discovered that transmural lesions could be achieved consistently only by using a temperature setting of 80°C to 85°C for 120 seconds. We did not publish this information separately, but part of it is lost in the detail of some experimental work on a model of atrial flutter in sheep.12 This feedback, however, does not seem to have been relayed to other users of this technology. When the Australasian multicenter study got underway, rigorous attempts were made to ensure that the probe was used accurately, appropriately, and safely, with fixed parameters. Because the goal is creation of transmural scars created by coagulation of atrial tissue, it might be time to replace the term "RFA" with "radiofrequency coagulation" when used surgically.
In this context, while transmural scars are being created, the potential for collateral damage, such as esophageal injury, exists, as reported by Gillinov and colleagues31 and Mohr and associates.32 The lack of major tissue perforations in this multicenter experience, which included the learning curve, might be partly due to the rigorous adherence of the various surgeons to the protocol. The reinforcement by representatives of the probe manufacturers also helped maintain safety standards.
In a very timely editorial, Cox33 reminded us that there were too many variables, ranging from lesion patterns to the kind of energy types and settings, in the adaptation of a variety of energy sources, including RFA, in the race to replace the maze procedure. It has been difficult to delineate the benefits of one technique over another. In addition, the majority of the reports seem to emanate from one center that specializes in the application of that particular technique. Table 5 compares patient profiles, mortalities, and numbers of patients undergoing concomitant procedures in the recent literature, citing large, single-center maze experiences and one single-center radiofrequency experience for comparison.
|
The longer-term follow-up from the Australasian registry is important in this respect because it will focus on the results of a specific pattern of lesions that have been created by fairly uniform settings of radiofrequency energy in multiple centers. The early results with this technique across Australia and New Zealand have been safe and satisfactory. Similar to the experience in Berlin,35 our follow-up showed a steady improvement over time in recovery of sinus rhythm. This also suggests that early susceptibility to AF after cardiac surgery is not significantly changed by RFA. Interestingly, the kind of left atrial lesions did not significantly alter the recovery of sinus rhythm. Patients with preoperative paroxysmal AF showed trends toward better recovery of sinus rhythm, and this might again be a reflection of smaller atria with focal triggers.
"Atrial transport function" is a term used freely in most publications about the maze procedure and its modifications. There is a great emphasis on transmitral Doppler velocities across the mitral valve.36,37 However, as most echocardiocardiologists point out, these velocities and ratios are imprecise markers of atrial function and might be quite fallacious in the setting of prosthetic mitral valves or mitral valve repairs.38 The left atrial appendage is overseen or obliterated, and the emptying velocity of the appendage, which is often used to measure atrial transport, cannot be used. Some echocardiologists have suggested using atrial diastolic and systolic volumes. However, echocardiographic assessment of these volumes can be variable and very operator dependant. These measurements would be more accurate when assessed by means of magnetic resonance imaging. These are areas that need more discussion and investigation. The difference in E/A ratios in the endocardial versus epicardial group might reflect the difference in atrial size in the 2 groups. The other important feature in the follow-up of the Australasian registry patients was the lack of any significant thromboembolic complications, which could be a clinical surrogate for atrial transport function.
Finally, maintaining a registry of this size requires a significant financial commitment and a good registry co-coordinator in addition to regular surgical audits and the cooperation of surgeons. It is our sincere hope that this registry can be continued in the present form at least for the next few years to establish the long-term results of surgical RFA/radiofrequency coagulation in the treatment of AF.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionDiscussionReferences |
|---|
As Dr Raman pointed out, the 2 principles to which all the participants adhered in this study were that the pattern of lesions would correspond as closely as possible to those of either the maze or radial procedures, both of which are proved procedures, and that every effort would be made to be certain that every portion of every lesion was transmural.
By adhering to these 2 fundamental principles, the results could have been predicted, I think, because every previous study that has taken this approach has resulted in essentially the same results. This strongly implies, though, that as long as the lesion pattern is correct and the lesions are transmural, the particular method for creating those lesions is immaterial to the outcome. In other words, any energy source or technique that is capable of creating the correct lesion pattern and is uniformly transmural will result in a cure of AF in essentially all patients.
I have 2 questions for Dr Raman. Why do you think the results improved with time, because that has not been the experience in most other studies? Second, you note that there was no AF on long-term follow-up, but did those patients ever experience episodes of atrial flutter?
I congratulate Dr Raman and his associates on a well-designed and important study.
Dr Raman. Thank you very much, Dr Cox. It is a privilege to actually have you discuss this paper because you served as an inspiration for us to actually set up this protocol. I will answer the second question first. In terms of flutter, we had one patient who had one episode of flutter late, and it resolved spontaneously. This was asymptomatic flutter, and fortunately, the patient was actually reviewed at that time by his cardiologist and local doctor. On subsequent Holter monitoring, he was found to be in sinus rhythm, and therefore he did not actually need further medications or specific intervention. A few years ago this man ran marathons. He was in his 70s, and he went back to running up hills. He said he used to experience tachycardia or felt palpitations coming on when he ran up a hill, but he did not have this experience postoperatively, and I think this was a chance finding that he actually had flutter, which was relatively slow.
The first question concerns the improvement in AF over time. I think almost definitely that this procedure does nothing to reduce the propensity toward early postoperative AF. We all know that postoperative AF in patients undergoing any cardiac surgical procedure is about 30%, and despite all the different manipulations, such as magnesium infusions and ß-blockade, this still seems a bugbear. We did not see a difference. Early in the experience, we were very enthusiastic, and some of the surgeons applied these procedures in all kinds of patients. I think they have actually been a lot more selective later on. As our experience grew, we have also been more aggressive in using cardioversion early on, which might account for the improvement in the recovery of sinus rhythm. The other interesting thing, and this is an observation that we have seen, is that patients who have been followed up for 3 months who seem to be in fibrillation and then at the 6-month mark they have reverted to sinus rhythm. I know that this is anecdotal, but I have at least 3 patients with that, and other people have had similar experiences. I cannot explain it, but I think they seem to stabilize electrically with the passage of time.
Dr Joao Melo (Lisbon, Portugal). Congratulations on a beautiful paper. I am happy to see those results, which are much better than most reported results, including ours. But to understand whether we are comparing the same type of patients, I noticed that 25% of your patients were not in chronic AF. Therefore, it is important to know how many of your patients with mitral valve disease were part of this group of patients. My first question is to know whether this a selected population. My second question regards the issue of pacemakers because it is also very striking that you did your lines on the Cox maze concept and you are reporting no need for postoperative pacemakers. Can you elaborate on the reason for that finding?
Dr Raman. As I said, this is a reporting of what we have seen with this probe, and I cannot speak for specific indications in other centers, but all patients that had documented AF and came up initially for mitral valve surgery and later on for coronary artery surgery or early valve replacement surgery were enrolled, and as they got more familiar with the technique, they started getting more confident with it. I cannot tell you with the AF how many of them had mitral valve disease, but we will look at that carefully.
Dr Ralph Damiano (St Louis, Mo). I would like to echo both Dr Cox's and Dr Melo's congratulations on a beautiful study. One of the objectives of the maze procedure, or any type of surgery for AF, is to allow patients to discontinue antiarrhythmic drugs, which have a high incidence of side effects. I noticed you did not mention that, but I was wondering what percentage of your patients were both in sinus rhythm and free of antiarrhythmic drugs and what percentage were also free of chronic anticoagulation?
Dr Raman. I knew someone would ask that. As I said, this, again, is a work in evolution. When we first started off, we did not use any antiarrhythmics as a routine, but there were suggestions from others, particularly electrophysiologists, that using low-dose amiodarone perioperatively might help. What we have recommended, and not all the surgeons adhere to this because amiodarone does provoke strong feelings among surgeons, is that we put patients on 200 mg/d amiodarone postoperatively for the first 3 months, but that has not been strictly adhered to. In terms of long-term data, none of the 15 patients who are out close to 24 months are on any antiarrhythmic medications. In terms of anticoagulation, 30% at 3 months were on anticoagulation, 20%of them were for a mechanical valve, because, as in the United States, we see some amount of rheumatic disease. It is not as much as other places, but there is a definite proportion of patients who have mechanical valves. Out at 6 months it decreases to 20%, but at the 12-month mark with our first group of patients, 3 of them had mechanical valves. Therefore, 25% of that first group of patients that have gone beyond 12 months are on anticoagulation.
Regarding the other question that Dr Melo asked about pacemakers, we did have 4 pacemakers, so it is not zero, and I think as we follow these patients up further we might see, as Dr Damiano presented in his series yesterday, an increasing incidence of pacemakers for sinus bradycardia.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. P. Beukema, H. T. Sie, A. R. Ramdat Misier, P. P. H.M. Delnoy, H. J.J. Wellens, and A. Elvan Intermediate to long-term results of radiofrequency modified maze procedure as an adjunct to open-heart surgery. Ann. Thorac. Surg., November 1, 2008; 86(5): 1409 - 1414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. P. Beukema, H. T. Sie, A. R. R. Misier, P. P. Delnoy, H. J.J. Wellens, and A. Elvan Predictive factors of sustained sinus rhythm and recurrent atrial fibrillation after a radiofrequency modified Maze procedure Eur. J. Cardiothorac. Surg., October 1, 2008; 34(4): 771 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-Q. Lai, J.-H. Li, J.-W. Li, S.-D. Xu, Y. Luo, and Z.-G. Zhang Concomitant irrigated monopolar radiofrequency ablation of atrial fibrillation in adults with congenital heart disease Interactive CardioVascular and Thoracic Surgery, February 1, 2008; 7(1): 80 - 83. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Deneke, K. Khargi, B. Lemke, T. Lawo, M. Lindstaedt, A. Germing, T. Brodherr, L. Bosche, A. Mugge, A. Laczkovics, et al. Intra-operative cooled-tip radiofrequency linear atrial ablation to treat permanent atrial fibrillation Eur. Heart J., December 1, 2007; 28(23): 2909 - 2914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Doty, D. B. Doty, K. W. Jones, J. H. Flores, M. Mensah, B. B. Reid, S. E. Clayson, G. Snow, E. Righter, and R. C. Millar Comparison of standard Maze III and radiofrequency Maze operations for treatment of atrial fibrillation J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1037 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Jahangiri, G. Weir, K. Mandal, I. Savelieva, and J. Camm Current strategies in the management of atrial fibrillation. Ann. Thorac. Surg., July 1, 2006; 82(1): 357 - 364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Barnett and N. Ad Surgical ablation as treatment for the elimination of atrial fibrillation: A meta-analysis J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 1029 - 1035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Sanchez-Quintana, J. A. Cabrera, V. Climent, J. Farre, M. C. de Mendonca, and S. Y. Ho Anatomic Relations Between the Esophagus and Left Atrium and Relevance for Ablation of Atrial Fibrillation Circulation, September 6, 2005; 112(10): 1400 - 1405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ninet, X. Roques, R. Seitelberger, C. Deville, J. L. Pomar, J. Robin, O. Jegaden, F. Wellens, E. Wolner, C. Vedrinne, et al. Surgical ablation of atrial fibrillation with off-pump, epicardial, high-intensity focused ultrasound: Results of a multicenter trial J. Thorac. Cardiovasc. Surg., September 1, 2005; 130(3): 803 - 803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zangrillo, G. Crescenzi, G. Landoni, S. Benussi, M. Crivellari, F. Pappalardo, E. Dorigo, C. Pappone, and O. Alfieri The Effect of Concomitant Radiofrequency Ablation and Surgical Technique (Repair Versus Replacement) on Release of Cardiac Biomarkers During Mitral Valve Surgery Anesth. Analg., July 1, 2005; 101(1): 24 - 29. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
