Surgical Treatment of Atrial Fibrillation Using Argon-Based Cryoablation During Concomitant Cardiac Procedures
Background— The development of ablative energy sources has simplified the surgical treatment of atrial fibrillation (AF) during concomitant cardiac procedures. We report our results using argon-based endocardial cryoablation for the treatment of AF in patients undergoing concomitant cardiac procedures.
Methods and Results— Sixty-three patients with AF who were undergoing concomitant cardiac procedures had the same left atrial endocardial lesion set using a flexible argon-based cryoablative device. Mean age was 65.1±1.3 years. Sixty-two percent had permanent AF, whereas 38% had paroxysmal AF. Mean duration of AF was 30.5±4.8 months. Mean left atrial diameter was 5.5±0.1 cm. Mean ejection fraction was 45±1.4%. All endocardial lesions were performed for 1 minute once tissue temperature reached −40°C. Follow-up echocardiograms were obtained to determine freedom from AF. Kaplan-Meier analysis demonstrated an 88.5% freedom from AF rate at 12 months. Ablation time was 16.8±0.6 minutes. There were no in-hospital deaths and no strokes. Twelve patients (19%) required postoperative permanent pacemaker placement.
Conclusions— Cryoablation using this flexible argon-based device for the treatment of AF during concomitant cardiac procedures was safe and effective, with 88.5% of patients free from AF at 12 months.
Atrial fibrillation (AF) is the most common of all sustained arrhythmias and is present in approximately 2.2 million adults in the United States.1 The incidence of AF increases with age, and it is estimated that at least 10% of the patients over 80 years of age suffer from this disorder.2 AF is responsible for a substantial morbidity caused by thromboembolic events, hemodynamic compromise, rate-related cardiomyopathy, and anticoagulant-related bleeding. Additionally, AF is associated with up to a 1.9-fold increased risk of mortality in individuals affected.3
The Cox-Maze III procedure, which creates “cut and sew” lesions, is the gold standard for the surgical treatment of AF. This procedure is based on the accepted principle that AF is caused by multiple reentrant circuits within the atria. The Cox-Maze III procedure effectively blocks the propagation of these reentrant circuits into dead ends and allows the normal propagation of sinoatrial node stimuli.4,5 The efficacy of this procedure demonstrates a greater than 95% freedom from AF rate at a mean follow-up of over 5 years.6 The Cox-Maze III procedure has been shown to greatly diminish the risk of stroke.7 Despite its success, the Cox-Maze III technique has failed to gain wide spread use because of its complexity, prolonged cardiopulmonary bypass and ischemic times, and overall morbidity.
New insights into the pathogenesis of AF, specifically the importance of the pulmonary veins in the initiation of paroxysmal AF, hastened the development of simpler approaches to the surgical treatment of AF.8 Several investigators using a variety of energy sources demonstrated that by focusing on posterior left atrium and pulmonary veins, acceptable success rates can be achieved.9–12 Two major strategies have evolved: (1) Reduction/elimination of the amount of “cut and sew,” and (2) development of a safe, effective, and simple technique to treat AF that is appropriate for all patients with a history of AF who are undergoing cardiac surgery. In this study, we report our results using argon-based cryoablation on patients with a history of AF who are undergoing concomitant cardiac procedures at New York Presbyterian Hospital-Weill Cornell Medical Center.
All data were collected prospectively from patients with AF who were undergoing concomitant cardiac surgical and cryoablation procedures. All procedures were performed at The New York Presbyterian Hospital-Weill Cornell Medical Center between December 2002 and October 2004. Patients were included if they were undergoing open cardiac surgery and had either paroxysmal AF for 6 months more greater or permanent AF for 6 months or more. Patients in paroxysmal AF were defined as having spontaneous, self-limiting episodes of AF for a period of 6 months or more. Permanent AF was defined constant AF despite previous attempts at electrical or pharmacological cardioversion.
In this study, age, sex, duration of atrial fibrillation, and types of concomitant procedures were identified. Immediate postoperative rhythms as well as rhythms at discharge were studied. Follow-up electrocardiograms (ECGs) at 1, 3, 6, and 12 months were used to determine freedom from AF. Kaplan-Meier analysis was used to determine freedom from AF beyond the first 3 postoperative months. Only those patients in normal sinus rhythm (NSR) at 3 months were considered in the analysis, as they were at risk to develop AF/atrial flutter (AFl).
Before the procedure, the left atrial diameter and ejection fraction were determined from the intraoperative echocardiogram. The ablation procedures were performed with a Food and Drug Administration-approved flexible cryoprobe (SurgiFrost, CryoCath/Endocare Inc) using an argon-based cooling system. This technology consists of a malleable 60-mm probe that uses argon as its cooling source and is capable of reaching temperatures of −160°C. This temperature is measured at a single point within the inner bellows of the probe. All lesions were completed after cardiopulmonary bypass and cardioplegic arrest. The left atrium was opened using either a superior septal incision or left atrial incision. All patients had the same left atrial lesion set. The ablation was performed for 1 minute after the probe temperature reached −40°C (with minimum temperature reaching between −120°C and −160°C). If a superior septal incision was used, the lesion set consisted of circumferential linear cryolesions around the endocardial surface of the right pulmonary veins. If a left atrial incision was used, then a semicircular cryolesion that effectively connects to both ends of the atriotomy was performed to complete a ring around the right superior and inferior pulmonary veins. Two semicircular cryolesions were then made around the endocardial surface of the left pulmonary veins, thereby creating an ablation ring around both the left superior and inferior pulmonary veins. A connecting lesion from the right inferior pulmonary vein to the left inferior pulmonary vein was then made. The base of the left atrial appendage was ablated, and a connecting lesion from the appendage to the superior left pulmonary vein was created. A final lesion was made from the left inferior pulmonary vein to the P3 region of the mitral annulus. The opening to the left atrial appendage was then over-sewn (Figure 1).
A lesion from the inferior vena cava to the tricuspid valve annulus (isthmus lesion) was created in patients who were undergoing a right atriotomy for concomitant procedures such as tricuspid valve repair, atrial septal defect (ASD) repair, or a superior septal approach to the mitral valve, or who had a history of AFl. It was our intent to incorporate the isthmus lesion in these select patients with right atrial pathology to prevent postoperative AFl or atrial tachycardia.
All patients were received anticoagulation therapy with coumadin for a period of at least 3 months, regardless of rhythm. Target international normalized ratio range was 2.5 to 3.0 in the absence of a mechanical valve. Amiodarone was used selectively in patients who developed postoperative AF and was usually discontinued after 1 month at the discretion of the cardiologist.
After hospital discharge, patients were evaluated at 1, 3, 6, and 12 months after surgery by ECGs, records of referring cardiologists, telephone interviews, and direct office contact. Patients free from AF for 3 months underwent 24-hour Holter monitoring to identify asymptomatic paroxysmal AF that may not have been present on the ECG. If the 24-hour Holter monitoring did not demonstrate paroxysmal AF, an echocardiogram was recommended to evaluate atrial transport. If there was adequate evidence of atrial transport, then coumadin was discontinued at the discretion of the cardiologist. If a mechanical valve was present, the patient remained on life-long coumadin therapy.
Patients with AF or AFl 3 months after surgery were referred to an electrophysiologist. Subsequent electric cardioversion or percutaneous radiofrequency (RF) catheter ablation was performed at the electophysiologists’ and primary cardiologists’ discretion.
Data were collected and evaluated with Microsoft Access 2000, and analyzed with SPSS Version 12.02 and Microsoft Excel spreadsheet 2002. Continuous variables are presented as the mean±standard error of the mean (SEM) and compared using a 2-tailed unpaired t test. Clinical profiles between groups were compared using the χ2 test. A probability value less than 0.05 was considered significant.
Between December 2002 and October 2004, 63 patients with AF underwent left atrial cryoablation procedures during concomitant open-heart operations. The mean age of these patients was 65.1±1.3 years. There were nearly an equal number of men and women (54% men). The mean duration of AF was 30.5±4.8 months. Thirty-nine patients (62%) had permanent AF and 24 (38%) had paroxysmal AF. Mean left atrial size was 5.5±0.1 cm. Ejection fraction was 45.3±1.4% (Table 1). One patient had 2 attempts at percutaneous RF pulmonary vein isolation before undergoing cardiac surgery. Two patients had preexisting pacemakers.
Although all patients underwent left atrial cryoablation, 18 patients (25%) had an additional right atrial isthmus cryoablation. Thirteen patients had lone mitral valve repair and 15 patients had mitral valve repair with other concomitant procedures. There were 19 lone mitral valve replacements and 8 mitral valve replacements included other concomitant procedures. In addition, there were 3 ascending aortic arch reconstructions requiring aortic valve replacement conduits under hypothermic circulatory arrest, 2 coronary artery bypass grafts, 2 atrial septal defect repairs, and 2 tricuspid repairs (Table 2). Six patients had previous cardiac surgery. Cardiopulmonary bypass and cross-clamp times were 119.4±3.6 and 86.3±3.3 minutes, respectively. Average ablation time was 16.8±0.6 minutes. Postoperative hospital stay was 10.5±0.8 days. There were no in-hospital postoperative deaths. To date, there have been no neurological events such as thromboembolic strokes or transient ischemic attacks. Mean follow-up time was 11.4 months.
Three patients died during the follow-up period between 3 months and 12 months after surgery. All deaths appeared to be unrelated to the cryoablation procedure. One patient died of sepsis after surgery for necrotizing fasciitis, 1 patient died of pneumonia, and a third patient died of unknown causes. At last follow-up, 2 of these patients were in sinus rhythm and 1 patient was in AF. These patients were excluded from further analysis after their deaths.
Thirty-three patients (52%) left the operating room in sinus rhythm, wheras 29 patients (46%) required epicardial AV-pacing for junctional rhythm or sinus bradycardia. One patient (1.6%) was in AF at the conclusion of the operation. During the hospital stay, 38 patients (60%) had at least 1 episode of AF/AFl. At the time of discharge, 48 patients (76%) were free from AF, 12 (19%) were in AF, and 3 (5%) were in AFl or atrial tachycardia. After the third postoperative month, 54 patients were free from AF/AFl. In these patients, Kaplan-Meier analysis demonstrated an 88.5% freedom from AF rate after 12 months (Figure 2). There was no difference in outcome when comparing patients who underwent left atrial (LA) ablation versus LA and right atrial (RA) ablation. LA size and duration of AF were not predictors of heart failure (P>0.05).
Elective electrical cardioversion was performed in 8 of 63 patients (13%) for postoperative AF/AFl (Figure 3). Of these 8 patients, 5 were successfully converted to NSR and have remained free from AF/AFl. Two of the remaining 3 patients underwent percutaneous RF catheter ablation for refractory AFl. The first of these patients underwent a mitral repair/ASD repair and left and right atrial cryoablation for permanent AF and was discharged in AFl. Six months after discharge, a right atrial mapping study demonstrated the source of AFl to correlate to the inferior vena cava (IVC) cannulation site. After successful RF catheter ablation, the patient has remained in sinus rhythm. The second patient developed recurrent AFl 3 months after an ASD repair with right and left atrial cryoablation for permanent AF. Right atrial mapping also revealed AFl emanating from the IVC cannulation site. RF catheter ablation was initially successful in restoring sinus rhythm; however, AFl recurred 2 months later. A repeat percutaneous ablation was unable to establish NSR, and the patient remains in AFl. The third patient has not undergone ablation and remains in AF.
Holter Monitoring and of Use of Perioperative Antiarrhythmics
Initially in our protocol, patients in sinus rhythm at 3 months were to undergo Holter monitoring to check for asymptomatic paroxysmal AF that may not have been present on ECG. Poor compliance, however, resulted in approximately 20% of patients in sinus rhythm having Holter monitoring. None of those patients had evidence of paroxysmal AF.
Eleven (17%) patients were taking amiodarone on admission and 33 (52%) were taking amiodarone at discharge. After a mean follow-up of 11.4 months, the number of patients taking amiodarone was 3.4% (2 of 58).
Postoperative Pacemaker Placement
Twelve patients (19%) underwent permanent pacemaker (PPM) placement for postoperative symptomatic bradyarrhythmias. Six patients received PPMs for sick sinus syndrome and the remaining 6 patients received PPMs for symptomatic second-degree heart block. We performed a subset analysis to identify which patients were more likely to require PPM. Both age (P=0.010) and superior septal incision (P=0.041) were found to correlate with the need for postoperative PPM implantation.
One patient required re-exploration for mediastinal bleeding in the immediate postoperative period. A second patient returned 2 weeks after discharge with a large symptomatic pericardial effusion requiring operative drainage. A third patient underwent an open superior mesenteric artery embolectomy for small bowel infarction on postoperative day 2. The patient was in normal sinus rhythm at the time of this complication. No source for the embolus was identified, and the patient was found to have a hypercoagulable syndrome, lupus anticoagulant. This patient received anticoagulants and was discharged from the hospital in sinus rhythm on postoperative day 48. A fourth patient was readmitted 2 weeks fter a mitral valve repair for moderate to severe mitral regurgitation and underwent a subsequent mitral valve replacement.
Since the evolution of new ablative technologies and an improved understanding in the pathogenesis of AF, there has been a renewed interest in the surgical treatment of AF in patients who are already undergoing cardiac procedures. Several authors have demonstrated a wide range of success with cure rates between 60% and 90% using a variety of energy sources during endocardial ablation.9,11–14
In this study, we present the largest group of patients to undergo endocardial cryoablation using a flexible argon-based device during concomitant cardiac procedures in the surgical treatment of AF. Our results compare favorably with other published reports using a left atrial endocardial approach with hyperthermic energy sources such as radiofrequency and microwave.9,12,13 We chose to use this argon-based cryoablation system for its relative advantages over hyperthermic energy sources. These advantages include (1) the ability to cool to −160°C and create lesions in a timely fashion, (2) preservation of underlying tissue architecture, (3) less potential for thrombus formation, (4) sticking of the probe during freezing, which facilitates positioning, and (5) less risk of adjacent tissue injury (ie, circumflex coronary artery, esophagus).15 There have been no published reports of esophageal injury, such as those reported with unipolar radiofrequency energy, using cryoablation.16 This is particularly relevant during reoperative procedures, where dissection of the heart from adhesions can increase morbidity, especially in the presence of pre-existing bypass grafts. In our series, 6 patients underwent reoperative procedures with no additional morbidity.
Doll et al11 recently published results from 28 patients in whom the same argon-based cryoablation device was used during concomitant cardiac procedures. The procedures were performed either through a traditional sternotomy or via a mini-thoracotomy, with 74% of patients in sinus rhythm at 6 months. Although our results are similarly favorable, it should be noted that all patients in our study underwent sternotomy and left atrial cryoablation using the same left atrial lesion-set, with a mean follow up of 11.4 months. We do acknowledge, however, that the wide variation of concomitant procedures makes the data difficult to interpret. It was our intent to treat as many patients as possible with a history of AF who were to undergo a cardiac procedure. It was also our purpose to demonstrate that regardless of the complexity of the concomitant procedure, the addition of the LA cryoablation could be done in a timely manner, be efficacious, and add little or no additional morbidity.
LA size and duration of AF have been identified in previous reports as predictors of success rate after ablation procedures.17–19 It is therefore important to point out that the preoperative duration of AF in our report was about 2.5 years, which is somewhat shorter than that observed in other studies.9,11–14 In analyzing our data to identify predictors of failure, we could not corroborate the relationship between LA size or duration of AF with postoperative AF.
Because of the poor compliance with postoperative Holter monitoring, we subsequently removed it from our protocol. Although none of the patients studied were found to have paroxysmal AF, our results would be more robust if a greater proportion had undergone Holter monitoring.
Of the first 25 patients, 20 were discharged taking amiodarone. Only 13 out of the next 38 patients were discharged taking amiodarone. Although this may impact favorably on 3-month results, it is unlikely to have had a significant effect on the long-term results, as only 3.4% of the patients continued taking amiodarone at mean follow-up of 11.4 months.
The addition of a right atrial cryoablation or “isthmus lesion” (inferior vena cava to tricuspid annulus) did not correlate to a statistically significant difference in freedom from AF/AFl when compared with patients who only had left atrial cryoablation. This was consistent with the work of Deneke et al,20 who demonstrated there was no statistically significant difference in outcome in patients who underwent a left atrial ablation compared with those who underwent a bi-atrial ablation using a cooled-tip RF device. We did not randomize which patients would get right atrial ablation. In fact, patients were selected to receive right atrial ablation (1) in the presence of “right sided” pathology and right atrial dilatation, such as an atrial septal defect, moderate to severe tricuspid regurgitation, (2) during a superior septal approach to the mitral valve, or (3) if there was a history of atrial flutter. These were the groups of patients who were most likely to benefit from the addition of an isthmus cryolesions, as evidenced by our relatively low incidence of atrial tachycardia postoperatively. The addition of this isthmus lesion, however, should not be confused with a full right-sided Maze lesion set. Because a full right-sided lesion set was not performed, one cannot expect to see difference in postoperative AF. Another factor that must be taken into consideration was that the number of patients that received a bi-atrial cryoablation was small compared with the number of patients who received a left atrial cryoablation. It was not the purpose of this study to determine whether right and left atrial ablation correlated to a higher success rate than only left atrial cryoablation. To address such a question, each patient would need to be randomized to either undergo a left atrial cryoablation alone or a bi-atrial cryoablation, and outcome would need to be assessed in a blinded manner.
In the present series, 2 patients required postoperative percutaneous RF catheter ablation for atrial flutter, despite having undergone a right isthmus ablation. The focus of the flutter was a scar in the region of the IVC that correlated to the venous cannulation site. Although mostly described after congenital heart operations, post-surgical incisional tachycardias are also seen after operations for acquired heart disease, including left atrial ablative procedures for atrial fibrillation.21 In their series of 41 patients, Usui et al21 reported 4 patients (9.8%) with postoperative AFl who required postoperative percutaneous RF catheter ablation. Two of these patients were found to have AFl as a result of an atriotomy scar. As a result, it is recommended that atriotomies be placed longitudinally toward the lower aspect of the RA when performing RA ablations. An additional lesion is also recommended between the atriotomy and the IVC to block the reentry loops caused by atriotomy scars. We have since incorporated these modifications into all right atrial ablations.
The disadvantages of this argon-based cryoablation system are (1) the fact that lesions are not visible once they have thawed and (2) concern over the ability to reliably create transmural lesions during a closed-chest beating-heart procedure because of heat from intracardiac blood flow. Further developments in this system may allow for its use in a minimally invasive beating-heart approach.
The most important aspect to this report is that the surgical treatment of AF using this flexible argon-based cryoablation device was safe and effective during a wide variety of concomitant cardiac procedures. On the basis of these findings, all patients with any history of atrial fibrillation who are undergoing a cardiac surgical procedure should be considered for a concomitant ablation procedure.
Dr Mack has received a research grant from and is a consultant for CryoCath.
Benjamin EJ, Wolf PA, D’Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham heart study. Circulation. 1998; 98: 946–952.
Giata F, Gallotti R, Calo L, Manasse E, Riccardi R, Garberoglio L, Nicolini F, Scaglione M, Di Donna P, Caponi D, Francisosi G. Limited posterior left atrial cryoablation in patients with chronic atrial fibrillation undergoing valvular heart surgery. J Americ Coll Cardiol. 2000; 36: 159–166.
Melo J, Adragao P, Neves J. Surgery for atrial fibrillation using radiofrequency catheter ablation: assessment of results at one year. Eur J Cardiothorac Surg. 1999; 15: 851–855.
Yoshihara F, Toshio N, Yoshikado S, Hino J, Kobayashi J, Minatoya K, Bando K, Kosaki Y, Horio T, Suga S, Kawano Y, Matsuoka H, Yutani C, Matsuo H, Kitamura S, Ohe T, Kangawa K. Plasma atrial natriuretic peptide concentration inversely correlates with left atrial collagen volume fraction in patients with atrial fibrillation: plasma ANP as a possible biochemical marker to predict the outcome of the maze procedure. J Am Coll Cardiol. 2002; 39: 288–294.
Deneke T, Khargi K, Grewe PH, von Dryander S, Kuschkowitz F, Lawo T, Muller KM, Laczkovics A, Lemke B. Left atrial versus bi-atrial Maze operation using intraoperatively cooled-tip radiofrequency ablation in patients undergoing open-heart surgery: safety and efficacy. J Am Coll Cardiol. 2002; 39: 1644–1650.