Early Experience With the Cryoablation Balloon Procedure for the Treatment of Atrial Fibrillation by an Experienced Radiofrequency Catheter Ablation Center

Download a PDF of this article.

Abstract: Background. Radiofrequency (RF) catheter ablation has provided an effective method for treating drug-refractory symptomatic atrial fibrillation. Recently, a cryoablation balloon approach has also received approval. The purpose of this study was to compare RF catheter ablation to cryoablation for the treatment of atrial fibrillation with respect to safety, immediate efficacy, and effects on procedural and fluoroscopy times. In addition, actual procedural costs were compared. Methods. This study was approved by the Winthrop University Hospital Institutional Review Board to retrospectively examine cryoablation with the Arctic Front Cardiac CryoAblation balloon catheter (Medtronic, Inc) and compare it to RF catheter ablation for the treatment of drug-refractory symptomatic atrial fibrillation. Patient and procedural characteristics as well as immediate success were compared. Immediate failure was defined as incomplete pulmonary vein isolation of all veins. Results. A total of 124 procedures (62 RFs and 62 cryoablations) were performed from December 2010 through July 2012. The cryoablation procedure took longer to perform than RF (171 ± 61 minutes vs 126 ± 49 minutes, respectively; P<.0001). There was no difference in fluoroscopy times between the two groups (29 ± 20 minutes for RF vs 32 ± 18 minutes for cryoablation; P=.39). The infusion of protamine following procedures was much more common in the cryoablation group (30 patients vs 2 patients in the RF group; P<.0001). The immediate success rate was 93.5% with RF ablation vs 96.7% with cryoablation (P=NS). There was not a significant difference in complications between the two approaches. The cost for each procedure was $24,391.88 ± 4826.77 for RF and $31,874.02 ± 8349.70 for cryoablation (P<.0001). Conclusion. Cryoablation provides an additional and alternative approach to RF ablation for the treatment of symptomatic drug-refractory atrial fibrillation with comparable immediate success and complications. It is synergistic with RF and permits the ability to tackle the entire gamut of atrial fibrillation (ie, paroxysmal and persistent). This study showed no decrease in procedural or fluoroscopy times with our early experience. One significant limitation with cryoablation is the cost. Cryoablation resulted in over $7000 extra cost to the hospital per procedure. The clinical benefits achieved by this additional cost warrant further investigation.

J INVASIVE CARDIOL 2013;25(6):288-292

Key words: atrial fibrillation, cryoablation, radiofrequency ablation

______________________________________

For many years, symptomatic atrial fibrillation has been successfully treated via radiofrequency (RF) catheter ablation in which pulmonary vein isolation (PVI) formed the cornerstone of treatment. The most common PVI procedure was performed using RF energy. More recently, cryoablation with the Arctic Front Cardiac CryoAblation balloon catheter (Medtronic, Inc) received approval by the United States Food and Drug Administration (FDA) as another method besides RF ablation to treat symptomatic drug-refractory paroxysmal atrial fibrillation. This novel technology is used specifically to perform pulmonary vein isolation. A single probe, which includes an inflatable balloon, is placed across the atrial septum into the left atrium.¹ This system pumps refrigerant (N₂O) into the balloon, while contacting the pulmonary vein orifice.¹ Contrast is injected into the pulmonary vein in order to demonstrate a good seal and physical contact with the balloon. A 4-minute freeze is typically applied to each pulmonary vein followed by a thaw and then another freeze.¹

There have been few trials that have compared a “real-world” clinical experience with the cryoablation balloon technique to RF catheter ablation since the system received FDA approval. The purpose of this study was to compare RF catheter ablation to cryoablation for the treatment of atrial fibrillation with respect to safety, immediate efficacy, and effects on procedural and fluoroscopy times. In addition, actual procedural costs were compared.

Methods

This study considered procedures starting in December 2010 through July 2012 and only included operators who performed cryoablation (4 operators). Cryoablation began in October 2011 and these procedures were matched to RF ablations that immediately predated this newer method. These techniques were evaluated in patients with paroxysmal atrial fibrillation (recurrent episodes that self-terminate in <7 days), persistent atrial fibrillation (recurrent episodes that last >7 days), and long-standing persistent atrial fibrillation (recurrent episodes that last longer than 1 year) including initial and repeat procedures. Patient and procedural characteristics were compared including procedural and fluoroscopy (radiation exposure) times, completeness of pulmonary vein isolations, complications, and actual procedural costs. Data were obtained from a variety of hospital- and office-based electronic medical record systems, the hospital’s finance department, as well as the physical electrophysiology log sheets in which fluoroscopy times were recorded. Complications were cross-checked against those reported to hospital-wide quality improvement.

All patients included in this trial were required to have symptomatic atrial fibrillation and failed or couldn’t tolerate antiarrhythmic drug therapy. Patients who were not adequately anticoagulated and in atrial fibrillation underwent transesophageal echocardiography prior to the procedure to rule out left atrial thrombus. Most patients who were on warfarin therapy underwent the procedure on therapeutic anticoagulation. Patients not on therapy and in sinus rhythm were started on therapeutic anticoagulation after the procedure. Informed consent was obtained in all patients and each received deep sedation administered by an anesthesiologist. Computer axial tomographic angiography was performed in all patients without significant renal insufficiency and used for three-dimensional reconstruction of the left atrium prior to each procedure. During the procedure, patients received intravenous heparin to a target activated clotting time (ACT) of 350 to 400 seconds. Imaging was performed with intracardiac echocardiography (to assist with the transeptal punctures and monitor for pericardial effusion), single-plane fixed fluoroscopy, and non-fluoroscopic three-dimensional electroanatomical imaging (Ensite NavX; St. Jude Medical) in all patients. During cryoablation of the right pulmonary veins, the operator performed continuous pacing of the phrenic nerve in order to detect the diaphragmatic contraction. Any loss of contraction prompted immediate therapy termination in order to prevent diaphragm paralysis. After the procedure, protamine was occasionally administered at the discretion of the operator. Procedure start (percutaneous access) and end times were determined independent of the electrophysiologist and recorded by the nursing and anesthesia team. The radiology technician recorded fluoroscopy times. 

Immediate success was defined as demonstrated complete isolation of all pulmonary veins by the end of the procedure as indicated by complete entrance and exit block. Anything short of this was deemed a procedural failure. Additional RF ablation procedures, including atrial flutter, continuous fragmented electrical activity, and linear lines such as those involving the superior vena cava, coronary sinus, mitral valve isthmus, and left atrial roof, were performed as indicated at the discretion of the operator. 

Statistical analysis. Patient characteristics, procedural data, and outcomes were compared using Fisher’s exact test and Student’s unpaired t-test such that P≤.05 was arbitrarily determined to be statistically significant. All data are expressed as mean ± the standard deviation.

Results 

This study examined patients at Winthrop University Hospital who underwent atrial fibrillation ablation by operators trained in cryoablation therapy. The RF pool included all PVI patients from December 16 2010 up to October of 2011, in order to achieve an equal number of RF-only cases matched to cryoablation. No RF-only ablations since the introduction of cryoablation were considered. This study consisted of a total of 124 procedures (62 RFs and 62 cryoablations). Pure RF cases typically received two separate transseptal punctures (one for mapping and one for ablation); cryoablation procedures only had one puncture performed. 

Table 1 shows the patient characteristics of the RF and cryoablation patients. The two groups had comparable ages (60 ± 8 years for RF vs 61 ± 9 years for cryoablation; P=NS), body surface mass indices, and left atrial sizes. There was a higher percentage of pure paroxysmal atrial fibrillation patients in the cryoablation group than in the RF group (67.7% vs 54.8%, respectively; P=.04). Note, there were only 3 long-standing persistent patients in this trial (1 in the RF group and 2 in the cryoablation group). There was no significant difference between the two groups with respect to valve disease (moderate or more severe), coronary artery disease, congestive heart failure, obstructive sleep apnea, left ventricular hypertrophy, and hypertension. There was a trend toward more pulmonary disease and male gender in the RF group, although this was not statistically significant. In the RF group, there was 1 pacemaker and 5 implantable defibrillators; in the cryoablation group, there were 0 pacemakers and 1 implantable defibrillator (P=NS). 

Table 2 shows procedural characteristics in the two groups. In general, cryoablation procedures took longer than those of RF (171 ± 61 minutes vs 126 ± 49 minutes, respectively; P<.0001). This included all patients regardless of atrial fibrillation type, procedural complexity, protamine infusion, or repeat procedure. The infusion of protamine following procedures was much more common in the cryoablation group (30 patients vs 2 patients, respectively; P<.0001). Even excluding the protamine infusion patients, however, the cryoablation still took longer than RF (155 ± 61 minutes vs 124 ± 48 minutes; P<.011). Importantly, there was no difference in fluoroscopy times between the two groups (29 ± 19 minutes for RF vs 32 ± 18 minutes for cryoablation; P=NS). The first 31 cryoablation cases were compared to the latter 31 in order to gauge the learning curve effects with respect to procedural times and fluoroscopy times. The procedural time was 185 ± 62 minutes for the first 31 cases vs 156 ± 59 minutes for the second 31 cases (P=.06). The fluoroscopy time was 36 ± 17 minutes for the first 31 cases vs 27 ± 17 minutes for the second 31 cases (P=.04). In addition, with respect to cryoablation efficacy, the two failures were in the first half of the cryoablation cases performed (procedures #17 and #24) and there were no failures in the latter half of the cryoablation procedures. There were significantly more first-time cyroablation procedures vs RF (85.5% vs 53.2%; P=.0002).

Table 3 shows the complexity of procedures performed in each group. Forty-nine RFs and 47 cryoablations had PVI alone. Twelve RFs and 15 cryoablations had either concomitant cavo-tricuspid isthmus ablation for atrial flutter, superior vena cava ablation, linear lines (coronary sinus, mitral valve isthmus, and/or left atrial roof), and/or ablation of continuous fragmented atrial electrical activity (CFE), and 1 had a slow pathway ablation to treat atrioventricular nodal re-entrant tachycardia. Seven cryoablations received additional RF applications in order to complete PVI (so called “touch-up” ablation). One had only a continuous fractionated electrical activity ablation with no PVI or lines created. In addition, 13 cryoablation patients received RF to treat right-sided isthmus dependent atrial flutter; 2 other cryoablation patients had concomitant RF left-sided lines created. 

The immediate success rate of PVI with RF was 93.4% (57 of 61 RF procedures, with 1 additional RF procedure already having all pulmonary veins isolated at onset); and 96.7% for cryoablation (P=NS). There were 4 failures in the RF group and 2 in the cryoablation group. Two of 4 failures in the RF group were the result of pulmonary veins that were either too small or stenotic to be cannulated. One RF failure was due to unreliable bidirectional block and case #4 was aborted due to hypotension attributed to anesthesia. In the cryoablation group, 1 failure was due to a right middle pulmonary vein that was too small to be cannulated and the other was due to an aborted case due to transient diaphragmatic paralysis.

There was a total of 5 complications in each group. RF complications were groin hematoma/bleeding (n = 3) and significant hypotension (n = 2; 1 with asystole/1 with small pericardial effusion). Cryoablation complications were groin hematoma/bleeding or pseudoaneurysm (n = 3); renal infarction in a patient in whom dabigatran was stopped for 4 doses prior to the procedure and restarted 6 hours after sheath removal (n = 1; note that patient did not receive protamine); transient injury to phrenic nerve that occurred during cryoablation of the right upper pulmonary vein, with diaphragm paralysis that resolved by the end of the procedure (n = 1). The latter patient had the procedure aborted and only had the left-sided pulmonary veins isolated (and therefore was classified as a procedural failure); this patient has had no symptoms of atrial fibrillation recurrence off of antiarrhythmic drugs after 147 days of follow-up. There was no evidence of pulmonary vein stenosis in either group. Additionally, there were a few patients referred for cryoablation who were unable to undergo this procedure due to an inability to receive contrast dye as required by the procedure; 1 patient had an intended cryoablation; however, the cryoballoon delivery sheath was unable to be advanced across the interatrial septum into the left atrium and a smaller sheath was advanced and RF ablation was successfully performed.

Winthrop University Hospital’s Finance Department provided us with the procedural costs associated with RF and cryoablation in 118 procedures (59 RFs and 59 cryoablations). The mean total cost to Winthrop University Hospital for each procedure was $24,391.88 ± 4826.77 for RF and $31,874.02 ± 8349.70 for cryoablation (P<.0001).

Discussion

This study retrospectively examined a “real-world” early experience at a university hospital in the United States with the only FDA-approved cryoablation balloon system to treat atrial fibrillation and compared it to an immediately preceding mature RF ablation. This study demonstrated comparable outcomes with respect to immediate success and procedural complications with cryoablation as compared to RF regardless of the atrial fibrillation classification. Unlike the RF group, the cryoablation group included the learning curve with this technique. Our study clearly demonstrated this cryoablation learning curve (by comparing the first 31 cryoablation procedures to the second 31 cryoablation procedures) with 100% success in the latter half of procedures (as compared to 93.5% in the first half of the procedures) plus shorter fluoroscopy times and a trend toward shorter procedural times.

Cryoablation procedures in general took longer than RF-only procedures. This may be related to the learning curve, the lengthy time for the recommended freeze-thaw-freeze applications (for 4 pulmonary veins, the applications alone typically took 32 minutes), time needed for additional RF “touch-up” procedures, and high protamine usage (48.3%; related to the large balloon introducer diameter of approximately 15 mm). Importantly, cryoablation with the cryoballoon itself does not preclude the use of RF and in many circumstances this combined approach permitted a complete ablation of a number of putative atrial fibrillation triggers. One major drawback of cryoablation is the cost. Our study obtained the actual costs levied against a particular procedure and found that cryoablation on the average was over $7000 more expensive per patient. This is slightly more than the cost of a disposable cryoballoon system for each case. Electroanatomical mapping and intracardiac echocardiography were still utilized in this study for cryoablation as they were RF. Perhaps the cost difference will decline with further experience and less reliance on ancillary technologies.

This study is similar in design to the British study by Kojodjojo and colleagues; however, our results are quite different.¹ Kojodjojo and colleagues looked at an entire cryoablation approach in 124 patients (from 2006 to 2009), but only compared the paroxysmal atrial fibrillation ablation patients who underwent cryoablation (90 patients) to RF (53 patients). They included only first-time studies, and essentially excluded patients who received more complicated ablative therapies such as linear lines and treatment of continuous fragmented electrical activity.¹ Their study demonstrated a 99% immediate success rate with RF as compared to 83% with cryoablation with respect to PVI. In addition, the procedural and fluoroscopy times were cut in half with cryoablation as compared to RF (P<.001).¹ Only the RF patients in their study underwent electroanatomical mapping, and it was unclear whether protamine administration was given and/or considered as part of the entire procedure. Their improved procedural and fluoroscopic findings with cryoablation were in contrast to our current study, which demonstrated statistically longer procedural times with cryoablation as compared to RF; and no improvement in fluoroscopy times between the two approaches. Most importantly, our study found similar immediate procedural success with cryoablation, whereas the Kojodjojo study appeared to have a lower immediate success rate, although no statistical analysis was performed. 

Linhart and coworkers reported on an early international experience in 2009 with cryotherapy as compared to RF ablation.² They performed a case control study using cryoablation (cryoballoon with or without standard non-balloon cryotherapy as a touch-up) in order to isolate the pulmonary veins.² They matched their cryoablation experience in 20 patients to 20 patients who underwent RF catheter ablation in order to treat drug-refractory atrial fibrillation. The cryoballoon method was nearly equivalent to RF energy and resulted in a 50% success rate as opposed to 45% for RF energy after 6 months of follow-up. The success rate with an entire cryoballoon plus non-balloon cryocatheter “touch-ups” increased the success rate to 66%.²

Study limitations. Our study has a number of significant limitations. First, it is a retrospective study and is merely looking back at our laboratory’s experience over nearly the past two and a half years. Second, there was no controlled approach and or follow-up. In addition, 4 different operators used different procedural judgments and follow-ups were employed, giving a “real-world” experience, but limiting the extent of our conclusions. There were no controlled standard follow-up times and monitoring modalities employed across the different operators. Two of the four operators had patients referred from outside the region and did not have longer-term data on many of those patients. Third, cryoablation was only recently approved for use in the United States and introduced into our laboratory in October 2011 and almost half the cryoablation patients did not have 3 months of follow-up. Fourth, postprocedural protamine administration was at the discretion of the operator, based on laboratory time and issues related to groin access. The protamine infusion added significant “in-laboratory” procedure time, which included at least 10 minutes for the infusion plus time for an ACT check and a lengthy in-laboratory groin access (typically 40 minutes).

Currently, the FreezeAF trial is a randomized controlled trial that is prospectively investigating the cryoballoon as compared to RF catheter ablation.³ This trial plans to enroll a total of 244 patients with drug-refractory paroxysmal atrial fibrillation.³ Each group is limited to their particular therapy modality with repeat therapies permitted after 6 months.³ The primary endpoint is absence of atrial fibrillation recurrence off of drug therapy and without procedural complications.

Other trials have reported on the potential for inflammatory, embolic, and neurologic effects regardless of cryoablation and RF approach.^4-6 These studies do not show any difference between the two approaches. Only 1 embolic event related to a procedure was noted in our study as a result of cryoablation — a renal infarction potentially related to stopping and restarting a direct thrombin inhibitor (dabigatran). There were no neurologic events observed.

Conclusion

Cryoablation with the FDA-approved cryoablation balloon system provides an additional and alternative approach to RF ablation for the treatment of symptomatic drug-refractory atrial fibrillation. Cryoablation is synergistic with RF and permits the ability to tackle the entire gamut of atrial fibrillation (ie, paroxysmal and persistent). Outcomes are comparable between RF and cryoablation; however, this study showed no decrease in procedural or fluoroscopy times. One significant limitation with cryoablation is the cost. Cryoablation resulted in over $7000 extra cost to the hospital per procedure. The clinical benefits achieved by this additional cost warrant further investigation. In addition, cryoablation cannot be performed in patients who are intolerant to contrast dye (ie, significant allergy history or renal dysfunction). A randomized prospective controlled trial (the FreezeAF trial) is underway and will hopefully answer many of the questions with respect to long-term outcomes and complications.

Acknowledgments. We gratefully acknowledge the help and assistance of Donald Brand, Director of Clinical Trials, for explaining the statistical analysis required in this study. The authors also wish to thank Eileen Dunne in the Finance Department at Winthrop University Hospital for providing the cost of the specific procedures included in the study.

References

1. Kojodjojo P, Davies WD. How to perform antral pulmonary venous isolation using the cryoballoon.  Heart Rhythm. 2011;8(9):1452-1456. 
2. Linhart M, Bellmann B, Mittmann-Braun E, et al. Comparison of cryoballoon and radiofrequency ablation of pulmonary veins in 40 patients with paroxysmal atrial fibrillation: a case-control study.  J Cardiovasc Electrophysiol. 2009;20(12):1343-1348.
3. Luik A, Merkel M, Hoeren D, Riexinger T, Kieser M, Schmitt C. Rationale and design of the FreezeAF trial: a randomized controlled noninferiority trial comparing isolation of the pulmonary veins with the cryoballoon catheter versus open irrigated radiofrequency ablation in patients with paroxysmal atrial fibrillation.  Am Heart J. 2010;159(4):555-560.e1.
4. Herrera Siklódy C, Deneke T, Hocini M, et al. Incidence of asymptomatic intracranial embolic events after pulmonary vein isolation: comparison of different atrial fibrillation ablation technologies in a multicenter study.  J Am Coll Cardiol. 2011;58(7):681.
5. Gaita F, Leclercq JF, Schumacher B, et al. Incidence of silent cerebral thromboembolic lesions after atrial fibrillation ablation may change according to technology used: comparison of irrigated radiofrequency, multipolar nonirrigated catheter and cryoballoon.  J Cardiovasc Electrophysiol. 2011;22(9):961-968.
6. Herrera Siklódy C, Arentz T, Minners J, et al. Cellular damage, platelet activation, and inflammatory response after pulmonary vein isolation: a randomized study comparing radiofrequency ablation with cryoablation.  Heart Rhythm. 2012 Feb;9(2):189-96.
7. Haïssaguerre M, Jaïs P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins.  N Engl J Med. 1998;339(10):659-666.