Share this article on:

Advancing the Understanding of Surgical Cryothermy

Cheema, Faisal H. MD*†; Roberts, Harold G. Jr MD*‡

Innovations: Technology & Techniques in Cardiothoracic & Vascular Surgery: November/December 2012 - Volume 7 - Issue 6 - p 387–388
doi: 10.1097/IMI.0b013e318284eb16

From the *Aegis Cardiovascular Research Foundation, Fort Lauderdale, FL USA; †Division of Cardiothoracic Surgery, College of Physicians and Surgeons of Columbia University, New York, NY USA; and ‡Jim Moran Cardiovascular Research Foundation at Holy Cross Hospital, Fort Lauderdale, FL USA.

Accepted for publication November 21, 2012.

Disclosures: Harold G. Roberts, Jr, MD, is a consultant to Medtronic, Inc, Minneapolis, MN USA, and Edwards Lifesciences, Inc, Irvine, CA USA. Faisal H. Cheema, MD, declares no conflict of interest.

Address correspondence and reprint requests to Faisal H. Cheema, MD, Division of Cardiothoracic Surgery, College of Physicians and Surgeons of Columbia University, MHB 7 GN 435, 177 Fort Washington Ave, New York, NY 10032 USA. E-mail:

Cutting-edge technology, coupled with unbridled enthusiasm from the industry, continues to fuel the way for the development of innovative devices in cardiovascular medicine. An area where this has been particularly significant is the interventional treatment of atrial fibrillation (AF). Although the maze procedure was introduced by Dr James L. Cox more than 25 years ago, beginning around 2000, numerous power sources were developed with the goal of achieving less invasive yet equally reliable alternatives to the cut-and-sew method. In this issue, there are two excellent studies by Weimar et al,1,2 both from Washington University, the birthplace of the maze procedure, that impart some science to the field of cardiac surgical ablation by focusing on evaluation of a novel ablation system based on cryothermy.

See accompanying articles on pages 403 and 410

Cryothermy is the only power source that can single-handedly reproduce all the lesions of the Cox-maze III. Unlike its heat-based counterparts, it does not destroy collagen, thus making it particularly safe for valve tissue and the coronary sinus. In the first study by Weimar et al,1 the authors evaluated a cryoablation system in a porcine model of chronic AF by reporting an efficacy comparison of two commercially available nitrous oxide–based cryoprobes (Cryo1 vs 3011 linear probe; AtriCure, Inc, West Chester, OH USA) in creating transmural endocardial lesions. Using a miniature swine beating heart model, both bipolar radiofrequency and cryothermy were used to create various lesions documented by electrical isolation. When evaluated 14 days later, transmurality was confirmed in 99% of the Cryo1 versus 98% of the 3011 linear probe cross-sections, with no differences in lesion width, depth, or atrial thickness.

In this in vivo study by Weimar and colleagues,1 the main goal was to answer a clinically relevant question by characterizing the performance of the two previously mentioned nitrous oxide–based probes. Their in vivo efficacy was examined, and both were found to reliably create acute and chronic transmural lesions. However, some questions still remain as to their comparative performance as nitrous oxide–based probes (∼−60°C) with other commercially available argon-based devices (∼−140°C). Perhaps, carefully designed future studies to address these may prove to be extremely beneficial for our field and put this debate to rest. Furthermore, it should be kept in mind that the shorter periods of follow-up in this model may overestimate the efficacy of lesions because of the additive effect of edema associated with the acute injury. Presumably, months later, some of these lesions could allow conduction when the effects of the acute injury have subsided. In addition, because there is some evidence in the literature that endocardial application of cryoablation may lead to embolic phenomena,3 future studies should be designed to investigate the incidence and the severity of acute or chronic emboli.

The fact that the probe penetrates to only about 4 to 5 mm is potentially problematic for the often thicker human atria. The human mitral isthmus is one of the thickest areas of the atrium, with wall thickness reaching up to 10 to 12 mm. Therefore, achieving transmurality around the mitral isthmus region with a mere 2-minute application using these probes is quite questionable. Hence, when using these probes for creating the Cox-maze III lesion set, much longer periods of freezing, beyond the suggested 1 to 2 minutes, may be required. This further illustrates the importance of coupling the endocardial freeze with the “mirror” epicardial lesion via the oblique sinus when constructing a hopefully transmural mitral isthmus lesion.

Although cryothermy can faithfully replicate the lesions of a complete cut-and-sew operation, epicardial application alone does not ensure transmurality because of the heat-sink phenomenon in the warm, beating heart. One of the many appealing aspects of cryothermy is that, if it is applied on one side of the tissue and frost is seen on the other side, it is, by and large, a reliably transmural lesion. We have recently reported our data4 investigating the safety of cryothermy, and what seems to be a promising early article demonstrates the lack of coronary artery injury associated with cryothermy. This reinforces the fact that, despite its often close proximity to the circumflex artery when the coronary sinus is ablated, a mirror lesion of the mitral isthmus is safe and mandatory to obtain successful treatment of long-standing, persistent AF.

Using bovine myocardium in an ex vivo setting, the second study2 aimed at comparing two different nitrous oxide cryoablation probes: a new malleable 10-cm aluminum cryoprobe versus the currently available rigid 3.5-cm copper linear probe. Various parameters were recorded such as tissue temperature in various temperature water baths measured at 4- and 5-mm tissue depths, time to reach −20°C, and radial ice formation around the probes. Ice formation increased significantly with lower water bath temperatures. The width and the depth of ice formation were significantly less for the rigid linear probe. The new malleable probe achieved significantly lower temperatures at the proximal compared with the distal end. Although the time to reach a temperature of −20°C for both probes was faster for the 4-mm than the 5-mm tissue depths, the study concluded that the new malleable probe achieved rapid, clinically relevant freezing in tissues up to 5-mm thick. Both probes require 86 seconds to freeze the tissue at 5 mm. This leaves only a narrow overlap of time for AtriCure’s recommendation of 2-minute freezes per application. Many human atrial tissues are significantly thicker than 5 mm, particularly in the mitral isthmus at the atrioventricular fat pad. Ablation times of longer than 2 minutes should be considered if tissue depth exceeds 5 mm and mirror lesions are not possible, such as the portion of the box lesion around the left pulmonary veins.

The authors should be congratulated for both of these elegant studies that investigate important clinical questions in the surgical treatment of AF, namely, the performance and the characteristics of these nitrous oxide–based cryoprobes. However, it should be kept in mind that human atrial tissue can be as thick as 7 to 10 mm, so the results of this study done on tissues 4 to 5 mm thick may not be applicable to all kinds of atrial tissue.5 In addition, the potential efficacy cannot be guaranteed with a 1-minute application with the current clinically available probe, in that −20°C is barely reached at 1 minute in 4-mm specimens. Perhaps, longer application time in an in vivo setting is needed to really determine whether these cryoprobes are truly clinically effective. When the next phase of studies is conducted, one should aim to shift from the current ex vivo model, which is compromised by some limitations, to a totally in vivo setting to better incorporate the physiologic milieu. Alternatively, one can always conduct this study in an ex vivo setting using the explanted human hearts collected from transplantation.

Finally, another very helpful aspect of this article is the fact that it has elicited and emphasized the importance of a uniform probe-tissue contact. Wrinkles or pleats between the probe and the atrial tissue allow collection of fluid in the valleys. Ice has a much lower thermal conductivity than aluminum does, thus potentially insulating the atrial tissue and decreasing the likelihood of a transmural lesion.

Again, we would like to congratulate the authors for their past and present excellent studies of power sources that address some of the relevant clinical concerns in the ever-changing device world for surgical AF ablation. The new probe studied in these articles undoubtedly adds to the armamentarium of currently available probes to achieve surgical AF ablation. This is hopefully the beginning of laboratory studies assessing the new generation of cryoprobes. Overall, both of these are seminal articles providing in depth in vivo and ex vivo analysis of the performance of nitrous-based cryoprobes. These predict the efficacy of the probes in creating uniform, linear, transmural lesions in most atrial tissues. However, judgment should be exercised when encountering particularly thick atria, as often seen in some patients with underlying cardiomyopathies and morbid obesity.

Back to Top | Article Outline


1. Weimar T, Lee AM, Ray S, Schuessler RB, Damiano RJ. Evaluation of a novel cryoablation system: in-vivo testing in a chronic porcine model. Innovations. 2012; 7: 410–416.
2. Weimar T, Lee AM, Ray S, Schuessler RB, Damiano RJ. Evaluation of a novel cryobation system: in-vitro testing of heat capacity and freezing temperatures. Innovations. 2012; 7: 403–409.
3. Sauren LD, la Meir M, de Roy L, et al.. Increased number of cerebral emboli during percutaneous endocardial pulmonary vein isolation versus a thoracoscopic epicardial approach. Eur J Cardiothorac Surg. 2009; 36: 833–837.
4. Cheema FH, Pervez MB, Mehmood M, et al.. Does cryomaze injure the circumflex artery? A preliminary search for occult post-procedure stenoses. Innovations. 2012; 7: 73–80.
5. Hall B, Jeevanantham V, Simon R, Filippone J, Vorobiof G, Daubert J. Variation in left atrial transmural wall thickness at sites commonly targeted for ablation of atrial fibrillation. J Interv Card Electrophysiol. 2006; 17: 127–132.
Copyright © 2012 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.