Key Points
- •
Atrial fibrillation increases the risk for stroke and death.
- •
Surgical endocardial cryoablation is considered to be safe, effective, and useful in eliminating atrial fibrillation in patients with concomitant valvular heart disease.
- •
Pulmonary vein isolation alone may be a good option in patients with paroxysmal atrial fibrillation and normal atrium undergoing surgery for valvular heart disease.
- •
The majority of surgical lesion sets proposed for persistent/permanent atrial fibrillation ablation involves ablation of the posterior part of the left atrium, the pulmonary veins, and the creation of linear lesions connecting the pulmonary veins to the mitral annulus.
- •
The “7 linear scheme” is the simplest, most effective, and least time-consuming ablation scheme in permanent atrial fibrillation.
- •
The advent and development of a minimally invasive surgical procedure without sternotomy or cardiopulmonary bypass will also allow treatment of patients with lone atrial fibrillation.
Atrial fibrillation (AF) is the most frequent supraventricular arrhythmia, with an overall prevalence rate of 4%. This increases with age, ranging from 3% to 5% in people older than 65 years and 9% in those older than 80. AF is more common with concomitant structural heart disease related to coronary artery disease, cardiomyopathies, heart failure, hypertension, and valvular heart disease. AF is associated with irregular heart beat, loss of synchronous atrioventricular contraction, and stasis in the left atrium, which increased the risk for stroke and death.
The two main therapeutic strategies for AF are rate and rhythm control. Rhythm control is more desirable because it not only relieves the patient’s symptoms, but reduces the risk for thromboembolic events and improves cardiac performance by restoring the synchronized atrial contraction. Antiarrhythmic drugs are generally a lifelong therapy, and they are challenging because of limited efficacy and potentially hazardous side effects. For these reasons, new nonpharmacologic therapies have been developed. The AF ablation treatment goals are (1) restore a regular rate, (2) improve heart performance, and (3) restore atrial contraction. In the early 1990s, the first surgical experiences in AF ablation provided the proof of concept and paved the way for catheter procedure and less complicated operating techniques.
Electrophysiology of Atrial Fibrillation
It may be useful to consider our understanding of the underlying pathophysiology of AF, to evaluate the various strategies for AF ablation. Historically, both rapidly firing foci and reentry have been implicated as potential mechanisms. Random reentry with the circuits constantly changing and seeking nonrefractory tissue and the concept of the “wavelength of reentry” were subsequently popularized. The wavelength is the distance traveled by the electrical impulse in one refractory period. Thus, the wavelength (product of the refractory period and the conduction velocity) is the shortest path length that can sustain reentry. Either reduction of refractoriness or slowing of conduction velocity, or both, can reduce the wavelength, thus facilitating AF. The number of reentrant circuits in the atria at one time thus depends also on atrial size. Subsequent mapping studies of the atria also showed that the probability of AF perpetuation increased with the number of reentrant wavelets.
Atrial mapping during AF in humans showed that certain regions had rapid and fractionated activity (presumably “drivers”), whereas other regions had slower, regular activity, presumably “passively” activated. Furthermore, permanent AF had more regions with a disorganized pattern than did paroxysmal AF.
The importance of local “focal” areas of rapid activity in initiating and maintaining AF has also been recently demonstrated. It is currently believed that AF can be mechanistically heterogeneous, and both areas of local rapid activity and abnormal substrate contribute.
Surgical Ablation: History
In 1980, Williams et al. described a technique to isolate the left atrium electrically from the remaining part of the heart to confine AF to the left atrium, whereas the rest of the heart is restored to sinus rhythm. Based on the observation that sustained AF requires a critical mass of tissue, Guiraudon et al. reported the isolation procedure of a strip of atrial tissue between the sinus node and the atrioventricular node: the so-called corridor procedure. Theoretically, after this operation, fibrillatory impulses are separated from the sinus impulses and the AF no longer influences the ventricular rate. In this way, the sinus impulse travels across the corridor, whereas the remaining atrial tissue remains in AF. It was observed that after this procedure the left atrium frequently remained in AF, whereas the right atrium did not, showing the importance of the left atrium in the genesis of AF. The pitfalls of these two procedures are the loss of the left atrial “kick,” with the consequent drawback in ventricular function, and the inadequate effect in the prevention of thromboembolic events.
Cox et al. successfully treated AF using the maze procedure, which consisted of extensive surgical dissection and compartmentalization of both the left and right atria ( Figure 19–1 A ). The aim of this procedure was to guide the cardiac impulse through a “maze” from the sinus node throughout the atria via a narrow channel incapable of sustaining AF. Retrospectively, it can be asserted that the efficacy of this procedure was due to the isolation of left atrium including the pulmonary veins (PVs) and the back of the left atrium. Initially, Cox reported on 185 patients with postoperative sinus rhythm conversion of 93% with a mortality rate of 2.2%. After manifold modifications, now the Cox-maze procedure consists of a precisely defined pattern of biatrial incisions (Cox-maze III procedure) to interrupt the multiple wavelet macroreentry circuits that perpetuate AF, and its success rate is 97% to 99%. Further experience confirmed a high success rate but a complication rate and perioperative mortality rates varying from 1.3% to 2.1%. For this reason, it has failed to gain widespread use in the surgery ablation of AF. Different lesion sets and ablative tools have been evaluated to increase results and reduce complications and operative time. The aim of these procedures is either to modify the substrate, eliminating “anchors” of reentry by linear lesions, or target focal triggers residing in most cases in the PVs. All are essentially simplified variants of the original Cox-maze procedure, and all have in common the involvement of the posterior part of the left atrium, which represents a region with a high degree of disorganization for the impulse, and the PVs. These studies, in general, share a number of limitations including relatively low numbers of patients, nonstandardized antiarrhythmic drug protocols, incomplete long-term follow-ups, variable follow-up duration, and disparate monitoring strategies for recurrences. Most patients in these studies had structural heart disease requiring additional procedures to be performed at the time of their partial maze procedure, including mitral valve replacement or repair and coronary artery bypass.
Cryoenergy Ablative Tools
The experience with cryoenergy in open-heart surgery ablation of the arrhythmogenic substrate is extensive. The use of cryoenergy ablation has unique advantages: (1) It creates more homogeneous lesions than radiofrequency ; (2) it leaves the atrial endocardium intact ; (3) it significantly reduces the risk for thromboembolism ; (4) it preserves tissue architecture and collagen structure; and (5) it is locally effective without spreading its effects to the surrounding tissues. Differently from radiofrequency, particularly serious and potentially fatal complications such as esophageal perforations have not been reported with cryoablation until now.
Indeed, the lesions created by cryoablation used endocardially on an arrested heart are transmural and clearly limited. After an acute phase of inflammation and tissue infiltration, which may be in itself arrhythmogenic, the lesion evolves into dense fibrotic tissue, without any further evolution in dilation or rupture, because of the typical tissue necrosis induced by cryo (cell death without denaturation of the matrix proteins).
There are several sources of cryothermal energy, which may be used in cardiac surgery. The older technology uses nitrous oxide and is manufactured by Cooper Surgical (Trumbull, CT), which was purchased in 2007 by AtriCure (Cincinnati, OH). This system has a number of different shapes and sized probes, not disposable and easy to handle ( Figure 19–2 A ). Usually in open-heart surgery, the ablation takes places by the means of the maze linear probe with a 3.5-cm-long freeze area and a 1.5-cm-diameter flat-face curved probe. Most recently, Endocare (Irvine, CA) and CryoCath Technologies (Montreal, Quebec, Canada) have developed an argon gas device. This technology was acquired recently and is distributed by ATS Medical (Minneapolis, Minn.). With CryoCath, a disposable and malleable 10-cm cryosurgical probe is available; the probe has an adjustable insulation sleeve for varying lengths of the ablation lines (see Figure 19–2 B ).
Although the two systems are based on the same physics principles, they differ on the overall capability: at 1 atmosphere of pressure, nitrous oxide is capable of reaching the temperature of −89°C, whereas argon can reach the theoretical temperature of −185°C. This involves a faster cooling of the tissue, and thus a deeper and wider progression of the lesion. Consequently, the application time in the most recent devices is reduced to 1 minute, compared with the 2 minutes of the former devices. A longer and pliable probe in the CryoCath allows the surgeon to perform the complete ablation of the posterior left atrial procedure increasing the cross-clamp time of 6 to 10 minutes.
The probe consists of a hollow shaft, an electrode tip, and an integrated thermocouple for distal temperature recording. A console houses the tank containing the cooling gas at liquefied pressure. This liquid is pumped under high pressure to the electrode through an inner lumen. Once the fluid reaches the electrode, it expands and vaporizes from the liquid to the gas phase, absorbing energy and resulting in a rapid cooling of the tissue. The gas is then aspirated by vacuum through a separate return lumen to the console. At the tissue–electrode interface, there is a well-defined line of frozen tissue, called an ice ball. There are cryoablation systems that use liquid nitrogen, CO 2 , and more recently, near-critical nitrogen in experimental settings.
Endocardial Cryoablation
The first hybrid approach (cryo and scalpel) was proposed by Sueda et al. that excluded the PVs by incision and cryosurgery, and added two cryolesions connecting the encircled pulmonary region to the mitral annulus (see Figure 19–1 B ). The cryoablation was delivered at −60°C for 2 minutes. This procedure ablated the electrical activation of the left posterior wall of the left atrium, which had the shortest fibrillatory cycle length of both atria. Indeed, Sueda et al. showed that in patients with chronic AF and mitral valve disease, there is a regular and repetitive activation of the left atrium, and it was proved that the areas with the shortest cycle length were at the base of left atrial appendage and the posterior wall lateral to the left PVs. Furthermore, 78% of patients with chronic AF and associated valvular disease remained in sinus rhythm after 18 months of follow-up, with a reduction in the left atrial diameter and some recovery of both right and left atrial contractility. Gaita et al. reported 32 patients with chronic AF who underwent cryoablation limited to posterior left atrium during valvular heart surgery using a dual-probe cardiac cryosurgical system. The two probes were used simultaneously, close to each other, to shorten the procedure duration and to reduce the risk for any gap within the ablation line. The lesion scheme represented an inverted U shape to isolate the posterior left atrium and simultaneously reach the mitral annulus (see Figure 19–1 C ). In this series, 90% of the patients remained in sinus rhythm at the 12-month follow-up, including one patient requiring additional radiofrequency ablation for atrial tachycardia and five patients still taking antiarrhythmic drugs.
At that time, different techniques, more or less extensive, were proposed, but an important and unresolved issue was the knowledge of the real electrophysiologic effects of the surgical lesions. In fact, performing an ablation scheme does not necessarily mean that completeness and transmurality of the lesions have been achieved. It may happen that the intended surgical target is not always reached. In fact, lesions may be nontransmural or incomplete with a consequent presence of gaps. Even when the intended aim was the pure modification of the substrate through the creation of linear lesions, it is difficult to claim that the results were totally unrelated to PV isolation. With the aim of defining the best surgical lesion scheme and clarifying the role of PVs isolation, a study was done to compare three different ablation schemes: (1) a reversed “U” linear cryoablation interconnecting the PV ostia and the right and left lower PVs down to the mitral annulus; (2) a “7” linear cryoablation interconnecting the PV ostia and the left lower PV only down to the mitral annulus (see Figure 19–1 D ); and (3) anatomic isolation of the PVs performed just around the PV ostia. At 24 months, sinus rhythm without antiarrhythmic drugs was present in 57% of the groups U and 7, and in only 20% of the PV isolation group. The Carto mapping after the procedure further showed that practically, the intended U shape was never reached given that at least endocardially it was not possible to completely isolate the area between the right inferior PV and the mitral annulus. In addition, the intended lesion was achieved by the surgeon only about 50% of the time. Notably, when the linear lesions connecting the four PVs and the mitral annulus (“7” scheme) were obtained and confirmed, sinus rhythm without any drugs was achieved over a long term in 86% of the cases, whereas sinus rhythm was achieved in only 25% of patients with a complete electrical isolation of the PVs. According to this evidence, in permanent AF with associated valvular heart disease, PV isolation alone is insufficient.
The importance of complete isolation of the encircled PV area was confirmed by Todd et al., who performed isolation of the PVs and posterior left atrium using a scalpel incision and cryosurgery in 14 patients with drug-refractory lone AF. A cryoablation line connecting the excluded region to the mitral ring was added. The efficacy of the electrical isolation procedure was verified after surgery. Two patients had early documented recurrent arrhythmia, with one showing a gap in the PV area encirclement subsequently completed by catheter ablation. During long-term follow-up (25.1 months), all patients remained in sinus rhythm.
Recently, several lesion sets aiming to limit the surgical trauma and the procedure time have been reported. In some cases, a biatrial approach was performed, whereas in others a limited procedure to the left atrium was performed. In some schemes, the en bloc isolation of the PVs is performed alone (see Figure 19–1 G ) or with additional lines from the isolated segment to the mitral annulus and to the left atrial appendage (see Figure 19–1 H ). Other techniques are based on the isolation of the PVs in two blocks, one for the left and one for the right PV; further lines can be added to connect the PV blocks, the left PV block to the posterior segment of the mitral valve and to the left atrial appendage (see Figure 19–1 E, F ). In general, since the amendment of the cut-and-sew techniques of the old Cox-maze I and introduction of several energy sources, the procedural time for the surgical ablation for AF is limited to 10 to 20 minutes according to the lesion set, with a reported success rate of 65% to 100% ( Table 19–1 ).
