|
|
||||||||
Division of Cardiac Surgery, S Raffaele University Hospital, via Olgettina 60, 20132 Milan, Italy
Received 26 September 2007; received in revised form 29 December 2007; accepted 1 January 2008.
* Corresponding author. (Email: stefano.benussi{at}hsr.it).
| Abstract |
|---|
|
|
|---|
Key Words: Atrial fibrillation Radiofrequency ablation Bipolar radiofrequency Cardiac surgery Mitral valve disease Left isthmus Mitral line
| 1. Introduction |
|---|
|
|
|---|
Reports of coronary injury related to AV groove ablation to perform the mitral connecting line have given rise to some concern about the hazardous nature of such ablation [6,7].
Bipolar radiofrequency clamps are to date the most reliably transmural ablation devices available to the surgeon [4,8,9]. Nevertheless, due to its clamping nature and its deep tissue penetration, bipolar radiofrequency has not been used to perform the mitral connecting line. So, the rising popularity of bipolar radiofrequency has led many surgeons to either treat patients with incomplete lesion sets, which is clinically not convenient [2,4], or to use an additional unipolar device to complete the mitral line, which is probably not as efficient, and surely not cost-effective.
We describe here our results with an original technique to perform a complete left set of lines, entailing the mitral line, using bipolar radiofrequency only.
| 2. Methods |
|---|
|
|
|---|
Sixty-seven patients (96%) had primary mitral valve disease (rheumatic in 42, degenerative in 23, functional in two) as the main indication to surgery. Preoperative data are summarised in Table 1 .
|
A series of 33 patients undergoing concomitant ablation of AF using two ablation devices, a bipolar and a unipolar RF handpiece, during open-heart surgery, 31/33 (94%) for mitral valve disease (rheumatic/degenerative = 20/13) from September 2004 to November 2006, was used as control group.
Male/female ratio (9/24), age (66 ± 9 years), NYHA class distribution (II/II/III: 1/21/55 patients), ejection fraction (53 ± 10%), left atrial diameter (54 ± 8 mm), type of AF (paroxysmal/persistent/permanent: 0/11/22 patients) did not differ from the study population. But the control group had a longer duration of persistent/permanent AF; median 60 months; interquartile range 36–117 (p = 0.007).
A transoesophageal echocardiography was performed within 24 h of surgery in all patients to rule out intracavitary thrombosis.
Informed written consent was signed by each patient before surgery.
2.1 Surgical technique
The specific anatomy of the circumflex and of the right coronary arteries was thoroughly considered at coronary angiogram, when available.
All ablations were performed using bipolar radiofrequency, the Cobra Bipolar (Estech Inc., Camino Ramon, CA) was used in 38 patients, the BP2 (Medtronic Inc., Minneapolis, MN) in 21, the Isolator (Atricure Inc., Cincinnati, OH) in 11.
The encircling ablations around the right and left PV orifices were performed epicardially, usually on cardiopulmonary bypass. Then coronary anatomy was inspected by lifting the fully decompressed beating heart. The terminal branch of the circumflex and of the right coronary artery were identified, and a 20 mm hypodermic needle was stuck in the posterior left atrial wall, just above the coronary sinus, to mark the middle of such coronary-free area of the AV groove (Fig. 1 ). Then, after aortic cross-clamping, the atrium was incised parallel to the Waterston groove. The projection of the needle on the posterior mitral annulus was identified (Fig. 2 ). The needle was removed. An ablation line was then carried out by clamping the left atrial wall in an endo-epicardial fashion, with the tip of the inner jaw clearly biting the posterior mitral annulus and the proximal portion of the clamp intersecting the atriotomy (Fig. 3 ). Another ablation was generally performed after slightly slipping the clamp backward, off the ventricular part of the groove which, being thicker, might prevent a proper contact of the jaws with the thin atrial wall. The coronary sinus, which lies totally on the atrial side of the AV groove, is thus circumferentially ablated. The resulting ablation might be inspected by lifting the back of the atrium (Fig. 4 ).
|
|
|
|
No ablation was generally carried out at the base of the appendage, based on the assumption that a watertight suture of its base would cause a fibrotic involution anyhow.
2.2 Postoperative management and follow-up
Perioperative antiarrhythmic prophylaxis with amiodarone was administered for 3–6 months in the absence of contraindications [10].
Standard 12 leads ECG, Holter and transthoracic echocardiography monitoring were performed 3, 6 and 12 months after operation.
2.3 Statistical analysis
Data are expressed as mean ± standard deviation unless stated otherwise. Non-normally distributed variables, as assessed by the Kolmogorov–Smirnov test, are expressed as median, interquartile range (25°–75° percentile) for continuous variables, and as N (%) for categorical variables.
Preoperative variables as listed in Table 1 and operative variables such as replacement versus reconstructive valve surgery, duration of cardiopulmonary bypass and of cardiac arrest and overall intensive care unit and hospital stay were analysed for a possible correlation with relapsing tachyarrhythmia after surgery.
Association between arrhythmia recurrence at each time point and all other variables was investigated by
2-test and by Mann–Whitney U-test as appropriate.
Preoperative variables, intraoperative variables (including cost of the ablation devices per patient) and clinical outcomes of the study group were assessed for differences versus the control group by means of
2-test or Student's t-test as appropriate.
| 3. Results |
|---|
|
|
|---|
|
Overall, the median intensive care unit stay was 2 days (interquartile range 1–3) and the median postoperative hospital stay was 7 days (interquartile range 5–9).
Two patients (3%) were implanted with a permanent pacemaker after surgery for AV block: one after mitral valve replacement in a heavily calcified mitral annulus and one after aortic valve replacement, mitral valve repair involving implantation of a rigid ring and tricuspid rigid ring annuloplasty.
Follow-up was 100% complete (mean 11 ± 5.6 months, median 10 months; interquartile range 6–13.5).
At discharge 45/70 patients (64%; 95% CI: 53%, 75%) were in normal sinus rhythm. Subsequent freedom from AF was 84% (95% CI: 75%, 93%, 56/67 patients) at 6 months and 81% (95% CI: 70%, 93%, 35/43 patients) at 1 year (Fig. 5 ).
|
After excluding patients with paroxysmal AF, patients with permanent AF had a lower success rate at 6 months than those with persistent AF (
2
= 6.73; p
= 0.009). None of the other preoperative or intraoperative variables correlated with rhythm outcome at any follow-up interval.
Among patients without a mechanical valve prosthesis, freedom from oral anticoagulation was 48% (22/42) at 6 months and 57% (12/21) at 12 months. No thromboembolic event was recorded. No patient died.
3.1 Control group
There was no significant difference between the study and the control groups in intra and postoperative clinical variables, namely cardiopulmonary bypass time (105 ± 19 min), aortic cross-clamp time (82 ± 17 min), bleeding (median 400 cc; interquartile range 300–600), intensive care unit (median: 2 days; interquartile range 1–4) and hospital stay (median: 7 days; interquartile range 6–9).
Instead, as predicted, the per patient cost of the ablation devices was nearly doubled: 2403 ± 17 \#8364; in the control group versus 1245 ± 50 \#8364; in the study group (p < 0.0001).
No difference was found in the sinus rhythm recovery rate, which was 55% (95% CI: 38%, 72%, 18/33 patients) at discharge, 76% (95% CI: 60%, 91%, 22/29 patients) at 6 months and 74% (95% CI: 57%, 81%, 20/27 patients) at 12 months. One patient (3%) in the control group experienced right flutter, while three patients (9%) had left flutter (p = ns vs study group).
| 4. Discussion |
|---|
|
|
|---|
This is of no small consequence since growing evidence both from transcatheter ablation [11,12] and from AF surgery [2,4,13] points to the relevance of completeness of the lesion set and of transmurality and continuity of the ablations in determining clinical outcome. Gaita and colleagues showed that the final set of lines is particularly crucial to rhythm outcome in patients undergoing heart surgery with permanent AF [13]. In fact those missing the connecting lines at postoperative electrophysiological study, had more than a 70% chance of relapsing AF or flutter. In a review of the initial experience with bipolar ablation at the Cleveland Clinic, Gillinov et al. has recently confirmed that, while PV isolation alone may relate to a good rhythm outcome in patients with paroxysmal AF, connecting lines are also necessary to prevent arrhythmia recurrence in those with permanent AF [2,3].
Continuity and transmurality of the scars strongly relates to the adopted ablation tool. Unipolar devices have proven effective in creating transmural ablation endocardially on the arrested heart [14–16]. Nevertheless, shortcomings like scarce penetration during epicardial application on the beating heart [17,18], collateral damage due to thermal spreading [19], and a poor predictability of completeness of the lesions due to patient specific and operator dependent factors [20], have limited their popularity.
Since its first introduction in clinical practice, bipolar radiofrequency appeared as the natural answer to most of such shortcomings. The clamping design of bipolar devices clears out heat sink due to convective cooling exerted endocardially by circulating blood, thus rendering transmurality of the lines highly predictable after both epicardial and endocardial ablation [4,8,9,21]. Temperature or impedance feedback featured by bipolar radiofrequency devices improves the reproducibility of complete tissue penetration. Collateral damage is extremely unlikely; only clamped tissue can be significantly heated by bipolar devices.
Due to these features bipolar radiofrequency has quickly gained popularity in the field of open-heart ablation. Clinical results are actually satisfactory, both in left atrial and in biatrial ablation procedures [2,4,5]. PV isolation is easily achieved by epicardial clamping on the antral portion of the left atrium. Connections between the encirclings are easily feasible by clamping the atrial wall in an endo-epicardial fashion through the atriotomy. Actually, such posterior connecting lines can also be carried out epicardially on the beating heart [22]. But performing the connecting line to the mitral annulus with a bipolar clamp has not been reported so far. Such reluctance is understandably due to the concern of injuring a major coronary vessel in the AV groove with a transmurally ablating device. In fact, coronary injury has been described after endocardial ablation with virtually all sorts of energy sources, comprising unipolar radiofrequency and cryoablation [6,7]. In our initial experience with bipolar radiofrequency, we also omitted such line and concentrated on the quality of the remaining ablations. This turned into a disturbingly high rate of left flutter, occurring in 7% of the patients [4].
We devised a technique, which allows performing all left atrial ablations with a single bipolar device. We had already underlined the importance of coronary anatomy to safely perform the mitral connecting line with unipolar radiofrequency [23]. Of course, tailoring accurately the route of the mitral line on the individual coronary anatomy is crucial with bipolar radiofrequency due to its penetration properties. For this reason, a tangible marker such as a needle turned out to be particularly useful. Other marking techniques can be adopted such as dyeing the epicardium of the AV groove with methylene blue or analogues.
Clamping the posterior mitral annulus is routinely feasible provided there is no massive calcification. We did not record any major complication related to AV groove ablation. In particular, the absence of any significant case of bleeding and of any case of postoperative myocardial ischaemia is particularly comforting.
Clinical results of our preliminary series were actually satisfactory with 81% of the patients in stable sinus rhythm 1 year after surgery. The only variable related to a worse AF freedom after ablation was permanent AF, as opposed to persistent AF (p = 0.009). Of note, the study population was composed almost exclusively (all but three) of patients with permanent or persistent AF, since those with paroxysmal AF tended to receive simplified lesion sets. Therefore, the low rate of post ablation left flutter, occurring in only 1/70 patients, is particularly significant. This accounts for the efficacy of the mitral connecting line in our series. In fact, based on our prior experience with bipolar radiofrequency in a comparable series of patients, the absence of the mitral connecting line would have led to a tenfold incidence of postoperative left flutter [4].
Other authors described using a second unipolar device to complete the mitral line [24]. This is, on one hand, not cost-effective especially in Europe where no reusable ablation device is available for cardiac use; on the other hand, ablations performed with any unipolar surgical device are not as reliably transmural as those achieved with bipolar radiofrequency [21]. In past years, we have also used a second unipolar radiofrequency device to complete the lesion set on the mitral annulus in a series of 33 patients undergoing bipolar ablation of the left atrium (control group), basically before devising a way to perform all the ablation with bipolar radiofrequency. By comparison with the study group, no significant difference was found in any clinical endpoint, possibly due, to a certain extent, to the small size of the control group. But interestingly, although not surprisingly, the per patient cost of the ablation devices in the study group was nearly halved (1245 ± 50 \#8364; vs 2403 ± 17 \#8364;; p < 0.0001) thanks to the described technique.
The final result of ablating up to the mitral annulus with bipolar radiofrequency entails circumferential ablation of the coronary sinus. Such ablation is considered necessary to increase AF cure rate and to prevent left flutter [3] and is generally not achievable by ablating endocardially with unipolar devices. Although we did not record any major complication with this technique, despite the near routine use of retrograde cardioplegia, it may be safer to avoid cannulation of the ablated coronary sinus in elderly patients or when tissue frailty is any way suspected.
The presence of a coronary-free area in the AV groove may be difficult to identify in the presence of a left dominant coronary system [23]. Actually in such anatomical context, most frequently the posterior mitral annulus may be safely clamped in the area of the posteromedial commissure. In this case the needle marker can be positioned in a spot corresponding to the terminal branch of the left coronary artery and once the left atrium is opened, the mitral annulus must be clamped carefully medial to the needle projection on the valve.
When mitral replacement is not planned, excessive burning of the mitral valve can be of concern with the described technique since the consequent scarring can possibly lead to retraction of the posterior leaflet. When possible, such burning is positioned in the prolapsing segment of the posterior leaflet, which is then resected. Otherwise, limiting the mitral clamping to annulus is preferable. However, the risk of mitral malfunction due to surgical ablation is reasonably low considering that it has never been reported so far whatever the source of energy. On the other side care must be taken to always involve at least the fibrous structure of the valve annulus since an incomplete connecting line may frequently be the cause of relapsing arrhythmia [25].
Preoperative angiogram is kept in mind when inspecting the epicardium for coronary anatomy. Nevertheless, more frequently than not, the coronary-free area can be safely identified also in patients with no preoperative angiogram. In such patients, when the specific coronary anatomy cannot be clearly identified by epicardial inspection, either (careful) use of a second unipolar device or aborting the mitral line is suggested. A preoperative multi-slice CT scan can possibly obviate the problem.
Adding a right set of lines would have possibly improved the sinus rhythm recovery rate, but bipolar RF ablation of the tricuspid annulus does not appear to be feasible with the described technique and the intercaval line may increase the risk of postoperative sinus node dysfunction (absent in our series).
In conclusion, anatomy based tailoring of the mitral line allows performing a complete left atrial ablation with bipolar radiofrequency. Such a technique is safe and cost-effective and yields satisfactory mid-term freedom from AF and left flutter.
| Acknowledgments |
|---|
| Footnotes |
|---|
Presented at the 21st Annual Meeting of the European Association for Cardio-thoracic Surgery, Geneva, Switzerland, September 16–19, 2007.

Disclosure: Stefano Benussi is consultant for Estech Inc. and for Atricure Inc. and has received lecture fees from St Jude Medical Inc., Medtronic Inc., Cryocath Inc., Edwards-Lifesciences Inc. Valerio Zerbi is supported by a research grant offered by Medtronic Inc. Ottavio Alfieri is co-owner of patent rights related to prosthetic mitral rings commercialised by Edwards-Lifesciences Inc.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Castella Reply to the Editor. J. Thorac. Cardiovasc. Surg., June 1, 2009; 137(6): 1574 - 1575. [Full Text] [PDF] |
||||
![]() |
S. Benussi, D. Corradi, and O. Alfieri Clamping the mitral and tricuspid annuli with bipolar devices. J. Thorac. Cardiovasc. Surg., June 1, 2009; 137(6): 1573 - 1574. [Full Text] [PDF] |
||||
![]() |
S. Benussi, A. Galanti, S. Nascimbene, A. Fumero, E. Dorigo, V. Zerbi, and O. Alfieri Complete right atrial ablation with bipolar radiofrequency. Ann. Thorac. Surg., May 1, 2009; 87(5): 1573 - 1576. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. O. von Oppell, N. Masani, P. O'Callaghan, R. Wheeler, G. Dimitrakakis, and S. Schiffelers Mitral valve surgery plus concomitant atrial fibrillation ablation is superior to mitral valve surgery alone with an intensive rhythm control strategy Eur. J. Cardiothorac. Surg., April 1, 2009; 35(4): 641 - 650. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Wang, X. Meng, H. Li, Y. Cui, J. Han, and C. Xu Prospective randomized comparison of left atrial and biatrial radiofrequency ablation in the treatment of atrial fibrillation Eur. J. Cardiothorac. Surg., January 1, 2009; 35(1): 116 - 122. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |