Abstract
Significant recurrence of atrial tachyarrhythmias are observed after the surgical Cox Maze procedure (CMP). We retrospectively enrolled 11 consecutive patients who had atrial tachyarrhythmias (ATAs) that recurred after a biatrial CMP and underwent catheter ablation. Information including the site of any incomplete lesions and the etiology of the clinical ATAs was shared with the surgical team as feedback. In a total of 11 patients, 12 clinical ATAs were identified. They consisted of 2 atrial fibrillations and 10 atrial tachycardias (ATs). In 6 patients, the CMP was performed after the beginning of this investigation. In a total of 10 ATs, we diagnosed 5 mitral annular flutters, 2 roof-dependent flutters, 1 pulmonary vein (PV)-reentrant AT, and 1 localized reentrant AT. A total of 6 patients had reconnected perimitral block lines. PV reconnections were observed in 3 and posterior wall (PW) residual conduction was also observed in 3 cases. However, no residual conduction of the pulmonary vein isolation (PVI) and only 1 residual conduction of the PW were observed in 5 patients who underwent their index surgery after the beginning of this investigation. This fact may implicate that sharing the information from the electrophysiological study of postsurgical ATAs with the surgical team may contribute to the refinement of the CMP in each facility.
<Learning objective: Reconduction of the surgical lesion is the major etiology of recurrence of atrial tachyarrhythmias after the surgical Cox Maze procedure. Although perimitral block line seemed to be the most frequent reconduction site, our study suggested that durable lesion of the PVI and the PW isolation could be achieved by improving surgical techniques by feedback from the electrophysiological team to the surgical team.>
Keywords: Atrial fibrillation, Atrial tachycardia, Catheter ablation, Maze procedure, Pulmonary vein isolation
Introduction
The surgical Cox Maze procedure (CMP) was the first therapeutic intervention for atrial fibrillation (AF) with a modification of the atrial conduction and established the prophylaxis of AF [1], [2], [3]. The contemporary Cox Maze IV (CMP-IV) utilizes a combination of incisional scars and ablation lesions created by radiofrequency or cryothermal energy [2]. The CMP-IV reduced the complexity and procedural time as compared to the Cox Maze III (CMP-III) using a “cut and sew” technique [2], [3]. On the other hand, the ablation lesions are not as reliable as incisional scar even if created under direct visualization, which causes a common problem of percutaneous catheter ablation: recurrence of atrial tachyarrhythmias (ATAs) due to reconnections of block lines [4]. In the present report, the residual conduction of surgical block lines in the left atria (LA) was investigated in patients who underwent catheter ablation for relapsed ATAs after the CMP.
Case report
We retrospectively enrolled 11 consecutive patients who had ATAs that recurred after a biatrial CMP and underwent catheter ablation from April 2011 to July 2015 in our institution. The ATAs included AF and atrial tachycardias (ATs). All patients gave their written consent prior to the procedure. The present study was approved by the domestic ethical committee.
Surgical procedure
From 1998 to 2013, all patients underwent a CMP-III/IV concomitantly with the index operation for the primary disease. CMP-III was performed as previously reported [5]. The technique we adopted for the CMP-IV during the period of this study entailed the use of a radiofrequency ablation device (AtriCure, Inc., Mason, OH, USA) and cryoablation device (142 Cryo Unit, Spembly Medical, Andover, UK). All patients underwent a bilateral pulmonary vein isolation (PVI) and ablation of the right lesion set on the beating heart using a bipolar RF ablation clamp; additional ablation using a unipolar device (cryoablation or RF ablation) was performed where necessary to complete the ablation lines. The left atrial lesion set, which was approached from the incised right pulmonary vein and through the amputated left atrial appendage, was ablated similarly but under an aortic cross clamp. The left and right pulmonary vein lesions were connected to create a box lesion on the left atrium. A linear lesion for the perimitral block line was made by a bipolar RF ablation from the right side of the left atrium to the posterior portion of the mitral annulus. An endocardial ablation was then performed for the mitral annular lesion for 120 s. Additional epicardial ablation using monopolar RF was performed to connect this ablation line to the coronary sinus at the end of the mitral isthmus lesion. We made an interatrial septal line. Other procedural details of our Maze procedure were similar to those described by Cox et al. [6], [7]. None of the patients required a deviation or abortion of the procedure in the present study.
Ablation procedure
All antiarrhythmic drugs were discontinued for five half-lives prior to the procedure. All electrocardiograms were recorded and stored on a workstation lab (RMS-4000/RMS-5000, NIHON KOHDEN, Tokyo, Japan). Local intracardiac potentials with location tags were also stored and analyzed on a non-fluoroscopic three-dimensional mapping system (CARTO XP/CARTO3, Biosense-Webster, Diamond Bar, CA, USA). The mapping and RF ablation catheters were all inserted from the femoral approach. An irrigated-tip ablation catheter (Thermocool, Biosense-Webster) was used in all cases. Conduction block of the PVI, roof line, bottom line, and perimitral line were examined before starting ablation if possible. The endpoint was to complete all block lines and the non-inducibility of any ATAs under atrial burst pacing with minimum cycle length of 200 ms. After an electrophysiological study, information including the site of any incomplete lesions and the etiology of the clinical ATAs was shared with the surgical team as feedback.
Results
The patient characteristics and surgical information are shown in Table 1. In a total of 11 patients enrolled, 12 clinical ATAs were identified. They consisted of 2 AFs and 10 ATs. The index CMPs were 10 CMP-IVs and 1 CMP-III. The ablation procedures were performed 40.4 ± 51.4 months after the index CMP. During the CMP-IVs, a combination of a linear RF device and pen-type cryothermal device was used in 9 cases and a cryothermal device was exclusively used in Case 7. In 6 patients (Cases 4, 5, 8–11), the CMP was performed after the beginning of this investigation in 2011.
Table 1.
Patient characteristics.
| Case | Sex | Age | Primary disease | AF type | Operation | LAD, mm | Maze method | Postoperative months |
|---|---|---|---|---|---|---|---|---|
| 1 | F | 60 | MR | Paroxysmal | MVR | 51 | Cox Ⅲ | 186 |
| 2 | F | 74 | MR | Paroxysmal | MVR | 63 | Cox IV | 6 |
| 3 | M | 64 | MR | Persistent | MVP | 44 | Cox IV | 48 |
| 4 | M | 51 | ASD | Persistent | ASDC | 56 | Cox IV | 60 |
| 5 | M | 69 | IHD | Persistent | CAB | 43 | Cox IV | 4 |
| 6 | F | 81 | MR | Persistent | MVP | 50 | Cox IV | 25 |
| 7 | M | 78 | IHD | Paroxysmal | CAB | 48 | Cox IV | 30 |
| 8 | M | 78 | MR | Persistent | MVP | 45 | Cox IV | 31 |
| 9 | M | 74 | MR | Persistent | MVR | 51 | Cox IV | 6 |
| 10 | M | 64 | MR | Persistent | MVP | 42 | Cox IV | 17 |
| 11 | F | 77 | AR, ischemic | Persistent | AVR/CAB | 52 | Cox IV | 31 |
AF, atrial fibrillation; AR, aortic regurgitation; ASDC, atrial septal defect closure; AVR, aortic valve replacement; CAB, coronary artery bypass; IHD, ischemic heart disease; LAD, left atrial diameter; MR, mitral regurgitation; MVP, mitral valvuloplasty; MVR, mitral valve replacement.
The diagnosis and characteristics of the clinical ATAs are shown in Table 2. We had no arrhythmias originating from the right atria. In the two AF cases, residual conduction of the PVI was observed and it was successfully re-isolated. In a total of 10 ATs, 9 ATs were on-going at the beginning of the ablation procedure. One AT of Case 4 was provoked after eliminating another AT. The 9 ATs except Case 5 were successfully mapped to elucidate the detailed activation. We diagnosed 5 mitral annular flutters (MAFs), 2 roof-dependent flutters (RDFs), 1 PV-reentrant AT, and 1 localized reentrant AT arising from the interatrial septum. We found the perimitral block line was the most dominant site of the residual conduction in this investigation. A total of 6 patients had reconnected perimitral block lines, while 5 exhibited MAFs as shown in Fig. 1. They were successfully treated by radiofrequency catheter ablation, however, attempts to complete the native surgical block line all failed and a novel block line was needed. We experienced no recurrences of ATAs except for in Case 5. PV reconnections were observed in 3 including 2 AF cases. As for Case 7, multiple gaps of the left PVI resulted in a PV-reentrant tachycardia. There was a PV reconnection even in Case 1 of a CMP-III. Posterior wall (PW) residual conduction was also observed in 3 cases in which 2 exhibited RDFs. No residual conduction of the PVI and only 1 residual conduction of the PW were observed in 5 patients who underwent their index surgery after the beginning of this investigation. There were no complications, including cardiac tamponade, cerebral infarctions, or bradycardia requiring a pacing device throughout the investigation.
Table 2.
Diagnosis of ATAs and reconnection sites.
| Case | Clinical ATA | TCL | Diagnosis | LA Reconnection |
FU(months) | Rec | ||
|---|---|---|---|---|---|---|---|---|
| PV | PW | Perimitral | ||||||
| 1 | PAF | N/A | PAF | RPV | − | − | 41 | − |
| 2 | PAF | N/A | PAF | LPV | − | − | 46 | − |
| 3 | Persistent AT | 243 | RDF | − | ＋ | − | 31 | − |
| 4* | Persistent AT | 264/250 | RDF/MAF | − | ＋ | ＋ | 20 | − |
| 5 | Persistent AT | 376 | Septal | − | − | − | 13 | − |
| 6 | Persistent AT | 250 | Septal reentry | − | − | ＋ | 12 | + |
| 7 | Persistent AT | 293 | PV reentry | L/RPV | ＋ | − | 21 | − |
| 8* | Persistent AT | 258 | MAF | − | − | ＋ | 21 | − |
| 9* | Persistent AT | 300 | MAF | − | − | ＋ | 12 | − |
| 10* | Persistent AT | 204 | MAF | − | − | ＋ | 14 | − |
| 11* | Persistent AT | 239 | MAF | − | − | ＋ | 6 | − |
AT, atrial tachycardia; ATA, atrial tachyarrhythmia; FU, period of follow-up; LA, left atrium; LPV, left pulmonary vein; MAF, mitral annular flutter; PAF, paroxysmal atrial fibrillation; PV, pulmonary vein; PW, posterior wall; RDF, roof-dependent flutter; Rec, recurrence; RPV, right pulmonary vein; TCL, tachycardia cycle length.
Indicating cases who underwent their index surgery after the start of the present investigation.
Fig. 1.
A representative case (Case 8) of clockwise mitral annular flutter. The color indicates local activation time. On the right panel, scattered scar area below right pulmonary veins might have reflected the perimitral block line and posterior wall isolation during index Maze procedure. The new block line was created at the posterolateral aspect of the left atrium.
AP, anteroposterior; PA, posteroanterior.
Discussion
In the present study, the perimitral block line was the most frequent reconduction site in patients with a prior CMP. Although it was uncertain whether the lesions reconducted after the CMPs or were initially created, these facts were identical to the results of previous reports [4], [8], [9]. Creating a transmural lesion in thick annular tissue should be the cornerstone during the CMP procedure. Indeed, residual conduction of perimitral block line seemed more dominant even after starting this investigation (because Cases 1 and 2 had no reconnection of perimitral line, surgical procedure Case 4 was not performed about the feedback). This might have been due to 1) spontaneously increased number of CMP after starting the present study 2) that repeated ablation around PV reduced time for perimitral block. As for PVs and PW, incomplete lesions could be considered mainly due to an inadequate application of the ablation energy. We found that in the cases that received a CMP-IV after feedback from the EP-team had no residual conduction of the PVs. It might have been due to an increased application of a sandwich-type ablation device during the CMP. In this regard, sharing the information from the EP study of postsurgical ATAs with the surgical team may contribute to the refinement of the CMP in each facility. However, because the initial 2 cases had only reconnection of PVs, CMP of Case 4 was not performed after the feedback about residual conduction of perimitral line. No postoperative ATs originating from the right atrium were observed during the study period at our facility. This fact was incompatible with the previous larger investigations reporting significant right atrial tachycardia [4], [8]. One of the reasons might be that a surgical LA access was routinely established by a right lateral incision, not by a superior transseptal approach at our facility.
Conclusion
We experienced 11 cases of post-CMP ATAs. Mitral annular lesions were the dominant residual conduction site as well as the substrate for clinical ATAs. Residual conduction of the linear lesion might be decreased by feedback from the EP-team to the surgical team.
Funding
None.
Conflict of interest
All authors declare no conflict of interest related to this study.
Acknowledgments
We thank Mr John Martin for his linguistic assistance in the preparation of this manuscript.
References
- 1.Cox J.L., Boineau J.P., Schuessler R.B., Ferguson T.B., Jr., Cain M.E., Lindsay B.D. Successful surgical treatment of atrial fibrillation. Review and clinical update. JAMA. 1991;266:1976–1980. [PubMed] [Google Scholar]
- 2.Weimar T., Schena S., Bailey M.S., Maniar H.S., Schuessler R.B., Cox J.L. The cox-maze procedure for lone atrial fibrillation: a single-center experience over 2 decades. Circ Arrhythm Electrophysiol. 2012;5:8–14. doi: 10.1161/CIRCEP.111.963819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Barnett S.D., Ad N. Surgical ablation as treatment for the elimination of atrial fibrillation: a meta-analysis. J Thorac Cardiovasc Surg. 2006;131:1029–1035. doi: 10.1016/j.jtcvs.2005.10.020. [DOI] [PubMed] [Google Scholar]
- 4.Wazni O.M., Saliba W., Fahmy T., Lakkireddy D., Thal S., Kanj M. Atrial arrhythmias after surgical maze: findings during catheter ablation. J Am Coll Cardiol. 2006;48:1405–1409. doi: 10.1016/j.jacc.2006.05.061. [DOI] [PubMed] [Google Scholar]
- 5.Cox J.L., Jaquiss R.D., Schuessler R.B., Boineau J.P. Modification of the maze procedure for atrial flutter and atrial fibrillation. II. Surgical technique of the maze III procedure. Thorac Cardiovasc Surg. 1995;110:485–495. doi: 10.1016/S0022-5223(95)70245-8. [DOI] [PubMed] [Google Scholar]
- 6.Cox J.L. A brief overview of surgery for atrial fibrillation. Ann Cardiothorac Surg. 2014;3:80–88. doi: 10.3978/j.issn.2225-319X.2014.01.05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Damiano R., Gaynor S.L. Atrial fibrillation ablation during mitral valve surgery using the Atricure™ device. Oper Tech Thorac Cardiovasc Surg. 2004;9:24–33. [Google Scholar]
- 8.Trumello C., Pozzoli A., Mazzone P., Nascimbene S., Bignami E., Cireddu M. Electrophysiological findings and long-term outcomes of percutaneous ablation of atrial arrhythmias after surgical ablation for atrial fibrillation. Eur J Cardiothorac Surg. 2016;49:273–280. doi: 10.1093/ejcts/ezv034. [DOI] [PubMed] [Google Scholar]
- 9.Enriquez A., Santangeli P., Zado E.S., Liang J., Castro S., Garcia F.C. Postoperative atrial tachycardias after mitral valve surgery: mechanisms and outcomes of catheter ablation. Heart Rhythm. 2017;14:520–526. doi: 10.1016/j.hrthm.2016.12.002. [DOI] [PubMed] [Google Scholar]