725
Atrial Arrhythmias Following Surgical AF Ablation: Electrophysiological Findings, Ablation Strategies, and Clinical Outcome YAN HUO, M.D.,∗ ROBERT SCHOENBAUER, M.D.,† SERGIO RICHTER, M.D.,† SASCHA ROLF, M.D.,† PHILIPP SOMMER, M.D.,† ARASH ARYA, M.D.,† ARDAWAN RASTAN, M.D.,‡ NICOLAS DOLL, M.D.,§ FRIEDRICH-WILHELM MOHR, M.D.,¶ GERHARD HINDRICKS, M.D.,† CHRISTOPHER PIORKOWSKI, M.D.,∗ and THOMAS GASPAR, M.D.∗ From the ∗ Department of Electrophysiology, Dresden University—Heart Center, Dresden, Germany; †Department of Electrophysiology, Leipzig University—Heart Center, Leipzig, Germany; ‡Department of Cardiac Surgery, Rotenburg a. d. Fulda, Germany; §Department of Cardiac Surgery, Sana Hospital, Stuttgart, Germany; and ¶Department of Cardiac Surgery, Leipzig University—Heart Center, Leipzig, Germany
Surgical Ablation Related Reentrant Tachycardia. Background: Intraoperative atrial fibrillation (AF) ablation during cardiac surgery is a well-established treatment. However, tachycardia mechanisms, ablation strategies, and long-term follow-up of atrial arrhythmias (AA) following intraoperative AF ablation (AFA) have not been previously studied in a large cohort of patients. Objective: Eighty-two patients (48 male, median age of 65 years) with symptomatic recurrence of AA following intraoperative AFA underwent radiofrequency catheter ablation. Methods: Regular atrial tachycardias (AT) were mapped using 3-dimensional (3D) color-coded entrainment/activation mapping and eliminated by linear ablation. Pulmonary vein (PV)-isolation (PVI) was achieved in patients with left atrium-PV (LAPV) conduction after AT elimination. Results: In 85 (83%) out of a total of 103 regular ATs, the entire reentrant circuits were localized perimitrally (n = 27), around PVs (left PV [LPV] or right PV [RPV]; n = 9), around left atrial appendage (LAA; n = 1), on left-sided atrial septum (n = 8), on atrioventricular nodal area (n = 1), on the posterior wall of LA (n = 1), along roof-septum-inferoposterior wall (n = 8), at coronary sinus ostium (n = 2), upper loop in RA (n = 1), and as cavotricuspid isthmus-dependent reentrant ATs (n = 27). Sixty-five (79%) patients received PVI. Noninducibility of any AT was reached at the end of all procedures. During a median follow-up time of 18 months, 69 patients (87%) were free of AA. Conclusion: Reentrant AT appears in the majority of patients with recurrence of AA following intraoperative AFA. Detailed 3D color-coded entrainment mapping was successfully obtained in the majority of patients suffering from reentrant AT after intraoperative AFA, facilitated the accurate identification of the entire reentrant circuit and selection of optimal ablation lines. (J Cardiovasc Electrophysiol, Vol. 25, pp. 725-738, July 2014) atrial fibrillation, atrial flutter, atrial tachycardia, catheter ablation, surgical ablation Introduction Intraoperative ablation of atrial fibrillation (AF) has been predominantly performed as a concomitant antiarrhythmic procedure.1-3 Recently, however, clinical and scientific interests of both electrophysiologists and cardiac surgeons have led to a revival of this procedure, not only as a concomitant practice, but also as a stand-alone therapy for AF.4-6 Patients with an advanced atrial electrical disease and those with mulY. Huo and R. Schoenbauer contributed equally. Other authors: No disclosures. Address for correspondence: Christopher Piorkowski, M.D., Department of Electrophysiology Dresden University—Heart Center, 01307 Dresden, Germany. Fax: +49-351-450-1902; E-mail: [email protected] Manuscript received 6 October 2013; Revised manuscript received 13 February 2014; Accepted for publication 17 February 2014. doi: 10.1111/jce.12406
tiple unsuccessful catheter-based interventions are thought to be potential candidates for antiarrhythmic surgery.4 While the optimal lesion pattern still remains a matter of debate, comparative studies have revealed the importance of left atrial ablations, including pulmonary vein (PV) isolation (PVI).2 It is important to note, however, that such operations can also suffer from failures and arrhythmia recurrences. Recurrence rate of atrial arrhythmias (AA) following intraoperative AF ablation (AFA) is highly dependent on patient selection and the source of energy used to perform the linear lesions. Given the current trends, it is expected that clinical electrophysiologists will be seeing more patients with AA following surgical ablation in the foreseeable future. Management of such patients requires a better understanding of the recurring arrhythmia’s underlying mechanism, the available treatment options, and the expected clinical outcomes after surgery.7-11 We aim to supplement the existing scientific evidence by describing electrophysiological findings, arrhythmia mechanisms, catheter ablation strategies, and clinical outcomes of AA following intraoperative AFA in a large cohort of patients.
726
Journal of Cardiovascular Electrophysiology
Vol. 25, No. 7, July 2014
Methods Patient Population We retrospectively analyzed 82 consecutive patients (48 male, median age 65 years, range 33–79) who underwent an electrophysiological study and radiofrequency (RF) catheter ablation (RFCA) as treatment for symptomatic recurrences of AA following intraoperative AFA. The patients were included from June 2006 to April 2012 at the Heart Center Leipzig in Leipzig, Germany. Twenty-two patients (27%) had a recurrence of AF, 52 patients (63%) had a recurrence of regular AT, and 8 patients (10%) had a recurrence of both AA, which were documented during follow-up after an intraoperative AFA procedure. Fifty-eight (71%) out of 82 patients had received a mitral valve repair (MVR, mitral valve replacement or reconstruction, including mechanical mitral valve replacement); 14 (17%) patients had received a tricuspid valve repair (TVR, tricuspid valve replacement/reconstruction); 13 (16%) patients had received an aortic valve replacement (AVR); 10 (12%) patients had been diagnosed with coronary artery disease, 5 of whom had also received a coronary artery bypass graft (CABG); 5 (6%) patients had dilated cardiomyopathy (DCM); 1 (1%) patient had received a ventricular septal defect (VSD) closure device; 2 (2%) patients had received an atrial septal defect (ASD) closure device; and 51 (62%) patients had a history of arterial hypertension. Fifteen (18%) patients had received an intraoperative ablation exclusively for a primary indication of symptomatic AF. Left ventricular ejection fraction (LVEF) was 57 ± 11% and left atrial diameter was 47 ± 7 mm (Table 1; Fig. 1) for all patients. Previous Intraoperative Ablation Procedure The intraoperative AFA procedures were performed as either RF ablation in 17 patients (21%) or cryoablation in 65 patients (79%). The surgical approach and the ablation technique have been published previously.7,12,13 Briefly, all patients were operated on using a video-assisted minimally invasive surgical approach via a right anterolateral minithoracotomy. The left atrium (LA) was incised parallel to the interatrial groove anterior to the right PVs, allowing access to the mitral annulus and exposure of the PV orifices. In the case of RF ablation, alternating current (500 kHz, modified HAT 200S; Osypka GmbH, Grenzach-Wyhlen, Germany) was delivered in a unipolar mode between the 10mm T-shaped tip electrode (Osypka GmbH) and a 10 × 16 cm external back-plate electrode. The 1st lesion line extended from the left inferior aspect of the mitral annulus to the left PVs. The second line connected the left inferior PV (LIPV) and superior PV (LSPV) orifices. A third line coupled the LSPV and right superior PV (RSPV). Then the RSPV and right inferior PV (RIPV) orifices were connected. Finally, the line at the left atrial roof was connected to the surgical incision to prevent “incisional reentry” (Fig. 2A). In the case of cryoablation, an argon gas refrigerant (SurgiFrost CryoCath, Irvine, CA, USA) was used to create the planned lesion set. A continuous lesion line was created extending from the inferior aspect of the posterior mitral leaflet to the LIPV. Separate lesion lines were created to iso-
TABLE 1 Clinical Characteristics in Patients with Intraoperative Ablation Due to AF Intraoperative Cryoablation (n = 65)
Intraoperative RFAblation (n = 17)
P Value
Age (year) Male LV-EF (%) LA-diameter (mm) LV-diameter (mm) Hypertension Dilated cardiomyopathy Coronary artery disease Diabetes mellitus Chronic heart failure Mitral valve repair Tricuspid valve repair Aortic valve replacement Coronary artery bypass graft Ventricular septal defect closure Atrial septal defect closure
65 (33–79) 37 (57%) 56 (23–81) 48 (35–80) 51 (37–68) 41 (63%) 4 (6%)
62 (37–79) 11 (65%) 65 (34–67) 44 (34–59) 51 (42–66) 11 (65%) 1 (6%)
0.227 0.564 0.034* 0.037* 0.718 0.961 0.956
8 (12%)
2 (12%)
0.935
8 (12%) 10 (15%) 53 (82%) 14 (22%) 12 (18%)
4 (24%) 3 (18%) 5 (29%) 0 (0%) 1 (6%)
0.258 0.841 0.001** 0.559 0.983
4 (6%)
1 (6%)
1.000
1 (2%)
0 (0%)
1.000
2 (3%)
0 (0%)
1.000
Documented AF during follow-up after intraoperative ablation Documented regular AT during follow-up after intraoperative ablation Documented AT/AF during follow-up after intraoperative ablation Period from intraoperative ablation to 1st recurrence (months)
11 (17%)
11 (65%)
0.001**
47 (72%)
5 (29%)
0.001**
7 (11%)
1 (6%)
0.001**
10 (0–85)
72 (6–147)
Median
15% indicated a focal mechanism.23,24 In ATs with significant cycle length variations, tachycardia overdrive pacing was performed to identify the site with the shortest PPI that was considered only to be in a close proximity to the AT focus.17 (2) Stable AT cycle lengths and consistent return cycle lengths in response to overdrive pacing at different pacing sites were indicative of a reentrant mechanism.22 (3) The presence of a significant
isoelectric line in between the P-waves on all 12 ECG leads indicated a focal origin.25,26 Linear Ablation, Validation of Ablation Lines, and Endpoint of Procedure In Patients with Initial Regular AT or Induced Regular AT During Procedure In patients with a macroreentrant circuit, optimal ablation lines were designed and classified into 4 types as follows: – Roof line (RL) between LSPV and RSPV on the roof of LA and posterior line (PL) between LIPV und RIPV on the posterior wall of LA: Complete linear block was validated by (i) demonstration of a corridor of doublepotentials along the ablation line during pacing of the
Huo et al. Surgical Ablation Related Reentrant Tachycardia
anterior LA. Anterior LA pacing could be achieved by pacing the high septal RA when activation proceeded over the anterior interatrial connections (i.e., Bachmann’s bundle) to the anterior-septal LA. (ii) Activation mapping to demonstrate an activation detour either during high-septal right atrial pacing when the latest LA activation was the posterior LA behind the RL (or beneath PL) or during pacing the posterior LA behind the RL (or beneath PL) when latest activation was the highseptal RA (among at least 2 pacing sites, proximal CS and high-septal RA). – Mitral isthmus line (MIL) between mitral annulus (3–5 o’clock LAO) to LIPV on the lateral wall of LA, septal line (SL) from mitral annulus (12 o’clock LAO) to RSPV on the septum of LA and anterior line (AL) from mitral annulus (12 o’clock LAO) to LSPV on the anterior wall of LA close to the edge of LAA: Complete linear block was validated by (i) demonstrating a corridor of double-potentials along the line during pacing lateral to the line, (ii) differential pacing techniques by pacing lateral to the line, shifting the pacing site from one side of the line to the other, and (iii) activation mapping to demonstrate an activation detour either during pacing the LA lateral to the ablation line (distal bipoles of CS catheter) or during high-septal right atrial pacing when the earliest LA activation was anteroseptal. – Cavotricuspid isthmus ablation: performed with an endpoint of bidirectional conduction block.27 – Multiple ablation lines: validation of ablation lines was achieved using the techniques described above and adding the “pace and ablate” technique,28 which offered a solid endpoint for linear ablation, such as box-isolation as a combination of RL and PL, or roof-anterior wall isolation as a combination of RL, SL, and AL.
729
not inducible during the electrophysiological procedure, validation of bidirectional block at the level of PV entrance was performed in SR and—if necessary—was achieved by catheter ablation. Subsequently, a voltage map was carried out to evaluate the previous surgical ablation lines and to localize untreated substrate defined as areas with peak-to-peak voltage amplitude
Atrial Arrhythmias Following Surgical AF Ablation: Electrophysiological Findings, Ablation Strategies, and Clinical Outcome YAN HUO, M.D.,∗ ROBERT SCHOENBAUER, M.D.,† SERGIO RICHTER, M.D.,† SASCHA ROLF, M.D.,† PHILIPP SOMMER, M.D.,† ARASH ARYA, M.D.,† ARDAWAN RASTAN, M.D.,‡ NICOLAS DOLL, M.D.,§ FRIEDRICH-WILHELM MOHR, M.D.,¶ GERHARD HINDRICKS, M.D.,† CHRISTOPHER PIORKOWSKI, M.D.,∗ and THOMAS GASPAR, M.D.∗ From the ∗ Department of Electrophysiology, Dresden University—Heart Center, Dresden, Germany; †Department of Electrophysiology, Leipzig University—Heart Center, Leipzig, Germany; ‡Department of Cardiac Surgery, Rotenburg a. d. Fulda, Germany; §Department of Cardiac Surgery, Sana Hospital, Stuttgart, Germany; and ¶Department of Cardiac Surgery, Leipzig University—Heart Center, Leipzig, Germany
Surgical Ablation Related Reentrant Tachycardia. Background: Intraoperative atrial fibrillation (AF) ablation during cardiac surgery is a well-established treatment. However, tachycardia mechanisms, ablation strategies, and long-term follow-up of atrial arrhythmias (AA) following intraoperative AF ablation (AFA) have not been previously studied in a large cohort of patients. Objective: Eighty-two patients (48 male, median age of 65 years) with symptomatic recurrence of AA following intraoperative AFA underwent radiofrequency catheter ablation. Methods: Regular atrial tachycardias (AT) were mapped using 3-dimensional (3D) color-coded entrainment/activation mapping and eliminated by linear ablation. Pulmonary vein (PV)-isolation (PVI) was achieved in patients with left atrium-PV (LAPV) conduction after AT elimination. Results: In 85 (83%) out of a total of 103 regular ATs, the entire reentrant circuits were localized perimitrally (n = 27), around PVs (left PV [LPV] or right PV [RPV]; n = 9), around left atrial appendage (LAA; n = 1), on left-sided atrial septum (n = 8), on atrioventricular nodal area (n = 1), on the posterior wall of LA (n = 1), along roof-septum-inferoposterior wall (n = 8), at coronary sinus ostium (n = 2), upper loop in RA (n = 1), and as cavotricuspid isthmus-dependent reentrant ATs (n = 27). Sixty-five (79%) patients received PVI. Noninducibility of any AT was reached at the end of all procedures. During a median follow-up time of 18 months, 69 patients (87%) were free of AA. Conclusion: Reentrant AT appears in the majority of patients with recurrence of AA following intraoperative AFA. Detailed 3D color-coded entrainment mapping was successfully obtained in the majority of patients suffering from reentrant AT after intraoperative AFA, facilitated the accurate identification of the entire reentrant circuit and selection of optimal ablation lines. (J Cardiovasc Electrophysiol, Vol. 25, pp. 725-738, July 2014) atrial fibrillation, atrial flutter, atrial tachycardia, catheter ablation, surgical ablation Introduction Intraoperative ablation of atrial fibrillation (AF) has been predominantly performed as a concomitant antiarrhythmic procedure.1-3 Recently, however, clinical and scientific interests of both electrophysiologists and cardiac surgeons have led to a revival of this procedure, not only as a concomitant practice, but also as a stand-alone therapy for AF.4-6 Patients with an advanced atrial electrical disease and those with mulY. Huo and R. Schoenbauer contributed equally. Other authors: No disclosures. Address for correspondence: Christopher Piorkowski, M.D., Department of Electrophysiology Dresden University—Heart Center, 01307 Dresden, Germany. Fax: +49-351-450-1902; E-mail: [email protected] Manuscript received 6 October 2013; Revised manuscript received 13 February 2014; Accepted for publication 17 February 2014. doi: 10.1111/jce.12406
tiple unsuccessful catheter-based interventions are thought to be potential candidates for antiarrhythmic surgery.4 While the optimal lesion pattern still remains a matter of debate, comparative studies have revealed the importance of left atrial ablations, including pulmonary vein (PV) isolation (PVI).2 It is important to note, however, that such operations can also suffer from failures and arrhythmia recurrences. Recurrence rate of atrial arrhythmias (AA) following intraoperative AF ablation (AFA) is highly dependent on patient selection and the source of energy used to perform the linear lesions. Given the current trends, it is expected that clinical electrophysiologists will be seeing more patients with AA following surgical ablation in the foreseeable future. Management of such patients requires a better understanding of the recurring arrhythmia’s underlying mechanism, the available treatment options, and the expected clinical outcomes after surgery.7-11 We aim to supplement the existing scientific evidence by describing electrophysiological findings, arrhythmia mechanisms, catheter ablation strategies, and clinical outcomes of AA following intraoperative AFA in a large cohort of patients.
726
Journal of Cardiovascular Electrophysiology
Vol. 25, No. 7, July 2014
Methods Patient Population We retrospectively analyzed 82 consecutive patients (48 male, median age 65 years, range 33–79) who underwent an electrophysiological study and radiofrequency (RF) catheter ablation (RFCA) as treatment for symptomatic recurrences of AA following intraoperative AFA. The patients were included from June 2006 to April 2012 at the Heart Center Leipzig in Leipzig, Germany. Twenty-two patients (27%) had a recurrence of AF, 52 patients (63%) had a recurrence of regular AT, and 8 patients (10%) had a recurrence of both AA, which were documented during follow-up after an intraoperative AFA procedure. Fifty-eight (71%) out of 82 patients had received a mitral valve repair (MVR, mitral valve replacement or reconstruction, including mechanical mitral valve replacement); 14 (17%) patients had received a tricuspid valve repair (TVR, tricuspid valve replacement/reconstruction); 13 (16%) patients had received an aortic valve replacement (AVR); 10 (12%) patients had been diagnosed with coronary artery disease, 5 of whom had also received a coronary artery bypass graft (CABG); 5 (6%) patients had dilated cardiomyopathy (DCM); 1 (1%) patient had received a ventricular septal defect (VSD) closure device; 2 (2%) patients had received an atrial septal defect (ASD) closure device; and 51 (62%) patients had a history of arterial hypertension. Fifteen (18%) patients had received an intraoperative ablation exclusively for a primary indication of symptomatic AF. Left ventricular ejection fraction (LVEF) was 57 ± 11% and left atrial diameter was 47 ± 7 mm (Table 1; Fig. 1) for all patients. Previous Intraoperative Ablation Procedure The intraoperative AFA procedures were performed as either RF ablation in 17 patients (21%) or cryoablation in 65 patients (79%). The surgical approach and the ablation technique have been published previously.7,12,13 Briefly, all patients were operated on using a video-assisted minimally invasive surgical approach via a right anterolateral minithoracotomy. The left atrium (LA) was incised parallel to the interatrial groove anterior to the right PVs, allowing access to the mitral annulus and exposure of the PV orifices. In the case of RF ablation, alternating current (500 kHz, modified HAT 200S; Osypka GmbH, Grenzach-Wyhlen, Germany) was delivered in a unipolar mode between the 10mm T-shaped tip electrode (Osypka GmbH) and a 10 × 16 cm external back-plate electrode. The 1st lesion line extended from the left inferior aspect of the mitral annulus to the left PVs. The second line connected the left inferior PV (LIPV) and superior PV (LSPV) orifices. A third line coupled the LSPV and right superior PV (RSPV). Then the RSPV and right inferior PV (RIPV) orifices were connected. Finally, the line at the left atrial roof was connected to the surgical incision to prevent “incisional reentry” (Fig. 2A). In the case of cryoablation, an argon gas refrigerant (SurgiFrost CryoCath, Irvine, CA, USA) was used to create the planned lesion set. A continuous lesion line was created extending from the inferior aspect of the posterior mitral leaflet to the LIPV. Separate lesion lines were created to iso-
TABLE 1 Clinical Characteristics in Patients with Intraoperative Ablation Due to AF Intraoperative Cryoablation (n = 65)
Intraoperative RFAblation (n = 17)
P Value
Age (year) Male LV-EF (%) LA-diameter (mm) LV-diameter (mm) Hypertension Dilated cardiomyopathy Coronary artery disease Diabetes mellitus Chronic heart failure Mitral valve repair Tricuspid valve repair Aortic valve replacement Coronary artery bypass graft Ventricular septal defect closure Atrial septal defect closure
65 (33–79) 37 (57%) 56 (23–81) 48 (35–80) 51 (37–68) 41 (63%) 4 (6%)
62 (37–79) 11 (65%) 65 (34–67) 44 (34–59) 51 (42–66) 11 (65%) 1 (6%)
0.227 0.564 0.034* 0.037* 0.718 0.961 0.956
8 (12%)
2 (12%)
0.935
8 (12%) 10 (15%) 53 (82%) 14 (22%) 12 (18%)
4 (24%) 3 (18%) 5 (29%) 0 (0%) 1 (6%)
0.258 0.841 0.001** 0.559 0.983
4 (6%)
1 (6%)
1.000
1 (2%)
0 (0%)
1.000
2 (3%)
0 (0%)
1.000
Documented AF during follow-up after intraoperative ablation Documented regular AT during follow-up after intraoperative ablation Documented AT/AF during follow-up after intraoperative ablation Period from intraoperative ablation to 1st recurrence (months)
11 (17%)
11 (65%)
0.001**
47 (72%)
5 (29%)
0.001**
7 (11%)
1 (6%)
0.001**
10 (0–85)
72 (6–147)
Median
15% indicated a focal mechanism.23,24 In ATs with significant cycle length variations, tachycardia overdrive pacing was performed to identify the site with the shortest PPI that was considered only to be in a close proximity to the AT focus.17 (2) Stable AT cycle lengths and consistent return cycle lengths in response to overdrive pacing at different pacing sites were indicative of a reentrant mechanism.22 (3) The presence of a significant
isoelectric line in between the P-waves on all 12 ECG leads indicated a focal origin.25,26 Linear Ablation, Validation of Ablation Lines, and Endpoint of Procedure In Patients with Initial Regular AT or Induced Regular AT During Procedure In patients with a macroreentrant circuit, optimal ablation lines were designed and classified into 4 types as follows: – Roof line (RL) between LSPV and RSPV on the roof of LA and posterior line (PL) between LIPV und RIPV on the posterior wall of LA: Complete linear block was validated by (i) demonstration of a corridor of doublepotentials along the ablation line during pacing of the
Huo et al. Surgical Ablation Related Reentrant Tachycardia
anterior LA. Anterior LA pacing could be achieved by pacing the high septal RA when activation proceeded over the anterior interatrial connections (i.e., Bachmann’s bundle) to the anterior-septal LA. (ii) Activation mapping to demonstrate an activation detour either during high-septal right atrial pacing when the latest LA activation was the posterior LA behind the RL (or beneath PL) or during pacing the posterior LA behind the RL (or beneath PL) when latest activation was the highseptal RA (among at least 2 pacing sites, proximal CS and high-septal RA). – Mitral isthmus line (MIL) between mitral annulus (3–5 o’clock LAO) to LIPV on the lateral wall of LA, septal line (SL) from mitral annulus (12 o’clock LAO) to RSPV on the septum of LA and anterior line (AL) from mitral annulus (12 o’clock LAO) to LSPV on the anterior wall of LA close to the edge of LAA: Complete linear block was validated by (i) demonstrating a corridor of double-potentials along the line during pacing lateral to the line, (ii) differential pacing techniques by pacing lateral to the line, shifting the pacing site from one side of the line to the other, and (iii) activation mapping to demonstrate an activation detour either during pacing the LA lateral to the ablation line (distal bipoles of CS catheter) or during high-septal right atrial pacing when the earliest LA activation was anteroseptal. – Cavotricuspid isthmus ablation: performed with an endpoint of bidirectional conduction block.27 – Multiple ablation lines: validation of ablation lines was achieved using the techniques described above and adding the “pace and ablate” technique,28 which offered a solid endpoint for linear ablation, such as box-isolation as a combination of RL and PL, or roof-anterior wall isolation as a combination of RL, SL, and AL.
729
not inducible during the electrophysiological procedure, validation of bidirectional block at the level of PV entrance was performed in SR and—if necessary—was achieved by catheter ablation. Subsequently, a voltage map was carried out to evaluate the previous surgical ablation lines and to localize untreated substrate defined as areas with peak-to-peak voltage amplitude
