International Journal of Cardiology 176 (2014) 48–54
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Electrophysiological characteristics of left atrial diverticulum in patients with atrial fibrillation: Electrograms, impedance and clinical implications Chen Tan a,⁎, Wei Han b, Xingpeng Liu c, Xuehong Hu a, Jianguo Liu a, Junyu Cui a, Junxia Li a a b c
Department of Cardiology, Beijing Military Region General Hospital of PLA, Beijing, China Department of Radiology, Beijing Military Region General Hospital of PLA, Beijing, China Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
a r t i c l e
i n f o
Article history: Received 30 March 2014 Received in revised form 15 May 2014 Accepted 24 June 2014 Available online 2 July 2014 Keywords: Atrial fibrillation Left atrial diverticulum Catheter ablation Impedance Electroanatomic mapping
a b s t r a c t Background: Left atrial diverticulum (LAD) is not rare in patients with atrial fibrillation (AF). Recent reports focused on its morphology however data on its electrophysiological characteristics are lacking. Our study aims to investigate the electrogram and impedance features of LAD. Methods: This study included 24 patients (mean age, 58.5 ± 10.7 years) with LAD undergoing catheter ablation for AF and 24 gender-and-age-matched individuals without LAD as controls. A bipolar LAD electroanatomic map was acquired in sinus rhythm from all study participants. Points were acquired for diverticulum in the LAD group and for corresponding areas in the control group. Electrogram deflections were counted, bipolar voltage and impedance were measured for each point, and average Δimpedance and highest Δimpedance were calculated. Results: A total of 234 points were collected in the two groups. In the LAD vs. control group, median (Q1, Q3) of electrogram deflections was 6 (5, 7) and 4 (4, 5) (P b 0.0001), respectively, voltage was not significantly different (1.58 ± 0.68 mV vs. 1.28 ± 0.65 mV, P = 0.10), and average Δimpedance was significantly higher in the LAD group (19.5 ± 9.0 Ω vs 3.9 ± 1.7 Ω, P b 0.0001). A cut-off value of 9.5 Ω for Δimpedance predicted LAD with sensitivity, specificity, and positive and negative predictive values of 83.5%, 92.8%, 92.1% and 84.9%, respectively. Conclusions: Electrogram was more fractionated and impedance was higher at LAD than in corresponding areas without LAD, which might help to differentiate LAD during catheter ablation for AF. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atrial fibrillation (AF) is the most common tachyarrhythmia, and catheter ablation is considered a reasonable therapy for paroxysmal AF [1]. Left atrial diverticulum (LAD) is common in AF patients (17% to 41% occurrence) [2–5], with most (74%–80%) [4,5] located at the anterosuperior left atrium in various shapes, namely cystiform, coneshaped, tubiform and irregular. Examination of an autopsied heart [6] revealed that, despite similar endocardial surface, LAD wall was much thinner than adjacent tissue [3] which might increase tamponade risk for catheter manipulation during mapping and complicate radiofrequency application in LAD. Little is known about the
⁎ Corresponding author at: Department of Cardiology, Beijing Military Region General Hospital of PLA, Dongsi Shitiao, Dongcheng District, Beijing 100700, China. Tel.: + 86 010 84008033; fax: +86 010 66721872. E-mail address: [email protected] (C. Tan).
http://dx.doi.org/10.1016/j.ijcard.2014.06.050 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
electrophysiological characteristics of LAD, which should be distinguished from those of abnormal left atrial wall without LAD to avoid excessive ablation in LAD. This study therefore investigates the electrogram and impedance features of LAD relative to those of corresponding areas without LAD in AF patients undergoing catheter radiofrequency ablation.
2. Methods 2.1. Patient population From February 2012 to November 2013, 91 patients with AF underwent multidetector computed tomography (MDCT) pulmonary venography and catheter radiofrequency ablation at our hospital. After excluding patients with congenital and/or valvular heart disease, 24 patients meeting the inclusion criteria (LAD detected by cardiac CT; and presence/restoration of sinus rhythm during ablation) were consecutively assigned to the LAD group while another 24 without LAD, matched with regard to gender and age (±2 years), served as the control group. The study was approved by the institutional review committee of our hospital, and all participants provided informed consent.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54 2.2. MDCT protocol Cardiac CT images were acquired on a 64-slice CT scanner (Discovery CT750 HD, GE Company, USA) using 0.625-mm slice thickness, 350-ms gantry rotation, 120 kV and 500–600 mA. A total of 50–70 mL of non-ionic contrast agent (iopamidol, 370 mg/mL; Bracco Sine Pharmaceutical Corp. Ltd, Shanghai, China) was injected via the antecubital vein followed by 40 mL of saline, both at a flow rate of 5.0 mL/s. Contrast agent tracking was used to synchronize scanning from the thorax entrance to the diaphragm. Acquired data were transferred to a workstation (AW4.5, GE Company, USA) for image processing, interpretation, and three-dimensional rendering generation. Images were reconstructed with 0.625-mm slice thickness by volume-rendered, multi-planar reformation, and maximum intensity projection reconstructions. LAD was defined as a protrusion from the heart cavity to outside the left atrial wall plane. For each diverticulum, orifice and cervix width and length were measured. According to a previous study [3], LAD shape was classified as: cystiform, if LAD had a broad orifice and a domelike cecum (i.e., body length/orifice width ratio b3); cone-shaped, if LAD had a broad orifice and looked like a cone; and tubiform, if LAD appeared as a long but small cavity (length/orifice width N3). 2.3. Diverticulum electrogram mapping The entire process described below was guided by CARTO XP or CARTO 3 (Biosense Webster, Diamond Bar, CA, USA). A bipolar LAD map was acquired in sinus rhythm as carefully as possible during the ablation procedure. If the patient was in atrial fibrillation, cardioversion to sinus rhythm was followed by a waiting period of at least 5 min. A copy of the cardiac computerized tomography taken prior to the procedure was integrated into the electroanatomic mapping system. For both groups, an ablation catheter (Navistar Thermocool, Diamond Bar, CA, Biosense Webster, USA) was used to collect points for electroanatomic mapping. Due to safety and accuracy of the result, all the ablations were avoided in an area including the diverticulum, of which the diameter was at least 10 mm. If a diverticulum was located close to the pulmonary veins, we left it outside the ablation line. Meanwhile, if the diverticulum was located on the left atrial roof, we tried to move the roof ablation line to a bit anterior or posterior to avoid delivering radiofrequency energy in the diverticulum. We collected points with evaluation of tissue contact based on the stable fluoroscopic motion of the catheter and stable electrogram morphology. Three to five points within a 10-mm-diameter area were acquired for the diverticulum according to its size and shape. For the control group, 3–5 points were also collected in a similar area to that in the LAD group. Number of deflections present in each electrogram was determined by manual counting as previously reported [6]. Electrograms with N5 deflections were defined as fragmented electrograms, while those with ≤5 deflections were considered normal electrograms. The percentage of fractionated signals was calculated by dividing the number of fractionated signals by the total number of signals at diverticulum in the LAD group or the corresponding area in the control group. 2.4. Voltage mapping Local voltage of the local electrogram was defined as the amplitude of peak positiveto-peak negative deflections. For each point, bipolar voltage was recorded with the combination of automated algorithms, and manually verified to validate the result. Low voltage was defined by the conventional b0.5 mV cut-off point for the atrium [7]. 2.4.1. Impedance mapping For each point acquired, impedance was recorded during mapping. Concurrently, average impedance of three points around the target area was recorded as the left atrial impedance. Δimpedance was impedance of each point minus the left atrial impedance. The
49
highest Δimpedance of each area was also recorded. Average Δimpedance was calculated by dividing the total Δimpedance by the number of points acquired in the LAD or in the corresponding area in the control group. 2.5. Re-induction or recurrence of atrial fibrillation after catheter ablation All study participants received high-dose isoproterenol (20–30 μg/min) after ablation to locate additional non-pulmonary vein trigger sites not previously present. Rapid pacing in the coronary sinus was also applied to induce atrial tachycardia arrhythmia. During the follow-up period, in patients with recurrence of AF or atrial tachycardia, recovery of potentials in pulmonary veins and association with tachycardia were evaluated first; other mechanisms for the tachycardia were sought thereafter. 2.6. Statistical analysis Continuous data are expressed as mean ± SD or median (25%, 75% interquartile) as appropriate, and categorical variables are presented as percentages. Continuous variables were compared using t-test or Mann–Whitney U test as appropriate, while Chi-square test or Fisher's exact test were used to compare categorical variables. A two-tailed P value of b0.05 indicated statistical significance. Logistic regression was used to explore the relationship between diverticulum presence and clinical features. Multivariate linear regression was used to confirm the association between electrophysiological characteristics and diverticulum presence.
3. Results 3.1. Patient characteristics Baseline characteristics of patients are shown in Table 1. There were no significant differences in age, gender, type of atrial fibrillation, left atrial size, left ventricular end diastolic diameter, left ventricular function and complications between the two groups. Logistic regression analysis showed no association between diverticulum occurrence and patient characteristics. In particular, LAD presence was unrelated to left atrial size and type of atrial fibrillation. 3.2. Diverticulum characteristics A total of 27 LAD were found in 24 of the 91 patients screened. Three patients had multiple diverticula. Twenty LAD (74.1%) were located in the anterosuperior wall, 4 (11.1%) in the septal wall, and only 3 (11.1%) in the inferior wall. LAD presented in three morphologies, including cystiform, cone-shaped and tubiform. 3.3. Diverticulum electrogram characteristics 234 points were collected in the two groups. In the diverticulum group, the most complex signal in diverticulum had 10 deflections (Fig. 1A), the median average deflection of electrograms was 6, and the fractionated signal proportion was 57.1 ± 29.5%. In the control group, the median average deflection of points was 4, and the
Table 1 Patient characteristics in the LAD and control groups.
Number of patients Age Male gender Type of AF Persistent Comorbidities Hypertension Diabetes Ischemia Echocardiography Left atrial size (mm) Left ventricular end diastolic diameter (mm) Left ventricular ejection fraction (%) AF = atrial fibrillation; LAD = left atrial diverticulum.
LAD group
Control group
P value
24 58.5 ± 10.7 17 (70.8%)
24 59.2 ± 10.4 17 (70.8%)
0.81 1
6 (25%)
9 (37.5%)
0.56
12 (50%) 3 (12.5%) 6 (25%)
16 (66.7%) 5 (20.8%) 7 (29.2%)
0.53 0.70 0.98
35.4 ± 3.81 46.1 ± 3.0 62.9 ± 4.68
37.5 ± 5.29 46.4 ± 5.03 64.5 ± 7.46
0.14 0.79 0.32
50
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
percentage of complex signal was 17.6 ± 14.2%, suggesting that most potentials had no more than 5 deflections (Fig. 1B). The electrogram was more fractionated in the diverticulum region than that in corresponding areas without diverticulum in the left atrium (P b 0.0001) (Table 2). Furthermore, multivariate linear regression analysis showed that only diverticulum presence was associated with multiple deflections (P b 0.0001).
Table 2 Comparison of characteristics of electrograms between AF patients with and without LAD.
Median of deflection number Percentage of complex signals (%) Mean voltage (mV)
LAD
Control
P value
6 (5, 7) 57.1 ± 29.5 1.58 ± 0.68
4 (4, 5) 17.6 ± 14.2 1.28 ± 0.65
b0.0001 b0.0001 0.10
3.4. Assessment of voltage The mean voltage in the LAD group (1.58 ± 0.68 mV) was not significantly different from that of the corresponding area in the control group (1.28 ± 0.65 mV, P = 0.10). Multivariate linear regression analysis showed no relationship between voltage and diverticulum presence, however, voltage amplitude correlated with hypertension (P = 0.016).
and one patient did not accept a second ablation. AF recurrence was not associated with diverticulum presence.
3.5. Assessment of impedance
The increased use of ablation in atrial fibrillation treatment has recently fueled recent research mainly focused on incidence and morphology [3,4] but not electrophysiological characteristics of atrial diverticula. In the current study, we found that (1) there were more fractionated signals at diverticulum; (2) the voltage of electrogram at diverticulum was not different from that of the corresponding area without diverticulum; and (3) the impedance of diverticulum was much higher than that of its adjacent area. In this study, we also determined the predictive Δimpedance cut-off of 9.5 Ω for left atrial diverticulum presence.
The highest Δimpedance was 35.3 ± 19.3 Ω in the LAD group, compared with 7.9 ± 3.2 Ω in the control group (P b 0.0001). The maximal Δimpedance in the diverticulum group was 90 Ω; the point was located in the deep site of a tubiform diverticulum (Fig. 2). In addition, impedance of point at the summit of cone-shape or cystiform diverticula was also very high (Figs. 3, 4). The average Δimpedance was also significantly higher than that of the control group (19.5 ± 9.0 Ω vs. 3.9 ± 1.7 Ω, P b 0.0001) (Fig. 5). In multivariate linear regression analysis, Δimpedance was significantly associated with diverticulum presence (P b 0.0001). Cut-off value of Δimpedance was 9.5 Ω by ROC curve statistic. The sensitivity, specificity and positive and negative predictive value of the cut-off impedance was 83.5%, 92.8%, 92.1% and 84.9%, respectively. 3.6. Relationship between diverticulum presence and atrial fibrillation In our study, since diverticulum was intact, we observed that tachycardia was unrelated to diverticulum presence in first ablation. During follow-up, AF recurred in 10 patients in the two groups (5 vs. 5, P = 1): in 6 patients AF recurred because of the recovery of pulmonary vein potentials; 3 patients had triggers in the superior vena cava;
4. Discussion 4.1. Main findings
4.2. Fractionated electrogram in diverticulum In our study, more fractionated signals were recorded in diverticulum and at sites close to the diverticular orifice. One of the possible underlying mechanisms for fractionated electrogram and arrhythmia is the presence of inhomogeneous muscle in diverticulum, which Pachon et al. [9] termed fibrillar myocardium. Fibrillar myocardium, mainly located in the LA roof and septum except for those close to the PV insertion in LA, entails many cell groups that form wave fronts at different phases and manifests as a polymorphic signal in endocardial electrogram; ablation for fractionated signals could help eliminate atrial fibrillation. A second possible mechanism is presented in another
Fig. 1. A: electrograms in a cone-shaped diverticulum in the left atrial anterosuperior wall acquired during electroanatomic mapping. The distance from the orifice of the right superior pulmonary vein to the diverticulum is 12.5 mm. All are fractionated electrograms. The most fractionated signal (asterisk) corresponds to the point located at the orifice of diverticulum. B: electrograms in the corresponding area of the left atrium acquired during electroanatomic mapping; the distance from the orifice of the right superior pulmonary vein to the area is 12.7 mm. Most points are normal electrograms. The deflection of the fractionated signal (asterisk) is 6.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
study in which complex electrogram means wavefront collision and not a substrate for arrhythmia [8]. In the said study, sinus rhythm fractionated signals were most prevalent on the LA septum (65%) and the anterior wall (49%), which are not completely consistent with locations in Pachon's study. The autopsy study by Igawa et al. [6] showed protrusion from the heart cavity to outside the left atrial wall and the presence of trabeculated myocardium in the diverticulum. Results of the present study support this second mechanism for complex electrogram in the diverticulum: most points were normal electrograms on the left atrial roof in the control group, and complex signal in diverticulum appears unrelated with AF. However, a case report [10] showed that diverticulum ablation terminated atrial fibrillation. In our study no radiofrequency energy was delivered in the diverticulum in any patient and no association was found between atrial fibrillation and diverticulum presence. During follow-up, reasons for recurrent AF included return of pulmonary potentials and nonpulmonary vein triggers such as the superior vena cava.
4.3. Voltage of electrograms Although Peng et al. [3] reported that the diverticular wall was thinner than the normal wall in the left atrium, and that diverticular wall construction did not include fibrotic tissue and was similar to that of normal atrial wall [6], our study did not document more low voltage potentials in the diverticular area as compared with corresponding areas
51
in the control group. Also voltage amplitude was not indicative of diverticulum presence. 4.4. High impedance of diverticulum Impedance monitoring is commonly used in the electrophysiology laboratory. During RF application, a 10–20% drop in local impedance often indicates a successful ablation, while sudden increases in impedance suggests carbonization on the catheter tip. During mapping, higher local impedance can be seen when the catheter is inserted into cardiac veins such as the pulmonary vein or coronary sinus [11–13] mainly because of two reasons: the particular tissue (e.g., lung) around the vein, leading to an increased resistive effect, and smaller blood volume in the vein, reducing its conductive effect. The highest Δimpedance between the vein and atrial tissue can be greater than 100 Ω [11]. The end of a diverticulum is a cecum, and cystiform, cone-shaped, and tubiform diverticula provide little space; therefore, blood volume in the diverticula is small leading to a remarkable increase in local impedance. In our study, Δimpedance differentiated the diverticulum from the normal wall in the left atrium with high sensitivity, specificity, and positive and negative predictive values. During mapping without atrial CT, sudden high local impedance combined with more fractionated signals suggests presence of a diverticulum in the atrial wall. More careful and gentler manipulation of diverticular areas might avoid cardiac tamponade; also, RF application to these areas should be avoided as much as possible.
Fig. 2. A: impedance of points in a tubiform diverticulum. The highest impedance in the deep site of diverticulum is 230 Ω, while that at the ostium of diverticulum is 140 Ω which is the same as that for the tissue surrounding the diverticulum. B: length and width of the tubiform diverticulum. C: three-dimensional CT reconstruction of the diverticulum.
52
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
Fig. 3. A: impedance of points at two cone-shaped diverticula located in the anterosuperior wall close to the right superior pulmonary vein, and in the inferior wall. The highest impedance in the deep part of anterosuperior diverticulum is 170 Ω, while that at the orifice of the interior diverticulum is 166 Ω which is also higher than that of the normal left atrial wall (145 Ω). B: three-dimensional CT reconstruction of the two diverticula. C: three-dimensional CT reconstruction clearly showed the diverticulum located in the inferior wall.
5. Limitations For safety and accuracy reasons, we collected points by evaluating tissue contact through catheter pressure feeling and catheter motion according to fluoroscopy. However, time and conditions between MDCT scanning and mapping varied among patients which does not ensure absolute accuracy in LAD orifice location; therefore, a certain range area (10 mm diameter) encompassing LAD was defined for point collection. Because the LAD wall is known to be thinner than that of adjacent LA, effective refractory period (ERP) and conductive velocity (CV) were not determined to avoid multiple stimulations in or around the diverticulum. Therefore we could not assess the relationship between the latter two electrophysiological characteristics of LAD and atrial fibrillation. In this study, the number of patients enrolled was small, limiting the power of multivariate linear regression and logistic regression analysis. However, significant differences were found between the groups. 6. Conclusions LAD is not uncommon among patients with AF undergoing ablation, and its potential risk for cardiac tamponade has attracted attention. This
is the first study documenting the basic electrophysiological features of LAD, namely higher impedance, complex signals and normal voltage. Change of impedance and deflection of potential during mapping can help distinguish possible diverticulum from the adjacent left atrial wall. Radiofrequency application should be avoided or decreased energy delivered at LAD.
Conflict of interest The authors report no relationships that could be construed as a conflict of interest.
Author contributions Chen Tan: research design; analysis and interpretation of data; drafting the paper; approval of the submitted and final versions. Wei Han: collect data; analysis of data. Xingpeng Liu: collect data; revising the paper critically. Xuehong Hu and Jianguo Liu: collect data; analysis of data. Junyu Cui and Junxia Li: interpretation of data.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
53
Fig. 4. A: impedance of points at a cystiform diverticulum. The highest impedance in the deep part of diverticulum is 170 Ω, the impedance at the orifice of diverticulum is 150 Ω which is also higher than that of the normal atrial wall (135 Ω). B, C: length and width of the tubiform diverticulum.
References
Fig. 5. Comparison of Δimpedance between the LAD and control groups. Both average Δimpedance and highest Δimpedance are significantly higher in the LAD vs. control group. * indicates P b 0.0001 for the comparison between groups.
[1] Calkins H, Kuck KH, Cappato R, et al. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design. Heart Rhythm 2012;9:632–96. [2] Wan YD, He Z, Zhang L, et al. The anatomical study of left atrium diverticulum by multi-detector row CT. Surg Radiol Anat 2009;31:191–8. [3] Peng LQ, Yu JQ, Yang ZG, et al. Left atrial diverticula in patients referred for radiofrequency ablation of atrial fibrillation: assessment of prevalence and morphologic characteristics by dual-source computed tomography. Circ Arrhythm Electrophysiol 2012;5:345–50. [4] De Ponti R, Lumia D, Marazzi R, et al. Left atrial diverticula in patients undergoing atrial fibrillation ablation: morphologic analysis and clinical impact. J Cardiovasc Electrophysiol 2013;24:1232–9. [5] Incedayi M, Öztürk E, Sonmez G, et al. The incidence of left atria diverticula in coronary CT angiography. Diagn Interv Radiol 2012;18:542–6. [6] Igawa O, Miake J, Adach M. The small diverticulum in the right anterior wall of the left atrium. Europace 2008;10:120. [7] Stiles MK, John B, Wong CX, et al. Paroxysmal lone atrial fibrillation is associated with an abnormal atrial substrate: characterizing the “second factor.”. J Am Coll Cardiol 2009;53:1182–91. [8] Saghy L, Callans DJ, Garcia F, et al. Is there a relationship between complex fractionated atrial electrograms recorded during atrial fibrillation and sinus rhythm fractionation? Heart Rhythm 2012;9:181–8. [9] Pachon MJC, Pachon MEI, Pachon MJC, et al. A new treatment for atrial fibrillation based on spectral analysis to guide the catheter RF-ablation. Europace 2004;6:590–601.
54
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
[10] Killeen RP, O'Connor SA, Keane D, Dodd JD. Ectopic focus in an accessory left atrial appendage: radiofrequency ablation of refractory atrial fibrillation. Circulation 2009;120:e60–2. [11] Pollak SJ, Seckel H, Monir J, Ebra G, Monir G. Detection of inadvertent catheter movement into the coronary sinus ostium or middle cardiac vein by real-time impedance monitoring prior to radiofrequency ablation in the right atrial posteroseptal region. J Interv Card Electrophysiol 2012;34:311–5.
[12] Lang CC, Gugliotta F, Santinelli V, et al. Endocardial impedance mapping during circumferential pulmonary vein ablation of atrial fibrillation differentiates between atrial and venous tissue. Heart Rhythm 2006;3:171–8. [13] Cheung P, Hall B, Chugh A, et al. Detection of inadvertent catheter movement into a pulmonary vein during radiofrequency catheter ablation real-time impedance monitoring. J Cardiovasc Electrophysiol 2004;15:674–8.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Electrophysiological characteristics of left atrial diverticulum in patients with atrial fibrillation: Electrograms, impedance and clinical implications Chen Tan a,⁎, Wei Han b, Xingpeng Liu c, Xuehong Hu a, Jianguo Liu a, Junyu Cui a, Junxia Li a a b c
Department of Cardiology, Beijing Military Region General Hospital of PLA, Beijing, China Department of Radiology, Beijing Military Region General Hospital of PLA, Beijing, China Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
a r t i c l e
i n f o
Article history: Received 30 March 2014 Received in revised form 15 May 2014 Accepted 24 June 2014 Available online 2 July 2014 Keywords: Atrial fibrillation Left atrial diverticulum Catheter ablation Impedance Electroanatomic mapping
a b s t r a c t Background: Left atrial diverticulum (LAD) is not rare in patients with atrial fibrillation (AF). Recent reports focused on its morphology however data on its electrophysiological characteristics are lacking. Our study aims to investigate the electrogram and impedance features of LAD. Methods: This study included 24 patients (mean age, 58.5 ± 10.7 years) with LAD undergoing catheter ablation for AF and 24 gender-and-age-matched individuals without LAD as controls. A bipolar LAD electroanatomic map was acquired in sinus rhythm from all study participants. Points were acquired for diverticulum in the LAD group and for corresponding areas in the control group. Electrogram deflections were counted, bipolar voltage and impedance were measured for each point, and average Δimpedance and highest Δimpedance were calculated. Results: A total of 234 points were collected in the two groups. In the LAD vs. control group, median (Q1, Q3) of electrogram deflections was 6 (5, 7) and 4 (4, 5) (P b 0.0001), respectively, voltage was not significantly different (1.58 ± 0.68 mV vs. 1.28 ± 0.65 mV, P = 0.10), and average Δimpedance was significantly higher in the LAD group (19.5 ± 9.0 Ω vs 3.9 ± 1.7 Ω, P b 0.0001). A cut-off value of 9.5 Ω for Δimpedance predicted LAD with sensitivity, specificity, and positive and negative predictive values of 83.5%, 92.8%, 92.1% and 84.9%, respectively. Conclusions: Electrogram was more fractionated and impedance was higher at LAD than in corresponding areas without LAD, which might help to differentiate LAD during catheter ablation for AF. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atrial fibrillation (AF) is the most common tachyarrhythmia, and catheter ablation is considered a reasonable therapy for paroxysmal AF [1]. Left atrial diverticulum (LAD) is common in AF patients (17% to 41% occurrence) [2–5], with most (74%–80%) [4,5] located at the anterosuperior left atrium in various shapes, namely cystiform, coneshaped, tubiform and irregular. Examination of an autopsied heart [6] revealed that, despite similar endocardial surface, LAD wall was much thinner than adjacent tissue [3] which might increase tamponade risk for catheter manipulation during mapping and complicate radiofrequency application in LAD. Little is known about the
⁎ Corresponding author at: Department of Cardiology, Beijing Military Region General Hospital of PLA, Dongsi Shitiao, Dongcheng District, Beijing 100700, China. Tel.: + 86 010 84008033; fax: +86 010 66721872. E-mail address: [email protected] (C. Tan).
http://dx.doi.org/10.1016/j.ijcard.2014.06.050 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
electrophysiological characteristics of LAD, which should be distinguished from those of abnormal left atrial wall without LAD to avoid excessive ablation in LAD. This study therefore investigates the electrogram and impedance features of LAD relative to those of corresponding areas without LAD in AF patients undergoing catheter radiofrequency ablation.
2. Methods 2.1. Patient population From February 2012 to November 2013, 91 patients with AF underwent multidetector computed tomography (MDCT) pulmonary venography and catheter radiofrequency ablation at our hospital. After excluding patients with congenital and/or valvular heart disease, 24 patients meeting the inclusion criteria (LAD detected by cardiac CT; and presence/restoration of sinus rhythm during ablation) were consecutively assigned to the LAD group while another 24 without LAD, matched with regard to gender and age (±2 years), served as the control group. The study was approved by the institutional review committee of our hospital, and all participants provided informed consent.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54 2.2. MDCT protocol Cardiac CT images were acquired on a 64-slice CT scanner (Discovery CT750 HD, GE Company, USA) using 0.625-mm slice thickness, 350-ms gantry rotation, 120 kV and 500–600 mA. A total of 50–70 mL of non-ionic contrast agent (iopamidol, 370 mg/mL; Bracco Sine Pharmaceutical Corp. Ltd, Shanghai, China) was injected via the antecubital vein followed by 40 mL of saline, both at a flow rate of 5.0 mL/s. Contrast agent tracking was used to synchronize scanning from the thorax entrance to the diaphragm. Acquired data were transferred to a workstation (AW4.5, GE Company, USA) for image processing, interpretation, and three-dimensional rendering generation. Images were reconstructed with 0.625-mm slice thickness by volume-rendered, multi-planar reformation, and maximum intensity projection reconstructions. LAD was defined as a protrusion from the heart cavity to outside the left atrial wall plane. For each diverticulum, orifice and cervix width and length were measured. According to a previous study [3], LAD shape was classified as: cystiform, if LAD had a broad orifice and a domelike cecum (i.e., body length/orifice width ratio b3); cone-shaped, if LAD had a broad orifice and looked like a cone; and tubiform, if LAD appeared as a long but small cavity (length/orifice width N3). 2.3. Diverticulum electrogram mapping The entire process described below was guided by CARTO XP or CARTO 3 (Biosense Webster, Diamond Bar, CA, USA). A bipolar LAD map was acquired in sinus rhythm as carefully as possible during the ablation procedure. If the patient was in atrial fibrillation, cardioversion to sinus rhythm was followed by a waiting period of at least 5 min. A copy of the cardiac computerized tomography taken prior to the procedure was integrated into the electroanatomic mapping system. For both groups, an ablation catheter (Navistar Thermocool, Diamond Bar, CA, Biosense Webster, USA) was used to collect points for electroanatomic mapping. Due to safety and accuracy of the result, all the ablations were avoided in an area including the diverticulum, of which the diameter was at least 10 mm. If a diverticulum was located close to the pulmonary veins, we left it outside the ablation line. Meanwhile, if the diverticulum was located on the left atrial roof, we tried to move the roof ablation line to a bit anterior or posterior to avoid delivering radiofrequency energy in the diverticulum. We collected points with evaluation of tissue contact based on the stable fluoroscopic motion of the catheter and stable electrogram morphology. Three to five points within a 10-mm-diameter area were acquired for the diverticulum according to its size and shape. For the control group, 3–5 points were also collected in a similar area to that in the LAD group. Number of deflections present in each electrogram was determined by manual counting as previously reported [6]. Electrograms with N5 deflections were defined as fragmented electrograms, while those with ≤5 deflections were considered normal electrograms. The percentage of fractionated signals was calculated by dividing the number of fractionated signals by the total number of signals at diverticulum in the LAD group or the corresponding area in the control group. 2.4. Voltage mapping Local voltage of the local electrogram was defined as the amplitude of peak positiveto-peak negative deflections. For each point, bipolar voltage was recorded with the combination of automated algorithms, and manually verified to validate the result. Low voltage was defined by the conventional b0.5 mV cut-off point for the atrium [7]. 2.4.1. Impedance mapping For each point acquired, impedance was recorded during mapping. Concurrently, average impedance of three points around the target area was recorded as the left atrial impedance. Δimpedance was impedance of each point minus the left atrial impedance. The
49
highest Δimpedance of each area was also recorded. Average Δimpedance was calculated by dividing the total Δimpedance by the number of points acquired in the LAD or in the corresponding area in the control group. 2.5. Re-induction or recurrence of atrial fibrillation after catheter ablation All study participants received high-dose isoproterenol (20–30 μg/min) after ablation to locate additional non-pulmonary vein trigger sites not previously present. Rapid pacing in the coronary sinus was also applied to induce atrial tachycardia arrhythmia. During the follow-up period, in patients with recurrence of AF or atrial tachycardia, recovery of potentials in pulmonary veins and association with tachycardia were evaluated first; other mechanisms for the tachycardia were sought thereafter. 2.6. Statistical analysis Continuous data are expressed as mean ± SD or median (25%, 75% interquartile) as appropriate, and categorical variables are presented as percentages. Continuous variables were compared using t-test or Mann–Whitney U test as appropriate, while Chi-square test or Fisher's exact test were used to compare categorical variables. A two-tailed P value of b0.05 indicated statistical significance. Logistic regression was used to explore the relationship between diverticulum presence and clinical features. Multivariate linear regression was used to confirm the association between electrophysiological characteristics and diverticulum presence.
3. Results 3.1. Patient characteristics Baseline characteristics of patients are shown in Table 1. There were no significant differences in age, gender, type of atrial fibrillation, left atrial size, left ventricular end diastolic diameter, left ventricular function and complications between the two groups. Logistic regression analysis showed no association between diverticulum occurrence and patient characteristics. In particular, LAD presence was unrelated to left atrial size and type of atrial fibrillation. 3.2. Diverticulum characteristics A total of 27 LAD were found in 24 of the 91 patients screened. Three patients had multiple diverticula. Twenty LAD (74.1%) were located in the anterosuperior wall, 4 (11.1%) in the septal wall, and only 3 (11.1%) in the inferior wall. LAD presented in three morphologies, including cystiform, cone-shaped and tubiform. 3.3. Diverticulum electrogram characteristics 234 points were collected in the two groups. In the diverticulum group, the most complex signal in diverticulum had 10 deflections (Fig. 1A), the median average deflection of electrograms was 6, and the fractionated signal proportion was 57.1 ± 29.5%. In the control group, the median average deflection of points was 4, and the
Table 1 Patient characteristics in the LAD and control groups.
Number of patients Age Male gender Type of AF Persistent Comorbidities Hypertension Diabetes Ischemia Echocardiography Left atrial size (mm) Left ventricular end diastolic diameter (mm) Left ventricular ejection fraction (%) AF = atrial fibrillation; LAD = left atrial diverticulum.
LAD group
Control group
P value
24 58.5 ± 10.7 17 (70.8%)
24 59.2 ± 10.4 17 (70.8%)
0.81 1
6 (25%)
9 (37.5%)
0.56
12 (50%) 3 (12.5%) 6 (25%)
16 (66.7%) 5 (20.8%) 7 (29.2%)
0.53 0.70 0.98
35.4 ± 3.81 46.1 ± 3.0 62.9 ± 4.68
37.5 ± 5.29 46.4 ± 5.03 64.5 ± 7.46
0.14 0.79 0.32
50
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
percentage of complex signal was 17.6 ± 14.2%, suggesting that most potentials had no more than 5 deflections (Fig. 1B). The electrogram was more fractionated in the diverticulum region than that in corresponding areas without diverticulum in the left atrium (P b 0.0001) (Table 2). Furthermore, multivariate linear regression analysis showed that only diverticulum presence was associated with multiple deflections (P b 0.0001).
Table 2 Comparison of characteristics of electrograms between AF patients with and without LAD.
Median of deflection number Percentage of complex signals (%) Mean voltage (mV)
LAD
Control
P value
6 (5, 7) 57.1 ± 29.5 1.58 ± 0.68
4 (4, 5) 17.6 ± 14.2 1.28 ± 0.65
b0.0001 b0.0001 0.10
3.4. Assessment of voltage The mean voltage in the LAD group (1.58 ± 0.68 mV) was not significantly different from that of the corresponding area in the control group (1.28 ± 0.65 mV, P = 0.10). Multivariate linear regression analysis showed no relationship between voltage and diverticulum presence, however, voltage amplitude correlated with hypertension (P = 0.016).
and one patient did not accept a second ablation. AF recurrence was not associated with diverticulum presence.
3.5. Assessment of impedance
The increased use of ablation in atrial fibrillation treatment has recently fueled recent research mainly focused on incidence and morphology [3,4] but not electrophysiological characteristics of atrial diverticula. In the current study, we found that (1) there were more fractionated signals at diverticulum; (2) the voltage of electrogram at diverticulum was not different from that of the corresponding area without diverticulum; and (3) the impedance of diverticulum was much higher than that of its adjacent area. In this study, we also determined the predictive Δimpedance cut-off of 9.5 Ω for left atrial diverticulum presence.
The highest Δimpedance was 35.3 ± 19.3 Ω in the LAD group, compared with 7.9 ± 3.2 Ω in the control group (P b 0.0001). The maximal Δimpedance in the diverticulum group was 90 Ω; the point was located in the deep site of a tubiform diverticulum (Fig. 2). In addition, impedance of point at the summit of cone-shape or cystiform diverticula was also very high (Figs. 3, 4). The average Δimpedance was also significantly higher than that of the control group (19.5 ± 9.0 Ω vs. 3.9 ± 1.7 Ω, P b 0.0001) (Fig. 5). In multivariate linear regression analysis, Δimpedance was significantly associated with diverticulum presence (P b 0.0001). Cut-off value of Δimpedance was 9.5 Ω by ROC curve statistic. The sensitivity, specificity and positive and negative predictive value of the cut-off impedance was 83.5%, 92.8%, 92.1% and 84.9%, respectively. 3.6. Relationship between diverticulum presence and atrial fibrillation In our study, since diverticulum was intact, we observed that tachycardia was unrelated to diverticulum presence in first ablation. During follow-up, AF recurred in 10 patients in the two groups (5 vs. 5, P = 1): in 6 patients AF recurred because of the recovery of pulmonary vein potentials; 3 patients had triggers in the superior vena cava;
4. Discussion 4.1. Main findings
4.2. Fractionated electrogram in diverticulum In our study, more fractionated signals were recorded in diverticulum and at sites close to the diverticular orifice. One of the possible underlying mechanisms for fractionated electrogram and arrhythmia is the presence of inhomogeneous muscle in diverticulum, which Pachon et al. [9] termed fibrillar myocardium. Fibrillar myocardium, mainly located in the LA roof and septum except for those close to the PV insertion in LA, entails many cell groups that form wave fronts at different phases and manifests as a polymorphic signal in endocardial electrogram; ablation for fractionated signals could help eliminate atrial fibrillation. A second possible mechanism is presented in another
Fig. 1. A: electrograms in a cone-shaped diverticulum in the left atrial anterosuperior wall acquired during electroanatomic mapping. The distance from the orifice of the right superior pulmonary vein to the diverticulum is 12.5 mm. All are fractionated electrograms. The most fractionated signal (asterisk) corresponds to the point located at the orifice of diverticulum. B: electrograms in the corresponding area of the left atrium acquired during electroanatomic mapping; the distance from the orifice of the right superior pulmonary vein to the area is 12.7 mm. Most points are normal electrograms. The deflection of the fractionated signal (asterisk) is 6.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
study in which complex electrogram means wavefront collision and not a substrate for arrhythmia [8]. In the said study, sinus rhythm fractionated signals were most prevalent on the LA septum (65%) and the anterior wall (49%), which are not completely consistent with locations in Pachon's study. The autopsy study by Igawa et al. [6] showed protrusion from the heart cavity to outside the left atrial wall and the presence of trabeculated myocardium in the diverticulum. Results of the present study support this second mechanism for complex electrogram in the diverticulum: most points were normal electrograms on the left atrial roof in the control group, and complex signal in diverticulum appears unrelated with AF. However, a case report [10] showed that diverticulum ablation terminated atrial fibrillation. In our study no radiofrequency energy was delivered in the diverticulum in any patient and no association was found between atrial fibrillation and diverticulum presence. During follow-up, reasons for recurrent AF included return of pulmonary potentials and nonpulmonary vein triggers such as the superior vena cava.
4.3. Voltage of electrograms Although Peng et al. [3] reported that the diverticular wall was thinner than the normal wall in the left atrium, and that diverticular wall construction did not include fibrotic tissue and was similar to that of normal atrial wall [6], our study did not document more low voltage potentials in the diverticular area as compared with corresponding areas
51
in the control group. Also voltage amplitude was not indicative of diverticulum presence. 4.4. High impedance of diverticulum Impedance monitoring is commonly used in the electrophysiology laboratory. During RF application, a 10–20% drop in local impedance often indicates a successful ablation, while sudden increases in impedance suggests carbonization on the catheter tip. During mapping, higher local impedance can be seen when the catheter is inserted into cardiac veins such as the pulmonary vein or coronary sinus [11–13] mainly because of two reasons: the particular tissue (e.g., lung) around the vein, leading to an increased resistive effect, and smaller blood volume in the vein, reducing its conductive effect. The highest Δimpedance between the vein and atrial tissue can be greater than 100 Ω [11]. The end of a diverticulum is a cecum, and cystiform, cone-shaped, and tubiform diverticula provide little space; therefore, blood volume in the diverticula is small leading to a remarkable increase in local impedance. In our study, Δimpedance differentiated the diverticulum from the normal wall in the left atrium with high sensitivity, specificity, and positive and negative predictive values. During mapping without atrial CT, sudden high local impedance combined with more fractionated signals suggests presence of a diverticulum in the atrial wall. More careful and gentler manipulation of diverticular areas might avoid cardiac tamponade; also, RF application to these areas should be avoided as much as possible.
Fig. 2. A: impedance of points in a tubiform diverticulum. The highest impedance in the deep site of diverticulum is 230 Ω, while that at the ostium of diverticulum is 140 Ω which is the same as that for the tissue surrounding the diverticulum. B: length and width of the tubiform diverticulum. C: three-dimensional CT reconstruction of the diverticulum.
52
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
Fig. 3. A: impedance of points at two cone-shaped diverticula located in the anterosuperior wall close to the right superior pulmonary vein, and in the inferior wall. The highest impedance in the deep part of anterosuperior diverticulum is 170 Ω, while that at the orifice of the interior diverticulum is 166 Ω which is also higher than that of the normal left atrial wall (145 Ω). B: three-dimensional CT reconstruction of the two diverticula. C: three-dimensional CT reconstruction clearly showed the diverticulum located in the inferior wall.
5. Limitations For safety and accuracy reasons, we collected points by evaluating tissue contact through catheter pressure feeling and catheter motion according to fluoroscopy. However, time and conditions between MDCT scanning and mapping varied among patients which does not ensure absolute accuracy in LAD orifice location; therefore, a certain range area (10 mm diameter) encompassing LAD was defined for point collection. Because the LAD wall is known to be thinner than that of adjacent LA, effective refractory period (ERP) and conductive velocity (CV) were not determined to avoid multiple stimulations in or around the diverticulum. Therefore we could not assess the relationship between the latter two electrophysiological characteristics of LAD and atrial fibrillation. In this study, the number of patients enrolled was small, limiting the power of multivariate linear regression and logistic regression analysis. However, significant differences were found between the groups. 6. Conclusions LAD is not uncommon among patients with AF undergoing ablation, and its potential risk for cardiac tamponade has attracted attention. This
is the first study documenting the basic electrophysiological features of LAD, namely higher impedance, complex signals and normal voltage. Change of impedance and deflection of potential during mapping can help distinguish possible diverticulum from the adjacent left atrial wall. Radiofrequency application should be avoided or decreased energy delivered at LAD.
Conflict of interest The authors report no relationships that could be construed as a conflict of interest.
Author contributions Chen Tan: research design; analysis and interpretation of data; drafting the paper; approval of the submitted and final versions. Wei Han: collect data; analysis of data. Xingpeng Liu: collect data; revising the paper critically. Xuehong Hu and Jianguo Liu: collect data; analysis of data. Junyu Cui and Junxia Li: interpretation of data.
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
53
Fig. 4. A: impedance of points at a cystiform diverticulum. The highest impedance in the deep part of diverticulum is 170 Ω, the impedance at the orifice of diverticulum is 150 Ω which is also higher than that of the normal atrial wall (135 Ω). B, C: length and width of the tubiform diverticulum.
References
Fig. 5. Comparison of Δimpedance between the LAD and control groups. Both average Δimpedance and highest Δimpedance are significantly higher in the LAD vs. control group. * indicates P b 0.0001 for the comparison between groups.
[1] Calkins H, Kuck KH, Cappato R, et al. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design. Heart Rhythm 2012;9:632–96. [2] Wan YD, He Z, Zhang L, et al. The anatomical study of left atrium diverticulum by multi-detector row CT. Surg Radiol Anat 2009;31:191–8. [3] Peng LQ, Yu JQ, Yang ZG, et al. Left atrial diverticula in patients referred for radiofrequency ablation of atrial fibrillation: assessment of prevalence and morphologic characteristics by dual-source computed tomography. Circ Arrhythm Electrophysiol 2012;5:345–50. [4] De Ponti R, Lumia D, Marazzi R, et al. Left atrial diverticula in patients undergoing atrial fibrillation ablation: morphologic analysis and clinical impact. J Cardiovasc Electrophysiol 2013;24:1232–9. [5] Incedayi M, Öztürk E, Sonmez G, et al. The incidence of left atria diverticula in coronary CT angiography. Diagn Interv Radiol 2012;18:542–6. [6] Igawa O, Miake J, Adach M. The small diverticulum in the right anterior wall of the left atrium. Europace 2008;10:120. [7] Stiles MK, John B, Wong CX, et al. Paroxysmal lone atrial fibrillation is associated with an abnormal atrial substrate: characterizing the “second factor.”. J Am Coll Cardiol 2009;53:1182–91. [8] Saghy L, Callans DJ, Garcia F, et al. Is there a relationship between complex fractionated atrial electrograms recorded during atrial fibrillation and sinus rhythm fractionation? Heart Rhythm 2012;9:181–8. [9] Pachon MJC, Pachon MEI, Pachon MJC, et al. A new treatment for atrial fibrillation based on spectral analysis to guide the catheter RF-ablation. Europace 2004;6:590–601.
54
C. Tan et al. / International Journal of Cardiology 176 (2014) 48–54
[10] Killeen RP, O'Connor SA, Keane D, Dodd JD. Ectopic focus in an accessory left atrial appendage: radiofrequency ablation of refractory atrial fibrillation. Circulation 2009;120:e60–2. [11] Pollak SJ, Seckel H, Monir J, Ebra G, Monir G. Detection of inadvertent catheter movement into the coronary sinus ostium or middle cardiac vein by real-time impedance monitoring prior to radiofrequency ablation in the right atrial posteroseptal region. J Interv Card Electrophysiol 2012;34:311–5.
[12] Lang CC, Gugliotta F, Santinelli V, et al. Endocardial impedance mapping during circumferential pulmonary vein ablation of atrial fibrillation differentiates between atrial and venous tissue. Heart Rhythm 2006;3:171–8. [13] Cheung P, Hall B, Chugh A, et al. Detection of inadvertent catheter movement into a pulmonary vein during radiofrequency catheter ablation real-time impedance monitoring. J Cardiovasc Electrophysiol 2004;15:674–8.
