IJCA-18036; No of Pages 8 International Journal of Cardiology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation☆ Hisanori Kanazawa a, Hiroshige Yamabe a,⁎, Koji Enomoto a, Junjiroh Koyama a, Kenji Morihisa a, Tadashi Hoshiyama a, Kunihiko Matsui b, Hisao Ogawa a a b

Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Chuo-ku, Kumamoto 860–8556, Japan Department of General Internal Medicine, Yamaguchi University Hospital, Ube, Japan

a r t i c l e

i n f o

Article history: Received 7 February 2014 Received in revised form 10 April 2014 Accepted 12 April 2014 Available online xxxx Keywords: atrial fibrillation pericardial fat complex fractionated atrial electrogram inflammation fibrosis

a b s t r a c t Background/Objectives: Pericardial fat (PF) and complex fractionated atrial electrogram (CFAE) are both associated with atrial fibrillation (AF). Therefore, we examined the relation between PF and CFAE area in AF. Methods: The study population included 120 control patients without AF and 120 patients with AF (80 paroxysmal AF and 40 persistent AF) who underwent catheter ablation. Total cardiac PF volume, representing all adipose tissue within the pericardial sac, was measured by contrast-enhanced computed tomography. The location and distribution of CFAE region were identified by left atrial endocardial mapping using a three-dimensional mapping system. We analyzed the significance of total cardiac PF volume and total area of CFAE region on AF, persistence of AF from paroxysmal to persistent form, and the relation between total cardiac PF volume and total CFAE area. We also evaluated the regional distribution of PF volume and CFAE area in five areas of the left atrium (LA). Results: Total cardiac PF volume correlated with AF (odds ratio [OR]: 1.024, p b 0.001). Total cardiac PF volume and total CFAE area were both independently associated with persistence of AF (OR: 1.018, p = 0.018, OR: 1.144, p = 0.002, respectively). Multivariate linear regression analysis identified total cardiac PF volume as a significant and independent determinant of total CFAE area (r = 0.488, p b 0.001). Furthermore, regional PF volume correlated with local CFAE area in an each LA area. Conclusions: PF volume correlated significantly with CFAE area in patients with AF. This finding suggests that PF is directly related to the progression of CFAE area and promotes the pathogenic process of AF. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Previous studies have reported a close relationship between inflammation and atrial fibrillation (AF) [1,2]. Cytokines are known to activate fibroblasts, with subsequent extracellular matrix deposition and fibrosis, resulting in the initiation and maintenance of AF [3,4]. Previous studies have shown that pericardial fat (PF) is highly associated with AF [5–7]. In addition, Wong et al. [6] showed that PF is also associated with the severity of AF and poorer outcomes of AF ablation. Furthermore, Shin et al. [7] reported that PF is independently associated with left atrial volume in AF patients, suggesting a close relation between PF and left atrial remodeling. Human PF correlates with increased expression of numerous inflammatory markers [8,9]. Thus, PF

Abbreviations: AF, atrial fibrillation; BMI, body mass index; BNP, brain natriuretic peptide; CFAE, complex fractionated atrial electrogram; CT, computed tomography; LA, left atrium; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; PF, pericardial fat; SR, sinus rhythm. ☆ Part of this study was presented at the 84th Scientific Session of the American Heart Association, Orlando, FL, November 12–16, 2011. ⁎ Corresponding author. Tel.: +81 96 373 5175; fax: +81 96 362 3256. E-mail address: [email protected] (H. Yamabe).

seems to increase the risk for the development of AF through the secretion of inflammatory mediators via paracrine or vasocrine pathway [9]. The complex fractionated atrial electrogram (CFAE) region is considered critical for the initiation and maintenance of AF, and associated with conduction slowing and heterogeneous activation [10,11]. Recent studies have shown that the genesis of CFAE region correlates with myocardial fibrosis, suggesting a relation between CFAE and inflammation of the atrial myocardium [12,13]. We hypothesized the presence of a close relationship between PF volume and the extent of CFAE, both associated with the development of AF. To test this hypothesis, we determined the significance of PF volume and the area of CFAE region on AF and the change from paroxysmal AF to persistent AF. Furthermore, we determined the relation between PF volume and the area of CFAE region in AF. 2. Methods 2.1. Study population The study population included 120 control patients without history of AF (sinus rhythm [SR] group) and 120 consecutive patients with AF (AF group) who underwent first-time catheter ablation for paroxysmal AF (n = 80) and persistent AF (n = 40), with no contraindication to contrast-enhanced computed tomography (CT). All antiarrhythmic

http://dx.doi.org/10.1016/j.ijcard.2014.04.135 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

2

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx

medications were discontinued five half-life periods before the ablation. Patients of the SR group underwent cardiac contrast CT for the evaluation of coronary artery. Patients of the AF group also underwent contrast-enhanced CT before the procedure to evaluate the threedimensional structure of the left atrium (LA). None of the patients of the SR and AF groups had evidence of structural heart disease including coronary artery disease. Classification of paroxysmal and persistent AF was defined according to the 2012 Heart Rhythm Society/ European Heart Rhythm Association/European Cardiac Arrhythmia Society Expert Consensus Statement on Catheter and Surgical Ablation of Atrial Fibrillation [14]. The study protocol was approved by the Hospital Human Research Committee and a signed an informed consent form was obtained from each patient before enrollment in the study.

2.2. CT and measurement of pericardial fat A 64-detector CT machine (Brilliance-64, Phillips Medical Systems, Cleveland, OH) was used with the following parameters: detector collimation 64 × 0.625 mm, table feed 19.7 mm/s, 0.2 helical pitch (beam pitch), and rotation time 0.4 seconds, tube current 900 mAs, and voltage 120 kVp, as reported previously [15,16]. During a single inspiratory breath-hold, imaging was performed in a craniocaudal direction on helical mode from the bifurcation of the pulmonary artery to the diaphragm with a retrospective electrocardiographic gating. The contrast material (iohexol [Omnipaque-350; Daiichi-Sankyo Pharmaceutical, Tokyo, Japan], which contained 350 mg iodine/mL) was administered using a mechanical power injector (Dual Shot; Nemoto-Kyorindo, Tokyo). To minimize the difference in left atrial enhancement among patients, we adopted the body weight-tailored contrast material dose of 1.2 mL/kg and the fixed injection duration of 9 seconds. The contrast material was administered intravenously and dynamic monitoring scans acquired at the level of the ascending aorta (the level of left main coronary trunk) to generate the patient-specific enhancement curve. For angiography, we selected a delay of 6 seconds after peak enhancement determined by the enhancement curve. The scanning time varied from 6 to 8 seconds. To obtain adequate gating and minimal motion artifact, an oral β-adrenergic blocker (metoprolol, 20 mg) was administered 1 hour before CT imaging in all patients. The scan raw data were reconstructed with 75% of the cardiac cycle or particular optimal phase, and the reconstructed image date of the CT was transferred to a workstation for post processing (ZIO M900, Amin/ZIO, Tokyo). PF volume was measured using a semiautomatic technique in three dimensions on the workstation, as reported previously [15,16] (Fig. 1). Readers blinded to the clinical data were required to trim along the pericardial sac using axial (panel A), coronal (panel B) and sagittal slices (panel C), and volume-rendered image (panels D and E). A slice 1 cm above the top level of the left atrial appendage was defined as the superior border of PF. Then, a predefined image display setting (window width, 150 Hounsfield units; window center, −120 Hounsfield units) was used to identify pixels that corresponded to adipose tissue [17]. Total cardiac PF volume was defined as any adipose tissue located within the pericardial sac.

2.3. Localization of CFAE region Endocardial mapping of the LA was performed using a three-dimensional mapping system (EnSite System 3000; St. Jude Medical, St. Paul, MN), as reported previously [10, 11,18,19]. A 9-Fr multi-electrode array catheter (EnSite Array; St. Jude Medical) was introduced into the LA from the right femoral 10.5-Fr sheath through transseptal approach, deployed on a 0.032-inch guide-wire with its distal tip fixed in the left superior pulmonary vein. A 7-Fr large-tip (8 mm in length) deflectable quadripolar electrode catheter (Japan Lifeline, Tokyo) was also inserted into the LA from the right femoral 8.5-Fr sheath. After constructing the geometry of LA, contact CFAE mapping during AF was performed before catheter ablation procedure in all patients. For patients who presented with sinus rhythm, AF was induced by rapid burst atrial pacing at a rate up to loss of 1:1 capture without isoproterenol or adenosine. In all patients, CFAE mapping was performed during sustained AF that lasted more than 10 minutes. When AF terminated, it was re-induced and the CFAE map was generated again from the start. The automatic “CFE-mean” detection algorithm of NavX DxL CFE software (St. Jude Medical, St. Paul, MN) was used for CFAE mapping [18]. Accurate identification of the CFAE area by this automatic “CFE-mean” detection algorithm has been confirmed previously [19]. At each LA site, deflections above peak-to-peak sensitivity were detected automatically and the mean AF cycle length was calculated by averaging the time interval between consecutive deflections over 6.0 seconds. The peakto-peak sensitivity was adjusted for each patient among 0.03 to 0.10 mV based on the electrical noise level, and the refractory period of 30 msec was used to avoid double counting of fractionated electrograms. The low-voltage identification of 0.05 mV was utilized to identify electrical scar, and the width criterion between maximum to minimum deflections of 10 msec was adopted to exclude wide deflections with a gentle slope and limit high frequency signals. Furthermore, electrograms with distance exceeding 7 mm from a geometry of LA (interior projection; 7 mm, exterior projection; 7 mm) were removed to prevent far-field electrograms. Based on the criteria proposed by Nademanee et al. [20], CFAE was defined as the continuous fractionated electrogram with a short cycle length b120 msec. The location of CFAE region was also validated by the virtual unipolar and virtual bipolar Laplacian electrograms (Fig. 2). The areas of whole LA and CFAE region were measured on a three-dimensional geometry of the LA, constructed by the EnSite system.

2.4. Correlation between regional distribution of PF and CFAE area The PF volume over the whole LA was defined as total LA PF volume (Fig. 3A). The CFAE area over the whole LA was defined as total CFAE area. The LA was divided into five areas: roof LA, anterior LA, lateral LA, septal LA and posterior LA (Fig. 3A and B), and the regional distribution of PF volume and CFAE area in each LA area was evaluated. Furthermore, the correlation between PF volume and CFAE area over the whole LA and in each of the five divided LA areas was examined.

Fig. 1. Measurement of PF on the contrast-enhanced CT. The pericardial fat (PF) (yellow) was initially defined by trimming the pericardial sac (red) using axial (panel A), coronal (panel B) and sagittal slices (panel C). Then, the volume-rendered image was created (panel D). Subsequently, the PF image (panel E) was obtained using a predefined image display setting, and total cardiac PF volume was measured.

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx

3

Fig. 2. Identification of the CFAE region by contact and noncontact mapping. Location of the CFAE regions (encircled by the solid line). Contact unipolar and bipolar electrograms and virtual unipolar and Laplacian bipolar electrograms in the CFAE region (panel A) and non-CFAE region (panel B). I and aVF = surface electrocardiogram leads; CS = coronary sinus; LAA = left atrial appendage; LIPV = left inferior pulmonary vein; LSPV = left superior pulmonary vein; MV = mitral valve; RIPV = right inferior pulmonary vein; RSPV = right superior pulmonary vein.

2.5. Statistical analysis Data are expressed as mean ± standard deviation or frequencies (percent). Continuous parameters were compared using the Student's t test for normally distributed data and the Mann–Whitney U test for parameters with skewed data distribution. Categorical

parameters were compared using the chi-square test. To examine whether any of the baseline parameters were independently associated with the presence of AF, and persistence of AF, multiple logistic regression analysis using parameters with p value b0.05 in the comparison of baseline characteristics, was conducted. Furthermore, to determine whether any of the baseline parameters were independently associated with the total

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

4

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx

Fig. 3. The five areas of the left atrium (LA). The pericardial fat (PF) volume over the whole LA (total LA PF volume) was extracted (panel A). Then, the total LA PF volume was divided into five areas (i.e., roof LA, anterior LA, lateral LA, septal LA and posterior LA) (panel A). The local CFAE area was also measured at these five divided LA area (panel B). LPV = left pulmonary vein; MV = mitral valve; RPV = right pulmonary vein. A: anterior view, P: posterior view.

CFAE area, multivariate linear regression analysis was conducted and included parameters with p value b0.05 in the univariate linear regression analysis. The correlation between regional PF volume and local CFAE area was evaluated using the Pearson's product– moment correlation coefficient or Spearman's rank correlation coefficient, depending on the data distribution. Differences among regional PF volume or local CFAE area in the five divided LA areas were analyzed using Kruskal–Wallis test, followed by Steel– Dwass multiple comparison. Statistical tests were performed using the SPSS software, version 19 (IBM, NY). A two-tailed p value b0.05 was considered statistically significant.

3. Results 3.1. Patient characteristics The baseline characteristics of patients of the SR and AF groups are listed in Table 1. Body mass index (BMI) and waist circumference were significantly larger in the AF group than SR group. Left atrial diameter (LAD) was significantly larger in the AF group than SR group. Left ventricular ejection fraction (LVEF) was significantly lower in the AF group than SR group. The level of brain natriuretic peptide (BNP) was significantly higher in the AF group than SR group. Furthermore, the total cardiac PF volume was significantly larger in the AF group than SR group (Table 1). A significantly larger proportion of patients of the AF group were hypertensive than the SR group. The use of calcium

channel blockers was not different between the two groups, whereas angiotensin-converting enzyme inhibitors, angiotensin II type 1 receptor blockers, β-blockers and antiarrhythmic drugs were significantly more commonly used in the AF group than SR group (Table 1). The baseline characteristics of patients of the paroxysmal and persistent AF groups are listed in Table 2. LAD and LA area were larger in the persistent AF than paroxysmal AF group. LVEF was significantly lower in the persistent AF than paroxysmal AF group. BNP level was significantly higher in the persistent AF than paroxysmal AF group. Furthermore, total cardiac PF volume and total CFAE area were significantly larger in the persistent AF than paroxysmal AF group (Table 2). There were no significant differences in cardiovascular risk factors and medications between the two AF groups (Table 2).

3.2. Relation between presence or persistence of AF and baseline characteristics To define the factors associated with the presence of AF, baseline parameters that were significantly different between the SR and AF groups (BMI, waist circumference, LAD, LVEF, BNP, total cardiac PF volume and prevalence of hypertension) (Table 1) were entered into

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx Table 1 Baseline characteristics of patients of the sinus rhythm (SR) and atrial fibrillation (AF) groups.

Age, years Gender, male BMI, kg/m2 Waist circumference, cm LAD, mm LVEF, % BNP, pg/ml hs-CRP, mg/dl eGFR, mL/min/1.73 m2 Total cardiac PF volume, ml Cardiovascular risk factors Hypertension Diabetes mellitus Dyslipidemia Smoking Medications Calcium channel blockers ACE-I/ARB β-Blockers Antiarrhythmic drugs

SR group (n = 120)

AF group (n = 120)

p Value

59.2 ± 12.6 93 (78) 23.4 ± 3.5 84.9 ± 9.7 33.9 ± 4.6 64.4 ± 4.6 17.9 ± 13.5 0.10 ± 0.14 75.3 ± 12.0 113.9 ± 32.5

60.4 ± 10.2 98 (82) 24.1 ± 3.1 88.4 ± 8.7 38.3 ± 5.1 62.2 ± 5.8 52.0 ± 51.3 0.12 ± 0.20 72.5 ± 12.2 148.8 ± 46.1

0.516⁎ 0.423† 0.010⁎ 0.003⁎⁎ b0.001⁎⁎ 0.006⁎ b0.001⁎ 0.821⁎ 0.064⁎ b0.001⁎

53 (44) 16 (13) 26 (22) 70 (58)

75 (63) 11 (9) 37 (31) 77 (64)

42 (35) 37 (31) 9 (8) 0 (0)

43 (36) 57 (48) 38 (32) 105 (88)

0.004† 0.307† 0.107† 0.354† 0.893† 0.008† b0.001† b0.001†

Values are mean ± standard deviation or n (%). ACE-I = angiotensin-converting enzyme inhibitors; ARB = angiotensin II type 1 receptor blockers; BMI = body mass index; BNP = brain natriuretic peptide; eGFR = estimated glomerular filtration rate; hs-CRP = high sensitive C-reactive protein; LAD = left atrial diameter; LVEF = left ventricular ejection fraction; PF = pericardial fat. ⁎⁎ p: by Student's t test. ⁎ p: by Mann–Whitney's U test. † p: by chi-square.

multiple logistic regression analysis. After adjustment for waist circumference, LVEF and prevalence of hypertension, the analysis identified BMI, LAD, BNP and total cardiac PF volume as independent and significant parameters associated with AF (Table 3). To define the factors associated with persistence of AF from paroxysmal to persistent form, the baseline parameters that were significantly different between the paroxysmal and persistent AF groups (LAD, LA

5

Table 3 Results of multiple logistic regression analysis.

AF and SR† BMI, kg/m2 Waist circumference, cm LAD, mm LVEF, % BNP, pg/ml Total cardiac PF volume, ml Hypertension Persistent and paroxysmal AF¶ LAD, mm LA area, cm2 LVEF, % BNP, pg/ml Total cardiac PF volume, ml Total CFAE area, cm2

Odds ratio (95% CI)

p Value

1.105 (1.016 to 1.202) Not selected 1.157 (1.068 to 1.254) Not selected 1.042 (1.024 to 1.061) 1.024 (1.014 to 1.034) Not selected

0.009 – b0.001 – b0.001 b0.001 –

Not selected Not selected 0.869 (0.790 to 0.956) 1.018 (1.006 to 1.030) 1.018 (1.003 to 1.032) 1.144 (1.052 to 1.244)

– – 0.004 0.003 0.018 0.002

Model chi-square test p b 0.001. CI = confidence interval. Other abbreviations as in Tables 1 and 2. † Predictive accuracy 78.8%. ¶ Predictive accuracy 85.8%.

area, LVEF, BNP, total cardiac PF volume and total CFAE area) (Table 2) were entered into multiple logistic regression analysis. After adjustment for LAD and LA area, the analysis identified LVEF, BNP, total cardiac PF volume and total CFAE area as independent and significant determinants of persistent AF (Table 3). 3.3. Relation between CFAE area and baseline characteristics Univariate linear regression analysis was used to examine the relation between total CFAE area and baseline characteristics (Table 4). The total CFAE area correlated positively and significantly with sex, BMI and waist circumference. Furthermore, the total CFAE area correlated positively and significantly with LA area, total cardiac PF volume, prevalence of dyslipidemia and smoking. There was a strongest positive correlation between the total CFAE area and total cardiac PF volume (R = 0.576, p b 0.001) (Table 4 and Fig. 4A).

Table 2 Baseline characteristics of patients with paroxysmal and persistent atrial fibrillation (AF).

Age, years Gender, male BMI, kg/m2 Waist circumference, cm AF duration, years LAD, mm LA area, cm2 LVEF, % BNP, pg/ml hs-CRP, mg/dl eGFR, mL/min/1.73 m2 Total cardiac PF volume, ml Total CFAE area, cm2 Cardiovascular risk factors Hypertension Diabetes mellitus Dyslipidemia Smoking Medications Calcium channel blockers ACE-I/ARB β-Blockers Antiarrhythmic drugs

Paroxysmal AF (n = 80)

Persistent AF (n = 40)

p Value

61.0 ± 10.8 62 (78) 23.8 ± 3.0 87.7 ± 7.9 5.3 ± 5.7 37.3 ± 4.6 130.4 ± 27.4 63.8 ± 5.2 40.8 ± 41.8 0.12 ± 0.21 73.2 ± 12.7 134.1 ± 37.6 20.0 ± 7.3

59.1 ± 8.7 36 (90) 24.9 ± 3.2 89.8 ± 9.9 5.8 ± 4.4 40.2 ± 5.6 142.0 ± 20.7 59.1 ± 5.8 74.5 ± 60.8 0.13 ± 0.18 71.3 ± 11.0 178.3 ± 47.9 28.3 ± 7.8

0.146⁎ 0.095† 0.071⁎⁎ 0.215⁎⁎ 0.198⁎ 0.003⁎⁎ 0.003⁎ b0.001⁎ b0.001⁎ 0.124⁎ 0.434⁎⁎ b0.001⁎⁎ b0.001⁎⁎

49 (61) 8 (10) 21 (26) 48 (60)

26 (65) 3 (8) 16 (40) 29 (73)

0.689† 0.468† 0.124† 0.178†

26 (33) 37 (46) 18 (23) 67 (84)

17 (43) 20 (50) 15 (38) 38 (95)

0.282† 0.698† 0.083† 0.079†

Values are mean ± standard deviation or n (%). CFAE = complex fractionated atrial electrogram; LA = left atrium. Other abbreviations as in Table 1. ⁎⁎ p: by Student's t test. ⁎ p: by Mann–Whitney's U test. † p: by chi-square.

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

6

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx

Table 4 Results of univariate linear regression analysis for the correlation between total CFAE area and baseline clinical features.

Age, years Gender, male BMI, kg/m2 Waist circumference, cm AF duration, years LAD, mm LA area, cm2 LVEF, % BNP, pg/ml hs-CRP, mg/dl eGFR, mL/min/1.73 m2 Total cardiac PF volume, ml Cardiovascular risk factors Hypertension Diabetes mellitus Dyslipidemia Smoking Medications Calcium channel blockers ACE-I/ARB β-Blockers Antiarrhythmic drugs Abbreviations as in Tables 1–3.

β (95% CI)

R

p Value

−0.117 (−0.266 to 0.031) 6.496 (2.740 to 10.252) 0.827 (0.359 to 1.296) 0.275 (0.106 to 0.444) −0.141 (−0.428 to 0.145) 0.274 (−0.020 to 0.567) 0.105 (0.049 to 0.161) −0.236 (−0.496 to 0.025) −0.002 (−0.031 to 0.028) −3.571 (−11.252 to 4.110) −0.092 (−0.216 to 0.033) 0.105 (0.078 to 0.132)

0.142 0.301 0.306 0.284 0.090 0.167 0.323 0.163 0.010 0.084 0.133 0.576

0.121 0.001 0.001 0.002 0.331 0.068 b0.001 0.075 0.915 0.359 0.147 b0.001

1.128 (−2.013 to 4.268) 2.341 (−2.922 to 7.605) 4.085 (0.871 to 7.300) 3.546 (0.435 to 6.658)

0.065 0.081 0.226 0.203

0.479 0.380 0.013 0.026

1.009 (−2.163 to 4.182) −0.448 (−3.498 to 2.603) −0.189 (−3.602 to 3.224) 0.399 (−4.208 to 5.006)

0.058 0.027 0.010 0.016

0.530 0.772 0.913 0.864

After adjustment for BMI, waist circumference, LA area, prevalence of dyslipidemia and smoking, multivariate linear regression analysis identified sex and total cardiac PF volume as independent and significant correlates with the total area of CFAE (R2 = 0.377, ANOVA p b 0.001) (Table 5). In particular, the standardized partial regression coefficient of total cardiac PF volume against total CFAE area was higher than that of sex (0.542 vs. 0.214), suggesting that total CFAE area correlates specifically with total cardiac PF volume (Table 5).

3.4. Correlation between regional distribution of PF and CFAE area The PF volume over the whole LA (total LA PF volume) was 31.3 ± 12.6 ml. PF volumes at the roof LA, anterior LA, lateral LA, septal LA and posterior LA were 9.1 ± 4.4, 8.8 ± 4.0, 6.3 ± 3.6, 4.2 ± 2.1 and 2.8 ± 1.9 ml, respectively. Regional PF volume at the roof LA and anterior LA were significantly larger than those at the lateral LA, septal LA and posterior LA, respectively (Fig. 5A). The CFAE area over the whole LA (total CFAE area) was 22.8 ± 8.4 cm2. The CFAE areas at the roof LA, anterior LA, lateral LA, septal LA and posterior LA were 8.3 ± 4.5, 7.9 ± 3.8, 2.9 ± 3.2, 2.5 ± 3.2 and 1.2 ± 2.2 cm2, respectively. The local CFAE areas at the roof LA and anterior LA were significantly larger than those at the lateral LA, septal LA and posterior LA, respectively (Fig. 5B). Therefore, the regional

Fig. 4. Correlation between PF volume and CFAE area. Correlation between total cardiac PF volume and total CFAE area (panel A) and between total LA PF volume and total CFAE area (panel B). Correlations between regional PF volume and local CFAE area for the five LA areas (panel C-G). **p: by Pearson's product–moment correlation coefficient; *p: by Spearman's rank correlation coefficient.

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx Table 5 Results of multivariate linear regression analysis for the correlation between total CFAE area and baseline clinical features.

Gender, male BMI, kg/m2 Waist circumference, cm LA area, cm2 Total cardiac PF volume, ml Dyslipidemia Smoking

β (95% CI)

Standardized β

p Value

4.625 (1.461 to 7.788) Not selected Not selected Not selected 0.099 (0.072 to 0.125) Not selected Not selected

0.214 – – – 0.542 – –

0.005 – – – b0.001 – –

R2 = 0.377, ANOVA p b 0.001. Abbreviations as in Tables 1–3.

distribution of PF volume was quite similar to that of CFAE area. In addition, there was a significant correlation between the total LA PF volume and total CFAE area (r = 0.554, p b 0.001) (Fig. 4B). Furthermore, there was a significant correlation between the regional PF volume and local CFAE area in the roof LA (r = 0.455, p b 0.001) (Fig. 4C), anterior LA (r = 0.448, p b 0.001) (Fig. 4D), lateral LA (r = 0.441, p b 0.001) (Fig. 4E), septal LA (r = 0.287, p = 0.001) (Fig. 4F) and posterior LA (r = 0.273, p = 0.003) (Fig. 4G).

7

area. These findings suggest that PF is directly related to the genesis of CFAE and may promote the pathogenic process of AF. 4.2. Atrial fibrillation and pericardial fat A close relationship between AF and PF in human has been shown recently [5–7]. Al Chekakie et al. [5] reported that AF patients had a larger PF volume compared with SR patients, and patients with persistent AF had a larger PF volume compared to patients with paroxysmal AF. Wong et al. [6] also reported that PF is associated with the presence of AF. Furthermore, they showed that PF is associated with the severity of AF and poorer outcomes after AF ablation [6]. In addition, Shin et al. [7] also showed a close relation between PF and AF. Since left atrial volume was independently associated with PF in AF patients, they suggested that PF is closely associated with left atrial remodeling in AF patients [7]. In the present study, multiple logistic regression analysis identified PF volume as an independent and significant factor associated with AF, in agreement with the results of recent studies [5,6]. Furthermore, our results showed that PF volume was also independently associated with the persistence of AF from paroxysmal to persistent form. These findings indicate that PF volume is closely associated not only with the onset of AF but also with the progression of AF morbidity, being consistent with the results of previous study [6].

4. Discussion

4.3. Inflammatory process of AF and PF

4.1. Major findings

Previous studies indicated that inflammation plays an important role in the pathogenesis of AF [3,4]. Nattel et al. [3] described inflammatory mediators as profibrotic molecules that accelerate fibroblast proliferation and differentiation into myofibroblasts, as well as collagen synthesis through various signaling pathways. In particular, myofibroblasts play a pivotal role in the fibrotic process by producing not only extracellular matrix proteins and protease, but also various growth factors, cytokines, and chemokines that stimulate fibroblast proliferation and differentiation, thereby providing positive feedback and perpetuating the fibrogenesis cascade [4]. These mechanisms enhance fibrosis caused by extracellular matrix accumulation, and atrial structural remodeling, which induces conduction slowing and perpetuation of AF, resulting in the development of AF. Meanwhile, Mazurek et al. [8] indicated that the above inflammatory mediators are also secreted from PF. Their analysis of patients with coronary artery disease demonstrated significantly higher expression levels of inflammatory cytokines in pericardial adipose tissue compared with their levels in the subcutaneous adipose tissue. Furthermore, Venteclef et al. [21] also demonstrated that human pericardial adipose tissue induces fibrosis of the atrial myocardium through the secretion of adipo-fibrokines. These findings indicate that PF promotes fibrosis and structural remodeling of LA through the secretion of chemokines and inflammatory cytokines, suggesting the involvement of PF in the pathogenic process of AF.

The present study demonstrated that PF volume correlates with AF. Furthermore, both PF volume and area of CFAE correlated independently with persistence of AF. In addition, PF volume correlated significantly with the area of CFAE in patients with AF. Also, there was a significant correlation between regional PF volume and local CFAE

4.4. Role of CFAE in maintenance of AF

Fig. 5. Distribution of regional PF volume and local CFAE area. Regional PF volume (panel A) and local CFAE area (panel B) at the roof LA, anterior LA, lateral LA, septal LA and posterior LA. **p: by Kruskal–Wallis tests; *p: by Steel–Dwass multiple comparisons.

With regard to the role of CFAE in the maintenance of AF, Nademanee et al. [20] postulated that CFAE represent either continuous reentry of the fibrillation waves into the same region or overlap of different wavelets entering the same region at different times. Subsequently, Yamabe et al. [10] reported that CFAE was generated by slow conduction and pivoting activation. They have also shown that wave break and wave fusion over the CFAE region sustained AF [10]. Others reported that additional CFAE ablation after pulmonary vein isolation increased the rate of long-term sinus rhythm maintenance [22,23]. The above studies indicate that the CFAE area plays an important role in the maintenance of AF. In the present study, the CFAE area was significantly larger in persistent AF than in paroxysmal AF. Furthermore, the CFAE area was independently associated with the persistence of AF from paroxysmal to

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135

8

H. Kanazawa et al. / International Journal of Cardiology xxx (2014) xxx–xxx

persistent form. Previous studies also showed that the CFAE area in persistent AF was larger than in paroxysmal AF [18,24]. Considered together, the above findings and the results of the present study suggest that the CFAE area is closely associated with persistence of AF. The larger area of CFAE in persistent AF may reflect the stabilized random wave propagation over the LA. 4.5. Genesis of CFAE and its relation to PF Previous studies indicated that increased density and length of collagenous septa reduce the conduction velocity and increase the amount of fractionation and asymmetry in the electrograms [12], suggesting that the formation of CFAE is consistently associated with regions of severe fibrosis [12]. Ashihara et al. [25] found that heterogeneous fibroblast proliferation in the myocardial sheet resulted in frequent spiral wave breakups, and bipolar electrograms recorded at the fibroblast proliferation area exhibited CFAE. Their results suggest a close relation between CFAE and the inflammatory process. On the other hand, the PF also secretes various inflammatory mediators and thus promotes the fibrotic cascade [8,21]. These results indicate that both PF and CFAE correlate with AF through a common inflammatory cascade. Although a close relationship between PF and AF has been shown in the previous studies [5–7], none of these studies have examined the relation between PF and CFAE in AF patients. In this study, sex and PF volume correlated significantly and independently with the area of CFAE in patients with AF. In particular, we found that PF volume had the strongest significant association with the CFAE area. In addition, there was a significant correlation between regional PF volume and the local CFAE area. These findings indicate that PF is closely related to the genesis of CFAE directly through the inflammatory process, and thus may promote the pathogenic process of AF. 5. Conclusions The present study demonstrated a significant correlation between PF volume and presence of AF. Furthermore, PF volume and the area of CFAE region were significantly associated with persistence of AF. In addition, PF volume correlated significantly with the area of CFAE region, suggesting that PF is directly related with the genesis of CFAE, resulting in the development of AF. References [1] Chung MK, Martin DO, Sprecher D, et al. C-reactive protein elevation in patients with atrial arrhythmias: inflammatory mechanisms and persistence of atrial fibrillation. Circulation 2001;104:2886–91. [2] Aviles RJ, Martin DO, Apperson-Hansen C, et al. Inflammation as a risk factor for atrial fibrillation. Circulation 2003;108:3006–10. [3] Nattel S, Burstein B, Dobrev D. Atrial remodeling and atrial fibrillation: mechanisms and implications. Circ Arrhythm Electrophysiol 2008;1:62–73.

[4] Yue L, Xie J, Nattel S. Molecular determinants of cardiac fibroblast electrical function and therapeutic implications for atrial fibrillation. Cardiovasc Res 2011;89: 744–53. [5] Al Chekakie MO, Welles CC, Metoyer R, et al. Pericardial fat is independently associated with human atrial fibrillation. J Am Coll Cardiol 2010;56:784–8. [6] Wong CX, Abed HS, Molaee P, et al. Pericardial fat is associated with atrial fibrillation severity and ablation outcome. J Am Coll Cardiol 2011;57:1745–51. [7] Shin SY, Yong HS, Lim HE, et al. Total and interatrial epicardial adipose tissues are independently associated with left atrial remodeling in patients with atrial fibrillation. J Cardiovasc Electrophysiol 2011;22:647–55. [8] Mazurek T, Zhang L, Zalewski A, et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003;108:2460–6. [9] Sacks HS, Fain JN. Human epicardial adipose tissue: a review. Am Heart J 2007; 153:907–17. [10] Yamabe H, Morihisa K, Tanaka Y, et al. Mechanisms of the maintenance of atrial fibrillation: role of the complex fractionated atrial electrogram assessed by noncontact mapping. Heart Rhythm 2009;6:1120–8. [11] Yamabe H, Morihisa K, Koyama J, Enomoto K, Kanazawa H, Ogawa H. Analysis of the mechanisms initiating random wave propagation at the onset of atrial fibrillation using noncontact mapping: role of complex fractionated electrogram region. Heart Rhythm 2011;8:1228–36. [12] Jacquemet V, Henriquez CS. Genesis of complex fractionated atrial electrograms in zones of slow conduction: a computer model of microfibrosis. Heart Rhythm 2009;6:803–10. [13] Liu X, Shi HF, Tan HW, Wang XH, Zhou L, Gu JN. Decreased connexin 43 and increased fibrosis in atrial regions susceptible to complex fractionated atrial electrograms. Cardiology 2009;114:22–9. [14] 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. [15] Konishi M, Sugiyama S, Sugamura K, et al. Association of pericardial fat accumulation rather than abdominal obesity with coronary atherosclerotic plaque formation in patients with suspected coronary artery disease. Atherosclerosis 2010;209:573–8. [16] Konishi M, Sugiyama S, Sato Y, et al. Pericardial fat inflammation correlates with coronary artery disease. Atherosclerosis 2010;213:649–55. [17] Rosito GA, Massaro JM, Hoffmann U, et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a communitybased sample: the Framingham Heart Study. Circulation 2008;117:605–13. [18] Jadidi AS, Duncan E, Miyazaki S, et al. Functional nature of electrogram fractionation demonstrated by left atrial high-density mapping. Circ Arrhythm Electrophysiol 2012;5:32–42. [19] Verma A, Novak P, Macle L, et al. A prospective, multicenter evaluation of ablating complex fractionated electrograms (CFEs) during atrial fibrillation (AF) identified by an automated mapping algorithm: acute effects on AF and efficacy as an adjuvant strategy. Heart Rhythm 2008;5:198–205. [20] Nademanee K, McKenzie J, Kosar E, et al. A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate. J Am Coll Cardiol 2004;43:2044–53. [21] Venteclef N, Guglielmi V, Balse E, et al. Human epicardial adipose tissue induces fibrosis of the atrial myocardium through the secretion of adipo-fibrokines. Eur Heart J Mar 22 2013. http://dx.doi.org/10.1093/eurheartj/eht099. [22] Verma A, Mantovan R, Macle L, et al. Substrate and trigger ablation for reduction of atrial fibrillation (STAR AF): a randomized, multicentre, international trial. Eur Heart J 2010;31:1344–56. [23] Li WJ, Bai YY, Zhang HY, et al. Additional ablation of complex fractionated atrial electrograms after pulmonary vein isolation in patients with atrial fibrillation: a meta-analysis. Circ Arrhythm Electrophysiol 2011;4:143–8. [24] Solheim E, Off MK, Hoff PI, Schuster P, Ohm OJ, Chen J. Characteristics and distribution of complex fractionated atrial electrograms in patients with paroxysmal and persistent atrial fibrillation. J Interv Card Electrophysiol 2010;28:87–93. [25] Ashihara T, Haraguchi R, Nakazawa K, et al. The role of fibroblasts in complex fractionated electrograms during persistent/permanent atrial fibrillation: implications for electrogram-based catheter ablation. Circ Res 2012;110:275–84.

Please cite this article as: Kanazawa H, et al, Importance of pericardial fat in the formation of complex fractionated atrial electrogram region in atrial fibrillation, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.04.135