2016 年 80 巻 8 号 p. 1726-1733
Background: Pericardial fat is correlated with the occurrence of atrial fibrillation or coronary atherosclerosis. However, the role of pericardial fat in ventricular arrhythmia remains unclear.
Methods and Results: Patients who had undergone dual-source computed tomography and 24-h Holter ECG were retrospectively enrolled. Quantification of the volume of pericardial fat surrounding the ventricles was analyzed using threshold attenuation of dual-source CT. The volume of pericardial fat was significantly different among those without ventricular premature beats (VPBs) in 24 h (n=28), those with occasional VPBs (n=54) and those with frequent VPBs (n=34) (12.5±6.1 cm3 vs. 14±8.9 cm3 vs. 29.9±17.3 cm3, P<0.001). In addition, the number of VPBs strongly correlated with the volume of total pericardial fat (R=0.501, P<0.001), right ventricular (RV) pericardial fat (R=0.539, P<0.001), and left ventricular pericardial fat (R=0.376, P<0.001). Multivariate logistic regression analysis showed that quartiles of RV localized pericardial fat significantly increased the risk of frequent VPBs (OR=3.2, P=0.047). Moreover, the number of VPBs in 24 h was significantly different among the patients with a fat volume within the 25th percentile, 25–75th percentile and 75th percentile.
Conclusions: Pericardial fat (especially RV pericardial fat) was associated with the frequency of VPBs, which suggests the arrhythmogenic potential of ventricular pericardial fat. (Circ J 2016; 80: 1726–1733)
Obesity is an important risk factor for death1 through increases in the incidence of heart failure, coronary artery disease, arrhythmia, and ventricular dysfunction.2,3 In addition, an increase in body mass index (BMI) is associated with a higher prevalence of new onset or sustained atrial fibrillation (AF).4,5 The high incidence of cardiovascular comorbidities and poor prognosis of patients with obesity have been suggested to arise from the adverse cardiovascular and pro-arrhythmic effects of adipose tissue surrounding either the epicardium or pericardium.6 Pericardial adipose tissue is the visceral fat deposit around the heart, and it consists of a visceral epicardial fat layer and parietal paracardial fat layer. Epicardial fat is the adipose tissue located between the myocardium and visceral pericardium, and the adipose tissue layer surrounding the parietal pericardium is called the paracardial fat layer.7
Clinical evidence indicates that obesity is associated with higher levels of high-sensitivity C-reactive protein (hsCRP),8 and the pericardial adipose tissue has been shown to release pro-inflammatory cytokines and mediators, which have been implicated in the pathogenesis of cardiovascular events,9 especially in patients with heart failure.10 Previous reports have also shown that ischemic patients have more pericardial fat11 and thicker peri-coronary artery adipose tissue.12 The volume of pericardial fat has also been positively correlated with traditional cardiovascular risk factors, and it potentially acts as a pro-inflammatory mediator in the pathogenesis of atherosclerosis.13–15 Moreover, patients with AF have been shown to have more and thicker peri-atrial adipose tissue than matched healthy controls.16 Furthermore, adipocytes in peri-atrial adipose tissue can release adipocytokines, and the interactions of adipocytokines and adipocytes-cardiomyocytes can produce cardiac electrical and structural modeling and induce arrhythmogenicity.
In patients with systolic heart failure, increasing pericardial fat has been associated with a higher incidence of ventricular tachyarrhythmia.17 In addition, a previous study showed that each 1 kg/m2 increase in BMI was associated with a significant 4% increased adjusted risk for exercise-induced ventricular arrhythmia.18,19 Although pericardial adiposity plays an important role in the genesis of atrial arrhythmia, it is not clear whether periventricular adipose tissue has an arrhythmogenic effect on the ventricles, or whether the relationship between obesity and ventricular arrhythmia is related to the arrhythmogenetic potential of periventricular adipose tissue. Pericardial fat can be assessed and well quantified by computed tomography (CT), and the adipose tissue surrounding the coronary arteries and pericardium as quantified by CT has been proven to correlate with the occurrence of coronary atherosclerosis and AF.12,16 Compared with conventional CT, dual-source CT can provide greater accuracy, lower radiation dose, better quality results for integrative analysis of coronary arteries, myocardial structure and adipose tissue differentiation.20 Therefore, the aim of the present study was to investigate whether periventricular adipose tissue detected by dual-source CT can modulate the occurrence of ventricular arrhythmia.
This study received Institutional Review Board approval and enrolled 116 individuals (79 men, 37 women; mean age 63±11 years) who had undergone 24-h Holter ECG monitoring and dual-source CT for suspected coronary artery disease or general checkup. The medical history, medication usage, weight, height, blood pressure, and laboratory assessments of each patient were recorded. Patients with episodes of AF during CT or Holter recording were excluded.
24-h Holter ECG Monitoring24-h Holter ECGs were recorded using a Philips Zymed Holter 1810 series system. The recordings were analyzed and confirmed for each patient by 2 cardiologists, and then the morphology, frequency, and pattern of ventricular premature beats (VPBs) were analyzed. The morphology of the VPBs was categorized into 3 groups: left bundle branch block (LBBB)-like VPBs, right bundle branch block (RBBB)-like VPBs and undetermined VPBs. The frequencies of the VPBs were also classified into 3 groups: frequent (>240 beats/24 h), occasional (≤240 and >0 beats/24 h), and none.
Dual-Source CTCT scanning was performed on a dual-source 128-slice CT scanner (Somatom Definition Flash, Siemens Medical Solutions, Germany) using 2 X-ray sources to generate the images. The scanning volume was confined to the area from the carina to the diaphragm. A gantry rotation time of 0.28 s resulted in a temporal resolution of 75 ms. We monitored the CT findings of the area of interest at the root of the ascending aorta as soon as the contrast medium had been injected.
Agatston ScoreThe Agatston score was evaluated by radiologists using a dedicated workstation for each coronary vessel in each patient. Coronary calcium was defined as an area of the coronary artery with an intensity >130 Hounsfield units (HU).
Periventricular FatPeriventricular fat was defined as any adipose tissue located within the visceral and parietal surfaces of the pericardium surrounding the left and right ventricles (LV, RV). The volume of periventricular fat was measured in cubic centimeters (cm3) using volume analysis software on a dedicated workstation by manually tracing the border of the pericardium once in every fourth 2.5-mm thick axial slice (Figures 1A,B). Right periventricular fat was defined as that bordering the upper level of the pulmonary valve to the lower level of the diaphragm. The adipose tissue surrounding both atria was ignored. We excluded the atrial pericardial fat identified as adipose tissue above visualization of the tricuspid and mitral valves. The thicker rim of the ventricular myocardium and trabeculation of intraventricular cavity were the markers separating atria and ventricles. Total periventricular fat was defined as the sum of the left and right periventricular fat. The number of slices that had to be traced manually ranged from 10 to 20 in each patient (Figure 1B), according to the size of the heart. Coronal and sagittal views were used to help modify the accuracy of manual tracing if necessary. The computer then automatically reconstructed a 3D image of the RV and LV according to the manual tracing of axial slices. Fat voxels were identified using threshold attenuation values of −30 to −190 HU (Figures 1C,D).21 Pericardial fat within the traced and selected volume was then automatically quantified by the software in cubic centimeters (Figure 1E). The manual pericardial fat tracing was blinded to the patient’s characteristics.
Pericardial fat volume identified by dual-source CT. (A,B) Manual tracing of the pericardium border once in every fourth 2.5-mm thick axial slice. (C,D) Pericardial fat around the right ventricle was identified by threshold attenuation values of −30 to −190 HU. (E) Three-dimensional images of pericardial fat for measurement of volume using computer software. HU, Hounsfield units.
Results are expressed as frequencies (per cent) or mean±standard deviation. We used the Kruskal-Wallis test to determine differences in continuous variables between groups. Post-hoc analysis was preformed using the Mann-Whitney U test, and one-way ANOVA was used if continuous variables were normally distributed. Differences in the proportions of categories were compared using the Pearson chi-square test. Correlations between VPBs and volume of pericardial fat were computed using bivariate analysis with Spearman rank correlation. Multivariate logistic regression was used to determine the predictors of frequent VPBs from different quartiles and location of pericardial fat and other clinical variables. A P value <0.05 was considered to be statistically significant. All statistical analyses and calculations were performed using SPSS version 19 (Chicago, IL, USA).
Table 1 summarizes the clinical characteristics and laboratory data of the different groups. More male patients had frequent VPBs and they had a higher incidence of hypertension, coronary artery disease, concomitant use of β-blockers, a larger BMI, LV volume, RV volume, and higher Agatston score than the patients without VPBs or with occasional VPBs. The patients with frequent VPBs also had significantly larger volumes of total, RV pericardial fat, and LV pericardial fat than those with occasional or no VPBs. In the patients (41%) with available hsCRP levels, there were similar values (0.24±0.13, 0.20±0.13, and 0.29±0.13 mg/dl, P>0.05) in the patients with no VPB (n=10), occasional VPBs (n=22) and frequent VPBs (n=15).
| VPBs=0 (n=28) |
0<VPBs≤240 (n=54) |
VPBs >240 (n=34) |
P value | |
|---|---|---|---|---|
| Age, years | 59±10 | 64±11 | 63±12 | 0.14 |
| Male, % | 46 | 65a | 91a,b | 0.001 |
| Hypertension, % | 32 | 72a | 89b | 0.001 |
| Chronic kidney disease, % | 7 | 9 | 11 | 0.82 |
| Type 2 diabetes mellitus, % | 7 | 18 | 17 | 0.37 |
| Dyslipidemia, % | 35 | 50 | 56 | 0.27 |
| Coronary artery disease, % | 25 | 59a | 56a | 0.009 |
| Congestive heart failure, % | 4 | 9 | 15 | 0.33 |
| Hypo- or hyperthyroidism, % | 4 | 4 | 3 | 0.98 |
| Smoking, % | 21 | 28 | 33 | 0.65 |
| β-blocker, % | 14 | 33 | 50a,b | 0.01 |
| Nondihydropyridine CCB, % | 10 | 7 | 11 | 0.79 |
| Dihydropyridine CCB, % | 14 | 35 | 21 | 0.09 |
| ACEI/ARB, % | 14 | 30 | 26 | 0.08 |
| Class IC AAD, % | 0 | 13 | 6 | 0.1 |
| Class III AAD, % | 3 | 0 | 5 | 0.4 |
| Statin, % | 35 | 51 | 53 | 0.31 |
| BMI, kg/m2 | 24±3 | 24±3 | 26±4a,b | 0.003 |
| LV ejection fraction, % | 71±6 | 71±6 | 69±8 | 0.119 |
| Total cholesterol, mg/dl | 181±32 | 172±36 | 173±47 | 0.6 |
| LDL, mg/dl | 107±29 | 96±27 | 97±35 | 0.266 |
| HDL, mg/dl | 48±12 | 50±14 | 51±14 | 0.753 |
| Triglyceride, mg/dl | 119±75 | 151±96 | 129±90 | 0.976 |
| Agatston score | 49±129 | 220±601a | 268±664b | 0.02 |
| RV volume, cm3 | 136±39 | 154±40a | 174±51a | 0.003 |
| RV free wall thickness, cm | 4.3±1.1 | 4.0±0.7 | 4.2±0.9 | 0.23 |
| LV volume, cm3 | 195±45 | 217±47 | 236±64a | 0.02 |
| LV free wall thickness, cm | 10.3±1.4 | 10.9±1.6 | 11.2±2.2 | 0.12 |
| Total pericardial fat, cm3 | 12.5±6.1 | 14±8.9 | 29.9±17.3a,b | <0.001 |
| RV pericardial fat, cm3 | 9.1±3.8 | 10.1±5.9 | 21.7±12.5a,b | <0.001 |
| LV pericardial fat, cm3 | 3.3±2.6 | 3.7±3.6 | 8.1±5.9a,b | <0.001 |
aP<0.05 compared with 0=VPBs, bP<0.05 compared with 0<VPBs≤240. AAD, anti-arrhythmic drug; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; BMI, body mass index; CCB, calcium-channel blocker; HDL, high-density lipoprotein; LBBB, left bundle branch block; LDL, low-density lipoprotein; LV, left ventricle; RBBB, right bundle branch block; VPB, ventricular premature beats.
We correlated the number of VPBs with pericardial fat (Figure 2), and found that the number of VPBs correlated well with total, RV and LV pericardial fat (Figures 2A–C). In the subgroup analysis, the number of VPBs correlated significantly with the volumes of total pericardial fat and RV pericardial fat in the patients with frequent VPBs (Figures 2G–I), especially RV pericardial fat, based on the highest R value to the correlation of RV pericardial fat and number of VPBs. We also evaluated the relationship between the morphology of VPBs and distribution of pericardial fat. In the patients with RBBB-like VPBs, the number of VPBs correlated well with total, RV and LV pericardial fat, with similar R values to the correlation with RV or total pericardial fat (Figure 3A). In the patients with LBBB-like VPBs, the number of VPBs correlated well with total, RV and LV pericardial fat, with the highest R value in the correlation with RV pericardial fat (Figure 3B).
Correlation between the number of ventricular premature beats (VPBs) and volume of pericardial fat. (A) Correlation between total VPBs and periventricular fat. (B) Correlation between total VPBs and RV pericardial fat. (C) Correlation between total VPBs and LV pericardial fat. (D) Correlation between total VPBs and pericardial fat in subgroup of occasional VPBs. (E) Correlation between total VPBs and RV pericardial fat in subgroup of occasional VPBs. (F) Correlation between total VPBs and LV pericardial fat in subgroup of occasional VPBs. (G) Correlation between total VPBs and pericardial fat in subgroup of frequent VPBs. (H) Correlation between total VPBs and RV pericardial fat in subgroup of frequent VPBs. (I) Correlation between total VPBs and LV pericardial fat in subgroup of frequent VPBs.
Correlation between the number of RBBB- or LBBB-like ventricular premature beats (VPBs) and volume of pericardial fat. Correlation between (A) RBBB-like VPBs and (B) LBBB-like VPBs with total (Right), RV (Middle) and LV (Left) pericardial fat around the ventricles. LBBB/RBBB, left/right bundle branch block.
Table 2 shows the clinical profiles of the patients with different volumes of pericardial fat; the differences in BMI, smokers and triglycerides were significant between the groups. The numbers of total VPBs, RBBB-like VPBs, LBBB-like VPBs were also significantly different between the groups categorized by percentiles. Total VPBs, RBBB-like VPBs and LBBB-like VPBs were most frequent and prevalent in the 75th percentile group of pericardial fat. The incidence of couplet VPBs and short run/non-sustained ventricular tachycardia was similar in the patients with different volumes of pericardial fat. There were significantly different hsCRP levels (0.15±0.10, 0.26±0.12, and 0.28±0.14 mg/dl, P=0.03) in the patients with 25th percentile (n=11), 25–75th percentile (n=21), and 75th percentile (n=15) of pericardial fat. The patients with pericardial fat in the 75th percentile had higher hsCRP than the patients with pericardial fat in the 25th percentile (P=0.04), but the hsCRP level was not significantly different in the patients with 25th percentile and 25–75th percentile pericardial fat (P=0.09). In addition, BMI correlated well with total pericardial fat (Figure 4).
| 25th percentile (n=29) |
25th–75th percentile (n=58) |
75th percentile (n=29) |
P value | |
|---|---|---|---|---|
| Age, years | 59±11 | 62±9 | 67±14 | 0.186 |
| Male, % | 41 | 34 | 17 | 0.120 |
| Hypertension, % | 62 | 52 | 55 | 0.658 |
| Chronic kidney disease, % | 3 | 10 | 13 | 0.385 |
| Type 2 diabetes mellitus, % | 13 | 12 | 24 | 0.327 |
| Dyslipidemia, % | 34 | 58 | 51 | 0.227 |
| Coronary artery disease, % | 52 | 50 | 48 | 0.966 |
| Hypo or hyper-thyroidism, % | 3 | 5 | 0 | 0.46 |
| Congestive heart failure, % | 3 | 10 | 13 | 0.405 |
| Smoking, % | 34 | 15a | 45a | 0.01 |
| β-blocker, % | 24 | 34 | 41 | 0.373 |
| Nondihydropyridine CCB, % | 10 | 7 | 13 | 0.592 |
| Dihydropyridine CCB, % | 31 | 24 | 24 | 0.764 |
| ACEI/ARB, % | 48 | 29 | 31 | 0.194 |
| Class IC AAD, % | 7 | 7 | 10 | 0.835 |
| Class III AAD, % | 10 | 5 | 7 | 0.668 |
| Statin, % | 41 | 48 | 55 | 0.576 |
| BMI, kg/m2 | 23±3 | 25±3a | 26±3a | <0.001 |
| LV ejection fraction, % | 72±5 | 70±6 | 68±9 | 0.219 |
| Total cholesterol, mg/dl | 162±31 | 178±34 | 179±51 | 0.129 |
| LDL, mg/dl | 89±25 | 103±30 | 101±36 | 0.139 |
| HDL, mg/dl | 52±12 | 50±15 | 49±12 | 0.567 |
| Triglyceride, mg/dl | 94±50 | 130±77a | 152±107a | 0.014 |
| Agatston score | 283±713 | 138±379 | 211±656 | 0.507 |
| Total VPBs | 72±226 | 186±363a | 1,250±1,837a,b | <0.001 |
| Couplet VPBs | 1±5 | 0.19±1.2 | 0.76±2.2 | 0.412 |
| Short run/non-sustained VT | 0 | 0.03±0.18 | 0.1±0.3 | 0.14 |
| LBBB-like VPBs | 8±22 | 100±235 | 352±610a,b | 0.008 |
| RBBB-like VPBs | 64±227 | 87±292 | 896±1,835a,b | 0.01 |
aP<0.05 compared with 25th percentile, bP<0.05 compared with 25th–75th percentile. Abbreviations as in Table 1.
Correlation between body mass index (BMI) and volume of pericardial fat.
The univariate logistic regression analysis showed that male sex (odds ratio [OR]=5.62, 95% confidence interval [CI]=1.81–17.4), BMI (OR=1.22, 95% CI=1.07–1.39), β-blocker use (OR=0.42, 95% CI=0.19–0.96), LV volume (OR=1.01, 95% CI=1.00–1.02), RV volume (OR=1.01, 95% CI=1.00–1.02), quartile of total pericardial fat (OR=3.34, 95% CI=2.04–5.48), quartile of RV localized pericardial fat (OR=4.13, 95% CI=2.4–7.13), and quartile of LV localized pericardial fat (OR=2.53, 95% CI=1.63–3.91) were associated with a higher risk of frequent VPBs. Moreover, multivariate logistic regression including significant clinical variables and quartiles of pericardial fat showed that only the quartile of localized RV pericardial fat was an independent predictor of frequent VPBs (OR=3.2, 95% CI=1.01–10, P=0.047). Quartile of total pericardial fat, quartile of LV localized pericardial fat and other clinical variables were insignificant predictors.
Pericardial fat has both paracrine and systemic effects. Adipocytokines such as leptin, resistin, interleukins 6 and 17, and tumor necrosis factor-α can induce adventitial neovascularization and accelerate the development of coronary atherosclerosis.22 Adipocytokines released from adipose tissues have pro-inflammatory activity,23 which may produce cardiac electrical abnormalities. Previous study showed that cardiac adipocytokines significantly decreased the membrane potential of delayed rectifier K+ outward currents.24 Some distinctive electrophysiological effects of pericardial fat could be considered for the pathogenesis of ventricular arrhythmia. In this study, we found that pericardial fat was significantly associated with the occurrence of VPBs. Moreover, RV pericardial fat, rather than total pericardial fat, LV pericardial fat, Agatston score, BMI or other clinical variables was a potential predictor of frequent VPBs, suggesting a local arrhythmogenic effect of RV pericardial fat. The patients with more pericardial fat had higher levels of hsCRP, which supported the potential inflammatory effect of periventricular adipose tissue.
The incidence and frequency of VPBs were associated more with RV pericardial fat than LV pericardial fat, total pericardial fat, BMI or Agatston score. The correlation between localized RV pericardial fat and VPBs (including RBBB- and LBBB-like VPBs) remained significant even after adjusting for other clinical factors. These findings suggested a site-specific influence of RV pericardial fat on the development of ventricular arrhythmia. The RV outflow tract is the most common site where ventricular arrhythmia is induced, and it contains distinctive cardiomyocytes with a high arrhythmogenic potential. The myocardium in the RV (and especially the RV outflow tract) is quite thin, and has a propensity to be affected by cytokines from local adipose tissues because both the volume and amount of RV pericardial fat are greater than in the LV.25 Moreover, the bioactive signaling inflammatory molecules that can induce systemic toxic effects can penetrate the RV more easily than the LV. On the other hand, fat depots such as pericardial fat may be the result of compensation of energy supply to the myocardium. Pericardial fat has the ability to fulfill the energy needs of arterial walls and heart muscles while avoiding lipotoxicity. Arrhythmogenicity requiring energy may be partially supplied by compensated localized pericardial fat accumulation. In patients with ventricular tachyarrhythmia, an arrhythmogenic substrate in the RV is more common than in the LV, so the effect of RV pericardial fat was more significant than that of LV pericardial fat and total pericardial fat in the current study. Although only a weak correlation between the morphology of the VPBs and the distribution of pericardial fat was shown in our study, the existence of a localized toxic effect from pericardial fat cannot be excluded because of the small sample size. LBBB- or RBBB-like VPBs may still be induced by the effect of localized RV or LV pericardial fat, respectively. The morphology of VPBs is determined by the origin of the abnormal electrical ventricular myocytes. In this study, we found that both RBBB- and LBBB-like VPBs correlated with total, RV and LV pericardial fat, which suggests that pericardial fat may have effects beyond local release. However, the different levels of correlation suggest that RV pericardial fat may have a stronger arrhythmogenic potential. In this study, we also found that the incidence of short run and non-sustained ventricular tachycardia showed a trend to be significantly different. Previous studies showed that the incidence of VPBs was associated with cardiovascular events and all-cause death.26,27 Furthermore, another study reported that these risks may be increased in patients with concurrent frequent VPBs and couplet VPBs as recorded by 48-h Holter ECG monitoring.28
Previous studies have reported that ventricular hypertrophy is associated with the incidence of ventricular arrhythmia, especially in patients with hypertrophic cardiomyopathy,29 arrhythmogenic RV cardiomyopathy30 and congestive heart failure.31 This arrhythmogenic substrate may be caused by structural and electrical remodeling.32,33 Although pericardial fat is a risk factor for changes in the thickness and function of cardiac muscles, no significant differences in LV or RV wall thickness were found between the VPB groups in this study (Table 1). This may be related to the distinctive population in this study, because most of the study individuals were relatively healthy. Therefore, the potential arrhythmogenesis of structural ventricular remodeling may not have played a significant role in the cause or comorbidity of ventricular arrhythmia in the current study. Long-term follow-up studies have found that frequent VPBs can be a risk factor for heart failure through the potential mechanism of asynchronous ventricular contraction or tachycardia-induced cardiomyopathy.34 A higher frequency of VPBs in patients with high volumes of pericardial fat may contribute to the known high rate of comorbid heart failure in these cases.17 However, there was no correlation between the number of VPBs and pericardial fat in the patients with occasional VPBs in this study (Figures 2D–F), which suggested that only a small amount of pericardial fat may exist in normal individuals. Thus, patients with occasional VPBs are relatively healthy.
Reducing body weight can help to reduce the volume and thickness of pericardial fat, and preliminary studies have indicated that substantial weight loss in severely obese patients following bariatric surgery or marked calorie intake reduction may be accompanied by a corresponding decrease in pericardial fat volume.35,36 Nakazato et al reported that a reduction in total body weight (>5%) was associated with stabilization or reduction in epicardial fat burden, whereas weight gain led to obvious epicardial fat progression according to CT measurements.37 Interestingly, in a clinical study investigating epicardial fat volume using MRI, a reduction in epicardial fat volumes did not correlate with a reduction in visceral abdominal fat or BMI, suggesting specific effects of weight loss on different visceral fat depots.38 In the current study, we found a significant association between pericardial fat and ventricular arrhythmia. In this study, pericardial fat was well correlated with BMI, which suggests that overweight or obese patients may accumulate more periventricular fat. Previous reports showed that pericardial fat influenced the pathogenesis of coronary atherosclerosis by promoting chronic inflammation and endothelial dysfunction.23 In addition to the local inflammatory effects of pericardial fat, ventricular arrhythmia or ventricular ectopy may also be induced by ventricular remodeling, myocardial ischemia or heart failure. The similar circulatory levels of hsCRP in the patients with different VPBs suggested that the inflammatory effects of periventricular adipose tissue may not fully account for the pathophysiology of ventricular arrhythmogenesis. Accordingly, it is expected that weight loss causing a reduction in adipose tissue around the ventricles may decrease the incidence of ventricular tachyarrhythmia.
Study LimitationsFirst, the study was retrospective in design, and the study population was relatively small. Second, the study populations were predominantly Asian, which may limit the applicability of the results to different ethnicities. Third, we cannot exclude the possibility that atrial pericardial fat was incompletely separated from the periventricular adipose tissue by manual tracing. Moreover, single ambulatory Holter monitoring may have a potential uncertainty in assessing the VPB burden because of the limited recording time, although this method is widely used in studies of VPBs with high reproducibility.27,28,39 Finally, the patients were recruited from a hospital rather than from the community, and it is unclear whether our findings can be applied to the general population.
We demonstrated that localized RV pericardial fat was independently associated with the frequency of ventricular arrhythmia. Systemic pro-arrhythmic effects induced by RV pericardial fat may play an important role in the development of ventricular arrhythmia. Further studies are necessary to investigate the cause-effect relationship and the possible mechanisms of pericardial fat-related VPBs.
This study was supported by grants from the Ministry of Science and Technology (MOST103-2314-B-038-041-MY2, and MOST104-2314-B-038-071-MY3), Taipei Medical University-Wan Fang Hospital (104swf02, 104-wf-eva-01, 104-wf-eva-06, 104-wf-eva-14, and 105-wf-eva-08).
None.
