Epicardial Adipose Tissue Accumulation Is Associated With Renal Dysfunction and Coronary Plaque Morphology on Multidetector Computed Tomography

Abstract

Background: Chronic kidney disease (CKD) is strongly associated with coronary artery disease (CAD), although the underlying pathophysiological mechanism remains unclear. Epicardial adipose tissue (EAT) has recently been recognized as an important source of various pro-inflammatory cytokines causing coronary atherosclerosis. This study investigated the relationship between CKD and EAT volume in association with high-risk plaque.

Methods and Results: The study included 275 patients with an estimated glomerular filtration rate (eGFR) ≥30 ml/min/1.73 m² who underwent multidetector computed tomography (MDCT) for the evaluation of CAD. Patients were classified, according to eGFR, into a CKD group (30≤eGFR<60 ml/min/1.73 m²) or a non-CKD group (eGFR ≥60 ml/min/1.73 m²). MDCT was used to assess coronary plaque morphology and EAT volume. One hundred and ten patients with CKD were more likely to be older, have higher prevalence of hypertension, lower serum HDL-C, higher serum CRP, and larger EAT volume, than those without CKD (all P<0.01). On multivariate analysis age, hypertension, and EAT volume were significantly associated with eGFR (all P<0.01). EAT volume was associated with the presence of high-risk plaque, independent of traditional CAD risk factors (P=0.003).

Conclusions: Patients with CKD had significantly increased EAT volume, which could be associated with the presence of high-risk plaque. (Circ J 2016; 80: 196–201)

Chronic kidney disease (CKD) is a global health concern,¹ and is an independent risk factor for coronary artery disease (CAD).^2–4 Even the early stage of CKD is associated with a worse cardiovascular prognosis in patients with and without known CAD.^5,6 The underlying mechanism, however, has not been fully elucidated. Epicardial adipose tissue (EAT) is a visceral adipose tissue located within the pericardial sac with close proximity to the coronary arteries. EAT accumulation is strongly associated with the severity and development of CAD, independent of systemic obesity and inflammation.^7–10 Multidetector computed tomography (MDCT) can now be used to accurately assess coronary atherosclerosis as well as to measure EAT volume.^{7–9,11–14} We hypothesized that EAT accumulation may be associated with coronary plaque morphology in patients with CKD. The purpose of this study was to investigate the relationship between renal function and MDCT-estimated EAT volume in association with the presence of high-risk plaque.

Editorial p 64

Methods

Subjects

We enrolled 275 patients (181 men; mean age, 65±10 years) who underwent MDCT for the evaluation of CAD. The exclusion criteria were history of open heart surgery or coronary revascularization, severe renal dysfunction (estimated glomerular filtration rate [eGFR] <30 ml/min/1.73 m², and poor MDCT image quality. One hundred and eighty-two patients (66%) were referred for MDCT for chest pain of varying degree. The remaining asymptomatic patients (34%) were referred for evaluation of CAD in the setting of multiple CAD risk factors, peripheral arterial disease, or cerebrovascular disease, or abnormal electrocardiogram (ECG) and echocardiography. Patients were classified, according to eGFR, into a CKD group (30≤eGFR<60 ml/min/1.73 m²) or a non-CKD group (eGFR ≥60 ml/min/1.73 m²).

The following factors were noted for each patient: hypertension (blood pressure ≥140/90 mmHg on repeated measurements or taking antihypertensive agents), hypercholesterolemia (serum total cholesterol ≥200 mg/dl, or statin treatment), diabetes (fasting plasma glucose >126 mg/dl, taking hypoglycemic drugs or insulin, or a combination of the two), and current smoking. Body mass index (BMI) was calculated using height and weight (kg/m²). Biochemistry was assessed using blood samples taken before MDCT. For the evaluation of renal function, we assessed eGFR using the Modification of Diet in Renal Disease equation.¹⁵ Serum low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald equation.¹⁶ In the case of triglycerides ≥400 mg/dl, serum LDL-C was measured using direct homogeneous assay (Sekisui Medical).¹⁷ The study was approved by the Institutional Review Board of Osaka Ekisaikai Hospital, and written informed consent was obtained before MDCT.

Scan Protocol and Image Reconstruction

All MDCT examinations were conducted using Somatom Sensation 64 (Siemens Medical System, Forchheim, Germany). The scan parameters were as follows: 64×0.6 mm collimation, tube voltage 120 kV, gantry rotation time 330 ms, and tube current 770–850 mAs. Each subject took 5 mg bisoprolol 2 h before MDCT if heart rate was >60 beats/min. Nitroglycerin spray is used immediately before MDCT to improve visualization of smaller-caliber coronary vessels through vasodilation. For the contrast-enhanced scans, 40–70 ml of non-ionic contrast agent (Omnipaque 350, Daiichi Sankyo, Tokyo, Japan) was injected i.v. at a flow rate of 4.0–5.5 ml/s followed by 30 ml of saline. All scans were performed during a single breath hold. The raw data were reconstructed using algorithms optimized for retrograde ECG-gated multislice spiral reconstruction. MDCT analysis was performed by experienced physicians blinded to other information.

CT Angiography Interpretation and EAT Measurement

MDCT parameters were assessed using SYNAPSE VINCENT (Fujifilm Medical, Tokyo, Japan). The coronary arteries were divided into 16 separate segments ≥1.5 mm in diameter as measured on MDCT angiogram, based on American Heart Association classification.¹⁸ Each segment was classified as interpretable or not. Patients were excluded when proximal or mid-segment or >3 segments were uninterpretable.¹⁹ In the case of plaque, plaque was also analyzed for the presence or absence of the following 2 high-risk features: positive remodeling (PR; remodeling index >1.1) and low-attenuation plaque (LAP).²⁰ LAP was defined as plaque with mean density <30 Hounsfield units (HU). High-risk plaque was defined as having PR and/or LAP.

EAT was defined as the adipose tissue located within the pericardial sac. To obtain EAT volume, there were 2 manual editing steps (steps 1 and 2) and 2 automated steps (steps 3 and 4). In step 1, 7–10 parallel and equidistant axial planes were extracted. The number of planes depended on the size of the heart. The upper slice limit was marked at the bifurcation of the pulmonary artery trunk and the lower slice limit was chosen as the last slice containing any portion of the heart. In step 2, in each plane, 8–12 control points were placed on the pericardium, and the software automatically generated a smooth closed pericardial contour as a region of interest. In step 3, within a smooth closed pericardial contour, the software automatically identified adipose tissue using the attenuation threshold –30 to –250 HU.^8,11 Finally, EAT volume was calculated as the sum of the EAT areas (step 4). Total EAT volume was calculated as the total sum of the EAT areas.

Statistical Analysis

Categorical variables are presented as frequencies, and continuous variables as mean±SD. Chi-squared test was used for comparing categorical variables and independent samples t-test or Mann-Whitney U-test for continuous variables as appropriate, according to the presence of normal distribution. Univariate and multivariate linear regression analysis was performed to investigate the relationship between eGFR and baseline variables including EAT volume. Univariate and multivariate logistic regression analysis was conducted to evaluate the impact of EAT volume on the presence of high-risk plaque. Differences were considered significant at P<0.05.

Inter-observer variability for EAT measurements was analyzed in 50 randomly selected subjects assessed by 2 independent blinded observers. Intra-observer variability was analyzed in another group of 50 subjects by the same observer at 2 different time points. The results were analyzed using both least squares fit linear regression analysis and Bland-Altman method. Statistical analysis was performed using JMP 10 (SAS Institute, Cary, NC, USA).

Results

Among the 275 patients, 110 (40%) were classified as CKD (30≤eGFR<60 ml/min/1.73 m²) and 165 (60%) as non-CKD (eGFR ≥60 ml/min/1.73 m²). Patients with CKD were more likely to be older (P<0.001), have a higher prevalence of hypertension (P<0.001), lower serum high-density lipoprotein cholesterol (HDL-C; P=0.001), and higher serum C-reactive protein (CRP; P=0.002), compared with the non-CKD patients (Table 1). Despite similar BMI, patients with CKD had significantly increased EAT volume than the non-CKD patients (111±41 vs. 81±29 ml, P<0.001). HDL-C was negatively associated with EAT volume (r=–0.25, P<0.001).

Table 1. Clinical Subject Characteristics vs. Presence of CKD

	CKD (n=110)	Non-CKD (n=165)	P-value
Age (years)	69±9	63±10	<0.001
Male	77 (70)	104 (63)	0.2
Hypertension	88 (80)	92 (56)	<0.001
Diabetes	45 (41)	60 (36)	0.4
Hypercholesterolemia	65 (59)	97 (59)	1.0
Smoking	28 (25)	48 (29)	0.5
BMI (kg/m2)	24.0±3.2	23.8±3.7	0.6
Medications
ACEI or ARB	65 (59)	69 (42)	0.005
β-blockers	23 (21)	33 (20)	0.9
CCB	44 (40)	48 (29)	0.06
Statin	53 (48)	85 (52)	0.6
Laboratory data
Glucose (mg/dl)	121±39	119±34	0.6
LDL-C (mg/dl)	114±35	117±37	0.6
HDL-C (mg/dl)	49±12	54±13	0.001
CRP (mg/L)	1.5±2.3	0.8±0.6	0.002
eGFR (ml/min/1.73 m2)	49±7	75±12	<0.001
EAT volume (ml)	111±41	81±29	<0.001

Data given as mean±SD or n (%). ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; CCB, calcium channel blockers; CKD, chronic kidney disease; CRP, C-reactive protein; EAT, epicardial adipose tissue; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

As shown in Figure A, there was a significant negative correlation between EAT volume and eGFR (r=–0.34, P<0.001), whereas no significant relationship was observed between BMI and eGFR (r=0.004, P=0.9; Figure B). On multivariate linear regression analysis age (standardized β=–0.30, P<0.001), hypertension (standardized β=–0.18, P=0.001) and EAT volume (standardized β=–0.28, P<0.001) were identified as independent factors for eGFR (Table 2). Of 275 patients, 44 patients (16%) had ≥1 plaque with PR and/or LAP. On multivariate logistic regression analysis hypercholesterolemia (P=0.02) and EAT volume (P=0.003) were significantly associated with the presence of high-risk plaque (Table 3). The multivariate logistic regression analysis was repeated including eGFR. As a result, similar results were obtained, although P-value was slightly changed: hypercholesterolemia (P=0.02) and EAT volume (P=0.01) were risk factors for high-risk plaque.

Figure.

Correlation between estimated glomerular filtration rate (eGFR) and (A) epicardial adipose tissue (EAT) volume, and (B) body mass index.

Table 2. Relationship Between eGFR and Clinical Variables

	Univariate		Multivariate
	Standardized β (95% CI)	P-value	Standardized β (95% CI)	P-value
Age (years)	−0.37 (−0.76 to −0.41)	<0.001	−0.30 (−0.66 to −0.29)	<0.001
Sex, male	0.07 (−0.81 to 3.3)	0.2	−0.01 (−2.2 to 1.8)	0.8
Hypertension	−0.27 (−6.7 to −2.8)	<0.001	−0.18 (−5.0 to −1.2)	0.001
Diabetes mellitus	−0.03 (−2.5 to 1.5)	0.6	−0.02 (−2.1 to 1.4)	0.7
Hypercholesterolemia	−0.004 (−2.0 to 1.9)	0.9	−0.03 (−2.3 to 1.2)	0.5
Smoking	0.12 (0.11 to 4.4)	0.04	0.04 (−1.4 to 2.8)	0.5
BMI (kg/m2)	0.004 (−0.54 to 0.58)	0.9	0.04 (−0.34 to 0.73)	0.5
EAT volume (ml)	−0.34 (−0.20 to −0.10)	<0.001	−0.28 (−0.17 to −0.08)	<0.001

Abbreviations as in Table 1.

Table 3. Significant Indicators of High-Risk Plaque

	Univariate		Multivariate
	OR (95% CI)	P-value	OR (95% CI)	P-value
Age (per 10 year change)	1.30 (0.93–1.85)	0.1	1.45 (098–2.23)	0.07
Sex (male)	1.41 (0.70–3.00)	0.3	1.36 (0.60–3.24)	0.5
Hypertension	3.20 (1.44–8.13)	0.003	2.14 (0.91–5.65)	0.08
Diabetes mellitus	1.20 (0.61–2.32)	0.6	1.07 (0.52–2.17)	0.8
Hypercholesterolemia	2.24 (1.11–4.86)	0.02	2.48 (1.13–5.86)	0.02
Smoking	1.70 (0.84–3.34)	0.1	1.91 (0.85–4.27)	0.1
BMI (kg/m2)	1.05 (0.96–1.15)	0.3	1.03 (0.92–1.15)	0.6
EAT volume (per 10 ml change)	1.16 (1.07–1.27)	<0.001	1.15 (1.05–1.26)	0.003

Abbreviations as in Table 1.

Excellent correlation was observed in inter- and intra-observer variability of EAT measurements (r=0.99 and r=0.99, respectively). According to Bland-Altman analysis, the 95% limits of agreement for the inter- and intra-observer variabilities in EAT volume were 1.3±9.0 ml and 0.9±7.8 ml (mean±1.96 SD, respectively).

Discussion

This study has demonstrated that CKD patients have significantly increased EAT volume compared with non-CKD patients, and that this is associated with high-risk plaque.

The worldwide rise in the number of CKD patients is threatening to reach epidemic proportions.¹ Cardiovascular events are the most common cause of mortality and morbidity in patients with CKD, rather than renal failure itself.^2–4 Furthermore, recent studies have shown that even mild renal disease should be considered a major risk factor for cardiovascular events.^5,6 The underlying pathophysiological mechanisms, however, are still not fully elucidated. Altered fat distribution is proposed to be a key contributory factor to atherosclerosis progression in patients with CKD of any degree.^21–23 Kim et al found a significant reverse correlation between visceral fat volume and eGFR in 929 patients with type 2 diabetes mellitus.²¹ Lamacchia et al showed increased para- and perirenal fat thickness in patients with CKD compared with those with preserved renal function on echocardiography.²² Odamaki et al demonstrated that 92 patients without diabetes who were treated with hemodialysis had significantly increased visceral fat and decreased subcutaneous fat compared with control subjects of similar BMI.²³ Furthermore, an experimental study also showed a shift of subcutaneous fat to ectopic fat in uninephrectomized rats as a renal dysfunction model.²⁴ Some recent studies investigated the impact of EAT accumulation on renal function and future cardiovascular events in patients with CKD.^25,26 Kerr et al found a positive correlation between urinary albumin to creatinine ratio and single-slice EAT volume in stage 3–5 CKD patients.²⁵ Cordeiro et al demonstrated that increased EAT volume was a predictor for future CAD events independent of traditional cardiovascular risk factors in 277 patients with stage 3–5 CKD during a median follow-up of 32 months.²⁶ In the present study, we found significantly increased EAT volume in CKD patients compared with non-CKD patients, which was associated with high-risk plaque.

The underlying mechanisms of increased EAT volume in CKD subjects are not fully characterized, but there are several possible explanations for this. First, the obvious links have been proved between CKD and metabolic syndrome, which is also strongly associated with EAT accumulation.^27,28 Furthermore, intra-abdominal visceral fat, which is an important component of metabolic syndrome, and EAT share the same origin from the splanchnopleuric mesoderm associated with the gut.²⁹ Recent studies have demonstrated an association between adipocyte-specific parameters such as adiponectin and leptin, and CKD.^30–32 Ohashi et al noted that urine albumin excretion, glomerular hypertrophy and interstitial fibrosis were significantly worse in the adiponectin knockout mice compared with wild type.³¹ Wolf et al showed that leptin mediated glomerular hypertrophy and sclerosis by stimulating glomerular endothelial and mesangial cell proliferation and type IV collagen production.³² Second, EAT is a source of inflammatory mediators³³ and modulates the intrinsic autonomic nervous system,³⁴ which would accelerate the progression of renal dysfunction. Indeed, in the present study serum CRP was significantly higher in CKD patients than in non-CKD patients. In addition, given that EAT accumulation was reported to be associated with left ventricular dysfunction,³⁵ it may indirectly affect renal function. Finally, patients with CKD developed protein malnutrition, catabolism, and protein loss causing reduced muscle mass and insulin resistance, which in turn increased visceral fat including EAT.³⁶ In the present study, there was a significant but weak correlation between BMI and EAT volume (r=0.29, P<0.001). Also, on multivariate analysis EAT volume was associated with eGFR, independent of BMI. These observations are consistent with frequent discordances between BMI and EAT volume.^8,37 Similar results are observed in the association between EAT accumulation and CAD. It remains unclear whether EAT accumulation is the cause or consequence of CKD. EAT seems to be the cause of atherosclerosis based on longitudinal experimental and clinical studies.^38,39 Such a longitudinal study on EAT and CKD is required in the future.

Study Limitations

The present study has several limitations. First, this was a single-center preliminary study, thus further large-scale studies are necessary to confirm the present findings. Furthermore, because we enrolled mild-moderate CKD subjects, the results may not be generalizable to subjects with severe renal function including hemodialysis. These limitations regarding study population may have influenced the lack of statistical association between eGFR and diabetes in this study. Second, given the cross-sectional design of the present study, we could not establish causality. Therefore, a longitudinal study is required to prove the prognostic impact of EAT accumulation in CKD patients. Third, elevated serum CRP in patients with CKD suggests that inflammation is an important factor in the association between EAT and CKD, but detailed mechanisms remain to be resolved. Measurement of specific markers, such as leptin and adiponectin, may help to answer this question.^30–32 Finally, histopathological analysis of EAT was not performed in this study. Several studies have suggested the potential usefulness of MDCT for the assessment of EAT biological characteristics.^40,41 Future studies are necessary to investigate qualitative differences in EAT between CKD and non-CKD patients.

Conclusions

There is a close association between CKD, EAT volume, and high-risk plaque. The present findings are consistent with the hypothesis of a local pathogenic effect of EAT on coronary plaque morphology in patients with CKD. Intervention targeting EAT accumulation may have the potential to prevent or inhibit CAD in patients with CKD.

Disclosures

The authors declare no conflicts of interest.

References

• 1.    Levey  AS,  Coresh  J,  Balk  E,  Kausz  AT,  Levin  A,  Steffes  MW, et al. National Kidney Foundation practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Ann Intern Med 2003; 139: 137–147.
• 2.   National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis 2002; 39: S1–S266.
• 3.    Ix  JH,  Shlipak  MG,  Liu  HH,  Schiller  NB,  Whooley  MA. Association between renal insufficiency and inducible ischemia in patients with coronary artery disease: The heart and soul study. J Am Soc Nephrol 2003; 14: 3233–3238.
• 4.    Sarnak  MJ,  Levey  AS,  Schoolwerth  AC,  Coresh  J,  Culleton  B,  Hamm  LL, et al. Kidney disease as a risk factor for development of cardiovascular disease: A statement from the American Heart Association Councils on kidney in cardiovascular disease, high blood pressure research, clinical cardiology, and epidemiology and prevention. Circulation 2003; 108: 2154–2169.
• 5.    Anavekar  NS,  McMurray  JJ,  Velazquez  EJ,  Solomon  SD,  Kober  L,  Rouleau  JL, et al. Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med 2004; 351: 1285–1295.
• 6.    Go  AS,  Chertow  GM,  Fan  D,  McCulloch  CE,  Hsu  CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296–1305.
• 7.    Rosito  GA,  Massaro  JM,  Hoffmann  U,  Ruberg  FL,  Mahabadi  AA,  Vasan  RS, et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: The Framingham Heart Study. Circulation 2008; 117: 605–613.
• 8.    Sarin  S,  Wenger  C,  Marwaha  A,  Qureshi  A,  Go  BD,  Woomert  CA, et al. Clinical significance of epicardial fat measured using cardiac multislice computed tomography. Am J Cardiol 2008; 102: 767–771.
• 9.    Cheng  VY,  Dey  D,  Tamarappoo  B,  Nakazato  R,  Gransar  H,  Miranda-Peats  R, et al. Pericardial fat burden on ECG-gated noncontrast CT in asymptomatic patients who subsequently experience adverse cardiovascular events. JACC Cardiovasc Imaging 2010; 3: 352–360.
• 10.    Iacobellis  G,  Lonn  E,  Lamy  A,  Singh  N,  Sharma  AM. Epicardial fat thickness and coronary artery disease correlate independently of obesity. Int J Cardiol 2011; 146: 452–454.
• 11.    Nakanishi  K,  Fukuda  S,  Tanaka  A,  Otsuka  K,  Sakamoto  M,  Taguchi  H, et al. Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation. Circ J 2012; 76: 2748–2754.
• 12.    Otsuka  K,  Fukuda  S,  Tanaka  A,  Nakanishi  K,  Taguchi  H,  Yoshikawa  J, et al. Napkin-ring sign on coronary CT angiography for the prediction of acute coronary syndrome. JACC Cardiovasc Imaging 2013; 6: 448–457.
• 13.    Urabe  Y,  Yamamoto  H,  Kitagawa  T,  Utsunomiya  H,  Tsushima  H,  Tatsugami  F, et al. Association between serum levels of n-3 polyunsaturated fatty acids and coronary plaque detected by coronary computed tomography angiography in patients receiving statin therapy. Circ J 2013; 77: 2578–2585.
• 14.    Kawai  H,  Motoyama  S,  Sarai  M,  Ito  H,  Takahashi  H,  Harigaya  H, et al. Adding coronary computed tomography angiography to invasive coronary angiography improves prediction of cardiac events. Circ J 2014; 78: 2735–2740.
• 15.    Levey  AS,  Bosch  JP,  Lewis  JB,  Greene  T,  Rogers  N,  Roth  D. A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999; 130: 461–470.
• 16.    Friedewald  WT,  Levy  RI,  Fredrickson  DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972; 18: 499–502.
• 17.    Zhang  B,  Kawachi  E,  Matsunaga  A,  Imaizumi  S,  Noda  K,  Uehara  Y, et al. Reactivity of direct assays for low-density lipoprotein (LDL) cholesterol toward charge-modified LDL in hypercholesterolemia. Circ J 2012; 76: 2241–2248.
• 18.    Austen  WG,  Edwards  JE,  Frye  RL,  Gensini  GG,  Gott  VL,  Griffith  LS, et al. A reporting system on patients evaluated for coronary artery disease: Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 1975; 51: 5–40.
• 19.    Pundziute  G,  Schuijf  JD,  Jukema  JW,  Boersma  E,  de Roos  A,  van der Wall  EE, et al. Prognostic value of multislice computed tomography coronary angiography in patients with known or suspected coronary artery disease. J Am Coll Cardiol 2007; 49: 62–70.
• 20.    Motoyama  S,  Kondo  T,  Sarai  M,  Sugiura  A,  Harigaya  H,  Sato  T, et al. Multislice computed tomographic characteristics of coronary lesions in acute coronary syndromes. J Am Coll Cardiol 2007; 50: 319–326.
• 21.    Kim  SR,  Yoo  JH,  Song  HC,  Lee  SS,  Yoo  SJ,  Kim  YD, et al. Relationship of visceral and subcutaneous adiposity with renal function in people with type 2 diabetes mellitus. Nephrol Dial Transplant 2011; 26: 3550–3555.
• 22.    Lamacchia  O,  Nicastro  V,  Camarchio  D,  Valente  U,  Grisorio  R,  Gesualdo  L, et al. Para- and perirenal fat thickness is an independent predictor of chronic kidney disease, increased renal resistance index and hyperuricaemia in type-2 diabetic patients. Nephrol Dial Transplant 2011; 26: 892–898.
• 23.    Odamaki  M,  Furuya  R,  Ohkawa  S,  Yoneyama  T,  Nishikino  M,  Hishida  A, et al. Altered abdominal fat distribution and its association with the serum lipid profile in non-diabetic haemodialysis patients. Nephrol Dial Transplant 1999; 14: 2427–2432.
• 24.    Zhao  HL,  Sui  Y,  Guan  J,  He  L,  Zhu  X,  Fan  RR, et al. Fat redistribution and adipocyte transformation in uninephrectomized rats. Kidney Int 2008; 74: 467–477.
• 25.    Kerr  JD,  Holden  RM,  Morton  AR,  Nolan  RL,  Hopman  WM,  Pruss  CM, et al. Associations of epicardial fat with coronary calcification, insulin resistance, inflammation, and fibroblast growth factor-23 in stage 3–5 chronic kidney disease. BMC Nephrol 2013; 14: 26.
• 26.    Cordeiro  AC,  Amparo  FC,  Oliveira  MA,  Amodeo  C,  Smanio  P,  Pinto  IM, et al. Epicardial fat accumulation, cardiometabolic profile and cardiovascular events in patients with stages 3–5 chronic kidney disease. J Intern Med 2015; 278: 77–87.
• 27.    Iacobellis  G,  Ribaudo  MC,  Assael  F,  Vecci  E,  Tiberti  C,  Zappaterreno  A, et al. Echocardiographic epicardial adipose tissue is related to anthropometric and clinical parameters of metabolic syndrome: A new indicator of cardiovascular risk. J Clin Endocrinol Metab 2003; 88: 5163–5168.
• 28.    Kurella  M,  Lo  JC,  Chertow  GM. Metabolic syndrome and the risk for chronic kidney disease among nondiabetic adults. J Am Soc Nephrol 2005; 16: 2134–2140.
• 29.    Sacks  HS,  Fain  JN. Human epicardial adipose tissue: A review. Am Heart J 2007; 153: 907–917.
• 30.    Yano  Y,  Hoshide  S,  Ishikawa  J,  Hashimoto  T,  Eguchi  K,  Shimada  K, et al. Differential impacts of adiponectin on low-grade albuminuria between obese and nonobese persons without diabetes. J Clin Hypertens 2007; 9: 775–782.
• 31.    Ohashi  K,  Iwatani  H,  Kihara  S,  Nakagawa  Y,  Komura  N,  Fujita  K, et al. Exacerbation of albuminuria and renal fibrosis in subtotal renal ablation model of adiponectin-knockout mice. Arterioscler Thromb Vasc Biol 2007; 27: 1910–1917.
• 32.    Wolf  G,  Hamann  A,  Han  DC,  Helmchen  U,  Thaiss  F,  Ziyadeh  FN, et al. Leptin stimulates proliferation and TGF-beta expression in renal glomerular endothelial cells: Potential role in glomerulosclerosis. Kidney Int 1999; 56: 860–872.
• 33.    Mazurek  T,  Zhang  L,  Zalewski  A,  Mannion  JD,  Diehl  JT,  Arafat  H, et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003; 108: 2460–2466.
• 34.    Oh  S,  Zhang  Y,  Bibevski  S,  Marrouche  NF,  Natale  A,  Mazgalev  TN. Vagal denervation and atrial fibrillation inducibility: Epicardial fat pad ablation does not have long-term effects. Heart Rhythm 2006; 3: 701–708.
• 35.    Cetin  M,  Kocaman  SA,  Durakoglugil  ME,  Erdogan  T,  Ergul  E,  Dogan  S, et al. Effect of epicardial adipose tissue on diastolic functions and left atrial dimension in untreated hypertensive patients with normal systolic function. J Cardiol 2013; 61: 359–364.
• 36.    Banerjee  D,  Recio-Mayoral  A,  Chitalia  N,  Kaski  JC. Insulin resistance, inflammation, and vascular disease in nondiabetic predialysis chronic kidney disease patients. Clin Cardiol 2011; 34: 360–365.
• 37.    Silaghi  A,  Piercecchi-Marti  MD,  Grino  M,  Leonetti  G,  Alessi  MC,  Clement  K, et al. Epicardial adipose tissue extent: Relationship with age, body fat distribution, and coronaropathy. Obesity 2008; 16: 2424–2430.
• 38.    Takaoka  M,  Nagata  D,  Kihara  S,  Shimomura  I,  Kimura  Y,  Tabata  Y, et al. Periadventitial adipose tissue plays a critical role in vascular remodeling. Circ Res 2009; 105: 906–911.
• 39.    Nakanishi  K,  Fukuda  S,  Tanaka  A,  Otsuka  K,  Jissho  S,  Taguchi  H, et al. Persistent epicardial adipose tissue accumulation is associated with coronary plaque vulnerability and future acute coronary syndrome in non-obese subjects with coronary artery disease. Atherosclerosis 2014; 237: 353–360.
• 40.    Pracon  R,  Kruk  M,  Kepka  C,  Pregowski  J,  Opolski  MP,  Dzielinska  Z, et al. Epicardial adipose tissue radiodensity is independently related to coronary atherosclerosis: A multidetector computed tomography study. Circ J 2011; 75: 391–397.
• 41.    Sato  A,  Aonuma  K. Role of cardiac multidetector computed tomography beyond coronary angiography. Circ J 2015; 79: 712–720.