Association of Local Epicardial Adipose Tissue Depots and Left Ventricular Diastolic Performance in Patients With Preserved Left Ventricular Ejection Fraction

Abstract

Background: Although full-volume quantification of epicardial adipose tissue (EAT) is a predictor of LV diastolic dysfunction (LVDD), how localized EAT depots are linked to LVDD remains unclear. We evaluated the effect of local EAT depots on LV diastolic function parameters in patients with preserved LV ejection fraction (LVEF).

Methods and Results: From 423 consecutive patients who underwent cardiac CT angiography, we recruited 252 with sinus rhythm and normal LVEF. The EAT volume index (EATV/body surface area) and the localized EAT thickness around the right coronary artery (EAT_RCA), left anterior descending artery (EAT_LAD), left circumflex artery (EAT_LCX), right ventricle (EAT_RV), left ventricle (EAT_LV), right atrium (EAT_RA), and left atrium (EAT_LA) were measured using cardiac CT. In the LVDD group (n=71), the EATV index (75±30 vs. 64±28 mL/m², P=0.010), EAT_LCX (10.7±3.8 vs. 9.4±3.4 mm, P=0.008), and EAT_LV (2.6±1.6 vs. 2.1±1.4 mm, P=0.024) were greater than in the non-LVDD group (n=181). In contrast, EAT_LCX and EAT_LV were markedly associated with decreased lateral e’ and increased lateral E/e’. Multiple regression analysis indicated that EAT_LCX and EAT_LV were strongly associated with LV diastolic function parameters.

Conclusions: Localized EAT depots are linked to altered mitral annular motion. Further study is warranted to clarify whether localized EAT depots are functionally linked to the clinical manifestations of LVDD.

Epicardial adipose tissue (EAT) has been widely studied and is regarded as metabolically active fat that is anatomically adjacent to the myocardium and coronary arteries.^1–3 The EAT volume (EATV), as well as visceral adipose tissue, is increased in obese patients and correlates well with the presence^4,5 and incidence⁶ of coronary artery disease (CAD), independent of the traditional CAD risk factors. Also, excessive accumulation of EAT might create a paracrine or mechanical burden on the coronary microcirculation and thus the myocardium.⁷ The EATV is reported to be an independent predictor of left ventricular (LV) remodeling and LV diastolic dysfunction (LVDD) in patients with metabolic syndrome,⁸ and in patients following myocardial infarction.⁹ Furthermore, the EATV correlates with the left atrial diameter, which is an indirect structural parameter of LVDD.⁸ Those studies focused on full-volume quantification of EAT and its effect on coronary atherosclerosis or myocardial function. However, whether localized EAT accumulation is linked to LV diastolic parameters and, if so, the mechanisms by which regional EAT results in LVDD remain unclear.

Editorial p 156

LVDD is defined as impaired LV relaxation, with or without reduced restoring forces and increased LV chamber stiffness, which lead to an increase in cardiac filling pressure.¹⁰ Thus, the most recent guidelines recommend the evaluation of patients with suspected LVDD using several indices, including average E/e’, septal and lateral e’ velocities, tricuspid regurgitation velocity, and left atrial volume index (LAVI).¹¹ At present, which local EAT depots are linked to which echocardiographic LV diastolic indicators remains unclear.

In this study, we evaluated the effect of local EAT depots on echocardiographic LV diastolic function parameters and LVDD diagnosis in patients with preserved LV ejection fraction (LVEF).

Methods

Study Participants

We retrospectively analyzed 423 consecutive Japanese patients who had undergone cardiac CT for suspected CAD between 2012 and 2015 at Tokushima University Hospital (Supplementary Figure). Subjects were divided into CAD (≥1 coronary artery branch stenosis ≥50%) and non-CAD groups. The major exclusion criteria included serum creatinine levels >1.5 mg/dL; class III or IV heart failure; iodine-based allergy; acute coronary events, stroke, or coronary revascularization within the preceding 3 months; overt liver disease; hypothyroidism; and severe valvular disease.¹² LA size is also thought to be an indicator of LVDD or LV filling pressure, although certainly other conditions, such as atrial fibrillation (AF), can result in LA enlargement in the absence of a significant increase in LA pressure. Therefore, we excluded patients with AF (n=150); after exclusion of LVEF <50% (n=16), and insufficient data (n=5), 252 patients were included in the full analysis.

Cardiovascular Risk Factors Assessment

Covariates (i.e., various clinical parameters) were obtained from electronic medical records. All participants provided written informed consent for cardiac CT after they were advised about radiation exposure-related risks and the possible complications associated with iodine-containing contrast agents. Hypertension was defined as blood pressure ≥140/90 mmHg or the current use of antihypertensive medication(s). Diabetes was defined as HbA_1c ≥6.5%, fasting plasma glucose levels >126 mg/dL, or the current use of antidiabetic medications. Dyslipidemia was defined as a total serum cholesterol level ≥220 mg/dL, low-density lipoprotein cholesterol level ≥140 mg/dL, serum triglyceride level >150 mg/dL, or serum high-density lipoprotein cholesterol level <40 mg/dL, and/or current use of antihyperlipidemia medications. Smoking was defined as past or current smoker; non-smoking was defined as never smoked.

EATV and Localized Fat Depot Quantification

Cardiac CT was performed using a 320-slice CT scanner (Aquilion One; Toshiba Medical Systems, Tokyo, Japan) having 0.275-ms rotation and 0.5/320/0.25 collimation.¹² CT images were acquired using a retrospective, non-helical ECG-triggered acquisition mode protocol (tube voltage, 120 kV; tube current, 450 mA×5 ms) with 5-mm slice thicknesses. All reconstructed CT image data were transferred to an offline workstation (Synapse Vincent, ver. 4.4, Fuji Film, Tokyo, Japan). EATV and local EAT thicknesses were measured as previously reported.^13–15 EAT thickness measurements were performed around the left anterior descending artery (EAT_LAD), right coronary artery (EAT_RCA), and left circumflex artery (EAT_LCX) as described;¹⁵ similar measurements were also made around the LV wall (EAT_LV), right ventricular wall (EAT_RV), left atrial wall (EAT_LA), and right atrial wall (EAT_RA). Representative measurements from patients with and without LVDD are shown in Figure 1.

Figure 1.

Representative measurements of localized epicardial adipose tissue (EAT) depots in patients with and without left ventricular diastolic dysfunction (LVDD). EAT surrounding (1) the left anterior descending artery (EAT_LAD), (2) right coronary artery (EAT_RCA), (3) left circumflex artery (EAT_LCX), (4) right atrial wall (EAT_RA), (5) right ventricular wall (EAT_RV), (6) left atrial wall (EAT_LA), and (7) left ventricular wall (EAT_LV). (A) On plain axial 4-chamber views, a region of interest for EAT measurements was manually placed (B) along the epicardium; (C) the EAT area was automatically acquired as a density range between −190 and −30 HU on cardiac computed tomography. Compared with a patient without LVDD (D), the patient with LVDD (E) shows increased EAT_LCX (7.8 vs. 17.9 mm), despite other EAT measurements being similar. BMI, body mass index; EATV, EAT volume; EF, ejection fraction; LAV, LA volume; LVMI, LV mass index; LVEDV, LV end-diastolic volume; LVESV, LV end-systolic volume; RA, right atrium; RV, right ventricle.

Echocardiographic Parameters

Echocardiography was performed using commercially available ultrasound diagnostic instruments in accordance with the guidelines issued by the American Society of Echocardiography (ASE).¹⁶ Transmitral flow (TMF) velocity was recorded from the apical long-axis or 4-chamber view; the ratio of the peak early diastolic (E) and the peak atrial systolic TMF velocities was calculated, if applicable. The deceleration time of the early TMF velocity was also measured. The mitral annular motion velocity pattern was recorded using the apical 4-chamber view with the sample volume located at the lateral or septal side of the mitral annulus, using pulsed tissue Doppler echocardiography. The mean peak early diastolic mitral annular velocity (e’) was measured on the septal and lateral sides, and the E to e’ ratio (E/e’) was calculated as a marker of LV filling pressure. In addition to these diastolic parameters, routine echocardiographic parameters were measured, including the LA dimension, LV end-diastolic dimension, and LV end-systolic dimension, using the M-mode or 2D echocardiogram. The LVEF was calculated from the apical 2- and 4-chamber views using the modified Simpson’s method. The LA volume, LAVI (LAV/body surface area), and LV mass index (LVMI=LV mass/body surface area) were calculated.¹⁶ Based on the 2016 ASE/European Association of Echocardiography/European Association of Cardiovascular Imaging (ASE/EAE/EACI) algorithm,¹¹ LVDD was defined as normal, indeterminate or abnormal in patients with LVEF ≥50% (normal) using the following 4 items: (1) average E/e’ ≥14; (2) septal e’ velocity <7 cm/s or lateral e’ velocity <10 cm/s; (3) tricuspid regurgitation velocity >2.8 m/s; (4) LAVI >34 mL/m².¹¹ In the current study, LVDD was redefined using a combination of the “indeterminate” and “abnormal” categories, because the participants mainly had mild LVDD. All Doppler recordings were performed during an end-expiratory breath hold and the mean value of 3 consecutive cardiac cycles was used in the analyses.

Statistical Analysis

Continuous variables are expressed as mean (± standard deviation) and 2 groups were compared using unpaired Student’s t-tests or the Mann-Whitney U-test. Categorical variables are summarized as frequencies and percentages and were compared using chi-square tests. Correlations between variables were determined using Pearson’s correlation coefficient tests. The effect of individual risk factors on septal e’ and E/e’ measurements, lateral e’ and E/e’ measurements, or on LVDD diagnosis were determined using multivariate regression models. All covariates, except EAT measurements, were chosen for their established or presumed influence on LV diastolic function. We constructed a model for septal and lateral e’ and E/e’ and LVDD by the following steps. In Model 1, age, male sex and body mass index (BMI) were selected. In Model 2, Model 1+established or presumed variables were selected (yes or no for smoking status, hyperlipidemia, hypertension, type 2 diabetes mellitus, and CAD). For Models 3–5, EAT indices were added to Model 2. It was assumed that the EATV index and respective values of EATV thickness were mutually correlated. Therefore, each EAT index was added in a hierarchical fashion to determine the stronger predictor of EAT indices. For Models 6–8, the LVMI was added to Models 3–5, because this index was assumed to affect both LVDD parameters and EATV.¹⁵ The optimal cutoff values for BMI, EATV index, EAT_LCX, and EAT_LV for predicting LVDD were identified using receiver-operating characteristic (ROC) curves. For all tests, statistical significance was set at P<0.05. All statistical analyses were performed using SPSS 21.0 for Windows (SPSS, Chicago, IL, USA).

Results

General Characteristics of Patients With and Without LVDD

General characteristics of patients with and without LVDD are shown in all age and <65- and ≥65-year-old groups (Table 1). In all age groups, the mean age of the included patients was higher in the LVDD group than in the non-LVDD group (70±12 vs. 64±14 years, P=0.001); the plasma concentration of B-type natriuretic peptide (BNP) was also higher in the LVDD group (89±106 vs. 38±42 pmol/L, P<0.001). Other parameters, including blood pressure, heart rate, anthropometric measurements, blood measurements, comorbidity prevalence and medications were comparable and the frequency of coronary artery stenosis ≥50% in the RCA, LAD and LCX did not significantly differ between the 2 groups. Among the echocardiographic parameters, LVMI, LAVI, interventricular septal thickness (IVT), and posterior wall thickness were significantly greater in the LVDD group than in the non-LVDD group. Among the EAT measures, the EATV index was significantly larger in the LVDD group (75±30 vs. 64±28 cm³/m², P=0.010); EAT_LCX and EAT_LV were also significantly larger in the LVDD group than in the non-LVDD (EAT_LV: 2.6±1.6 vs. 2.1±1.4 mm, P=0.024; EAT_LCX: 10.7±3.8 vs. 9.4±3.3 mm, P=0.008). In the subgroups <65 and ≥65 years of age, the mean age was comparable between the LVDD and non-LVDD groups, and BNP was still higher in the LVDD group. Other parameters, including blood pressure, heart rate, anthropometric and blood measurements, comorbidity and medications were comparable, except for 1-vessel disease in the <65 years age group. The echocardiographic parameters were similarly distributed in the LVDD and non-LVDD groups in both age subgroups. The EATV index, EAT_LCX and EAT_LV tended to be higher in the LVDD group as compared with the non-LVDD group, although the differences did not reach statistical significance.

Table 1. General Characteristics of Patients With and Without LVDD

Parameters	All	All		P value	<65 years old		P value	≥65 years old		P value
		LVDD−	LVDD+		LVDD−	LVDD+		LVDD−	LVDD+
n	252	181	71	–	79	15	–	102	56	–
Age (years)	66±13	64±14	70±12	0.001	52±10	53±9	0.794	74±5	75±7	0.158
Male sex, n (%)	137 (54)	104 (57)	33 (46)	0.087	43 (54)	6 (40)	0.401	61 (60)	26 (46)	0.115
Systolic BP (mmHg)	132±19	131±19	134±18	0.244	130±19	135±13	0.263	132±19	133±19	0.593
Diastolic BP (mmHg)	74±13	74±12	71±14	0.087	79±12	80±14	0.683	71±11	69±12	0.321
Heart rate (beats/min)	73±13	74±13	70±12	0.138	77±13	78±21	0.823	73±14	68±10	0.172
Anthropometry
Body weight (kg)	63±16	65±17	60±13	0.056	71±20	70±15	0.836	59±12	58±11	0.374
BMI (kg/m2)	24.90± 4.98	24.90± 5.32	24.91± 4.02	0.993	26±7	27±4	0.632	23.7±3.7	24.3±3.8	0.375
Hematology
Total cholesterol (mg/dL)	196±39	195±39	199±38	0.440	202±39	214±46	0.307	190±39	195±35	0.361
LDL cholesterol (mg/dL)	110±32	111±34	108±27	0.605	116±31	111±28	0.667	107±35	107±27	0.903
HDL cholesterol (mg/dL)	62±20	61±20	64±21	0.381	62±23	65±22	0.630	61±18	63±21	0.400
Triglycerides (mg/dL)	133±67	132±66	134±72	0.810	136±66	180±94	0.031	129±65	122±60	0.515
BNP (pmol/L)	53±71	38±42	89±106	<0.001	24±29	101±118	<0.001	49±49	85±104	0.008
Comorbidities
Smoking, n (%)	108 (43)	83 (46)	25 (35)	0.125	34 (43)	5 (33)	0.575	49 (48)	20 (36)	0.180
Hyperlipidemia, n (%)	189 (75)	134 (74)	55 (78)	0.571	59 (75)	14 (93)	0.177	75 (74)	41 (73)	1.000
Hypertension, n (%)	192 (76)	134 (74)	58 (82)	0.249	47 (59)	11 (73)	0.393	87 (85)	47 (84)	0.636
Diabetes mellitus, n (%)	77 (31)	60 (33)	17 (24)	0.146	24 (30)	4 (27)	1.000	36 (35)	13 (23)	0.150
BMI ≥25 kg/m2, n (%)	126 (50)	91 (50)	35 (49)	0.889	48 (61)	12 (80)	0.171	42 (41)	23 (41)	0.977
CAD, n (%)	137 (54)	94 (52)	43 (61)	0.199	26 (33)	7 (47)	0.389	68 (67)	36 (64)	0.861
RCA, n (%)*	81 (32)	61 (34)	20 (28)	0.454	21 (27)	3 (20)	0.575	40 (39)	17 (30)	0.302
LAD, n (%)*	96 (38)	64 (35)	32 (45)	0.254	20 (25)	2 (13)	0.500	44 (43)	29 (52)	0.547
LCX, n (%)*	78 (31)	54 (30)	24 (34)	0.546	13 (16)	2 (13)	0.748	41 (40)	22 (39)	0.942
1-vessel disease	56 (22)	35 (19)	21 (30)	0.076	8 (10)	6 (40)	0.009	27 (26)	15 (27)	0.942
2-vessel disease	44 (17)	33 (18)	11 (15)	0.613	8 (10)	1 (7)	0.667	25 (25)	10 (18)	0.347
3-vessel disease	37 (15)	26 (14)	11 (15)	0.813	10 (13)	0 (0)	0.357	16 (16)	11 (20)	0.512
Medications
Antihypertensive drugs	150 (60)	105 (58)	45 (63)	0.435	43 (54)	11 (73)	0.256	62 (61)	34 (61)	1.000
RAS inhibitors	72 (29)	56 (31)	16 (23)	0.187	21 (27)	4 (27)	0.984	35 (34)	12 (21)	0.095
Lipid-lowering drugs	120 (48)	84 (46)	36 (51)	0.628	33 (42)	10 (67)	0.094	51 (50)	26 (46)	0.739
Oral antidiabetic drugs	62 (25)	47 (30)	15 (21)	0.304	20 (25)	4 (27)	0.754	27 (26)	11 (20)	0.435
Echocardiography
Tricuspid regurgitation velocity (m/s)	2.23±0.32	2.17±0.27	2.39±0.39	<0.001	2.13±0.25	2.36±0.38	0.016	2.20±0.28	2.40±0.40	0.001
LA diameter (mm)	38.1±5.9	37.3±5.3	40.1±7.0	0.001	37.1±4.7	44.5±8.2	<0.001	37.5±5.7	39.0±6.2	0.130
LAV index (mL/m2)	29.2±10.5	25.4±7.3	40.4±10.4	<0.001	24.7±7.1	39.1±10.4	<0.001	25.9±7.4	40.7±10.5	<0.001
LV end-diastolic dimension (mm)	46.49± 5.22	46.50± 5.03	46.46± 5.70	0.962	47.98± 4.63	46.13± 5.01	0.123	45.35± 5.05	46.55± 5.91	0.182
LV end-systolic dimension (mm)	29.23± 5.09	29.33± 4.80	28.98± 5.79	0.627	30.73± 4.26	28.60± 7.37	0.627	28.24± 4.93	29.08± 5.37	0.321
IVS (mm)	8.8±1.9	8.5±1.3	9.5±2.7	<0.001	8.6±1.5	11.1±4.8	<0.001	8.5±1.2	9.0±1.6	0.023
LVPW (mm)	8.5±1.4	8.4±1.3	8.9±1.6	0.005	8.4±1.3	9.4±2.0	0.019	8.3±1.3	8.8±1.4	0.045
LVMI (g/m2)	85.2± 23.3	81.3± 20.7	95.2± 26.6	<0.001	80.7± 19.6	100.0± 37.3	0.005	81.7±21.6	94.0±23.5	0.001
LVEF (%)	64.9±5.4	65.0±5.1	64.7±6.2	0.639	64.0±5.6	65.7±6.0	0.292	65.8±4.5	64.4±6.2	0.110
Mitral peak E velocity (m/s)	65.5±20.1	63.2±16.4	71.4±28.8	0.005	67.8±19.3	78.9±48.2	0.132	59.6±12.7	69.3±20.9	<0.001
Mitral deceleration time (ms)	244.2± 72.1	242.8± 68.0	247.6± 82.1	0.633	219.1± 49.1	226.0± 69.8	0.646	260.3± 74.7	253.5± 84.7	0.607
Mitral peak A velocity (m/s)	81.7±21.3	77.3±19.7	92.9±21.3	<0.001	66.5±17.4	82.4±19.2	0.002	85.9±17.1	95.8±21.0	0.002
Mitral E/A	0.85±0.34	0.87±0.35	0.78±0.28	0.035	1.08±0.40	0.94±0.39	0.228	0.71±0.20	0.74±0.23	0.426
Lateral e’ (cm/s)	8.6±2.7	9.1±2.8	7.4±2.0	<0.001	10.3±3.0	7.3±2.2	<0.001	8.1±2.3	7.5±2.0	0.090
Septal e’ (cm/s)	6.6±2.5	7.0±2.5	5.5±2.0	<0.001	7.9±2.3	5.8±2.6	0.004	6.4±2.6	5.5±1.8	0.026
Septal E/e’	10.8±4.5	9.6±3.0	14.0±6.0	<0.001	8.9±2.8	14.4±7.1	<0.001	10.2±3.0	13.6±6.0	<0.001
Lateral E/e’	8.3±3.5	7.4±2.1	10.4±5.1	<0.001	6.9±1.9	11.6±6.1	<0.001	7.8±2.1	10.1±4.8	<0.001
Average E/e’	9.5±3.7	8.5±2.2	12.0±5.3	<0.001	7.9±2.1	13.0±6.0	<0.001	9.0±2.1	11.8±5.2	<0.001
EAT measurements
EATV (mL)	111.6± 51.8	108.7± 51.7	119.0± 51.7	0.153	102.5± 48.2	122.0± 70.6	0.188	113.4± 54.1	118.2± 46.2	0.574
EATV index (mL/m2)	67.2±29.0	64.3±28.2	74.7±29.5	0.010	56.7±22.4	69.4±38.8	0.081	70.2±30.8	76.1±26.8	0.232
EATRCA (mm)	16.4±4.6	16.0±4.3	17.3±5.3	0.052	15.8±3.9	16.0±3.9	0.810	16.2±4.6	17.6±5.6	0.092
EATLAD (mm)	6.1±2.4	5.9±2.4	6.4±2.4	0.212	5.7±2.6	6.0±2.9	0.655	6.1±2.3	6.5±2.3	0.409
EATLCX (mm)	9.8±3.5	9.4±3.3	10.7±3.8	0.008	8.9±2.9	10.2±3.7	0.117	9.8±3.6	10.9±3.9	0.099
EATRA (mm)	1.2±1.7	1.2±1.9	1.1±1.3	0.589	1.1±1.1	1.4±2.3	0.379	1.3±2.3	1.0±0.8	0.337
EATRV (mm)	2.9±2.2	2.9±2.3	3.0±1.9	0.633	2.9±2.6	3.1±2.1	0.731	2.8±2.0	3.0±1.9	0.671
EATLA (mm)	1.5±1.2	1.5±1.2	1.6±1.2	0.700	1.5±1.2	1.5±1.4	0.399	1.5±1.0	1.7±1.2	0.376
EATLV (mm)	2.2±1.4	2.1±1.4	2.6±1.6	0.024	2.0±1.3	2.5±1.2	0.129	2.2±1.4	2.6±1.7	0.150

Values are mean±SD or n (%). Statistical significance was tested by unpaired two-tailed t-test or chi-square test between LVDD− and LVDD+. *Coronary artery branch stenosis ≥50%. BMI, body mass index; BNP, B-type natriuretic peptide; BP, blood pressure; BSA, body surface area; EATV, epicardial adipose tissue volume; EATV index, EATV/BSA (mL/m²); IVS, interventricular septum; LA, left atrium; LAD, left anterior descending artery; LAV, left atrial volume; LCX, left circumflex artery; LVDD, left ventricular diastolic dysfunction; LVEF, LV ejection fraction; LVMI, left ventricular mass index; LVPW, left ventricular posterior wall thickness; RAS, renin-angiotensin system; RCA, right coronary artery.

Localized EAT Depots and LV Diastolic Function Parameters

The correlation coefficients between localized EAT depots and echocardiographic parameters in all age groups are shown in Table 2. Among them, EAT_LCX and EAT_LV were significantly associated with LVDD (Table 2). We therefore investigated whether these fat depots were related to the early diastolic velocity of the mitral annulus (septal e’ and lateral e’). For septal e’, age and LVMI were significant determinants (Table 3A, Model 8), whereas for lateral e’, age, hypertension, and EAT_LCX were determinants (Table 3B, Model 8). For septal E/e’, age, male gender, EAT_LCX, and LVMI were determinants (Table 3C, Model 8), and for lateral E/e’, age, male gender, and EAT_LCX were determinants (Table 3D, Model 8).

Table 2. Correlation Coefficients Between Parameters

Parameters	LVDD (yes or no)	Age (years)	Male sex (yes or no)	BMI (kg/m2)	Hypertension (yes or no)	Diabetes mellitus (yes or no)	CAD (yes or no)	Tricuspid regurgitation velocity (m/s)	LAV index (mL/m2)	LVMI (g/m2)	Septal e’ (cm/s)	Lateral e’ (cm/s)	Septal E/e’	Lateral E/e’	Average E/e’	EATV index (mL/m2)	EATRCA (mm)	EATLAD (mm)	EATLCX (mm)	EATRA (mm)	EATRV (mm)	EATLA (mm)	EATLV (mm)
LVDD (yes or no)	1.000	0.210**	−0.099	0.001	0.073	−0.090	0.081	0.298**	0.625**	0.269**	−0.265**	−0.272**	0.411**	0.391**	0.427**	0.162*	0.123	0.079	0.168**	−0.034	0.030	0.024	0.142*
Age (years)		1.000	0.053	−0.261**	0.231**	0.038	0.299**	0.159*	0.205**	−0.013	−0.328**	−0.477**	0.176**	0.209**	0.201**	0.266**	0.095	0.129*	0.158*	0.063	0.044	0.131*	0.127*
Male sex (yes or no)			1.000	0.098	−0.089	0.089	0.126*	−0.093	−0.047	0.121	0.077	0.051	−0.224**	−0.157*	−0.211**	−0.021	−0.049	−0.110	0.092	0.026	−0.051	0.092	0.008
BMI (kg/m2)				1.000	0.029	0.127*	0.002	−0.126	−0.015	0.076	0.075	−0.036	−0.054	0.011	−0.028	0.345**	0.311**	0.131*	0.324**	0.160*	0.286**	0.165**	0.344**
Hypertension (yes or no)					1.000	0.079	0.115	0.097	0.057	0.108	−0.179**	−0.302**	0.135*	0.158*	0.153*	0.158*	0.096	0.005	0.087	0.027	0.150*	0.089	0.072
Diabetes mellitus (yes or no)						1.000	0.056	−0.052	0.017	−0.085	−0.044	−0.046	−0.047	−0.025	−0.050	0.160*	0.005	−0.071	0.165**	0.214**	0.117	0.137*	0.098
CAD (yes or no)							1.000	0.057	0.018	0.031	−0.066	−0.136*	−0.025	0.029	−0.008	0.159*	0.069	0.201**	0.051	0.027	0.130*	0.070	0.202**
Tricuspid regurgitation velocity (m/s)								1.000	0.352**	0.150*	−0.033	−0.042	0.296**	0.295**	0.319**	−0.087	−0.075	−0.162*	0.098	0.019	−0.071	−0.055	−0.003
LAV index (mL/m2)									1.000	0.329**	−0.016	−0.026	0.181**	0.150*	0.176**	0.146*	0.112	0.027	0.188**	−0.045	0.072	0.012	0.106
LVMI (g/m2)										1.000	−0.191**	−0.104	0.204**	0.080	0.160*	0.053	0.068	−0.069	0.048	−0.024	−0.067	−0.067	−0.001
Septal e’ (cm/s)											1.000	0.600**	−0.593**	−0.224**	−0.469**	−0.190**	−0.081	−0.015	−0.134*	−0.067	−0.078	−0.039	−0.128*
Lateral e’ (cm/s)												1.000	−0.347**	−0.558**	−0.472**	−0.364**	−0.149*	−0.211**	−0.229**	−0.081	−0.244**	−0.168**	−0.276**
Septal E/e’													1.000	0.687**	0.940**	0.020	0.050	0.032	0.132*	0.087	0.023	−0.023	0.012
Lateral E/e’														1.000	0.895**	0.106	0.050	0.134*	0.174**	0.100	0.132*	0.057	0.111
Average E/e’															1.000	0.060	0.056	0.075	0.166**	0.099	0.081	0.017	0.060
EATV index (mL/m2)																1.000	0.487**	0.493**	0.528**	0.219**	0.443**	0.316**	0.523**
EATRCA (mm)																	1.000	0.312**	0.362**	0.113	0.243**	0.217**	0.355**
EATLAD (mm)																		1.000	0.254**	0.051	0.252**	0.136*	0.395**
EATLCX (mm)																			1.000	0.187**	0.252**	0.236**	0.373**
EATRA (mm)																				1.000	0.105	0.174**	0.135*
EATRV (mm)																					1.000	0.466**	0.533**
EATLA (mm)																						1.000	0.421**
EATLV (mm)																							1.000

**P<0.05 or **P<0.01. See Table 1 for all abbreviations.

Table 3. Multiple Regression Analysis to Estimate Septal e’, Lateral e’, Septal E/e’ and Lateral E/e’

A. Multiple regression analysis to estimate septal e’
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.117		0.139		0.153		0.148		0.146		0.19		0.184		0.183
P	<0.001		<0.001		<0.001		<0.001		<0.001		<0.001		<0.001		<0.001
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	−0.342	<0.001	−0.318	0.001	−0.266	<0.001	−0.300	<0.001	−0.293	<0.001	−0.276	<0.001	−0.308	<0.001	−0.302	<0.001
Male sex (yes or no)	0.093	0.128	0.006	0.929	−0.016	0.826	−0.004	0.959	0.010	0.897	0.042	0.571	0.053	0.472	0.066	0.373
BMI (kg/m2)	−0.034	0.592	−0.009	0.894	0.052	0.468	0.034	0.630	0.029	0.677	0.061	0.391	0.041	0.554	0.037	0.598
Smoking status (yes or no)			0.128	0.072	0.148	0.039	0.132	0.064	0.127	0.073	0.117	0.099	0.103	0.145	0.098	0.164
Hyperlipidemia (yes or no)			−0.059	0.354	−0.054	0.401	−0.063	0.324	−0.065	0.308	−0.027	0.672	−0.036	0.572	−0.037	0.555
Hypertension (yes or no)			−0.095	0.137	−0.093	0.146	−0.096	0.134	−0.093	0.146	−0.064	0.309	−0.067	0.288	−0.065	0.308
Type 2 diabetes mellitus (yes or no)			−0.023	0.713	−0.008	0.892	−0.017	0.783	−0.012	0.843	−0.043	0.487	−0.051	0.407	−0.047	0.446
CAD (yes or no)			0.024	0.712	0.027	0.679	0.042	0.526	0.021	0.745	0.018	0.781	0.032	0.622	0.013	0.839
EATV index (cm3/m2)					−0.142	0.047					−0.133	0.057
EATLV thickness (mm)							−0.107	0.113					−0.096	0.152
EATLCX thickness (mm)									−0.098	0.146					−0.086	0.196
LVMI (g/m2)											−0.194	0.002	−0.200	0.001	−0.197	0.002
B. Multiple regression analysis to estimate lateral e’
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.261		0.297		0.328		0.327		0.31		0.335		0.332		0.316
P	<0.001		<0.001		<0.001		<0.001		<0.001		<0.001		<0.001		<0.001
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	−0.524	<0.001	−0.49	<0.001	−0.414	<0.001	−0.455	<0.001	−0.459	<0.001	−0.419	<0.001	−0.460	<0.001	−0.465	<0.001
Male sex (yes or no)	0.096	0.082	0.050	0.450	0.016	0.806	0.033	0.611	0.053	0.414	0.042	0.526	0.058	0.382	0.079	0.237
BMI (kg/m2)	−0.173	0.003	−0.141	0.015	−0.048	0.447	−0.065	0.290	−0.091	0.147	−0.042	0.507	−0.062	0.313	−0.086	0.165
Smoking status (yes or no)			0.008	0.899	0.039	0.537	0.015	0.809	0.008	0.904	0.022	0.722	0.001	0.989	−0.008	0.901
Hyperlipidemia (yes or no)			−0.110	0.057	−0.100	0.076	−0.115	0.041	−0.118	0.040	−0.088	0.120	−0.103	0.069	−0.105	0.068
Hypertension (yes or no)			−0.169	0.003	−0.162	0.004	−0.169	0.003	−0.165	0.004	−0.148	0.009	−0.156	0.006	−0.152	0.008
Type 2 diabetes mellitus (yes or no)			0.007	0.895	0.026	0.635	0.015	0.781	0.021	0.711	0.012	0.824	0.002	0.976	0.006	0.913
CAD (yes or no)			0.040	0.499	0.047	0.413	0.070	0.232	0.036	0.530	0.041	0.476	0.064	0.268	0.031	0.589
EATV index (cm3/m2)					−0.211	0.001					−0.208	0.001
EATLV thickness (mm)							−0.193	0.001					−0.184	0.002
EATLCX thickness (mm)									−0.128	0.035					−0.120	0.047
LVMI (g/m2)											−0.069	0.208	−0.082	0.135	−0.076	0.172
C. Multiple regression analysis to estimate septal E/e’
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.086		0.096		0.098		0.096		0.115		0.155		0.152		0.167
P	<0.001		0.002		0.003		0.004		0.001		<0.001		<0.001		<0.001
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	0.195	0.002	0.197	0.005	0.216	0.004	0.198	0.005	0.159	0.025	0.231	0.001	0.212	0.002	0.174	0.012
Male sex (yes or no)	−0.237	<0.001	−0.211	0.005	−0.219	0.004	−0.212	0.005	−0.215	0.004	−0.293	<0.001	−0.285	<0.001	−0.284	<0.001
BMI (kg/m2)	0.021	0.738	0.016	0.805	0.039	0.599	0.019	0.793	−0.042	0.552	0.028	0.699	0.009	0.898	−0.051	0.452
Smoking status (yes or no)			0.005	0.945	0.012	0.872	0.005	0.943	0.006	0.928	0.053	0.458	0.046	0.521	0.044	0.528
Hyperlipidemia (yes or no)			0.034	0.602	0.036	0.581	0.034	0.605	0.043	0.500	0.003	0.967	0.001	0.994	0.010	0.874
Hypertension (yes or no)			0.075	0.246	0.076	0.239	0.075	0.248	0.071	0.263	0.039	0.542	0.037	0.560	0.035	0.573
Type 2 diabetes mellitus (yes or no)			−0.044	0.483	−0.039	0.538	−0.044	0.487	−0.061	0.332	0.002	0.972	−0.003	0.959	−0.021	0.738
CAD (yes or no)			−0.065	0.324	−0.064	0.335	−0.064	0.338	−0.060	0.357	−0.055	0.391	−0.053	0.418	−0.053	0.408
EATV index (cm3/m2)					−0.051	0.485					−0.063	0.374
EATLV thickness (mm)							−0.006	0.928					−0.023	0.737
EATLCX thickness (mm)									0.152	0.026					0.136	0.040
LVMI (g/m2)											0.241	<0.001	0.238	<0.001	0.233	<0.001
D. Multiple regression analysis to estimate lateral E/e’
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.079		0.106		0.106		0.110		0.128		0.115		0.118		0.136
P	<0.001		0.001		0.001		0.001		<0.001		0.001		0.001		<0.001
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	0.242	<0.001	0.238	0.001	0.240	0.001	0.224	0.001	0.195	0.006	0.246	0.001	0.230	0.001	0.201	0.004
Male sex (yes or no)	−0.178	0.004	−0.202	0.007	−0.203	0.007	−0.196	0.009	−0.208	0.005	−0.234	0.003	−0.226	0.003	−0.236	0.002
BMI (kg/m2)	0.092	0.151	0.077	0.241	0.080	0.273	0.048	0.503	0.014	0.843	0.073	0.320	0.043	0.540	0.008	0.903
Smoking status (yes or no)			0.119	0.099	0.120	0.100	0.117	0.105	0.122	0.086	0.139	0.059	0.134	0.064	0.139	0.053
Hyperlipidemia (yes or no)			0.088	0.174	0.088	0.174	0.090	0.164	0.097	0.128	0.073	0.259	0.075	0.246	0.084	0.195
Hypertension (yes or no)			0.078	0.223	0.078	0.223	0.078	0.221	0.074	0.243	0.062	0.338	0.063	0.333	0.059	0.355
Type 2 diabetes mellitus (yes or no)			−0.043	0.493	−0.042	0.504	−0.046	0.458	−0.063	0.313	−0.026	0.683	−0.030	0.629	−0.047	0.452
CAD (yes or no)			−0.051	0.441	−0.051	0.444	−0.062	0.351	−0.046	0.478	−0.045	0.499	−0.055	0.409	−0.041	0.528
EATV index (cm3/m2)					−0.007	0.923					−0.010	0.886
EATLV thickness (mm)							0.074	0.281					0.062	0.363
EATLCX thickness (mm)									0.166	0.014					0.159	0.019
LVMI (g/m2)											0.088	0.168	0.088	0.167	0.081	0.196

See Table 1 for abbreviations.

Determinants of e’ and E/e’ in Patients With and Without CAD

We individually compared the relationships between BMI, EATV index, EAT_LCX, and EAT_LV and LV diastolic parameters in all age groups with and without CAD (Figure 2). Lateral e’, not septal e’, predominantly correlated with the EATV index, EAT_LCX, and EAT_LV in patients with and without CAD. Between the subgroups without (LCX−) and with (LCX+) ≥50% stenosis in the LCX, EAT_LCX and echocardiographic parameters were compared (Figure 3), revealing that EAT_LCX and lateral e’ showed a borderline increase or decrease in LCX(+) as compared with LCX(−).

Figure 2.

Simple correlations between echocardiographic parameters and epicardial adipose tissue (EAT) measurements in patients with and without coronary artery disease (CAD). The simple correlations between body mass index (BMI), whole-heart EATV index, EAT thickness surrounding the left circumflex artery (EAT_LCX), and EAT_LV and the left atrial (LA) volume index, left ventricular (LV) mass index, lateral e’ and septal e’ are shown. Simple regression analysis was made separately in non-CAD (black circles and lines) and CAD patients (blue circles and lines). R and P values are shown.

Figure 3.

Epicardial adipose tissue (EAT) thickness surrounding the left circumflex artery (EAT_LCX) and echocardiographic parameters for left ventricular diastolic function in patients without (LCX−, open circle) or with LCX+, blue circle) ≥50% stenosis in the LCX. P values: unpaired t-test between LCX (−) vs. LCX (+).

Regression Analyses for Predicting LVDD (Tables 2,4)

Table 4. Multivariate Regression Analysis for Predicting LVDD

All
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.061		0.078		0.09		0.089		0.100		0.161		0.161		0.169
P	0.001		0.012		0.007		0.007		0.002		<0.001		<0.001		<0.001
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	0.236	<0.001	0.23	0.001	0.183	0.014	0.208	0.003	0.188	0.009	0.199	0.006	0.221	0.001	0.203	0.003
Male sex (yes or no)	−0.119	0.056	−0.065	0.387	−0.044	0.565	−0.054	0.469	−0.069	0.351	−0.120	0.110	−0.128	0.084	−0.141	0.055
BMI (kg/m2)	0.074	0.251	0.078	0.239	0.022	0.761	0.031	0.667	0.015	0.831	0.009	0.895	0.017	0.805	0.003	0.962
Smoking status (yes or no)			−0.078	0.282	−0.096	0.191	−0.082	0.257	−0.077	0.286	−0.055	0.437	−0.044	0.528	−0.039	0.573
Hyperlipidemia (yes or no)			0.028	0.669	0.023	0.725	0.032	0.625	0.038	0.554	−0.014	0.827	−0.006	0.929	0.001	0.992
Hypertension (yes or no)			0.011	0.860	0.008	0.899	0.012	0.851	0.008	0.901	−0.033	0.595	−0.030	0.637	−0.033	0.598
Type 2 diabetes mellitus (yes or no)			−0.11	0.083	−0.123	0.054	−0.115	0.069	−0.128	0.043	−0.080	0.195	−0.074	0.231	−0.086	0.164
CAD (yes or no)			0.036	0.593	0.032	0.626	0.018	0.793	0.041	0.536	0.040	0.530	0.026	0.693	0.047	0.460
EATV index (cm3/m2)					0.129	0.081					0.114	0.108
EATLV thickness (mm)							0.118	0.087					0.106	0.112
EATLCX thickness (mm)									0.165	0.016					0.147	0.026
LVMI (g/m2)											0.274	<0.001	0.280	<0.001	0.273	<0.001
<65 years old
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.018		0.053		0.083		0.070		0.083		0.165		0.174		0.187
P	0.643		0.780		0.578		0.706		0.573		0.075		0.088		0.058
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	0.047	0.656	0.026	0.814	−0.036	0.759	0.013	0.904	−0.007	0.950	0.040	0.706	0.034	0.751	0.014	0.899
Male sex (yes or no)	−0.124	0.250	−0.068	0.603	−0.037	0.777	−0.080	0.543	−0.085	0.516	−0.248	0.062	−0.252	0.058	−0.257	0.052
BMI (kg/m2)	0.079	0.465	0.043	0.708	−0.037	0.764	−0.007	0.954	−0.005	0.969	0.033	0.763	−0.003	0.976	−0.007	0.951
Smoking status (yes or no)			−0.053	0.680	−0.081	0.529	−0.067	0.605	−0.043	0.736	0.008	0.946	−0.006	0.963	0.014	0.911
Hyperlipidemia (yes or no)			0.120	0.301	0.128	0.267	0.094	0.426	0.113	0.326	0.056	0.612	0.039	0.729	0.053	0.633
Hypertension (yes or no)			0.099	0.365	0.092	0.395	0.076	0.490	0.104	0.333	0.015	0.888	0.001	0.998	0.022	0.833
Type 2 diabetes mellitus (yes or no)			−0.037	0.748	−0.084	0.473	−0.064	0.583	−0.081	0.486	0.059	0.594	0.038	0.738	0.020	0.863
CAD (yes or no)			0.083	0.477	0.034	0.777	0.072	0.538	0.092	0.429	0.149	0.187	0.137	0.225	0.153	0.171
EATV index (cm3/m2)					0.220	0.102
EATLV thickness (mm)							0.156	0.220					0.114	0.351
EATLCX thickness (mm)									0.192	0.098					0.162	0.141
LVMI (g/m2)											0.359	0.001	0.366	0.001	0.358	0.001
≥65 years old
	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
Adjusted R2	0.032		0.059		0.065		0.063		0.077		0.127		0.131		0.142
P	0.174		0.326		0.349		0.368		0.213		0.016		0.022		0.011
	β	P	β	P	β	P	β	P	β	P	β	P	β	P	β	P
Age (years)	0.116	0.147	0.109	0.178	0.105	0.195	0.102	0.212	0.113	0.161	0.127	0.105	0.120	0.129	0.131	0.095
Male sex (yes or no)	−0.117	0.142	−0.058	0.551	−0.043	0.663	−0.047	0.629	−0.063	0.514	−0.097	0.305	−0.086	0.365	−0.101	0.282
BMI (kg/m2)	0.078	0.325	0.100	0.228	0.051	0.599	0.069	0.449	0.017	0.857	0.070	0.382	0.039	0.658	−0.007	0.938
Smoking status (yes or no)			−0.104	0.274	−0.112	0.239	−0.103	0.279	−0.099	0.293	−0.081	0.379	−0.080	0.386	−0.077	0.401
Hyperlipidemia (yes or no)			0.009	0.916	0.005	0.956	0.015	0.860	0.029	0.733	−0.024	0.773	−0.018	0.831	−0.004	0.957
Hypertension (yes or no)			−0.063	0.448	−0.063	0.450	−0.054	0.515	−0.068	0.411	−0.086	0.283	−0.078	0.338	−0.091	0.258
Type 2 diabetes mellitus (yes or no)			−0.147	0.074	−0.151	0.067	−0.145	0.078	−0.159	0.053	−0.125	0.116	−0.123	0.122	−0.137	0.086
CAD (yes or no)			−0.001	0.986	0.007	0.930	−0.012	0.888	0.010	0.903	−0.009	0.910	−0.019	0.810	0.002	0.981
EATV index (cm3/m2)					0.089	0.361
EATLV thickness (mm)							0.071	0.438					0.072	0.418
EATLCX thickness (mm)									0.157	0.097					0.148	0.106
LVMI (g/m2)											0.268	0.001	0.268	0.001	0.264	0.001

See Table 1 for abbreviations.

We further evaluated the effect of localized EAT depots on LVDD diagnoses in all age. Simple regression analyses showed that most echocardiographic parameters (tricuspid regurgitation velocity, LAVI, LVMI, lateral and septal e’ velocities, lateral, septal and average E/e’) were associated with LVDD diagnosis (Table 2). Additionally, LVDD correlated with the EATV index, EAT_LCX, and EAT_LV but not with BMI or other EAT measures (EAT_RCA, EAT_LAD, EAT_RA, EAT_RV, or EAT_LA).

Multiple regression analyses investigated the associations between LVDD diagnosis and the EATV index, EAT_LCX, and EAT_LV in consideration of established or presumed LVDD covariates (Table 4). Age was a determinant of LVDD, after correcting for sex and BMI (Model 1). After considering smoking status, hypertension, dyslipidemia, and type 2 diabetes mellitus, a history of CAD did not result in a statistically significant increase in the corrected R² value (Model 2). After further addition of established or presumed LVDD covariates, the EATV index (Model 3) and EAT_LV thickness (Model 4) did not improve the model, but the addition of EAT_LCX improved the corrected R² value (Model 5). The addition of LVMI, a known LVDD risk factor, increased the corrected R² value in combination with the EATV index (Model 6) or EAT_LV (Model 7). As a result, the EAT_LCX and LVMI mutually and independently improved the model adjusted R² value (Model 8). Because aging was associated positively with the EATV index, EAT_LAD, EAT_LCX and EAT_LV and negatively with septal and lateral e’ and positively with septal and lateral E/e’ (Table 2), we compared the determinants for LVDD in age subgroups of <65 and ≥65 years (Table 4). In Model 5, EAT_LCX showed a borderline significance in both subgroups.

EATV Index and EAT_LCX Cutoffs for Detecting LVDD

The optimal cutoff points for predicting LVDD in the ROC curve analysis are shown in Table 5. The EATV index and EAT_LCX cutoff points for predicting LVDD in all patients were ≥69.7 mL/m² and ≥10.1 mm, respectively, showing significant power. However, neither BMI nor EAT_LV showed significant power for diagnosing LVDD. When subjects were segregated into the CAD (≥1 coronary artery branch stenosis ≥50%) and non-CAD groups, the cutoff points for predicting LVDD were larger in the CAD group than in the non-CAD group: EATV index, 64.0 vs. 61.2 mL/m² and EAT_LCX, 14.7 vs. 10.0 mm.

Table 5. Receiver-Operating Characteristic (ROC) Curves for Predicting LVDD

	BMI	EATV index	EATLCX	EATLV
All patients (n=252)
Cutoff	≥25.7 kg/m2	≥69.7 mL/m2	≥10.1 mm	≥1.7 mm
AUC	0.528	0.603	0.578	0.569
P value	0.272	<0.001	0.038	0.205
Patients without CAD (n=112)
Cutoff	≥24.0 kg/m2	≥61.2 mL/m2	≥10.0 mm	≥2.3 mm
AUC	0.507	0.630	0.575	0.591
P value	0.860	<0.001	0.107	0.170
Patients with CAD (n=140)
Cutoff	≥23.0 kg/m2	≥64.0 mL/m2	≥14.7 mm	≥2.3 mm
AUC	0.553	0.575	0.586	0.555
P value	0.126	0.030	0.043	0.320

Cutoff values based on body mass index (BMI), epicardial adipose tissue volume index (EATV index), epicardial adipose tissue (EAT) thickness surrounding the left circumflex artery (EAT_LCX) and left ventricle (EAT_LV) were calculated to obtain the maximal area under the curve (AUC). CAD, coronary artery disease.

Discussion

We investigated the relationship between localized EAT depots and LVDD in Japanese patients with suspected CAD and obtained 3 major observations. First, among the various echocardiographic parameters, EAT_LCX and EAT_LV were strongly associated with lateral e’ and lateral E/e’, suggesting that localized fat depots are directly linked to mitral annular motion and LVDD diagnosis. Second, when comparing the measures of the whole heart (EATV index) and localized fat depots (EAT_RCA, EAT_LAD, EAT_RA, EAT_RV, EAT_LA), stronger associations were observed between LVDD and EAT_LCX or EAT_LV, matching known LVDD risk factors. Third, among the various EAT measures, only EAT_LCX had a significant cutoff value for predicting LVDD, indicating that EAT_LCX could be a pathophysiologic indicator of LVDD.

Whole Heart and Localized EAT Depots and LVDD

Among the known LVDD risk factors, such as age, hypertension, diabetes mellitus, and smoking,¹⁷ obesity is important.^18–20 In the current study, there were no significant differences in BMI or the prevalence of BMI ≥25 kg/m² between the LVDD and non-LVDD groups (Table 1); however, the EATV index was larger in the LVDD group, as in previous studies.^8,9,21,22 Previous studies have used either the whole heart EATV^9,21 or nonspecific local EAT thicknesses^8,22 to establish correlations with LVDD. Meta-analysis revealed that EAT independently predicts diastolic function parameters over adiposity measures such as BMI, and visceral or subcutaneous adipose tissue area.²³ We further compared measures of the whole heart (EATV index) and localized fat depots (EAT_RCA, EAT_LAD, EAT_RA, EAT_RA, EAT_RV, EAT_LA, EAT_LV) with LVDD diagnosis. As a result, we found that EAT_LV and EAT_LCX more strongly correlated with LVDD diagnosis than the whole-heart measure of EATV (EATV index) or other localized EAT thicknesses (EAT_RCA, EAT_LAD, EAT_RA, EAT_RV, EAT_LA) (Tables 2,5). Although a simple correlation was observed between the EATV index and LVDD, it was lost after adjusting for cofounding factors (Table 4, Model 3) and LVMI, a well-known LVDD risk factor (Table 4, Model 6). This finding suggested that the EATV index reflects the LVDD caused by LV remodeling (LV hypertrophy) and/or other confounding factors. The present study is the first to report a strong association between localized EAT measures (EAT_LCX and EAT_LV) and LVDD diagnosis, and we further showed that the link was independent of systemic LVDD risk factors such as age, male sex, smoking, and obesity.^11,17–20

Influence of Localized Fat Depots on Echocardiographic Parameters

Currently, LVDD in patients with normal LVEF (≥50%) is defined as a combination of 4 items: (1) average E/e’ ≥14, (2) septal e’ velocity <7 cm/s or lateral e’ velocity <10 cm/s, (3) tricuspid regurgitation velocity >2.8 m/s, and (4) LAVI >34 mL/m², based on the 2016 ASE/EAE/EACI guideline.¹¹ We found that EAT_LCX and EAT_LV were strongly associated with lateral e’ and lateral E/e’, despite minimal or no link with other echocardiographic parameters (Tables 2,3). This finding suggested that localized fat deposition around the lateral mitral annulus (EAT_LCX and EAT_LV) is directly related to its impaired motion and thus to LVDD diagnosis. The potential mechanisms underlying the correlation between EAT and LVDD remain to be elucidated,^3,20 but may involve mechanical and paracrine processes.^1–3

Mechanically, atrial enlargement correlates with impaired diastolic filling in obese patients.^24,25 In the current study, the LAVI (a marker of atrial enlargement) was associated with the LVMI, EATV index, and EAT_LCX (Table 2), but not with EAT_RA or EAT_LA. Collectively, these findings suggested that impairment of the restoring LV force and active LV relaxation, partially affected by localized EAT accumulation (EAT_LCX and EAT_LV) and LV hypertrophy, but not by impaired lengthening load (LA dysfunction), could cause LVDD.¹⁰ Notably, the LVMI correlated with LVDD diagnosis, independent of EAT_LCX (Table 4, Model 8). The LVMI correlated well with septal e’ and septal E/e’ (Table 3A,C, Model 8), but not with lateral e’ or lateral E/e (Table 3B,D, Model 8). However, the LVMI did not correlate with BMI or any measure of EAT thickness (Table 2). Conversely, EAT_LCX correlated well with lateral e’ and lateral E/e’ (Table 3B,D, Model 8), but not with septal e’ or septal E/e’ (Table 3A,C, Model 8). Taken together, localized EAT depots (EAT_LCX), independent of LV remodeling (hypertrophy), appeared to be associated with LVDD diagnosis through impairment of the motion of the lateral mitral annulus. Aging associated positively with the EATV index, EAT_LAD, EAT_LCX and EAT_LV and negatively with septal and lateral e’ and positively with septal and lateral E/e’ (Table 2). We previously reported that aging was related to an increase in the EATV index,¹³ and the current study further clarified that aging also correlated to increases in local EAT measures. The fact that EAT_LCX showed a borderline significance for LVDD in both the <65 and ≥65 years age groups (Table 4) suggested a role of EAT accumulation independent of aging.

Conversely, paracrine processes could also manifest through EAT accumulation. EAT is a source of several bioactive molecules that might directly influence the myocardium.^1–3,26 Under metabolic and cardiovascular disease conditions, EAT can expand, becoming hypoxic and dysfunctional,^27,28 recruiting inflammatory cells²⁹ that lead to reductions in protective cytokines and increases in detrimental cytokines, resulting in impaired cardiac function. Iozzo et al reported that the entire fat mass surrounding the heart ranges from 100 g in healthy individuals to 400–800 g in some patients.⁷ This increased weight probably leads to a mechanical burden on cardiac muscle expansion and further deteriorates LV relaxation in LVDD patients. We previously demonstrated that excessive myocardial triglyceride levels cause oxidative stress and impaired cardiac function.³⁰ Specifically, excessive levels of EAT may lead to myocardial triglyceride accumulation and negatively effects such as LV overload and hypertrophy³¹ or cardiac lipotoxicity.² Myocardial lipid content has been reported to positively correlate with EATV²¹ and EAT thickness.³² This type of close association has also been shown between EAT accumulation and the coronary microcirculation and LVDD.³³ Accumulated EAT, as compared with normal amount of EAT, contains high levels of various adipocytokines,^26,27 and therefore, its pro-inflammatory characteristics could cause coronary atherosclerosis²⁶ and LVDD.²⁷

EATVI and EAT_LCX Cutoffs for Detecting LVDD

The EATV index and EAT_LCX cutoffs for predicting LVDD were statistically significant. When the patients were classified into non-CAD and CAD groups, the cutoff points for predicting LVDD were larger and more predictive in the CAD group than in the non-CAD group. Collectively, localized fat depots, estimated by the EATV index and EAT_LCX, could be linked to LVDD diagnosis at least partly via impaired lateral mitral annulus function (lateral e’).

Because the EATV index^13,14,26 and EAT thickness¹⁵ can be related to cardiac performance via promotion of coronary atherosclerosis, we compared the relationship between EAT measures and LV diastolic parameters in patients with and without CAD (Figure 2). Lateral e’ correlated with the EATV index, EAT_LCX, and EAT_LV in patients with and without CAD. Meanwhile, EAT_LCX and lateral e’ showed a borderline increase or decrease in LCX(+) as compared with LCX(−) (Figure 3). Combined, an increase in EAT_LCX and a parallel decrease in lateral e’ may be associated with a change in local LCX atherosclerosis, at least partly. As shown in Table 5, the cutoff value of the EATV index for predicting LVDD was identified in patients without CAD (P<0.001), but was only weakly detectable in patients with CAD (P=0.035). In contrast, the cutoff value of EAT_LCX was not detectable (P=0.170) in patients without CAD, but weakly detectable in patients with CAD (P=0.043). Taking all of the above, the EATV index may be related to LVDD via both atherosclerotic and non-atherosclerotic mechanisms, but EAT_LCX may be related to LVDD differently in patients with and without CAD. The difference in cutoff points of EAT_LCX (non-CAD 10.0 mm vs. CAD 14.7 mm) might be linked to the complex results.

Study Limitations

First, the study design was cross-sectional and conducted at a single center with a relatively small number of patients. Second, the patients consisted entirely of Japanese patients; therefore, the relevance of this study to other ethnic populations requires further research. Third, although P values were significant, the areas under the ROC curves were not large enough (Table 5). The poor performance could be related to potential confounding factors and several flaws in our study such as the small number of patients. Also, the ROC results cannot be broadly applied to patients without suspected CAD and the sensitivity and specificity of EAT_LCX measurements for diagnosing CAD need to be confirmed in a future larger study. Fourth, we did not consider the effect of patient medications or lifestyle on the EAT measures. Fifth, with the modification of the 2016 ASE/EAE/EACI algorithm, we defined LVDD as the combination of “indeterminate” and “abnormal” categories, limiting our results to moderate to severe LVDD.

Conclusions

Among the various echocardiographic parameters, EAT_LCX and EAT_LV were strongly associated with lateral e’ and lateral E/e’, suggesting that localized EAT depots might be directly linked to early deterioration of diastolic mitral annular motion and early mitral inflow over the early diastolic mitral valve velocity. Further study is warranted to clarify whether these EAT depots are functionally linked to the clinical manifestation of LVDD.

Acknowledgments

We are deeply grateful to Drs. Masafumi Harada, Shoichiro Takao, and the staff at the Department of Radiology, Tokushima University, for their cooperation in cardiac CT data acquisition. We are also grateful to the staff of the Ultrasound Examination Center, Tokushima University Hospital, for acquiring the echocardiographic parameters.

Disclosure

The authors declare no competing interests.

Funding

This work was supported by JSPS Grant-in-Aid for Scientific Research (No. 17K13037 to Y.H., nos. 20590651, 23591314, 20590542, 16K01823 to M. Shimabukuro, nos. 19H03654, 25670390 and 25293184 to M. Sata), by the Takeda Science Foundation (M. Sata) and by the Vehicle Racing Commemorative Foundation (M. Sata).

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-19-0793

References

• 1.   Sacks HS, Fain JN. Human epicardial adipose tissue: A review. Am Heart J 2007; 153: 907–917.
• 2.   Shimabukuro M. Cardiac adiposity and global cardiometabolic risk: New concept and clinical implication. Circ J 2009; 73: 27–34.
• 3.   Packer M. Epicardial adipose tissue may mediate deleterious effects of obesity and inflammation on the myocardium. J Am Coll Cardiol 2018; 71: 2360–2372.
• 4.   Konishi M, Sugiyama S, Sugamura K, Nozaki T, Ohba K, Matsubara J, 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: 573–578.
• 5.   Taguchi R, Takasu J, Itani Y, Yamamoto R, Yokoyama K, Watanabe S, et al. Pericardial fat accumulation in men as a risk factor for coronary artery disease. Atherosclerosis 2001; 157: 203–209.
• 6.   Ding J, Hsu FC, Harris TB, Liu Y, Kritchevsky SB, Szklo M, et al. The association of pericardial fat with incident coronary heart disease: The Multi-Ethnic Study of Atherosclerosis (MESA). Am J Clin Nutr 2009; 90: 499–504.
• 7.   Iozzo P, Lautamaki R, Borra R, Lehto HR, Bucci M, Viljanen A, et al. Contribution of glucose tolerance and gender to cardiac adiposity. J Clin Endocrinol Metab 2009; 94: 4472–4482.
• 8.   Park HE, Choi SY, Kim M. Association of epicardial fat with left ventricular diastolic function in subjects with metabolic syndrome: Assessment using 2-dimensional echocardiography. BMC Cardiovasc Disord 2014; 14: 3.
• 9.   Fontes-Carvalho R, Fontes-Oliveira M, Sampaio F, Mancio J, Bettencourt N, Teixeira M, et al. Influence of epicardial and visceral fat on left ventricular diastolic and systolic functions in patients after myocardial infarction. Am J Cardiol 2014; 114: 1663–1669.
• 10.   Opdahl A, Remme Espen W, Helle-Valle T, Lyseggen E, Vartdal T, Pettersen E, et al. Determinants of left ventricular early-diastolic lengthening velocity. Circulation 2009; 119: 2578–2586.
• 11.   Nagueh SF, Smiseth OA, Appleton CP, Byrd BF 3rd, Dokainish H, Edvardsen T, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: An update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016; 29: 277–314.
• 12.   Raff GL, Abidov A, Achenbach S, Berman DS, Boxt LM, Budoff MJ, et al. SCCT guidelines for the interpretation and reporting of coronary computed tomographic angiography. J Cardiovasc Comput Tomogr 2009; 3: 122–136.
• 13.   Dagvasumberel M, Shimabukuro M, Nishiuchi T, Ueno J, Takao S, Fukuda D, et al. Gender disparities in the association between epicardial adipose tissue volume and coronary atherosclerosis: A 3-dimensional cardiac computed tomography imaging study in Japanese subjects. Cardiovasc Diabetol 2012; 11: 106.
• 14.   Maimaituxun G, Shimabukuro M, Salim HM, Tabata M, Yuji D, Morimoto Y, et al. Gender-linked impact of epicardial adipose tissue volume in patients who underwent coronary artery bypass graft surgery or non-coronary valve surgery. PLoS One 2017; 12: e0177170.
• 15.   Maimaituxun G, Shimabukuro M, Fukuda D, Yagi S, Hirata Y, Iwase T, et al. Local thickness of epicardial adipose tissue surrounding the left anterior descending artery is a simple predictor of coronary artery disease: New prediction model in combination with Framingham Risk Score. Circ J 2018; 82: 1369–1378.
• 16.   Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: An update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2015; 28: 1–39.e14.
• 17.   Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, et al. 2009 Focused update incorporated into the ACC/AHA 2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in collaboration with the International Society for Heart and Lung Transplantation. J Am Coll Cardiol 2009; 53: e1–e90.
• 18.   Russo C, Jin Z, Homma S, Rundek T, Elkind MS, Sacco RL, et al. Effect of obesity and overweight on left ventricular diastolic function: A community-based study in an elderly cohort. J Am Coll Cardiol 2011; 57: 1368–1374.
• 19.   Kishi S, Armstrong AC, Gidding SS, Colangelo LA, Venkatesh BA, Jacobs DR Jr, et al. Association of obesity in early adulthood and middle age with incipient left ventricular dysfunction and structural remodeling: The CARDIA study (Coronary Artery Risk Development in Young Adults). JACC Heart Fail 2014; 2: 500–508.
• 20.   Rozenbaum Z, Topilsky Y, Khoury S, Pereg D, Laufer-Perl M. Association of body mass index and diastolic function in metabolically healthy obese with preserved ejection fraction. Int J Cardiol 2019; 277: 147–152.
• 21.   Kankaanpaa M, Lehto HR, Parkka JP, Komu M, Viljanen A, Ferrannini E, et al. Myocardial triglyceride content and epicardial fat mass in human obesity: Relationship to left ventricular function and serum free fatty acid levels. J Clin Endocrinol Metab 2006; 91: 4689–4695.
• 22.   Lin HH, Lee JK, Yang CY, Lien YC, Huang JW, Wu CK. Accumulation of epicardial fat rather than visceral fat is an independent risk factor for left ventricular diastolic dysfunction in patients undergoing peritoneal dialysis. Cardiovasc Diabetol 2013; 12: 127.
• 23.   Nerlekar N, Muthalaly RG, Wong N, Thakur U, Wong DTL, Brown AJ, et al. Association of volumetric epicardial adipose tissue quantification and cardiac structure and function. J Am Heart Assoc 2018; 7: e009975.
• 24.   Iacobellis G, Leonetti F, Singh N, A MS. Relationship of epicardial adipose tissue with atrial dimensions and diastolic function in morbidly obese subjects. Int J Cardiol 2007; 115: 272–273.
• 25.   Fox CS, Gona P, Hoffmann U, Porter SA, Salton CJ, Massaro JM, et al. Pericardial fat, intrathoracic fat, and measures of left ventricular structure and function: The Framingham Heart Study. Circulation 2009; 119: 1586–1591.
• 26.   Shimabukuro M, Hirata Y, Tabata M, Dagvasumberel M, Sato H, Kurobe H, et al. Epicardial adipose tissue volume and adipocytokine imbalance are strongly linked to human coronary atherosclerosis. Arterioscler Thromb Vasc Biol 2013; 33: 1077–1084.
• 27.   Greenstein AS, Khavandi K, Withers SB, Sonoyama K, Clancy O, Jeziorska M, et al. Local inflammation and hypoxia abolish the protective anticontractile properties of perivascular fat in obese patients. Circulation 2009; 119: 1661–1670.
• 28.   Pang C, Gao Z, Yin J, Zhang J, Jia W, Ye J. Macrophage infiltration into adipose tissue may promote angiogenesis for adipose tissue remodeling in obesity. Am J Physiol Endocrinol Metab 2008; 295: E313–E322.
• 29.   Henrichot E, Juge-Aubry CE, Pernin A, Pache JC, Velebit V, Dayer JM, et al. Production of chemokines by perivascular adipose tissue: A role in the pathogenesis of atherosclerosis? Arterioscler Thromb Vasc Biol 2005; 25: 2594–2599.
• 30.   Zhou YT, Grayburn P, Karim A, Shimabukuro M, Higa M, Baetens D, et al. Lipotoxic heart disease in obese rats: Implications for human obesity. Proc Natl Acad Sci U S A 2000; 97: 1784–1789.
• 31.   van der Meer RW, Rijzewijk LJ, Diamant M, Hammer S, Schar M, Bax JJ, et al. The ageing male heart: Myocardial triglyceride content as independent predictor of diastolic function. Eur Heart J 2008; 29: 1516–1522.
• 32.   Malavazos AE, Di Leo G, Secchi F, Lupo EN, Dogliotti G, Coman C, et al. Relation of echocardiographic epicardial fat thickness and myocardial fat. Am J Cardiol 2010; 105: 1831–1835.
• 33.   Nakanishi K, Fukuda S, Tanaka A, Otsuka K, Taguchi H, Shimada K. Relationships between periventricular epicardial adipose tissue accumulation, coronary microcirculation, and left ventricular diastolic dysfunction. Can J Cardiol 2017; 33: 1489–1497.