Advertisement
Canadian Journal of Cardiology

Relationships Between Periventricular Epicardial Adipose Tissue Accumulation, Coronary Microcirculation, and Left Ventricular Diastolic Dysfunction

Published:August 07, 2017DOI:https://doi.org/10.1016/j.cjca.2017.08.001

      Abstract

      Background

      Obesity is related to left ventricular (LV) diastolic dysfunction, although its pathophysiological mechanism remains unclear. Epicardial adipose tissue (EAT) is an ectopic fat with paracrine effects on coronary circulation and myocardium. We hypothesized that left ventricle–specific (periventricular) EAT may deteriorate diastolic function by impairing coronary microcirculation.

      Methods

      In protocol 1, 74 patients without obstructive narrowing of the left anterior descending artery on multidetector computed tomography (MDCT) underwent coronary flow reserve (CFR) examination to evaluate the relationship between EAT and coronary microcirculation. In protocol 2, 372 patients who underwent both MDCT and serial transthoracic Doppler echocardiographic (TTDE) examinations were enrolled to investigate the impact of periventricular EAT on changes in diastolic function. EAT volume was measured by MDCT. CFR and diastolic function were assessed by TTDE. Deterioration of LV diastolic function was defined as a ≥ 20% decrease in early diastolic mitral annular velocity.

      Results

      CFR was significantly correlated with periventricular EAT volume (r = −0.37; P = 0.001), but not with total EAT volume (r = −0.21; P = 0.071). Periventricular EAT volume (P = 0.010) was significantly associated with CFR independent of cardiovascular risk factors. Among the 372 patients who had serial TTDE examinations, the frequency of deteriorated LV diastolic function was lowest in the lower tertile of periventricular EAT, intermediate in the middle tertile, and highest in the upper tertile (12.9%, 21.0%, and 25.8%, respectively; P = 0.037). Age, diabetes mellitus, and periventricular EAT volume were significantly associated with deterioration of LV diastolic function (all P < 0.05).

      Conclusions

      This study demonstrated the close association of periventricular EAT with impaired CFR and deteriorated LV diastolic function.

      Résumé

      Contexte

      L’obésité est liée à une dysfonction diastolique du ventricule gauche (DDVG), même si l’on méconnaît son mécanisme physiopathologique. Le tissu adipeux épicardique (TAE) est un tissu graisseux ectopique qui a des effets paracrines sur la circulation coronaire et le myocarde. Nous avons postulé que la présence de TAE autour du ventricule gauche (périventriculaire) pouvait détériorer la fonction diastolique en altérant la microcirculation coronaire.

      Méthodologie

      Dans le cadre du protocole 1, 74 patients exempts de sténose obstructive de l’artère descendante antérieure gauche à la tomographie à multidétecteurs (TGMD) ont été soumis à un examen de la réserve coronaire pour évaluer le lien entre le TAE et la microcirculation coronaire. Dans le cadre du protocole 2, 372 patients, ayant subi une TGMD et une échocardiographie doppler transthoracique à balayage séquentiel, ont été inscrits à l’étude pour évaluer les effets du TAE périventriculaire sur les changements de la fonction diastolique. Le volume du TAE a été mesuré par TGMD. La réserve coronaire et la fonction diastolique ont été évaluées par échocardiographie doppler transthoracique. La détérioration de la fonction diastolique du ventricule gauche a été définie comme une diminution de ≥ 20 % de la vélocité annulaire mitrale diastolique précoce.

      Résultats

      On a observé une corrélation significative entre la réserve coronaire et le volume du TAE périventriculaire (r = −0,37; p = 0,001), mais aucune corrélation avec le volume total de TAE (r = −0,21; p = 0,071). Le volume du TAE périventriculaire (p = 0,010) a été associé de façon significative à la réserve coronaire sans égard aux facteurs de risque cardiovasculaire. Parmi les 372 patients soumis à des examens périodiques par échocardiographie doppler transthoracique, la fréquence de détérioration de la fonction diastolique du ventricule gauche était la plus faible dans le tertile inférieur du TAE périventriculaire, intermédiaire dans le tertile moyen, et la plus élevée dans le tertile supérieur (12,9 %, 21,0 % et 25,8 %, respectivement; p = 0,037). L’âge, le diabète et le volume du TAE périventriculaire étaient significativement associés à une détérioration de la fonction diastolique du ventricule gauche (p < 0,05 dans tous les cas).

      Conclusion

      Cette étude a montré l’existence d’un lien étroit entre le TAE périventriculaire et l’altération de la réserve coronaire et la détérioration de la fonction diastolique du ventricule gauche.
      Left ventricular (LV) diastolic dysfunction is well recognized as a risk factor for heart failure.
      • Redfield M.M.
      • Jacobsen S.J.
      • Burnett Jr., J.C.
      • et al.
      Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic.
      • Bhatia R.S.
      • Tu J.V.
      • Lee D.S.
      • et al.
      Outcome of heart failure with preserved ejection fraction in a population-based study.
      However, therapeutic strategies for LV diastolic dysfunction have not been established. For example, randomized trials have failed to show positive results for the improvement of long-term outcomes in patients with LV diastolic dysfunction.
      • Cleland J.G.
      • Tendera M.
      • Adamus J.
      • et al.
      The perindopril in elderly people with chronic heart failure (PEP-CHF) study.
      • Pitt B.
      • Pfeffer M.A.
      • Assmann S.F.
      • et al.
      Spironolactone for heart failure with preserved ejection fraction.
      These unsatisfactory results warrant a reappraisal of the cause of LV diastolic dysfunction. Obesity is reported as 1 of the major risk factors for LV diastolic dysfunction.
      • Wong C.Y.
      • O'Moore-Sullivan T.
      • Leano R.
      • et al.
      Alterations of left ventricular myocardial characteristics associated with obesity.
      • Russo C.
      • Jin Z.
      • Homma S.
      • et al.
      Effect of obesity and overweight on left ventricular diastolic function: a community-based study in an elderly cohort.
      • Share B.L.
      • La Gerche A.
      • Naughton G.A.
      • Obert P.
      • Kemp J.G.
      Young women with abdominal obesity have subclinical myocardial dysfunction.
      Epicardial adipose tissue (EAT) is an ectopic visceral adipose tissue located within the pericardial sac with close proximity to the coronary arteries and the myocardium.
      • Sacks H.S.
      • Fain J.N.
      Human epicardial adipose tissue: a review.
      • Iacobellis G.
      • Willens H.J.
      Echocardiographic epicardial fat: a review of research and clinical applications.
      EAT is increased in obese individuals
      • Iacobellis G.
      • Ribaudo M.C.
      • Assael F.
      • et al.
      Echocardiographic epicardial adipose tissue is related to anthropometric and clinical parameters of metabolic syndrome: a new indicator of cardiovascular risk.
      • Rosito G.A.
      • Massaro J.M.
      • Hoffmann U.
      • et al.
      Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: the Framingham Heart Study.
      • Ng A.C.
      • Goo S.Y.
      • Roche N.
      • van der Geest R.J.
      • Wang W.Y.
      Epicardial adipose tissue volume and left ventricular myocardial function using 3-dimensional speckle tracking echocardiography.
      and is associated with the presence and incidence of coronary artery disease (CAD) independent of traditional risk factors,
      • Taguchi R.
      • Takasu J.
      • Itani Y.
      • et al.
      Pericardial fat accumulation in men as a risk factor for coronary artery disease.
      • Sarin S.
      • Wenger C.
      • Marwaha A.
      • et al.
      Clinical significance of epicardial fat measured using cardiac multislice computed tomography.
      • Ding J.
      • Hsu F.C.
      • Harris T.B.
      • et al.
      The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA).
      • Cheng V.Y.
      • Dey D.
      • Tamarappoo B.
      • et al.
      Pericardial fat burden on ECG-gated noncontrast CT in asymptomatic patients who subsequently experience adverse cardiovascular events.
      indicating that EAT may have local effects on the coronary circulation and myocardium through a paracrine mechanism. Indeed, recent studies have demonstrated that the distribution of EAT is more strongly associated with coronary atherosclerosis and supraventricular arrhythmia than is total EAT volume.
      • Mahabadi A.A.
      • Reinsch N.
      • Lehmann N.
      • et al.
      Association of pericoronary fat volume with atherosclerotic plaque burden in the underlying coronary artery: a segment analysis.
      • Wang T.D.
      • Lee W.J.
      • Shih F.Y.
      • et al.
      Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity.
      • Wong C.X.
      • Abed H.S.
      • Molaee P.
      • et al.
      Pericardial fat is associated with atrial fibrillation severity and ablation outcome.
      • Nakanishi K.
      • Fukuda S.
      • Tanaka A.
      • et al.
      Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation.
      We hypothesized that EAT accumulation, especially ventricular-specific (periventricular) EAT, may be associated with coronary microvascular dysfunction. It is also reported that impaired myocardial microcirculation is associated with myocardial dysfunction.
      • Galderisi M.
      • Cicala S.
      • Caso P.
      • et al.
      Coronary flow reserve and myocardial diastolic dysfunction in arterial hypertension.
      • Pop-Busui R.
      • Kirkwood I.
      • Schmid H.
      • et al.
      Sympathetic dysfunction in type 1 diabetes: association with impaired myocardial blood flow reserve and diastolic dysfunction.
      Because LV diastolic dysfunction is considered an early manifestation of cardiac target-organ damage, we further hypothesized that ventricular-specific EAT accumulation may deteriorate LV diastolic function by impairing the coronary microcirculation. The intent of this study was to investigate the association of periventricular EAT with coronary microcirculation and diastolic function.

      Methods

      Protocol

      This study had 2 arms. The coronary flow reserve (CFR) arm included patients who underwent multidetector computed tomography (MDCT) and CFR examinations within 1 week to investigate the association of periventricular EAT with coronary microcirculation. The diastolic function arm included a relatively large number of patients and was used to determine the relation between periventricular EAT volume and changes in diastolic function. The study was approved by the ethics committee of Osaka Ekisaikai Hospital.

      Study population

      CFR arm

      We enrolled 100 consecutive patients who underwent MDCT examination for the evaluation of CAD. The exclusion criteria were nonsinus rhythm, history of myocardial infarction, coronary revascularization in the left anterior descending (LAD) artery, greater than mild valvular disease, cardiomyopathy, and chronic kidney disease. After the MDCT examinations, 18 patients with obstructive narrowing (≥ 50%) of the LAD artery based on CT angiography and 8 patients with poor image quality were excluded. Therefore, the final population of this arm included 74 patients (48 men; mean age, 68 ± 10 years). All patients underwent CFR examinations within 1 week after the MDCT examination.

      Diastolic function arm

      To evaluate the association of periventricular EAT accumulation with changes in diastolic function, 372 patients (247 men; mean age, 67 ± 10 years) who underwent MDCT examination and serial TTDE examinations with an interval of 2.0 ± 1.0 years (ranging from 0.5-5.0 years) were included in the diastolic function arm. The exclusion criteria were nonsinus rhythm, history of myocardial infarction, greater than mild valvular disease, cardiomyopathy, LV ejection fraction < 50%, chronic kidney disease, and obstructive CAD (≥ 50% luminal stenosis) on CT angiography. Baseline TTDE images were acquired at the time of the MDCT examination. The patients were divided into 3 tertiles according to periventricular EAT volume. In both arms, the following factors were noted for each patient: hypertension (blood pressure of ≥ 140/90 mm Hg on repeated measurements or treatment with antihypertensive agents), hypercholesterolemia (serum total cholesterol level ≥ 200 mg/dL or statin drug treatment), diabetes (fasting plasma glucose level > 126 mg/dL or treatment with hypoglycemic drugs or insulin, or a combination of both), body mass index (BMI), and current smoking status.

      Scanning protocol, image reconstruction, and EAT measurements

      All MDCT scans were conducted using SOMATOM Sensation 64 systems (Siemens Medical System, Forchheim, Germany). The scan parameters were: 64 × 0.6 mm collimation; tube voltage, 120 kV; gantry rotation time, 330 ms; and tube current, 770-850 mAs. Each patient took 5 mg of bisoprolol 2 hours before the MDCT examination if the heart rate was > 60 bpm. Nitroglycerin spray was used immediately before the MDCT examination to improve visualization of the smaller-caliber coronary vessels by vasodilation. For the contrast-enhanced scans, 40-70 mL of nonionic contrast agent (Omnipaque 350, Daiichi Sankyo, Tokyo, Japan) was injected intravenously at a flow rate of 4.0- .5 mL/s followed by 30 mL of saline. All scans were obtained during a single breath hold. All MDCT analyses were performed by experienced physicians blinded to other information. The coronary arteries were divided into 16 separate segments that were ≥ 1.5 mm in diameter as measured on the MDCT angiogram and were based on the American Heart Association classification. The severity of stenosis was classified as nonobstructive plaques (< 50% luminal stenosis) or obstructive plaques (≥ 50% luminal stenosis).
      To obtain EAT volume, 2 manual editing steps (steps 1 and 2) and 2 automatically performed steps (steps 3 and 4) were conducted. 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 (Fig. 1A). In step 2, 8-12 control points were placed on the pericardium in each plane, and the software automatically generated a smooth closed pericardial contour as a region of interest (Fig. 1B). In step 3, the software automatically identified adipose tissue within the smooth closed pericardial contour by using threshold attenuation values of −30 to −250 Hounsfield units (Fig. 1C).
      • Sarin S.
      • Wenger C.
      • Marwaha A.
      • et al.
      Clinical significance of epicardial fat measured using cardiac multislice computed tomography.
      • Nakanishi K.
      • Fukuda S.
      • Tanaka A.
      • et al.
      Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation.
      • Nakanishi K.
      • Fukuda S.
      • Tanaka A.
      • 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.
      • Kitagawa T.
      • Yamamoto H.
      • Sentani K.
      • et al.
      The relationship between inflammation and neoangiogenesis of epicardial adipose tissue and coronary atherosclerosis based on computed tomography analysis.
      Finally, in step 4, the EAT volume was calculated as the sum of the EAT area. EAT was subsequently divided into periatrial or periventricular EAT, defined as pericardial fat around the atria or the ventricles, respectively (Fig. 1, D and 1E).
      • Wong C.X.
      • Abed H.S.
      • Molaee P.
      • et al.
      Pericardial fat is associated with atrial fibrillation severity and ablation outcome.
      • Nakanishi K.
      • Fukuda S.
      • Tanaka A.
      • et al.
      Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation.
      All MDCT parameters were assessed using SYNAPSE VINCENT (Fujifilm Medical, Tokyo, Japan).
      Figure thumbnail gr1
      Figure 1Measurement of the epicardial adipose tissue (EAT) volume. (A) Cross-sectional cardiac slice displays the pericardium (yellow arrows). (B) Segmentation of the epicardial adipose tissue was achieved by tracing the pericardium in the axial view. (C) Within a smooth closed pericardial contour, the software automatically identified adipose tissue by using threshold attenuation values of −30 to −250 Hounsfield units (green area). EAT was subsequently divided into (D) periatrial and (E) periventricular EAT.

      TTDE examination

      Echocardiographic examinations were performed with the Vivid 7 (General Electric, Milwaukee, WI) or the Sequoia 512 (Siemens Medical Solution, Mountain View, CA) echocardiographic device. The dimensions of the cardiac chambers were measured in the standard manner. LV ejection fraction was obtained by using the Simpson methods from apical 4- and 2-chamber views.
      • Lang R.M.
      • Badano L.P.
      • Mor-Avi V.
      • 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.
      LV mass was calculated with a validated formula
      • Devereux R.B.
      • Reichek N.
      Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method.
      : LV mass = 0.8 (1.04 [(IVS + LVDD + PW)3 − LVDD3]) + 0.6 (where IVS = interventricular septum, LVDD = left ventricular end-diastolic volume, and PW = posterior wall thickness) and indexed for body surface area. Pulsed-wave Doppler examination of mitral inflow was performed to measure early peak velocity (E) and late peak velocity (A), and their ratio (E/A) was calculated. Peak early diastolic mitral annular velocity (e′) was also measured from tissue Doppler imaging in the septal mitral annulus. All variables were acquired with at least 3 beats and averaged. Deterioration of diastolic function was defined as a ≥ 20% decrease in e′ at the follow-up TTDE compared with baseline TTDE.
      For the CFR examination, a modified foreshortened 2-chamber view was applied to explore the flow in the distal portion of the LAD artery. The angle between colour flow and the Doppler beam was corrected if it was > 20%. Coronary blood flow velocity was estimated at baseline and after an intravenous infusion of adenosine triphosphate at a rate of 0.14 mg/kg/min for 2 minutes to produce hyperemia. The mean diastolic flow velocity was measured by tracing the contour of the spectral Doppler signal. CFR was calculated as the ratio of hyperemic to basal flow velocities. Each parameter of the CFR measurements was expressed as the average value of 3 cycles. TTDE examinations were performed by 2 expert ultrasonographers with > 5 years of experience in echocardiography who were blinded to all clinical information.

      Statistical analyses

      Categorical variables are presented as numbers and percentages and were compared using the χ2 test. Continuous variables are expressed as the mean ± standard deviation and were compared using paired/unpaired t tests or analysis of variance. Univariable and multivariable linear regression analyses were used to study the association of periventricular EAT volume with CFR. The variables associated with CFR at the P < 0.1 level were selected as independent variables for the multivariable linear regression analysis. Univariable and multivariable logistic regression analyses were used to evaluate the association between periventricular EAT volume and deterioration of diastolic function, adjusting for significant potential cofactors (variables with P < 0.1 in the univariable analysis) in the multivariable models. Differences were considered significant at P < 0.05. Interobserver variability for periventricular EAT volume, CFR, and e′ was analyzed in 20 randomly selected participants assessed by 2 independent blinded observers. Intraobserver variability was also analyzed in 20 participants by the same observer at 2 different time points. The results were analyzed by both Pearson correlation analysis and the Bland-Altman method. Statistical analyses were performed using JMP 10 software (SAS Institute, Cary, NC).

      Results

      Periventricular EAT and CFR

      CFR examinations were adequately performed in all patients. The patient characteristics are shown in Supplemental Table S1. There was no significant difference in the prevalence of cardiovascular risk factors between the CFR arm and the diastolic function arm. Mean age was 68 ± 10 years, and 48 patients were men. Mean CFR was 3.17 ± 0.82 and total and periventricular EAT volumes were 118.9 ± 43.2 mL and 85.9 ± 34.3 mL, respectively. No significant correlation was observed between CFR and total EAT volume (r = −0.21; P = 0.071) (Fig. 2A), whereas there was a significant weak correlation between CFR and periventricular EAT volume (r = −0.37; P = 0.001) (Fig. 2B). CFR was also significantly associated with e′ (r = 0.27; P = 0.022). Multivariable linear regression analysis identified age (standardized β, −0.27; P = 0.014) and periventricular EAT volume (standardized β, −0.29; P = 0.010) as independent factors for the CFR value (Table 1). The association between periventricular EAT and CFR persisted even after we replaced periventricular EAT volume by periventricular EAT volume indexed for body surface area (standardized β, −0.32; P = 0.007). When total EAT volume instead of periventricular EAT volume was entered into the multivariable analysis, total EAT volume was not associated with CFR (standardized β, −0.14; P = 0.198). A possible association between the presence of CAD in the right coronary artery or left circumflex coronary artery (or both) with CFR in the LAD artery and periventricular EAT volume was investigated. However, there was no significant difference in CFR value (3.26 ± 0.88 vs 3.11 ± 0.79; P = 0.466) and periventricular EAT volume (91.3 ± 40.7 mL vs 82.7 ± 30.1 mL; P = 0.302) between patients with (n = 27) and those without (n = 47) CAD in these 2 coronary arteries.
      Figure thumbnail gr2
      Figure 2Scatter plots between coronary flow reserve (CFR) and (A) total and (B) periventricular epicardial adipose tissue (EAT) volume.
      Table 1Univariable and multivariable linear regression analysis for CFR
      VariableUnivariableMultivariable
      Standardized β (95% CI)P valueStandardized β (95% CI)P value
      Age, y−0.34 (−0.05 to −0.01)0.003−0.27 (−0.04 to −0.005)0.014
      Sex, male0.06 (−0.15 to 0.25)0.639
      Hypertension0.02 (−0.19 to 0.23)0.858
      Diabetes mellitus−0.12 (−0.30 to 0.10)0.307
      Hypercholesterolemia0.12 (−0.09 to 0.29)0.307
      Smoking−0.19 (−0.39 to 0.03)0.098−0.20 (−0.38 to 0.004)0.055
      Body mass index, kg/m20.12 (−0.03 to 0.09)0.299
      SBP, mm Hg−0.07 (−0.02 to 0.008)0.537
      DBP, mm Hg−0.01 (−0.02 to 0.02)0.913
      HR, bpm−0.09 (−0.03 to 0.01)0.450
      RPP, per 100 beats, mm Hg/min−0.11 (−0.02 to 0.006)0.345
      LDL cholesterol, mg/dL−0.01 (−0.006 to 0.005)0.919
      HDL cholesterol, mg/dL−0.03 (−0.02 to 0.01)0.770
      LV ejection fraction, %0.07 (−0.03 to 0.05)0.566
      LV mass index, g/m2−0.15 (−0.02 to 0.004)0.212
      Periventricular EAT volume, mL−0.37 (−0.01 to −0.004)0.001−0.29 (−0.01 to −0.002)0.010
      CFR, coronary flow reserve; CI, confidence interval; DBP, diastolic blood pressure; EAT, epicardial adipose tissue; HDL, high-density lipoprotein; HR, heart rate; LDL, low-density lipoprotein; LV, left ventricular; RPP, rate pressure product (SBP × HR); SBP, systolic blood pressure.

      Periventricular EAT and LV diastolic function

      Patients were divided into tertiles according to the periventricular EAT volume measured by MDCT. There was no significant difference in baseline characteristics except for BMI (22.6 ± 2.7 kg/m2 vs 24.2 ± 3.0 vs 25.4 ± 3.4 kg/m2; P < 0.001) and serum high-density lipoprotein cholesterol level (56 ± 16 mg/dL vs 50 ± 13 mg/dL vs 50 ± 13 mg/dL; P = 0.001) (Table 2). In 2-dimensional echocardiography, the LV ejection fraction and LV mass index did not differ among the 3 periventricular EAT groups, whereas follow-up echocardiography showed that there were significant differences in both E/A ratio (0.93 ± 0.37 vs 0.86 ± 0.31 vs 0.83 ± 0.29; P = 0.042) and e′ (6.26 ± 1.95 vs 5.44 ± 1.50 vs 5.13 ± 1.28; P < 0.001) among the periventricular EAT tertiles. LV diastolic function, expressed as e′, and its changes weakly but significantly correlated with periventricular EAT (Fig. 3). Deterioration of diastolic function was observed in 74 (19.9%) patients. The frequency of deteriorated diastolic function was lowest in the lower tertile, intermediate in the middle tertile, and highest in the upper tertile of periventricular EAT volume (12.9% vs 21.0% vs 25.8%; P = 0.037) (Fig. 4). In the multivariable analysis, age (adjusted odds ratio [OR], 1.04; P = 0.010), diabetes mellitus (adjusted OR, 1.74; P = 0.041), and periventricular EAT volume (adjusted OR, 1.11 per 10 mL; P = 0.021) were significantly associated with deterioration of diastolic function (Table 3). When we used the periventricular EAT volume indexed by body surface area, the periventricular EAT volume index was still significantly associated with deterioration of diastolic function (adjusted OR, 1.20 per 10 mL/m2; P = 0.018).
      Table 2Comparison of clinical characteristics and echocardiographic parameters according to periventricular EAT volume in the diastolic function arm
      CharacteristicLower tertile (n = 124)Middle tertile (n = 124)Upper tertile (n = 124)P value
      Age, y65 ± 1067 ± 1068 ± 100.142
      Male sex, n (%)77 (62.1)85 (68.6)85 (68.6)0.463
      Hypertension, n (%)90 (72.6)94 (75.8)102 (82.3)0.184
      Diabetes, n (%)36 (29.0)53 (42.7)44 (35.5)0.079
      Hypercholesterolemia, n (%)63 (50.8)71 (57.3)80 (64.5)0.092
      Smoking, n (%)32 (25.8)38 (30.6)29 (23.4)0.420
      Body mass index, kg/m222.6 ± 2.724.2 ± 3.025.4 ± 3.4< 0.001
      Systolic blood pressure, mm Hg131 ± 17134 ± 17131 ± 160.568
      Diastolic blood pressure, mm Hg76 ± 1280 ± 1177 ± 110.171
      Heart rate, bpm71 ± 972 ± 1172 ± 100.980
      LDL cholesterol, mg/dL119 ± 36118 ± 37118 ± 330.991
      HDL cholesterol, mg/dL56 ± 1650 ± 1350 ± 130.001
      Baseline echocardiography
       LV end-diastolic diameter, mm43.9 ± 4.845.7 ± 4.244.8 ± 4.40.060
       LV ejection fraction, %62.6 ± 4.961.8 ± 5.161.8 ± 5.40.396
       LV mass index, g/m288.4 ± 19.892.9 ± 0.688.5 ± 15.70.261
       E wave, cm/s65.1 ± 17.062.0 ± 19.462.3 ± 19.10.344
       A wave, cm/s76.1 ± 18.976.6 ± 20.776.2 ± 18.50.975
       E/A ratio0.90 ± 0.310.85 ± 0.300.85 ± 0.290.319
       e′, cm/s6.08 ± 1.865.55 ± 1.515.62 ± 1.440.019
      Follow-up echocardiography
       E wave, cm/s64.0 ± 21.659.5 ± 18.157.2 ± 18.40.020
       A wave, cm/s73.1 ± 21.472.8 ± 18.771.7 ± 19.70.832
       E/A ratio0.93 ± 0.370.86 ± 0.310.83 ± 0.290.042
       e′, cm/s6.26 ± 1.955.44 ± 1.505.13 ± 1.28< 0.001
       Echocardiographic interval, y1.9 ± 1.02.0 ± 1.02.0 ± 1.00.852
      Total EAT volume, mL64.7 ± 19.997.8 ± 15.1142.3 ± 35.5< 0.001
      Periventricular EAT volume, mL38.2 ± 9.464.4 ± 6.898.8 ± 23.3< 0.001
      Values are mean ± standard deviation or n (%).
      A, late diastolic transmitral flow velocity; E, early diastolic transmitral flow velocity; e′, early diastolic mitral annular velocity; EAT, epicardial adipose tissue; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LV, left ventricular.
      Figure thumbnail gr3
      Figure 3(A) Left ventricular (LV) diastolic function and periventricular epicardial adipose tissue (EAT) and (B) LV diastolic function changes and periventricular EAT. e′, early diastolic mitral annular velocity.
      Figure thumbnail gr4
      Figure 4The frequency of left ventricular diastolic function deterioration in the lower, middle, and upper tertiles of periventricular epicardial adipose tissue (EAT) volume.
      Table 3Univariable and multivariable logistic regression analysis for deterioration of LV diastolic function
      VariableUnivariableMultivariable
      OR (95% CI)P valueOR (95% CI)P value
      Age, y1.04 (1.01-1.07)0.0051.04 (1.01-1.07)0.010
      Sex, male0.99 (0.58-1.72)0.971
      Hypertension1.01 (0.56-1.89)0.974
      Diabetes mellitus1.83 (1.09-3.07)0.0221.74 (1.02-2.96)0.041
      Hypercholesterolemia0.84 (0.50-1.40)0.501
      Smoking1.03 (0.57-1.80)0.928
      BMI, kg/m20.97 (0.90-1.05)0.490
      Change in BMI, %1.04 (0.99-1.09)0.156
      SBP, mm Hg1.02 (0.996-1.04)0.116
      DBP, mm Hg0.99 (0.96-1.02)0.541
      LDL cholesterol, mg/dL1.00 (0.99-1.004)0.348
      HDL cholesterol, mg/dL1.01 (0.99-1.03)0.311
      LV ejection fraction, %0.47 (0.14-1.62)0.230
      LV mass index, g/m21.00 (0.98-1.02)0.845
      Echocardiographic interval, y1.11 (0.86-1.41)0.426
      Periventricular EAT volume, 10 mL1.13 (1.04-1.23)0.0051.11 (1.02-1.21)0.021
      BMI, body mass index; CI, confidence interval; DBP, diastolic blood pressure; EAT, epicardial adipose tissue; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LV, left ventricular; OR, odds ratio; SBP, systolic blood pressure.
      Excellent correlations were observed in the inter- and intraobserver variabilities of EAT, CFR, and e′; r = 0.99 and r = 0.99 for periventricular EAT volume; r = 0.98 and r = 0.99 for CFR; r = 0.99 and r = 0.99 for e′, respectively. In the Bland-Altman analysis, the inter- and intraobserver variabilities were −2.2 ± 10.1 mL and 0.9 ± 9.4 mL for periventricular EAT volume, 0.03 ± 0.24 and −0.12 ± 0.15 for CFR, and 0.11 ± 0.38 cm/s and −0.04 ± 0.31 cm/s for e′ (mean ± 1.96 SD, respectively). The coefficient of variation of e′ at different time points was 2.3%.

      Discussion

      The present study demonstrated for the first time, to our knowledge, that periventricular EAT accumulation is significantly associated with impaired CFR and the deterioration of LV diastolic function.

      Periventricular EAT and coronary microcirculation

      EAT is an ectopic visceral adipose tissue that directly contacts the myocardium and coronary arteries.
      • Sacks H.S.
      • Fain J.N.
      Human epicardial adipose tissue: a review.
      • Iacobellis G.
      • Willens H.J.
      Echocardiographic epicardial fat: a review of research and clinical applications.
      EAT accumulation is reported to be associated with coronary atherosclerosis and unfavorable cardiovascular outcomes independent of traditional CAD risk factors.
      • Taguchi R.
      • Takasu J.
      • Itani Y.
      • et al.
      Pericardial fat accumulation in men as a risk factor for coronary artery disease.
      • Sarin S.
      • Wenger C.
      • Marwaha A.
      • et al.
      Clinical significance of epicardial fat measured using cardiac multislice computed tomography.
      • Ding J.
      • Hsu F.C.
      • Harris T.B.
      • et al.
      The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA).
      • Cheng V.Y.
      • Dey D.
      • Tamarappoo B.
      • et al.
      Pericardial fat burden on ECG-gated noncontrast CT in asymptomatic patients who subsequently experience adverse cardiovascular events.
      Furthermore, recent investigations have suggested that the distribution of EAT is more strongly associated with coronary atherosclerosis than is total EAT volume.
      • Mahabadi A.A.
      • Reinsch N.
      • Lehmann N.
      • et al.
      Association of pericoronary fat volume with atherosclerotic plaque burden in the underlying coronary artery: a segment analysis.
      • Wang T.D.
      • Lee W.J.
      • Shih F.Y.
      • et al.
      Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity.
      Mahabadi et al.
      • Mahabadi A.A.
      • Reinsch N.
      • Lehmann N.
      • et al.
      Association of pericoronary fat volume with atherosclerotic plaque burden in the underlying coronary artery: a segment analysis.
      showed that pericoronary fat volume was positively associated with the presence of plaque independent of total EAT volume in 78 participants. Wang et al.
      • Wang T.D.
      • Lee W.J.
      • Shih F.Y.
      • et al.
      Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity.
      also demonstrated that EAT thickness in the LAD artery was the best indicator of the coronary atherosclerotic burden compared with total EAT volume and cross-sectional area. Similar results were observed in the association between periatrial EAT volume and supraventricular arrhythmia.
      • Wong C.X.
      • Abed H.S.
      • Molaee P.
      • et al.
      Pericardial fat is associated with atrial fibrillation severity and ablation outcome.
      • Nakanishi K.
      • Fukuda S.
      • Tanaka A.
      • et al.
      Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation.
      CFR is dependent on the combined effects of epicardial coronary stenosis and microvascular dysfunction. This suggests that impaired CFR reflects the presence of coronary microvascular dysfunction in the absence of obstructive coronary artery narrowing.
      • Hirata K.
      • Amudha K.
      • Elina R.
      • et al.
      Measurement of coronary vasomotor function: getting to the heart of the matter in cardiovascular research.
      Previous echocardiographic studies showed a significant relationship between EAT thickness anterior to the right ventricle and CFR in some clinical settings.
      • Sade L.E.
      • Eroglu S.
      • Bozbas H.
      • et al.
      Relation between epicardial fat thickness and coronary flow reserve in women with chest pain and angiographically normal coronary arteries.
      • Yilmaz Y.
      • Kurt R.
      • Gurdal A.
      • et al.
      Circulating vaspin levels and epicardial adipose tissue thickness are associated with impaired coronary flow reserve in patients with nonalcoholic fatty liver disease.
      Sade et al.
      • Sade L.E.
      • Eroglu S.
      • Bozbas H.
      • et al.
      Relation between epicardial fat thickness and coronary flow reserve in women with chest pain and angiographically normal coronary arteries.
      demonstrated that among 68 female participants with chest pain and angiographically normal coronary arteries, 26 (40%) of them with impaired CFR exhibited increased EAT thickness compared with those with preserved CFR (5.4 ± 1.6 mm vs 3.9 ± 0.9 mm; P < 0.001). They also showed that EAT thickness > 4.5 mm was the best cutoff value for the detection of impaired CFR. Our observation is consistent with these studies, and we also demonstrated that periventricular EAT volume, but not total EAT volume, was associated with CFR. Our results suggest the possibility of a paracrine role of periventricular EAT on the coronary microcirculation.

      Periventricular EAT and diastolic function

      Obesity is an important risk factor for LV diastolic dysfunction,
      • Wong C.Y.
      • O'Moore-Sullivan T.
      • Leano R.
      • et al.
      Alterations of left ventricular myocardial characteristics associated with obesity.
      • Russo C.
      • Jin Z.
      • Homma S.
      • et al.
      Effect of obesity and overweight on left ventricular diastolic function: a community-based study in an elderly cohort.
      • Share B.L.
      • La Gerche A.
      • Naughton G.A.
      • Obert P.
      • Kemp J.G.
      Young women with abdominal obesity have subclinical myocardial dysfunction.
      which has been shown to be a predictor of heart failure development. However, the underlying mechanism linking obesity to LV diastolic dysfunction has not been fully elucidated. Recent studies showed the association between EAT accumulation and diastolic function in some clinical conditions.
      • Konishi M.
      • Sugiyama S.
      • Sugamura K.
      • et al.
      Accumulation of pericardial fat correlates with left ventricular diastolic dysfunction in patients with normal ejection fraction.
      • Fontes-Carvalho R.
      • Fontes-Oliveira M.
      • Sampaio F.
      • et al.
      Influence of epicardial and visceral fat on left ventricular diastolic and systolic functions in patients after myocardial infarction.
      Konishi et al.
      • Konishi M.
      • Sugiyama S.
      • Sugamura K.
      • et al.
      Accumulation of pericardial fat correlates with left ventricular diastolic dysfunction in patients with normal ejection fraction.
      demonstrated that total EAT volume was independently associated with diastolic function in 229 patients who underwent MDCT examinations for the evaluation of CAD. Fontes-Carvalho et al.
      • Fontes-Carvalho R.
      • Fontes-Oliveira M.
      • Sampaio F.
      • et al.
      Influence of epicardial and visceral fat on left ventricular diastolic and systolic functions in patients after myocardial infarction.
      showed that total EAT volume was associated with the presence of diastolic dysfunction in 225 patients with a history of myocardial infarction. However, because of the cross-sectional design of these studies, they could not establish a cause-effect relationship and its potential mechanisms. We demonstrated for the first time that periventricular EAT accumulation was significantly associated with deterioration of diastolic function during a follow-up in a larger sample of patients with preserved LV systolic function. Our results suggest that increased periventricular EAT volume is 1 of the important causes of diastolic dysfunction and that impaired CFR may be involved in the association between periventricular EAT and diastolic dysfunction, which can provide useful information for the understanding of the pathophysiological mechanism of diastolic dysfunction. We also showed that there were significant differences in the E wave at the time of follow-up among the periventricular EAT tertiles, suggesting that periventricular EAT may affect LV relaxation but not LV compliance. Periventricular EAT may serve as a potential therapeutic target for diastolic dysfunction.
      Changes in BMI were not associated with deterioration of diastolic function in this study, although a nonsignificant trend was present. Previous studies showed that BMI reduction by bariatric surgery or a very low-calorie diet is associated with decreased EAT thickness and improves diastolic function in severely obese patients.
      • Willens H.J.
      • Byers P.
      • Chirinos J.A.
      • et al.
      Effects of weight loss after bariatric surgery on epicardial fat measured using echocardiography.
      • Iacobellis G.
      • Singh N.
      • Wharton S.
      • Sharma A.M.
      Substantial changes in epicardial fat thickness after weight loss in severely obese subjects.
      • Martin J.
      • Bergeron S.
      • Pibarot P.
      • et al.
      Impact of bariatric surgery on N-terminal fragment of the prohormone brain natriuretic peptide and left ventricular diastolic function.
      • Fenk S.
      • Fischer M.
      • Strack C.
      • et al.
      Successful weight reduction improves left ventricular diastolic function and physical performance in severe obesity.
      The discordance between our results and previous studies may be at least partly dependent on differences in patient characteristics and therapeutic interventions. Considering the stronger association of visceral adipose tissue with diastolic function rather than that of BMI in cross-sectional studies,
      • Ammar K.A.
      • Redfield M.M.
      • Mahoney D.W.
      • et al.
      Central obesity: Association with left ventricular dysfunction and mortality in the community.
      • Wu C.K.
      • Yang C.Y.
      • Lin J.W.
      • et al.
      The relationship among central obesity, systemic inflammation, and left ventricular diastolic dysfunction as determined by structural equation modeling.
      EAT volume reduction may have a more favourable effect on diastolic function rather than BMI reduction. Future studies are needed to evaluate whether a more aggressive pharmacologic treatment intervention and risk factor control such as body weight loss and subsequent EAT reduction can improve CFR and diastolic function.
      • Willens H.J.
      • Byers P.
      • Chirinos J.A.
      • et al.
      Effects of weight loss after bariatric surgery on epicardial fat measured using echocardiography.
      • Iacobellis G.
      • Singh N.
      • Wharton S.
      • Sharma A.M.
      Substantial changes in epicardial fat thickness after weight loss in severely obese subjects.
      • Sato T.
      • Kameyama T.
      • Ohori T.
      • Matsuki A.
      • Inoue H.
      Effects of eicosapentaenoic acid treatment on epicardial and abdominal visceral adipose tissue volumes in patients with coronary artery disease.
      Furthermore, longitudinal studies are also needed to conclude that impaired CFR deteriorates diastolic function even though a previous cross-sectional study demonstrated that impaired myocardial microcirculation is significantly associated with diastolic dysfunction,
      • Galderisi M.
      • Cicala S.
      • Caso P.
      • et al.
      Coronary flow reserve and myocardial diastolic dysfunction in arterial hypertension.
      which is consistent with our observations.

      Study limitations

      Our study has several limitations. This study included different patient cohorts (the CFR arm and the diastolic function arm), but no significant difference was observed in the prevalence of cardiovascular risk factors between the 2 arms. Because of the observational design of this study, we cannot conclude that periventricular EAT accumulation causes diastolic dysfunction through impaired CFR, which was measured by TTDE in the LAD artery alone; therefore, the association of EAT with microcirculation in other coronary vessels cannot be evaluated. Because the LAD artery runs along the anterior interventricular groove and we cannot accurately distinguish the peri-LV and peri–right ventricular EAT, both of which are adjacent to the LAD artery, we were unable to evaluate whether fat around the right ventricle affects the LV microcirculation. Furthermore, although we used the same protocol of an intravenous infusion of adenosine triphosphate for 2 minutes as in previous studies,
      • Otsuka R.
      • Watanabe H.
      • Hirata K.
      • et al.
      Acute effects of passive smoking on the coronary circulation in healthy young adults.
      • Hozumi T.
      • Eisenberg M.
      • Sugioka K.
      • et al.
      Change in coronary flow reserve on transthoracic Doppler echocardiography after a single high-fat meal in young healthy men.
      infusion over 2 minutes may affect our observations. The diagnosis of CAD was dependent on CT angiography. However, previous studies demonstrated an excellent negative predictive value for CT angiography.
      • Budoff M.J.
      • Dowe D.
      • Jollis J.G.
      • et al.
      Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial.
      Although we carefully excluded the patients with obstructive CAD from the diastolic function arm, the presence of nonobstructive CAD may affect our observations. We evaluated deterioration of diastolic function based on the value of e′, which is a commonly used marker of diastolic function, but other diastolic markers. such as propagation flow velocity and LV torsion, were not evaluated in this study.
      • Oh J.K.
      • Park S.J.
      • Nagueh S.F.
      Established and novel clinical applications of diastolic function assessment by echocardiography.
      In addition, the correlations between periventricular EAT and diastolic parameters were significant but weak, although the association remained significant even after multivariable adjustments. This may be partially explained by the short follow-up period (mean, 2.0 years). Furthermore, the patients enrolled in our study were relatively elderly and have a high prevalence of cardiovascular risk factors. The control status of these cardiovascular risk factors during the follow-up may also affect the changes in diastolic parameters and periventricular EAT volume. There was no significant association of diabetes and hypertension with CFR. Hypertension, which is an established risk factor for diastolic dysfunction, was not associated with deterioration of diastolic function, although elevated systolic blood pressure tended to be associated with deterioration of diastolic function (P = 0.116). These results may partially result from the relatively small number of patients included in the study, the lack of evaluation of the severity of diabetes and hypertension, the effect of treatments such as oral medications, or a combination.

      Conclusions

      Periventricular EAT accumulation was significantly associated with impaired coronary microcirculation and the deterioration of LV diastolic function.

      Disclosures

      The authors have no conflicts of interest to disclose.

      Supplementary Material

      References

        • Redfield M.M.
        • Jacobsen S.J.
        • Burnett Jr., J.C.
        • et al.
        Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic.
        JAMA. 2003; 289: 194-202
        • Bhatia R.S.
        • Tu J.V.
        • Lee D.S.
        • et al.
        Outcome of heart failure with preserved ejection fraction in a population-based study.
        N Engl J Med. 2006; 355: 260-269
        • Cleland J.G.
        • Tendera M.
        • Adamus J.
        • et al.
        The perindopril in elderly people with chronic heart failure (PEP-CHF) study.
        Eur Heart J. 2006; 27: 2338-2345
        • Pitt B.
        • Pfeffer M.A.
        • Assmann S.F.
        • et al.
        Spironolactone for heart failure with preserved ejection fraction.
        N Engl J Med. 2014; 370: 1383-1392
        • Wong C.Y.
        • O'Moore-Sullivan T.
        • Leano R.
        • et al.
        Alterations of left ventricular myocardial characteristics associated with obesity.
        Circulation. 2004; 110: 3081-3087
        • Russo C.
        • Jin Z.
        • Homma S.
        • 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
        • Share B.L.
        • La Gerche A.
        • Naughton G.A.
        • Obert P.
        • Kemp J.G.
        Young women with abdominal obesity have subclinical myocardial dysfunction.
        Can J Cardiol. 2015; 31: 1195-1201
        • Sacks H.S.
        • Fain J.N.
        Human epicardial adipose tissue: a review.
        Am Heart J. 2007; 153: 907-917
        • Iacobellis G.
        • Willens H.J.
        Echocardiographic epicardial fat: a review of research and clinical applications.
        J Am Soc Echocardiogr. 2009; 22: 1311-1319
        • Iacobellis G.
        • Ribaudo M.C.
        • Assael F.
        • 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
        • Rosito G.A.
        • Massaro J.M.
        • Hoffmann U.
        • 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
        • Ng A.C.
        • Goo S.Y.
        • Roche N.
        • van der Geest R.J.
        • Wang W.Y.
        Epicardial adipose tissue volume and left ventricular myocardial function using 3-dimensional speckle tracking echocardiography.
        Can J Cardiol. 2016; 32: 1485-1492
        • Taguchi R.
        • Takasu J.
        • Itani Y.
        • et al.
        Pericardial fat accumulation in men as a risk factor for coronary artery disease.
        Atherosclerosis. 2001; 157: 203-209
        • Sarin S.
        • Wenger C.
        • Marwaha A.
        • et al.
        Clinical significance of epicardial fat measured using cardiac multislice computed tomography.
        Am J Cardiol. 2008; 102: 767-771
        • Ding J.
        • Hsu F.C.
        • Harris T.B.
        • et al.
        The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA).
        The Am J Clin Nutr. 2009; 90: 499-504
        • Cheng V.Y.
        • Dey D.
        • Tamarappoo B.
        • 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
        • Mahabadi A.A.
        • Reinsch N.
        • Lehmann N.
        • et al.
        Association of pericoronary fat volume with atherosclerotic plaque burden in the underlying coronary artery: a segment analysis.
        Atherosclerosis. 2010; 211: 195-199
        • Wang T.D.
        • Lee W.J.
        • Shih F.Y.
        • et al.
        Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity.
        Atherosclerosis. 2010; 213: 279-287
        • Wong C.X.
        • Abed H.S.
        • Molaee P.
        • et al.
        Pericardial fat is associated with atrial fibrillation severity and ablation outcome.
        J Am Coll Cardiol. 2011; 57: 1745-1751
        • Nakanishi K.
        • Fukuda S.
        • Tanaka A.
        • et al.
        Peri-atrial epicardial adipose tissue is associated with new-onset nonvalvular atrial fibrillation.
        Circ J. 2012; 76: 2748-2754
        • Galderisi M.
        • Cicala S.
        • Caso P.
        • et al.
        Coronary flow reserve and myocardial diastolic dysfunction in arterial hypertension.
        Am J Cardiol. 2002; 90: 860-864
        • Pop-Busui R.
        • Kirkwood I.
        • Schmid H.
        • et al.
        Sympathetic dysfunction in type 1 diabetes: association with impaired myocardial blood flow reserve and diastolic dysfunction.
        J Am Coll Cardiol. 2004; 44: 2368-2374
        • Nakanishi K.
        • Fukuda S.
        • Tanaka A.
        • 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
        • Kitagawa T.
        • Yamamoto H.
        • Sentani K.
        • et al.
        The relationship between inflammation and neoangiogenesis of epicardial adipose tissue and coronary atherosclerosis based on computed tomography analysis.
        Atherosclerosis. 2015; 243: 293-299
        • Lang R.M.
        • Badano L.P.
        • Mor-Avi V.
        • 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
        • Devereux R.B.
        • Reichek N.
        Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method.
        Circulation. 1977; 55: 613-618
        • Hirata K.
        • Amudha K.
        • Elina R.
        • et al.
        Measurement of coronary vasomotor function: getting to the heart of the matter in cardiovascular research.
        Clin Sci (Lond). 2004; 107: 449-460
        • Sade L.E.
        • Eroglu S.
        • Bozbas H.
        • et al.
        Relation between epicardial fat thickness and coronary flow reserve in women with chest pain and angiographically normal coronary arteries.
        Atherosclerosis. 2009; 204: 580-585
        • Yilmaz Y.
        • Kurt R.
        • Gurdal A.
        • et al.
        Circulating vaspin levels and epicardial adipose tissue thickness are associated with impaired coronary flow reserve in patients with nonalcoholic fatty liver disease.
        Atherosclerosis. 2011; 217: 125-129
        • Konishi M.
        • Sugiyama S.
        • Sugamura K.
        • et al.
        Accumulation of pericardial fat correlates with left ventricular diastolic dysfunction in patients with normal ejection fraction.
        J Cardiol. 2012; 59: 344-351
        • Fontes-Carvalho R.
        • Fontes-Oliveira M.
        • Sampaio F.
        • 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
        • Willens H.J.
        • Byers P.
        • Chirinos J.A.
        • et al.
        Effects of weight loss after bariatric surgery on epicardial fat measured using echocardiography.
        Am J Cardiol. 2007; 99: 1242-1245
        • Iacobellis G.
        • Singh N.
        • Wharton S.
        • Sharma A.M.
        Substantial changes in epicardial fat thickness after weight loss in severely obese subjects.
        Obesity. 2008; 16: 1693-1697
        • Martin J.
        • Bergeron S.
        • Pibarot P.
        • et al.
        Impact of bariatric surgery on N-terminal fragment of the prohormone brain natriuretic peptide and left ventricular diastolic function.
        Can J Cardiol. 2013; 29: 969-975
        • Fenk S.
        • Fischer M.
        • Strack C.
        • et al.
        Successful weight reduction improves left ventricular diastolic function and physical performance in severe obesity.
        Int Heart J. 2015; 56: 196-202
        • Ammar K.A.
        • Redfield M.M.
        • Mahoney D.W.
        • et al.
        Central obesity: Association with left ventricular dysfunction and mortality in the community.
        Am Heart J. 2008; 156: 975-981
        • Wu C.K.
        • Yang C.Y.
        • Lin J.W.
        • et al.
        The relationship among central obesity, systemic inflammation, and left ventricular diastolic dysfunction as determined by structural equation modeling.
        Obesity. 2012; 20: 730-737
        • Sato T.
        • Kameyama T.
        • Ohori T.
        • Matsuki A.
        • Inoue H.
        Effects of eicosapentaenoic acid treatment on epicardial and abdominal visceral adipose tissue volumes in patients with coronary artery disease.
        J Atheroscler Thromb. 2014; 21: 1031-1043
        • Otsuka R.
        • Watanabe H.
        • Hirata K.
        • et al.
        Acute effects of passive smoking on the coronary circulation in healthy young adults.
        JAMA. 2001; 286: 436-441
        • Hozumi T.
        • Eisenberg M.
        • Sugioka K.
        • et al.
        Change in coronary flow reserve on transthoracic Doppler echocardiography after a single high-fat meal in young healthy men.
        Ann Intern Med. 2002; 136: 523-528
        • Budoff M.J.
        • Dowe D.
        • Jollis J.G.
        • et al.
        Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial.
        J Am Coll Cardiol. 2008; 52: 1724-1732
        • Oh J.K.
        • Park S.J.
        • Nagueh S.F.
        Established and novel clinical applications of diastolic function assessment by echocardiography.
        Circ Cardiovasc Imaging. 2011; 4: 444-455