Abstract
Aims
This study aimed to identify the impact of increased epicardial adipose tissue (EAT) and its regional distribution on cardiac function in patients with diastolic dysfunction.
Methods and results
Sixty‐eight patients with exertional dyspnoea (New York Heart Association ≥II), preserved ejection fraction (≥50%), and diastolic dysfunction (E/e′ ≥ 8) underwent rest and stress right heart catheterization, transthoracic echocardiography, and cardiovascular magnetic resonance (CMR). EAT volumes were depicted from CMR short‐axis stacks. First, the impact of increased EAT above the median was investigated. Second, the association of ventricular and atrial EAT with myocardial deformation at rest and during exercise stress was analysed in a multivariable regression analysis. Patients with high EAT had higher HFA‐PEFF and H2FPEFF scores as well as N‐terminal prohormone of brain natriuretic peptide levels (all P < 0.048). They were diagnosed with manifest heart failure with preserved ejection fraction (HFpEF) more frequently (low EAT: 37% vs. high EAT: 64%; P = 0.029) and had signs of adverse remodelling indicated by higher T1 times (P < 0.001). No differences in biventricular volumetry and left ventricular mass (all P > 0.074) were observed. Patients with high EAT had impaired atrial strain at rest and during exercise stress, and impaired ventricular strain during exercise stress. Regionally increased EAT was independently associated with functional impairment of the adjacent chambers.
Conclusions
Patients with diastolic dysfunction and increased EAT show more pronounced signs of diastolic functional failure and adverse structural remodelling. Despite similar morphological characteristics, patients with high EAT show significant cardiac functional impairment, in particular in the atria. Our results indicate that regionally increased EAT directly induces atrial functional failure, which represents a distinct pathophysiological feature in HFpEF.
Keywords: Diastolic dysfunction, HFpEF, Epicardial fat, Cardiovascular magnetic resonance, Cardiac function
Introduction
Epicardial adipose tissue (EAT) has emerged as an important risk factor in multiple diseases, while it seems to accelerate cardiac functional deterioration and increase the probability of adverse events. 1 , 2 , 3 , 4 Studies revealed that patients with increased EAT show haemodynamic alterations with functional impairment of both ventricles, as well as increased cardiac filling pressures at rest and during exercise stress with concomitantly reduced exercise capacity. 5 , 6 , 7 , 8 Furthermore, EAT directly impacts local cardiac structure 4 and is suspected to cause inflammation and remodelling, both playing a role in the development and progression of cardiovascular disease. 9 , 10 , 11
The subsequent cardiac remodelling was observed to induce typical pathophysiological and morphological conditions observed in patients with diastolic dysfunction and heart failure with preserved ejection fraction (HFpEF). 5 , 12 This risk for diastolic dysfunction due to increased EAT volume was found to be present in early adolescence and otherwise healthy individuals. 13 , 14
Albeit those first findings offering an enhanced understanding of the pathogenesis of diastolic dysfunction and HFpEF, a detailed exploration of the implications of regional epicardial fat on the function of the associated cardiac chambers is lacking.
We hypothesized that the total amount and the distribution of EAT around distinct myocardial regions within patients with diastolic dysfunction would contribute to more severe functional impairment and structural changes.
Methods
Patient cohort
Sixty‐eight patients with dyspnoea at exertion [New York Heart Association (NYHA) ≥II], preserved ejection fraction (EF) (≥50%), and signs of diastolic dysfunction (E/e′ ≥ 8) were recruited after initial referral for echocardiographic evaluation of dyspnoea within the HFpEF‐Stress trial. 15 During screening, exclusion criteria comprised other causes of dyspnoea including pulmonary disease identified on spirometry (FEV1 < 80%), coronary artery disease, or cardiovascular disease including cardiomyopathies, ischaemic, non‐ischaemic, or valvular heart diseases, as well as typical contraindications for cardiovascular magnetic resonance (CMR). 15 Patients were diagnosed with HFpEF according to thresholds of pulmonary capillary wedge pressure (PCWP) ≥15 mmHg at rest or ≥25 mmHg during exercise stress right heart catheterization (RHC). 16 Rest and stress echocardiographic assessments were conducted simultaneously to RHC and followed by a CMR within 24 h. 15
Right heart catheterization and echocardiographic assessments
Using a standardized 5‐W‐increasing exercise ramp protocol on a bicycle ergometer in supine position, patients had to achieve and maintain heart rates between 100 and 110 b.p.m. 15 , 17
Assessments of pulmonary artery (PA) pressure, PCWP, cardiac index (CI), and pulmonary vascular resistance were conducted during RHC.
Simultaneously to RHC, transthoracic echocardiographic assessments at rest and during exercise stress comprised measurements in apical long‐axis (LAX) and parasternal short‐axis (SAX) views. 15 Ventricular and atrial dimensions were obtained, as well as functional parameters for systolic and diastolic function using appropriate techniques such as M‐mode, pulsed‐wave, and continuous‐wave Doppler as suggested by current guidelines. 16 Global longitudinal strain (GLS) at rest and during exercise stress was measured by echocardiographic speckle tracking of the left ventricle in longitudinal two‐, three‐, and four‐chamber views. 15
Cardiovascular magnetic resonance
CMR images were obtained on a 3.0 T MAGNETOM Skyra (Siemens Healthcare, Erlangen, Germany) using a 32‐channel cardiac surface receiver coil. For post‐processing, a commercially available software was used (Medis, QMass®, Medical Imaging Systems, Leiden, The Netherlands). 18
During the CMR survey, patients were exposed to exercise stress using the same ramp protocol as for RHC before on a CMR‐compatible supine ergometer (Lode, Leiden, The Netherlands).
Volumetric measures were obtained from balanced steady‐state free precession (bSSFP) cine‐sequence SAX stacks with full coverage of the atria and ventricles [25 frames per cardiac cycle, echo time (TE) of 1.5 ms, repetition time (TR) of 55 ms, flip angle of 55°, and slice thickness of 7 mm with a 7.7 mm inter‐slice gap].
Real‐time (RT) images at rest and during exercise stress were acquired using heavily under‐sampled bSSFP sequences and iterative reconstruction as described by Uecker et al. 19
Manual LAX strains for all four chambers (left atrium/left ventricle/right atrium/right ventricle) of the heart were assessed using OsiriX MD (Pixmeo SARL, Bernex, Switzerland). 15
For T1 measurements, a septal region of interest (ROI) was manually delineated in a single midventricular SAX slice obtained using a 5(3)3 Modified Look‐Locker Inversion Recovery sequence [field of view (FOV) of 360 × 306.6 mm2, in‐plane resolution of 1.41 × 1.41 × 8 mm3, TR of 280 ms, TE of 1.12 ms, inversion time (TI) of 180 ms, flip angle of 35°, and bandwidth of 1085 Hz/pixel with total acquisition of 11 heart beats]. 20
Calculation of epicardial adipose tissue
EAT volume was calculated as previously proposed and validated. 1 , 5 Prior to the quantification of EAT volumes, the presence of EAT was validated in pre‐contrast and post‐contrast T1 maps. Hereby, T1 times of subcutaneous fat and suspected epicardial fat were compared by manually drawn ROIs. EAT was defined as the adipose tissue between the outer layer of the myocardium and the visceral layer of the pericardium and manually delineated in end‐diastolic SAX slices from the most basal slice around the atria up to the apical slices of the ventricle (cf. Figure 1 ). The localization of the mitral valve was used to differentiate between atrial and ventricular EAT. The total EAT volume per patient was computed by the modified Simpson rule. 21
Figure 1.
Delineation of epicardial adipose tissue (EAT). The segmentation of atrial EAT in a four‐chamber view (A) and short‐axis stack (B). The segmentation of ventricular EAT in a four‐chamber view (C) and short‐axis stack (D).
The calculation of EAT was performed blinded to patient characteristics and other functional and morphological results.
Statistical analysis
All statistical analyses were performed using SPSS Version 28 (IBM, Armonk, NY, USA) and GraphPad Prism 9 (GraphPad Software, CA, USA). A Shapiro–Wilk test was used to test for normal distribution. Categorical variables are displayed as frequencies and percentages and were compared by the χ 2 test. Continuous variables are shown as median with corresponding inter‐quartile ranges (IQRs) and were compared using the nonparametric Mann–Whitney U test.
Associations between markers for diastolic dysfunction and EAT were tested with multiple univariable linear regressions. PCWP and E/e′ as conventional diastolic functional parameters and the CMR‐derived long axial strain (LAS) of all four individual chambers at rest and during exercise stress were tested for association with total and regional EAT using a multivariable linear regression analysis. Patient‐specific characteristics and risk factors potentially influencing the presence of EAT [age, sex, body mass index (BMI), hyperlipoproteinaemia (HLP), and diabetes] were included in the analysis. In categorical variables, association was tested for the presence of HLP, the presence of diabetes, and male sex. P values below 0.05 were considered statistically significant.
Results
Baseline characteristics
For further assessments, the cohort was dichotomized at the median of the total EAT volume for comparisons between the groups with high EAT (>60.7 mL/m2) and low EAT (≤60.7 mL/m2) (cf. Figure 2 ).
Figure 2.
Main findings in patients with increased epicardial adipose tissue (EAT) volumes. The patient cohort was dichotomized into patients with low and high EAT volumes. Patients with high EAT volumes revealed specific morphological and functional characteristics as diagnosed by right heart catheterization (RHC) and cardiovascular magnetic resonance (CMR). HFpEF, heart failure with preserved ejection fraction; LV, left ventricular.
Baseline characteristics can be appreciated from Table 1 . Patients with increased EAT volumes had a higher prevalence of atrial fibrillation (P = 0.045) and revealed higher HFA‐PEFF and H2FPEFF scores (HFA‐PEFF: 3 vs. 4; P = 0.004 and H2FPEFF: 3 vs. 5; P = 0.017). Furthermore, patients with high EAT volumes were diagnosed with manifest HFpEF more frequently using thresholds of RHC (low EAT: 37% vs. high EAT: 64%; P = 0.029) (see Figure 2 ).
Table 1.
Baseline characteristics
| Parameter | Low EAT (n = 35) | High EAT (n = 33) | P value |
|---|---|---|---|
| Age (years) | 68 (60–73) | 69 (66–73) | 0.256 |
| BMI (kg/m2 BSA) | 27.5 (26.1–33.2) | 28.4 (26.9–32.8) | 0.384 |
| NYHA (II/III) | 25/10 | 23/10 | 0.867 |
| Female | 21 (60) | 23 (70) | 0.403 |
| Diabetes, n (%) | 4 (11) | 6 (18) | 0.432 |
| AHT, n (%) | 27 (77) | 27 (82) | 0.634 |
| AF, n (%) | 7 (20) | 14 (42) | 0.045* |
| HLP, n (%) | 19 (54) | 23 (70) | 0.191 |
| Smoking, n (%) | 3 (9) | 6 (18) | 0.242 |
| HFA‐PEFF score | 3 (2–5) | 4 (4–6) | 0.004* |
| H2FPEFF score | 3 (2–5) | 5 (3–6) | 0.017* |
| HFpEF, n (%) | 13 (37) | 21 (64) | 0.029* |
AF, atrial fibrillation; AHT, arterial hypertension; BMI, body mass index; BSA, body surface area; EAT, epicardial adipose tissue; H2FPEFF score, body mass index >30 kg/m2, ≥2 antihypertensive drugs, atrial fibrillation, pulmonary artery systolic pressure >35 mmHg, >60 years, and E/e′ > 9; HFA‐PEFF score, Heart Failure Association diagnostic algorithm score; HFpEF, heart failure with preserved ejection fraction; HLP, hyperlipoproteinaemia; NYHA, New York Heart Association class.
Patient baseline characteristics after dichotomization at the median of 60.7 mL/m2 total EAT.
P values below 0.05 were considered statistically significant.
Impact of increased total epicardial adipose tissue on laboratory markers, functional and structural parameters by right heart catheterization, echocardiography, and cardiovascular magnetic resonance
As presented in Table 2 and Figure 3 , patients with higher EAT volume had increased levels of the N‐terminal prohormone of brain natriuretic peptide (NT‐proBNP) (P = 0.048), while echocardiographic measurements did not reveal significant differences between both groups (P > 0.124).
Table 2.
Laboratory markers, echocardiography, and right heart catheterization
| Parameter | Low EAT (n = 35) | High EAT (n = 33) | P value |
|---|---|---|---|
| Laboratory markers | |||
| NT‐proBNP (ng/L) | 107.5 (58.1–186.5) | 136.6 (77.2–495.0) | 0.048* |
| Creatinine (mg/dL) | 0.9 (0.8–1.0) | 0.8 (0.7–1.0) | 0.461 |
| hsCRP (mg/L) | 1.49 (0.67–2.71) | 1.48 (0.81–4.2) | 0.318 |
| Echocardiography | |||
| E/e′ rest | 10.4 (8.2–12.5) | 10.8 (9.3–13.0) | 0.124 |
| E/e′ stress | 12.0 (10.0–14.4) | 13.8 (10.3–16.3) | 0.449 |
| LAVI (mL/m2) | 37.5 (28.4–50.3) | 38.5 (36.1–47.4) | 0.311 |
| PAPsys (mmHg) | 23.9 (21.0–28.0) | 24.8 (21.6–33.0) | 0.380 |
| LV‐GLS rest (%) | −15.9 (−19.8 to −13.3) | −15.4 (−18.2 to −12.2) | 0.345 |
| LV‐GLS stress (%) | −14.9 (−19.6 to −12.7) | −15.6 (−17.3 to −13.6) | 1.0 |
| Right heart catheterization | |||
| PCWP rest (mmHg) | 9.0 (7.0–12.0) | 11.0 (10.0–16.5) | 0.043* |
| PCWP stress (mmHg) | 22.0 (13.0–26.0) | 25.0 (19.5–28.0) | 0.055 |
| PA rest (mmHg) | 19.0 (16.0–21.0) | 20.0 (17.0–24.5) | 0.091 |
| PA stress (mmHg) | 36.0 (26.0–42.0) | 42.0 (36.5–47.0) | 0.005* |
| CI rest (L/m2 BSA) | 3.0 (2.7–3.6) | 2.7 (2.4–3.0) | 0.017* |
| CI stress (L/m2 BSA) | 5.6 (4.8–6.5) | 5.1 (4.0–6.1) | 0.078 |
| PVR rest (Wood units) | 1.5 (1.1–1.8) | 1.7 (1.2–2.5) | 0.130 |
| PVR stress (Wood units) | 1.3 (1.1–1.7) | 1.7 (1.3–2.5) | 0.030* |
BSA, body surface area; CI, cardiac index; E, passive mitral inflow; e′, septal and lateral mitral annulus velocity; EAT, epicardial adipose tissue; GLS, global longitudinal strain; hsCRP, high‐sensitivity C‐reactive protein; LAVI, left atrial volume index; LV, left ventricular; NT‐proBNP, N‐terminal prohormone of brain natriuretic peptide; PA, pulmonary artery; PAPsys, pulmonary artery systolic pressure; PCWP, pulmonary capillary wedge pressure; PVR, pulmonary vascular resistance.
Laboratory markers, echocardiography, and right heart catheterization assessments of the patients after dichotomization at the median of 60.7 mL/m2 total EAT.
P values below 0.05 were considered statistically significant.
Figure 3.
Comparison of N‐terminal prohormone of brain natriuretic peptide (NT‐proBNP), pulmonary capillary wedge pressure (PCWP), and T1 times within patients with high and low epicardial adipose tissue (EAT). Displayed are the median NT‐proBNP, PCWP, and T1 times with inter‐quartile ranges within patients with high EAT (>60.7 mL/m2) and low EAT (<60.7 mL/m2). P values below 0.05 were considered statistically significant.
During RHC, patients with increased EAT volume had increased PCWP at rest (P = 0.043) with the same trend for PCWP during stress (P = 0.055) (see Figure 3 ). Additionally, they revealed concordantly increased PA pressure during exercise stress in particular (P = 0.005) paired with a higher pulmonary vascular resistance (P = 0.030). The total cardiac output was reduced in the high EAT volume group at rest (P = 0.017) with the same trend for the CI during exercise stress (P = 0.078). Detailed results can be obtained from Table 2 .
As measured in CMR, in patients with increased total EAT volume, no difference in right ventricular (RV) and left ventricular (LV) end‐diastolic volume (EDV) and end‐systolic volume (ESV), as well as LV mass, could be observed (all P > 0.074) compared with patients with lower EAT volumes (see Figure 2 and Table 3 ).
Table 3.
Cardiovascular magnetic resonance imaging
| Parameter | Low EAT (n = 35) | High EAT (n = 33) | P value |
|---|---|---|---|
| Left ventricle | |||
| LV mass (g/m2 BSA) | 55.6 (50.7–71.4) | 58.1 (50.7–68.1) | 0.878 |
| LV EDV (mL/m2 BSA) | 72.4 (61.1–78.6) | 65.0 (56.3–74.0) | 0.074 |
| LV ESV (mL/m2 BSA) | 20.1 (16.4–26.4) | 20.3 (14.1–24.8) | 0.439 |
| LVEF (%) | 69.3 (66.1–75.9) | 68.6 (65.1–76.3) | 0.985 |
| LV SV (mL/m2 BSA) | 51.3 (43.2–56.0) | 45.0 (40.6–51.9) | 0.041* |
| LV LAS rest (%) | 14.2 (11.7–16.7) | 13.7 (11.7–14.8) | 0.089 |
| LV LAS stress (%) | 18.7 (15.3–21.6) | 15.6 (12.7–18.3) | 0.002* |
| T1 time (ms) | 1189 (1167–1211) | 1224 (1196–1264) | <0.001* |
| Right ventricle | |||
| RV EDV (mL/m2 BSA) | 67.8 (58.5–75.8) | 62.5 (53.7–69.6) | 0.133 |
| RV ESV (mL/m2 BSA) | 22.8 (18.8–28.2) | 20.6 (16.3–25.7) | 0.175 |
| RV SV (mL/m2 BSA) | 44.0 (39.0–51.8) | 42.4 (36.4–48.9) | 0.390 |
| RV EF (%) | 64.8 (62.2–68.4) | 66.7 (59.6–71.7) | 0.540 |
| RV LAS rest (%) | 26.0 (21.8–27.5) | 24.6 (20.1–28.7) | 0.372 |
| RV LAS stress (%) | 28.8 (25.6–31.4) | 24.8 (20.3–28.7) | 0.006* |
| Left atrium | |||
| LAVI (mL/m2 BSA) | 42.7 (29.7–54.0) | 41.0 (33.2–51.3) | 0.830 |
| LA LAS rest (%) | 21.6 (17.2–26.6) | 16.9 (13.1–21.5) | 0.016* |
| LA LAS stress (%) | 25.9 (18.9–31.5) | 18.4 (13.8–27.0) | 0.006* |
| Right atrium | |||
| RAVI (mL/m2 BSA) | 47.4 (40.0–61.6) | 48.1 (40.5–55.7) | 0.969 |
| RA LAS rest (%) | 31.0 (28.2–35.4) | 24.7 (21.9–31.0) | 0.002* |
| RA LAS stress (%) | 37.0 (30.3–41.5) | 32.0 (26.0–37.6) | 0.032* |
BSA, body surface area; CMR, cardiovascular magnetic resonance; EAT, epicardial adipose tissue; EDV, end‐diastolic volume; EF, ejection fraction; ESV, end‐systolic volume; LA, left atrial; LAS, long axial strain; LAVI, left atrial volume index; LV, left ventricular; LVEF, left ventricular ejection fraction; RA, right atrial; RAVI, right atrial volume index; RV, right ventricular; SV, stroke volume.
Comprehensive CMR of patients after dichotomization at the median of 60.7 mL/m2 total EAT.
P values below 0.05 were considered statistically significant.
In the high EAT volume group, ventricular strain of the right ventricle and the left ventricle was impaired during exercise stress (right ventricle: P = 0.006; left ventricle: P = 0.002) but not at rest (right ventricle: P = 0.372; left ventricle: P = 0.089).
Meanwhile, strain of the right atrium and left atrium was impaired at rest already (right atrium: P = 0.002; left atrium: P = 0.016) and remained lower during exercise stress in patients with high EAT volumes (right atrium: P = 0.032; left atrium: P = 0.006).
Patients with higher EAT volumes had signs of increased myocardial fibrosis as indicated by longer T1 times (P < 0.001) (see Figure 3 ). The detailed results of the CMR analysis can be depicted in Table 3 .
Association of total, atrial, and ventricular epicardial adipose tissue with diastolic functional parameters and regional cardiac strain in cardiovascular magnetic resonance
EAT was majorly located in front of the ventricles compared with the atria (44.9 vs. 16.8 mL/m2; P < 0.001) (cf. Figure 4 ).
Figure 4.
Distribution of epicardial adipose tissue (EAT) in front of the ventricles and atria. Displayed is the median EAT with the inter‐quartile range of the ventricles and the atria in the overall cohort.
In multivariable regression analysis, increasing total EAT volume was independently associated with higher PCWP during exercise stress (P = 0.029) but not at rest (P = 0.310). For echocardiography, increasing total EAT volume showed a trend towards higher E/e′ at rest (P = 0.081), but no association was found during exercise stress (P = 0.593) (see Table 4 ). Furthermore, increasing total EAT volume was independently associated with impaired LV and left atrial (LA) LAS at rest and during exercise stress (all P < 0.039). RV LAS only revealed an association with total EAT volume during exercise stress, while impaired right atrial (RA) LAS was associated with higher total EAT volume at rest (P < 0.001) and showed a strong trend for an association during exercise stress (P = 0.050) (cf. Table 5 ).
Table 4.
Multivariable regression analysis of the association of regional cardiac strain with total, ventricular, and atrial epicardial adipose tissue and patient characteristic in the overall cohort
| PCWP rest | PCWP stress | E/e′ rest | E/e′ stress | |||||
|---|---|---|---|---|---|---|---|---|
| Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | |
| Total EAT | 0.117 | 0.310 | 0.254 | 0.029* | 0.206 | 0.081 | 0.078 | 0.593 |
| Age | 0.405 | 0.002* | 0.409 | 0.002* | 0.282 | 0.029* | 0.095 | 0.552 |
| BMI | 0.449 | <0.001* | 0.269 | 0.030* | 0.044 | 0.726 | −0.198 | 0.209 |
| Diabetes | 0.001 | 0.996 | 0.047 | 0.684 | −0.050 | 0.673 | 0.031 | 0.832 |
| HLP | −0.004 | 0.973 | −0.067 | 0.555 | −0.048 | 0.682 | −0.028 | 0.849 |
| Sex | −0.010 | 0.932 | −0.084 | 0.461 | −0.262 | 0.026* | −0.261 | 0.078 |
BMI, body mass index; EAT, epicardial adipose tissue; HLP, hyperlipoproteinaemia; PCWP, pulmonary capillary wedge pressure.
Multivariable regression analysis of the association of PCWP and E/e′ with total EAT and patient characteristics in the overall cohort. Data are presented as the standardized association coefficient β with referring P values.
P values below 0.05 were considered statistically significant.
Table 5.
Multivariable regression analysis of the association of regional cardiac strain with total, ventricular, and atrial epicardial adipose tissue and patient characteristic in the overall cohort
| Ventricles | LV LAS rest | LV LAS stress | RV LAS rest | RV LAS stress | ||||
|---|---|---|---|---|---|---|---|---|
| Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | |
| Total EAT | −0.281 | 0.020* | −0.326 | 0.008* | −0.099 | 0.428 | −0.280 | 0.026* |
| Age | −0.062 | 0.637 | −0.029 | 0.826 | 0.246 | 0.074 | −0.228 | 0.093 |
| BMI | −0.182 | 0.160 | −0.155 | 0.226 | 0.055 | 0.680 | −0.150 | 0.256 |
| Diabetes | 0.263 | 0.034* | 0.164 | 0.177 | 0.018 | 0.888 | −0.112 | 0.371 |
| HLP | −0.019 | 0.874 | 0.133 | 0.265 | 0.147 | 0.237 | −0.117 | 0.341 |
| Sex | −0.193 | 0.111 | 0.211 | 0.078 | −0.285 | 0.024* | 0.081 | 0.506 |
| Ventricular EAT | −0.298 | 0.014* | −0.277 | 0.023* | −0.120 | 0.332 | −0.254 | 0.042* |
| Age | −0.088 | 0.497 | −0.064 | 0.626 | 0.229 | 0.084 | −0.241 | 0.072 |
| BMI | −0.188 | 0.145 | −0.159 | 0.222 | 0.022 | 0.866 | −0.136 | 0.307 |
| Diabetes | 0.267 | 0.030* | 0.160 | 0.195 | 0.024 | 0.851 | −0.111 | 0.379 |
| HLP | −0.026 | 0.828 | 0.119 | 0.325 | 0.127 | 0.308 | −0.125 | 0.313 |
| Sex | −0.214 | 0.077 | 0.192 | 0.115 | −0.299 | 0.019* | 0.063 | 0.616 |
| Atria | LA LAS rest | LA LAS stress | RA LAS rest | RA LAS stress | ||||
|---|---|---|---|---|---|---|---|---|
| Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | Standardized coefficient β | P value | |
| Total EAT | −0.284 | 0.018* | −0.274 | 0.039* | −0.425 | <0.001* | −0.243 | 0.050 |
| Age | −0.253 | 0.052 | −0.319 | 0.015* | −0.146 | 0.263 | −0.275 | 0.043* |
| BMI | −0.327 | 0.011* | −0.291 | 0.023* | −0.172 | 0.180 | −0.122 | 0.350 |
| Diabetes | 0.186 | 0.122 | 0.022 | 0.851 | −0.112 | 0.356 | −0.116 | 0.350 |
| HLP | 0.021 | 0.855 | −0.013 | 0.914 | 0.076 | 0.525 | −0.115 | 0.348 |
| Sex | −0.065 | 0.576 | 0.101 | 0.392 | −0.037 | 0.753 | 0.130 | 0.288 |
| Atrial EAT | −0.272 | 0.027* | −0.277 | 0.023* | −0.342 | 0.010* | −0.201 | 0.118 |
| Age | −0.225 | 0.091 | −0.283 | 0.033* | −0.106 | 0.437 | −0.249 | 0.071 |
| BMI | −0.313 | 0.015* | −0.279 | 0.028* | −0.119 | 0.367 | −0.082 | 0.535 |
| Diabetes | 0.166 | 0.165 | 0.008 | 0.945 | −0.149 | 0.240 | −0.137 | 0.274 |
| HLP | 0.021 | 0.858 | −0.008 | 0.948 | 0.077 | 0.538 | −0.105 | 0.401 |
| Sex | −0.030 | 0.799 | 0.137 | 0.247 | 0.017 | 0.890 | 0.167 | 0.186 |
BMI, body mass index; EAT, epicardial adipose tissue; HLP, hyperlipoproteinaemia; LA, left atrial; LAS, long axial strain; LV, left ventricular; RA, right atrial; RV, right ventricular.
Multivariable regression analysis of the association of atrial and ventricular long‐axis strain with total, atrial, and ventricular EAT and patient characteristics in the overall cohort. Data are presented as the standardized association coefficient β with referring P values.
P values below 0.05 were considered statistically significant.
For regional EAT, increasing ventricular EAT volume was independently associated with impaired LV strain at rest and during exercise stress (P < 0.023). Meanwhile, in the right ventricle, higher ventricular EAT volume was independently associated with decreasing RV LAS during exercise stress only (P = 0.042) (see Table 5 ). In turn, atrial EAT volume was independently associated with decreasing atrial strain at rest and during exercise stress (P < 0.027), besides for RA LAS during exercise stress (0.118) (see Table 5 ).
Discussion
Numerous studies have identified EAT as an important risk factor in cardiovascular disease, 2 , 3 , 5 while it promotes the development of diastolic dysfunction and HFpEF. 8 This study was now able to show that patients with preexisting diastolic dysfunction and increased EAT volume show pronounced signs of diastolic failure and adverse remodelling. Despite similar morphological characteristics, functional impairment is stronger in patients with higher EAT volumes. In particular, atrial function is impaired to a higher degree than ventricular cardiac function, and regional EAT in front of the ventricles and atria is independently associated with the functional impairment of the associated cardiac chambers. Those results strengthen the need for a more detailed classification and characterization of patients with diastolic dysfunction, as distinct phenotypes might require different handling.
Cardiac dyspnoea is mostly caused by pulmonary congestion due to compromised cardiac output by either reduced ejection or reduced filling. EAT was associated with diastolic dysfunction but not with systolic functional impairment in multiple studies. 22 , 23 The underlying pathomechanisms comprise two interacting theories. On the one hand, EAT restricts the heart in its function and particularly diminishes diastolic relaxation capacities such as commonly observed in pericardial effusion. 24 On the other hand, EAT is known to induce myocardial inflammation, either by migration of fat tissue from the epicardium directly into the myocardium or by proinflammatory mediators of fat tissue in direct regional relation to the myocardium. 24 The inflammatory processes are often followed by subsequent remodelling, which is mostly hypertrophic. 25
Both described mechanisms, the restriction and the inflammation, impose typical conditions for diastolic dysfunction including higher filling pressures of the left ventricle hand in hand with a higher workload for ventricular filling and subsequent backward failure, which typically cause pulmonary congestion and dyspnoea. 26 , 27
All patients in this cohort suffered from clinically apparent dyspnoea at exertion and incident echocardiographic signs of diastolic dysfunction. However, patients with higher volumes of EAT showed more pronounced signs of diastolic failure and higher rates of manifest HFpEF.
Despite significantly differing amounts of EAT volumes from around 20 to 120 mL/m2 in other HFpEF study populations, 1 , 8 HFpEF patients with high EAT volume always had impaired haemodynamics, out of proportional exercise intolerance and worse outcomes compared with patients with low EAT volume. 1 , 7
Importantly, at this stage, we could not observe differences in the ventricular volumetry and mass between the groups with high and low EAT volumes, whereas functional differences were already present.
Even though the conventional echocardiographic diastolic functional parameter E/e′ did not reveal an association with increased EAT volumes, exercise stress PCWP as the reference standard of diastolic functional assessment showed an independent association with increasing EAT volumes, pointing out to a potential link of both pathologies. Furthermore, impaired LV and LA LAS as non‐invasive CMR parameters, which were previously linked to the diagnosis of diastolic dysfunction and HFpEF, 15 provided further evidence for a potential interconnection of EAT volumes and diastolic dysfunction.
During the in‐depth analysis of the underlying functional impairment, this study now precisely discriminated that patients with higher total EAT volume had impaired biventricular diastolic function during exercise stress only, while atrial strain was reduced at rest already. This could be explained by lower muscular strength of the atria with less capacities for the compensation of additional external pressure load induced by EAT. At the same time, atrial distensibility is diminished by the additive volume and pressure afterload of the left ventricle 7 and inflammatory reactions of the vulnerable atrial myocardium causing incremental damage and atrial functional impairment in advance to ventricular decompensation. 28 , 29
This adds to the important role of the atrium in diastolic dysfunction.
Non‐invasively quantified atrial function in diastolic dysfunction and HFpEF was confirmed as an important parameter in general for the diagnosis and prognostication in HFpEF as it is a key element in the disease's pathophysiology. 15 , 30 , 31 On the one hand, atrial EAT could aggravate commonly known haemodynamic challenges by further restriction and might limit the compensatory function of the atrium in ventricular diastole. 32
On the other hand, those impairments by atrial EAT align with the assumption of the development of an atrial cardiomyopathy, which nurtures LA malfunction as a central role in diastolic dysfunction. 29 , 33
As we now observe LV myocardial remodelling in the cohort with increased EAT volume, an early atrial decompensation in the disease progression associated with increased atrial EAT volume could help to identify patients before adverse remodelling occurs. However, the timely sequence of the development of EAT and fibrosis has yet to be investigated.
The distinct influences of increased EAT volumes and its regional parts yield important implications. The haemodynamic and functional alterations on ventricular and atrial function without morphological changes underline the need for an enhanced characterization and differentiation within the inhomogeneous and broad patient group suffering from diastolic dysfunction. Discriminating patients with increased EAT volume and particularly increased atrial EAT volume could help to identify patients at risk for further functional deterioration. As sodium dependent glucose co‐transporter 2 (SGLT‐2) inhibitors were already observed to reduce EAT, 34 , 35 an improved phenotyping and the early identification of patients at risk could even guide those medical interventions to interfere in potentially reversible stages of heart failure.
Even though this has to be prospectively evaluated in future research, there could be a chance to modify the progression of early diastolic failure in patients with distinctive EAT patterns by simple lifestyle changes and early medical treatment.
Limitations
Patients included in this study were highly selected and might not represent the general population of patients with diastolic dysfunction. Furthermore, the trial was a monocentric study conducted in an experienced core lab, which limits its reproducibility in other centres. 14 The quantification of EAT volumes can be biased by partial volume effects, even though a highly standardized post‐processing evaluation should reduce the probability for individual errors. The impact of EAT might be subject to further comorbidities or lifestyle factors, which have not been investigated in this study. Therefore, a direct association of EAT with diastolic dysfunction cannot be finally proven.
The small number of included patients potentially reduces the statistical accuracy and implies that at least some of the results could only be reported as a trend without statistical significance. Furthermore, this study mainly focuses on the associations between EAT and diastolic dysfunction. The pathophysiological interaction should be investigated in future prospectively randomized studies.
Conclusions
Increased EAT volumes in patients with dyspnoea and signs of diastolic dysfunction is associated with more severe cardiac diastolic failure and adverse remodelling and promotes the development of manifest HFpEF. Despite similar ventricular morphology, important differences in cardiac function were observed between patients with high and low EAT volumes. In patients with higher EAT volume, ventricular diastolic function could be compensated longer than atrial function, which was impaired at rest already. In this context, regional EAT seems to directly contribute to the deterioration of the associated chambers and points out to distinct phenotypes of patients with diastolic failure. Further investigations of those EAT phenotypes and their chronological appearance within heart failure development could enhance the understanding of the disease progression and help to identify potentially reversible stages for intervention before significant adverse remodelling occurs.
Conflict of interest
None declared.
Funding
The study was funded by the German Centre for Cardiovascular Research (Deutsches Zentrum für Herz‐Kreislaufforschung; DZHK‐17).
Acknowledgements
Open Access funding enabled and organized by Projekt DEAL.
Schulz, A. , Backhaus, S. J. , Lange, T. , Evertz, R. , Kutty, S. , Kowallick, J. T. , Hasenfuß, G. , and Schuster, A. (2024) Impact of epicardial adipose tissue on cardiac function and morphology in patients with diastolic dysfunction. ESC Heart Failure, 11: 2013–2022. 10.1002/ehf2.14744.
References
- 1. van Woerden G, van Veldhuisen DJ, Manintveld OC, van Empel VPM, Willems TP, de Boer RA, et al. Epicardial adipose tissue and outcome in heart failure with mid‐range and preserved ejection fraction. Circ Heart Fail 2022;15:e009238. doi: 10.1161/CIRCHEARTFAILURE.121.009238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Mahabadi AA, Berg MH, Lehmann N, Kälsch H, Bauer M, Kara K, et al. Association of epicardial fat with cardiovascular risk factors and incident myocardial infarction in the general population: The Heinz Nixdorf Recall Study. J Am Coll Cardiol 2013;61:1388‐1395. doi: 10.1016/j.jacc.2012.11.062 [DOI] [PubMed] [Google Scholar]
- 3. Eberhard M, Stocker D, Meyer M, Kebernik J, Stähli BE, Frauenfelder T, et al. Epicardial adipose tissue volume is associated with adverse outcomes after transcatheter aortic valve replacement. Int J Cardiol 2019;286:29‐35. doi: 10.1016/j.ijcard.2019.01.068 [DOI] [PubMed] [Google Scholar]
- 4. Backhaus SJ, Schulz A, Lange T, Schmidt‐Schweda LS, Evertz R, Kowallick J, et al. Real‐time cardiovascular magnetic resonance imaging for non‐invasive characterisation of heart failure with preserved ejection fraction: Final outcomes of the HFpEF stress trial. Clin Res Cardiol 2024;113:496‐508. doi: 10.1007/s00392-023-02363-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. van Woerden G, van Veldhuisen DJ, Gorter TM, van Empel VPM, Hemels MEW, Hazebroek EJ, et al. Importance of epicardial adipose tissue localization using cardiac magnetic resonance imaging in patients with heart failure with mid‐range and preserved ejection fraction. Clin Cardiol 2021;44:987‐993. doi: 10.1002/clc.23644 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gorter TM, van Woerden G, Rienstra M, Dickinson MG, Hummel YM, Voors AA, et al. Epicardial adipose tissue and invasive hemodynamics in heart failure with preserved ejection fraction. JACC: Heart Failure 2020;8:667‐676. doi: 10.1016/j.jchf.2020.06.003 [DOI] [PubMed] [Google Scholar]
- 7. Koepp KE, Obokata M, Reddy YNV, Olson TP, Borlaug BA. Hemodynamic and functional impact of epicardial adipose tissue in heart failure with preserved ejection fraction. JACC Heart Fail 2020;8:657‐666. doi: 10.1016/j.jchf.2020.04.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. 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. doi: 10.1161/JAHA.118.009975 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Zhou M, Wang H, Chen J, Zhao L. Epicardial adipose tissue and atrial fibrillation: Possible mechanisms, potential therapies, and future directions. Pacing Clin Electrophysiol 2020;43:133‐145. doi: 10.1111/pace.13825 [DOI] [PubMed] [Google Scholar]
- 10. Iacobellis G. Epicardial adipose tissue in contemporary cardiology. Nat Rev Cardiol 2022;19:593‐606. doi: 10.1038/s41569-022-00679-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Konwerski M, Gąsecka A, Opolski G, Grabowski M, Mazurek T. Role of epicardial adipose tissue in cardiovascular diseases: A review. Biology (Basel) 2022;11:355. doi: 10.3390/biology11030355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Li C, Qin D, Hu J, Yang Y, Hu D, Yu B. Inflamed adipose tissue: A culprit underlying obesity and heart failure with preserved ejection fraction. Front Immunol 2022;13: doi: 10.3389/fimmu.2022.1072573 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Yang M‐C, Liu H‐K, Tsai C‐C, Su Y‐T, Wu J‐R. Epicardial adipose tissue was highly associated with reduction in left ventricular diastolic function as early as in adolescence. Acta Cardiol Sin 2022;38:601‐611. doi: 10.6515/ACS.202209_38(5).20220331B [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. de Wit‐Verheggen VHW, Altintas S, Spee RJM, Mihl C, van Kuijk SMJ, Wildberger JE, et al. Pericardial fat and its influence on cardiac diastolic function. Cardiovasc Diabetol 2020;19:129. doi: 10.1186/s12933-020-01097-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Backhaus SJ, Lange T, George EF, Hellenkamp K, Gertz RJ, Billing M, et al. Exercise stress real‐time cardiac magnetic resonance imaging for noninvasive characterization of heart failure with preserved ejection fraction: The HFpEF‐Stress trial. Circulation 2021;143:1484‐1498. doi: 10.1161/CIRCULATIONAHA.120.051542 [DOI] [PubMed] [Google Scholar]
- 16. Pieske B, Tschöpe C, de Boer RA, Fraser AG, Anker SD, Donal E, et al. How to diagnose heart failure with preserved ejection fraction: The HFA‐PEFF diagnostic algorithm: A consensus recommendation from the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Eur Heart J 2019;40:3297‐3317. doi: 10.1093/eurheartj/ehz641 [DOI] [PubMed] [Google Scholar]
- 17. Erdei T, Smiseth OA, Marino P, Fraser AG. A systematic review of diastolic stress tests in heart failure with preserved ejection fraction, with proposals from the EU‐FP7 MEDIA study group. Eur J Heart Fail 2014;16:1345‐1361. doi: 10.1002/ejhf.184 [DOI] [PubMed] [Google Scholar]
- 18. Gertz RJ, Lange T, Kowallick JT, Backhaus SJ, Steinmetz M, Staab W, et al. Inter‐vendor reproducibility of left and right ventricular cardiovascular magnetic resonance myocardial feature‐tracking. PLoS ONE 2018;13:e0193746. doi: 10.1371/journal.pone.0199489 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Uecker M, Zhang S, Voit D, Karaus A, Merboldt KD, Frahm J. Real‐time MRI at a resolution of 20 ms. NMR Biomed 2010;23:986‐994. doi: 10.1002/nbm.1585 [DOI] [PubMed] [Google Scholar]
- 20. Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP. Modified Look‐Locker inversion recovery (MOLLI) for high‐resolution T1 mapping of the heart. Magn Reson Med 2004;52:141‐146. doi: 10.1002/mrm.20110 [DOI] [PubMed] [Google Scholar]
- 21. Flüchter S, Haghi D, Dinter D, Heberlein W, Kühl HP, Neff W, et al. Volumetric assessment of epicardial adipose tissue with cardiovascular magnetic resonance imaging. Obesity (Silver Spring) 2007;15:870‐878. [DOI] [PubMed] [Google Scholar]
- 22. Wu C‐K, Lee JK, Hsu JC, Su MYM, Wu YF, Lin TT, et al. Myocardial adipose deposition and the development of heart failure with preserved ejection fraction. Eur J Heart Fail 2020;22:445‐454. doi: 10.1002/ejhf.1617 [DOI] [PubMed] [Google Scholar]
- 23. Jin X, Hung CL, Tay WT, Soon D, Sim D, Sung KT, et al. Epicardial adipose tissue related to left atrial and ventricular function in heart failure with preserved versus reduced and mildly reduced ejection fraction. Eur J Heart Fail 2022;24:1346‐1356. doi: 10.1002/ejhf.2513 [DOI] [PubMed] [Google Scholar]
- 24. van Woerden G, van Veldhuisen DJ, Westenbrink BD, de Boer RA, Rienstra M, Gorter TM. Connecting epicardial adipose tissue and heart failure with preserved ejection fraction: Mechanisms, management and modern perspectives. Eur J Heart Fail 2022;24:2238‐2250. doi: 10.1002/ejhf.2741 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Anthony SR, Guarnieri AR, Gozdiff A, Helsley RN, Phillip Owens A III, Tranter M. Mechanisms linking adipose tissue inflammation to cardiac hypertrophy and fibrosis. Clin Sci (Lond) 2019;133:2329‐2344. doi: 10.1042/CS20190578 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Schulz A, Schuster A. Visualizing diastolic failure: Non‐invasive imaging‐biomarkers in patients with heart failure with preserved ejection fraction. EBioMedicine 2022;86:104369. doi: 10.1016/j.ebiom.2022.104369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Schulz A, Beuthner BE, Böttiger ZM, Gersch SS, Lange T, Gronwald J, et al. Epicardial adipose tissue as an independent predictor of long‐term outcome in patients with severe aortic stenosis undergoing transcatheter aortic valve replacement. Clin Res Cardiol 2024. doi: 10.1007/s00392-024-02387-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Nalliah CJ, Bell JR, Raaijmakers AJA, Waddell HM, Wells SP, Bernasochi GB, et al. Epicardial adipose tissue accumulation confers atrial conduction abnormality. J Am Coll Cardiol 2020;76:1197‐1211. doi: 10.1186/s12872-022-02703-9 [DOI] [PubMed] [Google Scholar]
- 29. Wong CX, Ganesan AN, Selvanayagam JB. Epicardial fat and atrial fibrillation: Current evidence, potential mechanisms, clinical implications, and future directions. Eur Heart J 2017;38:1294‐1302. doi: 10.1093/eurheartj/ehw045 [DOI] [PubMed] [Google Scholar]
- 30. Backhaus SJ, Rösel SF, Schulz A, Lange T, Hellenkamp K, Gertz RJ, et al. RT‐CMR imaging for noninvasive characterization of HFpEF: Medium‐term outcomes of the HFpEF stress trial. Cardiovasc Imaging 2022;15:943‐945. doi: 10.1016/j.jcmg.2021.11.013 [DOI] [PubMed] [Google Scholar]
- 31. Schulz A, Mittelmeier H, Wagenhofer L, Backhaus SJ, Lange T, Evertz R, et al. Assessment of the cardiac output at rest and during exercise stress using real‐time cardiovascular magnetic resonance imaging in HFpEF‐patients. Int J Cardiovasc Imaging 2024. doi: 10.1007/s10554-024-03054-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Hamlin SK, Villars PS, Kanusky JT, Shaw AD. Role of diastole in left ventricular function, II: Diagnosis and treatment. Am J Crit Care 2004;13:453‐466. doi: 10.1111/vru.13326 [DOI] [PubMed] [Google Scholar]
- 33. Shah SJ, Borlaug BA, Chung ES, Cutlip DE, Debonnaire P, Fail PS, et al. Atrial shunt device for heart failure with preserved and mildly reduced ejection fraction (REDUCE LAP‐HF II): A randomised, multicentre, blinded, sham‐controlled trial. Lancet 2022;399:1130‐1140. doi: 10.1016/S0140-6736(22)00016-2 [DOI] [PubMed] [Google Scholar]
- 34. Mullens W, Martens P. Empagliflozin‐induced changes in epicardial fat: The centerpiece for myocardial protection? JACC: Heart Failure 2021;9:590‐593. doi: 10.1016/j.jchf.2021.05.006 [DOI] [PubMed] [Google Scholar]
- 35. Masson W, Lavalle‐Cobo A, Nogueira JP. Effect of SGLT2‐inhibitors on epicardial adipose tissue: A meta‐analysis. Cell 2021;10:2150. doi: 10.3390/cells10082150 [DOI] [PMC free article] [PubMed] [Google Scholar]