The value of epicardial adipose tissue (EAT) thickness as determined by echocardiography in cardiovascular risk assessment is not well understood. The aim of this study was to determine the associations between EAT thickness and Framingham risk score, carotid intima media thickness, carotid artery plaque, and computed tomographic coronary calcium score in a primary prevention population.
Patients presenting for cardiovascular preventive care ( n = 356) who underwent echocardiography as well as carotid artery ultrasound and/or coronary calcium scoring were included.
EAT thickness was weakly correlated with Framingham risk score. The prevalence of carotid plaque was significantly greater in those with EAT thickness ≥5.0 mm who either had low Framingham risk scores or had body mass indexes ≥25 kg/m 2 , compared with those with EAT thickness <5.0 mm. No significant association between EAT thickness and carotid intima-media thickness or coronary calcium score existed.
EAT thickness ≥5.0 mm may identify an individual with a higher likelihood of having detectable carotid atherosclerosis.
Higher body mass index (BMI) is predictive of cardiovascular events. Furthermore, the distribution of adipose tissue, especially visceral adipose tissue, is associated with coronary artery disease and cardiovascular events. Epicardial adipose tissue (EAT) is visceral fat adjacent to the heart and lies between the myocardium and the visceral pericardial tissue.
Visceral fat is most accurately quantified by magnetic resonance imaging, or computed tomographic (CT) imaging. However, these methods can be expensive and time consuming and, in the case of CT imaging, can expose patients to radiation. Additionally, patients may have other contraindications for these tests, such as medical devices (i.e., pacemakers with magnetic resonance imaging). Waist circumference is a practical and inexpensive means to obtain an estimate of visceral fat, but this measure is confounded by subcutaneous fat and reproducibility.
Transthoracic echocardiography is commonly performed in individuals with cardiovascular risk factors and can accurately assess EAT. EAT is easily visualized in standard views on the right ventricular free wall during systole and diastole. EAT measured in diastole by echocardiography is associated with increased left ventricular mass, metabolic syndrome, endothelial dysfunction, and the presence and severity of coronary artery disease. EAT measured during systole is associated with insulin resistance, and EAT thicknesses of 9.5 mm in men and 7.5 mm in women predict the presence of metabolic syndrome. EAT measured by CT imaging does correlate with cardiovascular events independent of traditional risk factors.
The relationship of echocardiographic measures of EAT to current cardiovascular risk assessment tools in a primary prevention population is not well understood. The purpose of this study was to define the relationship of EAT measured by echocardiography to Framingham risk score (FRS), carotid intima-media thickness (CIMT), carotid artery plaque, and coronary artery calcium score (CACS) in a population being evaluated for primary prevention of cardiovascular disease.
The study was a retrospective review of 356 asymptomatic patients aged 25 to 80 years (mean age, 55 ± 9 years; 71% men). Patients were referred to the Mayo Clinic (Scottsdale, AZ) between 2004 and 2008 by primary care physicians or subspecialists for cardiovascular risk evaluation. Individuals were evaluated by internists and cardiologists with subsequent testing for subclinical atherosclerosis and echocardiography done at the discretion of the evaluating physician. The most common indications for the assessment of CACS or CIMT were intermediate FRS, family history of premature coronary artery disease, or high levels of a single risk factor for coronary artery disease (i.e., serum lipids). Subjects were included in the analysis if they had undergone echocardiography and had undergone CACS measurement and/or carotid artery ultrasound for the evaluation of CIMT or plaque. The FRS for use in primary care was used in the analysis. The FRS uses age, gender, total and high-density lipoprotein cholesterol, systolic blood pressure, high blood pressure treatment, smoking, and diabetes status. Subjects were stratified by their calculated 10-year incidence of cardiovascular events as low (<10%), intermediate (10%–20%), or high (>20%). Exclusion criteria were documented cardiovascular disease, prior cerebrovascular accident, use of exogenous steroids, and end-stage renal disease. Subjects with high FRS (>20%) were excluded from the analysis, because this population requires aggressive management of cardiovascular risk factors without further risk stratification and did not represent a significant portion of our overall population. Demographic, morphometric, and clinical variables, including age, gender, height, weight, BMI, body surface area, systolic and diastolic blood pressure, smoking history, history of hypertension, and the presence or absence of diabetes were recorded. Biochemical data including total cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides, homocysteine, high-sensitivity C-reactive protein, lipoprotein(a), fasting blood glucose, glycosylated hemoglobin, and FRS were recorded into an electronic database for each subject. We performed a review of each patient’s clinical chart to extract information and to calculate the FRS.
Each subject underwent detailed transthoracic two-dimensional, M-mode, Doppler, and tissue Doppler echocardiography using standardized techniques per American Society of Echocardiography guidelines. Echocardiograms were uploaded to an online system (ProSolv, Indianapolis, IN). Echocardiograms were interpreted by experienced echocardiologists. Readers were blinded to the subjects’ clinical and demographic data. EAT was interpreted by two echocardiologists. Every third consecutive study was read by both physicians (intraclass correlation coefficient, 0.90). EAT was measured on the free wall of the right ventricle from both parasternal long-axis and short-axis views at end-diastole in three cardiac cycles ( Figure 1 ). The maximum value at any site was measured, and the average value was considered.
Electron-beam CT imaging (GE, Fairfield, CT) was used to evaluate coronary artery calcification. CT scout images were obtained to determine the location of the heart, and then a prospective electrocardiographically gated scan with a 3-mm slice thickness from the level of the carina through the bottom of the heart was performed. When coronary calcium was detected, an automated program based on the Agatston method was used to compute the CT calcium score. The CT calcium score percentiles were based on the scores of age-matched and sex-matched control patients.
The carotid arteries were imaged with a Siemens Sequoia ultrasonography system (Siemens Medical Solutions USA, Inc., Mountain View, CA) with an 8-MHz to 15-MHz linear-array transducer. A depth of 4 cm was used. The examination included a thorough scan of the extracranial carotid arteries to detect carotid plaque, defined either as a thickening of the intima-media of ≥1.5 mm and >50% of the surrounding intima-media ( Figure 2 ). In the absence of identified plaque, the CIMT of the distal 1 cm of the far wall of the common carotid artery was measured using a semiautomated border detection program. The mean CIMT was calculated by averaging three measurements of CIMT at each of three scan planes (anterior, lateral, and posterior) from both the right and left common carotid arteries (18 total measurements). Using data from the Atherosclerosis Risk in Communities study, the Bogalusa Heart Study, and the Cardiovascular Health Study as referenced by Stein et al. , age-matched, gender-matched, and race-matched quartile from the composite CIMT of the left and right common carotid arteries were calculated. If the quartile was discrepant between the right and left common carotid arteries, the highest quartile was assigned. Our laboratory’s absolute mean interobserver difference for CIMT measurement was 0.0186 ± 0.0244 mm, and interobserver variability was 2.9 ± 3.8%.
Data are summarized as mean ± SD for continuous variables and as frequency (percentage) for categorical variables. Correlations between EAT and variables of interest were assessed using a linear and quadratic regression method and Spearman’s ρ coefficient. Two-sample t tests and one-way analysis of variance were used to compare continuous variables between two groups and more than two groups, respectively. Fisher’s exact tests and Pearson’s χ 2 tests were used to compare categorical data. P values <.05 were considered statistically significant. All analyses were performed using SAS version 9.1.3 (SAS Institute Inc., Cary, NC).
The study population ( n = 356) is described in Table 1 . The mean EAT thickness in the cohort was 4.7 ± 1.5 mm. The mean EAT thickness in men was 4.8 ± 1.6mm and that in women 4.7 ± 1.4 mm. The mean CIMT was 0.67 ± 0.14 mm (0.68 ± 0.15 mm in men, 0.63 ± 0.12 mm in women). Carotid plaque was present in 124 subjects. Fifty-four percent (192 of 356) had carotid plaque or CIMT greater than the 75th percentile for age, race, and gender. CT calcium scores were obtained in 140 of the 356 subjects. In the overall cohort, the mean CT calcium score was 205 ± 574 and trended higher in men (254 ± 645) than women (57 ± 200) ( P = .053).
|Variable||All||Men||Women||P (men vs women)|
|Subjects||356||253 (71%)||103 (29%)|
|Age (y)||55 ± 9||54 ± 10||57 ± 11||.017|
|BMI (kg/m 2 )||27.9 ± 4.7||28.7 ± 4.4||26 ± 5.1||<.0001|
|Tobacco use||.027 ∗|
|Never||236 (66%)||171 (67%)||65 (63%)|
|Former||94 (19%)||59 (23%)||35 (34%)|
|Current||26 (7%)||23 (9%)||3 (3%)|
|Hypertension||129 (36%)||91 (36%)||38 (37%)||.90|
|Diabetes||23 (6%)||18 (7%)||5 (5%)||.49|
|Fasting glucose (mg/dL)||104 ± 31||106 ± 28||100 ± 37||.12|
|Total cholesterol (mg/dL)||190 ± 41||186 ± 41||198 ± 43||.018|
|HDL cholesterol (mg/dL)||58 ± 17||53 ± 15||68 ± 18||<.0001|
|LDL cholesterol (mg/dL)||110 ± 49||111 ± 54||109 ± 38||.68|
|EAT thickness (mm)||4.7 ± 1.5||4.8 ± 1.6||4.7 ± 1.4||.73|
|CIMT (mm)||0.67 ± 0.14||0.68 ± 0.15||0.63 ± 0.12||.01|
|Plaque present||124 (35%)||92 (37%)||32 (31%)||.39|
|CACS||205 ± 574||254 ± 645||57 ± 200||.053|
A correlation existed between FRS and EAT ( r = 0.21, P < .0001). This relationship remained significant after adjustment for BMI, low-density lipoprotein, triglycerides, homocysteine, high-sensitivity C-reactive protein, lipoprotein(a), fasting blood glucose, and glycosylated hemoglobin. A difference in EAT thickness was observed between low-FRS and intermediate-FRS groups (4.5 ± 1.4 vs 5.5 ± 1.6 mm, respectively, P = .0011).
EAT did not correlate with CIMT ( r = 0.07, P = .243) or CT calcium scores ( r = 0.01, P = .873). No significant difference existed between EAT and CT calcium score quartiles ( P = .97; Figure 3 ). Subdividing CT calcium scores by age and gender percentile groups did not demonstrate EAT differences. EAT thickness did differ significantly between those with CIMT less than the 75th percentile thickness by age and gender ( n = 192) and subjects at the 75th percentile or higher ( n = 161) (4.5 ± 1.3 vs 4.9 ± 1.6 mm, P = .0091; Figure 4 ). A statistically significant difference in EAT thickness was observed when subjects were stratified by the presence or absence of carotid plaque (5.2 ± 1.6 vs 4.5 ± 1.4 mm, respectively, P = .0001; Figure 5 ). In the overweight and obese group ( n = 255; BMI ≥25 mg/kg 2 ), EAT thickness differed significantly when participants were stratified by the presence or absence of plaque (5.4 ± 1.7 mm [ n = 97] vs 4.6 ± 1.5 mm [ n = 158], respectively, P = .0003). EAT thickness also differed significantly when the overweight and obese patients were divided by CIMT above and below the 75th percentile for age and gender (5.1 ± 1.7 mm [ n = 149] vs 4.6 ± 1.4 mm [ n = 106], respectively, P = .005).
Because groups with carotid plaque had a mean EAT thickness ≥5.0 mm, we sought to identify the prevalence of carotid plaque between groups divided by this value ( Figure 6 ). In the entire cohort, plaque was present in 46% (64 of 137) of those with EAT thicknesses ≥5.0 mm (hazard ratio, 2.26; 95% confidence interval [CI], 1.44–3.53; P = .0004) and in 28% (61 of 218) of those with EAT thicknesses <5.0 mm ( P = .0003). A trend toward significance was noted when a threshold of 75th percentile CIMT was used in place of carotid plaque ( P = .08). FRS groups were subdivided by EAT thickness of 5.0 mm. In the low-FRS group, carotid plaque was present in 39% of subjects (28 of 72) with EAT thicknesses ≥5.0 mm (hazard ratio, 2.11; 95% CI, 1.15–3.89; P = .016) and 23% of subjects (34 of 145) with EAT thicknesses <5.0 mm ( P = .025). No significant difference in prevalence of carotid plaque was seen in the intermediate-risk group when divided by an EAT thickness threshold of 5.0 mm. The population was divided at a BMI of 25 kg/m 2 , and both groups were further subdivided by an EAT thickness threshold of 5.0 mm. Carotid plaque incidence was 49% (53 of 108) in the overweight and obese group with EAT thicknesses ≥5.0mm (hazard ratio, 2.28; 95% CI, 1.36–3.82; P = .0018) compared with 30% (44 of 148) in the group with EAT thicknesses <5.0 mm ( P < .0026). In the group with BMIs <25 kg/m 2 , the incidence of carotid plaque was 38% (11 of 29) in the group with EAT thicknesses ≥5.0 mm compared with 25% (17 of 71) in the group with EAT thicknesses <5.0 mm ( P = .22). Further evaluation of different thresholds of EAT thickness (4.0, 6.0, 7.0, and 8.0 mm) failed to perform differently than the 5.0-mm cutoff.