Pericardial Adipose Tissue Determined by Dual Source CT Is a Risk Factor for Coronary Atherosclerosis
Objectives— Pericardial fat as a visceral fat depot may be involved in the pathogenesis of coronary atherosclerosis. To gain evidence for that concept we sought to investigate the relation of pericardial fat volumes to risk factors, serum adiponectin levels, inflammatory biomarkers, and the quantity and morphology of coronary atherosclerosis.
Methods and Results— Using Dual source CT angiography pericardial fat volume and coronary atherosclerosis were assessed simultaneously. Plaques were classified as calcified, mixed, and noncalcified, and the number of affected segments served as quantitative score. Patients with atherosclerotic lesions had significant larger PAT volumes (226 cm3±92 cm3) than patients without atherosclerosis (134 cm3±56 cm3; P>0.001). No association was found between BMI and coronary atherosclerosis. PAT volumes >300 cm3 were the strongest independent risk factor for coronary atherosclerosis (odds ratio 4.1; CI 3.63 to 4.33) also significantly stronger compared to the Framingham score. We furthermore demonstrated that elevated PAT volumes are significantly associated with low adiponectin levels, low HDL levels, elevated TNF-α levels, and hsCRP.
Conclusion— In the present study we demonstrated that elevated PAT volumes are associated with coronary atherosclerosis, hypoadiponectinemia, and inflammation and represent the strongest risk factor for the presence of atherosclerosis and may be important for risk stratification and monitoring.
There is growing evidence that regional visceral fat distribution may contribute to an unfavorable metabolic and cardiovascular risk profile.1,2 In patients with obesity, insulin resistance, diabetes, and hyperlipidemia visceral fat hypertrophies and transforms into a multifunctional organ that produces and secretes multiple endocrine and paracrine factors promoting inflammation, neovascularization, and oxidative stress, features that also characterize atherosclerosis.3 Pericardial fat as a local visceral fat depot with close proximity to coronary arteries may serve as a source of inflammatory cytokines and cells that may locally enhance systemic proatherogenic effects via outside to inside signaling.4,5 Thus it may be a specific parameter indicating an unfavorable cardio-metabolic state and may be used for risk stratification. To date, however, only little attention has focused on this regional fat depot located around the heart and its relation to cardiovascular risk factors, and the quantity and composition of coronary atherosclerosis is not well studied yet.
Multi-slice CT is a noninvasive tool that allows to reliably assess both obstructive and nonobstructive subclinical coronary artery disease in an earlier stage than invasive angiography.6–9 Based on density measurements, plaques can be further characterized in noncalcified, mixed, and calcified plaques.7 By using the same scan data this tool furthermore allows to quantify the exact pericardial fat volume.9 We thus sought to assess the relation of pericardial fat volume to cardiovascular risk factors, levels of inflammatory cytokines, adiponectin, and to the extent and the phenotype of coronary atherosclerosis.
For detailed methods, please see the supplemental materials (available online at http://atvb.ahajournals.org).
From March 2006 until August 2007 we included 286 consecutive patients who underwent dual-source multi-slice CT coronary angiography with an intermediate pretest likelihood for coronary artery disease. Further patient characteristics are given in Table 1.
CT coronary angiography was performed using a Siemens Definition scanner (Siemens Medical Solutions), which uses 2 X-ray sources for image generation.
Atherosclerotic plaques were classified as calcified, mixed, or noncalcified as described previously by our group. Based on the number of diseased segments a plaque score for each patient was calculated.10
Following the publications of Stary et al, Mautner et al, and most recently Bamberg et al, we defined existence of exclusively noncalcified plaques in a patient as early stage atherosclerosis.11–14
Peri- and Epicardial Fat Assessment Protocol
The pericardial fat volume was measured in cm3 using the Volume Analysis software tool of our cardiac workstation (Siemens, Leonardo). We defined pericardial fat as epicardial fat plus paracardial fat.
In a subset of 120 patients both epi- and pericardial fat volume was determined. Our analysis revealed an excellent correlation (Pearson correlation coefficient r=0.97 between those two fat depots and an equal correlation of both fat depots with the number of atherosclerotic plaque r=0.40 versus 0.41). Because epicardial fat volume was less reproducible (interobserver variability 15%) compared to PAT volume (8%), only the latter was used for further analysis. In our study, the acquisition of an additional data set was not necessary to determine PAT volume; the data set of coronary CTA could be used.
DSCT angiography could be performed in all 286 patients. Adequate image quality for evaluation of coronary plaques could be obtained in 264 patients. The remaining 22 patients with insufficient image quality because of motion artifacts (n=17) or insufficient opacification were excluded from the study. In all 264 study patients, PAT volume could be determined.
Morphology of Coronary Plaques
In 56 of 264 (21%) patients coronary plaques could be excluded, 42 (16%) patients showed only noncalcified plaques, 110 (42%) patients revealed a mixture of all plaque types, and 56 (21%) patients exclusively calcified plaques.
Relation of PAT Volume to Age, Gender, Risk Factors, and Coronary Plaques
We found a significant increase of PAT volume with age in both men and women from 148±77 cm3 in patients under 40 years to 252±118 cm3 for patients above 70 years (P<0.05). In all age groups women showed a significantly lower PAT volume compared to men. Patients with cardiovascular risk factors showed a significantly higher PAT volume compared to patients without cardiovascular risk factors (222±110 cm3 versus 187±71 cm3, P<0.05). We also found an increase of PAT volume with an increasing number of risk factors: from 187±71 cm3 in patients without any risk factors up to 226±119 cm3 in patients with 3 or more risk factors (P<0.05). PAT volume correlated with BMI (r=0.45, P<0.0001) and with the number of diseased coronary segments (r=0,44, P<0.0001). The correlation of PAT volume and number of plaques remained statistically significant even after adjusting for BMI (Table 2). There was no correlation between the presence or the number of atherosclerotic segments and BMI (r=0.12, P=0.13).
Patients with any coronary plaque showed a significant higher PAT volume compared to patients without coronary plaques (226±97 cm3 compared to 134±56 cm3, P<0.01). This difference was persistent independent of patients age (Figure 1). A significant difference in PAT volume between patients with noncalcified, calcified, or a mixture of all plaque types could not be observed, although patients with exclusively noncalcified plaques had a significantly lower plaque burden (Table 3). All plaque types revealed a similar and significant correlation to PAT-volume (Table 3).
The relative risk for the presence of coronary plaques in dependence of PAT-Volume and risk factors is given in Table 4a and 4⇓b. The area under the ROC curve to discriminate patients with atherosclerotic plaques from those without plaque was significantly higher for PAT volume compared to the Framingham score (0.82 versus 0.64, P<0.02, Figure 2).
Relation of PAT Volume to Serum Lipids, Inflammation, and Adiponectin
Using the spearman correlation coefficient a significant correlation between PAT volume, Adiponectin (inverse correlation), HDL (inverse correlation), TNF-α, and hsCRP could be observed (Table 4c). No correlation was found between PAT volume and IL 6 and LDL.
The results of the present study demonstrate that PAT volume is strongly associated with the presence and extent of coronary atherosclerosis. This relation is independent of body mass index. Compared to all traditional risk factors including the Framingham Score, PAT volumes >300 cm3 represented the strongest predictor for the presence of coronary atherosclerosis. We further found a significant association of PAT volumes to hypoadiponectinemia and systemic inflammation. Patients revealing only noncalcified plaques already reveal significantly elevated PAT volumes, indicating that PAT volume accumulation may precede plaque calcification and the development of mature atherosclerotic plaques in general. Thus PAT volume quantification may be used in addition to calcium scoring to identify patients with coronary artery disease even in the absence of coronary calcium.
Measurements of PAT Volume
It is widely accepted that the amount of visceral fat is an important cardiovascular risk factor. Because quantification of visceral abdominal fat is difficult, waist circumference is clinically used as a surrogate marker. Waist circumference, however, could be confounded by large amounts of subcutaneous fat, and a number of studies have demonstrated that waist circumference reflects both subcutaneous and visceral fat amount with only moderate correlations to intraabdominal fat.15 Thus efforts are made to develop better measures of visceral adiposity.16 In the context of coronary artery disease PAT is attributable to its location of particular interest. Thus peri- and epicardial fat was quantified in recent studies by different imaging tools. Iacobellis et al have shown using echocardiography that the thickness of epicardial fat anteriorly to the right ventricle is highly correlated to the amount of visceral adipose tissue determined by MRI.16 They further demonstrated that epicardial fat thickness correlated well with PAT volume determined by MRI. In two additional studies using CT, PAT volume was highly correlated to visceral abdominal fat.9,17
By echocardiography the thickness of PAT anterior to the right ventricle is generally used as a surrogate marker for the entire PAT volume. Iacobellis et al have shown that these measurements correlate well with PAT volume measurements by MRI.15 The advantage of echocardiography is the fact that it can be easily performed and is less demanding compared to MRI and CT. The disadvantage is the inability to determine the exact PAT volume, as the distribution of PAT around the heart is variable, single measurements of fat-thickness at one location only may significantly under- or overestimate the real PAT volume. Multi-slice CT allows to visualize coronary arteries in a detailed manner and allows to detect and characterize atherosclerotic plaques even in subclinical stages. Because of the submillimeter resolution it is the most accurate tool to quantify PAT volume. Therefore it offers the unique opportunity to simultaneously assess both the presence and the extent of coronary atherosclerosis and the magnitude of PAT volume. MSCT exams are, however, associated with radiation exposure. In our study a protocol was used for CT coronary angiography with an average dose of 9.4 msv. However, radiation dose for CT angiography can be minimized down to 2.5 msv using the sequential scanning technique, and it is expected that wider detector coverage radiation may be reduced below 1.5 msv. PAT volume, however, can also be determined with less sophisticated scan protocols that are used for calcium scoring. The radiation for this type of scan is ≈0.5 msv, which equals 6 chest X-ray examinations and is 10% of the annual natural background radiation in southern Germany (5 msV per year).
PAT Volume and Its Relation to Cardiovascular Risk Factors
Recent studies have demonstrated that epicardial and PAT volume is related to cardiovascular risk factors indicating the metabolic syndrome. Iacobellis et al demonstrated an independent and significant relation of epicardial fat thickness assessed by echocardiography to fasting insulin levels and diastolic blood pressure.8 In addition to these findings we now demonstrated that PAT volume increases with the number of risk factors. In the multivariate analysis it is significantly correlated to hypertension, diabetes, age, and hypercholesterolemia, and thus it reflects the clustering effect of multiple risk factors in the individual patient and is a good indicator for the metabolic syndrome.
PAT Volume and Its Relation to Coronary Atherosclerosis
So far there are only a few studies investigating the association of PAT volume and coronary atherosclerosis, although there is striking evidence that visceral fat is adversely related to cardiovascular risk. In a recent study by Jeong et al,18 PAT volume was the best predictor for angiographic disease severity compared to waist circumference or visceral abdominal fat volume.18 Ahn et al have demonstrated that patients with significant obstructive coronary artery disease have significantly more epicardial fat determined by echocardiography than patients with non significant coronary artery disease.17 These results are in line with the observations of our present study. Patients with evidence for any plaque on Dual Source CT-angiography revealed on average significantly higher PAT volumes than patients without coronary lesions. Increasing PAT volumes increase the risk for the presence of coronary plaques substantially. The ROC analysis revealed that 95% of patients with PAT volumes >300 cm3 had detectable coronary atherosclerosis. Compared to traditional risk factors PAT volumes >300 cm3 represent the strongest independent risk factor for the presence of coronary plaques (odds ratio 4.1). Interestingly, we found absolute no association between BMI and presence or extent of coronary atherosclerosis, so that this parameter seems to be not useful in predicting coronary artery disease.
Association of PAT Volume and Plaque Composition
In addition to plaque burden we examined for the first time the association of plaque composition and PAT volume. Prior studies using MSCT have demonstrated that the number of noncalcified plaques is associated with unstable coronary artery disease, and it was suggested that these lesions may indicate an elevated disease activity.19–21 Bamberg et al recently demonstrated that noncalcified plaques on MSCT are a feature of early stage atherosclerosis and that their number decreases with age whereas the number of calcified lesions increases.11 Unfortunately CT does not yet allow an accurate and reliable further classification of noncalcified lesions in fibrous and lipid-rich vulnerable lesions, so that we only classified in calcified, mixed, and noncalcified lesions.7 We found no difference of PAT-Volumes between patients with only noncalcified, mixed, or only calcified lesions. All types of plaques individually and independently correlated significantly with PAT-volume with similar correlation coefficients. However, patients with exclusively noncalcified plaques were significantly younger and had a significantly lower plaque burden than patients with calcified lesions. These findings indicate that PAT-Volume accumulation is already present before plaque calcification occurs and it may also precede development of atherosclerosis in general. This speculation is supported by a study of Lear et al who observed that abdominal visceral fat accumulation was more strongly associated with carotid intimal thickening than to plaque number and plaque area. They hypothesized that visceral adipose tissue–derived proatherogenic factors are responsible for early endothelial dysfunction and predisposition for atherosclerosis.22 Thus it may be used additionally to calcium scoring as a parameter indicating an increased cardiovascular risk, and it may indicate an elevated disease activity even in patients with lower plaque burden. However, whether these cross-sectional observations will translate into a higher rate of complications in prospective trials has to be proven.
Visceral fat is able to produce large amounts of proinflammatory chemokines like TNF-α, IL-6, free fatty acids, or plasminogen activator inhibitor-1 (PAI-1). All of them are also involved in atherosclerosis and thrombosis. Adiponectin is a hormone that is exclusively synthesized by adipocytes and has antiatherosclerotic properties. The release of adiponectin is reduced by TNF-α in states like obesity and insulin resistance. We assume that the correlation of PAT volume and systemic inflammation and hypoadiponectinemia, observed in our study, is driven by the fact that PAT volume is a surrogate marker for the entire visceral fat burden. Nevertheless, because of its location around the coronary vessels, evidence suggests that PAT locally enhances systemic atherogenic effects, accelerating initiation and progression of coronary atherosclerosis. In patients undergoing bypass surgery it could be demonstrated that PAT reveals a pathological adipokine and cytokine profile as well as a large number of inflammatory cells, which could not be observed in subcutaneous fat.23 The exact mechanism on how cyto- and chemokines originating from pericardial adipocytes enter the vessel wall and accelerate and initiate atherosclerosis is not clear yet. These inflammatory mediators may exert their effect through diffusion and direct adventitial contact. In a porcine model adventitial treatment with proinflammatory factors resulted in early stage atherosclerosis and migration of inflammatory cells into the vessel wall.24,25 Another mechanism may be a direct communication of the vasa vasorum with epicardial adipocytes. The concept of this kind of local signaling is supported by some interesting observations in human studies: (1) There is a closer correlation and association of PAT to coronary atherosclerosis than of abdominal visceral fat to coronary atherosclerosis.17 (2) In the human myocardial bridge where visceral fat around the coronary vessels is totally absent, no atherosclerotic plaque development can be observed. (3) By using intravascular ultrasound it could be demonstrated that plaques develop most frequently with a pericardial spatial orientation suggesting a permissive role of pericardial fat.26 (4) In coronary arteries that on autopsy revealed plaques with large necrotic cores more macrophages in periadventitial fat could be observed than in vessels without lipidcore plaques.4,27,28
Limitations and Clinical Perspectives
The findings of our study identify PAT volume as a strong and independent risk factor for coronary artery disease. By using CT this local fat depot can be easily quantified even on native CA scoring scans with an acceptable radiation exposure. Future prospective trials, however, are needed to prove the incremental prognostic impact of this parameter over CA scoring and traditional risk factors. As we have found some evidence that pathological PAT accumulation may precede the development of atherosclerotic plaque and indicates a proatherogenic state, this fat depot may be an attractive target for antiatherosclerotic therapies, and future research should also address the effect of different drugs and weight modifying strategies on PAT volume.
Sources of Funding
This work was supported by grants BR 2151/4-1 (to U.C.B.) and LE 1350/2-1 (to M.L.) from the Deutsche Forschungsgemeinschaft.
Alexander Leber is an employee for Siemens AG Healthcare.
M.G., A.B., and A.W.L. contributed equally to this study.
Received September 3, 2008; revision accepted January 30, 2009.
Leber AW, Knez A, von Ziegler F, Becker A, Nikolaou K, Paul S, Wintersperger B, Reiser M, Becker CR, Steinbeck G, Boekstegers P. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol. 2005; 46: 147–154.
Leber AW, Knez A, Becker A, Becker C, von Ziegler F, Nikolaou K, Rist C, Reiser M, White C, Steinbeck G, Boekstegers P. Accuracy of multidetector spiral computed tomography in identifying and differentiating the composition of coronary atherosclerotic plaques: a comparative study with intracoronary ultrasound. J Am Coll Cardiol. 2004; 43: 1241–1247.
Gorter PM, van Lindert AS, de Vos AM, Meijs MF, van der Graaf Y, Doevendans PA, Prokop M, Visseren FL. Quantification of epicardial and peri-coronary fat using cardiac computed tomography; reproducibility and relation with obesity and metabolic syndrome in patients suspected of coronary artery disease. Atherosclerosis. 2008; 197: 896–903.
Leber AW, Becker A, Knez A, von Ziegler F, Sirol M, Nikolaou K, Ohnesorge B, Fayad ZA, Becker CR, Reiser M, Steinbeck G, Boekstegers P. Accuracy of 64-slice computed tomography to classify and quantify plaque volumes in the proximal coronary system: a comparative study using intravascular ultrasound. J Am Coll Cardiol. 2006; 47: 672–677.
Bamberg F, Dannemann N, Shapiro MD, Seneviratne SK, Ferencik M, Butler J, Koenig W, Nasir K, Cury RC, Tawakol A, Achenbach S, Brady TJ, Hoffmann U. Association between cardiovascular risk profiles and the presence and extent of different types of coronary atherosclerotic plaque as detected by multidetector computed tomography. Arterioscler Thromb Vasc Biol. 2008; 28: 568–574.
Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1995; 92: 1355–1374.
Iacobellis G, Ribaudo MC, Assael F, Vecci E, Tiberti C, Zappaterreno A, Di Mario U, Leonetti F. 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.
Ahn SG, Lim HS, Joe DY, Kang SJ, Choi BJ, Choi SY, Yoon MH, Hwang GS, Tahk SJ, Shin JH. Relationship of epicardial adipose tissue by echocardiography to coronary artery disease. Heart. 2008; 94: e7.
Leber AW, Knez A, White CW, Becker A, von Ziegler F, Muehling O, Becker C, Reiser M, Steinbeck G, Boekstegers P. Composition of coronary atherosclerotic plaques in patients with acute myocardial infarction and stable angina pectoris determined by contrast-enhanced multislice computed tomography. Am J Cardiol. 2003; 91: 714–718.
Lear SA, Humphries KH, Kohli S, Frohlich JJ, Birmingham CL, Mancini GB. Visceral adipose tissue, a potential risk factor for carotid atherosclerosis: results of the Multicultural Community Health Assessment Trial (M-CHAT). Stroke. 2007; 38: 2422–2429.
Shimokawa H, Morishige K, Miyata K, Kandabashi T, Eto Y, Ikegaki I, Asano T, Kaibuchi K, Takeshita A. Long-term inhibition of Rho-kinase induces a regression of arteriosclerotic coronary lesions in a porcine model in vivo. Cardiovasc Res. 2001; 51: 169–177.
Miyata K, Shimokawa H, Kandabashi T, Higo T, Morishige K, Eto Y, Egashira K, Kaibuchi K, Takeshita A. Rho-kinase is involved in macrophage-mediated formation of coronary vascular lesions in pigs in vivo. Arterioscler Thromb Vasc Biol. 2000; 20: 2351–2358.
Prati F, Arbustini E, Labellarte A, Sommariva L, Pawlowski T, Manzoli A, Pagano A, Motolese M, Boccanelli A. Eccentric atherosclerotic plaques with positive remodelling have a pericardial distribution: a permissive role of epicardial fat? A three-dimensional intravascular ultrasound study of left anterior descending artery lesions. Eur Heart J. 2003; 24: 329–336.