Abdominal visceral adipose tissue (VAT) is considered a source of inflammatory mediators and a receptacle of inflammatory cells and has been linked with prevalence and incidence of atherosclerosis and its complications in the general population [1–7]. More recently another form of visceral fat, epicardial adipose tissue (EAT), has been shown to be associated with the presence of coronary artery disease and incident cardiovascular events [8–10]. For the majority of their course the coronary arteries are embedded in EAT and it is of interest that intramyocardial coronary artery segments (protected from exposure to epicardial fat) never develop atherosclerosis [11,12]. This induced some investigators to suggest that EAT, as an inflammatory milieu, may sustain the development of atherosclerosis via paracrine mechanisms . HIV is a state of heightened inflammation, and cardiovascular disease events are increasing in the HIV population exposed chronically to antiretroviral therapies (ARTs) [14–18]. Numerous reports have been published in the general population, whereas a single, small report on EAT in HIV-infected patients has been released so far . Lo et al. showed that EAT was greater in HIV patients than controls and it was associated with metabolic alterations; the investigators, however, did not find an association of EAT with subclinical atherosclerosis. Screening for coronary artery calcium (CAC) has been utilized in the general population for almost two decades to refine assessment of cardiovascular risk . The presence of CAC signals the presence of atherosclerosis with an almost 100% specificity. As a marker of atherosclerosis, CAC has been demonstrated to predict cardiovascular events with greater accuracy than traditional risk factors. In particular, a CAC score greater than 100 has been identified in several publications as a marker of substantially increased risk . Accordingly, the aims of this study were to evaluate the relationship between EAT and CAC in HIV patients and to evaluate the association between EAT and HIV infection, ART and lipodystrophy.
This was a cross-sectional observational study of 876 consecutive HIV-infected patients recruited in an outpatient clinic at the University of Modena and Reggio Emilia, Italy, between January 2006 and June 2007. The multidisciplinary team caring for these patients includes infectious disease specialists, cardiologists, endocrinologists, radiologists, nutritionists, personal trainers, psychologists, and plastic surgeons. A signed informed consent to participate in this study was obtained from each patient. Inclusion criteria were: serologically documented HIV-1 infection, age more than 18 years, at least 18 months of ART exposure, and, for patients with established diagnoses of hyperlipidemia and hyperglycemia, stable lipid-lowering and diabetes therapy for at least 6 months. Patients were excluded if they reported or had documented evidence of any of the following cardiovascular conditions: previous myocardial infarction, stroke, coronary artery by-pass surgery or angioplasty, and peripheral vascular disease. Demographic and clinical data, including duration of HIV infection, prior opportunistic diseases (Centers for Disease Control classification), antiretroviral therapy history, and lifestyle, were obtained from medical chart review. Smoking was assessed at entry with the aid of a questionnaire. Hypertension was defined as a supine blood pressure greater than 160 mmHg systolic and/or greater than 90 mmHg diastolic or as chronic therapy with antihypertensive medications.
Measurement of metabolic and anthropometric parameters
Insulin resistance was calculated using the homeostasis model assessment equation: HOMA-IR = [fasting insulin (mU/ml) × fasting glucose (mmol/l)/22.5]. CD4+/μl (most recent value and the lowest value ever reported, known as CD4+ nadir), total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, apolipoprotein A and B, glucose and insulin levels were measured at entry after an overnight fast. Diabetes mellitus was defined as actively taking hypoglycemic drugs or a fasting glucose level less than 126 mg/dl. The presence of the metabolic syndrome was defined according to the criteria proposed by the Adult Treatment Panel III (ATP-III) . The Framingham risk score was calculated for each patient according to the equations proposed by ATP-III .
All patients underwent a physical examination and lipodystrophy assessment. Lipodystrophy was defined using the HIV Outpatient Study (HOPS) definition, with anthropomorphic categorizations of lipoatrophy, lipohypertrophy, and mixed form .
The following anthropometric measurements were made on the same day that serum chemistries were drawn: waist circumference, BMI calculated as weight in kilograms divided by the square of height in meters, and visceral abdominal adipose tissue volume (VAT), using a single-slice abdominal CT scan at the level of the L4 vertebra as per standard protocol .
HIV-related history and immunovirological parameters
Plasma HIV-1 RNA levels and cumulative exposure to non-nucleoside reverse transcriptase inhibitors (NNRTIs), nucleoside reverse transcriptase inhibitors (NRTIs), and protease inhibitors were recorded. HIV viral load was classified as undetectable if less than 40 copies/ml were present. Previous AIDS diagnosis was defined according to the Centers for Disease Control group ‘C’ category . Because NRTIs are the mainstay of all ART regimen, cumulative exposure to NRTI was considered a surrogate for ART cumulative exposure.
Imaging of coronary artery calcium and epicardial adipose tissue
All patients underwent cardiac CT imaging with a Volume CT 64-slice scanner (GE Medical Systems, Milwaukee, Wisconsin, USA). All images were obtained during a single breath hold using 320 mAs and 140 kV. Image acquisition was prospectively triggered at 80% of the R–R interval on the surface electrocardiogram. A section thickness of 2.5 mm, a field of view of 20 cm2, and a matrix of 512 × 512 were used to reconstruct the raw image data, yielding a nominal pixel size of 0.39 mm2 and a voxel of 0.4 mm3. Images were then transferred to an off-line workstation that enabled CAC quantification using the ‘Smart Score’ software (GE Medical Systems). The CAC score was calculated according to the Agatston method, as previously described . Because the CAC score was not normally distributed and its predictive power has been shown to be significantly increased for values greater than 100, this variable was dichotomized as greater or smaller than 100. Neither quadratic nor logarithmic transformation normalized the distribution of the CAC score. For the purpose of analyzing the distribution of EAT with respect to CAC, the population was divided in the following groups: CAC score between 0 and 9, between 10 and 100, and greater than 100. EAT was measured on the same images acquired to detect CAC; therefore, there was no additional radiation exposure to the patients. Epicardial fat was identified on CT as a hypodense rim surrounding the myocardium and limited by pericardium. Quantification of EAT volume was performed on a standalone workstation (Advantage; GE Medical Systems) with dedicated software (Volume Viewer; GE Medical Systems) as described by Gorter et al.. The visceral pericardium was traced manually from the mid-left atrium to the left ventricular apex interpolating slices of 10 mm thickness, and all extrapericardial tissue was excluded. These images were then segmented using an attenuation threshold ranging between −190 and −30 HU providing the EAT volume in each slice. This effectively excluded myocardium, coronary arteries, coronary calcium, aorta, and blood pool. The individual EAT volumes at each level were then summed to determine the ‘total EAT volume’. The reproducibility of this method is excellent as reported by Nichols et al. and Gorter et al.. The total estimated radiation dose for the chest CT was 1.1 mSv.
Normally distributed variables (including EAT measurements) were reported as mean and standard deviation (SD). Non-normally distributed variables (including CAC score) were reported as median and interquartile range (IQR). Kruskal–Wallis and Mann–Whitney U tests were used to assess association between lipodystrophy phenotypes and EAT volume. Factors independently associated with EAT were explored in a multivariable backward stepwise linear regression analysis including: sex, age, BMI, VAT, waist circumference, nadir CD4+ count, current CD4+ count, triglycerides, total cholesterol, HOMA-IR, smoking, hypertension, duration of HIV infection, and cumulative exposure to NNRTIs, NRTIs, protease inhibitors. Multivariable logistic regression was used to evaluate associations between CAC greater than 100 and EAT. A CAC score threshold greater than 100 was chosen because in prior publications in the general population it identified patients at increased risk of cardiovascular events [21,30]. Such association, however, has not been made in HIV patients yet. Other covariates were included when significantly associated in univariate analyses. The ability of measures of general (BMI) and focal adiposity (waist circumference, VAT, and EAT) to predict CAC greater than 100 was compared using receiver operating characteristic (ROC) curves. The significance of the difference of the area under the ROC curves was tested using the roccomp command on Stata. Statistical significance was set at a P less than 0.05. Statistical analyses were performed using the software package STATA 10.1 Intercooled Version for Mac (StataCorp, College Station, Texas, USA).
Table 1 shows the clinical characteristics of the 876 patients enrolled. They were all ART experienced, the majority were men (68%), the mean age was 47.2 ± 8 years, and the mean known duration of HIV infection was long (183 ± 68 months). The CAC score was not normally distributed with a high prevalence of 0 scores; the median CAC score was 0 (IQR 0–11), and the mean score was 40 + 155; 77 (8.8%) patients had a CAC score greater than 100. The prevalence of the metabolic syndrome was 16%, and EAT was significantly larger in patients with the metabolic syndrome than those without (median 87 vs. 70 cm3, P value <0.001). Lipodystrophy was diagnosed in 538 (61%) patients. Lipoatrophy phenotype was present in 251 (29%), lipohypertrophy in 71 (8%), and the mixed form in 216 (25%). Figure 1 depicts the median values and IQR of EAT volume in different phenotypes of lipodystrophy and in its absence. A significant increase in EAT was seen comparing absence of lipodystrophy (64.6 cm3) and lipoatrophy (64.9 cm3) with mixed form (83.4 cm3) and fat accumulation lipodystrophy phenotypes. A similar statistically significant trend was noted for VAT (data not shown).
These trends were similar for men and women, although the median EAT volume in men was larger than in women (80 cm3 IQR 55–103 vs. 61 cm3 IQR 42–82, P < 0.001).
Table 2 shows a list of independent predictors of EAT by stepwise selection. Factors independently associated with EAT were: age [β = 0.6, confidence interval (CI) 0.2–1.0], male sex (β = 6.6, CI 0.5–12.7), VAT (β = 0.12, CI 0.08–0.17), waist circumference (β = 0.7, CI 0.04–1.3), current CD4+ (β = 0.6, CI 0.1–1.2; per 50 cells increase), total cholesterol (β = 0.1, CI 0.02–0.15), cumulative exposure to ART (β = 0.05, CI 0.00–0.11; per months).
As CAC was not normally distributed, the population was divided into the following groups: CAC score between 0 and 9, between 10 and 100, and greater than 100. The EAT volume significantly increased proportional to CAC from 73.6 (35) cm3, to 91.5 (42) cm3 and 98.3 (50) cm3, respectively, in CAC score groups 0–9, 10–100 and greater than 100 (P < 0.001) (Fig. 2).
Among several measurements of adiposity, EAT was the closest predictor of CAC greater than 100, as shown in Fig. 3. Nevertheless, there was no statistical difference between the ROC areas (P = 0.527).
A logistic regression model was used to identify predictors of CAC greater than 100. The following variables were independently associated with CAC greater than 100: EAT volume, age, male sex, and diabetes mellitus (Table 3).
This cross-sectional study showed a clear association between EAT volume, some HIV-specific factors such as current CD4+ cell count, and both central fat accumulation and mixed lipodystrophy phenotypes. Furthermore, we were able to demonstrate an association between EAT and CAC greater than 100, a marker of subclinical atherosclerosis and increased cardiovascular risk. The latter association has been reported previously in the general population [8,21,30] but not in HIV patients. In the only study of EAT in HIV patients published so far, the authors did not find an association between EAT and coronary artery plaque .
Compared to the study recently published by Lo et al., we included a larger number of patients (876 vs. 78), but we did not include controls. Although the general clinical characteristics of the patients were similar (age, sex, hypertension, diabetes mellitus, etc.), several anthropometric measures were different: BMI, waist circumference, and EAT were substantially larger in the North American study compared to our study. Hence, our results may not be generalizable to a US patient population and appear to be more consistent with the epidemiological characteristics of western European countries.
The close correlation of EAT with VAT is suggestive of the potential utility of EAT as a marker of cardiovascular risk in HIV. Visceral fat is a source of inflammatory mediators and may host a large number of highly pro-atherogenic mediators and inflammatory cells. Because EAT and VAT share the same embryological origin, it is conceivable that EAT may be as strong a risk marker of cardiovascular risk as VAT. In fact, in our study we were able to demonstrate that EAT was the best predictor of CAC greater than 100 compared to other anthropometric measures and was positively associated with CAC greater than 100 after adjustment for age, sex, and diabetes. As confirmatory evidence of the importance of this new marker of risk, EAT has been linked with incident cardiovascular events in the general population , although its role in HIV-infected patients awaits confirmation.
We further found an association between current CD4+ cell count and cumulative exposure to ART with EAT. These findings raise the question of whether EAT may represent the expression of a direct ART toxicity, that is the result of metabolic alterations induced by these therapies, and/or a manifestation of the potential pathological effects of immune-reconstitution. Indeed, reconstituted CD4+ lymphocytes may behave in a pro-inflammatory and pro-atherogenic way  rather than being solely protective. T-lymphocytes are increased in the adipose tissue of obese individuals, potentially playing a role in obesity-related inflammation [31–33]. The purported noxious activity of reconstituted CD4+ and CD8+ lymphocytes will need to be explored in future studies that include measures of the immunological activity of these cells. We did not find an association of EAT with glycemia and insulin levels which was instead shown by Lo et al. However, as remarked above, our patients had a substantially smaller EAT volume, BMI, and waste circumference and may have been less prone to develop alterations of glucose metabolism.
A few limitations of this study should be mentioned: ours was a cross-sectional analysis and as such no cause and effect can be inferred from the associations found. As typical of all cross-sectional studies our analyses were most likely affected by a survival bias, that is we measured EAT only in a sample of patients alive at a specific point in time and the ones who had died previously may have had a more or less extensive accumulation of EAT. We did not include controls from the general population to compare the extent of EAT. A few advantages are also worth being noted; we studied a very large group of HIV-infected patients, well characterized from the point of view of their cardiovascular risk profile and free of any prior cardiovascular history. We measured EAT on chest CT images obtained for the purpose of measuring CAC, therefore administering no additional radiation to the patients and at no extra cost. If EAT were found to be a good marker of risk in future studies, it may become more cost effective to simply measure it with transthoracic echocardiography, although this technique is less accurate than CT [8,34].
We conclude that EAT shows promise as a marker of cardiovascular risk in HIV patients, although this will need to be investigated in prospective studies. The clear advantage of measuring EAT is that in the same CT imaging session, one can obtain information on CAC and EAT without additional radiation exposure that would be necessary if one intended to measure VAT as a gauge of cardiovascular risk. Future studies will need to investigate the contribution of immune-reconstitution to the development of systemic inflammation, EAT, and atherosclerosis.
Author contributions: G.G.: directed the metabolic unit at the University of Modena and Reggio Emilia; conceived research question and organized research group; wrote initial draft and several of the subsequent reviews; provided intellectual support related to the field of HIV and metabolic dysfunction as well as cardiovascular risk in HIV. R.S.: data collection, EAT measurement. S.Z.: data management, statistical analysis, manuscript writing, critical review of manuscript drafts and final version. G.O.: data collection, manuscript writing. F.C.: data collection, manuscript writing. G.L.: data collection, EAT measurement. G.B.: data collection, EAT measurement. P.B.: data collection, manuscript writing. R.R.: imaging expertise; critical review of manuscript drafts and final version. M.G.M.: imaging expertise; critical review of manuscript drafts and final version; intellectual support for the research group. P.R.: conceived research question; directed acquisition and interpretation of all imaging modalities; wrote the initial draft and provided critical review of subsequent manuscript drafts and final version; intellectual support for the research group.
1. Hansen T, Ahlstrom H, Soderberg S, Hulthe J, Wikstrom J, Lind L, et al
. Visceral adipose tissue, adiponectin levels and insulin resistance are related to atherosclerosis as assessed by whole-body magnetic resonance angiography in an elderly population. Atherosclerosis 2009; 205:163–167.
2. Lear S, Humphries K, Kohli S, Frohlich J, Birmingham C, Mancini G. 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.
3. Zamboni M, Armellini F, Sheiban I, De Marchi M, Todesco T, Bergamo-Andreis IA, et al
. Relation of body fat distribution in men and degree of coronary narrowings in coronary artery disease. Am J Cardiol 1992; 70:1135–1138.
4. Folsom AR, Kaye SA, Sellers TA, Hong CP, Cerhan JR, Potter JD, et al
. Body fat distribution and 5-year risk of death in older women. JAMA 1993; 269:483–487.
5. Folsom AR, Prineas RJ, Kaye SA, Munger RG. Incidence of hypertension and stroke in relation to body fat distribution and other risk factors in older women. Stroke 1990; 21:701–706.
6. Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjöström L. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12-year follow-up of participants in the population study of women in Gothenburg, Sweden. BMJ (Clin Res Ed) 1984; 289:1257–1261.
7. Nakamura T, Kobayashi H, Yanagi K, Nakagawa T, Nishida M, Kihara S, et al
. Importance of intra-abdominal visceral fat accumulation to coronary atherosclerosis in heterozygous familial hypercholesterolaemia. Int J Obes Relat Metab Disord 1997; 21:580–586.
8. Eroglu S, Sade LE, Yildirir A, Bal U, Ozbicer S, Ozgul AS, et al
. Epicardial adipose tissue thickness by echocardiography is a marker for the presence and severity of coronary artery disease. Nutr Metab Cardiovasc Dis 2009; 19:211–217.
9. Gorter PM, de Vos AM, van der Graaf Y, Stella PR, Doevendans PA, Meijs MFL, et al
. Relation of epicardial and pericoronary fat to coronary atherosclerosis and coronary artery calcium in patients undergoing coronary angiography. Am J Cardiol 2008; 102:380–385.
10. Ding J, Hsu F-C, Harris TB, Liu Y, Kritchevsky SB, Szklo M, et al
. The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA). Am J Clin Nutr 2009; 90:499–504.
11. Robicsek F, Thubrikar MJ. The freedom from atherosclerosis of intramyocardial coronary arteries: reduction of mural stress – a key factor. Eur J Cardiothorac Surg 1994; 8:228–235.
12. Ishii T, Asuwa N, Masuda S, Ishikawa Y. The effects of a myocardial bridge on coronary atherosclerosis and ischaemia. J Pathol 1998; 185:4–9.
13. Mazurek T, Zhang L, Zalewski A, Mannion JD, Diehl JT, Arafat H, et al
. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003; 108:2460–2466.
14. Melendez M, McNurlan M, Mynarcik D, Khan S, Gelato M. Endothelial adhesion molecules are associated with inflammation in subjects with HIV
disease. Clin Infect Dis 2008; 46:775–780.
15. Calza L, Pocaterra D, Pavoni M, Colangeli V, Manfredi R, Verucchi G, et al
. Plasma levels of VCAM-1, ICAM-1, E-Selectin, and P-Selectin in 99 HIV
-positive patients versus 51 HIV
-negative healthy controls. J Acquir Immune Defic Syndr 2009; 50:430–432.
16. Calmy A, Gayet-Ageron A, Montecucco F, Nguyen A, Mach F, Burger F, et al
increases markers of cardiovascular risk: results from a randomized, treatment interruption trial. AIDS 2009; 23:929–939.
17. Hsue P, Hunt P, Schnell A, Kalapus S, Hoh R, Ganz P, et al
. Role of viral replication, antiretroviral therapy, and immunodeficiency in HIV
-associated atherosclerosis. AIDS 2009; 23:1059–1067.
18. Friis-Moller N, Reiss P, Sabin C, Weber R, Monforte A, El-Sadr W, et al
. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med 2007; 356:1723–1735.
19. Lo J, Abbara S, Rocha-Filho JA, Shturman L, Wei J, Grinspoon SK. Increased epicardial adipose tissue volume in HIV
-infected men and relationships to body composition and metabolic parameters. AIDS 2010; 24:2127–2130.
20. Achenbach S, Raggi P. Imaging of coronary atherosclerosis by computed tomography. Eur Heart J 2010; 31:1442–1448.
21. Arad Y, Goodman K, Roth M, Newstein D, Guerci A. Coronary calcification, coronary disease risk factors, C-reactive protein, and atherosclerotic cardiovascular disease events: the St. Francis Heart Study. J Am Coll Cardiol 2005; 46:158–165.
22. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP). Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).JAMA
23. Third Report of the National Cholesterol Education Program (NCEP). Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report.Circulation
24. Lichtenstein K, Ward D, Moorman A, Delaney K, Young B, Palella F, et al
. Clinical assessment of HIV
-associated lipodystrophy in an ambulatory population. AIDS 2001; 15:1389–1398.
25. Guaraldi G, Zona S, Alexopoulos N, Orlando G, Carli F, Ligabue G, et al
. Coronary aging in HIV
-infected patients. Clin Infect Dis 2009; 49:1756–1762.
26. From the Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults.JAMA
27. Agatston A, Janowitz W, Hildner F, Zusmer N, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990; 15:827–832.
28. Gorter PM, van Lindert ASR, de Vos AM, Meijs MFL, van der Graaf Y, Doevendans PA, et al
. 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.
29. Nichols JH, Samy B, Nasir K, Fox CS, Schulze PC, Bamberg F, et al
. Volumetric measurement of pericardial adipose tissue from contrast-enhanced coronary computed tomography angiography: a reproducibility study. J Cardiovasc Comput Tomogr 2008; 2:288–295.
30. Budoff M, Nasir K, McClelland R, Detrano R, Wong N, Blumenthal R, et al
. Coronary calcium predicts events better with absolute calcium scores than age–sex–race/ethnicity percentiles: MESA (Multi-Ethnic Study of Atherosclerosis). J Am Coll Cardiol 2009; 53:345–352.
31. Libby P, Okamoto Y, Rocha VZ, Folco E. Inflammation in atherosclerosis: transition from theory to practice. Circ J 2010; 74:213–220.
32. Wu H, Ghosh S, Perrard XD, Feng L, Garcia GE, Perrard JL, et al
. T-cell accumulation and regulated on activation, normal T cell expressed and secreted upregulation in adipose tissue in obesity. Circulation 2007; 115:1029–1038.
33. Yang H, Youm Y-H, Vandanmagsar B, Ravussin A, Gimble JM, Greenway F, et al
. Obesity increases the production of proinflammatory mediators from adipose tissue T cells and compromises TCR repertoire diversity: implications for systemic inflammation and insulin resistance. J Immunol 2010; 185:1836–1845.
34. Iacobellis G. Echocardiographic epicardial fat: a new tool in the white coat pocket. Nutr Metab Cardiovasc Dis 2008; 18:519–522.