European Heart Journal - Cardiovascular Imaging Advance Access published April 20, 2015 European Heart Journal – Cardiovascular Imaging doi:10.1093/ehjci/jev085
Coronary heart disease risk factors, coronary artery calcification and epicardial fat volume in the Young Finns Study Olli Hartiala 1*, Costan G. Magnussen 1,2, Marco Bucci 3, Sami Kajander 3, Juhani Knuuti 3, Heikki Ukkonen 4, Antti Saraste3,4, Irina Rinta-Kiikka 5, Sakari Kainulainen 6, Mika Ka¨ho¨nen 7, Nina Hutri-Ka¨ho¨nen 8, Tomi Laitinen 9, Terho Lehtima¨ki 10, Jorma S.A. Viikari 4, Jaakko Hartiala 11, Markus Juonala 1,4,12, and Olli T. Raitakari 1,11
Received 12 December 2014; accepted after revision 18 March 2015
Aims
We investigated associations of pre-clinical coronary heart disease (CHD), adolescence and adulthood CHD risk factors, and epicardial fat volume (EFV), which is thought to influence CHD pathology. ..................................................................................................................................................................................... Methods EFV and coronary calcium scores were quantified using computed tomography imaging for 557 subjects from the Cardioand results vascular Risk in Young Finns Study in 2007. CHD risk marker levels were assessed repeatedly from 1980 to 2007. Carotid intima-media thickness (cIMT), carotid distensibility, and brachial flow-mediated dilatation were measured by vascular ultrasound in 2007. Increased EFV was cross-sectionally associated with male sex, increased waist circumference, body-mass index (BMI), cIMT, metabolic syndrome prevalence, levels of apolipoprotein B, total cholesterol, low-density lipoprotein cholesterol, triglycerides, C-reactive protein, blood pressure, insulin, and fasting glucose, as well as ever smoking, alcoholic intake, and lower high-density lipoprotein cholesterol (HDL-C), carotid distensibility and physical activity in adulthood. In BMI-adjusted analyses, only apolipoprotein B, ever smoking, alcohol intake and metabolic syndrome prevalence were independently associated with EFV. In adolescence, skinfold thickness, BMI, and insulin levels were higher and HDL-C lower with increasing EFV. Subjects in the lowest vs. highest quarter of EFV had consistently lower BMI across the early life-course. ..................................................................................................................................................................................... Conclusion Associations of CHD risk markers with EFV were attenuated after multivariable adjustment. We found no evidence of increased EFV being independently associated with pre-clinical atherosclerosis. EFV was most strongly associated with BMI and waist circumference. Subjects with higher EFV had consistently higher BMI from age 12 suggesting that lifelong exposure to higher BMI influences the development of EFV.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atherosclerosis † Epicardial fat † Risk factors † Coronary artery calcium
Introduction Epicardial fat is visceral fat that surrounds the heart and the coronary arteries. The concentration of inflammatory markers is increased in epicardial fat tissue compared with that of subcutaneous fat, and in
patients with coronary heart disease (CHD) compared with those without.1 – 4 Epicardial fat has been suggested to influence CHD development due to its proximity to the coronary arteries and lack of a physical barrier between it and the coronary artery walls and myocardium. Indeed, previous studies have found increased epicardial fat to
* Corresponding author: Tel: +358 40 7084794; Fax: +358 2 3337270, E-mail: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: [email protected].
Downloaded from by guest on May 18, 2015
1 Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 2Menzies Research Institute Tasmania, University of Tasmania, Hobart, TAS 7005, Australia; 3Turku PET Center, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 4 Department of Medicine, University of Turku and Division of Medicine, Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 5Department of Clinical Radiology, Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 6Department of Radiology, Kuopio University Hospital, Puijonlaaksontie, 70210 Kuopio, Finland; 7Department of Clinical Physiology, University of Tampere and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 8Department of Pediatrics, University of Tampere and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 9Department of Clinical Physiology and Nuclear Medicine, University of Eastern Finland and Kuopio University Hospital, Puijonlaaksontie, 70210 Kuopio, Finland; 10Department of Clinical Chemistry, Fimlab Laboratories and University of Tampere School of Medicine and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 11Department of Clinical Physiology and Nuclear Medicine, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; and 12Murdoch Childrens Research Institute, Flemington Road, Parkville VIC 3052, Australia
Page 2 of 8 be associated with incident CHD events,5 – 7 presence and severity of CHD,8 total atherosclerotic plaque burden,9 coronary artery calcium,10 – 14 carotid intima-media thickness (cIMT) and traditional risk factor levels,6,12 and to be increased among obese patients and those with type 11 and 215 diabetes and metabolic syndrome (MetS).16,17 However, controversy still remains as to whether epicardial fat volume (EFV) is a truly independent fat deposit or a marker of overall adiposity. In a number of studies, associations between CHD risk factor levels and EFV have been lost after adjustment for measures of overall adiposity, such as body-mass index (BMI) or waist circumference.8,10,12 We sought to examine the correlations between CHD risk factors and data from carotid ultrasound and coronary artery calcium (CAC) measurements with EFV quantified from multidetector computed tomography (CT) in a population-based sample of middle-aged adults who were first screened for clinical risk factors as adolescents. To our knowledge, no previous studies have examined the association of adolescence risk factors on adult EFV.
Methods Participants The Cardiovascular Risk in Young Finns Study is an ongoing follow-up of atherosclerosis risk factors of Finnish children and young adults carried
O. Hartiala et al.
out in all five Finnish university cities with medical schools (Helsinki, Kuopio, Oulu, Tampere, and Turku) and their rural surrounds. The first cross-sectional survey was conducted in 1980 on 3596 participants aged 3 – 18 years. Follow-up surveys were conducted in 1983, 1986, 2001, and 2007. In 2007, 2204 individuals of the original cohort participated.18 In 2008, a cardiac CT study to measure CAC was conducted among 589 individuals, then aged 40– 46 years. This was a convenience sample; the CT study was offered to all of the 802 participants from the three oldest cohorts who resided in the three centres (Turku, Tampere, and Kuopio) with CAC imaging capability, with the attendance rate equalling 73%. EFV was quantified for 557 subjects with interpretable CT scans (Figure 1). Participants gave written informed consent and the study was approved by local ethics committees.
Clinical characteristics The BMI was calculated as weight, kg/(height, m)2. Blood pressure was measured using a standard mercury sphygmomanometer in 1980 and a random zero sphygmomanometer in 2007. Smoking habits, alcohol intake, and physical activity were inquired with a questionnaire. Smoking was defined as daily smoking in adolescence and/or in adulthood. Ever smoking was defined as a history of daily smoking at any time. Physical activity index was calculated as a metabolic equivalent index by assessing the duration, intensity, and frequency of physical activity, including leisure-time physical activity and commuting.19 Subjects reported their weekly consumption of cans or bottles (1/3 L) of beer, glasses (12 cL) Downloaded from by guest on May 18, 2015
Figure 1 Flow chart of the study sample from 1980 to 2007. Analyses were repeated after excluding 70 participants with regular antihypertensive, lipid-lowering, or diabetes medication with similar results as seen in the final study population.
Page 3 of 8
Association of CHD risk factors, CAC, and EFV
of wine, and shots (4 cL) of strong alcohol. Alcohol intake per day was calculated from the total weekly volume of different beverages consumed. Information on food consumption was assessed with food frequency questionnaires.
Biochemical analyses Venous blood samples were taken after an overnight fast. Standard methods were used to determine glucose, total cholesterol, high-density lipoprotein cholesterol (HDL-C), triglyceride, Apo-B, and Apo-A1 concentrations. Low-density lipoprotein cholesterol (LDL-C) was calculated indirectly using the Friedewald formula.18 In 1980, serum insulin was measured using a modification of the immunoassay method and in 2007, by microparticle enzyme immunoassay kit. Serum high-sensitivity C-reactive protein (CRP) was analysed with a turbidimetric immunoassay kit. HOMA-index was calculated as (insulin × fasting glucose)/22.5.20 Subjects were considered to have type 2 diabetes if they: (i) had a fasting glucose ≥7.0 mmol/L (≥125 mg/dL), (ii) reported receiving oral hypoglycaemic agents and/or insulin and did not have type 1 diabetes, or (iii) reported a history of physician-diagnosed DM2.21 To identify subjects with MetS, we used the modification of the original National Cholesterol Education Programme definition by a joint expert group of the National Heart, Lung, and Blood Institute and the American Heart Association.22
Computed tomography imaging
Ultrasound imaging Ultrasound studies were performed by trained physicians and sonographers following standardized protocols in five centres. Mean IMT was derived using a minimum of four measurements at end diastole from the posterior (far) wall of the left carotid artery approximately 10 mm proximal to the carotid bifurcation. Measurements were made by one experienced reader (M.J.) who was blinded to the subjects’ clinical
Statistical analysis The EFV, EPFV, and TTFV were not normally distributed, therefore, a log-transformation was used. Additionally, subjects were divided into sex-specific quartiles based on their EFVs. Associations between fat volumes and risk factors were assessed using multivariable analysis of variance models adjusted for age and sex. A multivariable regression model including sex, age, BMI, Apo-B, and ever smoking was finally created to establish the magnitude of the individual risk factors’ effect on EFV. Age and sex interactions were examined and they were nonsignificant. Using multi-level mixed modelling with maximum-likelihood estimation, we compared the BMI trajectory (from years 1980, 1983, 1986, 2001, and 2007) as a function of age for two groups; those with EFV in the lowest quarter in 2007 vs. those with EFV in the highest quarter in 2007. This approach allows for missing data (assuming missing data are random) and takes into account correlations between repeated measures on the same individual. We fitted interaction terms between EFV group (lowest vs. upper quarter) and time to compare the trajectory of BMI between EFV groups, allowing us to determine the age at which differences in BMI were apparent. An unstructured covariance matrix was used in all models. A P-value of ,0.05 was considered statistically significant. All computations were performed using SAS version 9.3 or STATA version 13.1.
Results Risk factor levels among those who participated and those who did not are shown in Supplementary data online, Table S1. Seventy participants were either on antihypertensive (n ¼ 62), lipid-lowering medication (n ¼ 19), oral medication for diabetes (n ¼ 2), or used injected insulin (n ¼ 4). No substantial difference in results was observed after these participants were excluded from the analyses (data not shown). Mean EFV was 79.5 + 34.3 cm3, mean EPFV was 106.8 + 69.7 cm3, and mean TTFV was 186.3 + 96.9 cm3. The fat volumes were larger in men: 246.6 + 103.2 vs. 139.6 + 58.5 cm3 for TTFV, 153.8 + 75.9 vs. 70.4 + 33.8 cm3 for EPFV, and 92.8 + 36.1 vs. 69.1 + 29.0 cm3 for EFV (all sex differences P , 0.0001). EFV and EPFV were positively correlated (correlation coefficient 0.70 overall, 0.73 for women and 0.65 for men, all P , 0.0001). Detailed study characteristics are shown in Table 1. Waist circumference, BMI, levels of Apo-B, total cholesterol, LDL-C, triglycerides,
Downloaded from by guest on May 18, 2015
The CT scans were performed with a GE Discovery VCT 64-slice CT/ PET device (Turku), a Philips Brilliance 64-slice CT device (Tampere), and a Siemens Somaton Sensation 16-slice CT device (Kuopio). The average radiation dose received by the subjects was 1.43 mSv. CAC scores were calculated using the Agatston method.23 The coefficient of variation for intraobserver measurements was 4.0%. A phantom with deposits of known calcium concentration was scanned twice using three projections at the three study centres; the coefficient of variation between the phantom scans was 3.9%. Presence of CAC was defined as an Agatston score of 1 or greater. Full details of the study protocol have been published previously.24 Total thoracic fat, extra-pericardial fat and EFVs (TTFV, EPFV and EFV, respectively) in cubic centimetre were calculated using a GE Advantage Workstation (version 4.4) as described by Mahabadi et al. 7 Fat volumes were assessed semi-automatically by first manually tracing regions of interest on non-consecutive slices and making necessary adjustments on the computed interpolations. The analyser then traced the pericardial sac on the images. Within the region of interest, fat was defined as voxels within a window of 2195 to 245 Hounsfield units. Total thoracic fat was defined as any fat tissue located within the thorax from the level of the right pulmonary artery to the diaphragm and from the chest wall to the descending aorta as well as any fat tissue inside the pericardial sac. EFV was defined as fat tissue within the pericardial sac. EPFV was defined as the difference between TTFV and EFV. One reader blinded to participant details performed all measurements. In total, 32 subjects’ scans were non-interpretable due to missing data. Of the three fat volumes, EFV was chosen as the primary outcome because of its reported associations with CHD development.
characteristics. An ultrasound imaging device with a high-resolution system (Sequoia 512; Acuson, CA, USA) and 13.0 MHz linear-array transducer was used. The coefficient of variation between visits was 6.4%. To assess carotid distensibility, the best-quality cardiac cycle was selected from a continuous 5-s image file. The common carotid diameter was measured at least twice during end diastole. Ultrasound and concomitant brachial blood pressure measurements were used to calculate Cdist: Cdist ¼ [(Ds 2 Dd)/Dd]/(Ps 2 Pd), where Dd is diastolic diameter, Ds systolic diameter, Ps systolic blood pressure, and Pd diastolic blood pressure. To assess brachial flow-mediated dilatation, the left brachial artery diameter was measured at rest and during reactive hyperaemia. Increased flow was induced by inflation of a pneumatic tourniquet placed around the forearm to a pressure of 250 mmHg for 4.5 min followed by release. The average of three measurements at rest and 40, 60, and 80 s after cuff release was used to derive maximum flow-mediated dilatation. The maximal vessel diameter in scans after reactive hyperaemia was expressed as the percentage relative to resting scan. Full details of the methods have been described previously.18,25
Page 4 of 8
Table 1
O. Hartiala et al.
Adult risk factor and pre-clinical CHD marker levels by sex-specific EFV quartiles All subjects (n 5 557)
1st quartile (114b cm3; n 5 139)
P-valuec
P-valued
............................................................................................................................................................................... 43.6 41.8 + 2.5
43.5 41.8 + 2.4
43.6 41.8 + 2.4
43.6 41.7 + 2.5
43.9 41.9 + 2.5
,0.0001 0.29
Waist circumference (cm) BMI (kg/m2)
92.1 + 13.3
82.3 + 9.3
88.6 + 10.0
93.4 + 9.3
103.4 + 13.8
,0.0001
26.9 + 5.0
23.6 + 2.8
25.5 + 3.4
27.1 + 3.5
31.3 + 5.9
,0.0001
Apo-B (g/L)
1.1 + 0.3
0.9 + 0.2
1.1 + 0.3
1.1 + 0.3
1.2 + 0.3
,0.0001
0.009
Apo-A1 (g/L) Total cholesterol (mmol/L) LDL-C (mmol/L)
1.6 + 0.2 5.2 + 1.0
1.6 + 0.3 4.9 + 0.9
1.6 + 0.3 5.2 + 0.9
1.6 + 0.2 5.3 + 0.9
1.6 + 0.2 5.3 + 1.0
0.11 ,0.0001
0.054
3.2 + 0.8
3.3 + 0.7
3.3 + 0.8
3.3 + 0.8
3.3 + 0.9
0.001
0.07
HDL-C (mmol/L)
1.3 + 0.3
1.4 + 0.3
1.3 + 0.3
1.3 + 0.3
1.3 + 0.3
,0.0001
0.29
Triglycerides (mmol/L)
1.5 + 1.0
1.1 + 0.6
1.4 + 0.9
1.6 + 1.0
1.8 + 1.1
,0.0001
0.08
C-reactive protein (mg/L)
2.1 + 5.2
1.8 + 8.7
1.5 + 2.1
2.0 + 3.1
3.2 + 4.3
0.003
0.50
Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
123.5 + 16.0
121.1 + 17.3
122.1 + 15.0
122.8 + 15.3
127.8 + 15.4
,0.0001
0.51
77.4 + 12.3
74.9 + 12.7
75.9 + 12.2
76.9 + 11.6
81.7 + 11.7
,0.0001
0.36
Ever smoking (%)
50.58
41.7
54.0
55.4
51.1
0.044
0.0008
Alcoholic intake (units/day)
1.0 + 1.2
0.8 + 1.0
0.9 + 1.1
1.1 + 1.3
1.2 + 1.4
0.003
0.002
Physical activity index
19.3 + 22.5
22.3 + 21.6
21.3 + 24.5
17.3 + 21.4
16.7 + 22.1
0.02
0.29
Insulin (mU/L) Fasting glucose mmol/L HOMA-indexe
9.6 + 10.4 5.4 + 0.7
6.6 + 7.3 5. 3 + 0.6
7.8 + 10.3 5.3 + 0.5
9.6 + 7.2 5.5 + 0.7
14.1 + 13.7 5.6 + 0.9
,0.0001 ,0.0001
0.88 0.84
2.5 + 4.2
1.7 + 2.2
2.0 + 3.3
2.4 + 2.2
3.9 + 6.8
,0.0001
0.86
Fruit intake (g/day)
221.1 + 199.5
249.7 + 204.5
234.8 + 247.8
186.0 + 147.4
215.0 + 184.4
0.37
Vegetable intake (g/day)
277.9 + 181.5
275.6 + 149.3
274.6 + 174.4
265.5 + 175.2
294.4 + 217.8
0.11
Type 2 diabetes (%)
2.3
1.4
0.7
2.1
5.0
0.06
MetS prevalence (%)
22.4
4.0
13.9
23.8
46.6
,0.001
CAC prevalence, %
18.7
16.7
20.7
20.0
17.3
0.61
Intima-media thickness (mm)
0.66 + 0.10
0.64 + 0.10
0.67 + 0.10
0.66 + 0.09
0.68 + 0.10
,0.0001
0.43
Carotid distensibility (%/10 mmHg) Flow-mediated dilation (%)
1.68 + 0.62
1.83 + 0.58
1.61 + 0.61
1.66 + 0.66
1.63 + 0.62
0.002
0.89
8.80 + 4.39
8.29 + 4.40
8.86 + 4.31
9.47 + 4.26
8.61 + 4.55
0.60
Data are mean + SD or proportions. a Range for women. b Range for men. c P-values for multivariable models including age and sex where applicable. d P-values for multivariable models including age, sex, and BMI where applicable. e Calculated as (insulin × fasting glucose)/22.5.
,0.0001
0.01
Downloaded from by guest on May 18, 2015
Male gender (%) Age
Page 5 of 8
Association of CHD risk factors, CAC, and EFV
Table 2
Adolescent risk factor levels by sex-specific EFV quartiles in the Young Finns Study Overall (n 5 557)
1st quartile (n 5 138)
2nd quartile (n 5 140)
3rd quartile (n 5 140)
4th quartile (n 5 139)
P-valuea
,0.0001
P-valueb
............................................................................................................................................................................... Male gender (%)
43.6
43.5
43.6
43.6
43.9
Age
14.8 + 2.5
14.8 + 2.4
14.8 + 2.4
14.7 + 2.5
14.9 + 2.5
0.29
Skinfold thickness (mm) BMI (kg/m2)
29.8 + 13.4 19.7 + 3.0
26.8 + 12.1 18.9 + 2.6
27.9 + 10.8 19.5 + 2.5
30.4 + 13.7 19.5 + 2.6
34.1 + 15.5 20.7 + 3.7
,0.0001 ,0.0001
Apo-B (g/L)
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.45
Apo-A1 (g/L) Total cholesterol (mmol/L)
1.5 + 0.2 5.1 + 0.9
1.5 + 0.2 5.1 + 0.8
1.5 + 0.3 5.1 + 0.8
1.5 + 0.2 5.1 + 1.0
1.5 + 0.2 5.1 + 0.9
0.44 0.69
LDL-C (mmol/L)
3.3 + 0.8
3.2 + 0.7
3.3 + 0.7
3.2 + 0.9
3.3 + 0.8
0.91
HDL-C (mmol/L) Triglycerides (mmol/L)
1.5 + 0.3 0.7 + 0.3
1.6 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.4
0.053 0.37
C-reactive protein (mg/L)
1.1 + 3.4
0.9 + 2.6
1.0 + 3.8
1.0 + 3.0
1.3 + 3.9
0.34
Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
116.5 + 10.7 70.51 + 9.7
116.1 + 11.4 70.5 + 9.4
117.0 + 9.9 71.1 + 9.9
114.8 + 9.7 69.8 + 10.1
118.1 + 11.5 70.6 + 9.5
0.20 0.79
Insulin (mU/L)
13.1 + 5.8
12.2 + 5.3
12.8 + 5.4
13.0 + 5.4
14.4 + 6.8
0.0019
Smoking (%) Ever smoking (%)
16.8 43.4
13.1 40.3
18.7 49.6
22.8 44.4
12.5 39.3
0.42 0.99
Fruit intake (units/week)
6.4 + 2.9
6.7 + 2.8
6.5 + 2.8
6.4 + 3.0
6.1 + 2.8
0.26
Vegetable intake (units/week) Physical activity index
5.6 + 2.9 8.9 + 1.8
5.4 + 3.0 8.9 + 1.9
5.8 + 3.0 9.1 + 1.8
5.6 + 2.9 8.8 + 1.8
5.6 + 2.9 8.8 + 1.8
0.43 0.43
,0.0001
0.37
CRP, systolic and diastolic blood pressure, insulin, and fasting glucose as well former smoking, alcoholic intake, HOMA-index, MetS prevalence, and cIMT increased, while HDL-C levels, physical activity index, and carotid distensibility decreased across EFV quartiles. Adolescent skinfold thickness, BMI, and insulin levels were higher with increasing EFV (Table 2). However, only sex, adult waist circumference, BMI, Apo-B, former smoking, MetS prevalence, and alcohol intake remained statistically significant after multivariable adjustment. Prevalence of CAC was not different across quartiles. In addition, the severity of CAC as quantified by the Agatston score was not associated with EFV (data not shown). In a multivariable regression model including sex, age, waist circumference, Apo-B and a history of smoking, all risk factors except age showed a positive correlation with EFV (Table 3). Figure 2 shows the trajectories for BMI for the highest and lowest quarters of EFV and Supplementary data online, Table S2 shows the raw BMI values for all EFV quarters. These data show that those in the lowest vs. the highest quarter of EFV at the 2007 follow-up had consistently lower BMI across the early life-course. Participants in the highest quarter of EFV in 2007 had, on average, greater mean levels of BMI (2.2 kg/m2, 95%CI, 1.5 – 2.9, P , 0.001) across the lifecourse. The gap between groups increased as participants aged, particularly after adolescence (Figure 2). For example, at age 12 the gap between the two groups was 1.8 kg/m2 and at age 45, it was 9.0 kg/m2.
Table 3 Multivariable regression analysis for the cross-sectional prediction of adult EFV Standardized estimate (+ + SE)
P-value
0.29 + 0.034 0.045 + 0.032
,0.0001 0.15
0.56 + 0.035
,0.0001
0.096 + 0.037 0.11 + 0.032
0.008 0.0007
................................................................................ Sex Age BMI Apo-B Ever smoking
Discussion Some previous studies have suggested that increased EFV may play a role in the development of CHD. Therefore, we examined the associations between CHD risk markers and EFV in population of young and middle-aged adults. We found that CHD risk markers were less favourable with increasing EFV but the associations were attenuated in multivariable adjustment. Therefore, we found no evidence that increased EFV would be independently associated with markers of pre-clinical atherosclerosis, including increased CAC prevalence, in our population. EFV was most strongly associated with BMI and waist circumference. Subjects with higher EFV had higher BMI from age 12 and this association became more pronounced later in life
Downloaded from by guest on May 18, 2015
Data are mean + SD or proportions. a P-values for multivariable models including age and sex where applicable. b P-values for multivariable models including age, sex, and BMI where applicable.
Page 6 of 8
Figure 2 Comparison of the BMI life-course trajectory between participants in the lowest (white circles) vs. the highest (black boxes) sex-specific quarter of EFV in adulthood. *P , 0.001.
and found that those with stiffer carotid arteries at age 36 years had greater levels of blood pressure and central fatness. These findings add to the evidence base that life-long exposure to risk factors is an important factor for the development of CHD-related changes. According to our results, however, other adolescence risk factors than BMI do not seem to play a role in the development of EFV. This is in contrast to our previous findings from the same study population concerning CAC24 and carotid atherosclerosis.25 Unlike several previous studies,9,12 – 14 we could not demonstrate an association between EFV and CAC prevalence or severity after adjustment for sex, age, and waist circumference. Our study confirms the finding by Bucci et al. 33 in subjects with a moderate pretest likelihood of CAD; the correlations between EFV and EPFV and CAC disappeared when correcting for age and sex. Similar results were found by Mahabadi et al. 6 in the Heinz Nixdorf Recall Study, in which they concluded that shared risk factors ultimately accounted for the linkage of epicardial fat and CAC, and Pracon et al. 34 who found that epicardial adipose tissue radiodensity, but not volume, was associated with CAC. In addition, Alexopoulos et al. 11 found that EFV was greater in patients with mixed or non-calcified plaques compared with those with calcified plaques and Gorter et al. 35 found that in patients with suspected CHD, EFV was not associated with CAC overall but in those with low BMI, an association was found. A recent meta-analysis found no difference in EFV between groups of increasing CAC scores.36 A large proportion of studies showing an association between EFV and CAC included selected populations with either known or suspected CHD and were therefore possibly biased. In the Framingham Study, subjects were considerably older and had more prevalent CAC than in our study.12 It is possible that epicardial fat influences CHD development through shared risk factors with CAC. We additionally found that higher cIMT and decreased carotid distensibility were associated with EFV when adjusted for age and sex, however, in multivariable models these associations did not prevail. A few studies have addressed the possible associations between EFV and carotid artery ultrasound measurements. Sengul et al. 18 found that EFV measured by ultrasound was associated with increased IMT in patients with MetS. In the MultiEthnic Study of Atherosclerosis, carotid stiffness,37 but not IMT,38 was associated with EFV even after adjustment for BMI and waist circumference and CHD risk factors. We were also able to link EFV measurements with brachial artery endothelial-dependent flowmediated vasodilation capacity, and found no association. Therefore, we found no evidence that increased EFV would play an independent role in the development of early structural and functional changes of atherosclerosis. For this reason, we find it unlikely that measurement of EFV in clinical practice would bring additional information of CHD risk over and above that gained from CAC measurement alone. A strength of this study is that the study population is a nonselected population thus representative of the general population. Owing to the longitudinal design of this study, bias due to differential loss to follow-up is a potential concern. However, we have previously shown those lost to follow-up do not differ with respect to the major risk factors at baseline18 and attrition in our study has been substantially less when compared with other similar studies.39 Moreover, we found no significant differences between the subset that participated at clinics in 2007 but did not attend cardiac CT scans and those that attended both. Collectively therefore, we expect bias due to
Downloaded from by guest on May 18, 2015
suggesting that life-long exposure to higher BMI plays a role in the development of EFV. Previously, a number of studies have established independent associations between risk factors including waist circumference, male sex, LDL-C and HDL-C levels, diabetes and impaired fasting glucose, MetS and current as well as former smoking, and EFV.6,8,12,15 – 17,26 In many cases, however, the associations have been attenuated or even lost their statistical significance after adjustment for adiposity measures, such as BMI, waist circumference or visceral adipose tissue volume.8,10,12 Similarly in our study, risk factor levels were less favourable across increasing EFV quartiles but for most risk factors, the differences were not statistically significant after adjustment for BMI. This is in agreement with a recent study, where associations with EFV were not statistically significant after adjustment for visceral abdominal adipose tissue in non-obese patients.27 This may indicate that the association of epicardial fat and risk factors is at least partly explained by a strong association between other adiposity measures and EFV. In addition, prevalence of MetS increased across EFV quartiles, which could be expected given the strong association of adiposity measures with EFV in our study. This finding is in agreement with previous studies.16,17,28 To our knowledge, no other follow-up study has been conducted over adolescence risk factors or life-long exposure to risk factors and EFV in adulthood. We and others have, however, shown that adverse childhood or adolescence risk factor levels are predictive of CAC24,29 and carotid atherosclerosis25,30 in adulthood. Schusterova et al. 31 found epicardial adipose tissue thickness measured by echocardiography in overweight children to be associated with blood pressure, triglycerides, uric acid, HDL-C, apo-B, and alanine aminotransferase but it was not a stronger indicator of cardiometabolic risk than BMI alone. In our study, a significant difference in BMI was observed already at age 12 years between those with low vs. high EFV in adulthood. This difference became more pronounced later in life, especially after adolescence, and continued into mid-adulthood. Ferreira et al. 32 used a similar life-course analysis approach regarding carotid stiffness
O. Hartiala et al.
Association of CHD risk factors, CAC, and EFV
differential loss to follow-up to be minimal. Visceral or subcutaneous adipose tissue was not measured because the field of view in the CT scans only covered the heart with varying marginal superiorly and inferiorly and so relations with EFV and these measures could not be studied. Rather, an indirect marker for abdominal adipose tissue, waist circumference, was used. Another limitation is that the IMT and FMD measurements were performed operator-dependently instead of utilizing an automated or semi-automated operatorindependent system. Finally, we are unable to conclude concerning the direction of effect between BMI and EFV. For example, our lifecourse models posit that the BMI separation among the extremes of EFV was already apparent at age 12 years. Because we did not have EFV data from across the life-course we are unable to discount that EFV separation occurred before, after, or at the same time as BMI. In conclusion, most risk associations between risk factor levels and EFV were attenuated in multivariable analyses. Sex, BMI, and waist circumference measured in adulthood were most strongly associated with EFV; Apo-B, a history of smoking, and alcohol intake showed a weaker association with EFV. BMI in adolescence and throughout young adulthood was higher across increasing EFV. No evidence was found that increased EFV would be independently associated with increased CAC prevalence or other structural and functional markers of pre-clinical atherosclerosis.
Supplementary data are available at European Heart Journal – Cardiovascular Imaging online.
Acknowledgements The expert technical assistance in the statistical analyses by Irina Lisinen and Ville Aalto is gratefully acknowledged. Conflict of interest: None declared.
Funding The Cardiovascular Risk in Young Finns Study was financially supported by the Academy of Finland (grants 117787, 121584, 126925, 124282, 129378, 41071), the Social Insurance Institution of Finland, the Turku University Foundation, Kuopio, Tampere and Turku University Hospital Medical Funds, Special Federal Grants for University Hospitals, the Juho Vainio Foundation, Paavo Nurmi Foundation, the Finnish Foundation of Cardiovascular Research, Orion-Farmos Research Foundation, Emil Aaltonen Foundation (T.L), Tampere Tuberculosis Foundation (T.L), Turku University Foundation and the Finnish Cultural Foundation. CGM is supported through a National Health and Medical Research Council Early Career Fellowship (APP1037559).
References 1. Baker AR, Silva NF, Quinn DW, Harte AL, Pagano D, Bonser RS et al. Human epicardial adipose tissue expresses a pathogenic profile of adipocytokines in patients with cardiovascular disease. Cardiovasc Diabetol 2006;5:1. 2. 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– 6. 3. Hirata Y, Kurobe H, Akaike M, Chikugo F, Hori T, Bando Y et al. Enhanced inflammation in epicardial fat in patients with coronary artery disease. Int Heart J 2011;52: 139 –42.
4. Shimabukuro M, Hirata Y, Tabata M, Dagvasumberel M, Sato H, Kurobe H et al. Epicardial adipose tissue volume and adipocytokine imbalance are strongly linked to human coronary atherosclerosis. Arterioscler Thromb Vasc Biol 2013;33:1077 –84. 5. Ding J, Hsu FC, 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. 6. Mahabadi AA, Berg MH, Lehmann N, Ka¨lsch 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 –95. 7. Mahabadi AA, Massaro JM, Rosito GA, Levy D, Murabito JM, Wolf PA et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: The Framingham heart study. Eur Heart J 2009;30: 850 –6. 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;19:211–7. 9. Bettencourt N, Toschke AM, Leite D, Rocha J, Carvalho M, Sampaio F et al. Epicardial adipose tissue is an independent predictor of coronary atherosclerotic burden. Int J Cardiol 2012;158:26 –32. 10. Yerramasu A, Dey D, Venuraju S, Anand DV, Atwal S, Corder R et al. Increased volume of epicardial fat is an independent risk factor for accelerated progression of sub-clinical coronary atherosclerosis. Atherosclerosis 2012;220:223 – 30. 11. Alexopoulos N, McLean DS, Janik M, Arepalli CD, Stillman AE, Raggi P. Epicardial adipose tissue and coronary artery plaque characteristics. Atherosclerosis 2010; 210:150–4. 12. Rosito GA, Massaro JM, Hoffmann U, Ruberg FL, Mahabadi AA, Vasan RS et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: The Framingham heart study. Circulation 2008;117:605 –13. 13. Iwasaki K, Matsumoto T, Aono H, Furukawa H, Samukawa M. Relationship between epicardial fat measured by 64-multidetector computed tomography and coronary artery disease. Clin Cardiol 2011;34:166 –71. 14. Kim S, Chung J, Kwon B, Song S, Choi W. The associations of epicardial adipose tissue with coronary artery disease and coronary atherosclerosis. Int Heart J 2014;55: 197 –203. 15. Wang CP, Hsu HL, Hung WC, Yu TH, Chen YH, Chiu CA et al. Increased epicardial adipose tissue (EAT) volume in type 2 diabetes mellitus and association with metabolic syndrome and severity of coronary atherosclerosis. Clin Endocrinol (Oxf) 2009; 70:876–82. 16. Pierdomenico S, Pierdomenico A, Neri M, Cuccurullo F. Epicardial adipose tissue and metabolic syndrome in hypertensive patients with normal body weight and waist circumference. Am J Hypertens 2011;24:1245 –9. 17. Sengul C, Cevik C, Ozveren O, Oduncu V, Sunbul A, Akgun T et al. Echocardiographic epicardial fat thickness is associated with carotid intima-media thickness in patients with metabolic syndrome. Echocardiography 2011;28:853 –8. 18. Raitakari OT, Juonala M, Ro¨nnemaa T, Keltikangas-Ja¨rvinen L, Ra¨sa¨nen L, Pietika¨inen M et al. Cohort profile: the cardiovascular risk in Young Finns Study. Int J Epidemiol 2008;37:1220 –6. 19. Telama R, Yang X, Viikari J, Va¨lima¨ki I, Wanne O, Raitakari O. Physical activity from childhood to adulthood: a 21-year tracking study. Am J Prev Med 2005;28:267–73. 20. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: 412 –9. 21. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetic Med 1998;15:539 –53. 22. Grundy S, Brewer HB, Cleeman J, Smith S, Lenfant C. Definition of metabolic syndrome: report of the National heart, lung, and blood institute/American Heart Association conference on scientific issues related to definition. Circulation 2004;109: 433 –8. 23. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827 –32. 24. Hartiala O, Magnussen CG, Kajander S, Knuuti J, Ukkonen H, Saraste A et al. Adolescence risk factors are predictive of coronary artery calcification at middle age: the cardiovascular risk in Young Finns study. J Am Coll Cardiol. 2012;60:1364 –70. 25. Raitakari OT, Juonala M, Ka¨ho¨nen M, Taittonen L, Laitinen T, Ma¨ki-Torkko N et al. Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: the cardiovascular risk in Young Finns study. J Am Med Assoc 2003; 290:2277 –83. ¨ zben B, Yavuz D, Deyneli O, Aydın H, Tarcin O ¨ . et al. Epicardial adipose 26. Yazıcı D, O tissue thickness in type 1 diabetic patients. Endocrine 2011;40:250 –5.
Downloaded from by guest on May 18, 2015
Supplementary data
Page 7 of 8
Page 8 of 8 27. De Larochellie`re E, Coˆte´ J, Gilbert G, Bibeau K, Ross MK, Dion-Roy V et al. Visceral/ epicardial adiposity in nonobese and apparently healthy young adults: association with the cardiometabolic profile. Atherosclerosis 2014;234:23 –9. 28. Yorgun H, Canpolat U, Hazırolan T, Ates¸ AH, Sunman H, Dural M et al. Increased epicardial fat tissue is a marker of metabolic syndrome in adult patients. Int J Cardiol 2013;165:308 – 13. 29. Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA et al. Coronary risk factors measured in childhood and young adult life are associated with coronary artery calcification in young adults: The Muscatine study. J Am Coll Cardiol 1996;27: 277 –84. 30. Li S, Chen W, Srinivasan SR, Bond MG, Tang R, Urbina EM et al. Childhood cardiovascular risk factors and carotid vascular changes in adulthood: The Bogalusa Heart study. J Am Med Assoc 2003;290:2271 –6. 31. Schusterova I, Leenen FHH, Jurko A, Sabol F, Takacova J. Epicardial adipose tissue and cardiometabolic risk factors in overweight and obese children and adolescents. Pediatr Obes 2014;9:63–70. 32. Ferreira I, van de Laar RJ, Prins M, Twisk J, Stehouwer C. Carotid stiffness in young adults: a life-course analysis of its early determinants: The Amsterdam growth and health longitudinal study. Hypertension 2012;59:54– 61. 33. Bucci M, Joutsiniemi E, Saraste A, Kajander S, Ukkonen H, Saraste M et al. Intrapericardial, but not extrapericardial, fat is an independent predictor of impaired
O. Hartiala et al.
34.
35.
36.
37.
38.
39.
hyperemic coronary perfusion in coronary artery disease. Arterioscler Thromb Vasc Biol 2011;31:211 –8. Pracon R, Kruk M, Kepka C, Pregowski J, Opolski MP, Dzielinska Z et al. Epicardial adipose tissue radiodensity is independently related to coronary atherosclerosis. A multidetector computed tomography study. Circ J 2011;75:391 –7. Gorter PM, de Vos AM, van der Graaf Y, Stella PR, Doevendans PA, Meijs MF 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 –5. Xu Y, Cheng X, Hong K, Huang C, Wan L. How to interpret epicardial adipose tissue as a cause of coronary artery disease: a meta-analysis. Coron Artery Dis 2012;23: 227 –33. Brinkley TE, Hsu FC, Carr JJ, Hundley WG, Bluemke DA, Polak JF et al. Pericardial fat is associated with carotid stiffness in the Multi-Ethnic Study of Atherosclerosis. Nutr Metab Cardiovasc Dis 2011;21:332 –8. Soliman E, Ding J, Hsu F, Carr JJ, Polak J, Goff D. Association between carotid intimamedia thickness and pericardial fat in the Multi-Ethnic Study of Atherosclerosis (MESA). J Stroke Cerebrovasc Dis 2010;19:58–65. Dwyer T, Sun C, Magnussen C, Raitakari O, Schorck N, Venn A et al. Cohort profile: The International Childhood Cardiovascular Cohort (i3C) consortium. Int J Epidemiol 2013;42:86– 96.
Downloaded from by guest on May 18, 2015
Coronary heart disease risk factors, coronary artery calcification and epicardial fat volume in the Young Finns Study Olli Hartiala 1*, Costan G. Magnussen 1,2, Marco Bucci 3, Sami Kajander 3, Juhani Knuuti 3, Heikki Ukkonen 4, Antti Saraste3,4, Irina Rinta-Kiikka 5, Sakari Kainulainen 6, Mika Ka¨ho¨nen 7, Nina Hutri-Ka¨ho¨nen 8, Tomi Laitinen 9, Terho Lehtima¨ki 10, Jorma S.A. Viikari 4, Jaakko Hartiala 11, Markus Juonala 1,4,12, and Olli T. Raitakari 1,11
Received 12 December 2014; accepted after revision 18 March 2015
Aims
We investigated associations of pre-clinical coronary heart disease (CHD), adolescence and adulthood CHD risk factors, and epicardial fat volume (EFV), which is thought to influence CHD pathology. ..................................................................................................................................................................................... Methods EFV and coronary calcium scores were quantified using computed tomography imaging for 557 subjects from the Cardioand results vascular Risk in Young Finns Study in 2007. CHD risk marker levels were assessed repeatedly from 1980 to 2007. Carotid intima-media thickness (cIMT), carotid distensibility, and brachial flow-mediated dilatation were measured by vascular ultrasound in 2007. Increased EFV was cross-sectionally associated with male sex, increased waist circumference, body-mass index (BMI), cIMT, metabolic syndrome prevalence, levels of apolipoprotein B, total cholesterol, low-density lipoprotein cholesterol, triglycerides, C-reactive protein, blood pressure, insulin, and fasting glucose, as well as ever smoking, alcoholic intake, and lower high-density lipoprotein cholesterol (HDL-C), carotid distensibility and physical activity in adulthood. In BMI-adjusted analyses, only apolipoprotein B, ever smoking, alcohol intake and metabolic syndrome prevalence were independently associated with EFV. In adolescence, skinfold thickness, BMI, and insulin levels were higher and HDL-C lower with increasing EFV. Subjects in the lowest vs. highest quarter of EFV had consistently lower BMI across the early life-course. ..................................................................................................................................................................................... Conclusion Associations of CHD risk markers with EFV were attenuated after multivariable adjustment. We found no evidence of increased EFV being independently associated with pre-clinical atherosclerosis. EFV was most strongly associated with BMI and waist circumference. Subjects with higher EFV had consistently higher BMI from age 12 suggesting that lifelong exposure to higher BMI influences the development of EFV.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atherosclerosis † Epicardial fat † Risk factors † Coronary artery calcium
Introduction Epicardial fat is visceral fat that surrounds the heart and the coronary arteries. The concentration of inflammatory markers is increased in epicardial fat tissue compared with that of subcutaneous fat, and in
patients with coronary heart disease (CHD) compared with those without.1 – 4 Epicardial fat has been suggested to influence CHD development due to its proximity to the coronary arteries and lack of a physical barrier between it and the coronary artery walls and myocardium. Indeed, previous studies have found increased epicardial fat to
* Corresponding author: Tel: +358 40 7084794; Fax: +358 2 3337270, E-mail: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: [email protected].
Downloaded from by guest on May 18, 2015
1 Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 2Menzies Research Institute Tasmania, University of Tasmania, Hobart, TAS 7005, Australia; 3Turku PET Center, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 4 Department of Medicine, University of Turku and Division of Medicine, Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; 5Department of Clinical Radiology, Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 6Department of Radiology, Kuopio University Hospital, Puijonlaaksontie, 70210 Kuopio, Finland; 7Department of Clinical Physiology, University of Tampere and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 8Department of Pediatrics, University of Tampere and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 9Department of Clinical Physiology and Nuclear Medicine, University of Eastern Finland and Kuopio University Hospital, Puijonlaaksontie, 70210 Kuopio, Finland; 10Department of Clinical Chemistry, Fimlab Laboratories and University of Tampere School of Medicine and Tampere University Hospital, Teiskontie, 33520 Tampere, Finland; 11Department of Clinical Physiology and Nuclear Medicine, University of Turku and Turku University Hospital, Kiinamyllynkatu, 20520 Turku, Finland; and 12Murdoch Childrens Research Institute, Flemington Road, Parkville VIC 3052, Australia
Page 2 of 8 be associated with incident CHD events,5 – 7 presence and severity of CHD,8 total atherosclerotic plaque burden,9 coronary artery calcium,10 – 14 carotid intima-media thickness (cIMT) and traditional risk factor levels,6,12 and to be increased among obese patients and those with type 11 and 215 diabetes and metabolic syndrome (MetS).16,17 However, controversy still remains as to whether epicardial fat volume (EFV) is a truly independent fat deposit or a marker of overall adiposity. In a number of studies, associations between CHD risk factor levels and EFV have been lost after adjustment for measures of overall adiposity, such as body-mass index (BMI) or waist circumference.8,10,12 We sought to examine the correlations between CHD risk factors and data from carotid ultrasound and coronary artery calcium (CAC) measurements with EFV quantified from multidetector computed tomography (CT) in a population-based sample of middle-aged adults who were first screened for clinical risk factors as adolescents. To our knowledge, no previous studies have examined the association of adolescence risk factors on adult EFV.
Methods Participants The Cardiovascular Risk in Young Finns Study is an ongoing follow-up of atherosclerosis risk factors of Finnish children and young adults carried
O. Hartiala et al.
out in all five Finnish university cities with medical schools (Helsinki, Kuopio, Oulu, Tampere, and Turku) and their rural surrounds. The first cross-sectional survey was conducted in 1980 on 3596 participants aged 3 – 18 years. Follow-up surveys were conducted in 1983, 1986, 2001, and 2007. In 2007, 2204 individuals of the original cohort participated.18 In 2008, a cardiac CT study to measure CAC was conducted among 589 individuals, then aged 40– 46 years. This was a convenience sample; the CT study was offered to all of the 802 participants from the three oldest cohorts who resided in the three centres (Turku, Tampere, and Kuopio) with CAC imaging capability, with the attendance rate equalling 73%. EFV was quantified for 557 subjects with interpretable CT scans (Figure 1). Participants gave written informed consent and the study was approved by local ethics committees.
Clinical characteristics The BMI was calculated as weight, kg/(height, m)2. Blood pressure was measured using a standard mercury sphygmomanometer in 1980 and a random zero sphygmomanometer in 2007. Smoking habits, alcohol intake, and physical activity were inquired with a questionnaire. Smoking was defined as daily smoking in adolescence and/or in adulthood. Ever smoking was defined as a history of daily smoking at any time. Physical activity index was calculated as a metabolic equivalent index by assessing the duration, intensity, and frequency of physical activity, including leisure-time physical activity and commuting.19 Subjects reported their weekly consumption of cans or bottles (1/3 L) of beer, glasses (12 cL) Downloaded from by guest on May 18, 2015
Figure 1 Flow chart of the study sample from 1980 to 2007. Analyses were repeated after excluding 70 participants with regular antihypertensive, lipid-lowering, or diabetes medication with similar results as seen in the final study population.
Page 3 of 8
Association of CHD risk factors, CAC, and EFV
of wine, and shots (4 cL) of strong alcohol. Alcohol intake per day was calculated from the total weekly volume of different beverages consumed. Information on food consumption was assessed with food frequency questionnaires.
Biochemical analyses Venous blood samples were taken after an overnight fast. Standard methods were used to determine glucose, total cholesterol, high-density lipoprotein cholesterol (HDL-C), triglyceride, Apo-B, and Apo-A1 concentrations. Low-density lipoprotein cholesterol (LDL-C) was calculated indirectly using the Friedewald formula.18 In 1980, serum insulin was measured using a modification of the immunoassay method and in 2007, by microparticle enzyme immunoassay kit. Serum high-sensitivity C-reactive protein (CRP) was analysed with a turbidimetric immunoassay kit. HOMA-index was calculated as (insulin × fasting glucose)/22.5.20 Subjects were considered to have type 2 diabetes if they: (i) had a fasting glucose ≥7.0 mmol/L (≥125 mg/dL), (ii) reported receiving oral hypoglycaemic agents and/or insulin and did not have type 1 diabetes, or (iii) reported a history of physician-diagnosed DM2.21 To identify subjects with MetS, we used the modification of the original National Cholesterol Education Programme definition by a joint expert group of the National Heart, Lung, and Blood Institute and the American Heart Association.22
Computed tomography imaging
Ultrasound imaging Ultrasound studies were performed by trained physicians and sonographers following standardized protocols in five centres. Mean IMT was derived using a minimum of four measurements at end diastole from the posterior (far) wall of the left carotid artery approximately 10 mm proximal to the carotid bifurcation. Measurements were made by one experienced reader (M.J.) who was blinded to the subjects’ clinical
Statistical analysis The EFV, EPFV, and TTFV were not normally distributed, therefore, a log-transformation was used. Additionally, subjects were divided into sex-specific quartiles based on their EFVs. Associations between fat volumes and risk factors were assessed using multivariable analysis of variance models adjusted for age and sex. A multivariable regression model including sex, age, BMI, Apo-B, and ever smoking was finally created to establish the magnitude of the individual risk factors’ effect on EFV. Age and sex interactions were examined and they were nonsignificant. Using multi-level mixed modelling with maximum-likelihood estimation, we compared the BMI trajectory (from years 1980, 1983, 1986, 2001, and 2007) as a function of age for two groups; those with EFV in the lowest quarter in 2007 vs. those with EFV in the highest quarter in 2007. This approach allows for missing data (assuming missing data are random) and takes into account correlations between repeated measures on the same individual. We fitted interaction terms between EFV group (lowest vs. upper quarter) and time to compare the trajectory of BMI between EFV groups, allowing us to determine the age at which differences in BMI were apparent. An unstructured covariance matrix was used in all models. A P-value of ,0.05 was considered statistically significant. All computations were performed using SAS version 9.3 or STATA version 13.1.
Results Risk factor levels among those who participated and those who did not are shown in Supplementary data online, Table S1. Seventy participants were either on antihypertensive (n ¼ 62), lipid-lowering medication (n ¼ 19), oral medication for diabetes (n ¼ 2), or used injected insulin (n ¼ 4). No substantial difference in results was observed after these participants were excluded from the analyses (data not shown). Mean EFV was 79.5 + 34.3 cm3, mean EPFV was 106.8 + 69.7 cm3, and mean TTFV was 186.3 + 96.9 cm3. The fat volumes were larger in men: 246.6 + 103.2 vs. 139.6 + 58.5 cm3 for TTFV, 153.8 + 75.9 vs. 70.4 + 33.8 cm3 for EPFV, and 92.8 + 36.1 vs. 69.1 + 29.0 cm3 for EFV (all sex differences P , 0.0001). EFV and EPFV were positively correlated (correlation coefficient 0.70 overall, 0.73 for women and 0.65 for men, all P , 0.0001). Detailed study characteristics are shown in Table 1. Waist circumference, BMI, levels of Apo-B, total cholesterol, LDL-C, triglycerides,
Downloaded from by guest on May 18, 2015
The CT scans were performed with a GE Discovery VCT 64-slice CT/ PET device (Turku), a Philips Brilliance 64-slice CT device (Tampere), and a Siemens Somaton Sensation 16-slice CT device (Kuopio). The average radiation dose received by the subjects was 1.43 mSv. CAC scores were calculated using the Agatston method.23 The coefficient of variation for intraobserver measurements was 4.0%. A phantom with deposits of known calcium concentration was scanned twice using three projections at the three study centres; the coefficient of variation between the phantom scans was 3.9%. Presence of CAC was defined as an Agatston score of 1 or greater. Full details of the study protocol have been published previously.24 Total thoracic fat, extra-pericardial fat and EFVs (TTFV, EPFV and EFV, respectively) in cubic centimetre were calculated using a GE Advantage Workstation (version 4.4) as described by Mahabadi et al. 7 Fat volumes were assessed semi-automatically by first manually tracing regions of interest on non-consecutive slices and making necessary adjustments on the computed interpolations. The analyser then traced the pericardial sac on the images. Within the region of interest, fat was defined as voxels within a window of 2195 to 245 Hounsfield units. Total thoracic fat was defined as any fat tissue located within the thorax from the level of the right pulmonary artery to the diaphragm and from the chest wall to the descending aorta as well as any fat tissue inside the pericardial sac. EFV was defined as fat tissue within the pericardial sac. EPFV was defined as the difference between TTFV and EFV. One reader blinded to participant details performed all measurements. In total, 32 subjects’ scans were non-interpretable due to missing data. Of the three fat volumes, EFV was chosen as the primary outcome because of its reported associations with CHD development.
characteristics. An ultrasound imaging device with a high-resolution system (Sequoia 512; Acuson, CA, USA) and 13.0 MHz linear-array transducer was used. The coefficient of variation between visits was 6.4%. To assess carotid distensibility, the best-quality cardiac cycle was selected from a continuous 5-s image file. The common carotid diameter was measured at least twice during end diastole. Ultrasound and concomitant brachial blood pressure measurements were used to calculate Cdist: Cdist ¼ [(Ds 2 Dd)/Dd]/(Ps 2 Pd), where Dd is diastolic diameter, Ds systolic diameter, Ps systolic blood pressure, and Pd diastolic blood pressure. To assess brachial flow-mediated dilatation, the left brachial artery diameter was measured at rest and during reactive hyperaemia. Increased flow was induced by inflation of a pneumatic tourniquet placed around the forearm to a pressure of 250 mmHg for 4.5 min followed by release. The average of three measurements at rest and 40, 60, and 80 s after cuff release was used to derive maximum flow-mediated dilatation. The maximal vessel diameter in scans after reactive hyperaemia was expressed as the percentage relative to resting scan. Full details of the methods have been described previously.18,25
Page 4 of 8
Table 1
O. Hartiala et al.
Adult risk factor and pre-clinical CHD marker levels by sex-specific EFV quartiles All subjects (n 5 557)
1st quartile (114b cm3; n 5 139)
P-valuec
P-valued
............................................................................................................................................................................... 43.6 41.8 + 2.5
43.5 41.8 + 2.4
43.6 41.8 + 2.4
43.6 41.7 + 2.5
43.9 41.9 + 2.5
,0.0001 0.29
Waist circumference (cm) BMI (kg/m2)
92.1 + 13.3
82.3 + 9.3
88.6 + 10.0
93.4 + 9.3
103.4 + 13.8
,0.0001
26.9 + 5.0
23.6 + 2.8
25.5 + 3.4
27.1 + 3.5
31.3 + 5.9
,0.0001
Apo-B (g/L)
1.1 + 0.3
0.9 + 0.2
1.1 + 0.3
1.1 + 0.3
1.2 + 0.3
,0.0001
0.009
Apo-A1 (g/L) Total cholesterol (mmol/L) LDL-C (mmol/L)
1.6 + 0.2 5.2 + 1.0
1.6 + 0.3 4.9 + 0.9
1.6 + 0.3 5.2 + 0.9
1.6 + 0.2 5.3 + 0.9
1.6 + 0.2 5.3 + 1.0
0.11 ,0.0001
0.054
3.2 + 0.8
3.3 + 0.7
3.3 + 0.8
3.3 + 0.8
3.3 + 0.9
0.001
0.07
HDL-C (mmol/L)
1.3 + 0.3
1.4 + 0.3
1.3 + 0.3
1.3 + 0.3
1.3 + 0.3
,0.0001
0.29
Triglycerides (mmol/L)
1.5 + 1.0
1.1 + 0.6
1.4 + 0.9
1.6 + 1.0
1.8 + 1.1
,0.0001
0.08
C-reactive protein (mg/L)
2.1 + 5.2
1.8 + 8.7
1.5 + 2.1
2.0 + 3.1
3.2 + 4.3
0.003
0.50
Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
123.5 + 16.0
121.1 + 17.3
122.1 + 15.0
122.8 + 15.3
127.8 + 15.4
,0.0001
0.51
77.4 + 12.3
74.9 + 12.7
75.9 + 12.2
76.9 + 11.6
81.7 + 11.7
,0.0001
0.36
Ever smoking (%)
50.58
41.7
54.0
55.4
51.1
0.044
0.0008
Alcoholic intake (units/day)
1.0 + 1.2
0.8 + 1.0
0.9 + 1.1
1.1 + 1.3
1.2 + 1.4
0.003
0.002
Physical activity index
19.3 + 22.5
22.3 + 21.6
21.3 + 24.5
17.3 + 21.4
16.7 + 22.1
0.02
0.29
Insulin (mU/L) Fasting glucose mmol/L HOMA-indexe
9.6 + 10.4 5.4 + 0.7
6.6 + 7.3 5. 3 + 0.6
7.8 + 10.3 5.3 + 0.5
9.6 + 7.2 5.5 + 0.7
14.1 + 13.7 5.6 + 0.9
,0.0001 ,0.0001
0.88 0.84
2.5 + 4.2
1.7 + 2.2
2.0 + 3.3
2.4 + 2.2
3.9 + 6.8
,0.0001
0.86
Fruit intake (g/day)
221.1 + 199.5
249.7 + 204.5
234.8 + 247.8
186.0 + 147.4
215.0 + 184.4
0.37
Vegetable intake (g/day)
277.9 + 181.5
275.6 + 149.3
274.6 + 174.4
265.5 + 175.2
294.4 + 217.8
0.11
Type 2 diabetes (%)
2.3
1.4
0.7
2.1
5.0
0.06
MetS prevalence (%)
22.4
4.0
13.9
23.8
46.6
,0.001
CAC prevalence, %
18.7
16.7
20.7
20.0
17.3
0.61
Intima-media thickness (mm)
0.66 + 0.10
0.64 + 0.10
0.67 + 0.10
0.66 + 0.09
0.68 + 0.10
,0.0001
0.43
Carotid distensibility (%/10 mmHg) Flow-mediated dilation (%)
1.68 + 0.62
1.83 + 0.58
1.61 + 0.61
1.66 + 0.66
1.63 + 0.62
0.002
0.89
8.80 + 4.39
8.29 + 4.40
8.86 + 4.31
9.47 + 4.26
8.61 + 4.55
0.60
Data are mean + SD or proportions. a Range for women. b Range for men. c P-values for multivariable models including age and sex where applicable. d P-values for multivariable models including age, sex, and BMI where applicable. e Calculated as (insulin × fasting glucose)/22.5.
,0.0001
0.01
Downloaded from by guest on May 18, 2015
Male gender (%) Age
Page 5 of 8
Association of CHD risk factors, CAC, and EFV
Table 2
Adolescent risk factor levels by sex-specific EFV quartiles in the Young Finns Study Overall (n 5 557)
1st quartile (n 5 138)
2nd quartile (n 5 140)
3rd quartile (n 5 140)
4th quartile (n 5 139)
P-valuea
,0.0001
P-valueb
............................................................................................................................................................................... Male gender (%)
43.6
43.5
43.6
43.6
43.9
Age
14.8 + 2.5
14.8 + 2.4
14.8 + 2.4
14.7 + 2.5
14.9 + 2.5
0.29
Skinfold thickness (mm) BMI (kg/m2)
29.8 + 13.4 19.7 + 3.0
26.8 + 12.1 18.9 + 2.6
27.9 + 10.8 19.5 + 2.5
30.4 + 13.7 19.5 + 2.6
34.1 + 15.5 20.7 + 3.7
,0.0001 ,0.0001
Apo-B (g/L)
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.9 + 0.2
0.45
Apo-A1 (g/L) Total cholesterol (mmol/L)
1.5 + 0.2 5.1 + 0.9
1.5 + 0.2 5.1 + 0.8
1.5 + 0.3 5.1 + 0.8
1.5 + 0.2 5.1 + 1.0
1.5 + 0.2 5.1 + 0.9
0.44 0.69
LDL-C (mmol/L)
3.3 + 0.8
3.2 + 0.7
3.3 + 0.7
3.2 + 0.9
3.3 + 0.8
0.91
HDL-C (mmol/L) Triglycerides (mmol/L)
1.5 + 0.3 0.7 + 0.3
1.6 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.3
1.5 + 0.3 0.7 + 0.4
0.053 0.37
C-reactive protein (mg/L)
1.1 + 3.4
0.9 + 2.6
1.0 + 3.8
1.0 + 3.0
1.3 + 3.9
0.34
Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
116.5 + 10.7 70.51 + 9.7
116.1 + 11.4 70.5 + 9.4
117.0 + 9.9 71.1 + 9.9
114.8 + 9.7 69.8 + 10.1
118.1 + 11.5 70.6 + 9.5
0.20 0.79
Insulin (mU/L)
13.1 + 5.8
12.2 + 5.3
12.8 + 5.4
13.0 + 5.4
14.4 + 6.8
0.0019
Smoking (%) Ever smoking (%)
16.8 43.4
13.1 40.3
18.7 49.6
22.8 44.4
12.5 39.3
0.42 0.99
Fruit intake (units/week)
6.4 + 2.9
6.7 + 2.8
6.5 + 2.8
6.4 + 3.0
6.1 + 2.8
0.26
Vegetable intake (units/week) Physical activity index
5.6 + 2.9 8.9 + 1.8
5.4 + 3.0 8.9 + 1.9
5.8 + 3.0 9.1 + 1.8
5.6 + 2.9 8.8 + 1.8
5.6 + 2.9 8.8 + 1.8
0.43 0.43
,0.0001
0.37
CRP, systolic and diastolic blood pressure, insulin, and fasting glucose as well former smoking, alcoholic intake, HOMA-index, MetS prevalence, and cIMT increased, while HDL-C levels, physical activity index, and carotid distensibility decreased across EFV quartiles. Adolescent skinfold thickness, BMI, and insulin levels were higher with increasing EFV (Table 2). However, only sex, adult waist circumference, BMI, Apo-B, former smoking, MetS prevalence, and alcohol intake remained statistically significant after multivariable adjustment. Prevalence of CAC was not different across quartiles. In addition, the severity of CAC as quantified by the Agatston score was not associated with EFV (data not shown). In a multivariable regression model including sex, age, waist circumference, Apo-B and a history of smoking, all risk factors except age showed a positive correlation with EFV (Table 3). Figure 2 shows the trajectories for BMI for the highest and lowest quarters of EFV and Supplementary data online, Table S2 shows the raw BMI values for all EFV quarters. These data show that those in the lowest vs. the highest quarter of EFV at the 2007 follow-up had consistently lower BMI across the early life-course. Participants in the highest quarter of EFV in 2007 had, on average, greater mean levels of BMI (2.2 kg/m2, 95%CI, 1.5 – 2.9, P , 0.001) across the lifecourse. The gap between groups increased as participants aged, particularly after adolescence (Figure 2). For example, at age 12 the gap between the two groups was 1.8 kg/m2 and at age 45, it was 9.0 kg/m2.
Table 3 Multivariable regression analysis for the cross-sectional prediction of adult EFV Standardized estimate (+ + SE)
P-value
0.29 + 0.034 0.045 + 0.032
,0.0001 0.15
0.56 + 0.035
,0.0001
0.096 + 0.037 0.11 + 0.032
0.008 0.0007
................................................................................ Sex Age BMI Apo-B Ever smoking
Discussion Some previous studies have suggested that increased EFV may play a role in the development of CHD. Therefore, we examined the associations between CHD risk markers and EFV in population of young and middle-aged adults. We found that CHD risk markers were less favourable with increasing EFV but the associations were attenuated in multivariable adjustment. Therefore, we found no evidence that increased EFV would be independently associated with markers of pre-clinical atherosclerosis, including increased CAC prevalence, in our population. EFV was most strongly associated with BMI and waist circumference. Subjects with higher EFV had higher BMI from age 12 and this association became more pronounced later in life
Downloaded from by guest on May 18, 2015
Data are mean + SD or proportions. a P-values for multivariable models including age and sex where applicable. b P-values for multivariable models including age, sex, and BMI where applicable.
Page 6 of 8
Figure 2 Comparison of the BMI life-course trajectory between participants in the lowest (white circles) vs. the highest (black boxes) sex-specific quarter of EFV in adulthood. *P , 0.001.
and found that those with stiffer carotid arteries at age 36 years had greater levels of blood pressure and central fatness. These findings add to the evidence base that life-long exposure to risk factors is an important factor for the development of CHD-related changes. According to our results, however, other adolescence risk factors than BMI do not seem to play a role in the development of EFV. This is in contrast to our previous findings from the same study population concerning CAC24 and carotid atherosclerosis.25 Unlike several previous studies,9,12 – 14 we could not demonstrate an association between EFV and CAC prevalence or severity after adjustment for sex, age, and waist circumference. Our study confirms the finding by Bucci et al. 33 in subjects with a moderate pretest likelihood of CAD; the correlations between EFV and EPFV and CAC disappeared when correcting for age and sex. Similar results were found by Mahabadi et al. 6 in the Heinz Nixdorf Recall Study, in which they concluded that shared risk factors ultimately accounted for the linkage of epicardial fat and CAC, and Pracon et al. 34 who found that epicardial adipose tissue radiodensity, but not volume, was associated with CAC. In addition, Alexopoulos et al. 11 found that EFV was greater in patients with mixed or non-calcified plaques compared with those with calcified plaques and Gorter et al. 35 found that in patients with suspected CHD, EFV was not associated with CAC overall but in those with low BMI, an association was found. A recent meta-analysis found no difference in EFV between groups of increasing CAC scores.36 A large proportion of studies showing an association between EFV and CAC included selected populations with either known or suspected CHD and were therefore possibly biased. In the Framingham Study, subjects were considerably older and had more prevalent CAC than in our study.12 It is possible that epicardial fat influences CHD development through shared risk factors with CAC. We additionally found that higher cIMT and decreased carotid distensibility were associated with EFV when adjusted for age and sex, however, in multivariable models these associations did not prevail. A few studies have addressed the possible associations between EFV and carotid artery ultrasound measurements. Sengul et al. 18 found that EFV measured by ultrasound was associated with increased IMT in patients with MetS. In the MultiEthnic Study of Atherosclerosis, carotid stiffness,37 but not IMT,38 was associated with EFV even after adjustment for BMI and waist circumference and CHD risk factors. We were also able to link EFV measurements with brachial artery endothelial-dependent flowmediated vasodilation capacity, and found no association. Therefore, we found no evidence that increased EFV would play an independent role in the development of early structural and functional changes of atherosclerosis. For this reason, we find it unlikely that measurement of EFV in clinical practice would bring additional information of CHD risk over and above that gained from CAC measurement alone. A strength of this study is that the study population is a nonselected population thus representative of the general population. Owing to the longitudinal design of this study, bias due to differential loss to follow-up is a potential concern. However, we have previously shown those lost to follow-up do not differ with respect to the major risk factors at baseline18 and attrition in our study has been substantially less when compared with other similar studies.39 Moreover, we found no significant differences between the subset that participated at clinics in 2007 but did not attend cardiac CT scans and those that attended both. Collectively therefore, we expect bias due to
Downloaded from by guest on May 18, 2015
suggesting that life-long exposure to higher BMI plays a role in the development of EFV. Previously, a number of studies have established independent associations between risk factors including waist circumference, male sex, LDL-C and HDL-C levels, diabetes and impaired fasting glucose, MetS and current as well as former smoking, and EFV.6,8,12,15 – 17,26 In many cases, however, the associations have been attenuated or even lost their statistical significance after adjustment for adiposity measures, such as BMI, waist circumference or visceral adipose tissue volume.8,10,12 Similarly in our study, risk factor levels were less favourable across increasing EFV quartiles but for most risk factors, the differences were not statistically significant after adjustment for BMI. This is in agreement with a recent study, where associations with EFV were not statistically significant after adjustment for visceral abdominal adipose tissue in non-obese patients.27 This may indicate that the association of epicardial fat and risk factors is at least partly explained by a strong association between other adiposity measures and EFV. In addition, prevalence of MetS increased across EFV quartiles, which could be expected given the strong association of adiposity measures with EFV in our study. This finding is in agreement with previous studies.16,17,28 To our knowledge, no other follow-up study has been conducted over adolescence risk factors or life-long exposure to risk factors and EFV in adulthood. We and others have, however, shown that adverse childhood or adolescence risk factor levels are predictive of CAC24,29 and carotid atherosclerosis25,30 in adulthood. Schusterova et al. 31 found epicardial adipose tissue thickness measured by echocardiography in overweight children to be associated with blood pressure, triglycerides, uric acid, HDL-C, apo-B, and alanine aminotransferase but it was not a stronger indicator of cardiometabolic risk than BMI alone. In our study, a significant difference in BMI was observed already at age 12 years between those with low vs. high EFV in adulthood. This difference became more pronounced later in life, especially after adolescence, and continued into mid-adulthood. Ferreira et al. 32 used a similar life-course analysis approach regarding carotid stiffness
O. Hartiala et al.
Association of CHD risk factors, CAC, and EFV
differential loss to follow-up to be minimal. Visceral or subcutaneous adipose tissue was not measured because the field of view in the CT scans only covered the heart with varying marginal superiorly and inferiorly and so relations with EFV and these measures could not be studied. Rather, an indirect marker for abdominal adipose tissue, waist circumference, was used. Another limitation is that the IMT and FMD measurements were performed operator-dependently instead of utilizing an automated or semi-automated operatorindependent system. Finally, we are unable to conclude concerning the direction of effect between BMI and EFV. For example, our lifecourse models posit that the BMI separation among the extremes of EFV was already apparent at age 12 years. Because we did not have EFV data from across the life-course we are unable to discount that EFV separation occurred before, after, or at the same time as BMI. In conclusion, most risk associations between risk factor levels and EFV were attenuated in multivariable analyses. Sex, BMI, and waist circumference measured in adulthood were most strongly associated with EFV; Apo-B, a history of smoking, and alcohol intake showed a weaker association with EFV. BMI in adolescence and throughout young adulthood was higher across increasing EFV. No evidence was found that increased EFV would be independently associated with increased CAC prevalence or other structural and functional markers of pre-clinical atherosclerosis.
Supplementary data are available at European Heart Journal – Cardiovascular Imaging online.
Acknowledgements The expert technical assistance in the statistical analyses by Irina Lisinen and Ville Aalto is gratefully acknowledged. Conflict of interest: None declared.
Funding The Cardiovascular Risk in Young Finns Study was financially supported by the Academy of Finland (grants 117787, 121584, 126925, 124282, 129378, 41071), the Social Insurance Institution of Finland, the Turku University Foundation, Kuopio, Tampere and Turku University Hospital Medical Funds, Special Federal Grants for University Hospitals, the Juho Vainio Foundation, Paavo Nurmi Foundation, the Finnish Foundation of Cardiovascular Research, Orion-Farmos Research Foundation, Emil Aaltonen Foundation (T.L), Tampere Tuberculosis Foundation (T.L), Turku University Foundation and the Finnish Cultural Foundation. CGM is supported through a National Health and Medical Research Council Early Career Fellowship (APP1037559).
References 1. Baker AR, Silva NF, Quinn DW, Harte AL, Pagano D, Bonser RS et al. Human epicardial adipose tissue expresses a pathogenic profile of adipocytokines in patients with cardiovascular disease. Cardiovasc Diabetol 2006;5:1. 2. 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– 6. 3. Hirata Y, Kurobe H, Akaike M, Chikugo F, Hori T, Bando Y et al. Enhanced inflammation in epicardial fat in patients with coronary artery disease. Int Heart J 2011;52: 139 –42.
4. Shimabukuro M, Hirata Y, Tabata M, Dagvasumberel M, Sato H, Kurobe H et al. Epicardial adipose tissue volume and adipocytokine imbalance are strongly linked to human coronary atherosclerosis. Arterioscler Thromb Vasc Biol 2013;33:1077 –84. 5. Ding J, Hsu FC, 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. 6. Mahabadi AA, Berg MH, Lehmann N, Ka¨lsch 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 –95. 7. Mahabadi AA, Massaro JM, Rosito GA, Levy D, Murabito JM, Wolf PA et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: The Framingham heart study. Eur Heart J 2009;30: 850 –6. 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;19:211–7. 9. Bettencourt N, Toschke AM, Leite D, Rocha J, Carvalho M, Sampaio F et al. Epicardial adipose tissue is an independent predictor of coronary atherosclerotic burden. Int J Cardiol 2012;158:26 –32. 10. Yerramasu A, Dey D, Venuraju S, Anand DV, Atwal S, Corder R et al. Increased volume of epicardial fat is an independent risk factor for accelerated progression of sub-clinical coronary atherosclerosis. Atherosclerosis 2012;220:223 – 30. 11. Alexopoulos N, McLean DS, Janik M, Arepalli CD, Stillman AE, Raggi P. Epicardial adipose tissue and coronary artery plaque characteristics. Atherosclerosis 2010; 210:150–4. 12. Rosito GA, Massaro JM, Hoffmann U, Ruberg FL, Mahabadi AA, Vasan RS et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: The Framingham heart study. Circulation 2008;117:605 –13. 13. Iwasaki K, Matsumoto T, Aono H, Furukawa H, Samukawa M. Relationship between epicardial fat measured by 64-multidetector computed tomography and coronary artery disease. Clin Cardiol 2011;34:166 –71. 14. Kim S, Chung J, Kwon B, Song S, Choi W. The associations of epicardial adipose tissue with coronary artery disease and coronary atherosclerosis. Int Heart J 2014;55: 197 –203. 15. Wang CP, Hsu HL, Hung WC, Yu TH, Chen YH, Chiu CA et al. Increased epicardial adipose tissue (EAT) volume in type 2 diabetes mellitus and association with metabolic syndrome and severity of coronary atherosclerosis. Clin Endocrinol (Oxf) 2009; 70:876–82. 16. Pierdomenico S, Pierdomenico A, Neri M, Cuccurullo F. Epicardial adipose tissue and metabolic syndrome in hypertensive patients with normal body weight and waist circumference. Am J Hypertens 2011;24:1245 –9. 17. Sengul C, Cevik C, Ozveren O, Oduncu V, Sunbul A, Akgun T et al. Echocardiographic epicardial fat thickness is associated with carotid intima-media thickness in patients with metabolic syndrome. Echocardiography 2011;28:853 –8. 18. Raitakari OT, Juonala M, Ro¨nnemaa T, Keltikangas-Ja¨rvinen L, Ra¨sa¨nen L, Pietika¨inen M et al. Cohort profile: the cardiovascular risk in Young Finns Study. Int J Epidemiol 2008;37:1220 –6. 19. Telama R, Yang X, Viikari J, Va¨lima¨ki I, Wanne O, Raitakari O. Physical activity from childhood to adulthood: a 21-year tracking study. Am J Prev Med 2005;28:267–73. 20. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: 412 –9. 21. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetic Med 1998;15:539 –53. 22. Grundy S, Brewer HB, Cleeman J, Smith S, Lenfant C. Definition of metabolic syndrome: report of the National heart, lung, and blood institute/American Heart Association conference on scientific issues related to definition. Circulation 2004;109: 433 –8. 23. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827 –32. 24. Hartiala O, Magnussen CG, Kajander S, Knuuti J, Ukkonen H, Saraste A et al. Adolescence risk factors are predictive of coronary artery calcification at middle age: the cardiovascular risk in Young Finns study. J Am Coll Cardiol. 2012;60:1364 –70. 25. Raitakari OT, Juonala M, Ka¨ho¨nen M, Taittonen L, Laitinen T, Ma¨ki-Torkko N et al. Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: the cardiovascular risk in Young Finns study. J Am Med Assoc 2003; 290:2277 –83. ¨ zben B, Yavuz D, Deyneli O, Aydın H, Tarcin O ¨ . et al. Epicardial adipose 26. Yazıcı D, O tissue thickness in type 1 diabetic patients. Endocrine 2011;40:250 –5.
Downloaded from by guest on May 18, 2015
Supplementary data
Page 7 of 8
Page 8 of 8 27. De Larochellie`re E, Coˆte´ J, Gilbert G, Bibeau K, Ross MK, Dion-Roy V et al. Visceral/ epicardial adiposity in nonobese and apparently healthy young adults: association with the cardiometabolic profile. Atherosclerosis 2014;234:23 –9. 28. Yorgun H, Canpolat U, Hazırolan T, Ates¸ AH, Sunman H, Dural M et al. Increased epicardial fat tissue is a marker of metabolic syndrome in adult patients. Int J Cardiol 2013;165:308 – 13. 29. Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA et al. Coronary risk factors measured in childhood and young adult life are associated with coronary artery calcification in young adults: The Muscatine study. J Am Coll Cardiol 1996;27: 277 –84. 30. Li S, Chen W, Srinivasan SR, Bond MG, Tang R, Urbina EM et al. Childhood cardiovascular risk factors and carotid vascular changes in adulthood: The Bogalusa Heart study. J Am Med Assoc 2003;290:2271 –6. 31. Schusterova I, Leenen FHH, Jurko A, Sabol F, Takacova J. Epicardial adipose tissue and cardiometabolic risk factors in overweight and obese children and adolescents. Pediatr Obes 2014;9:63–70. 32. Ferreira I, van de Laar RJ, Prins M, Twisk J, Stehouwer C. Carotid stiffness in young adults: a life-course analysis of its early determinants: The Amsterdam growth and health longitudinal study. Hypertension 2012;59:54– 61. 33. Bucci M, Joutsiniemi E, Saraste A, Kajander S, Ukkonen H, Saraste M et al. Intrapericardial, but not extrapericardial, fat is an independent predictor of impaired
O. Hartiala et al.
34.
35.
36.
37.
38.
39.
hyperemic coronary perfusion in coronary artery disease. Arterioscler Thromb Vasc Biol 2011;31:211 –8. Pracon R, Kruk M, Kepka C, Pregowski J, Opolski MP, Dzielinska Z et al. Epicardial adipose tissue radiodensity is independently related to coronary atherosclerosis. A multidetector computed tomography study. Circ J 2011;75:391 –7. Gorter PM, de Vos AM, van der Graaf Y, Stella PR, Doevendans PA, Meijs MF 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 –5. Xu Y, Cheng X, Hong K, Huang C, Wan L. How to interpret epicardial adipose tissue as a cause of coronary artery disease: a meta-analysis. Coron Artery Dis 2012;23: 227 –33. Brinkley TE, Hsu FC, Carr JJ, Hundley WG, Bluemke DA, Polak JF et al. Pericardial fat is associated with carotid stiffness in the Multi-Ethnic Study of Atherosclerosis. Nutr Metab Cardiovasc Dis 2011;21:332 –8. Soliman E, Ding J, Hsu F, Carr JJ, Polak J, Goff D. Association between carotid intimamedia thickness and pericardial fat in the Multi-Ethnic Study of Atherosclerosis (MESA). J Stroke Cerebrovasc Dis 2010;19:58–65. Dwyer T, Sun C, Magnussen C, Raitakari O, Schorck N, Venn A et al. Cohort profile: The International Childhood Cardiovascular Cohort (i3C) consortium. Int J Epidemiol 2013;42:86– 96.
Downloaded from by guest on May 18, 2015
