If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The West German Heart and Vascular Center, Department of Cardiology and Vascular Medicine, University Clinic Essen, University of Duisburg-Essen, Essen, Germany
The West German Heart and Vascular Center, Department of Cardiology and Vascular Medicine, University Clinic Essen, University of Duisburg-Essen, Essen, Germany
The West German Heart and Vascular Center, Department of Cardiology and Vascular Medicine, University Clinic Essen, University of Duisburg-Essen, Essen, Germany
Corresponding author. Department of Cardiology and Vascular Medicine, West German Heart and Vascular Center Essen, University of Duisburg-Essen, Hufelandstr. 55, 45147, Essen, Germany.
The West German Heart and Vascular Center, Department of Cardiology and Vascular Medicine, University Clinic Essen, University of Duisburg-Essen, Essen, Germany
CT-derived pericoronary fat volume and attenuation are inversely correlated.
•
Pericoronary fat volume is increased around coronary segments with culprit lesions.
•
CT-derived pericoronary fat attenuation is not different around culprit lesions.
•
Pericoronary fat volume but not attenuation may predict acute coronary syndromes.
Abstract
Background and aims
We aimed to determine the association of pericoronary adipose tissue (PCAT) volume and attenuation with culprit lesions in the underlying coronary segment in patients with acute myocardial infarction.
Methods
In patients with myocardial infarction, PCAT volume and attenuation surrounding the following segments were manually traced from non-contrast CT imaging: LM, proximal and mid-segment of LAD, RCA, and LCX. PCAT volume and attenuation surrounding culprit and non-culprit lesions were compared. Odds ratios (OR) and 95% confidence intervals (CI) were calculated per 1 standard deviation increase in PCAT volume/attenuation.
Results
We included 46 subjects (mean age 64.4 ± 16.4 years, 71% male) with acute myocardial infarction. PCAT volume around the right coronary artery was higher compared to left coronary segments, while PCAT attenuation decreased from proximal to distal segments. PCAT volume surrounding culprit lesions was higher compared to segments without culprit lesion (4.90 ± 3.07 ml vs. 2.33 ± 2.63 ml, p < 0.0001), whereas the attenuation was not different (−84.8 ± 9.4 HU vs. −84.2 ± 9.9 HU, p = 0.77). In univariate regression analysis, PCAT volume was significantly associated with the probability of presence of culprit lesions (OR [95% CI]: 3.10 [1.84–5.22], p < 0.0001). Associations remained stable upon adjustment for risk factors (3.34 [1.81–6.15], p < 0.0001). PCAT attenuation was not relevantly different around culprit lesions (unadjusted: 0.94 [0.63–1.40], p = 0.77, risk factor adjusted: 1.00 [0.61–1.64], p = 0.996).
Conclusions
In patients with acute myocardial infarction, PCAT volume is strongly and independently associated with culprit lesions in the underlying coronary segments, whereas PCAT attenuation does not relevantly differentiate surrounding coronary segments with and without culprit lesions.
Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: the Framingham Heart Study.
Association of epicardial fat with cardiovascular risk factors and incident myocardial infarction in the general population: the Heinz Nixdorf Recall Study.
Association of epicardial adipose tissue with progression of coronary artery calcification is more pronounced in the early phase of atherosclerosis: results from the Heinz Nixdorf recall study.
Persistent epicardial adipose tissue accumulation is associated with coronary plaque vulnerability and future acute coronary syndrome in non-obese subjects with coronary artery disease.
]. Recent data suggests that epicardial fat attenuation in addition to volume may be linked with coronary artery disease, as it may reflect increased vascularization as well as higher concentration of mitochondria and is correlated with local and systemic inflammatory markers [
]. An increase in the amount of epicardial adipose is linked to the release of pro-inflammatory markers and cytokines from the fat tissue, which is suggested to locally promote atherosclerosis development [
]. Besides overall EAT, pericoronary adipose tissue (PCAT) has gained further interest, as it may more locally impact plaque development in the underlying vasculature [
In the present manuscript we aimed to determine the association of pericoronary fat volume and attenuation with the presence of culprit lesions in the underlying coronary segments in patients with acute myocardial infarction undergoing cardiac CT imaging [
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.
We retrospectively included a total of 50 consecutive patients with myocardial infarction, who received cardiac CT as part of clinical workup during index hospitalization before or after revascularization between 2010 and 2015. Indications for CT imaging were based on the clinical course of the patients and included CTA for suspect CAD, evaluation of aortic aneurysms, or quantification of coronary plaque burden, while no CT examinations were performed specifically for research purposes. Median time between invasive coronary angiography and CT imaging was 7 days. Patients with prior bypass surgery or other open-heart surgery were excluded from our analysis. Further, patients with an incomplete clinical information from available hospital records, not allowing the evaluation of clinical course or assessment of risk factor profile, or insufficient CT image quality for further analysis have also been excluded (n = 4). For better comprehension of our study process, a study flow-chart can be found as Supplementary Fig. 1. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and the study was approved by the institutional review board.
2.2 CT-imaging and fat quantification
CT imaging was acquired using dual source computed tomography (Siemens Somatom Definition or Definition Flash, Siemens, Forchheim, Germany). Non-contrast enhanced CT-scan for the assessment of coronary artery calcification, prospectively triggered at 70% of the RR-interval with 3 mm slices thickness, was followed by CT angiography. Divergences in scan parameters are caused by patient characteristics and scanner type as the present data are based on a retrospective analysis. However, tube voltage was set at 120 kV for all non-contrast CT scans.
Epicardial as well as PCAT volume and attenuation were quantified offline from non-contrast CT images in all patients, using a dedicated workstation (Aquarius Workstation, TeraRecon, Foster City, CA, USA). The assessment of regional PCAT volume and attenuation based on manual delineation of regions of interest at of 7 coronary segment (total of 322 coronary segments) in axial planes (Left main = 5 mm proximal to bifurcation, proximal LAD = 5 mm distal from bifurcation, mid LAD = 5 mm distal from origin of the first diagonal branch, proximal LCX = 5 mm distal from bifurcation, mid/distal LCX = 5 mm distal from origin of the first obtuse marginal branch, proximal RCA = 5 mm distal from the ostium, mid RCA = in the middle of the descending part of the RCA). Within the region of interest, fat was defined as pixels within a window of −195 to −45 HU. PCAT volume was automatically calculated by the workstation after 3D reconstruction (Fig. 1). PCAT attenuation was calculated as mean Hounsfield units of all pixels of fat surrounding one coronary segment, which were counted as PCAT volume. Due to insufficient CT image quality, which did not allow further analysis of PCAT volume and attenuation, we excluded a total of 4 coronary segments (1 mid LAD, 1 mid LCX, 2 proximal LCX). In addition to PCAT volume and attenuation, also total epicardial adipose tissue (EAT) volume and attenuation was quantified as previously described [
]. In brief, manual tracing of the pericardial sac from the right pulmonary artery to the apex as outer border in axial planes was performed and pixels with −195 to −45 HU were accounted as fat. All epicardial and periocoronary fat measurements were performed by a highly experienced reader, blinded to the clinical presentation and risk factor profile of the patients. Intra-observer variability was performed in 20 randomly selected segments demonstrated excellent reproducibility (ICC: 0.95, 95% CI 0.90–0.97, p < 0.001).
Fig. 1Pericoronary fat was quantified from cardiac CT examination.
(A) The pericoronary adipose tissue was manually traced as region of interest. Within this region of interest, pixels between −195 and −45 Hounsfield Units were accounted as fat. (B) After 3-dimensional reconstruction, PCAT volume was calculated by summation of all pixels accounted as fat. PCAT attenuation was defined as mean Hounsfield Units of all fat pixels of the EAT volume.
Risk factors and clinical diagnoses of patients were obtained from all available hospital records. Systolic and diastolic blood pressure was assessed from admission records. Hypertension was defined as systolic blood pressure of >140 mmHg, diastolic blood pressure of >90 mmHg, or the intake of antihypertensive medication. Total-, HDL- and LDL-cholesterol were quantified using standardized enzymatic measures and were recorded from the same hospital stay. Diabetes was defined based on fasting glucose levels, HbA1c levels, and medication. Positive family history of premature coronary heart disease, lipid lowering medication and active smoking were obtained as documented by the treating physicians. The presence of myocardial infarction and culprit lesions was based on discharge letters and coronary angiography reports. If a culprit lesion was observed in invasive coronary angiography, patients were evaluated as type I myocardial infarction. If however not obstructive coronary artery disease was observed in coronary angiography, patients were evaluated as type-II myocardial infarction, according to the current ESC definition [
]. Most patients (n = 40) were studied after invasive coronary angiography, while only 6 CT exams were performed before. From the patients with type-I myocardial infarction, 21 patients received CT after revascularization, while 6 patients received CTA first.
2.4 Statistical analysis
Continuous variables are depicted as mean ± standard deviation, binary variables as frequency (%). Differences in baseline characteristics for patients classified as type-I and type-II myocardial infarction were compared using 2-sided t-test for continuous variables and Fischers-Exact Test for binary variables. PCAT volume was right skewed, whereas the attenuation was normally distributed. Correlation of PCAT volume and attenuation were compared using Spearmen correlation coefficient. Differences between PCAT volume in different coronary segments were compared using Mann-Whitney-U-test, whereas t-test was used for PCAT attenuation. Logistic regression analysis was used to determine the association of PCAT volume and attenuation with the presence of culprit lesions in the underlying coronary segment in unadjusted and risk factor adjusted models. Risk factor adjustment was performed as following: (1) unadjusted, (2) age, gender, and BMI adjusted, and (3) age, gender, BMI, systolic blood pressure, antihypertensive medication, LDL-and HDL-cholesterol, lipid-lowering medication, diabetes, active smoking, and positive family history for premature CVD events. For regression analyses, PCAT volume was log-transformed to the base of 10 to adjust for its skewed distribution. For regression analyses, effect sizes were calculated per each standard deviation of log transformed PCAT volume as well as pericardial fat attenuation. All analyses were performed using SPSS 25 (IBM, Armonk, USA, version 25 for windows). A p-value of <0.05 indicated statistical significance.
3. Results
We included 46 subjects (mean age 64.4 ± 16.4 years, 71% male) and 318 coronary segments in our analysis. Of those, 26 subjects had a culprit lesion detected on invasive coronary angiography, while 20 subjects were considered as type-II myocardial infarction according to current ESC guidelines. Mean EAT volume was 123.1 ± 68.1 ml and mean EAT attenuation was −87.1 ± 5.3 HU. Patients with type-I myocardial infarction had a trend towards lower HDL than subjects with type-II myocardial infarction. Further baseline characteristics were not relevantly different between subjects with type-I and type-II myocardial infarction. Details of the baseline characteristics for the overall cohort and stratified by type of myocardial infarction is depicted in Table 1. Mean PCAT volume and attenuation were 2.5 ± 2.8 ml and −84.2 ± 9.9 HU. PCAT volume and attenuation were modestly inversely correlated (r = −0.45, p < 0.0001, Fig. 2).
Table 1Patient characteristics for the overall cohort and stratified by type of myocardial infarction (type-I vs. type-II myocardial infarction).
3.1 Distribution of PCAT volume and attenuation over the coronary tree
Highest PCAT volume was observed in proximal RCA, followed by mid RCA (LM: 0.8 ± 0.6 ml, prox LAD: 1.7 ± 1.4 ml, mid LAD: 2.7 ± 2 ml, prox LCX: 1.3 ± 0.9 ml, mid LCX: 1.8 ± 1.9 ml, prox RCA: 5.8 ± 4 ml, mid RCA: 3.7 ± 3 ml, Fig. 3A). Overall, PCAT volume was significantly higher in RCA than in the other coronary arteries (Fig. 3A). In contrast, PCAT volume was not significantly different among LM, LAD, and LCX (detailed data not shown). For the LAD, PCAT volume was 3,42 ± 3.07 ml in segments with culprit lesions, whereas mean PCAT volume for all LAD segments without culprit lesions was 2.03 ± 1.52 ml (p-value for difference: 0.023). Likewise, we observed a trend towards higher PCAT volume surrounding culprit lesions in the RCA (5.68 ± 2.85 ml) compared to RCA segments without culprit lesions (4.57 ± 3.83 ml), however, not reaching statistical significance (p = 0.26). Lowest fat attenuation was detected in mid LAD and highest fat attenuation in LM.
Fig. 3Mean pericoronary fat volume (A) and attenuation (B), stratified by coronary segment.
The number of culprit lesions (CL) per segment is shown on the x-axis. p-values for PCAT volume comparison between RCA and LM, RCA and LAD and RCA and LCX are depicted top down.
Overall, PCAT attenuation was lower in distal segments compared to proximal segments (prox LAD: −82.3 ± 8.7 HU vs. mid LAD: −89.9 ± 8.4 HU, p < 0.0001, prox LCX: −81.5 ± 9.49 vs. mid LCX: 84.1 ± 10.57, p = 0.219, prox RCA: −86.3 ± 9.3 vs. mid RCA -87.0 ± 9.8, p = 0.745, Fig. 3B).
3.2 Pericoronary fat and culprit lesion
PCAT volume surrounding coronary segments with culprit lesions was higher compared to coronary segments without culprit lesions (4.9 ± 3.1 vs. 2.3 ± 2.6 ml, p < 0.0001, Fig. 4A). In contrast, PCAT attenuation was not significantly different around segments with and without culprit lesions (−84.8 ± 9.4 vs −84.2 ± 9.9HU, p = 0.77, Fig. 4B). Comparing the 21 patients with CTA after revascularization with those with CTA first, PCAT attenuation surrounding the culprit lesions were not significantly different (−83.83 HU vs. −88.74 HU, for patients with CT after vs. before revascularization, p = 0.3). Also, pericoronary fat volume did not significantly differ, stratifying patients with type-I myocardial infarction by time of CTA with respect to coronary revascularization therapy (5.29 ml vs. 3.25 ml, for patients with CT after vs. before revascularization, p = 0.2). In univariate regression analysis, increase in PCAT volume by one standard deviation was associated with 3-fold increased odds for presence of culprit lesions in the underlying coronary segment (Table 2). Stepwise adjustment for age, gender, and BMI and ancillary adjusting for traditional risk factors did not relevantly influence the association of PCAT volume with the presence of culprit lesions. Additionally, controlling for EAT volume or coronary segment further did not relevantly change the results. In contrast, in univariate and risk factor adjusted regression analysis no relevant link between PCAT attenuation and the presence of culprit lesion could be demonstrated (Table 2).
Fig. 4Pericoronary fat volume (A) and attenuation (B) surrounding coronary segments with and without culprit lesions.
Table 2Logistic regression analysis for the association of pericoronary fat volume and attenuation with the presence of culprit lesion in the underlying coronary segment.
MV adjustment includes age, gender, BMI, systolic blood pressure, antihypertensive medication, LDL- and HDL-cholesterol, lipid-lowering medication, diabetes, active smoking, and positive family history for premature CVD events.
MV adjustment includes age, gender, BMI, systolic blood pressure, antihypertensive medication, LDL- and HDL-cholesterol, lipid-lowering medication, diabetes, active smoking, and positive family history for premature CVD events.
EAT volume was added to the model for the association of pericoronary fat volume, whereas EAT attenuation was added to the model for the association of pericoronary fat attenuation.
MV adjustment includes age, gender, BMI, systolic blood pressure, antihypertensive medication, LDL- and HDL-cholesterol, lipid-lowering medication, diabetes, active smoking, and positive family history for premature CVD events.
+ coronary segment
3.12 (1.51–6.45)
0.002
1.16 (0.69–1.97)
0.58
Pericoronary fat volume was log-transformed to adjust for its skewed distribution. OR (95%CI) are depicted per 1 standard deviation change in pericoronary fat volume and attenuation.
a MV adjustment includes age, gender, BMI, systolic blood pressure, antihypertensive medication, LDL- and HDL-cholesterol, lipid-lowering medication, diabetes, active smoking, and positive family history for premature CVD events.
b EAT volume was added to the model for the association of pericoronary fat volume, whereas EAT attenuation was added to the model for the association of pericoronary fat attenuation.
The present manuscript describes (1) the distribution and differences of PCAT volume and attenuation over the coronary tree, (2) a strong and independent association of PCAT volume with the presence of culprit lesions in the underlying coronary segment while (3) PCAT attenuation did not relevantly differentiate surrounding coronary segments with and without culprit lesions. Together with the existing literature, our results suggest that local measures of epicardial fat volume - in contrast to fat attenuation - as derived from non-contrast enhanced cardiac CT may play an emerging role in coronary atherosclerosis, leading to myocardial infarction.
Recent data suggests an important role of PCAT in the pathogenesis of coronary atherosclerosis [
Adipocyte-derived plasma protein adiponectin acts as a platelet-derived growth factor-BB-binding protein and regulates growth factor-induced common postreceptor signal in vascular smooth muscle cell.
Adipocyte-derived plasma protein adiponectin acts as a platelet-derived growth factor-BB-binding protein and regulates growth factor-induced common postreceptor signal in vascular smooth muscle cell.
]. The activity of NF kappa B enhances with the decrease of adiponectin. This tends to result in an increase in TNF alpha which leads to a local increase of inflammation [
]. The mismatch of anti- and pro-inflammatory mediators as well as cytokines from pericardial fat is assumed to have an influence on the underlying coronary segment [
]. Increasing body weight tends to result in accumulation of adipose tissue surrounding the heart and coronary arteries and lipid accumulation within cardiomyocytes. Derangements in the function of PCAT are leading to an infiltration of macrophages and upregulation of inflammatory adipokines [
]. In comparison to subcutaneous and visceral regions, perivascular adipocytes are showing a lower expression of adipocytic differentiation related genes, which suggests that perivascular adipocytes exist in a more primitive adipocytic state [
]. These findings are supported by the fact that the inflammatory activity of PCAT is higher in patients with coronary artery disease (CAD) than in non-CAD controls and that it is independently associated with coronary stenosis [
PET/CT evaluation of (18)F-FDG uptake in pericoronary adipose tissue in patients with stable coronary artery disease: independent predictor of atherosclerotic lesions' formation?.
The decrease of PCAT attenuation from proximal to distal segments has been previously observed and is confirmed by our results, suggesting that variations in CT density of PCAT are influenced by partial volume effects and image interpolation rather than just metabolic activity or tissue composition differences [
The previously reported association between PCAT volume and atherosclerosis goes in hand with our result that PCAT volume surrounding coronary segments with culprit lesions was significantly higher than in segments without culprit lesions [
]. Our results support the hypothesis that perivascular fat depots may have a local role in atherosclerosis development, leading to subsequent myocardial infarction. In addition, recent studies detected a significantly lower PCAT attenuation for normal versus atherosclerotic coronary segments [
], in contrast to our results, where no correlation between PCAT attenuation and culprit lesions was described. While according to larger studies, coronary atherosclerosis most frequently occurs in the proximal LAD, we observed a higher frequency of culprit lesions in the RCA. While this effect may have occurred by chance, given the small absolute number of culprit lesions in our study, also the difference in clinical symptom of patients with myocardial infraction and culprit lesions in the RCA, may have led to a different distribution in our study. However, further studies comparing PCAT volume and attenuation with local and systemic metabolic and inflammatory profiles are needed to confirm these results.
4.1 Limitations
Our analysis is limited by cohort size, which has restricted our statistical power. This was caused by the low number of patients with acute myocardial infarction undergoing cardiac CT imaging as part of clinical workup. Definition of coronary segments was precisely performed based on side branches and anatomical definitions to allow for best correlation of segments based on coronary angiography and cardiac CT, however, we were unable to detect the location of the culprit lesions within one segment by CT images, which may have biased our results towards the null. Previous stent implantation for revascularization of culprit lesions in patients with type-I myocardial infarction may have impacted PCAT measures due to beam hardening and photon scatter caused by the dense stent material. In addition, remodeling of the coronary artery may have influenced the PCAT. While we observed no statistically significant difference in PCAT volume and attenuation in patients with prior stenting compared to patients that received cardiac CT first, further studies are needed to assess the influence of stent-implantation on PCAT volume and attenuation. In addition to that, coronary anatomy shows differences in run and occurrence, which also may have higher influence related to our sample size. Further studies in larger cohorts need to confirm our results. Lastly, the pathophysiological relation between PCAT volume and myocardial infarction can only be hypothesized and also needs to be evaluated in proper studies.
4.2 Conclusion
In patients with acute myocardial infarction, PCAT volume is independently associated with culprit lesions in the underlying coronary segments, whereas PCAT attenuation does not relevantly differentiate surrounding coronary segments with and without culprit lesions. Our results suggest that local measures of epicardial fat volume as derived from non-contrast enhanced cardiac CT are a stronger marker of risk for myocardial infarction as compared to fat attenuation.
Conflicts of interest
The authors declared they do not have anything to disclose regarding conflict of interest with respect to this manuscript.
Appendix A. Supplementary data
The following is the supplementary data related to this article:
Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: the Framingham Heart Study.
Association of epicardial fat with cardiovascular risk factors and incident myocardial infarction in the general population: the Heinz Nixdorf Recall Study.
Association of epicardial adipose tissue with progression of coronary artery calcification is more pronounced in the early phase of atherosclerosis: results from the Heinz Nixdorf recall study.
Persistent epicardial adipose tissue accumulation is associated with coronary plaque vulnerability and future acute coronary syndrome in non-obese subjects with coronary artery disease.
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.
Adipocyte-derived plasma protein adiponectin acts as a platelet-derived growth factor-BB-binding protein and regulates growth factor-induced common postreceptor signal in vascular smooth muscle cell.
PET/CT evaluation of (18)F-FDG uptake in pericoronary adipose tissue in patients with stable coronary artery disease: independent predictor of atherosclerotic lesions' formation?.
30278-8&id=gr1.jpg){kind=link}
30278-8&id=gr2.jpg){kind=link}
30278-8&id=gr3.jpg){kind=link}
30278-8&id=gr4.jpg){kind=link}
30278-8&id=mmcfigs1.jpg){kind=link}