RPS 1403a - Epicardial fat, coronary calcifications and cardiovascular risk stratification: what's new in cardiac imaging?

Purpose:

Anthropomorphic phantoms containing calcifications are often used in coronary calcium scoring (CCS) studies. Manually quantifying CCS is highly time consuming. Accordingly, we aimed to develop and validate a fully automatic method for CCS in a commercially available phantom using two commonly used computer-programming languages.

Methods and materials:

An anthropomorphic thorax phantom with cardiac calcification insert containing 9 cylindrical calcifications was used. An extension ring was used to simulate larger patients. Scans were acquired on an electron-beam tomography system and 5 state-of-the-art CT systems from 4 major vendors using routine protocols with 130 and 120 kVp, respectively, 3 mm slice thickness and filtered back-projection. To simulate interscan variability, each acquisition was repeated 5 times after repositioning the phantom. The robustness of CCS quantification-algorithms was assessed by evaluating a variety of settings on one CT-system. CCS was determined with vendor-specific software and quantified as an Agatston score. We developed 2 fully automated algorithms in 2 programming languages: MATLAB and Python. Besides CCS, image quality analysis functions were implemented into the algorithm (standard-deviations of a water-equivalent region-of-interest and noise power spectrum). To assess CCS accuracy of the algorithms, CCS was compared with their values obtained with the vendor-specific software using Bland-Altman and intraclass correlation analyses.

Results:

Both algorithms achieved high agreements with vendor-specific software for Agatston scores (mean ± 95% confidence-interval for MATLAB and Python, respectively: -0.3% ± 4.0% and -0.1% ± 3.8%; ICCs were excellent for MATLAB and Python: 1.00 (1.00-1.00) for both). Output of the algorithms from one scan was given in <5 seconds.

Conclusion:

We developed two algorithms to fully automatically quantify calcifications and evaluate image quality in a commercially available phantom with excellent agreement compared to vendor-specific software.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

Coffee is one of the most popular beverages in the world and even small health effects may have a large impact on public health. Therefore, we sought to determine the relationship between coffee consumption and cerebral small vessel disease (SVD), cardiac function, or fat depots, using a whole-body magnetic resonance imaging (MRI) protocol.

Methods and materials:

MRI markers of SVD, cardiac function, and fat depots were obtained in a population-based study cohort without cardiovascular disease. A blended approach was used to estimate habitual coffee consumption. Associations of coffee intake with total brain volume, white matter lesions, cerebral microbleeds, total (TAT) and visceral adipose tissue (VAT), hepatic proton density fat fraction (PDFF_hepatic), and cardiac function were assessed by linear regression.

Results:

In a sample of 132 women (mean age 56.3±8.8years) and 168 men (mean age 56.2±9.3years), habitual coffee consumption was positively associated with MR-based cardiac function parameters including late diastolic filling rate, stroke volume (p<0.01 each), and ejection fraction (p<0.05) when adjusting for age, sex, smoking, hypertension, diabetes, LDL, triglycerides, cholesterol, and alcohol consumption. Coffee consumption was negatively associated with VAT when adjusting for age, sex, smoking, hypertension, diabetes, LDL, triglycerides, and cholesterol (p<0.05), but not after additional adjustment for alcohol consumption. There was no significant association between coffee consumption and early diastolic filling rate, end-diastolic and end-systolic volume, TAT, PDFF_hepatic, or SVD parameters.

Conclusion:

We found a significant positive and independent association between coffee intake and MRI-based systolic and diastolic cardiac function. Coffee consumption also seems to be inversely associated with VAT, but not independent of alcohol consumption. A significant association between coffee consumption and SVD was not observed.

Limitations:

No difference between coffee preparation method or type of coffee was made.

Ethics committee approval

Ethic committee approved.

Funding:

No funding was received for this work.

Purpose:

Essential hypertension (EH) is identified as one of the most important risk factors for cardiovascular morbidity in industrialised countries. Epicardial adipose tissue (EAT) is considered to be an important factor in the development of cardiovascular diseases. The pathophysiological mechanisms of EAT remain unclear to this day. Diverging results are presented regarding the association between EAT and EH. Moreover, there has been insufficient research on EAT association with different stages of EH. Our aim was to evaluate the association between size of EAT depots and the risk of EH, taking into account its stage.

Methods and materials:

There were 258 non-obese adult patients investigated with different cardiovascular diseases, 157 of whom had EH and 101 who did not. Patients with EH were classified to 3 groups according to the stage of hypertension. Volume and thickness of EAT depots on ventricular free walls (6 locations) and grooves (5 locations) were measured using cardiac magnetic resonance tomography images, and then compared between groups. For the prediction of EH, a regression model was constructed.

Results:

Overall, EAT thickness (in all locations) and volume were higher among hypertensive patients (p<0.001). Mean EAT thickness and volume were larger in the stage 1 hypertension group than among normotensive patients (p<0.05). Furthermore, the volume of EAT was lower in the stage 1 than in the stage 2 hypertension group (p<0.05). However, the EAT depots did not differ between severe and the estage 2 hypertension group. EAT volume, patients’ age, body mass index, and dyslipidaemia status were independent with EH associated factors in a predictive model.

Conclusion:

Significant differences in EAT accumulation between hypertensive and normotensive patients show that EAT measurements may serve as a risk indicator for EH as well as a predictor of hypertension severity.

Limitations:

n/a

Ethics committee approval

Ethics committee approval provided.

Funding:

No funding was received for this work.

Purpose:

To compare the prognostic value of CAD-RADS classification with the calcium score relative risk.

Methods and materials:

With a series population from a single-centre study of 1,918 pts (1,167 M, 751 F, mean age 63yrs) who underwent CA-MDCT from April 2008 to September 2019, we reconstructed and scored each exam, classifying the detected lesions and plaques according to the CAD-RADS system. 1,850/1,918 pts underwent additional pre-contrast coronary artery calcium protocol (Ca score). We determined the calcium burden of atherosclerotic plaques in these exams, according to Agatston method, and the relative risk of coronary events adjusted for patients age and gender. A correlation between the Agatston score and CAD-RADS was obtained using a Spearman test.

Results:

Patients with significant stenosis (CAD-RADS 3) showed an extensive degree of coronary artery calcification (average value of Ca score: 1,052), with a discrete to moderate relative risk (76%).

Patients with an occluded vessel (CAD-RADS 5) had a higher degree of coronary artery calcification (average value of Ca score: 1,929) and a moderate to high relative risk (87%).

Good correlation was recorded between calcium score and CAD-RADS, in case of Ca score 0 and CAD-RADS 0-1 and CAD-RADS 4 and 5 and Ca score severe (Rho 0.9 and 0.86 respectively P< 0.001), significant but a lower correlation in case of a moderate Ca score and CAD-RADS 3 (rho 0-67 p<0.001).

Conclusion:

CAD-RADS better describes the overall plaque extension and degree of significance than a Ca score.

Limitations:

Study population is not homogeneous for risk factors.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

A wide left coronary artery bifurcation angle has been reported to be associated with the severity of coronary artery disease, but it may not reflect complex three-dimensional vessel morphology. This study aimed to utilise statistical shape analysis to categorise and quantify the shape patterns of left coronary artery bifurcation, and correlate the shape changes to coronary calcium scores.

Methods and materials:

Coronary CT angiography obtained from 406 subjects with intermediate cardiovascular risks were retrieved from a cross-modality multicentre cardiovascular imaging database. The centrelines of the left coronary artery adjacent to bifurcation was extracted and manually corrected. Statistical shape analysis of these centerlines was performed and the most significant principal components were determined by a scree test. The correlations between shape quantification results and coronary calcium scores were examined with the Pearson correlation coefficient. Patients were stratified into a high-calcium score group (Agatston score>100) and a low calcium score group.

Results:

The scree test yielded 5 most significant shape patterns, explaining 76% of changes. The derived principal components were used on 178 subjects with available clinical information. There was a weak correlation: the calcium scores of LAD and LCX increased with wider bifurcating angle and shorter LM length (PC#1) (Pearson correlation coefficient: 0.18 for LM and 0.17 for LCx). The calcium score was independent of a simple acute turn at proximal LCX (PC#3). The differences in shape patterns were not significant between the high-calcium score group and the low-calcium score group.

Conclusion:

Statistical shape analysis provided comprehensive shape change pattern observation. The calcium score of the left coronary artery was weakly correlated with LM length and bifurcating angle, and was independent of simple acute proximal turn of LCX.

Limitations:

Shape analysis is subjected to intepretation.

Ethics committee approval

Aproved by NTUH research ethic committee.

Funding:

No funding was received for this work.

Purpose:

To determine alcohol-associated cardiac remodeling and sex-specific differences by cardiac magnetic resonance imaging (CMR) in a low-risk population-based sample.

Methods and materials:

In the Study of Health in Pomerania, 1,134 voluntaries (median age 51 years, 55.7% men) underwent gadolinium-enhanced CMR between 2008 and 2012. In a cross-sectional approach, left ventricular (LV) and left atrial (LA) structure and function were related to self-reported alcohol intake over the past month in multivariable linear regression models adjusted for a priori specified clinical and socio-demographic confounders.

Results:

75% of the investigated population consumed less than recently recommended limits of 100 g/week. Only in men was alcohol intake associated with a linearly greater LV mass (β_fullmodel=0.140(0.049;0.23) d (95%-CI); p=0.003), cardiac output (β_fullmodel=0.0093(0.002;0.02) l min^-1g^-1 d; p=0.008), and a lower systemic vascular resistance (β_fullmodel=-0.020(-0.04;-0.002) mmhg l^-1min g^-1d; p=0.030). In contrast, alcohol intake was associated with greater LV wall-stress (β_fullmodel=1.21(0.3;2.1) mmHg g^-1d; p=0.008) in women only. For neither men nor women was alcohol intake related to LV end-diastolic volume, LA maximal volume, LV ejection fraction, or late gadolinium enhancement.

Conclusion:

Even in a low-risk drinking population, alcohol intake for men is dose-dependently associated with peripheral vasodilation and a greater LV mass, independent from hypertension. In contrast, despite unchanged geometry, women’s hearts are exposed to greater LV wall-stress. Cardiac structure and function is altered below currently recommended risk-thresholds for alcohol intake.

Limitations:

The character of an observational study limits the ability to confirm causal relationships. Regression-dilution bias may arise from using alcohol consumption during the past month as a proxy for usual/cumulative consumption.

Ethics committee approval

All individuals gave written informed consent and the study was approved by Ethics Committee of the University of Greifswald.

Funding:

This study is supported by the German Federal State of Mecklenburg-West-Pomerania.

Purpose:

To assess the relationship between the exposure to environmental tobacco smoke (ETS) estimated using the SHSES scale and epicardial adipose tissue thickness (EATT) in patients with hypertension.

Methods and materials:

The study included 96 people with hypertension (69.32±9.54 years). All patients were assessed for exposure to ETS based on the SHSES scale and 128-slice coronary computed tomography (coroCT) was performed. Based on the SHSES scoring criterion, subgroups of subjects were identified: A) without ETS exposure (0 SHSES points, n=50), B) low ETS exposure (1-3 SHSES points, n=11), C) moderate ETS exposure (4-7 SHSES points, n=20), and D) high ETS exposure (8-11 SHSES points, n=17). Based on the images obtained in coroCT, EATT was measured. The maximum measurement taken along the anterior free wall of the right ventricle was taken as EATT.

Results:

In the studied group, EATT was 5.75±1.85 mm. EATT was statistically significantly higher in subgroup D than in subgroups A and B (A: 5.28±1.64 mm, B: 5.04±2.64 mm, D: 7.04±2.64 mm, p_A-D and p_B-D<0.05). There was a positive linear correlation between exposure to ETS expressed by the number of points on the SHSES scale and EATT (r=0.44, p<0.05). In the regression analysis, it has been documented that higher SHSES scores, higher BMI, and higher systolic and diastolic blood pressure values are independent risk factors for higher EATT, while the use of ACE inhibitors and β-blockers are independent protection factors against higher EATT.

Conclusion:

In patients with hypertension, there is a positive relationship between exposure to environmental tobacco smoke estimated using the SHSES scale and epicardial adipose tissue thickness.

Limitations:

A single-centre study.

Ethics committee approval

Local Bioethical Committee.

Funding:

No funding was received for this work.

Purpose:

The purpose of this study was to assess the effects of MetS and ectopic fat deposition on LV structure changes in subclinical type 2 diabetes mellitus (T2DM).

Methods and materials:

53 T2DM patients and 20 healthy controls were prospectively enrolled. All of them performed CMR (3.0-T, Siemens Medical Solutions, Germany). We excluded patients with known cardiovascular disease and cardiovascular-related symptoms. We divided patients into MetS group and non-MetS group. Single-voxel ¹H-MRSS was performed to detect the myocardial triglyceride content using the post-processing software (jMRUI, version 6.0).

Results:

Myocardial TG content was significantly higher in T2DM patients compared with controls (1.54±0.63 vs. 0.61±0.22 %, p < 0.001). GLS was lower in the MetS group than the non-MetS group and normal controls (all p<0.001). The MetS group had a significantly reduced upslope value (2.10±1.19 vs. 2.93±0.78, p<0.001) and an increased TTM value (36.09±14.57 vs. 24.77±11.01, p<0.001) than the controls. A correlation existed between the myocardium TG content and GLS (r = 0.339, p <0.05) and TTM (r = 0.415, p <0.05). ROC analysis showed that 1.56 was the optimal cutoff values of myocardium TG content that predicted the risk of LV deformation (AUC=0.71).

Conclusion:

T2DM patients with MetS will cause more serious damage to the myocardium and myocardial TG content is negatively correlated with LVGLS and perfusion.

Limitations:

The sample size was too small. A single-centre study, therefore an inclusion bias.

Ethics committee approval

/a

Funding:

This work was supported by the 1·3·5 project for disciplines of excellence, West
China Hospital, Sichuan University (ZYGD18013) and the National Natural Science Foundation of China (81471721, 8147172281641169, 81771887 and 81771897).

Purpose:

Improvements of CT technology over the last decade allow for improved acquisition and reconstruction methods of coronary artery calcium (CAC) quantification. Currently, CAC-CT requires a relatively high radiation dose and inter- and intra-scan reproducibility is suboptimal. This study aimed to optimise the standard for CAC-quantification using low-kVp protocols with improved reproducibility on state-of-the-art CT systems from two major vendors.

Methods and materials:

An anthropomorphic thoracic phantom with cardiac calcification insert containing 9 cylindrical calcifications and 2 additional extension rings was used. Images were acquired with new-generation CT systems from two vendors using routine protocols and a variety of tube voltages (120-80 kVp), tube currents (≤75% dose reduction), slice thicknesses (3/2.5 and 1/1.25 mm), and reconstruction techniques (filtered back projection and iterative reconstruction (IR)). Every protocol was scanned 5 times after repositioning of the phantom to asses reproducibility. Calcifications were quantified as Agatston and volume scores. Routine scores were compared to other scores using Friedman’s tests and post-hoc Wilcoxon signed-rank tests with Bonferroni correction.

Results:

Reducing tube potential to 100 kVp, radiation dose with 25%, and slice thickness to 1/1.25 mm, respectively, combined with high IR-levels, resulted in similar total Agatston scores (P>0.05). Total volume slightly reduced to values closer to the physical volume (P>0.05). Moreover, intrascanner CAC-score variability decreased with 1-9%. Analysis per calcification showed that variability and CAC-scores decreased in higher density calcifications. In lower density calcifications, CAC-scores increased due to increasing HU-values. On average, 6/9 lesions were detected with traditional slice thickness, while 7/9 were detected with thin-slice images.

Conclusion:

This multivendor study showed protocol-optimisation resulted in reduced radiation doses without affecting total Agatston scores, volume scores closer to the physical volume, and improved reproducibility.

Limitations:

/a

Ethics committee approval

/a

Funding:

No funding was received for this work.