Influence of reconstruction kernels on the accuracy of CT-derived fractional flow reserve.

OBJECTIVES: We evaluated the influence of image reconstruction kernels on the diagnostic accuracy of CT-derived fractional flow reserve (FFRCT) compared to invasive FFR in patients with coronary artery disease. METHODS: Sixty-nine patients, in whom coronary CT angiography was performed and who were further referred for invasive coronary angiography with FFR measurement via pressure wire, were retrospectively included. CT data sets were acquired using a third-generation dual-source CT system and rendered with medium smooth (Bv40) and sharp (Bv49) reconstruction kernels. FFRCT was calculated on-site using prototype software. Coronary stenoses with invasive FFR ≤ 0.80 were classified as significant. Agreement between FFRCT and invasive FFR was determined for both reconstruction kernels. RESULTS: One hundred analyzed vessels in 69 patients were included. Twenty-five vessels were significantly stenosed according to invasive FFR. Using a sharp reconstruction kernel for FFRCT resulted in a significantly higher correlation with invasive FFR (r = 0.74, p < 0.01 vs. r = 0.58, p < 0.01; p = 0.04) and a higher AUC in ROC curve analysis to correctly identify/exclude significant stenosis (AUC = 0.92 vs. AUC = 0.82 for sharp vs. medium smooth kernel, respectively, p = 0.02). A FFRCT value of ≤ 0.8 using a sharp reconstruction kernel showed a sensitivity of 88% and a specificity of 92% for detecting ischemia-causing lesions, resulting in a diagnostic accuracy of 91%. The medium smooth reconstruction kernel performed worse (sensitivity 60%, specificity 89%, accuracy 82%). CONCLUSION: Compared to invasively measured FFR, FFRCT using a sharp image reconstruction kernel shows higher diagnostic accuracy for detecting lesions causing ischemia, potentially altering decision-making in a clinical setting. KEY POINTS: • Image reconstruction parameters influence the diagnostic accuracy of simulated fractional flow reserve derived from coronary computed tomography angiography. • Using a sharp kernel image reconstruction algorithm delivers higher diagnostic accuracy compared to medium smooth kernel image reconstruction (gold standard invasive fractional flow reserve).

Carotid plaque composition by CT angiography in asymptomatic subjects: a head-to-head comparison to ultrasound.

OBJECTIVES: To describe carotid plaque composition by computed tomography angiography (CTA) in asymptomatic subjects and to compare this to carotid plaque assessment by ultrasound, coronary plaques by coronary CTA, and inflammatory biomarkers in plasma. METHODS: Middle-aged asymptomatic men, n = 43, without known cardiovascular disease and diabetes were included. Plaques in coronary and carotid arteries were evaluated using CTA. Total plaque volumes and plaque composition were assessed by a validated plaque analysis software. The 60% centile cut point was used to divide the population into low or high carotid total plaque volumes. The occurrence of carotid plaques and intima-media thickness (IMT) was estimated by ultrasound. RESULTS: Carotid plaque by ultrasound was undiagnosed in 13 of 28 participants (46%) compared to CTA. Participants having carotid plaques by ultrasound had significantly higher absolute volumes of all CTA-defined carotid plaque subtypes and a higher fraction of calcified plaque. A high carotid total plaque volume was independently associated with age (adjusted odds ratio (OR) 1.41 [95% confidence interval (CI) 1.14-1.74], p = 0.001), IMT (adjusted OR 2.26 [95% CI 1.10-4.65], p = 0.03), and D-dimer (adjusted OR 8.86 [95% CI 1.26-62.37], p = 0.03). All coronary plaque features were significantly higher in participants with a high carotid total plaque volume. CONCLUSION: The occurrence of carotid plaques in asymptomatic individuals is underestimated by ultrasound compared to plaque assessment by CTA. Carotid plaque composition by CTA is different in individuals with and without carotid plaques by ultrasound. KEY POINTS: • The occurrence of carotid plaques by ultrasound was underestimated in 46% of participants who had plaques by carotid CTA. • Participants with carotid plaques by ultrasound had higher volumes of all plaque subtypes and a higher calcified plaque component as determined by carotid CTA compared to participants without carotid plaques by ultrasound. • A high carotid total plaque volume was independently associated with age, intima-media thickness, and D-dimer.

The Munich-Transarterial Chemoembolisation Score Holds Superior Prognostic Capacities Compared to TACE-Tailored Modifications of 9 Established Staging Systems for Hepatocellular Carcinoma.

BACKGROUND/AIMS: The recently proposed Munich-transarterial chemoembolisation-score (M-TACE) was tailored to suit hepatocellular Carcinoma (HCC) patients evaluated for TACE. M-TACE outperformed the established HCC-staging-systems and successfully passed external validation. Modifications of staging-systems through the rearrangement of stages or by adding prognostic factors are methods of improving prognostic power. M-TACEs performance compared to scores modified this way should be tested. METHODS: Seven well-known HCC staging-systems (including Cancer of the Liver Italian Program-score [CLIP] and Barcelona Clinic liver cancer [BCLC]) and 2 TACE-specific scores (Selection for Transarterial Chemoembolisation Treatment [STATE] and Hepatoma Arterial embolisation Prognostic [HAP]) were rearranged in a cohort of 186 TACE-patients through score-point-analysis and subsequent linking of non-significant adjacent score-points. Additionally, a new score was constructed by combining the top established staging-system in TACE patients (CLIP-TACE) and the prognostic parameter with the highest hazard ratio for death in the TACE-cohort [C-reactive protein (CRP)]. Additionally, the TACE-tailored-scores were applied to an external TACE-cohort (n = 71). -Results: Rearrangement resulted in optimal stratification and monotonicity. CLIP-TACE demonstrated the best prognostic capability of all rearranged scores (c-index 0.668, AIC 1294) and the addition of CRP yielded further prognostic improvement (c-index 0.680, AIC 1289). However, superiority over M-TACE could not be achieved by any of the new scores in the internal and external cohort. CONCLUSION: M-TACE outperforms TACE-tailored modifications of all relevant HCC-staging-systems. Prospective validation of M-TACE to promote its role as the preferred staging-system for TACE-patients is therefore justified.

Integrated prediction of lesion-specific ischaemia from quantitative coronary CT angiography using machine learning: a multicentre study.

OBJECTIVES: We aimed to investigate if lesion-specific ischaemia by invasive fractional flow reserve (FFR) can be predicted by an integrated machine learning (ML) ischaemia risk score from quantitative plaque measures from coronary computed tomography angiography (CTA). METHODS: In a multicentre trial of 254 patients, CTA and invasive coronary angiography were performed, with FFR in 484 vessels. CTA data sets were analysed by semi-automated software to quantify stenosis and non-calcified (NCP), low-density NCP (LD-NCP, < 30 HU), calcified and total plaque volumes, contrast density difference (CDD, maximum difference in luminal attenuation per unit area) and plaque length. ML integration included automated feature selection and model building from quantitative CTA with a boosted ensemble algorithm, and tenfold stratified cross-validation. RESULTS: Eighty patients had ischaemia by FFR (FFR ≤ 0.80) in 100 vessels. Information gain for predicting ischaemia was highest for CDD (0.172), followed by LD-NCP (0.125), NCP (0.097), and total plaque volumes (0.092). ML exhibited higher area-under-the-curve (0.84) than individual CTA measures, including stenosis (0.76), LD-NCP volume (0.77), total plaque volume (0.74) and pre-test likelihood of coronary artery disease (CAD) (0.63); p < 0.006. CONCLUSIONS: Integrated ML ischaemia risk score improved the prediction of lesion-specific ischaemia by invasive FFR, over stenosis, plaque measures and pre-test likelihood of CAD. KEY POINTS: • Integrated ischaemia risk score improved prediction of ischaemia over quantitative plaque measures • Integrated ischaemia risk score showed higher prediction of ischaemia than standard approach • Contrast density difference had the highest information gain to identify lesion-specific ischaemia.

Diagnostic accuracy of low and high tube voltage coronary CT angiography using an X-ray tube potential-tailored contrast medium injection protocol.

OBJECTIVES: To compare the diagnostic accuracy between low-kilovolt peak (kVp) (≤ 100) and high-kVp (> 100) third-generation dual-source coronary CT angiography (CCTA) using a kVp-tailored contrast media injection protocol. METHODS: One hundred twenty patients (mean age = 62.6 years, BMI = 29.0 kg/m2) who underwent catheter angiography and CCTA with automated kVp selection were separated into two cohorts (each n = 60, mean kVp = 84 and 117). Contrast media dose was tailored to the kVp level: 70 = 40 ml, 80 = 50 ml, 90 = 60 ml, 100 = 70 ml, 110 = 80 ml, and 120 = 90 ml. Contrast-to-noise ratio (CNR) was measured. Two observers evaluated image quality and the presence of significant coronary stenosis (> 50% luminal narrowing). RESULTS: Diagnostic accuracy (sensitivity/specificity) with ≤ 100 vs. > 100 kVp CCTA was comparable: per patient = 93.9/92.6% vs. 90.9/92.6%, per vessel = 91.5/97.8% vs. 94.0/96.8%, and per segment = 90.0/96.7% vs. 90.7/95.2% (all P > 0.64). CNR was similar (P > 0.18) in the low-kVp vs. high-kVp group (12.0 vs. 11.1), as ws subjective image quality (P = 0.38). Contrast media requirements were reduced by 38.1% in the low- vs. high-kVp cohort (53.6 vs. 86.6 ml, P < 0.001) and radiation dose by 59.6% (4.3 vs. 10.6 mSv, P < 0.001). CONCLUSIONS: Automated tube voltage selection with a tailored contrast media injection protocol allows CCTA to be performed at ≤ 100 kVp with substantial dose reductions and equivalent diagnostic accuracy for coronary stenosis detection compared to acquisitions at > 100 kVp. KEY POINTS: • Low-kVp coronary CT angiography (CCTA) enables reduced contrast and radiation dose. • Diagnostic accuracy is comparable between ≤ 100 and > 100 kVp CCTA. • Image quality is similar for low- and high-kVp CCTA. • Low-kVp image acquisition is facilitated by automated tube voltage selection. • Tailoring contrast injection protocols to the automatically selected kVp-level is feasible.

Computed Tomography-derived Characterization of Pericoronary, Epicardial, and Paracardial Adipose Tissue and Its Association With Myocardial Ischemia as Assessed by Computed Fractional Flow Reserve.

BACKGROUND: Increased pericoronary adipose tissue (PCAT) attenuation derived from coronary computed tomography (CT) angiography (CTA) relates to coronary inflammation and cardiac mortality. We aimed to investigate the association between CT-derived characterization of different cardiac fat compartments and myocardial ischemia as assessed by computed fractional flow reserve (FFRCT). METHODS: In all, 133 patients (median 64 y, 74% male) with coronary artery disease (CAD) underwent CTA including FFRCT measurement followed by invasive FFR assessment (FFRINVASIVE). CT attenuation and volume of PCAT were quantified around the proximal right coronary artery (RCA), left anterior descending artery (LAD), and left circumflex artery (LCX). Epicardial adipose tissue (EAT) and paracardial adipose tissue (PAT; all intrathoracic adipose tissue outside the pericardium) were quantified in noncontrast cardiac CT datasets. RESULTS: Median FFRCT was 0.86 [0.79, 0.91] and median FFRINVASIVE was 0.87 [0.81, 0.93]. Subjects with the presence of myocardial ischemia (n=26) defined by an FFRCT-threshold of ≤0.75 showed significantly higher RCA PCAT attenuation than individuals without myocardial ischemia (n=107) (-75.1±10.8 vs. -81.1±10.6 HU, P=0.011). In multivariable analysis adjusted for age, body mass index, sex and risk factors, increased RCA PCAT attenuation remained a significant predictor of myocardial ischemia. Between individuals with myocardial ischemia compared with individuals without myocardial ischemia, there was no significant difference in the volume and CT attenuation of EAT and PAT or in the PCAT volume of RCA, LAD, and LCX. CONCLUSIONS: Increased RCA PCAT attenuation is associated with the presence of myocardial ischemia as assessed by FFR, while PCAT volume, EAT, and PAT are not.

Computed tomography-based pericoronary adipose tissue attenuation in patients undergoing TAVR: a novel method for risk assessment.

Objectives: This study aims to assess the attenuation of pericoronary adipose tissue (PCAT) surrounding the proximal right coronary artery (RCA) in patients with aortic stenosis (AS) and undergoing transcatheter aortic valve replacement (TAVR). RCA PCAT attenuation is a novel computed tomography (CT)-based marker for evaluating coronary inflammation. Coronary artery disease (CAD) in TAVR patients is common and usually evaluated prior to intervention. The most sensible screening method and consequential treatment approach are unclear and remain a matter of ceaseless discussion. Thus, interest remains for safe and low-demand predictive markers to identify patients at risk for adverse outcomes postaortic valve replacement. Methods: This single-center retrospective study included patients receiving a standard planning CT scan prior to TAVR. Conventional CAD diagnostic tools, such as coronary artery calcium score and significant stenosis via invasive coronary angiography and coronary computed tomography angiography, were determined in addition to RCA PCAT attenuation using semiautomated software. These were assessed for their relationship with major adverse cardiovascular events (MACE) during a 24-month follow-up period. Results: From a total of 62 patients (mean age: 82 ± 6.7 years), 15 (24.2%) patients experienced an event within the observation period, 10 of which were attributed to cardiovascular death. The mean RCA PCAT attenuation was higher in patients enduring MACE than that in those without an endpoint (-69.8 ± 7.5 vs. -74.6 ± 6.2, P = 0.02). Using a predefined cutoff of >-70.5 HU, 20 patients (32.3%) with high RCA PCAT attenuation were identified, nine (45%) of which met the endpoint within 2 years after TAVR. In a multivariate Cox regression model including conventional CAD diagnostic tools, RCA PCAT attenuation prevailed as the only marker with significant association with MACE (P = 0.02). After dichotomization of patients into high- and low-RCA PCAT attenuation groups, high attenuation was related to greater risk of MACE (hazard ration: 3.82, P = 0.011). Conclusion: RCA PCAT attenuation appears to have predictive value also in a setting of concomitant AS in patients receiving TAVR. RCA PCAT attenuation was more reliable than conventional CAD diagnostic tools in identifying patients at risk for MACE .

Differences of inflammatory cytokine profile in patients with vulnerable plaque: A coronary CTA study.

BACKGROUND AND AIMS: Various pro- and anti-inflammatory biomarkers are involved in the process of atherosclerosis. We analyzed the association of different biomarkers with coronary plaque volume and vulnerable plaque subcomponents. METHODS: In 301 patients undergoing coronary CT angiography (CTA), total coronary plaque volume (TPV) and subcomponents including non-calcified plaque volume (NCPV) and vulnerable plaque burden were quantified using semi-automated software. Serum was analyzed for various cytokines. RESULTS: Out of 301 patients, 207 (69%) were male. The mean age was 59 ± 10 years. Patients were divided using the median of TPV, NCPV and vulnerable plaque burden. In univariable analysis, patients with high TPV, high NCPV and high vulnerable plaque burden showed significant higher serum levels for IFNƔ, IL-1a, -2, -4, -10 and -17 and significant lower levels for IL-8 and MCP-1 (all p < 0.05). Multivariable analysis showed positive associations between high vulnerable plaque burden, IL-1a (OR 2.60, p = 0.001) and Eotaxin (OR 1.89, p = 0.020), and inverse association to MCP-1 (OR 0.33, p < 0.001), independent of age, gender and CVRF. In exploratory subanalyses, patients with presence of atherosclerosis (n = 247; 82%) showed significantly higher levels of IL-17 in all subgroups with high vulnerable plaque burden, irrespective of overall plaque volume (all p < 0.001). CONCLUSIONS: The cytokine profile significantly differs between patients with high and low coronary plaque volume. IL-1a and IL-17 seem to play a major proatherogenic role in vulnerable plaque formation, whereas MCP-1 paradoxically portends protective effects. Longitudinal studies with serial cytokine testing are needed to identify potential targets for therapeutic interventions.

Radiomics-Based Precision Phenotyping Identifies Unstable Coronary Plaques From Computed Tomography Angiography.

OBJECTIVES: The aim of this study was to precisely phenotype culprit and nonculprit lesions in myocardial infarction (MI) and lesions in stable coronary artery disease (CAD) using coronary computed tomography angiography (CTA)-based radiomic analysis. BACKGROUND: It remains debated whether any single coronary atherosclerotic plaque within the vulnerable patient exhibits unique morphology conferring an increased risk of clinical events. METHODS: A total of 60 patients with acute MI prospectively underwent coronary CTA before invasive angiography and were matched to 60 patients with stable CAD. For all coronary lesions, high-risk plaque (HRP) characteristics were qualitatively assessed, followed by semiautomated plaque quantification and extraction of 1,103 radiomic features. Machine learning models were built to examine the additive value of radiomic features for discriminating culprit lesions over and above HRP and plaque volumes. RESULTS: Culprit lesions had higher mean volumes of noncalcified plaque (NCP) and low-density noncalcified plaque (LDNCP) compared with the highest-grade stenosis nonculprits and highest-grade stenosis stable CAD lesions (NCP: 138.1 mm3 vs 110.7 mm3 vs 102.7 mm3; LDNCP: 14.2 mm3 vs 9.8 mm3 vs 8.4 mm3; both Ptrend < 0.01). In multivariable linear regression adjusted for NCP and LDNCP volumes, 14.9% (164 of 1,103) of radiomic features were associated with culprits and 9.7% (107 of 1,103) were associated with the highest-grade stenosis nonculprits (critical P < 0.0007) when compared with highest-grade stenosis stable CAD lesions as reference. Hierarchical clustering of significant radiomic features identified 9 unique data clusters (latent phenotypes): 5 contained radiomic features specific to culprits, 1 contained features specific to highest-grade stenosis nonculprits, and 3 contained features associated with either lesion type. Radiomic features provided incremental value for discriminating culprit lesions when added to a machine learning model containing HRP and plaque volumes (area under the receiver-operating characteristic curve 0.86 vs 0.76; P = 0.004). CONCLUSIONS: Culprit lesions and highest-grade stenosis nonculprit lesions in MI have distinct radiomic signatures compared with lesions in stable CAD. Within the vulnerable patient may exist individual vulnerable plaques identifiable by coronary CTA-based precision phenotyping.

Deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction: an international multicentre study.

BACKGROUND: Atherosclerotic plaque quantification from coronary CT angiography (CCTA) enables accurate assessment of coronary artery disease burden and prognosis. We sought to develop and validate a deep learning system for CCTA-derived measures of plaque volume and stenosis severity. METHODS: This international, multicentre study included nine cohorts of patients undergoing CCTA at 11 sites, who were assigned into training and test sets. Data were retrospectively collected on patients with a wide range of clinical presentations of coronary artery disease who underwent CCTA between Nov 18, 2010, and Jan 25, 2019. A novel deep learning convolutional neural network was trained to segment coronary plaque in 921 patients (5045 lesions). The deep learning network was then applied to an independent test set, which included an external validation cohort of 175 patients (1081 lesions) and 50 patients (84 lesions) assessed by intravascular ultrasound within 1 month of CCTA. We evaluated the prognostic value of deep learning-based plaque measurements for fatal or non-fatal myocardial infarction (our primary outcome) in 1611 patients from the prospective SCOT-HEART trial, assessed as dichotomous variables using multivariable Cox regression analysis, with adjustment for the ASSIGN clinical risk score. FINDINGS: In the overall test set, there was excellent or good agreement, respectively, between deep learning and expert reader measurements of total plaque volume (intraclass correlation coefficient [ICC] 0·964) and percent diameter stenosis (ICC 0·879; both p<0·0001). When compared with intravascular ultrasound, there was excellent agreement for deep learning total plaque volume (ICC 0·949) and minimal luminal area (ICC 0·904). The mean per-patient deep learning plaque analysis time was 5·65 s (SD 1·87) versus 25·66 min (6·79) taken by experts. Over a median follow-up of 4·7 years (IQR 4·0-5·7), myocardial infarction occurred in 41 (2·5%) of 1611 patients from the SCOT-HEART trial. A deep learning-based total plaque volume of 238·5 mm3 or higher was associated with an increased risk of myocardial infarction (hazard ratio [HR] 5·36, 95% CI 1·70-16·86; p=0·0042) after adjustment for the presence of deep learning-based obstructive stenosis (HR 2·49, 1·07-5·50; p=0·0089) and the ASSIGN clinical risk score (HR 1·01, 0·99-1·04; p=0·35). INTERPRETATION: Our novel, externally validated deep learning system provides rapid measurements of plaque volume and stenosis severity from CCTA that agree closely with expert readers and intravascular ultrasound, and could have prognostic value for future myocardial infarction. FUNDING: National Heart, Lung, and Blood Institute and the Miriam & Sheldon G Adelson Medical Research Foundation.

Imaging of the Pericoronary Adipose Tissue (PCAT) Using Cardiac Computed Tomography: Modern Clinical Implications.

Modern coronary computed tomography angiography (CTA) is the gold standard to visualize the epicardial adipose tissue (EAT) and pericoronary adipose tissue (PCAT). The EAT is a metabolic active fat depot enclosed by the visceral pericardium and surrounds the coronary arteries. In disease states with increased EAT volume and dysfunctional adipocytes, EAT secretes an increased amount of adipocytokines and the resulting imbalance of proinflammatory and anti-inflammatory mediators potentially causes atherogenic effects on the coronary vessel wall in a paracrine way ("outside-to-inside" signaling). These EAT-induced atherogenic effects are reported to increase the risk for the development of coronary artery disease, myocardial ischemia, high-risk plaque features, and future major adverse cardiac events. Coronary inflammation plays a key role in the development and progression of coronary artery disease; however, its noninvasive detection remains challenging. In future, this clinical dilemma might be changed by the CTA-derived analysis of the PCAT. On the basis of the concept of an "inside-to-outside" signaling between the inflamed coronary vessel wall and the surrounding PCAT recent evidence demonstrates that PCAT computed tomography attenuation especially around the right coronary artery derived from routine CTA is a promising imaging biomarker and "sensor" to noninvasively detect coronary inflammation. This review summarizes the biological and technical principles of CTA-derived PCAT analysis and highlights its clinical implications to improve modern cardiovascular prevention strategies.

High levels of eicosapentaenoic acid are associated with lower pericoronary adipose tissue attenuation as measured by coronary CTA.

BACKGROUND AND AIMS: Higher pericoronary adipose tissue (PCAT) attenuation, a novel marker of inflammation in coronary CT angiography (CTA), has been shown to indicate increased cardiac mortality. Supplementation of eicosapentaenoic acid (EPA) has been shown to decrease cardiovascular death. Whether blood levels of n-3 fatty acids are associated with differences in PCAT attenuation is unknown. METHODS: This is a cross-sectional analysis including 64 symptomatic patients who underwent coronary CTA. PCAT attenuation was measured in Hounsfield Units (HU) around the proximal 40 mm of the right coronary artery using semi-automated software. Erythrocyte membrane fatty acid composition was analyzed using gas chromatography. Individual fatty acids were expressed as a percentage of total identified fatty acids. RESULTS: The patient cohort was divided into two groups using the median PCAT attenuation of -78.1 HU (each n = 32). No differences were seen in age, sex, BMI or traditional cardiovascular risk factors (CVRF) between groups (all p > 0.05). In univariable analysis, significantly higher values of EPA (1.00% [0.78; 1.26] vs. 0.78% [0.63; 0.99]; p = 0.02) were seen in patients with lower PCAT attenuation. All other fatty acids showed no differences (all p > 0.05). Moreover, a significant negative correlation was seen between PCAT attenuation and EPA (CC: 0.38; p = 0.002). In multivariable analysis, an inverse association of EPA with PCAT attenuation existed (ß = -0.31, p = 0.017), independent of age, gender, BMI and number of CVRF (all p > 0.1). CONCLUSIONS: High levels of EPA are associated with lower PCAT attenuation on coronary CTA. This may indicate a different composition of pericoronary adipose tissue, potentially caused by a lower degree of coronary inflammation.

Pericoronary adipose tissue CT attenuation and its association with serum levels of atherosclerosis-relevant inflammatory mediators, coronary calcification and major adverse cardiac events.

BACKGROUND: Increased attenuation of pericoronary adipose tissue (PCAT) around the right coronary artery (RCA) derived from coronary CTA might detect coronary inflammation. We investigated a potential association between RCA PCAT attenuation and serum levels of atherosclerosis-relevant cytokines and MACE (coronary revascularization, myocardial infarction and/or cardiac death). METHODS: Blood samples of 293 clinically stable individuals (59.0 â€‹± â€‹9.8 years, 69% males) were analyzed for atherosclerosis-relevant cytokines including interleukin (IL)-2, IL- 4, IL-6, IL-7, IL-8, IL-10, IL-13, IL-15, IL-17, TNF-a, IP-10, CRP, MCP-1, MIP-1a, Eotaxin and GM-CSF. Subjects also underwent coronary calcium scoring (CCS) followed by CTA. PCAT CT attenuation was measured around the RCA using semi-automated software. Increased RCA PCAT attenuation was defined as PCAT attenuation above the 75th percentile (>-73.5 HU). To assess MACE, 232 individuals were followed for a mean duration of 9.6 â€‹± â€‹2.1 years. RESULTS: In patients with increased RCA PCAT attenuation the serum levels of MCP-1 were increased (p â€‹< â€‹0.01), whereas levels of anti-inflammatory mediators IL-4 and -13 were significantly reduced (each p â€‹< â€‹0.05). Adipocytokine MCP-1 (r â€‹= â€‹0.23, p â€‹< â€‹0.01) and pro-inflammatory mediator IL-7 (r â€‹= â€‹0.12, p â€‹= â€‹0.04) showed a mild positive correlation with RCA PCAT attenuation, whereas anti-inflammatory mediators Il-4, -10 and -13 correlated inversely (each r < -0.12, each p â€‹< â€‹0.05). 40/232 patients experienced MACE during follow-up. In multivariable Cox regression analysis increased RCA PCAT attenuation was shown to be an independent predictor of MACE (HR 2.01, p â€‹= â€‹0.044). CONCLUSIONS: Increased RCA PCAT CT attenuation shows a weak association with serum levels of selected atherosclerosis-relevant inflammatory biomarkers. Increased RCA PCAT attenuation is an independent predictor of MACE and may potentially guide future prevention strategies in stable patients.

Ethnic differences in coronary anatomy, left ventricular mass and CT-derived fractional flow reserve.

BACKGROUND: Studies have observed higher incidence of cardiovascular mortality in South Asians (SA), and lower prevalence in East Asians (EA), compared with Caucasians. These observations are not entirely explained by ethnic differences in cardiovascular risk factors and mechanistic factors such as variations in cardiac anatomy and physiology may play a role. This study compared ethnic differences in CT-assessed left ventricular (LV) mass, coronary anatomy and non-invasive fractional flow reserve (FFRCT). METHODS: Three-hundred symptomatic patients (age 59 ± 7.9, male 51%) underwent clinically-mandated CT-coronary-angiography (CTA) were matched for age, gender, BMI and diabetes (100 each ethnicity). Assessment of coronary stenosis, luminal dimensions and vessel dominance was performed by independent observers. LV mass, coronary luminal volume and FFRCT were quantified by blinded core-laboratory. A sub-analysis was performed on patients (n = 187) with normal/minimal disease (0-25% stenosis). RESULTS: Stenosis severity was comparable across ethnic groups. EA demonstrated less left-dominant circulation (2%) compared with SA (8.2%) and Caucasians (10.1%). SA compared with EA and Caucasians demonstrated smallest indexed LV mass, coronary luminal volumes and dimensions. EA compared with Caucasians had comparable indexed LV mass, coronary luminal dimensions and highest luminal volumes. The latter was driven by higher prevalence of right-dominance including larger and longer right posterior left ventricular artery. FFRCT in the left anterior descending artery (LAD) was lowest in SA (0.87) compared with EA (0.89; P = 0.009) and Caucasians (0.89; P < 0.001), with no difference in other vessels. All observed differences were consistent in patients with minimal disease. CONCLUSION: This single-centre study identified significant ethnic differences in CT-assessed LV mass, coronary anatomy and LAD FFRCT. These hypotheses generating results may provide a mechanistic explanation for ethnic differences in cardiovascular outcomes and require validation in larger cohorts.

Machine learning integration of circulating and imaging biomarkers for explainable patient-specific prediction of cardiac events: A prospective study.

BACKGROUND AND AIMS: We sought to assess the performance of a comprehensive machine learning (ML) risk score integrating circulating biomarkers and computed tomography (CT) measures for the long-term prediction of hard cardiac events in asymptomatic subjects. METHODS: We studied 1069 subjects (age 58.2 ± 8.2 years, 54.0% males) from the prospective EISNER trial who underwent coronary artery calcium (CAC) scoring CT, serum biomarker assessment, and long-term follow-up. Epicardial adipose tissue (EAT) was quantified from CT using fully automated deep learning software. Forty-eight serum biomarkers, both established and novel, were assayed. An ML algorithm (XGBoost) was trained using clinical risk factors, CT measures (CAC score, number of coronary lesions, aortic valve calcium score, EAT volume and attenuation), and circulating biomarkers, and validated using repeated 10-fold cross validation. RESULTS: At 14.5 ± 2.0 years, there were 50 hard cardiac events (myocardial infarction or cardiac death). The ML risk score (area under the receiver operator characteristic curve [AUC] 0.81) outperformed the CAC score (0.75) and ASCVD risk score (0.74; both p = 0.02) for the prediction of hard cardiac events. Serum biomarkers provided incremental prognostic value beyond clinical data and CT measures in the ML model (net reclassification index 0.53 [95% CI: 0.23-0.81], p < 0.0001). Among novel biomarkers, MMP-9, pentraxin 3, PIGR, and GDF-15 had highest variable importance for ML and reflect the pathways of inflammation, extracellular matrix remodeling, and fibrosis. CONCLUSIONS: In this prospective study, ML integration of novel circulating biomarkers and noninvasive imaging measures provided superior long-term risk prediction for cardiac events compared to current risk assessment tools.

Pericoronary adipose tissue and quantitative global non-calcified plaque characteristics from CT angiography do not differ in matched South Asian, East Asian and European-origin Caucasian patients with stable chest pain.

PURPOSE: South Asian (SA) have been observed to have higher cardiovascular mortality rates compared to East Asians (EA) and Caucasians. Pericoronary adipose tissue (PCAT) attenuation around the right coronary artery (RCA) from coronary CT angiography (CTA) has been associated with coronary inflammation and cardiac death. We aimed to investigate i) the relationship between plaque characteristics and PCAT attenuation and ii) to assess gender and ethnic differences in PCAT attenuation using a matched cohort of SA, EA and Caucasians. METHOD: Three-hundred symptomatic patients who underwent CTA were matched for age, gender, BMI and diabetes (100 in each ethnic group). Semi-automated software was used to quantify the total volumes and burden of non-calcified plaque (NCP), low-density non-calcified plaque (LD-NCP) and calcified plaque (CP) in blinded core-lab analysis. PCAT CT attenuation was measured around the RCA (10-50 mm from RCA ostium), the most standardized model for PCAT analysis. RESULTS: The total volumes and burden of NCP, LD-NCP and CP were comparable in the ethnic groups (each p > 0.05). PCAT attenuation was higher in patients with coronary plaque. PCAT attenuation correlated with the total volumes and burden of NCP, LD-NCP and CP (r>0.17; p < 0.003). Within the RCA this correlation persisted only for NCP features (r>0.39;p < 0.001). Males showed higher PCAT attenuation (p < 0.001). PCAT attenuation was similar between Caucasian, EA and SA (p = 0.32). CONCLUSIONS: PCAT CT attenuation correlated most with its surrounded NCP features further highlighting its role as surrogate measure of coronary inflammation. As coronary plaque burden and RCA PCAT attenuation did not differ between ethnic groups, causes of increased cardiac mortality in South Asians needs further investigations.

Deep Learning-Based Quantification of Epicardial Adipose Tissue Volume and Attenuation Predicts Major Adverse Cardiovascular Events in Asymptomatic Subjects.

BACKGROUND: Epicardial adipose tissue (EAT) volume (cm3) and attenuation (Hounsfield units) may predict major adverse cardiovascular events (MACE). We aimed to evaluate the prognostic value of fully automated deep learning-based EAT volume and attenuation measurements quantified from noncontrast cardiac computed tomography. METHODS: Our study included 2068 asymptomatic subjects (56±9 years, 59% male) from the EISNER trial (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) with long-term follow-up after coronary artery calcium measurement. EAT volume and mean attenuation were quantified using automated deep learning software from noncontrast cardiac computed tomography. MACE was defined as myocardial infarction, late (>180 days) revascularization, and cardiac death. EAT measures were compared to coronary artery calcium score and atherosclerotic cardiovascular disease risk score for MACE prediction. RESULTS: At 14±3 years, 223 subjects suffered MACE. Increased EAT volume and decreased EAT attenuation were both independently associated with MACE. Atherosclerotic cardiovascular disease risk score, coronary artery calcium, and EAT volume were associated with increased risk of MACE (hazard ratio [95%CI]: 1.03 [1.01-1.04]; 1.25 [1.19-1.30]; and 1.35 [1.07-1.68], P<0.01 for all) and EAT attenuation was inversely associated with MACE (hazard ratio, 0.83 [95% CI, 0.72-0.96]; P=0.01), with corresponding Harrell C statistic of 0.76. MACE risk progressively increased with EAT volume ≥113 cm3 and coronary artery calcium ≥100 AU and was highest in subjects with both (P<0.02 for all). In 1317 subjects, EAT volume was correlated with inflammatory biomarkers C-reactive protein, myeloperoxidase, and adiponectin reduction; EAT attenuation was inversely related to these biomarkers. CONCLUSIONS: Fully automated EAT volume and attenuation quantification by deep learning from noncontrast cardiac computed tomography can provide prognostic value for the asymptomatic patient, without additional imaging or physician interaction.

Machine learning to predict the long-term risk of myocardial infarction and cardiac death based on clinical risk, coronary calcium, and epicardial adipose tissue: a prospective study.

AIMS: Our aim was to evaluate the performance of machine learning (ML), integrating clinical parameters with coronary artery calcium (CAC), and automated epicardial adipose tissue (EAT) quantification, for the prediction of long-term risk of myocardial infarction (MI) and cardiac death in asymptomatic subjects. METHODS AND RESULTS: Our study included 1912 asymptomatic subjects [1117 (58.4%) male, age: 55.8 ± 9.1 years] from the prospective EISNER trial with long-term follow-up after CAC scoring. EAT volume and density were quantified using a fully automated deep learning method. ML extreme gradient boosting was trained using clinical co-variates, plasma lipid panel measurements, risk factors, CAC, aortic calcium, and automated EAT measures, and validated using repeated 10-fold cross validation. During mean follow-up of 14.5 ± 2 years, 76 events of MI and/or cardiac death occurred. ML obtained a significantly higher AUC than atherosclerotic cardiovascular disease (ASCVD) risk and CAC score for predicting events (ML: 0.82; ASCVD: 0.77; CAC: 0.77, P < 0.05 for all). Subjects with a higher ML score (by Youden's index) had high hazard of suffering events (HR: 10.38, P < 0.001); the relationships persisted in multivariable analysis including ASCVD-risk and CAC measures (HR: 2.94, P = 0.005). Age, ASCVD-risk, and CAC were prognostically important for both genders. Systolic blood pressure was more important than cholesterol in women, and the opposite in men. CONCLUSIONS: In this prospective study, machine learning used to integrate clinical and quantitative imaging-based variables significantly improves prediction of MI and cardiac death compared with standard clinical risk assessment. Following further validation, such a personalized paradigm could potentially be used to improve cardiovascular risk assessment.

Quantification of epicardial adipose tissue by cardiac CT: Influence of acquisition parameters and contrast enhancement.

PURPOSE: While computed tomography (CT) is frequently used to quantify epicardial adipose tissue (EAT), the effect of different acquisition parameters on EAT volume has not been systematically reported. We assessed the influence of low-voltage acquisition and contrast enhancement on EAT quantification. METHOD: Two independent cohorts (100 and 127 patients) referred for routine coronary CT were included. One cohort received a low-voltage and a standard voltage non-contrast acquisition (120 and 100 kV), the other cohort underwent non-contrast and contrast-enhanced CT. EAT volume was quantified using a semi-automated analysis software. Whereas the lower EAT threshold was consistently set at -190 Hounsfield Units (HU), different upper thresholds for EAT were analyzed. Bland-Altman analysis was used to analyze the agreement of EAT volume between scans with different acquisition parameters. We referred to a non-enhanced 120 kV acquisition with an upper threshold of -30 HU. RESULTS: Mean EAT volume was 159 ±â€¯76 ml as measured in 120 kV non-contrast data sets with an upper threshold of -30 HU. For 100 kV data sets, an upper threshold of -40 HU showed the best correlation (r = 0.961, p < 0.05). Significant overestimation was found for upper thresholds of -20 and -30 HU and significant underestimation for -50 HU. In non-contrast vs. contrast-enhanced acquisitions, there was a significant underestimation of EAT volume for contrast-enhanced scans (mean difference 31 ml, 95% limits of agreement 27 to -89 ml). CONCLUSIONS: CT-based EAT volume quantification in low-voltage and contrast-enhanced images is feasible. However, adjustment of the upper threshold for detection of fat is mandatory.

Deep learning-based stenosis quantification from coronary CT Angiography.

BACKGROUND: Coronary computed tomography angiography (CTA) allows quantification of stenosis. However, such quantitative analysis is not part of clinical routine. We evaluated the feasibility of utilizing deep learning for quantifying coronary artery disease from CTA. METHODS: A total of 716 diseased segments in 156 patients (66 ± 10 years) who underwent CTA were analyzed. Minimal luminal area (MLA), percent diameter stenosis (DS), and percent contrast density difference (CDD) were measured using semi-automated software (Autoplaque) by an expert reader. Using the expert annotations, deep learning was performed with convolutional neural networks using 10-fold cross-validation to segment CTA lumen and calcified plaque. MLA, DS and CDD computed using deep-learning-based approach was compared to expert reader measurements. RESULTS: There was excellent correlation between the expert reader and deep learning for all quantitative measures (r=0.984 for MLA; r=0.957 for DS; and r=0.975 for CDD, p<0.001 for all). The expert reader and deep learning method was not significantly different for MLA (median 4.3 mm2 for both, p=0.68) and CDD (11.6 vs 11.1%, p=0.30), and was significantly different for DS (26.0 vs 26.6%, p<0.05); however, the ranges of all the quantitative measures were within inter-observer variability between 2 expert readers. CONCLUSIONS: Our deep learning-based method allows quantitative measurement of coronary artery disease segments accurately from CTA and may enhance clinical reporting.