Improving the assessment of left ventricular diastolic dysfunction by including left atrial strain in the algorithm: LA Strain and LV diastolic dysfunction assessment.

BACKGROUND: The latest guidelines on echocardiographic assessment of left ventricular diastolic dysfunction (LVDD) leave a significant proportion of patients with LVDD status undetermined. We aimed to examine the implication of an alternative algorithm incorporating left atrial (LA) strain as a tiebreaker on the indeterminate LVDD category. METHODS AND RESULTS: We included 823 patients who underwent echocardiography and cardiac MRI within 7 days. LVDD was assessed by echocardiography following contemporary guidelines and an alternative algorithm including LA reservoir strain as a tie-breaker. LVDD was examined for its association with LV myocardial scar burden by cardiac MRI, and a composite outcome. 275 (33%) patients had LVDD, of whom 119 had advanced grades of LVDD (grade II-III), and 117 (14%) had indeterminate LVDD grade. When LA strain was applied at cut points of 18%, 24% and 35%, subjects were reclassified as normal or LVDD dependent accordingly. Reclassification allowed a similar outcome risk-stratification as the current guidelines. CONCLUSIONS: LA reservoir strain improved LVDD assessment by eliminating indeterminate status/grade while maintaining the same effective outcome stratification as current guidelines.

Association of pulmonary transit time by cardiac magnetic resonance with heart failure hospitalization in a large prospective cohort with diverse cardiac conditions.

BACKGROUND: Longer pulmonary transit time (PTT) is closely associated with hemodynamic abnormalities. However, the implications on heart failure (HF) risk have not been investigated broadly in patients with diverse cardiac conditions. In this study we examined the long-term risk of HF hospitalization associated with longer PTT in a large prospective cohort with a broad spectrum of cardiac conditions. METHODS: All subjects were prospectively recruited to undergo cardiac magnetic resonance (CMR). The dynamic images of first-pass perfusion were acquired to assess peak-to-peak pulmonary transit time (PTT) which was subsequently normalized to RR interval duration. The risk of HF was examined using Cox proportional hazards models adjusted for baseline confounding risk factors. RESULTS: Among 506 consecutively consented patients undergoing clinical cardiac MR with diverse cardiac conditions, the mean age was 63 ± 14 years and 373 (73%) were male. After a mean follow up duration of 4.5 ± 3.0 years, 70 (14%) patients developed hospitalized HF and of these 6 died. A normalized PTT ≥ 8.2 was associated with a significantly increased adjusted HF hazard ratio of 3.69 (95% CI 2.02, 6.73). The HF hazard ratio was 1.26 (95% CI 1.18, 1.33) for each 1 unit increase in PTT which was higher among those preserved (1.70, 95% CI 1.20, 2.41) compared to those with reduced left ventricular ejection fraction (< 50%) (1.18, 95% CI 1.09, 1.27). PTT remained a significant risk factor of hospitalized HF after additional adjustment for N-terminal pro-hormone brain natriuretic peptide (NT-proBNP) or left ventricular global longitudinal strain with additionally demonstrated incremental model improvement through likelihood ratio testing. CONCLUSIONS: Our findings support the role of PTT in assessing HF risk among patients with broad spectrum of cardiac conditions with reduced as well as preserved ejection fraction. Longer PTT duration is an incremental risk factor for HF when baseline global longitudinal strain and NT-proBNP are taken into consideration.

Clinical implications of compromised 82Rb PET data acquisition.

BACKGROUND: We wished to document the prevalence and quantitative effects of compromised 82Rb PET data acquisitions on myocardial flow reserve (MFR). METHODS AND RESULTS: Data were analyzed retrospectively for 246 rest and regadenoson-stress studies of 123 patients evaluated for known or suspected CAD. An automated injector delivered pre-determined activities of 82Rb. Automated quality assurance algorithms identified technical problems for 7% (9/123) of patients. Stress data exhibited 2 instances of scanner saturation, 1 blood peak detection, 1 blood peak width, 1 gradual patient motion, and 2 abrupt patient motion problems. Rest data showed 1 instance of blood peak width and 2 abrupt patient motion problems. MFR was lower for patients with technical problems flagged by the quality assurance algorithms than those without technical problems (1.5 ± 0.5 versus 2.1 ± 0.7, P = 0.01), even though rest and stress ejection fraction, asynchrony and relative myocardial perfusion measures were similar for these two groups (P > 0.05), suggesting that MFR accuracy was adversely affected by technical errors. CONCLUSION: It is important to verify integrity of 82Rb data to ensure MFR computation quality.

Angiotensin-Converting Enzyme Gene Polymorphism and Severe Lung Injury in Patients with Coronavirus Disease 2019.

Coronavirus disease 2019 has markedly varied clinical presentations, with most patients being asymptomatic or having mild symptoms. However, severe acute respiratory disease, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is common and associated with mortality in patients who require hospitalization. The etiology of susceptibility to severe lung injury remains unclear. Angiotensin II, converted by angiotensin-converting enzyme (ACE) from angiotensin I and metabolized by ACE 2 (ACE2), plays a pivotal role in the pathogenesis of lung injury. ACE2 is identified as an essential receptor for SARS-CoV-2 to enter the cell. The binding of ACE2 and SARS-CoV-2 leads to the exhaustion and down-regulation of ACE2. The interaction and imbalance between ACE and ACE2 result in an unopposed angiotensin II. Considering that the ACE insertion (I)/deletion (D) gene polymorphism contributes to the ACE level variability in general population, in which mean ACE level in DD carriers is approximately twice that in II carriers, we propose a hypothesis of genetic predisposition to severe lung injury in patients with coronavirus disease 2019. It is plausible that the ACE inhibitors and ACE receptor blockers may have the potential to prevent and to treat the acute lung injury after SARS-CoV-2 infection, especially for those with the ACE genotype associated with high ACE level.

The association of reduced left ventricular strains with increased extracellular volume and their collective impact on clinical outcomes.

BACKGROUND: Myocardial fibrosis and left ventricular (LV) longitudinal strain are independently associated with adverse clinical outcomes. However, the relationship between tissue properties and strain indices as well as their collective impact on outcomes are yet to be fully elucidated. We aim to investigate the relationship between LV global longitudinal strain (GLS), global circumferential strain (GCS) and global radial strain (GRS) with extracellular volume (ECV) and their collective impact. METHODS: Consecutive patients referred for clinical cardiovascular magnetic resonance (CMR) due to cardiomyopathy were prospectively enrolled. All patients underwent CMR with T1 mapping. ECV was calculated incorporating native and post-contrast T1 as well as hematocrit. LV GLS, GCS, and GRS were assessed by feature tracking. Hazard ratios and Kaplan-Meier curves were produced to assess the association between strains and T1 mapping indices with a composite outcome of all-cause mortality and hospitalized heart failure. RESULTS: The study consisted of 259 patients with mixed referring diagnoses of non-ischemic/ischemic cardiomyopathy and 21 normal controls. Decreased GLS, GCS and GRS were associated with increased ECV, increased native T1, and reduced post-contrast T1 in a dose dependent manner when T1 or ECV was in the abnormal range. After a mean follow-up of 31 ± 23 months, 41 events occurred including 37 heart failure admissions and 4 deaths. Kaplan-Meier plots demonstrated that reduced strains were associated with reduced event-free survival predominantly in patients with increased ECV (≥ 28.3%). The worst outcome was among those with both reduced strains and increased ECV. In the multivariable models, increased ECV, reduced post-contrast T1 and reduced strains in all 3 directions remained predictors of outcome risk, respectively. CONCLUSION: Our findings highlight the intrinsic link between altered CMR tissue properties and impaired myocardial mechanical performance and additionally demonstrate improved risk stratification by characterizing tissue property among patients with reduced strain.

Early onset of left ventricular regional asynchrony in arteries with sub-clinical stenosis.

BACKGROUND: Asynchrony has been reported to be a marker of ischemic-induced left ventricular dysfunction, the magnitude of which correlates with extent of epicardial coronary disease. We wished to determine whether normal-appearing arterial territories with mild degrees of asynchrony have lower 82Rb PET absolute myocardial blood flow (MBF) and/or lower myocardial flow reserve (MFR). METHODS AND RESULTS: Data were examined retrospectively for 105 patients evaluated for known/suspected CAD who underwent rest/regadenoson-stress 82Rb PET/CT and quantitative coronary angiography. Rest and stress absolute MBF and MFR were quantified from first-pass 82Rb PET curves. Regional relative myocardial perfusion summed stress score (SSS), summed rest score (SRS), regional phase bandwidth (BW), and regional semi-quantitative asynchrony visual scores of (Asynch) were assessed. We found that in apparently normal arteries (SSS < 4, SRS < 4 and stenosis < 70%), those with abnormally low MFR < 2.0 compared to those with MFR ≥ 2.0 had larger phase BW (186 ± 79° vs 158 ± 67°, P = .02), and more visually apparent Asynch (5.7 ± 4.2 vs 3.9 ± 3.6, P = .02), which was associated with increasing stenosis values (ρ = 0.44, P < .0001). CONCLUSION: A subgroup of coronary territories with normal relative perfusion and normal or non-obstructive coronary disease may have reduced MFR, which is signaled physiologically by a mild degree of left ventricular asynchrony.

The Prognostic Value of Left Atrial Global Longitudinal Strain and Left Atrial Phasic Volumes in Patients Undergoing Transcatheter Valve Implantation for Severe Aortic Stenosis.

INTRODUCTION: The changes and the prognostic implications of left atrial (LA) volumes (LAV), LA function, and vascular load in patients undergoing transcatheter aortic valve implantation (TAVI) for severe aortic stenosis (AS) are less known. METHODS: We enrolled 150 symptomatic patients (mean age 82 ± 8 years, 58% female, and pre-TAVI aortic valve area 0.40 ± 0.19 cm/m2) with severe AS who underwent 2D transthoracic echocardiography and 2D speckle tracking echocardiography at average 21 ± 35 days before and 171 ± 217 days after TAVI. The end point was a composite of new onset of atrial fibrillation, hospitalization for heart failure and all-cause death (major adverse cardiac events [MACE]). RESULTS: After TAVI, indexed maximal LA volume and minimum volume of the LA decreased by 2.1 ± 10 mL/m2 and 1.6 ± 7 mL/m2 (p = 0.032 and p = 0.011, respectively), LA function index increased by 6.8 ± 11 units (p < 0.001), and LA stiffness decreased by 0.38 ± 2.0 (p = 0.05). No other changes in the LA phasic volumes, emptying fractions, and vascular load were noted. Post-TAVI, both left atrial and ventricular global peak longitudinal strain improved by about 6% (p = 0.01 and 0.02, respectively). MACE was reached by 37 (25%) patients after a median follow-up period of 172 days (interquartile range, 20-727). In multivariable models, MACE was associated with both pre- and post-TAVI LA global peak longitudinal strain (hazard ratio [HR] 0.75, CI 0.59-0.97; and HR 0.77, CI 0.60-1.00, per 5 percentage point units, respectively), pre-TAVI LV global endocardial longitudinal strain (HR 1.37, CI 1.02-1.83 per 5 percentage point units), and with most of the LA phasic volumes. CONCLUSION: Within 6 months after TAVI, there is reverse LA remodeling and an improvement in LA reservoir function. Pre- and post-TAVI indices of LA function and volume remain independently associated with MACE. Larger studies enrolling a greater diversity of patients may provide sufficient evidence for the utilization of these imaging biomarkers in clinical practice.

Relationship of 82Rb PET territorial myocardial asynchrony to arterial stenosis.

OBJECTIVE: 82Rb PET/CT rest/regadenoson-stress data enable quantification of left ventricular rest and stress function, perfusion, and asynchrony. Our study was conducted to determine which parameters best identify patients with multi-vessel disease (MVD) and individual stenosed arteries. METHODS: PET/CT data were reviewed retrospectively for 105 patients referred for evaluation of CAD, who also underwent angiography. % arterial stenosis was determined quantitatively at a core laboratory. Severe stenosis was defined as ≥ 70%, and MVD as 2 or more stenosed arteries. Segmental MBF was calculated from first-pass data for arterial territories. Regional rest and stress systolic and diastolic asynchrony (Asynch) scores were determined from visual examination of phase polar maps. RESULTS: 65 vessels had stenoses ≥ 70%. 15 patients had MVD. ROC area under curve (ROC AUC) for identifying patients with MVD was 83% for Asynch and 73% for MFR. ROC AUC for identifying individual arterial territories with stenoses ≥ 70% was 81% and 72% for Asynch and MFR. CONCLUSION: 82Rb PET/CT accurately identified patients with MVD and individual stenosed territories, with regional asynchrony measurements contributing significantly to identify patients with CAD.

The Association between Aortic Valve Weight, Echocardiographic Indices, and All-Cause Death in 1,046 Patients Undergoing Surgical Aortic Valve Replacement for Aortic Stenosis.

BACKGROUND: Aortic valve weight (AVW), a flow independent measure of aortic stenosis (AS) severity, is reported to have heterogeneous associations with the echocardiographic variables used for AS evaluation. Controversy exists regarding its impact on survival after aortic valve replacement (AVR). OBJECTIVE: We sought to determine the association between AVW with echocardiographic measures of AS severity and all-cause mortality after surgical AVR. METHODS: One thousand and forty-sixconsecutive patients underwent surgical AVR for AS, the excised valves were weighed, and an echocardiogram was done before surgery. RESULTS: Males had heavier valves than females, for both absolute and body surface are (BSA)-indexed values (2.78 ± 1.23 vs. 2.08 ± 0.68 g, p < 0.001; and 1.38 ± 0.61 vs. 1.19 ± 0.41 g/m2, p < 0.001, respectively). In a restricted cohort of 634 patients with isolated severe AS and normal ejection fraction, the correlations of AVW with echocardiographic variables of AS were modest, the strongest being with the dimensionless index (r = -0.27 and -0.26 for male and female, both p < 0.01). Stratified by stroke volume index and mean gradient (MG), no associations were found in the low-gradient groups (i.e., MG <40 mmHg). At a median follow-up of 3.5 years, there were only 244 deaths in the entire cohort. Mortality was not related to AVW, except in females who displayed an inverse relationship (HR = 0.67; 95% CI 0.47-0.95) only when it was analyzed as a continuous variable. CONCLUSIONS: The weak correlation between AVW with the echocardiographic indices of AS may reflect its complex pathophysiology, heterogeneous hemodynamics, and possible pitfalls in the current echocardiographic methods used in clinical practice. The prognostic value of AVW after AVR warrants further evaluation.

Prevalence and prognosis of ischaemic and non-ischaemic myocardial fibrosis in older adults.

Aims: Non-ischaemic cardiomyopathies (NICM) can cause heart failure and death. Cardiac magnetic resonance (CMR) detects myocardial scar/fibrosis associated with myocardial infarction (MI) and NICM with late gadolinium enhancement (LGE). The aim of this study was to determine the prevalence and prognosis of ischaemic and non-ischaemic myocardial fibrosis in a community-based sample of older adults. Methods and results: The ICELAND-MI cohort, a substudy of the Age, Gene/Environment Susceptibility Reykjavik (AGES-Reykjavik) study, provided a well-characterized population of 900 subjects after excluding subjects with pre-existing heart failure. Late gadolinium enhancement CMR divided subjects into four groups: MI (n = 211), major (n = 54) non-ischaemic fibrosis (well-established, classic patterns, associated with myocarditis, infiltrative cardiomyopathies, or pathological hypertrophy), minor (n = 238) non-ischaemic fibrosis (remaining localized patterns not meeting major criteria), and a no LGE (n = 397) reference group. The primary outcome was time to death or first heart failure hospitalization. During a median follow-up of 5.8 years, 192 composite events occurred (115 deaths and 77 hospitalizations for incident heart failure). After inverse probability weighting, major non-ischaemic fibrosis [hazard ratio (HR) 3.2, P < 0.001] remained independently associated with the primary endpoint, while MI (HR 1.4, P = 0.10) and minor non-ischaemic LGE (HR 1.2, P = 0.39) did not. Major non-ischaemic fibrosis was associated with a poorer outcome than MI (HR = 2.3, P = 0.001) in the adjusted analysis. Conclusion: Major non-ischaemic patterns of myocardial fibrosis portended worse prognosis than no fibrosis/scar in an older community-based cohort. Traditional risk factors largely accounted for the effect of MI and minor non-ischaemic LGE.

Water-fat separation and parameter mapping in cardiac MRI via deep learning with a convolutional neural network.

BACKGROUND: Water-fat separation is a postprocessing technique most commonly applied to multiple-gradient-echo magnetic resonance (MR) images to identify fat, provide images with fat suppression, and to measure fat tissue concentration. Recently, Numerous advancements have been reported. In contrast to early methods, the process of water-fat separation has become complicated due to multiparametric analytic models, optimization methods, and the absence of a unified framework for diverse source data. PURPOSE: To determine the feasibility and performance of MRI water-fat separation and parametric mapping via deep learning (DL) with a range of inputs. STUDY TYPE: Retrospective data usage. POPULATION/SUBJECTS: Ninety cardiac MR examinations from normal control, acute, subacute, and chronic myocardial infarction subjects were obtained, providing 1200 multiple gradient-echo acquisitions. FIELD STRENGTH/SEQUENCE: 1.5 T/2D multiple gradient-echo pulse sequence ASSESSMENT: Ground-truth training and validation water-fat separation were obtained using a graph cut method with R2 *, off-resonance correction, and a multipeak fat spectrum. U-Net DL training with single and multiecho, complex, and magnitude inputs were compared using quantitative and three-observer subjective analysis. STATISTICAL TESTS: DL methods' image structural similarity, and quantitative proton density fat fraction (PDFF), R2 *, and off-resonance quantitative values were statistically compared with the GraphCut reference standard using Student's t-test and Pearson's correlation. RESULTS: Myocardial fat deposition in chronic myocardial infarction and intramyocardial hemorrhage in acute myocardial infarction were well visualized in the DL results. Predicted values for R2 *, off-resonance, water, and fat signal intensities were well correlated with a conventional model-based water fat separation (R2 ≥ 0.97, P < 0.001) with appropriate inputs. DL parameter maps had a 14% higher signal-to-noise ratio (P < 0.001) when compared with a conventional method. DATA CONCLUSION: DL water-fat separation is feasible with a wide range of inputs, while R2 * and off-resonance mapping requires multiple echoes and complex images. With appropriate inputs, DL provides quantitative and subjective results comparable to conventional model-based methods. LEVEL OF EVIDENCE: 1 Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2019;50:655-665.

A comparison of both DENSE and feature tracking techniques with tagging for the cardiovascular magnetic resonance assessment of myocardial strain.

BACKGROUND: Myocardial strain is increasingly recognized as an important assessment for myocardial function. In addition, it also improves outcome prediction. However, there is lack of standardization in strain evaluation by cardiovascular magnetic resonance (CMR). In this study we compared strain values using multiple techniques and multiple vendor products. METHODS: Prospectively recruited patients with cardiomyopathy of diverse etiology (N = 77) and healthy controls (N = 10) underwent CMR on a 1.5 T scanner. Tagging, displacement encoding with stimulated echoes (DENSE) and balanced stead state free precession cine imaging were acquired on all subjects. A single matched mid left ventricular (LV) short axis plane was used for the comparisons of peak circumferential (Ecc) and radial strain (Err) and a 4-chamber view for longitudinal strain (Ell). Tagging images were analyzed using harmonic phase (HARP) and displacement encoding with stimulated echoes (DENSE) images using a proprietary program. Feature tracking (FT) was evaluated using 3 commercially available software from Tomtec Imaging Systems, Cardiac Image Modeller (CIM), and Circle Cardiovascular Imaging. Tagging data were used as reference. Statistic analyses were performed using paired t-test, intraclass correlation coefficient (ICC), Bland Altman limits of agreement and coefficient of variations. RESULTS: Average LV ejection fraction was 50% (range 32 to 62%). Regional LV wall motion abnormalities were present in 48% of the analyzed planes. The average Ecc was - 13 ± 4%, - 13 ± 4%, - 16 ± 6%, - 10 ± 3% and - 14 ± 4% for tagging, DENSE, Tomtec, CIM and Circle, respectively, with the best agreement seen in DENSE and Circle with tagging. The Err was highly varied with poor agreement across the techniques, 32 ± 24%, 40 ± 28%, 47 ± 26%, 64 ± 33% and 23 ± 9% for tagging, DENSE, Tomtec, CIM and Circle, respectively. The average Ell was - 14 ± 4%, - 8 ± 3%, - 13 ± 5%, - 11 ± 3% and - 12 ± 4% for tagging, DENSE, Tomtec, CIM and Circle, respectively with the best agreement seen in Tomtec and Circle with tagging. In the intra- and inter-observer agreement analysis the reproducibility of each technique was good except for Err by HARP. CONCLUSIONS: Small but important differences are evident in Ecc and Ell comparisons among vendors while large differences are seen in Err assessment. Our findings suggest that CMR strain values are technique and vendor dependent. Hence, it is essential to develop reference standard from each technique and analytical product for clinical use, and to sequentially compare patient data using the same software.

The Utility of Flow Rate Compared with Left Ventricular Stroke Volume Index in the Hemodynamic Classification of Severe Aortic Stenosis with Preserved Ejection Fraction.

OBJECTIVES: To substitute the stroke volume index (SVi) with flow rate (FR) in the hemodynamic classification of severe aortic stenosis (AS) with preserved ejection fraction (EF), in order to evaluate its prognostic value. METHODS: A total of 529 patients (78.8 ± 9.8 years old, 44.1% males) with isolated severe AS (aortic valve area, AVA < 1 cm2), EF ≥50%, in sinus rhythm, who underwent transthoracic echocardiography, were stratified by FR (≥/< 200 mL/s) and mean pressure gradient (MG) (≥/< 40 mm Hg): FRnormal/MGhigh, FRlow/MGhigh, FRnormal/MGlow, and FRlow/MGlow. RESULTS: Aortic valve replacement was more frequently performed in the FRnormal/MGhigh than in the FRlow/MGlow group (69.3 vs. 47%, respectively, p < 0.0001), yielding a similar survival benefit across all four groups. Over a median follow-up of 51 ± 29 months, there were 249 deaths. In highly adjusted models, the FRlow/MGlow group had a higher all-cause mortality (HR = 1.7, 95% CI: 1.1-2.6, p = 0.02) than patients with FRnormal/MGhigh. FR had a stronger association with AVA than SVi (r = 0.51 vs. 0.41, respectively, p = 0.0002), and a similar predictive value for death (AUC = 0.57 and 0.58, respectively, p = 0.88). CONCLUSIONS: The FRlow/MGlow subset of AS is associated with the worst prognosis, and FR is not superior to SVi in the hemodynamic classification of severe AS.

SwinUNet: a multiscale feature learning approach to cardiovascular magnetic resonance parametric mapping for myocardial tissue characterization.

Objective.Cardiovascular magnetic resonance (CMR) can measure T1 and T2 relaxation times for myocardial tissue characterization. However, the CMR procedure for T1/T2 parametric mapping is time-consuming, making it challenging to scan heart patients routinely in clinical practice. This study aims to accelerate CMR parametric mapping with deep learning.Approach. A deep-learning model, SwinUNet, was developed to accelerate T1/T2 mapping. SwinUNet used a convolutional UNet and a Swin transformer to form a hierarchical 3D computation structure, allowing for analyzing CMR images spatially and temporally with multiscale feature learning. A comparative study was conducted between SwinUNet and an existing deep-learning model, MyoMapNet, which only used temporal analysis for parametric mapping. The T1/T2 mapping performance was evaluated globally using mean absolute error (MAE) and structural similarity index measure (SSIM). The clinical T1/T2 indices for characterizing the left-ventricle myocardial walls were also calculated and evaluated using correlation and Bland-Altman analysis.Main results. We performed accelerated T1 mapping with ≤4 heartbeats and T2 mapping with 2 heartbeats in reference to the clinical standard, which required 11 heartbeats for T1 mapping and 3 heartbeats for T2 mapping. SwinUNet performed well in all the experiments (MAE < 50 ms, SSIM > 0.8, correlation > 0.75, and Bland-Altman agreement limits < 100 ms for T1 mapping; MAE < 1 ms, SSIM > 0.9, correlation > 0.95, and Bland-Altman agreement limits < 1.5 ms for T2 mapping). When the maximal acceleration was used (2 heartbeats), SwinUNet outperformed MyoMapNet and gave measurement accuracy similar to the clinical standard.Significance. SwinUNet offers an optimal solution to CMR parametric mapping for assessing myocardial diseases quantitatively in clinical cardiology.

A machine learning-derived risk score to predict left ventricular diastolic dysfunction from clinical cardiovascular magnetic resonance imaging.

Introduction: The evaluation of left ventricular diastolic dysfunction (LVDD) by clinical cardiac magnetic resonance (CMR) remains a challenge. We aimed to train and evaluate a machine-learning (ML) algorithm for the assessment of LVDD by clinical CMR variables and to investigate its prognostic value for predicting hospitalized heart failure and all-cause mortality. Methods: LVDD was characterized by echocardiography following the ASE guidelines. Eight demographic and nineteen common clinical CMR variables including delayed enhancement were used to train Random Forest models with a Bayesian optimizer. The model was evaluated using bootstrap and five-fold cross-validation. Area under the ROC curve (AUC) was utilized to evaluate the model performance. An ML risk score was used to stratify the risk of heart failure hospitalization and all-cause mortality. Results: A total of 606 consecutive patients underwent CMR and echocardiography within 7 days for cardiovascular disease evaluation. LVDD was present in 303 subjects by echocardiography. The performance of the ML algorithm was good using the CMR variables alone with an AUC of 0.868 (95% CI: 0.811-0.917), which was improved by combining with demographic data yielding an AUC 0.895 (95% CI: 0.845-0.939). The algorithm performed well in an independent validation cohort with AUC 0.810 (0.731-0.874). Subjects with higher ML scores (>0.4121) were associated with increased adjusted hazard ratio for a composite outcome than subjects with lower ML scores (1.72, 95% confidence interval 1.09-2.71). Discussion: An ML algorithm using variables derived from clinical CMR is effective in identifying patients with LVDD and providing prognostication for adverse clinical outcomes.

Epicardial fat modifies the relationship between coronary calcium score and all-cause mortality: The St. Francis Heart Study.

Objective: Epicardial fat is associated with cardiovascular risk factors and adverse outcomes. However, it is not clear if epicardial fat remains to be a mortality risk when coronary calcium score (CAC) is taken into account. Methods: We studied the 1005 participants from the St. Francis Heart Study who were apparently healthy with CAC scores at 80th percentile or higher for age and gender, randomly assigned to placebo or statin therapy. At baseline, lipid profiles and non-contrast CT images were obtained where the epicardial fat volume was analyzed. Likelihood ratio testing was used to assess the additional prognostic value of epicardial fat to CAC for the risk of all-cause mortality. Results: Increased epicardial fat volume was associated with higher CAC. For each unit increase in lnCAC, the average epicardial fat volume increased by 3.34 mL/m2. After a mean follow-up period of 17 years, 179 (18%) participants died. Increased epicardial fat volume was associated with an adjusted hazard ratio of 1.11 (95% CI: 1.02 to 1.20) predicting all-cause mortality. In the stratified analysis testing strata of epicardial fat and CAC, those with increased epicardial fat and increased CAC had the highest risk of death. Compared with a model containing lnCAC and traditional risk factors, a model additionally containing epicardial fat volume yielded a better model fit (likelihood ratio test p < 0.001). Conclusion: Increased epicardial fat volume is associated with increased all-cause mortality risk. In addition, it portends incremental prognostic value to CAC score in mortality prediction.

Diagnostic Performance and Clinical Impact of Photon-Counting Detector Computed Tomography in Coronary Artery Disease.

BACKGROUND: Photon-counting detector-computed tomography (PCD-CT) has emerged as a promising technology, offering improved spatial resolution. OBJECTIVES: This study aimed to evaluate the clinical impact and diagnostic performance of PCD-CT vs conventional energy-integrating detector computed tomography (EID-CT) for obstructive coronary artery disease (CAD). METHODS: From 2022 to 2023, we retrospectively identified 7,833 consecutive patients who underwent clinically indicated coronary computed tomography angiography (CCTA) at a single center, with either PCD-CT (n = 3,876; NAEOTOM Alpha [Siemens Healthineers]) or EID-CT (n = 3,957; Revolution Apex 256 [GE HealthCare] or Aquilion ONE ViSION 320 [Canon Medical Systems]) scanners. Subsequent invasive coronary angiography (ICA) and percutaneous or surgical revascularization were performed as part of routine clinical care. Among those referred for ICA after coronary CTA, the presence of obstructive CAD in each vessel was determined by coronary CTA (severe stenosis on visual assessment per the Coronary Artery Disease Reporting and Data System) and ICA (≥50% diameter stenosis on quantitative coronary angiography) in a blinded fashion. The diagnostic performance of EID-CT and PCD-CT was compared by using quantitative coronary angiography as the reference standard. RESULTS: Patients who underwent PCD-CT were less frequently referred to subsequent ICA than those undergoing EID-CT (9.9% vs 13.1%; P < 0.001). Among those who underwent ICA, revascularization was more frequently performed in the PCD-CT group than in the EID-CT group (43.4% vs 35.5%; P = 0.02). In the vessel-level analysis (n = 1,686), specificity (98.0% vs 93.0%; P < 0.001), positive predictive value (83.3% vs 63.0%; P = 0.002), and diagnostic accuracy (97.2% vs 92.8%; P < 0.001) were improved by PCD-CT. Sensitivity (90.9% vs 90.7%; P = 0.95) and negative predictive value (98.9% vs 98.7%; P = 0.83) for obstructive CAD were similar between the PCD-CT and EID-CT groups, respectively. CONCLUSIONS: PCD-CT exhibited excellent diagnostic performance for detecting obstructive CAD. Compared with patients undergoing conventional EID-CT, fewer patients were referred to ICA after PCD-CT, but those referred were more likely to undergo revascularization.

Patient-Specific Myocardial Infarction Risk Thresholds From AI-Enabled Coronary Plaque Analysis.

BACKGROUND: Plaque quantification from coronary computed tomography angiography has emerged as a valuable predictor of cardiovascular risk. Deep learning can provide automated quantification of coronary plaque from computed tomography angiography. We determined per-patient age- and sex-specific distributions of deep learning-based plaque measurements and further evaluated their risk prediction for myocardial infarction in external samples. METHODS: In this international, multicenter study of 2803 patients, a previously validated deep learning system was used to quantify coronary plaque from computed tomography angiography. Age- and sex-specific distributions of coronary plaque volume were determined from 956 patients undergoing computed tomography angiography for stable coronary artery disease from 5 cohorts. Multicenter external samples were used to evaluate associations between coronary plaque percentiles and myocardial infarction. RESULTS: Quantitative deep learning plaque volumes increased with age and were higher in male patients. In the combined external sample (n=1847), patients in the ≥75th percentile of total plaque volume (unadjusted hazard ratio, 2.65 [95% CI, 1.47-4.78]; P=0.001) were at increased risk of myocardial infarction compared with patients below the 50th percentile. Similar relationships were seen for most plaque volumes and persisted in multivariable analyses adjusting for clinical characteristics, coronary artery calcium, stenosis, and plaque volume, with adjusted hazard ratios ranging from 2.38 to 2.50 for patients in the ≥75th percentile of total plaque volume. CONCLUSIONS: Per-patient age- and sex-specific distributions for deep learning-based coronary plaque volumes are strongly predictive of myocardial infarction, with the highest risk seen in patients with coronary plaque volumes in the ≥75th percentile.