Journal of Cardiovascular Magnetic Resonance最新文献

Background: Dysregulated myocardial calcium handling has been demonstrated in ischemic, non-ischemic and diabetic cardiomyopathy. Manganese-enhanced MRI (MEMRI) provides a unique method to quantify in-vivo myocardial calcium uptake but no studies have so far utilized MEMRI in patients with aortic stenosis (AS). We sought to: 1) determine whether myocardial calcium uptake is perturbed in people with severe AS, and 2) assess change in calcium uptake following aortic valve replacement (AVR).

Methods: In this prospective, pilot, case-control study, adults with severe AS underwent MEMRI before and after AVR. A group of healthy controls were also recruited. The primary outcome was the rate of manganese uptake (Ki) as assessed by Patlak modeling to act as a surrogate of myocardial calcium uptake. Comparison of Ki between groups was adjusted for age, body mass index (BMI) and systolic blood pressure.

Results: Twenty-eight controls and ten subjects with severe AS (age 72 [61-75] years, 8 male, 7 symptomatic, valve area 0.81 [0.74-1.0] cm²) were recruited, with seven returning for repeat scans post-AVR. AS patients had higher BMI and blood pressure, and a greater incidence of hyperlipidemia compared to controls. Baseline left ventricular (LV) volumes were similar between the groups, but the AS patients had higher indexed left ventricular mass. Global longitudinal strain and peak early diastolic strain rate were lower in the AS group. There was no significant difference in Ki between patients with severe AS and controls (7.09 [6.33-8.99] vs. 8.15 [7.54-8.78] mL/100g of tissue/min, P=0.815). Following AVR, there was regression in indexed LV mass (68 [51-79] to 49 [47-65] g/m², P=0.018) and mass-volume ratio (0.94 [0.80-1.13] to 0.74 [0.71-0.82] g/mL, P=0.028) but no change in Ki was seen (7.35 [6.81-8.96] to 7.11 [6.16-8.01] mL/100 g of tissue/min, P=0.499).

Conclusions: Despite clear features of adverse LV remodeling and systolic dysfunction, patients with severe AS demonstrated no alteration in calcium uptake at baseline compared to controls. Moreover, AVR led to reverse LV remodeling but no notable change in calcium uptake was seen. This may suggest that altered myocardial calcium handling does not play a significant pathophysiological role in AS.

Background: Four-dimensional cardiovascular magnetic resonance flow imaging (4D flow CMR) plays an important role in assessing cardiovascular diseases. However, the manual or semi-automatic segmentation of aortic vessel boundaries in 4D flow data introduces variability and limits the reproducibility of aortic hemodynamics visualization and quantitative flow-related parameter computation. This paper explores the potential of deep learning to improve 4D flow CMR segmentation by developing models for automatic segmentation and analyzes the impact of the training data on the generalization of the model across different sites, scanner vendors, sequences, and pathologies.

Methods: The study population consists of 260 4D flow CMR datasets, including subjects without known aortic pathology, healthy volunteers, and patients with bicuspid aortic valve (BAV) examined at different hospitals. The dataset was split to train segmentation models on subsets with different representations of characteristics, such as pathology, gender, age, scanner model, vendor, and field strength. An enhanced three-dimensional U-net convolutional neural network (CNN) architecture with residual units was trained for time-resolved two-dimensional aortic cross-sectional segmentation. Model performance was evaluated using Dice score, Hausdorff distance, and average symmetric surface distance on test data, datasets with characteristics not represented in the training set (model-specific), and an overall evaluation set. Standard diagnostic flow parameters were computed and compared with manual segmentation results using Bland-Altman analysis and interclass correlation.

Results: The representation of technical factors, such as scanner vendor and field strength, in the training dataset had the strongest influence on the overall segmentation performance. Age had a greater impact than gender. Models solely trained on BAV patients' datasets performed well on datasets of healthy subjects but not vice versa.

Conclusion: This study highlights the importance of considering a heterogeneous dataset for the training of widely applicable automatic CNN segmentations in 4D flow CMR, with a particular focus on the inclusion of different pathologies and technical aspects of data acquisition.

Background: There are few meta-analyses examining the prognostic value of right ventricular ejection fraction (RVEF) for a specific type of cardiovascular disease (CVD). The aim of this study was to compare the association of cardiac magnetic resonance (CMR)-derived RVEF with adverse outcomes for several specific types of CVD, using a robust Bayesian model-averaged meta-analysis.

Methods: Three databases were searched for CMR articles reporting hazard ratios (HRs) of RVEF restricted to a specific type of CVD. For each specific type of CVD, Bayesian model-averaged meta-analyses with and without publication bias adjustments were conducted to evaluate the strength of evidence for RVEF according to the Bayes factor (BF).

Results: Among 108 articles (21,166 patients) analyzing 11 CVD types, pooled HR for 5% reduction in RVEF assessed by publication bias-unadjusted, Bayesian model-averaged meta-analysis offered moderate or strong evidence of an association with outcomes for all types of CVD (HR: 1.07-1.37, BF₁₀: 4.3-9.6 * 10⁷). In contrast, a robust Bayesian model-averaged meta-analysis, adjusted for publication bias, found moderate or strong evidence in favor of an association of RVEF with outcomes only in hypertrophic cardiomyopathy (HR: 1.19, 95% credible interval (CrI): 0.98-1.42, BF₁₀: 5.0), dilated cardiomyopathy (HR: 1.16, 95% CrI: 1-1.22, BF₁₀: 23.3), pulmonary hypertension (HR: 1.05, 95% CrI: 1-1.12, BF₁₀: 3.0), and aortic stenosis (HR: 1.15, 95% CrI: 0.97-1.34, BF₁₀: 4.2). There was weak evidence for an association of RVEF with adverse outcomes in seven other CVDs.

Conclusion: In a Bayesian meta-analysis adjusted for publication bias, there was moderate or strong evidence for an association of RVEF with outcomes for only four CVDs. Additional data may strengthen evidence regarding other CVDs.

Background: Left ventricular (LV) diastolic function is a key determinant of cardiac output and impairments of diastolic function can lead to heart failure. Assessment of diastolic function is challenging due to several factors, including the load dependence of ventricular filling. We developed a method using cardiovascular magnetic resonance (CMR) imaging to model the untwisting motion of the LV as a viscoelastic damped oscillator to derive myocardial torsional modulus (µ) and frictional damping characteristics, and hypothesized that the torsional modulus would correlate with invasive measures of LV stiffness.

Methods: Twenty-two participants who underwent invasive left heart catheterization (LHC) and CMR for the evaluation of chest pain were evaluated. µ and damping constants were determined by solving a system of equations using CMR-measured LV geometrical and angular displacement data during diastole. Time constant of pressure decay τ and chamber stiffness β were measured from invasive LHC and CMR-derived volume data as comparison metrics of diastolic function.

Results: µ was correlated with chamber stiffness constant β and time constant of pressure decay τ, derived from invasive measurement (R = 0.78, p < 0.001, and R = 0.51, p = 0.014, respectively). µ was also correlated with pre-A-wave diastolic pressure (0.67, p = 0.001).

Conclusion: We propose a new method to objectively evaluate diastolic relaxation properties of the LV. This method may have promise to replace invasive, catheter-based assessment of diastolic function.

Background: Aortic blood flow characterization by four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) is increasingly performed in aneurysm research. A limited number of studies have established normal values that can aid the recognition of abnormal flow at an early stage. This study aims to establish additional sex-specific and age-dependent reference values for flow-related parameters in a large cohort of healthy adults.

Methods: Two hundred and twelve volunteers were included, and 191 volunteers completed the full study protocol. All underwent 4D flow CMR of the entire aorta. Quantitative values for velocity, vorticity, helicity, as well as total, circumferential, and axial wall shear stress (WSS) were determined for the aortic root (AoR), ascending aorta (AAo), aortic arch, descending aorta (DAo), suprarenal aorta, and infrarenal aorta. Vorticity and helicity were indexed for segment volume (mL).

Results: The normal values were estimated per sex and age group, where significant differences between males (M) and females (F) were found only for specific age groups. More specifically, the following variables were significantly different after applying the false discovery rate correction for multiple testing: 1) velocity in the AAo and DAo in the 60-70 years age group (mean ± SD: (M) 47.0 ± 8.2 cm s^-1 vs (F) 38.4 ± 6.9 cm s^-1, p = 0.001 and, (M) 55.9 ± 9.9 cm s^-1 vs (F) 46.5 ± 5.5 cm s^-1, p = 0.002), 2) normalized vorticity in AoR in the 50-59 years age group ((M) 27,539 ± 5042 s^-1 mL^-1 vs (F) 30,849 ± 7285 s^-1 mL^-1, p = 0.002), 3) axial WSS in the Aao in the 18-29 age group ((M) 1098 ± 203 mPa vs (F) 921 ± 121 mPa, p = 0.002). Good to strong negative correlations with age were seen for almost all variables, in different segments, and for both sexes.

Conclusion: This study describes reference values for aortic flow-related parameters acquired by 4D flow MRI. We observed limited differences between males and females. A negative relationship with age was seen for almost all flow-related parameters and segments.

Background: Fabry disease (FD) is an X-linked inherited lysosomal storage disease that is caused by deficient activity of the enzyme alpha-galactosidase A. Cardiovascular magnetic resonance (CMR) imaging can detect cardiac sphingolipid accumulation using native T1 mapping. The kidneys are often visible in cardiac CMR native T1 maps; however, it is currently unknown if the maps can be used to detect sphingolipid accumulation in the kidneys of FD patients. Therefore, the aim of this study was to evaluate if cardiac-dedicated native T1 maps can be used to detect sphingolipid accumulation in the kidneys.

Methods: FD patients (n = 18, 41 ± 10 years, 44% (8/18) male) and healthy subjects (n = 38, 41 ± 16 years, 47% (18/38) male) were retrospectively enrolled. Native T1 maps were acquired at 1.5T using modified Look-Locker inversion recovery research sequences. Native T1 values were measured by manually delineating regions of interest (ROI) in the renal cortex, renal medulla, heart, spleen, blood, and liver. Endo- and epicardial borders were delineated in the myocardium and averaged across all slices. Blood ROIs were placed in the left ventricular blood pool in the midventricular slice.

Results: There were no differences in native T1 between the FD patients and the healthy subjects in the renal cortex (1034 ± 88 ms vs 1056 ± 59 ms, p = 0.29), blood (1614 ± 111 ms vs 1576 ± 100 ms, p = 0.22), spleen (1143 ± 45 ms vs 1132 ± 70 ms, p = 0.54), or liver (568 ± 49 ms vs 557 ± 47 ms, p = 0.41). Native myocardial T1 was lower in FD patients compared to healthy subjects (951 ± 79 vs 1006 ± 38, p<0.01), and higher in the renal medulla (1635 ± 144 vs 1514 ± 81, p<0.01).

Conclusion: Compared to healthy subjects, patients with FD and cardiac involvement showed no differences in native T1 of the renal cortex. FD patients had higher native T1 in the renal medulla, which is not totally explained by differences in blood native T1 but may reflect a hyperfiltration state in the development of renal failure. The findings suggest that sphingolipid accumulation in the renal cortex in FD patients could not be detected with cardiac-dedicated research native T1 maps.

Background: Right ventricular (RV) dyssynchrony or post systolic contraction (PSC) causes inefficient pumping and has not been investigated as a prognostic marker in pulmonary arterial hypertension (PAH). The objective was to investigate if RV dyssynchrony and PSC are prognostic markers of transplantation-free survival in PAH and if multiple RV views improve prognostication.

Methods: Patients with PAH undergoing cardiovascular magnetic resonance between 2003 and 2021 were included. For strain analysis, endocardial end-diastolic RV contours were delineated in RV three-chamber (RV3ch), four-chamber (4ch), and midventricular short-axis (SAX) slice. RV dyssynchrony was defined as the standard deviation of time to peak strain in the walls from one (4ch), two (4ch and SAX), or three views (4ch, SAX, and RV3ch). PSC was defined as peak strain occurring after pulmonary valve closure. Outcome was defined as death or lung transplantation.

Results: One hundred and one patients (58 ± 19 years, 66% (67/101) women) were included. Median follow-up was 37 [51] months. There were 60 events (55 deaths and 5 lung transplantations). Outcome was associated with RV dyssynchrony from three views and with RV strain in 4ch. An increase in RV dyssynchrony-in three views-by 1% was associated with a 10% increased risk of lung transplantation or death. There was no association between outcome and RV dyssynchrony in one or two views nor with PSC.

Conclusion: RV dyssynchrony in three views was associated with outcome in PAH, whereas assessing dyssynchrony from one or two views and PSC was not. This implies that assessment of multiple instead of single RV views could potentially be used for prognostication in PAH.

Aims: Myocardial inflammation is increasingly detected noninvasively by tissue mapping with cardiovascular magnetic resonance (CMR). Intraindividual agreement with endomyocardial biopsy (EMB) or markers of myocardial injury, high-sensitive cardiac troponin (hs-cTnT) in patients with clinically suspected viral myocarditis is incompletely understood.

Methods: Prospective multicenter study of consecutive patients with clinically suspected myocarditis who underwent blood testing for hs-cTnT, CMR, and EMB as a part of diagnostic workup. EMB was considered positive based on immunohistological criteria in line with the European Society of Cardiology (ESC) definitions. CMR diagnoses employed tissue mapping using sequence-specific cut-off for native T1 and T2 mapping; active inflammation was defined as T1 ≥2 standard deviation (SD) and T2 ≥2 SD above the mean of normal range. Hs-cTnT of greater than 13.9 ng/L was considered significant.

Results: A total of 114 patients (age (mean ± SD) 54 ± 16, 65% males) were included, of which 79 (69%) had positive EMB criteria, 64 (56%) CMR criteria, and a total of 58 (51%) positive troponin. Agreement between EMB and CMR diagnostic criteria was poor (CMR vs ESC: area under the curve (AUC): 0.51 (0.39-0.62)). The agreement between a significant hs-cTnT rise and CMR-based diagnosis of myocarditis was good (AUC: 0.84 (0.68-0.92); p < 0.001), but poor for EMB (0.50 (0.40-0.61). Hs-cTnT was significantly associated with native T1 and T2, high-sensitive C-reactive protein, and N-terminal pro-hormone brain natriuretic peptide (r = 0.37, r = 0.35, r = 0.30, r = 0.25; p < 0.001), but not immunohistochemical criteria or viral presence.

Conclusion: In clinically suspected viral myocarditis, all diagnostic approaches reflect the pathophysiological elements of myocardial inflammation; however, the differing underlying drivers only partially overlap. The EMB and CMR diagnostic algorithms are neither interchangeable in terms of interpretation of myocardial inflammation nor in their relationship with myocardial injury.

Background: To investigate the ability of a delayed respiratory-navigated, electrocardiographically-gated three-dimensional inversion recovery-prepared fast low-angle shot (3D IR FLASH) sequence to evaluate the lower airways in children undergoing routine cardiovascular magnetic resonance (CMR).

Methods: This retrospective study included pediatric patients (0-18 years) who underwent clinical CMR where a delayed 3D IR FLASH sequence was performed between July 2020 and April 2021. The airway image quality and extent of lower airway visibility were graded by two blinded readers using a four-point ordinal scale (0-3). Lower airway anatomical variants and abnormalities were recorded.

Results: One hundred and eighty patients were included with a median age of 11.7 (4.6-15.3) years. Fifty-one of 180 (28%) were under general anesthesia. Overall, the median grading of airway image quality was 3 (2-3) and the extent of lower airway visibility was 3 (3-3). Interrater agreement was almost perfect (κ = 0.867 and κ = 0.956, respectively). Image quality correlated with extent of lower airway visibility (r = 0.62, p < 0.01). Delayed 3D IR FLASH was able to characterize the segmental bronchi in 137/180 (76%) and lobar bronchi in 172/180 (96%) of patients. Lower airway abnormalities were identified in 37/180 (21%) of patients and 33/129 (26%) with congenital heart disease (CHD). Identified abnormalities included tracheobronchial branching anomalies in 6/180 (3%), abnormal tracheobronchial situs in 6/180 (3%), and extrinsic vascular compression in 25/180 (14%).

Conclusion: Delayed 3D IR FLASH has excellent performance for evaluation of the lower airway anatomy and can simultaneously assess for myocardial late gadolinium enhancement. Lower airway abnormalities are not infrequently seen in children undergoing routine CMR for CHD.