Journal of Cardiovascular Magnetic Resonance最新文献

Background: Right heart catheterisation (RHC) with pulmonary capillary wedge pressure (PCWP) assessment is the reference standard for diagnosis of heart failure with preserved ejection fraction (HFpEF), remains however largely underused. Different approaches for non-invasive PCWP calculation have been proposed. However, as left atrial strain (LA Es) and volume index (ESVi) emerge as a key-criteria in HFpEF, we sought to investigate them for PCWP calculation.

Methods: The derivation population consisted of patients referred to RHC and cardiovascular magnetic resonance (CMR) imaging who were enrolled in a prospective monocentric registry. Patients were classified by RHC according to current guideline recommendations. The external validation population consisted of patients included in the HFpEF-Stress trial who underwent exercise-stress RHC and CMR with follow-up after 4 years for hospitalised cardiovascular events. Performance of strain-derived PCWP was compared to a published LA volume (LAV) and LV mass (LVM) derived method.

Results: The derivation population consisted of n=209 patients, n=123 underwent exercise-stress RHC (n=55 without PH, n=72 pre-capillary, n=27 combined post- and pre-capillary pulmonary hypertension (CpcPH), n=15 isolated post-capillary pulmonary hypertension (IpcPH), n=34 exercise and n=6 unclassified PH). Linear regressions models identified the following formulae for functional PCWPrest 10.304-0.095*Es+0.098*ESVi and functional PCWPstress 24.666-0.251*Es+0.056*ESVi calculation. The validation population consisted of n=74 patients (n=15 without, n=5 pre-capillary, n=8 CpcPH, n=10 IpcPH and n=32 exercise PH with n=4 remaining unclassified). Functional PCWPrest (11.8) and RHC-derived PCWPrest (11mmHg) were statistically similar (p=0.285) and showed moderate correlation (r=0.53, p<0.001). Functional PCWPrest (AUC 0.80) and PCWPstress (AUC 0.85) accurately identified HFpEF patients, were superior to LAV/LVM based PCWP (AUC 0.67, p≤0.002) and showed prognostic implications (HR 1.37 (1.16-1.62) and 1.29 (1.14-1.46), p<0.001).

Conclusions: Functional PCWP may aide in the identification of post-capillary involvement in PH and HFpEF superiorly compared to morphology-derived PCWP and shows prognostic implications.

Background: Extracellular volume fraction (ECV) is an independent predictor of mortality in aortic stenosis (AS). ECV can be measured using myocardial T1 maps acquired before and after contrast administration. Standard ECV measurements do not consider the limited rate of water exchange (WX) between cardiomyocytes and the extracellular matrix which can result in underestimated ECV at higher contrast agent concentrations.

Objectives: The objective was to estimate ECV in patients with severe AS before and after surgical aortic valve replacement (AVR) using a 2-site exchange model (2SXM) that also enables estimates of the intracellular lifetime of water (τ_ic; an indicator of the minor diameter of the cardiomyocytes).

Methods: 20 patients (67±6 years) with severe AS, referred for AVR, underwent MRI on a 3 T MR system before and 6 months after AVR. T1 measurements were made using a multiparametric saturation-recovery single-shot acquisition before and at four time points post-injection of contrast agent. A 2SXM and standard linear model (LM) were used to estimate ECV and, when combined with indexed left ventricular mass (LVMI), to calculate cell and matrix volumes, (LVMI × (1-ECV)/1.05) and (LVMI × ECV/1.05), respectively. The 2SXM model was also used to estimate τ_ic.

Results: Data were acquired before and 174 (157 to 267) days after AVR. LVMI reduced following AVR, from 78±15 g/m² to 63±11 g/m² (p<0.001). ECV estimates increased from 22±3% to 28±5% (p<0.001) using the LM compared to 28±5% to 32±4% (p = 0.005) using the 2SXM. Indexed cell volume decreased from 58±12 cm³/m² to 43±9 cm³/m² (p<0.001; LM) and from 54±12 cm³/m² to 41±8 cm³/m² (p<0.001; 2SXM). Indexed matrix volume did not change significantly by either method (LM, 16±4 cm³/m² to 17±3 cm³/m²; 2SXM, 20±5 cm³/m² to 19±3 cm³/m²). τ_ic decreased from 0.21±0.12 s to 0.12±0.09 s (p = 0.007).

Conclusion: Cellular hypertrophy regressed 6 months following AVR; the extracellular matrix volume did not change significantly. τ_ic decreased post-AVR, indicating that the reduction in cell volume can be largely attributed to a reduction in cardiomyocyte diameter.

Background: Free-running (FR) cardiac MRI enables free-breathing ECG-free fully dynamic 5D (3D spatial+cardiac+respiration dimensions) imaging but poses significant challenges for clinical integration due to the volume of data and complexity of image analysis. Existing segmentation methods are tailored to 2D cine or static 3D acquisitions and cannot leverage the unique spatial-temporal wealth of FR data.

Purpose: To develop and validate a deep learning (DL)-based segmentation framework for isotropic 3D+cardiac cycle FR cardiac MRI that enables accurate, fast, and clinically meaningful anatomical and functional analysis.

Methods: Free-running, contrast-free bSSFP acquisitions at 1.5T and contrast-enhanced GRE acquisitions at 3T were used to reconstruct motion-resolved 5D datasets. From these, the end-expiratory respiratory phase was retained to yield fully isotropic 4D datasets. Automatic propagation of a limited set of manual segmentations was used to segment the left and right ventricular blood pool (LVB, RVB) and left ventricular myocardium (LVM) on reformatted short-axis (SAX) end-systolic (ES) and end-diastolic (ED) images. These were used to train a 3D nnU-Net model. Validation was performed using geometric metrics (Dice similarity coefficient [DSC], relative volume difference [RVD]), clinical metrics (ED and ES volumes, ejection fraction [EF]), and physiological consistency metrics (systole-diastole LVM volume mismatch and LV-RV stroke volume agreement). To assess the robustness and flexibility of the approach, we evaluated multiple additional DL training configurations such as using 4D propagation-based data augmentation to incorporate all cardiac phases into training.

Results: The main proposed method achieved automatic segmentation within a minute, delivering high geometric accuracy and consistency (DSC: 0.94 ± 0.01 [LVB], 0.86 ± 0.02 [LVM], 0.92 ± 0.01 [RVB]; RVD: 2.7%, 5.8%, 4.5%). Clinical LV metrics showed excellent agreement (ICC > 0.98 for EDV/ESV/EF, bias < 2mL for EDV/ESV, < 1% for EF), while RV metrics remained clinically reliable (ICC > 0.93 for EDV/ESV/EF, bias < 1mL for EDV/ESV, < 1% for EF) but exhibited wider limits of agreement. Training on all cardiac phases improved temporal coherence, reducing LVM volume mismatch from 4.0% to 2.6%.

Conclusion: This study validates a DL-based method for fast and accurate segmentation of whole-heart free-running 4D cardiac MRI. Robust performance across diverse protocols and evaluation with complementary metrics that match state-of-the-art benchmarks supports its integration into clinical and research workflows, helping to overcome a key barrier to the broader adoption of free-running imaging.

Background: Endomyocardial biopsy (EMB) is the standard invasive method for monitoring acute cardiac allograft rejection (ACAR); however, non-invasive alternatives are increasingly proving to be dependable.

Objectives: We aimed to identify and validate dependable cardiovascular magnetic resonance (CMR) strain indices for ACAR detection.

Methods: We analyzed 160 CMR scans, including long- and short-axis cines, as well as T1/T2 maps from 54 transplant recipients. Uniparametric and multiparametric models integrating left ventricular strain metrics and tissue characteristics were developed to classify histological rejection grades (0, 1 R, ≥2 R) and evaluate therapeutic response.

Results: Regression analysis using generalized linear mixed-models identified significant differences between rejection groups, with global radial strain (GRS) (z-value = 3.1, p = 0.002) and global circumferential strain (GCS) (z-value = 2.5 p<0.008) outperforming global longitudinal strain (GLS) in discriminating ≥2 R from 1 R rejection. Diagnostic performance for detecting ≥2 R rejection was excellent, particularly for GCS (AUC = 0.852, negative predictive value [NPV] = 98.3%) and GRS (AUC = 0.826, NPV = 95.8% (95.8/100)), with enhanced accuracy in the anterolateral mid-basal segments (AUC>0.886, NPV>97.9%). Strain metrics effectively monitored recovery post-therapy for ≥2 R rejection, showing significant improvements (GRS Δ24.5±7.1%, GCS Δ15.9±4.6%, GLS Δ27.4±11.8%, all p<0.02). Furthermore, as strained-based detection of ≥2 R rejection correlated with increases in edema detected using T1/T2 mapping (all p<0.001), integrating strain with T1/T2 mapping significantly enhanced diagnostic accuracy, with T2+GRS (AUC = 0.931, NPV = 98.2) and T1+T2+GCS (AUC = 0.943, NPV = 97.5) as the most effective models.

Conclusion: Segmental CMR strain analysis demonstrates excellent diagnostic accuracy and negative predictive value for detecting high-grade ACAR and monitoring post-therapy recovery. This non-invasive approach, particularly when integrated with multiparametric models combining global strain and tissue mapping, has the potential to reduce reliance on invasive EMBs for ACAR surveillance in cardiac transplant recipients.

Background: Quantitative perfusion cardiovascular magnetic resonance (QP-CMR) allows the generation of pixel-wise myocardial blood flow (MBF) maps using model-based deconvolution with several models including Tofts, modified-Tofts, and Fermi function models. However, the accuracy of pixel-wise MBF mapping has not been fully investigated in humans. The aim of this study was to evaluate the accuracy of advanced QP-CMR using ¹⁵O-water positron emission tomography (PET) as a reference.

Methods: Thirty-nine patients (29 men, 68±11years) with known or suspected coronary artery disease underwent both CMR including stress and rest QP-CMR and ¹⁵O-water PET at a median interval of 13 days. QP-CMR was performed using dual-sequence technique and a single bolus of gadolinium contrast agent during adenosine triphosphate stress and at rest. MBF maps were generated using three different model-based deconvolution techniques as follows: Tofts, modified-Tofts, and Fermi function models. Agreement of MBF and myocardial perfusion reserve (MPR) between QP-CMR and ¹⁵O-water PET was evaluated using Pearson's correlation, Bland-Altman analysis, and intraclass correlation (ICC). The ability of CMR-derived stress MBF and MPR to detect PET-defined abnormal myocardial perfusion (stress MBF ≤2.3 mL/min/g and MPR ≤2.5) was evaluated by receiver operating characteristic (ROC) analysis.

Results: CMR-derived MBF showed a good linear correlation with ¹⁵O-water PET-derived MBF in each of the Tofts, modified-Tofts, and Fermi function models (r = 0.776, 0.752, 0.784, respectively; p<0.001 each) at the patient level. Bland-Altman analysis demonstrated measurement biases for MBF between CMR and ¹⁵O-water PET of 0.31±0.70, 0.05±0.63, and 0.26±0.68 mL/min/g for the Tofts, modified-Tofts, and Fermi function models, respectively. ICCs were 0.734, 0.747, and 0.750, respectively. The area under the ROC curves for stress MBF derived from the Tofts and Fermi function models (0.921 and 0.914, respectively) was significantly higher than that derived from the modified-Tofts model (0.861; p = 0.003 for both). However, there was no significant difference between the Tofts and Fermi function models (p = 0.618).

Conclusion: Advanced QP-CMR using three different model-based deconvolution techniques demonstrated strong agreement with ¹⁵O-water PET. Of these techniques, the Fermi function and Tofts models were more effective in detecting abnormal myocardial perfusion as determined by ¹⁵O-water PET. Considering our results, the model complexity, and its technical availability, the Fermi function model may possess a practical advantage.

Background: Commercial 0.55T low-field magnetic resonance imaging (MRI) systems have recently become available, offering the potential to enhance global accessibility to MRI. T1ρ mapping is an emerging quantitative cardiac MR imaging technique capable of detecting myocardial disease without the need for contrast administration. However, experience with cardiac T1ρ mapping at low-field strength remains limited. This study aims to develop and validate an efficient, free-breathing three-dimensional (3D) high-resolution adiabatic T1ρ mapping sequence for non-contrast whole-heart tissue characterization at 0.55T.

Methods: The proposed 3D T1ρ mapping research sequence acquires four interleaved volumes with different contrast weightings using saturation and adiabatic spin-lock preparation pulses, and a 3-parameter fitting method is used to calculate T1ρ maps. Two-dimensional (2D) image navigators are acquired for non-rigid motion-compensated image reconstruction, enabling 100% respiratory scan efficiency. Phantom and in-vivo experiments in 10 healthy volunteers were conducted to evaluate the accuracy and precision of the proposed 3D sequence in comparison with 2D T1ρ mapping sequences.

Results: Phantom T1ρ values measured using the proposed 3D sequence showed strong agreement with the 2D reference (R² = 0.997), demonstrating high accuracy and reduced sensitivity to heart rate variations. In-vivo experiments in healthy subjects demonstrated that the proposed sequence is feasible for acquiring whole-heart T1ρ maps with 2 mm isotropic resolution in an efficient scan time of 6.6±0.5 min. The mean myocardial T1ρ value obtained with the 3D sequence was slightly higher than that of a conventional 2D breath-hold sequence (112.8±16.7 vs. 106.1±15.1%, p<0.01), while coefficient of variation (CV) was slightly lower (10.2±5.2 vs. 11.4±4.4%, p = 0.02).

Conclusion: The proposed sequence enables 3D free-breathing high-resolution adiabatic T1ρ mapping and shows promising potential for non-contrast whole-heart tissue characterization at 0.55T.

Purpose: Free-running five-dimensional (5D) whole-heart magnetic resonance imaging (MRI) simplifies image acquisition by eliminating the need for external gating, breath-holding, and prospective scan planning. However, it remains vulnerable to patient movement in pediatric populations, which may require sedation or general anesthesia. We present a retrospective motion correction approach using the automatic respiratory and bulk patient motion correction (ACROBATIC) framework to detect, estimate, and correct for bulk motion, thereby improving image quality in pediatric cardiac MRI.

Methods: Free-running Ferumoxytol-enhanced three-dimensional (3D) radial gradient-echo (GRE) data from 210 pediatric patients were manually categorized by the amount of bulk motion within each acquisition, based on retrospective reconstructions. From this cohort, 25 cases with the highest and 25 with the lowest detected bulk motion were selected, forming the moving and reference cohorts, respectively, for subsequent analysis and evaluation of the proposed framework. Respiratory motion was estimated using focused navigation. Bulk motion events were automatically detected from the variation in repeated radial readouts. The data were divided into four-dimensional (4D) arrays with timepoints spanning single respiratory cycles and reconstructed into retrospective real-time images using compressed sensing. Bulk motion was corrected via 3D rigid registration and poorly aligned images were excluded using an outlier-rejection algorithm. Final reconstruction was performed using a previously established 5D cardiac and respiratory motion-resolved compressed sensing approach. ACROBATIC's performance was evaluated by a Dice coefficient (automatic detection), sharpness metrics at the blood-myocardium interface and within the pulmonary vessels, as well as qualitative grading by two expert reviewers.

Results: The ACROBATIC framework successfully differentiated between moving and non-moving patients relative to manual evaluation (Dice = 0.96). Image sharpness significantly improved after motion correction, for analyses of the blood-myocardium interfaces and pulmonary veins. Expert evaluations supported the quantitative findings with average grade improvements of 0.44 and 0.54, respectively for Reviewer 1 and Reviewer 2.

Conclusion: The ACROBATIC framework effectively reduces motion-related artifacts in pediatric cardiac MRI, particularly in patients with significant movement. This method supports the broader goal of achieving high-quality, dynamic whole-heart imaging in children without the need for sedation or general anesthesia.

Background: Cardiovascular magnetic resonance is promising for non-invasive assessment of various cardiac diseases with the ability to provide multi-contrast images, including late gadolinium enhancement (LGE) for myocardial tissue characterization and coronary magnetic resonance angiography (CMRA) for anatomical imaging. However, LGE and CMRA are usually acquired separately in clinical routine with unmatched spatial resolution and slice positions. In this proof of concept study, we aim to achieve a one-stop imaging of 3D gray-blood phase-sensitive inversion recovery (PSIR) LGE and 3D CMRA by proposing a free-breathing simultaneous Gray-Blood and Bright-blOOd phase SensiTive inversion recovery (GB-BOOST) sequence.

Methods: The proposed research sequence acquires two interleaved 3D volumes with inversion recovery and T2 preparation pulses to obtain gray-blood PSIR and CMRA, respectively. Two-dimensional image navigator (iNAV) is performed before the acquisition of each volume to detect respiratory motion, enabling free-breathing acquisition with 100% respiratory scan efficiency. The GB-BOOST framework is compatible with both Dixon gradient echo (GRE) and balanced steady-state free precession (bSSFP) sequences for the application at 3T and 1.5T. In-vivo validation experiments included in total 23 patients for GB-BOOST, which were performed on either a 3T or a 1.5T clinical scanner. The performance of the proposed sequence was compared with clinical 2D gray-blood PSIR and free-breathing 3D CMRA.

Results: GB-BOOST was successfully performed on all 23 patients and was able to efficiently acquire intrinsically co-registered 3D PSIR and CMRA images with 1.2 mm³ resolution in 9.4±1.3 min. Compared with 2D gray-blood PSIR, 3D PSIR GB-BOOST had comparable scar area detection performance without significant differences in image contrast of scar-to-blood (0.42±0.40 vs. 0.30±0.43, p = 0.38), scar-to-myocardium (1.09±0.27 vs. 1.02±0.32, p = 0.30), and blood-to-myocardium (0.67±0.19 vs. 0.72±0.23, p = 0.56). Compared with single-contrast 3D CMRA sequence, 3D T2prep GB-BOOST showed comparable image quality and quantitative vessel metrics of coronary arteries.

Conclusion: The proposed GB-BOOST sequence can achieve simultaneous co-registered 3D whole-heart gray-blood PSIR and CMRA in a single scan with image contrast and image quality comparable with separately acquired images.

Background: Wilson's disease (WD) causes intracellular copper accumulation due to a genetic defect in the copper-transporting protein ATP7B. Cardiac involvement has been reported even in young WD patients; however, pathophysiological mechanisms remain unclear. This study aimed to comprehensively assess the myocardium in WD patients without cardiac symptoms using multiparametric cardiovascular magnetic resonance imaging (CMR), including quantitative stress perfusion mapping and strain analysis.

Methods: WD patients and healthy volunteers underwent multiparametric 1.5T CMR, including cine, native T1, native T2, extracellular volume (ECV), adenosine stress perfusion mapping, and late gadolinium enhancement (LGE) imaging. Left and right ventricle (LV, RV) mass and volumes, global native T1, native T2, ECV, rest and stress perfusion, myocardial perfusion reserve (MPR), strain measures and liver native T1 were compared. LGE images were assessed visually. Disease type and duration, medications, and cardiovascular risk factors were recorded. Symptoms of myocardial ischemia were quantified with Seattle Angina Questionnaire-7.

Results: WD patients (n = 17, 34 [29-55] years, 8/17 (47%) female) and healthy volunteers (n = 17, 33 [29-52] years, 8/17 (47%) female, p = ns for both) were included. There were no differences in cardiovascular risk factors or medications. LV ejection fraction was lower in WD patients (57 [55-61] vs 62 [57-67] %, p = 0.02), and LV global circumferential strain was mildly worse (-18 [-19 to (-17)] vs -20 [-21 to (-18)] %, p = 0.005), otherwise there were no differences in LV or RV mass or function. WD patients had lower stress perfusion and MPR (2.95 [2.74-3.29] vs 3.81 [2.67-4.45] mL/min/g, and 3.3 [3.1-3.8] vs 5.0 [2.9-5.4]), while ECV was higher (29 [28-30] vs 26 [26-29] %), p<0.05 for all, but there were no other differences in multiparametric mapping results. ECV did not correlate with strain parameters. ECV was associated with WD and sex but not age (WD β = 2.58%, male sex β = -0.03%, model R² 0.41, p<0.05 for all). LGE was present in the RV insertion point in 12/17 (71%) of WD patients.

Conclusions: In this study, stable WD patients without apparent cardiac symptoms have early signs of diffuse fibrosis, coronary microvascular dysfunction, and worse systolic function. However, this study is limited by small sample size limiting further subgroup analysis, lack of both longitudinal clinical data and biopsies, not allowing for correlation of CMR findings to histopathology.

Background: Carotid atherosclerosis is a major contributor to the etiology of ischemic stroke. Although intraplaque hemorrhage (IPH) is known to increase stroke risk and plaque burden, its long-term effects on plaque dynamics remain unclear. This study aimed to evaluate the long-term impact of IPH on carotid plaque burden progression using deep learning-based segmentation on multi-contrast magnetic resonance vessel wall imaging (VWI).

Methods: 28 asymptomatic subjects with carotid atherosclerosis underwent an average of 4.7±0.6 VWI scans over 5.8±1.1 years. Deep learning pipelines were used to segment the carotid vessel walls and IPH. Bilateral plaque progression was analyzed using correlation coefficients and generalized estimating equations. Associations between IPH occurrence, IPH volume, and plaque burden (%WV) progression were evaluated using linear mixed-effect models.

Results: IPH was detected in 23/50 of the arteries at any time point. Of arteries without IPH at baseline, 11/39 developed new IPH that persisted, while 5/11 arteries with baseline IPH exhibited it throughout the study. Bilateral plaque growth was significantly correlated (r = 0.54, p<0.001), but this symmetry was weakened in cases with IPH (r = 0.1, p = 0.62). Moreover, IPH presence or development at any point was associated with a 2.3% absolute increase in %WV on average within the affected artery (p<0.001). The volume of IPH was also positively associated with increased %WV (p = 0.005).

Conclusion: Deep learning-based segmentation pipelines were utilized to identify IPH, quantify IPH volume, and measure their effects on carotid plaque burden during long-term follow-up. Findings demonstrated that IPH may persist for extended periods. While arteries without IPH demonstrated minimal progression under contemporary treatment, the presence of IPH and greater IPH volume significantly accelerated long-term plaque growth.