Journal of Cardiovascular Magnetic Resonance最新文献

Background: Structural and functional abnormalities of the left atrium (LA) predict adverse outcomes such as heart failure and mortality in many patients with heart disease. However, the effect of anthracyclines on LA structural and functional abnormalities remains incompletely characterized. Further, atorvastatin prevented the anthracycline-associated decline in the left ventricular ejection fraction; however, whether atorvastatin protects against anthracycline-associated impairment of LA structure and function is currently unknown.

Methods: In the STOP-CA randomized clinical trial, participants with lymphoma treated with anthracyclines were randomized to placebo (n=150) or atorvastatin (n=150) for 12 months. In post hoc analyses, CMR-derived LA volumetric and functional measurements (reservoir [GLS], conduit, and booster strain) were measured at baseline and 12 months using feature tracking (FT). The primary endpoint was the difference in the proportion of participants with a ≥1 SD decrease in LA GLS between the atorvastatin and placebo groups. The secondary endpoint was a ≥20% relative decrease in LA GLS. Other exploratory endpoints included volume indices and emptying fractions.

Results: Of 300 participants, 158 (mean age 51±16years, 48% female, 83 with atorvastatin) had paired CMR-derived LA strain and volumetric data at baseline and follow-up. Both groups had similar baseline characteristics and cancer treatment. All LA strain and volumetric measures were similar between the two groups at baseline. Among the placebo group, LA GLS decreased from baseline to follow-up (35.5±8.8 vs 32.4±8.2%, p=0.007). A ≥1 SD absolute decrease in LA GLS (8.8% units) was observed among 24% with atorvastatin and 28% with placebo (p=0.59). Similarly, a ≥20% relative decrease in GLS was observed in 25% vs 31% (p=0.48). Participants over 50 had an almost 10% (9.9%, 95% confidence interval: -18.75, -1.12) greater relative decrease in LA GLS with anthracyclines. There were no differences between cardiac hospitalization rates with a ≥1 SD absolute decrease (5% vs 8%, p=0.72) in LA GLS at 24 months. Among other indices of LA structure and function, the LA total emptying fraction also decreased from baseline to follow-up, with no differences between groups at follow-up.

Conclusion: Atorvastatin did not attenuate the decline in CMR-derived LA GLS among lymphoma patients undergoing anthracycline-based chemotherapy.

Clinical trial registration: NCT02943590; https://clinicaltrials.gov/study/NCT02943590.

Background: Cardiac diffusion tensor imaging (cDTI) is an emerging technique for microstructural characterization of the heart and has shown clinical potential in a range of cardiomyopathies. However, there is substantial variation reported for in vivo cDTI results across the literature, and sensitivity of cDTI to differences in imaging sites, scanners, acquisition protocols, and post-processing methods remains incompletely understood.

Methods: SIGNET is a prospective multi-center, observational study in traveling and non-traveling healthy volunteers. The study was initiated by the executive board of the Society of Cardiovascular Magnetic Resonance (SCMR) Cardiac Diffusion Special Interest Group (SIG) as a follow-up to a previous multi-center study on phantom validation of cardiac DTI and a recently published SCMR consensus statement on cardiac diffusion MRI. The study has been developed by the Project Management Committee in consultation with the SCMR cardiac diffusion SIG, which includes international experts in cardiac diffusion MRI. To date, more than 20 international institutions have engaged with the study, including sites that are new to cardiac DTI, making this the largest collaborative effort in the field.

Discussion: SIGNET will provide important information about the key sources of variation in cardiac DTI. This will help rationalize strategies for addressing and minimizing such variation. Harmonization of protocols in this and future studies will underpin efforts to translate cardiac DTI for clinical application.

Background: Identifying the cause of dyspnea (i.e., cardiac vs. non-cardiac) can be challenging in the absence of significant resting cardiac abnormalities. Exercise cardiovascular magnetic resonance (Ex-CMR) enables quantification of cardiac volumetric indices under physiological stress. Using Ex-CMR, we sought to develop a non-invasive imaging marker, referred to as the myocardial dynamic index (MDI), and to demonstrate its potential for evaluating cardiac dyspnea.

Methods: MDI is a metric derived from Ex-CMR work-volume loop model that integrates rest and stress left ventricular (LV) end-diastolic and end-systolic volumes with workload measured during supine exercise, while accounting for body size and LV mass. To evaluate MDI as a marker of cardiac dyspnea, we retrospectively analyzed data from a prospective multicenter study measuring MDI in patients with cardiac or non-cardiac dyspnea. All had invasive exercise testing before Ex-CMR. Cardiac dyspnea was defined by established invasive and non-invasive criteria, including HFpEF (early to advanced) and HFmrEF. Non-cardiac dyspnea patients had normal invasive hemodynamics and cardiac function. Univariable and multivariable logistic regression identified clinical and imaging predictors of cardiac dyspnea. A base model incorporating clinical and rest CMR variables was compared to a model that included the base model plus MDI. Diagnostic performance was assessed using receiver operating characteristic analysis and compared using the DeLong test. MDI scan/re-scan reproducibility over one year, inter- and intra-observer reproducibility, and correlation with VO₂ max were evaluated.

Results: Among 93 patients (66 with cardiac dyspnea, 27 with non-cardiac dyspnea), MDI was lower in patients with cardiac dyspnea (25.9±9.5 vs. 45.1±10.7 mL·W/g/m², p<0.0001). The base model included age, body mass index, NYHA class, and left atrial strain. In multivariable analysis, MDI emerged as the only independent predictor of cardiac dyspnea when added to the base model. Inclusion of MDI improved the AUC from 0.86 to 0.93 (p=0.012), while MDI alone yielded an AUC of 0.91. A strong correlation was observed between MDI and the VO₂ max index (r=0.84, p<0.0001). Reproducibility was excellent.

Conclusion: Ex-CMR MDI is independently associated with cardiac dyspnea and strongly correlates with the VO₂ max index. It aids in differentiating cardiac from non-cardiac dyspnea and provides incremental diagnostic value beyond conventional clinical and resting imaging parameters.

Background: Although balanced steady-state free-precession (bSSFP) cines provide excellent contrast for morpho-functional cardiac evaluation, the fluctuating myocardial cine signal intensity (mcSI) is rarely used diagnostically. These mcSI fluctuations were related to through-plane motion but the impact of this motion remains unclear. We aim to characterize the mid-ventricular pre- and postcontrast bSSFP cyclic mcSI fluctuations in healthy subjects and compare these to Bloch simulations incorporating through-plane motion.

Methods: Retrospectively-gated mid-ventricular short-axis cine bSSFP images from healthy subjects (n = 49) acquired at 1.5T pre- and early postcontrast were analyzed. First, the mcSI fluctuations during the heart cycle were determined and their timing compared to the radial myocardial motion. Next, pre- vs postcontrast differences were determined during systole, early-diastole, and late-diastole. Finally, Bloch simulations and acquisitions in a moving T1 phantom were performed to analyze the through-plane motion effect on the bSSFP and spoiled gradient echo (SGRE) mcSI.

Results: The bSSFP mcSI showed a three-peak pattern both pre- and postcontrast, corresponding to the contraction and relaxation phases. However, the mcSI peaks showed a time lag vs the times of maximum radial velocity that was larger for the systolic contraction than for the early or late-diastolic relaxation phases. In addition, the shape and amplitude of the systolic and early diastolic mcSI peaks changed significantly post- vs precontrast. Bloch simulations showed an in-vivo-like (regional) three-peak signal profile and similar changes for post- vs precontrast T1 levels. Finally, results in the moving phantom and accompanying simulations confirmed a slice-thickness-dependent time lag between the motion and mcSI profile in both bSSFP and SGRE.

Conclusion: In healthy subjects before and after contrast, the bSSFP mcSI variation during the heart cycle is characterized by a three-peak pattern associated with the contraction and relaxation phases. However, the delays in timing of these peaks vs the myocardial motion, as well as the differences between pre- and postcontrast, vary with the stage of the heart cycle. Bloch simulations suggest that these mcSI fluctuations are largely determined by the regional through-slice motion. A better understanding of these motion-induced contrast mechanisms may be beneficial to methods exploiting bSSFP mcSI.

Background: Late gadolinium enhancement imaging is the cornerstone of tissue characterization via cardiac magnetic resonance imaging. The contrast-enhancing effect of gadolinium is caused by a linear increase in tissue longitudinal R1 relaxation rates (R1 = 1/T1). The change in R1 of blood pre- and post-contrast (ΔR1_blood) is therefore a surrogate for the blood-pool gadolinium concentration, which in turn correlates linearly to the tissue gadolinium concentration. The total volume of distribution for gadolinium is the extracellular volume of the body, which differs with body composition, potentially leading to variations in blood-pool and tissue gadolinium concentrations.

Methods: This study is a hypothesis-generating secondary analysis of a dataset of 1098 patients who underwent contrast cardiovascular magnetic resonance between August 2014 and November 2020 at a tertiary center. ΔR1_blood was calculated from T1 relaxation time maps acquired before and approximately 15 min after application of 0.15 mmol/kg gadobutrol. Explorative data analysis and multiple linear regression were performed to assess the influence of body mass index (BMI), gender, age, cardiac index (CI), hematocrit (Hct), and left ventricular end-diastolic volume index (LVEDVi) on ΔR1_blood.

Results: In bivariate analysis, ΔR1_blood showed moderate correlation to BMI and weak correlation to LVEDVi, Hct, and CI. The correlation to BMI was higher in women (r = 0.52 at 1.5T and r = 0.47 at 3T) than in men (r = 0.27 at 1.5T and r = 0.37 at 3T). Multiple linear regression showed independent predictive value of BMI, BMI:gender, gender, CI, field strength (FS), and LVEDVi (R² = 0.268, P < 0.001), with BMI remaining the strongest individual predictor (b = 0.032 [0.025; 0.040], η² = 0.13, P < 0.001).

Conclusion: ΔR1_blood, a measurement of gadolinium contrast enhancement in the blood-pool and a surrogate of plasma C_Gd at the time of late enhancement imaging, showed moderate association with BMI, FS, and gender and weak association with LVEDVi and CI. Further research is necessary to assess the need for individualized gadolinium dosing.

Background: The use of parametric T1 mapping and T2 mapping cardiovascular magnetic resonance (CMR) in takotsubo cardiomyopathy has shown elevated native T1 and T2 relaxation times. In addition to native T1 and T2 mapping, a new native image parametric mapping method using T1 relaxation in the rotating frame (T1ρ) has shown potential to assess myocardial tissue characterization. This study aims to compare T1ρ with native T1 and T2 myocardial mapping in takotsubo cardiomyopathy.

Methods: T1ρ, T2, and native T1 relaxation times were obtained for 51 patients (96% (49/51) female, mean age 69) diagnosed with takotsubo cardiomyopathy and 16 healthy subjects (100% (16/16) female, mean age 41). The baseline scan for the takotsubo cohort was done within 3 weeks after symptom onset, with follow-up scans carried out on average 9 weeks after the baseline scan. Cardiac function and T1ρ, T2, native T1 maps of basal, mid, and apical segments were analyzed.

Results: A significant increase in T1ρ relaxation time was measured in mid and apical segments for the takotsubo baseline cohort compared to takotsubo follow-up cohort (p = 0.0006, p = 0.0011, respectively). A significant increase in T1ρ relaxation time was measured in mid and apical segments for the takotsubo baseline cohort compared to the healthy volunteer cohort (p < 0.0001, p < 0.0001, respectively). Significantly elevated T2 and native T1 relaxation were observed in basal (p = 0.0344, p = 0.0109, respectively), mid (p < 0.0001, p < 0.0001, respectively), and apical (p < 0.0001, p < 0.0001, respectively) segments for takotsubo baseline scans when compared to the takotsubo follow-up cohort. Significant increase in T2 and native T1 relaxation values was also observed in basal (p = 0.0038, p < 0.0001, respectively), mid (p < 0.0001, p < 0.0001, respectively), and apical (p < 0.0001, p < 0.0001, respectively) segments for takotsubo baseline cohort when compared to the healthy volunteer cohort.

Conclusion: In patients with takotsubo cardiomyopathy, T1ρ values were significantly elevated in the mid and apical segments, where edema is more pronounced. In contrast, both T2 and native T1 values were significantly increased across all three segments-basal, mid, and apical. Consequently, native T1 and T2 mapping showed superior ability to detect edema compared to T1ρ mapping.

Background: Understanding the influence of cardiovascular risk factors on longitudinal cardiac remodeling requires three-dimensional analysis of longitudinal shape changes beyond scalar indicators such as mass and volumes. The aim of this study is to determine trajectories of cardiovascular risk factor-related remodeling in a large cohort imaging study.

Methods: We examined 2521 participants (54% female, aged 60±9 years) of the multi-ethnic study of atherosclerosis (MESA) at baseline and after 10years. Myocardial remodeling was assessed by longitudinal left ventricular shape trajectories derived from cardiac magnetic resonance imaging using a statistical shape atlas. Penalized logistic regression was used to examine the associations between trajectory scores and cardiovascular risk factors, after adjustment for sex and age at baseline. Multivariate regression was used to determine independent shape changes associated with each risk factor.

Results: Between baseline and follow-up, there was a higher prevalence of hypertension (18.4%), antihypertensive medication usage (21.6%), statin usage, and treated diabetes mellitus (8.9%); all p<0.05. Longitudinal shape trajectory scores had stronger associations with obesity, high blood pressure, hypertension medication, and diabetes mellitus, than mass and volume changes (p<0.05). Multivariate regression showed independent longitudinal changes in wall thickening with obesity (13% increase), smoking (11% decrease), and high systolic blood pressure (5.6% increase), with distinct regional variations.

Conclusion: Trajectories of cardiovascular risk factor-related longitudinal remodeling can be examined using shape atlases. In addition to global changes, each risk factor is associated with a distinct regional remodeling of the myocardium.

Background: Although a recently developed wideband perfusion sequence has shown diagnostically acceptable image quality and accurate myocardial blood flow (MBF) quantification at rest in patients with cardiac implanted electronic devices, its performance during vasodilator stress remains unproven. This study aims to determine whether the sequence produces diagnostically acceptable image quality during stress and is capable of quantitatively detecting abnormal stress MBF and myocardial perfusion reserve (MPR) in patients with implanted cardiodefibrillators (ICDs).

Methods: We enrolled 29 patients with an ICD (mean age=63±15years, 17 males, 12 females) and 11 control patients (mean age=50±17years, 6 males, 5 females; negative coronary artery disease; negative stress perfusion CMR; and no cardiac event one year post CMR) with an ICD taped below the left clavicle to mimic image artifacts. Both groups underwent imaging using a six-fold accelerated wideband perfusion sequence during adenosine stress and at rest. Images were reconstructed using a compressed sensing framework. Two clinical readers independently graded the following three categories on a 5-point Likert scale (1: worst, 3: clinically acceptable, 5: best): conspicuity of wall enhancement, noise, and artifact. Pixel-wise stress-rest MBF maps were quantified for both global and segmental analysis. MPR was calculated as the ratio of mean stress to rest MBFs.

Results: The median summed visual score was above the acceptable cut-point (>9.0) and not significantly different between the two groups. Both mean global and segmental stress MBF and MPR were significantly lower (p<0.05) in the ICD patient group (global MBF=1.79±0.50 mL/g/min; global MPR=2.11±0.53) compared to the control group (global MBF=2.92±0.52 mL/g/min; global MPR=3.28±0.57), while rest MBF showed no significant difference (global MBF=0.88±0.18 mL/g/min in the patient group vs. 0.92±0.13 mL/g/min in the control group).

Conclusion: This study demonstrates the feasibility of using a six-fold accelerated wideband perfusion pulse sequence, which provides diagnostically acceptable image quality during stress and is sensitive for detecting abnormal stress MBF and MPR in patients with ICDs.