Cardiovascular magnetic resonance feature tracking in small animals - a preliminary study on reproducibility and sample size calculation

Post-Myocardial Infarction Remodeling and Hyperkinetic Remote Myocardium in Sheep Measured by Cardiac MRI Feature Tracking

BACKGROUND

Cardiac MRI feature tracking (FT) allows objective assessment of segmental left ventricular (LV) function following a myocardial infarction (MI), but its utilization in sheep, where interventions can be tested, is lacking.

PURPOSE

To apply and validate FT in a sheep model of MI and describe post-MI LV remodeling.

STUDY TYPE

Animal model, longitudinal.

ANIMAL MODEL

Eighteen lambs (6 months, male, n = 14; female, n = 4; 25.2 ± 4.5 kg).

FIELD STRENGTH/SEQUENCE

Two-dimensional balanced steady-state free precession (bSSFP) and 3D inversion recovery fast low angle shot (IR-FLASH) sequences at 3 T.

ASSESSMENT

Seven lambs underwent test-retest imaging to assess FT interstudy reproducibility. MI was induced in the remaining 11 by coronary ligation with MRI being undertaken before and 15 days post-MI. Injury size was measured by late gadolinium enhancement (LGE) and LV volumes, LV mass, ejection fraction (LVEF), and wall thickness (LVWT) were measured, with FT measures of global and segmental radial, circumferential, and longitudinal strain.

STATISTICAL TESTS

Sampling variability, inter-study, intra and interobserver reproducibility were assessed using Pearson's correlation, Bland-Altman analyses, and intra-class correlation coefficients (ICC). Diagnostic performance of segmental strain to predict LGE was assessed using receiver operating characteristic curve analysis. Significant differences were considered P < 0.05.

RESULTS

Inter-study reproducibility of FT was overall good to excellent, with global strain being more reproducible than segmental strain (ICC = 0.89-0.98 vs. 0.77-0.96). MI (4.0 ± 3.7% LV mass) led to LV remodeling, as evident by significantly increased LV volumes and LV mass, and significantly decreased LVWT in injured regions, while LVEF was preserved (54.9 ± 6.9% vs. 55.6 ± 5.7%; P = 0.778). Segmental circumferential strain (CS) correlated most strongly with LGE. Basal and mid- CS increased significantly, while apical CS significantly decreased post-MI.

DATA CONCLUSION

FT is reproducible and compensation by hyperkinetic remote myocardium may manifest as overall preserved global LV function.

EVIDENCE LEVEL

N/A TECHNICAL EFFICACY: Stage 2.

Cardiovascular magnetic resonance imaging for sequential assessment of cardiac fibrosis in mice: technical advancements and reverse translation

Cardiovascular magnetic resonance (CMR) imaging has become an essential technique for the assessment of cardiac function and morphology, and is now routinely used to monitor disease progression and intervention efficacy in the clinic. Cardiac fibrosis is a common characteristic of numerous cardiovascular diseases and often precedes cardiac dysfunction and heart failure. Hence, the detection of cardiac fibrosis is important for both early diagnosis and the provision of guidance for interventions/therapies. Experimental mouse models with genetically and/or surgically induced disease have been widely used to understand mechanisms underlying cardiac fibrosis and to assess new treatment strategies. Improving the appropriate applications of CMR to mouse studies of cardiac fibrosis has the potential to generate new knowledge, and more accurately examine the safety and efficacy of antifibrotic therapies. In this review, we provide 1) a brief overview of different types of cardiac fibrosis, 2) general background on magnetic resonance imaging (MRI), 3) a summary of different CMR techniques used in mice for the assessment of cardiac fibrosis including experimental and technical considerations (contrast agents and pulse sequences), and 4) provide an overview of mouse studies that have serially monitored cardiac fibrosis during disease progression and/or therapeutic interventions. Clinically established CMR protocols have advanced mouse CMR for the detection of cardiac fibrosis, and there is hope that discovery studies in mice will identify new antifibrotic therapies for patients, highlighting the value of both reverse translation and bench-to-bedside research.

Application of Magnetic Resonance Strain Analysis Using Feature Tracking in a Myocardial Infarction Model

This study validates the usefulness of myocardial strain analysis with cardiac cine magnetic resonance imaging (MRI) by evaluating the changes in the cardiac function and myocardial strain values longitudinally in a myocardial disease model. Six eight-week-old male Wistar rats were used as a model of myocardial infarction (MI). Cine images were taken in the short axis, two-chamber view longitudinal axis, and four-chamber view longitudinal axis directions in rats 3 and 9 days after MI and in control rats, with preclinical 7-T MRI. The control images and the images on days 3 and 9 were evaluated by measuring the ventricular ejection fraction (EF) and the strain values in the circumferential (CS), radial (RS), and longitudinal directions (LS). The CS decreased significantly 3 days after MI, but there was no difference between the images on days 3 and 9. The two-chamber view LS was -9.7 ± 2.1% at 3 days and -13.9 ± 1.4% at 9 days after MI. The four-chamber view LS was -9.9 ± 1.5% at 3 days and -11.9 ± 1.3% at 9 days after MI. Both the two- and four-chamber LS values were significantly decreased 3 days after MI. Myocardial strain analysis is, therefore, useful for assessing the pathophysiology of MI.

Reproducibility of Systolic Strain in Mice Using Cardiac Magnetic Resonance Feature Tracking of Black-Blood Cine Images

PURPOSE

Mouse models are widely utilized to enhance our understanding of cardiac disease. The goal of this study is to investigate the reproducibility of strain parameters that were measured in mice using cardiac magnetic resonance (CMR) feature-tracking (CMR42, Canada).

METHODS

We retrospectively analyzed black-blood CMR datasets from thirteen C57BL/6 B6.SJL-CD45.1 mice (N = 10 female, N = 3 male) that were imaged previously. The circumferential, longitudinal, and radial (Ecc, Ell, and Err, respectively) parameters of strain were measured in the mid-ventricular region of the left ventricle. Intraobserver and interobserver reproducibility were assessed for both the end-systolic (ES) and peak strain.

RESULTS

The ES strain had larger intraclass correlation coefficient (ICC) values when compared to peak strain, for both the intraobserver and interobserver reproducibility studies. Specifically, the intraobserver study showed excellent reproducibility for all three ES strain parameters, namely, Ecc (ICC 0.95, 95% CI 0.83-0.98), Ell (ICC 0.90, 95% CI 0.59-0.97), and Err (ICC 0.92, 95% CI 0.73-0.97). This was also the case for the interobserver study, namely, Ecc (ICC 0.92, 95% CI 0.60-0.98), Ell (ICC 0.76, 95% CI 0.33-0.93), and Err (ICC 0.93, 95% CI 0.68-0.98). Additionally, the coefficient of variation values were all < 10%.

CONCLUSION

The results of this preliminary study showed excellent reproducibility for all ES strain parameters, with good to excellent reproducibility for the peak strain parameters. Moreover, all ES strain parameters had larger ICC values than the peak strain. In general, these results imply that feature-tracking with CMR42 software and black-blood cine images can be reliably used to assess strain patterns in mice.

Low-dose dobutamine cardiovascular magnetic resonance segmental strain study of early phase of intramyocardial hemorrhage rats

Background

This study investigates the segmental myocardial strain of the early phase of intramyocardial hemorrhage (IMH) caused by reperfused myocardial infarction (MI) in rats by low-dose dobutamine (LDD) cardiovascular magnetic resonance (CMR) feature-tracking.

Methods

Nine sham rats and nine rats with 60-min myocardial ischemia followed by 48-h reperfusion were investigated using CMR, including T2*-mapping sequence and fast imaging with steady-state precession (FISP)–cine sequence. Another FISP–cine sequence was acquired after 2 min of dobutamine injection; the MI, IMH, and Non-MI (NMI) areas were identified. The values of peak radial strains (PRS) and peak circumferential strains (PCS) of the MI, IMH and NMI segments were acquired. The efficiency of PRS and PCS (EPRS and EPCS, respectively) were calculated on the basis of the time of every single heartbeat.

Results

The PRS, PCS, EPRS, and EPCS of the sham group increased after LDD injection. However, the PRS, PCS, EPRS, and EPCS of the IMH segment did not increase. Moreover, the PRS and PCS of the MI and NMI segments did not increase, but the EPRS and EPCS of these segments increased. The PRS, PCS, EPRS, and EPCS of the IMH segment were lower than those of the MI and NMI segments before and after LDD injection, but without a significant difference between MI segment and NMI segment before and after LDD injection.

Conclusions

LDD could help assess dysfunctions in segments with IMH, especially using the efficiency of strain. IMH was a crucial factor that decreased segmental movement and reserved function.

Pulmonary arterial banding in mice may be a suitable model for studies on ventricular mechanics in pediatric pulmonary arterial hypertension

BACKGROUND

The role of interventricular mechanics in pediatric pulmonary arterial hypertension (PAH) and its relation to right ventricular (RV) dysfunction has been largely overlooked. Here, we characterize the impact of maintained pressure overload in the RV-pulmonary artery (PA) axis on myocardial strain and left ventricular (LV) mechanics in pediatric PAH patients in comparison to a preclinical PA-banding (PAB) mouse model. We hypothesize that the PAB mouse model mimics important aspects of interventricular mechanics of pediatric PAH and may be beneficial as a surrogate model for some longitudinal and interventional studies not possible in children.

METHODS

Balanced steady-state free precession (bSSFP) cardiovascular magnetic resonance (CMR) images of 18 PAH and 17 healthy (control) pediatric subjects were retrospectively analyzed using CMR feature-tracking (FT) software to compute measurements of myocardial strain. Furthermore, myocardial tagged-CMR images were also analyzed for each subject using harmonic phase flow analysis to derive LV torsion rate. Within 48 h of CMR, PAH patients underwent right heart catheterization (RHC) for measurement of PA/RV pressures, and to compute RV end-systolic elastance (RV_Ees, a measure of load-independent contractility). Surgical PAB was performed on mice to induce RV pressure overload and myocardial remodeling. bSSFP-CMR, tagged CMR, and intra-cardiac catheterization were performed on 12 PAB and 9 control mice (Sham) 7 weeks after surgery with identical post-processing as in the aforementioned patient studies. RV_Ees was assessed via the single beat method.

RESULTS

LV torsion rate was significantly reduced under hypertensive conditions in both PAB mice (p = 0.004) and pediatric PAH patients (p < 0.001). This decrease in LV torsion rate correlated significantly with a decrease in RV_Ees in PAB (r = 0.91, p = 0.05) and PAH subjects (r = 0.51, p = 0.04). In order to compare combined metrics of LV torsion rate and strain parameters principal component analysis (PCA) was used. PCA revealed grouping of PAH patients with PAB mice and control subjects with Sham mice. Similar to LV torsion rate, LV global peak circumferential, radial, and longitudinal strain were significantly (p < 0.05) reduced under hypertensive conditions in both PAB mice and children with PAH.

CONCLUSIONS

The PAB mouse model resembles PAH-associated myocardial mechanics and may provide a potential model to study mechanisms of RV/LV interdependency.

Quantification of Biventricular Myocardial Strain Using CMR Feature Tracking: Reproducibility in Small Animals

Myocardial strain is a well-validated parameter for evaluating myocardial contraction. Cardiovascular magnetic resonance myocardial feature tracking (CMR-FT) is a novel method for the quantitative measurements of myocardial strain from routine cine acquisitions. In this study, we investigated the influence of temporal resolution on tracking accuracy of CMR-FT and the intraobserver, interobserver, and interstudy reproducibilities for biventricular strain analysis in mice from self-gated CMR at 11.7 T. 12 constitutive nexilin knockout (Nexn-KO) mice, heterozygous (Het, N = 6) and wild-type (WT, N = 6), were measured with a well-established self-gating sequence twice within two weeks. CMR-FT measures of biventricular global and segmental strain parameters were derived. Interstudy, intraobserver, and interobserver reproducibilities were investigated. For the assessment of the impact of the temporal resolution for the outcome in CMR-FT, highly oversampled semi-4 chamber and midventricular short-axis data were acquired and reconstructed with 10 to 80 phases per cardiac cycle. A generally reduced biventricular myocardial strain was observed in Nexn-KO Het mice. Excellent intraobserver and interobserver reproducibility was achieved in all global strains (ICC range from 0.76 to 0.99), where global right ventricle circumferential strain (RCSSAX) showed an only good interobserver reproducibility (ICC 0.65, 0.11-0.89). For interstudy reproducibility, left ventricle longitudinal strain (LLSLAX) was the most reproducible measure of strain (ICC 0.90, 0.71-0.97). The left ventricle radial strain (LRSSAX) (ICC 0.50, 0.10-0.83) showed fair reproducibility and RCSSAX (ICC 0.36, 0.14-0.74) showed only poor reproducibility. In general, compared with global strains, the segmental strains showed relatively lower reproducibility. A minimal temporal resolution of 20 phases per cardiac cycle appeared sufficient for CMR-FT strain analysis. The analysis of myocardial strain from high-resolution self-gated cine images by CMR-FT provides a highly reproducible method for assessing myocardial contraction in small rodent animals. Especially, global LV longitudinal and circumferential strain revealed excellent reproducibility of intra- and interobserver and interstudy measurements.

Serelaxin Improves Regional Myocardial Function in Experimental Heart Failure: An In Vivo Cardiac Magnetic Resonance Study

Background Animal studies demonstrated that serelaxin lessens fibrosis in heart failure. This study assessed its effect on myocardial deformation using cardiac magnetic resonance and elucidated its relationship to gene regulation and histology in a mouse heart failure model. Methods and Results C57BL/6J mice were subjected to SHAM (n=4) or transverse aortic constriction (TAC). At week 10, TAC mice were randomized to receive either serelaxin (0.5 mg/kg per day; n=11) or vehicle (n=13) for 4 weeks. Cardiac magnetic resonance imaging was performed at baseline and repeated at the end of the study (week 14). Cine images were used to calculate left ventricular (LV) global longitudinal, circumferential, and radial strain. Hearts were examined for histology and gene expression. Compared with SHAM, mice 10 weeks after TAC showed increased LV mass with significant decreases in LV deformation parameters, indicating subclinical deterioration of myocardial function. At week 14, TAC mice given serelaxin demonstrated significant improvements in all LV strain parameters and no decrease in LV stroke volume and ejection fraction compared with TAC mice given vehicle. A significant positive correlation between global circumferential strain and the extent of myocardial fibrosis was found, and global circumferential strain correlated significantly with the expression of heart failure genes in serelaxin-treated mice. Conclusions Serelaxin improved cardiac magnetic resonance-derived myocardial deformation parameters as well as histomorphometric and gene expression findings in mice with heart failure. Cardiac magnetic resonance-derived myocardial mechanics correlate with histology and gene expression, stressing its utilization in myocardial remodeling.

Cardiovascular magnetic resonance feature tracking in pigs: a reproducibility and sample size calculation study

Cardiovascular magnetic resonance feature tracking (CMR-FT) is a novel technique for non-invasive assessment of myocardial motion and deformation. Although CMR-FT is standardized in humans, literature on comparative analysis from animal models is scarce. In this study, we measured the reproducibility of global strain under various inotropic states and the sample size needed to test its relative changes in pigs. Ten anesthetized healthy Landrace pigs were investigated. After baseline (BL), two further steps were performed: (I) dobutamine-induced hyper-contractility (Dob) and (II) verapamil-induced hypocontractility (Ver). Global longitudinal (GLS), circumferential (GCS) and radial strain (GRS) were assessed. This study shows a good to excellent inter- and intra-observer reproducibility of CMR-FT in pigs under various inotropic states. The highest inter-observer reproducibility was observed for GLS at both BL (ICC 0.88) and Ver (ICC 0.79). According to the sample size calculation for GLS, a small number of animals could be used for future trials.

Effect of comprehensive initial training on the variability of left ventricular measures using fast-SENC cardiac magnetic resonance imaging

Cardiac magnetic resonance (CMR) is becoming the imaging modality of choice in multicenter studies where highly reproducible measurements are necessary. The purpose of this study was to assess the effect of comprehensive initial training on reproducibility of quantitative left ventricular (LV) parameters estimated using strain-encoded (SENC) imaging. Thirty participants (10 patients with heart failure (HF) and preserved LV ejection fraction (HFpEF), 10 patients with HF and reduced LV ejection fraction (HFrEF) and 10 healthy volunteers) were examined using fast-SENC imaging. Four observers with different experience in non-invasive cardiac imaging completed comprehensive initial training course and were invited to perform CMR data analysis. To assess agreement between observers, LV volumes, mass, ejection fraction (LVEF), global longitudinal strain (GLS) and global circumferential strain (GCS) were estimated using dedicated software (MyoStrain, USA). To test intraobserver agreement data analysis was repeated after 4 weeks. SENC imaging and analysis were fast and were completed in less than 5 minutes. LV end-diastolic volume index (LVEDVi), LVEF and strain were significantly lower in HFpEF patients than in healthy volunteers (p = 0.019 for LVEDVi; p = 0.023 for LVEF; p = 0.004 for GLS and p < 0.001 for GCS). All LV functional parameters were further reduced in HFrEF. Excellent interobserver agreement was found for all LV parameters independently of the level of experience. The reproducibility of LV mass was lower, especially at the intraobserver level (ICC 0.91; 95% CI 0.74–0.96). LV volumetric and functional parameters derived using fast-SENC imaging, are highly reproducible. The appropriate initial training is relevant and allows to achieve highest concordance in fast-SENC measurements.

Correlation between left ventricular myocardial strain and left ventricular geometry in healthy adults: a cardiovascular magnetic resonance-feature tracking study

This study was aimed to investigate the correlation between left ventricular (LV) myocardial strain and LV geometry in healthy adults using cardiovascular magnetic resonance-feature tracking (CMR-FT). 124 gender-matched healthy adults who underwent healthy checkup using CMR cine imaging were retrospectively analyzed. Peak global radial, circumferential, longitudinal strain (GRS, GCS and GLS) for left ventricle were measured. LV geometry was assessed by the ratio of LV mass (LVM) and end-diastolic volume (EDV). GRS, GCS and GLS were 34.18 ± 6.71%, - 22.17 ± 2.28%, - 14.76 ± 2.39% for men, and 33.40 ± 6.95%, - 22.49 ± 2.27%, - 15.72 ± 2.36% for women. Multiple linear regression showed that LVM/EDV was associated with decreased GLS (β = - 0.297, p = 0.005), but was not significantly associated with GRS and GCS (both p > 0.05). There was an increase in the magnitude of GRS, GCS and GLS with advancing age (β = 0.254, β = 0.466 and β = 0.313, all p < 0.05). Greater BMI was associated with decreased GRS, GCS and GLS (β = - 0.232, β = - 0. 249 and β = - 0.279, all p < 0.05). In conclusion, compared with GRS and GCS, GLS is more sensitive to assess LV concentric remodeling in healthy adults. GRS, GCS and GLS are all independently positively associated with age and negatively associated with BMI. Sex-based LV strain reference values for healthy Chinese adults are established.

Validation of simple measures of aortic distensibility based on standard 4-chamber cine CMR: a new approach for clinical studies

Objective

Aortic distensibility (AD) represents a well-established parameter of aortic stiffness. It remains unclear, however, whether AD can be obtained with high reproducibility in standard 4-chamber cine CMR images of the descending aorta. This study investigated the intra- and inter-observer agreement of AD based on different angles of the aorta and provided a sample size calculation of AD for future trials.

Methods

Thirty-one patients underwent CMR. Angulation of the descending aorta was performed to obtain strictly transversal and orthogonal cross-sectional aortic areas. AD was obtained both area and diameter based.

Results

For area-based values, inter-observer agreement was highest for 4-chamber AD (ICC 0.97; 95% CI 0.93–99), followed by orthogonal AD (ICC 0.96; 95% CI 0.91–98) and transversal AD (ICC 0.93; 95% CI 0.80–97). For diameter-based values, agreement was also highest for 4-chamber AD (ICC 0.97; 95% CI 0.94–99), followed by orthogonal AD (ICC 0.96; 95% CI 0.92–98) and transversal AD (ICC 0.91; 95% CI 0.77–96). Bland–Altman plots confirmed a small variation among observers. Sample size calculation showed a sample size of 12 patients to detect a change in 4-chamber AD of 1 × 10⁻³ mmHg⁻¹ with either the area or diameter approach.

Conclusion

AD measurements are highly reproducible and allow an accurate and rapid assessment of arterial compliance from standard 4-chamber cine CMR.

Graphic abstract

Electronic supplementary material

The online version of this article (10.1007/s00392-019-01525-8) contains supplementary material, which is available to authorized users.

Evaluation of a commercial multi-dimensional echocardiography technique for ventricular volumetry in small animals

BACKGROUND

The assessment of ventricular volumes using conventional echocardiography methods is limited with regards to the need of geometrical assumptions. In the present study, we aimed to evaluate a novel commercial system for three-dimensional echocardiography (3DE) in preclinical models by direct comparison with conventional 1D- and 2D-echocardiography (1DE; 2DE) and the gold-standard technique magnetic resonance imaging (MRI). Further, we provide a standard operating protocol for image acquisition and analysis with 3DE.

METHODS

3DE was carried out using a 30 MHz center frequency transducer coupled to a Vevo®3100 Imaging System. We evaluated under different experimental conditions: 1) in vitro phantom measurements served as controlled setting in which boundaries were clearly delineated; 2) a validation cohort composed of healthy C57BL/6 J mice and New Zealand Obese (NZO) mice was used in order to validate 3DE against cardiac MRI; 3) a standard mouse model of pressure overload induced-heart failure was investigated to estimate the value of 3DE.

RESULTS

First, in vitro volumetry revealed good agreement between 3DE assessed volumes and the MRI-assessed volumes. Second, cardiac volume determination with 3DE showed smaller mean differences compared to cardiac MRI than conventional 1DE and 2DE. Third, 3DE was suitable to detect reduced ejection fractions in heart failure mice. Fourth, inter- and intra-observer variability of 3DE showed good to excellent agreement regarding absolute volumes in healthy mice, whereas agreement rates for the relative metrics ejection fraction and stroke volume demonstrated good to moderate observer variabilities.

CONCLUSIONS

3DE provides a novel method for accurate volumetry in small animals without the need for spatial assumptions, demonstrating a technique for an improved analysis of ventricular function. Further validation work and highly standardized image analyses are required to increase reproducibility of this approach.

Guidelines for measuring cardiac physiology in mice

Cardiovascular disease is a leading cause of death, and translational research is needed to understand better mechanisms whereby the left ventricle responds to injury. Mouse models of heart disease have provided valuable insights into mechanisms that occur during cardiac aging and in response to a variety of pathologies. The assessment of cardiovascular physiological responses to injury or insult is an important and necessary component of this research. With increasing consideration for rigor and reproducibility, the goal of this guidelines review is to provide best-practice information regarding how to measure accurately cardiac physiology in animal models. In this article, we define guidelines for the measurement of cardiac physiology in mice, as the most commonly used animal model in cardiovascular research.

Listen to this article’s corresponding podcast at http://ajpheart.podbean.com/e/guidelines-for-measuring-cardiac-physiology-in-mice/.