RPS 603a - Cardiomyopathies: functional assessment and deep phenotyping

Purpose:

To characterise by radiomic analyses the injured myocardial tissue, shown via late gadolinium enhancement MR imaging, of patients with ischaemic (ISd) and non-ischaemic (nISd) myocardial disease.

Methods and materials:

Patients with ISd and nISd who underwent a contrast-enhanced cardiac MR from January 2018 to May 2019 were included in this retrospective study. One radiologist expert in cardiovascular imaging, blind to the clinical information, applied a 3 mm standardised region of interest (i.e. using the phase-sensitive inversion recovery images) on the scar of the myocardial tissue of each patient, using a 3D slicer. 56 radiomic features belonging to three categories were extracted: first-order statistics (FOS), grey-level co-occurrence matrix (GLCM), and grey-level run length matrix (GLRLM). The student’s t-test was used to compare the two groups of patients for each radiomic feature (p<0.05). The accuracy of the variables showing a statistically significant difference was assessed by receiver operating characteristic curves.

Results:

10 patients with ISd (1 female; mean age±SD 60.6±10.8 yrs) and 11 nISd (3 females, mean age±SD 51.8±21.4 yrs) were examined. For 41 out of the 56 investigated features, a significant difference between the two groups occurred (8 FOS, 17 GLCM, and 16 GLRLM; p<0.05 each). The area under the curve was excellent only for lmc1 (AUC=.982) and good only for minimum and the tenth percentile (respectively AUC=.773 and AUC=.782). All other features showed a low accuracy.

Conclusion:

3 features can be considered as good radiomic markers of ischaemic injury. Radiomic analyses support that ischaemic injuries differ from non-ischaemic lesions, despite both appearing visually very similar on late gadolinium enhanced MR sequence.

Limitations:

Future studies on a larger population are necessary to confirm this finding and its clinical implications.

Ethics committee approval

Ethics committee approval obtained.

Funding:

No funding was received for this work.

Purpose:

Involvement of the right ventricle is associated with a poor prognosis in several cardiomyopathies because of the increased risk for arrhythmias and subsequent sudden cardiac arrest. In Fabry’s disease (FD), the myocardium is affected by the accumulation of sphingolipids. Current diagnostic tests for cardiac involvement in FD focus on the left ventricle. The purpose of this retrospective study was to evaluate right ventricular involvement in FD by CMR.

Methods and materials:

We compared 61 FD patients and 56 healthy volunteers using CMR at 3T. T1 mapping, mass, and global strains were acquired for both ventricles using dedicated software. For group analysis, FD patients were divided by the severity of cardiac involvement into three stages.

Results:

RV mass rose with increasing accumulation of sphingolipids. All feature tracking strain parameters of the right ventricle showed significant differences in comparison with healthy volunteers (p<0.0001). The most altered parameter was global right ventricular longitudinal strain with a mean of -25.3% (± 8.4) for healthy volunteers and -17.1% (±5.8) in FD. Split by the different stages of cardiac involvement, the decrease of strain parameters rose with a higher stage of the disease. Notably, even in stage I, there was already an elevation of RV mass and reduction of LV global longitudinal strain.

Conclusion:

CMR parameters of the right ventricle show significant differences in all disease stages compared to healthy volunteers. RV changes (mass and strain) seem to occur earlier compared to an elevation of left ventricular mass. Therefore, analysis of the RV should be routinely included in the detailed assessment of patients with Fabry’s disease.

Limitations:

A single-centre study.

Ethics committee approval

The study was approved by the local ethics committee (837.196.13/837.477.14).

Funding:

No funding was received for this work.

Purpose:

To evaluate manually measured global longitudinal strain (GLS) in MRI scans as a technique to detect myocardial kinesia impairment in patients affected by hypertrophic cardiomyopathy.

Methods and materials:

We retrospectively evaluated 36 patients who underwent cardiac magnetic resonance imaging (MRI) between 2011 and 2019. The study population has been divided into two groups: the first group (control group) was composed of 16 patients who resulted negative for any detectable pathology and the second group was composed of 20 patients affected by hypertrophic cardiomyopathy (HCM) with preserved ejection fraction. Two radiologists independently measured global longitudinal strain (GLS) by tracing endocardial border lengths in routinely acquired cine images on vertical (VLA) and horizontal long axis (HLA). We calculated the mean values of GLS for each group and Student’s test was used to compare the means as proof that they were significantly different. Possible correlations between GLS and functional parameters were also analysed (e.g. ejection fraction, volumes, and presence of fibrosis).

Results:

The mean GLS values we obtained concerning the first (6 M, mean ejection fraction 63%) and the second group (13 M, mean ejection fraction 72%) were 22,9 ± 2,3 % and 16 ± 2,1%, respectively. The GLS values regarding the second group were significantly lower than the first group (p<0.01). GLS did not correlate with other parameters, such as the presence of fibrosis replacement or left ventricle volumes.

Conclusion:

We demonstrated that manually measured GLS is a useful tool to detect longitudinal kinesia impairment in patients affected by hypertrophic cardiomyopathy. Thus, the parameter we measured allows to rapidly estimate myocardial longitudinal kinesia without using any dedicated software.

Limitations:

A small study population and heterogeneous types of HCM.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To evaluate the feasibility of dual-energy CT (DECT)-based iodine quantification to estimate myocardial extracellular volume fraction (ECV) in patients with and without cardiomyopathy (CM), and assess its ability to distinguish healthy myocardial tissue from cardiomyopathic, with the goal of defining a threshold ECV value for disease detection.

Methods and materials:

10 healthy subjects and 60 patients with CM (mean age 66.4±9.4; 59 males; 40 ischemic and 20 non-ischemic CM) underwent late-iodine-enhanced DECT-imaging. Myocardial iodine maps were obtained using 3-material decomposition. ECV was estimated from haematocrit levels and the iodine maps, using the AHA 16-segment model of the left ventricle. ROC curve analysis was performed with corresponding AUC, along with Youden’s index assessment, to establish a threshold for CM detection.

Results:

The median ECV for ischaemic CM, non-ischaemic CM, and healthy myocardium were 36.9% (32.4-41.1), 38.3% (33.7-43.0), and 25.4% (22.9-27.3), respectively. Healthy myocardium showed significantly lower median ECV values compared to ischaemic and non-ischaemic CM (p<0.001). From Youden’s index analysis, an ECV>29.48% would indicate the presence of CM (sensitivity=90.3; specificity=90.3). The AUC for the ROC curve was 0.950 (p<0.001).

Conclusion:

Our study findings resulted in a promising threshold ECV value that could facilitate the differentiation between healthy and diseased myocardium, and highlights the potential of this DECT methodology to detect cardiomyopathic tissue. The threshold found is not intended to be used as an absolute diagnostic test, but rather as a tool that could provide additional value for the characterisation of myocardial tissue.

Limitations:

A retrospective single-centre study with limited availability of healthy-subjects' contrast-enhanced DECT delayed scans.

Ethics committee approval

Approved by the Institutional Review Board and written informed consent was obtained from all subjects.

Funding:

No funding was received for this work.