Early segmental relaxation abnormalities in hypertrophic cardiomyopathy for differential diagnostic of patients with left ventricular hypertrophy

Deep Learning for Discrimination of Hypertrophic Cardiomyopathy and Hypertensive Heart Disease on MRI Native T1 Maps

BACKGROUND

Native T1 and radiomics were used for hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD) differentiation previously. The current problem is that global native T1 remains modest discrimination performance and radiomics requires feature extraction beforehand. Deep learning (DL) is a promising technique in differential diagnosis. However, its feasibility for discriminating HCM and HHD has not been investigated.

PURPOSE

To examine the feasibility of DL in differentiating HCM and HHD based on T1 images and compare its diagnostic performance with other methods.

STUDY TYPE

Retrospective.

POPULATION

128 HCM patients (men, 75; age, 50 years ± 16) and 59 HHD patients (men, 40; age, 45 years ± 17).

FIELD STRENGTH/SEQUENCE

3.0T; Balanced steady-state free precession, phase-sensitive inversion recovery (PSIR) and multislice native T1 mapping.

ASSESSMENT

Compare HCM and HHD patients baseline data. Myocardial T1 values were extracted from native T1 images. Radiomics was implemented through feature extraction and Extra Trees Classifier. The DL network is ResNet32. Different input including myocardial ring (DL-myo), myocardial ring bounding box (DL-box) and the surrounding tissue without myocardial ring (DL-nomyo) were tested. We evaluate diagnostic performance through AUC of ROC curve.

STATISTICAL TESTS

Accuracy, sensitivity, specificity, ROC, and AUC were calculated. Independent t test, Mann-Whitney U-test and Chi-square test were adopted for HCM and HHD comparison. P < 0.05 was considered statistically significant.

RESULTS

DL-myo, DL-box, and DL-nomyo models showed an AUC (95% confidential interval) of 0.830 (0.702-0.959), 0.766 (0.617-0.915), 0.795 (0.654-0.936) in the testing set. AUC of native T1 and radiomics were 0.545 (0.352-0.738) and 0.800 (0.655-0.944) in the testing set.

DATA CONCLUSION

The DL method based on T1 mapping seems capable of discriminating HCM and HHD. Considering diagnostic performance, the DL network outperformed the native T1 method. Compared with radiomics, DL won an advantage for its high specificity and automated working mode.

LEVEL OF EVIDENCE

4 TECHNICAL EFFICACY STAGE: 2.

Incremental predictive value of the combined use of the neutrophil-to-lymphocyte ratio and systolic blood pressure difference after successful drug-eluting stent implantation

BACKGROUND

Previous work has highlighted the importance of the neutrophil-to-lymphocyte ratio (NLR) and the difference in the ward-to-catheterization laboratory systolic blood pressure (ΔSBP) in prognostic stratification after acute coronary syndrome. However, there is paucity of data regarding the added value of combining these two variables to predict 5-year major clinical outcomes after percutaneous coronary intervention.

METHODS

A total of 1188 patients were classified into four groups according to the NLR andΔSBP (high vs. low) using cutoffs derived from an analysis of receiver operating characteristic curves. A NLR > 3.0 and aΔSBP > 25 mmHg were considered high values. The primary endpoint was the composite of all-cause death, cardiac death, and non-fatal myocardial infarction. The secondary endpoint was the composite of target lesion revascularization, target vessel revascularization, and incidence of cerebrovascular accidents.

RESULTS

The incidence of the primary endpoint was significantly higher in the high NLR and ΔSBP group than in the other three groups (2.2% vs. 4.7% vs. 4.3% vs. 13.2%, p < 0.001). The incidence of the secondary endpoint was similar among the four groups. Incorporation of high NLR and high ΔSBP into a model with conventional and meaningful clinical and procedural risk factors increased the C-statistics in predicting the primary endpoint (0.575 to 0.635, p = 0.002).

CONCLUSIONS

The power to predict the primary endpoint after drug-eluting stent implantation at the 5-year follow-up was improved by combining NLR and ΔSBP.

Association of NT-proBNP and hs-cTnT with Imaging Markers of Diastolic Dysfunction and Focal Myocardial Fibrosis in Hypertrophic Cardiomyopathy

Serum biomarkers such as N-terminal prohormone of the brain natriuretic peptide (NT-proBNP) and cardiac troponins are elevated in patients with hypertrophic cardiomyopathy (HCM). At present, it is not clear if these markers are associated with distinct clinical alterations in HCM, such as left ventricular hypertrophy, outflow tract obstruction, myocardial fibrosis and/or diastolic dysfunction (DD), which are associated with adverse cardiovascular outcome. Here we evaluate the association of NT-proBNP and high sensitivity cardiac troponin T (hs-cTnT) to a variety of cardiac imaging parameters in HCM patients in a multivariable regression analysis. This retrospective cross-sectional study included 366 HCM patients who underwent transthoracic echocardiography (TTE), 218 of whom also obtained cardiovascular magnetic resonance (CMR) to assess focal myocardial fibrosis by LGE. Multivariable regression analyses revealed the strongest association of the DD parameters E/E′ mean and E/E′ septal with NT-proBNP (b = 0.06, 95%-CI [0.05−0.07], p < 0.001, R2 = 0.28; b = 0.08, 95%-CI [0.06−0.1], p < 0.001, R2 = 0.25) and LGE size showed the strongest association with hs-cTnT (b = 0.20, 95%-CI [0.15−0.24], p < 0.001, R2 = 0.28). This study indicates that NT-proBNP and hs-cTnT are associated with structural and functional alterations in HCM. NT-proBNP is a stronger predictor for DD, while hs-cTnT is associated with the extent of focal myocardial fibrosis. Both biomarkers might be useful in the diagnostic procedure in addition to imaging parameters.

Lost in translation: Interpreting cardiac muscle mechanics data in clinical practice

Current inotropic therapies improve systolic function in heart failure patients but also elicit undesirable side effects such as arrhythmias and increased intracellular Ca²⁺ transients. In order to maintain myocyte Ca²⁺ homeostasis, the increased cytosolic Ca²⁺ needs to be actively transported back to sarcoplasmic reticulum leading to depleted ATP reserves. Thus, an emerging approach is to design sarcomere-based treatments to correct impaired contractility via a direct and allosteric modulation of myosin's intrinsic force-generating behavior -a concept that potentially avoids the "off-target" effects. To achieve this goal, various biophysical approaches are utilized to investigate the mechanistic impact of sarcomeric modulators but information derived from diverse approaches is not fully integrated into therapeutic applications. This is in part due to the lack of information that provides a coherent connecting link between biophysical data to in vivo function. Hence, our ability to clearly discern the drug-mediated impact on whole-heart function is diminished. Reducing this translational barrier can significantly accelerate clinical progress related to sarcomere-based therapies by optimizing drug-dosing and treatment duration protocols based on information obtained from biophysical studies. Therefore, we attempt to link biophysical mechanical measurements obtained in isolated cardiac muscle and in vivo contractile function.

Early segmental relaxation abnormalities in hypertrophic cardiomyopathy for differential diagnostic of patients with left ventricular hypertrophy

BACKGROUND

Hypertrophic cardiomyopathy (HCM) is characterized by asymmetric left ventricular hypertrophy (LVH). However, clinical signs can be subtle and differentiation from other causes of LVH is challenging.

HYPOTHESIS

As diastolic dysfunction (DD) is an early sign in HCM, we aimed to find regional changes in relaxation pattern for differentiation from other entities of LVH.

METHODS

In 148 patients (81 HCM, 55 arterial hypertension (AHT), 12 Fabry disease) and 63 healthy controls, relaxation patterns were assessed using regional tissue Doppler imaging. In 42 HCM patients, myocardial mass and fibrosis were quantified by cardiac magnetic resonance imaging and correlated with relaxation parameters.

RESULTS

In HCM the septal to lateral isovolumic relaxation time (s/l IVRT) ratio was higher (1.5 ± 0.4) compared with AHT (1.1 ± 0.2), Fabry disease (1.0 ± 0.1), and controls (1.1 ± 0.2; P < 0.001), showing 77% sensitivity and 79% specificity to discriminate HCM-related LVH from other entities. The s/l IVRT ratio was independent of global DD in HCM (HCM with DD: 1.5 ± 0.5, n = 52; HCM without DD: 1.5 ± 0.3, n = 29) and remained significantly different from other entities in a subgroup of HCM patients with maximum wall thickness < 20 mm (s/l ratio: 1.5 ± 0.5, n = 28). Regional IVRT did not correlate with the corresponding segmental myocardial mass or amount of fibrosis in cardiac magnetic resonance imaging.

CONCLUSIONS

HCM patients show a prolonged septal IVRT irrespective of the extent of LVH and even before developing global DD. The s/l IVRT ratio is significantly higher in HCM compared with AHT or Fabry disease, thus establishing segmental IVRT analysis as a potential parameter for differential diagnosis in LVH.