Value of layer-specific strain distribution patterns in hypertrophied myocardium from different etiologies

Evaluation of left ventricular function of pregnant women with autoimmune diseases using speckle tracking echocardiography

Pregnancy can cause fluctuations in autoimmune diseases (AD) women, which may aggravate the cardiac damage. Maternal heart function is very important for maternal and fetal health. Therefore, early and accurate evaluation of the heart function of AD pregnant women is necessary. This study was aimed to evaluate the left ventricular (LV) function of AD pregnant women using two-dimensional speckle tracking echocardiography (2DSTE). A total of 96 subjects, including 26 non-pregnant AD patients (AD group), 33 AD pregnant women (AD-P group), and 37 healthy pregnant women (H-P group) were conducted. Baseline clinical and conventional echocardiography characteristics of all the subjects were collected. The 2DSTE was performed to acquire layer-specific strain parameters of LV. Compared with H-P group, AD-P group showed no significant differences in GLSmid and GLSepi. However, the GLSendo (24.10 [22.30 to 25.40] vs. 21.70 [19.05 to 25.15], P = 0.023) and ΔGLS (5.50 [4.80 to 6.00] vs. 4.90 [4.20 to 5.80], P = 0.017) were decreased, while the PSD (27 [23 to 32] vs. 32 [24 to 44], P = 0.014) was increased. At the segmental level, there was no significant difference in apex LSmid and LSepi between the two groups, while the AD-P group showed transmural dysfunction in basal and middle segments, and the LSendo in apex segments (32.84 [28.34 to 34.25] vs. 27.97 [21.87 to 33.61], P = 0.021) were significantly decreased. Compared with AD group, AD-P group showed no significant difference in ΔGLS, PSD, and GLS parameters of three layers. For the segmental level, there were no significant differences in the LSepi of the apex segment and the LS in three layers of the basal and middle segments between the two groups, while LSendo (32.69 [29.13 to 35.53] vs. 27.97 [21.87 to 33.60], P = 0.017) and LSmid (24.70 [22.24 to 27.78] vs. 21.32 [16.91 to 26.11], P = 0.023) in apex segments were significantly lower in AD-P group. The anti-SSA/Ro antibody were positive independently correlated with PSD. In conclusion, layer-specific strain parameters and PSD by 2DSTE provide an accurate and reproducible measurement of myocardial function. There are subclinical LV myocardial dysfunction in AD pregnant women. Besides, the positive of anti-SSA/Ro antibody maybe associated with LV myocardial dysfunction.

Endocardial versus whole-myocardial tracking global longitudinal strain analysis in patients with hypertrophic cardiomyopathy: A preliminary comparative study

BACKGROUND AND OBJECTIVES

We investigated whether the feasibility of left ventricular (LV) global longitudinal strain (GLS) in hypertrophic cardiomyopathy (HCM) varies according to the methodology (e.g. endocardial vs. whole myocardial tracking techniques).

METHODS

We retrospectively analyzed 111 consecutive patients with HCM (median age, 58 years; male, 68.5%) who underwent both transthoracic echocardiography (TTE) and cardiac magnetic resonance imaging (apical 29.7%, septal 33.3%, and diffuse or mixed 37.0%). TTE-whole myocardial and TTE-endocardial GLS were measured and compared in terms of association with late gadolinium enhancement (LGE) extent and discrimination performance for extensive LGE (>15% of the LV myocardium).

RESULTS

Although TTE-whole myocardial and TTE-endocardial GLS were significantly correlated, absolute TTE-endocardial GLS values (19.3 [16.2-21.9] %) were higher than TTE-whole myocardial GLS values (13.3[10.9-15.6] %, p<0.001). Both TTE-derived GLS parameters were significantly correlated with the LGE extent and independently associated with extensive LGE (odds ratio [OR] 1.30, p = 0.022; and OR 1.24, p = 0.013, respectively). Discrimination performance for extensive LGE was comparable between TTE-whole myocardial and TTE-endocardial GLS (area under the curve [AUC], 0.747 and 0.754, respectively, pdifference = 0.610). However, among patients with higher LV mass index (>70 g/m2), only TTE-whole myocardial GLS correlated with LGE extent and was independently associated with extensive LGE (OR 1.35, p = 0.042), while TTE-endocardial GLS did not. Additionally, TTE-whole myocardial GLS had better discrimination performance for extensive LGE than TTE-endocardial GLS (AUC, 0.705 and 0.668, respectively, pdifference = 0.006).

CONCLUSION

TTE-derived GLS using either the endocardial or whole myocardial tracking technique is feasible in patients with HCM. However, in those with severe hypertrophy, TTE-whole myocardial GLS is better than TTE-endocardial GLS.

Layer-specific strain echocardiography may reflect regional myocardial impairment in patients with hypertrophic cardiomyopathy

Our study aimed to determine whether layer-specific strain (LSS) could reflect regional myocardial impairment in patients with hypertrophic cardiomyopathy (HCM). The study enrolled 50 patients with HCM and 30 age-matched healthy controls. Transmural gradient of longitudinal strain (TGLS), defined as the difference between the longitudinal strain of the endocardium and epicardium in a left ventricular segment, was used to reflect layer-specific myocardial impairment. Negative TGLS was consistently observed in healthy controls. The TGLS was relatively consistent within the basal, middle, and apical levels in healthy controls,but showed a significant gradient from the base towards the apex. In patients with HCM, the hypertrophic segments had significantly higher TGLS than the relatively normal segments or healthy controls at all 3 levels (0.14 % ± 3.48 % vs. -2.65 % ± 4.44 % vs. -2.17 % ± 1.66 % for basal, - 0.72 % ± 3.71 % vs. -4.02 % ± 4.00 % vs. -3.58 % ± 2.29 % for middle, and - 8.69 % ± 7.96 % vs. -11.44 % ± 6.65 % vs. -10.04 % ± 3.20 % for apex). Abnormal TGLS, defined as positive TGLS, in patients with HCM was associated with chest pain. In receiver operating characteristic curve analysis, a large area of abnormal TGLS (> 4 segments) had moderate accuracy for predicting chest pain (sensitivity, 73.3 %; specificity, 70.0 %). TGLS, a novel LSS derived parameter, may reflect regional myocardial impairment in patients with HCM.

Myocardial contractility recovery following acute pressure unloading after transcatheter aortic valve intervention (TAVI) in patients with severe aortic stenosis and different left ventricular geometry: a multilayer longitudinal strain echocardiographicanalysis

Aim of the present study was to describe the left ventricular longitudinal strain (LS) in all myocardial layers in patients with severe aortic stenosis (AS), preserved left ventricular ejection fraction (LVEF) in different LV geometry and to compare LS analysis before and early after acute LV unloading provided by transcatheter aortic valve implantation (TAVI). 68 patients were enrolled. LS was measured from the endocardial layer (Endo-LS), epicardial layer (Epi-LS) and full thickness of myocardium (Transmural-LS) before and after TAVI. Patients were divided in two groups accordingly with relative wall thickness (RWT): concentric LV hypertrophy (cLVH) vs eccentric LV hypertrophy (eLVH). Less impaired values of LS at baseline were observed, in all layers, in patients with cLVHas compared to patients with eLVH (Endo-LS was - 13.2 ± 2 vs - 11.1±3 %, p = 0.041; Epi-LS was - 11.8 ± 1.8 vs - 9.9 ± 3 %, p = 0.043; Transmural-LS was - 12.3 ± 1.8 vs - 10.49 ± 3.3 %, p = 0.02, respectively). A significant improvement in endocardial LS (Endo-LS) after TAVI was detected only in cLVH(- 13 ± 2 vs - 14 ± 2, p = 0.011). Our findings documented that concentric LVH had better basal strain function and showed a better myocardial recovery after TAVI compared to eLVH.

Automated Recognition of Regional Wall Motion Abnormalities Through Deep Neural Network Interpretation of Transthoracic Echocardiography

BACKGROUND

Automated interpretation of echocardiography by deep neural networks could support clinical reporting and improve efficiency. Whereas previous studies have evaluated spatial relationships using still frame images, we aimed to train and test a deep neural network for video analysis by combining spatial and temporal information, to automate the recognition of left ventricular regional wall motion abnormalities.

METHODS

We collected a series of transthoracic echocardiography examinations performed between July 2017 and April 2018 in 2 tertiary care hospitals. Regional wall abnormalities were defined by experienced physiologists and confirmed by trained cardiologists. First, we developed a 3-dimensional convolutional neural network model for view selection ensuring stringent image quality control. Second, a U-net model segmented images to annotate the location of each left ventricular wall. Third, a final 3-dimensional convolutional neural network model evaluated echocardiographic videos from 4 standard views, before and after segmentation, and calculated a wall motion abnormality confidence level (0-1) for each segment. To evaluate model stability, we performed 5-fold cross-validation and external validation.

RESULTS

In a series of 10 638 echocardiograms, our view selection model identified 6454 (61%) examinations with sufficient image quality in all standard views. In this training set, 2740 frames were annotated to develop the segmentation model, which achieved a Dice similarity coefficient of 0.756. External validation was performed in 1756 examinations from an independent hospital. A regional wall motion abnormality was observed in 8.9% and 4.9% in the training and external validation datasets, respectively. The final model recognized regional wall motion abnormalities in the cross-validation and external validation datasets with an area under the receiver operating characteristic curve of 0.912 (95% CI, 0.896-0.928) and 0.891 (95% CI, 0.834-0.948), respectively. In the external validation dataset, the sensitivity was 81.8% (95% CI, 73.8%-88.2%), and specificity was 81.6% (95% CI, 80.4%-82.8%).

CONCLUSIONS

In echocardiographic examinations of sufficient image quality, it is feasible for deep neural networks to automate the recognition of regional wall motion abnormalities using temporal and spatial information from moving images. Further investigation is required to optimize model performance and evaluate clinical applications.