Left ventricular ejection fraction and Global Longitudinal Strain variability between methodology and experience

A deep multi-stream model for robust prediction of left ventricular ejection fraction in 2D echocardiography

We propose a deep multi-stream model for left ventricular ejection fraction (LVEF) prediction in 2D echocardiographic (2DE) examinations. We use four standard 2DE views as model input, which are automatically selected from the full 2DE examination. The LVEF prediction model processes eight streams of data (images + optical flow) and consists of convolutional neural networks terminated with transformer layers. The model is made robust to missing, misclassified and duplicate views via pre-training, sampling strategies and parameter sharing. The model is trained and evaluated on an existing clinical dataset (12,648 unique examinations) with varying properties in terms of quality, examining physician, and ultrasound system. We report [Formula: see text] and mean absolute error = 4.0% points for the test set. When evaluated on two public benchmarks, the model performs on par or better than all previous attempts on fully automatic LVEF prediction. Code and trained models are available on a public project repository .

Longitudinal Strain Patterns in Stress (Takotsubo) Cardiomyopathy: Evidence of Global Myocardial Injury and Incomplete Recovery

Stress cardiomyopathy develops after abrupt sympathetic stimulation, likely from catecholamine-induced myocardial toxicity. The evolution of myocardial strain during and after an episode have not been previously characterized. We aimed to determine whether preexisting contractile abnormalities may explain the observed regional dysfunction during an acute episode and to investigate the persistence of strain abnormalities after clinical recovery. We identified patients who were diagnosed with stress cardiomyopathy and had an echocardiogram performed before their episode, during their episode, and within 1 year after. The diagnosis was confirmed based on the absence of obstructive coronary lesions. Left ventricular (LV) longitudinal strain was calculated using speckle-tracking software and compared between baseline, episode, and follow-up echocardiograms. The LV strain analysis was performed on 23 patients. The LV ejection fraction was 64 ± 8.7% at baseline, 45 ± 12% during the episode, and 5 9 ± 10% after a median follow-up of 46 days. The LV global longitudinal strain was 24 ± 4.7% at baseline, 11 ± 4.9% during the episode, and 19 ± 4.6% after the follow-up. The mean ejection fraction (p <0.01) and global longitudinal strain (p <0.001) remained below baseline levels at follow-up. Longitudinal strain was reduced (<18%) in 80 ± 23% of myocardial segments during an episode and 41 ± 21% of myocardial segments at follow-up. During the acute episode, 35 ± 6% of the abnormal segments were in the base, outside of the region of ballooning. Our findings suggests that stress cardiomyopathy is associated with global rather than regional myocardial injury and that contractile abnormalities persist after clinical improvement. These findings challenge our previous understanding of stress cardiomyopathy and may guide future pathophysiologic understanding of this complex disease.

Effect of maximum exercise on left ventricular deformation and its correlation with cardiopulmonary exercise capacity in competitive athletes

BACKGROUND

Global longitudinal strain (GLS) and global myocardial work index (GWI) allow early detection of subclinical changes in left ventricular (LV) systolic function. The aim of the study was to investigate the immediate effects of maximum physical exercise by different exercise testing methods on early post exercise LV deformation parameters in competitive athletes and to analyze their correlation with cardiopulmonary exercise capacity.

METHODS

To reach maximum physical exercise, cardiopulmonary exercise testing (CPET) was performed by semi-recumbent ergometer in competitive handball players (n = 13) and by treadmill testing in competitive football players (n = 19). Maximum oxygen uptake (VO2max) indexed to body weight (relative VO2max) was measured in all athletes. Transthoracic echocardiography and blood pressure measurements were performed at rest and 5 min after CPET in all athletes. GLS, GWI and their changes before and after CPET (ΔGLS, ΔGWI) were correlated with (relative) VO2max.

RESULTS

In handball and football players, GLS and GWI did not differ significantly before and after CPET. There were no significant correlations between GLS and relative VO2max, but moderate correlations were found between ΔGWI and relative VO2max in handball (r = 0.631; P = 0.021) and football players (r = 0.592; P = 0.008). Furthermore, handball (46.7 ml/min*kg ± 4.7 ml/min*kg vs. 37.4 ml/min*kg ± 4.2; P = 0.004) and football players (58.3 ml/min*kg ± 3.7 ml/min*kg vs. 49.7 ml/min*kg ± 6.8; P = 0.002) with an increased ΔGWI after CPET showed a significant higher relative VO2max.

CONCLUSION

Maximum physical exercise has an immediate effect on LV deformation, irrespective of the used testing method. The correlation of relative VO2max with ΔGWI in the early post exercise period, identifies ΔGWI as an echocardiographic parameter for characterizing the current individual training status of athletes.

Longitudinal strain analysis for assessment of early cardiotoxicity during anthracycline treatment in childhood sarcoma: A single center experience

BACKGROUND

The growing population of long-term childhood cancer survivors encounter a substantial burden of cardiovascular complications. The highest risk of cardiovascular complications is associated with exposure to anthracyclines and chest radiation. Longitudinal cardiovascular surveillance is recommended for childhood cancer patients; however, the optimal methods and timing are yet to be elucidated.

AIMS

We aimed to investigate the feasibility of different echocardiographic methods to evaluate left ventricular systolic function in retrospective datasets, including left ventricular ejection fraction (LVEF), fractional shortening (FS), global longitudinal strain (GLS) and longitudinal strain (LS) as well as the incidence and timing of subclinical left ventricular dysfunction detected by these methods.

METHODS AND RESULTS

A retrospective longitudinal study was performed with re-analysis of longitudinal echocardiographic data, acquired during treatment and early follow-up, including 41 pediatric sarcoma patients, aged 2.1-17.8 years at diagnosis, treated at Astrid Lindgren Children's Hospital, Stockholm, Sweden, during the period 2010-2021. All patients had received treatment according to protocols including high cumulative doxorubicin equivalent doses (≥250 mg/m2 ). In 68% of all 366 echocardiograms, LS analysis was feasible. Impaired LS values (<17%) was demonstrated in >40%, with concomitant impairment of either LVEF or FS in 20% and combined impairment of both LVEF and FS in <10%. Importantly, there were no cases of abnormal LVEF and FS without concomitant LS impairment.

CONCLUSION

Our findings demonstrate feasibility of LS in a majority of echocardiograms and a high incidence of impaired LS during anthracycline treatment for childhood sarcoma. We propose inclusion of LS in pediatric echocardiographic surveillance protocols.

Clinical Validation of an Artificial Intelligence-Based Tool for Automatic Estimation of Left Ventricular Ejection Fraction and Strain in Echocardiography: Protocol for a Two-Phase Prospective Cohort Study

BACKGROUND

Echocardiography (ECHO) is a type of ultrasonographic procedure for examining the cardiac function and morphology, with functional parameters of the left ventricle (LV), such as the ejection fraction (EF) and global longitudinal strain (GLS), being important indicators. Estimation of LV-EF and LV-GLS is performed either manually or semiautomatically by cardiologists and requires a nonnegligible amount of time, while estimation accuracy depends on scan quality and the clinician's experience in ECHO, leading to considerable measurement variability.

OBJECTIVE

The aim of this study is to externally validate the clinical performance of a trained artificial intelligence (AI)-based tool that automatically estimates LV-EF and LV-GLS from transthoracic ECHO scans and to produce preliminary evidence regarding its utility.

METHODS

This is a prospective cohort study conducted in 2 phases. ECHO scans will be collected from 120 participants referred for ECHO examination based on routine clinical practice in the Hippokration General Hospital, Thessaloniki, Greece. During the first phase, 60 scans will be processed by 15 cardiologists of different experience levels and the AI-based tool to determine whether the latter is noninferior in LV-EF and LV-GLS estimation accuracy (primary outcomes) compared to cardiologists. Secondary outcomes include the time required for estimation and Bland-Altman plots and intraclass correlation coefficients to assess measurement reliability for both the AI and cardiologists. In the second phase, the rest of the scans will be examined by the same cardiologists with and without the AI-based tool to primarily evaluate whether the combination of the cardiologist and the tool is superior in terms of correctness of LV function diagnosis (normal or abnormal) to the cardiologist's routine examination practice, accounting for the cardiologist's level of ECHO experience. Secondary outcomes include time to diagnosis and the system usability scale score. Reference LV-EF and LV-GLS measurements and LV function diagnoses will be provided by a panel of 3 expert cardiologists.

RESULTS

Recruitment started in September 2022, and data collection is ongoing. The results of the first phase are expected to be available by summer 2023, while the study will conclude in May 2024, with the end of the second phase.

CONCLUSIONS

This study will provide external evidence regarding the clinical performance and utility of the AI-based tool based on prospectively collected ECHO scans in the routine clinical setting, thus reflecting real-world clinical scenarios. The study protocol may be useful to investigators conducting similar research.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/44650.

The impact of upright posture on left ventricular deformation in athletes

Besides LV ejection fraction (LVEF), global longitudinal strain (GLS) and global myocardial work index (GWI) are increasingly important for the echocardiographic assessment of left ventricular (LV) function in athletes. Since exercise testing is frequently performed on a treadmill, we investigated the impact of upright posture on GLS and GWI. In 50 male athletes (mean age 25.7 ± 7.3 years) transthoracic echocardiography (TTE) and simultaneous blood pressure measurements were performed in upright and left lateral position. LVEF (59.7 ± 5.3% vs. 61.1 ± 5.5%; P = 0.197) was not affected by athletes' position, whereas GLS (- 11.9 ± 2.3% vs. - 18.1 ± 2.1%; P < 0.001) and GWI (1284 ± 283 mmHg% vs. 1882 ± 247 mmHg%; P < 0.001) were lower in upright posture. Longitudinal strain was most frequently reduced in upright posture in the mid-basal inferior, and/or posterolateral segments. Upright posture has a significant impact on LV deformation with lower GLS, GWI and regional LV strain in upright position. These findings need to be considered when performing echocardiography in athletes.

Assessment of left ventricular function by spatio-temporal image correlation in fetuses with fetal growth restriction

AIM

To compare the evaluation of left ventricular function by spatio-temporal image correlation (STIC) between fetal growth restriction (FGR) fetuses and normal fetuses.

METHODS

Forty-two FGR fetuses and 50 normal fetuses with gestational age ranging from 28 to 35 weeks, were chosen for the study group and control group, respectively. The fetal heart was acquired using the STIC modality, beginning with a four-chamber view. A 7.5-12.5 s acquisition time and 20-35°angle of the acquisition were used for the acquisition. The resulting STIC dataset was saved for offline analysis. Ventricular volumes were measured using the Virtual Organ Computer-aided Analysis (VOCAL) mode, where the observer defines the contours of the ventricle and traces the endocardia. Stroke volume (SV) = end diastolic volume (EDV)-end systolic volume (ESV) and ejection fraction (EF) = SV/EDV × 100%. The data of the two groups were analyzed.

RESULTS

(1) SV increased with fetal growth in both groups and was positively correlated with gestational age (p < .01), whereas EF remained constant throughout gestation and had no correlation with gestational age (p > .05). (2) There was no difference found in EF between the two groups, (p > .05), SV was significantly lower in FGR group than those in the normal group (p < .01).

CONCLUSION

The STIC is a precise method for calculating fetal ventricular volume changes and functions. Reduced SV occurred at the initial stage of fetal deterioration before the discovery of abnormal EF in FGR fetuses, indicating cardiac dysfunction. SV could be a sensitive indicator of cardiac dysfunction. The use of EF to assess fetal cardiac function is not perfect.