Deep Learning-Based Electrocardiogram Analysis Predicts Biventricular Dysfunction and Dilation in Congenital Heart Disease

Using Deep-Learning Algorithms to Simultaneously Identify Right and Left Ventricular Dysfunction From the Electrocardiogram

To analyze myocardial fibrosis in dilated cardiomyopathy (DCM) patients with no late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR) using T1 mapping and extracellular volume (ECV) and investigate the potential correlation with left ventricular (LV) dilation and dysfunction. The study included 41 DCM patients without LGE and 79 healthy controls. T1 and ECV were compared between the two groups using multivariable logistic regression analysis. The correlations between histological and functional parameters were evaluated using Pearson's correlation. Mean native myocardial T1 and ECV were significantly higher in the DCM group compared to controls (p ≤ 0.001, respectively). Multivariable logistic regression revealed that ECV (mean, minimum), LV ejection fraction (LVEF), and LV end-diastolic diameter (LVEDD) were independent discriminators for LGE-negative DCM; the area under the curve (AUC) of LVEF, LVEDD, ECV mean, and ECV minimum were 0.97, 0.96, 0.88, and 0.68, respectively. In the DCM group, LVEDD and LVEF were positively and negatively correlated with ECV, respectively. LVEDV index and LV end-systolic volume (LVESV) index were positively correlated with native-T1 and ECV, and the absolute value of LV global strain had a negative correlation with ECV. Early myocardial fibrosis in DCM could be detected by prolonged native T1 and elevated ECV despite the absence of LGE on CMR. Moreover, the change of histological characteristics of myocardium in DCM was correlated with LV dilation and dysfunction. • At an early stage, patients with DCM may have myocardial fibrosis despite the absence of LGE. • T1 mapping and ECV are efficient methods for early detection of potential myocardial fibrosis. • Increased native T1 and ECV are correlated with left ventricular dilation and dysfunction in LGE-negative DCM patients.

Ambulatory simultaneous venoarterial extracorporeal membrane oxygenation and temporary percutaneous left ventricular assist device bridge to heart transplantation

In adults with valvular aortic stenosis (AS), valve replacement is recommended in the presence of symptoms and severely reduced aortic valve area (AVA).1 In such patients, valve replacement improves symptoms and survival, even in the setting of left ventricular (LV) dysfunction. LV dysfunction in severe AS is usually due to afterload mismatch; valve replacement relieves the afterload excess imposed by the stenotic valve and improves LV performance. 2 However, a subset of patients with “ severe” AS, LV dysfunction, and low-transvalvular gradient has been reported to have a relatively high operative mortality and poor prognosis.2–4⇓⇓ This clinical scenario has been termed “low-flow, low-gradient AS.” Accurate assessment of AVA in such patients is difficult because (1) calculated AVA is directly proportional to forward stroke volume, and (2) the Gorlin constant varies at low-flow states. 5–7⇓⇓ Some patients with low-flow, low-gradient AS have a reduced AVA as a result of inadequate forward stroke volume rather than anatomic stenosis, a situation analogous to reduced anterior mitral leaflet excursion in dilated cardiomyopathy where there is not enough forward flow to fully open the valve. For example, Cannon et al 8 showed that some patients with low-gradient AS were found to have only mild AS at surgery despite a Gorlin AVA indicating critical AS. Obviously, surgical therapy is unlikely to benefit such patients because their primary pathology is a cardiomyopathy. On the other hand, patients with severe anatomic AS may benefit from valve replacement despite the increased operative risk associated with a low-flow, low-gradient hemodynamic state. The recent American College of Cardiology/American Heart Association (ACC/AHA) guidelines for managing valvular heart disease recommends hemodynamic evaluation of low-flow, low-gradient AS using dobutamine echocardiography to distinguish patients with fixed anatomic AS from those with flow-dependent (“ relative”) AS in patients with LV …

Ventricular dysfunction and dilation in severe sepsis and septic shock: Relation to endothelial function and mortality

To assess the diagnostic accuracy of an automated algorithm to detect left ventricular (LV) dilatation on non-ECG gated CT, using cardiac magnetic resonance (CMR) as reference standard. Consecutive patients with contrast-enhanced CT thorax and CMR within 31 days (2016-2020) were analysed (n = 84). LV dilatation was defined against age-, sex- and body surface area-specific values for CMR. CTs underwent automated artificial intelligence(AI)-derived analysis that segmented ventricular chambers, presenting maximal LV diameter and volume. Area under the receiver operator curve (AUC-ROC) analysis identified CT thresholds with ≥90% sensitivity and highest specificity and ≥90% specificity with highest sensitivity. Youden's Index was used to identify thresholds with optimised sensitivity and specificity. Automated diameter analysis was feasible in 92% of cases (77/84; 45 men, age 61 ± 14 years, mean CT to CMR interval 10 ± 8 days). Relative to CMR as a reference standard, 45% had LV dilatation. In males, an automated LV diameter measurement of ≥55.5 mm was ≥90% specific for CMR-defined LV dilatation (positive predictive value (PPV) 85.7%, negative predictive value (NPV) 61.2%, accuracy 68.9%). In females, an LV diameter of ≥49.7 mm was ≥90% specific for CMR-defined LV dilatation (PPV 66.7%, NPV 73.1%, accuracy 71.9%). AI CT volumetry data did not significantly improve AUC performance. Fully automated AI-derived analysis LV dilatation on routine unselected non-gated contrast-enhanced CT thorax studies is feasible. We have defined thresholds for the detection of LV dilatation on CT relative to CMR, which could be used to routinely screen for dilated cardiomyopathy at the time of CT. We show, for the first time, that a fully-automated AI-derived analysis of maximal LV chamber axial diameter on non-ECG-gated thoracic CT is feasible in unselected real-world cases and that the derived measures can predict LV dilatation relative to cardiac magnetic resonance imaging, the non-invasive reference standard for determining cardiac chamber size. We have derived sex-specific cut-off values to screen for LV dilatation on routine contrast-enhanced thoracic CT. Future work should validate these thresholds and determine if technology can alter clinical outcomes in a cost-effective manner.

Introduction: Secondary mitral regurgitation (SMR) is related with poor clinical outcome. Impact of right ventricular (RV) dysfunction with or without left ventricular (LV) dysfunction in SMR patients remains unclear. Hypothesis: This study aimed to investigate the prevalence of isolated LV, RV, and biventricular dysfunction, and to determine their prognostic implication in moderate-to-severe SMR without surgical and interventional treatment. Methods: A total of 1090 consecutive moderate-to-severe SMR patients without surgical and interventional treatment hospitalized in our hospital center between 2009 and 2018 (aged 75±12years, 60.4% male) were included. Clinical and echocardiographic characteristics were analyzed. All patients completed at least 1-year clinical follow-up. The primary endpoint was defined as all-cause death. Results: Mean left ventricular ejection fraction (LVEF) was 44.6±16.2%, and 59.3% of patients had LVEF <50% (defined as LV dysfunction). RV dysfunction was defined as a reduced tricuspid annular plane excursion (TAPSE<17mm) or an increased systolic pulmonary artery disease (sPAP>40mmHg). All-cause mortality was 47.8% [follow-up period: median 23 (8-40) months]. Patients were divided into 4 subgroups: 1) preserved biventricular function: n=136 (12.5%); 2) isolated LV dysfunction: n=97 (8.9%); 3) isolated RV dysfunction: n=308 (28.3%); 4) biventricular dysfunction: n=549 (50.4%). The mortality in above groups was 27.2%, 36.1%, 50.0% and 53.7%, respectively. Multivariable survival analysis showed that isolated LV dysfunction, isolated RV dysfunction, or biventricular dysfunction were all independently associated with increased all-cause mortality, after adjustment for age, sex and other clinical covariates associated with mortality. Conclusions: Compared to patients with preserved biventricular function or isolated LV dysfunction, RV and biventricular dysfunction are associated with significantly higher all-cause mortality in SMR patients without surgical/interventional treatment. This might constitute important implications for planning and timing of therapy.

Cardiac magnetic resonance (CMR)-identified late gadolinium enhancement (LGE), representing regional fibrosis, is often used to predict ventricular arrhythmia risk in nonischemic cardiomyopathy (NICM). However, LGE is more closely correlated with sustained monomorphic ventricular tachycardia (SMVT) than ventricular fibrillation (VF). We characterized CMR findings of ventricular LGE in VF survivors. We examined consecutively resuscitated VF survivors undergoing contrast-enhanced 1.5T CMR between 9/2007 and 7/2016. We excluded coronary artery disease, hypertrophic cardiomyopathy, amyloid, sarcoid, arrhythmogenic right ventricular cardiomyopathy, and channelopathy. Preexisting implantable cardioverter-defibrillator (ICD) was a CMR contraindication. VF patients were divided into three groups: (1) NICM, (2) left ventricular (LV) dilatation with normal LV ejection fraction (LVEF), and (3) normal LV size and LVEF. Two groups of NICM patients with and without SMVT were examined for comparison. We analyzed 87VF patients, and found that LGE was seen in 8/22 (36%) with NICM (LVEF 38±11%, LV end-diastolic volume index [LVEDVI] 134±68mL/BSA), 11/40 (28%) with LV dilatation and normal LVEF (LVEDVI 103±17mL/BSA), 4/25 (16%) with normal LV size and LVEF. Incidence of LGE in NICM patients without prior ventricular tachycardia/VF (LVEF 36±12%, LVEDVI 141±46mL/body surface area [BSA]) was 117/277 and was not lower than those with VF and NICM (42%vs 36%; P=0.59). By contrast, 22/37 NICM patients with SMVT (LVEF 42±11%, LVEDVI 123±48mL/BSA) were LGE-positive (59% NICM-SMVT vs 36% NICM-VF; P=0.04). Most VF survivors with a diagnosis of NICM did not have LGE on CMR and would not have met primary prevention ICD criteria based on LVEF. Absence of LGE may not portend a benign prognosis in NICM. Novel strategies for determining SCD risk in this cohort are required.

Funding Acknowledgements Type of funding sources: None. Introduction Since the SARS-CoV2 break-out many reports focused on the role of cardiovascular system involvement in the patient’s outcome(1). However, few descriptions exist including only patients with severe ARDS admitted to general ICU. Methods We enrolled adult patients with severe ARDS related to COVID-19 infection consecutively admitted to ICU between February 2020 and March 2021. Acute cardiac injury was defined as a rise and fall of cTnI above the 99th percentile (i.e., ≥ 47 ng/L)(2). Hs-TnI was measured at admission and during ICU stay. Echocardiographic examinations were performed at ICU admission and at the time of worst cTnI. Left ventricular (LV) and right ventricular (RV) systolic dysfunction were defined respectively as LV ejection fraction (LVEF) < 50% and as tricuspid annular plane systolic excursion (TAPSE) < 16 mm. Results A total of 148 patients were included in the analysis. Acute cardiac injury was present in 65 (44.9%) cases at admission. Forty-seven (21.8%) patients had isolated depressed LVEF, 13 (9.0%) had isolated RV systolic dysfunction and 24 (16,4%) presented with biventricular systolic dysfunction. Eithy-two (56.1%) had a worsening of hs-TNI values during the admission of whom a consensual worsening of LVEF and TAPSE was observed in respectively 32 (21.8%) and 28 (19.2%) patients. Cardiac injury at ICU admission yielded an increased risk of in-hospital mortality (OR 1.30 [95% CI 0.71 – 2.38]) (Figure 1 – left panel) as well as the presence of impaired echocardiography (OR 1.85 [95% CI 1.01 – 3.40]) (Figure 1 – right panel). Worst hs-TNI correlated with worst procalcitonin (PCT) values (p <0.001 r20.81) in case of on-going septic shock and 85 % of patients with RV dysfunction had either evidence of pulmonary thrombus at CT scan or high inspiratory peak pressure (> 30 cmH2O) at mechanical ventilation (respectively, p < 0.001 and p <0.01). Bivariable analysis adjusted for the inverse probability of having an impaired echocardiographic exam confirmed the increased risk of in-hospital mortality in patients with cardiac injury (OR 2.01 [95% CI 1.28 – 3.15]) and in presence of altered echocardiography with RV dysfunction yielding a higher risk (OR 2.42 [95% CI 1.29 – 4.55]) compared with biventricular impairment (OR 1.86 [95% CI 0.87 – 4.00]) and isolated LV dysfunction (OR 1.30 [95% CI 0.77 – 2.19]) (Figure 2). Conclusion The use of echocardiography for the detection of cardiac dysfunction, has shown a greater correlation with mortality than the identification of acute cardiac damage using laboratory parameters. Of all the parameters obtained with the echocardiography, right ventricular dysfunction demonstrates the greatest correlation with mortality compared to left or biventricular dysfunction. However, the major cause of LV dysfunction in our population occurred during septic shock while RV dysfunction seemed to be due to pulmonary embolism and/or acute cor pulmonale ARDS-related.

BackgroundCardiovascular disease (CVD) is a major complication in patients on maintenance hemodialysis (MHD). While the effect of left ventricular dysfunction on patients on MHD is well known, the association between right ventricular (RV) dysfunction and their prognosis remains poorly understood.MethodsWe conducted a retrospective cohort study of 149 patients on MHD (median age, 72 years; male, 67%) between 1 April 2022 and 31 March 2024. All subjects underwent transthoracic two-dimensional and Doppler echocardiographic examination. RV dysfunction was determined when the tricuspid annular plane systolic excursion (TAPSE) was under 16 mm. Kaplan–Meier survival analysis was used to compare survival between patients with and without RV dysfunction. Multivariate Cox regression analyses were used to estimate the effects of RV dysfunction on all-cause mortality, including the effect of confounding factors such as age, sex, previous CVD events, and left ventricular (LV) dysfunction.ResultsIn total, 20 out of 149 patients on MHD (13%) had RV dysfunction. Indexes of their LV function (such as the LV ejection fraction and E/e' ratio) were similar to those of patients without RV dysfunction. During the follow-up period, 23 patients died (7 with RV dysfunction [without LV dysfunction], 3 with LV dysfunction [without RV dysfunction], 2 with biventricular dysfunction, and 11 without any ventricular dysfunction). Kaplan–Meier analyses showed significantly lower survival among patients with RV dysfunction (p < 0.01). The multivariate Cox regression analysis revealed that RV dysfunction (without LV dysfunction) was independently associated with all-cause mortality (hazard ratio: 3.44; 95% confidence interval: 1.28–9.23) after adjusting for confounders.ConclusionsRV dysfunction is a significant risk factor for all-cause mortality among patients on MHD.

Treatment of Diabetes in People with Heart Failure

Left ventricular (LV) dilatation is an important prognostic factor in patients with aortic regurgitation (AR). Although current guidelines recommend the use of LV end-systolic diameter index (LVESDi) to indicate the need for intervention, recent studies suggested that LV end-systolic volume index (LVESVi) may more accurately characterize LV remodelling.The present study aims to evaluate, in a multi-centre setting, whether combining LV linear and volumetric measures could improve risk stratification. A total of 1070 patients (56 ± 18 years, 65% male) with significant AR were included. Cut-off values of 20 mm/m2 for LVESDi and 45 mL/m2 for LVESVi were used to identify the following groups: no-significant LV dilatation (n = 485), when both LVESDi and LVESVi were below the cut-off values; discordant LV dilatation (n = 279) if only one positive criterium was present; and concordant LV dilatation (n = 306) when both LVESDi and LVESVi were enlarged. The primary endpoint was all-cause mortality. During a median follow-up of 7.4 (IQR, 4.5-11) years, 168 patients (16%) died, and 484 (45%) underwent aortic valve surgery (AVS). Patients with concordant LV dilatation showed the worst 10-year survival (P < 0.001). Discordant (HR 2.066, 95% CI 1.295-3.298; P = 0.002) or concordant LV dilatation (HR 2.759, 95% CI 1.616-4.710; P < 0.001) was independently associated with higher mortality compared with patients with no-significant LV dilatation after adjusting for relevant clinical and echocardiographic variables and regardless of AR severity. However, both groups showed greater benefit from AVS. LV dilatation, either concordant or discordant, was also independently associated with outcome in asymptomatic patients and those with left ventricular ejection fraction > 55%. In patients with significant AR, the presence of LV dilatation detected by linear and/or volumetric measures was independently associated with increased mortality.

Information about a real patient is presented in stages (boldface type) to expert clinicians (Dr James M

Mitral annular disjunction (MAD) has increasingly been recognized as a marker for adverse cardiovascular events in Marfan syndrome (MFS). As recent adult data links MFS with left ventricular (LV) dilation and reduced ejection fraction (LVEF), we hypothesized that MAD may be associated with LV dilation in pediatric MFS patients. A retrospective analysis was performed among MFS patients < 19years old at initial cardiac MRI (CMR). MAD and mitral valve prolapse (MVP) were assessed by CMR or most proximate echo. CMR-derived left ventricular end-diastolic (LVEDV) and end-systolic (LVESV) volumes were measured. Indexed volumes, absolute and indexed z-scores, and LVEF were calculated. The combined volume load from mitral and aortic regurgitation was indexed to LV stroke volume, allowing exclusion of patients with greater than mild volume load or prior MV intervention. MAD association with LV volumes and z-scores was then assessed. Forty-two patients were analyzed (median age 13.5years old, IQR [10.9, 15.3]). MAD was present in 28 patients (66.7%), and MVP was present in 13 patients (31.0%). Absolute LVEDV z-score was > 2 in 35.7% of patients, LVESV z-score was > 2 in 42.9%, and LVEF was < 55% in 45.2%. In multivariable analysis including MVP, MAD remained independently associated with elevated absolute LVESV z-score > 2 (RR 3.88, 95% CI 1.02-14.69, p = 0.046). MAD was associated with CMR-derived volume-load-independent LV dilation among pediatric MFS patients. Prospective studies are needed to further understand this association and its relationship with LV dilation over time.

Is the measurement of left ventricular ejection fraction the proper end point for cell therapy trials? An analysis of the effect of bone marrow mononuclear stem cell administration on left ventricular ejection fraction after ST-segment elevation myocardial infarction when evaluated by cardiac magnetic resonance imaging