Search

Background: Left ventricular end-diastolic pressure (LVEDP) is a key indicator of cardiac health. The gold-standard method of measuring LVEDP is invasive intra-cardiac catheterization. Echocardiography is used for non-invasive estimation of left ventricular (LV) filling pressures; however, correlation with invasive LVEDP is variable. We sought to use machine learning (ML) algorithms to predict elevated LVEDP (>20 mmHg) using clinical, echocardiographic, and biomarker parameters., Methods: We identified a cohort of 460 consecutive patients from the Cleveland Clinic, without atrial fibrillation or significant mitral valve disease who underwent transthoracic echocardiography within 24 hours of elective heart catheterization between January 2008 and October 2010. We included patients' clinical (e.g., heart rate), echocardiographic (e.g., E/e'), and biomarker [e.g., N-terminal brain natriuretic peptide (NT-proBNP)] profiles. We fit logistic regression (LR), random forest (RF), gradient boosting (GB), support vector machine (SVM), and K-nearest neighbors (KNN) algorithms in a 20-iteration train-validate-test workflow and measured performance using average area under the receiver operating characteristic curve (AUROC). We also predicted elevated tau (>45 ms), the gold-standard parameter for LV diastolic dysfunction, and performed multi-class classification of the patients' cardiac conditions. For each outcome, LR weights were used to identify clinically relevant variables., Results: ML algorithms predicted elevated LVEDP (>20 mmHg) with good performance [AUROC =0.761, 95% confidence interval (CI): 0.725-0.796]. ML models showed excellent performance predicting elevated tau (>45 ms) (AUROC =0.832, 95% CI: 0.700-0.964) and classifying cardiac conditions (AUROC =0.757-0.975). We identified several clinical variables [e.g., diastolic blood pressure, body mass index (BMI), heart rate, left atrial volume, mitral valve deceleration time, and NT-proBNP] relevant for LVEDP prediction., Conclusions: Our study shows ML approaches can robustly predict elevated LVEDP and tau. ML may assist in the clinical interpretation of echocardiographic data., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (https://cdt.amegroups.com/article/view/10.21037/cdt-24-128/coif). The authors have no conflicts of interest to declare., (2024 AME Publishing Company. All rights reserved.)

Background: Infective endocarditis (IE) is associated with high morbidity and mortality. Following an initial negative transesophageal echocardiogram (TEE), high clinical suspicion warrants repeat examination. We evaluated the diagnostic performance of contemporary TEE imaging for IE., Methods: This retrospective cohort study included patients ≥18 years old undergoing ≥2 TEEs within 6 months, with confirmed diagnosis of IE based on Duke criteria, 70 in 2011 and 172 in 2019, were included. We compared the diagnostic performance of TEE for IE in 2019 versus 2011. The primary endpoint was the sensitivity of initial TEE to detect IE., Results: Sensitivity of the initial TEE to detect endocarditis was 85.7% versus 95.3%, in 2011 and 2019, respectively (P=0.01). On multivariable analysis, initial TEE more frequently detected IE in 2019, compared to 2011 [odds ratio (OR): 4.06, 95% confidence intervals (CIs): 1.41-11.71, P=0.01]. Improved diagnostic performance was driven by improved detection of prosthetic valve infective endocarditis (PVIE), sensitivity 70.8% in 2011 versus 93.7% (P=0.009) in 2019. In 2019, TEEs more frequently utilized probes with higher frame rates/resolution, than 2011 (P<0.001). Three dimensional (3D) technology was utilized in 97.2% of initial TEEs in 2019, compared to 70.5% in 2011 (P<0.001)., Conclusions: Contemporary TEE was associated with improved diagnostic performance for endocarditis, driven by improved sensitivity for PVIE., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-22-431/coif). BM received Education grant $5000 from Healthcare Delivery and Implementation Science Center (separate project) and was supported by the National Center for Advancing Translational Sciences of the NIH under Award Number TL1TR002344. This grant contributed to drafting the manuscript, data analysis, manuscript submission, and manuscript revision. The other authors have no conflicts of interest to declare., (2023 Cardiovascular Diagnosis and Therapy. All rights reserved.)

Background: Left ventricular global longitudinal strain (LVGLS), circumferential strain (LVGCS) and radial strain (LVGRS) are echocardiographic parameters with wide clinical applicability. However, the thresholds for abnormal left ventricular (LV) strains, particularly the lower limits of normal (LLN), are not well established. This meta-analysis determined the mean and LLN of two- (2D) and three-dimensional (3D) LV strain in healthy subjects and factors that influence strain measurements., Methods: We searched PubMed, Embase and Cochrane databases until 31 December 2019 for studies reporting left ventricular (LV) global strain in at least 50 healthy subjects. We pooled means and LLNs of 2D and 3D LV strain using random-effects models, and performed subgroup and meta-regression analysis for LVGLS., Results: Forty-four studies were eligible totaling 8,910 subjects. The pooled means and LLNs (95% confidence intervals) were -20.1% (-20.7%, -19.6%) and -15.4% (-16.0%, -14.7%) respectively for 2D-LVGLS; -21.9% (-23.4%, -20.3%) and -15.3% (-16.9%, -13.8%) respectively for 2D-LVGCS; and 48.4% (43.8%, 53.0%) and 25.5% (17.8%, 33.1%) respectively for 2D-LVGRS. All pooled analyses demonstrated significant heterogeneity, and means and LLNs of and 3D-LV strains differed marginally from 2D. Only vendor software was associated with differences in pooled means and LLN of 2D-LVGLS., Conclusions: In conclusion, pooled means and LLNs of 2D- and 3D-LV global strain parameters in healthy subjects were reported. Based on the pooled LLNs, thresholds for abnormal, borderline and normal strains can be defined, such as less negative than -14.7%, between -14.7% and -16.0% and more negative than -16.0% respectively for 2D-LVGLS, and 2D-LVGLS values are only affected by vendor software., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at http://dx.doi.org/10.21037/cdt-20-711). TKMW received a clinical and research fellowship grant from the National Heart Foundation of New Zealand (number 1775). The other authors have no conflicts of interest to declare., (2020 Cardiovascular Diagnosis and Therapy. All rights reserved.)

Background: Assessment of right ventricular (RV) function plays an important role in patients with cardiopulmonary disease, and current guidelines recommend parameters including tricuspid annular plane systolic excursion (TAPSE) and right ventricular systolic excursion velocity (RVS') to assess RV longitudinal function. We assessed the hypothesis that the previously undescribed motion of RV longitudinal rotation (RVLR) is an independent predictor of both TAPSE and RVS'., Methods: We assessed a series of 100 consecutive patients with pulmonary hypertension (PH) undergoing echocardiography. Patients with left ventricular (LV) dilation and dysfunction were excluded. Standard RV parameters were determined using established guidelines, while RVLR and right ventricular global longitudinal strain (RVGLS) measurements were performed using 2-dimensional (2D) speckle tracking technique., Results: Mean peak RVLR measured -4.2±3.7 degrees. By convention, negative values implied clockwise motion. In a multiple linear regression model, TAPSE could be predicted from a combination of RVLR and RVGLS (R=0.56, P<0.001). A similar relationship was found for RVS' which could also be predicted from a combination of RVLR and RVGLS (R=0.52, P<0.001). While no association was found between RVLR and RV size, estimated RV systolic pressure (RVSP) or the presence of a pericardial effusion, a mild correlation was noted between RVLR and QRS duration (R=0.25, P=0.01)., Conclusions: RVLR is an independent predictor of TAPSE and RVS'. Awareness of this motion should be considered in the interpretation of TAPSE and RVS' values as markers of RV systolic function, as abnormal RVLR may account for exaggerated values, particularly in patients with PH and RV dysfunction., Competing Interests: Conflicts of Interest: The authors have no conflicts of interest to declare.

Background: Despite preserved left ventricular ejection fraction (LVEF), patients with significant primary mitral regurgitation (MR) often have reduced exercise capacity. In asymptomatic patients with ≥3+ primary MR undergoing rest-stress echocardiography (RSE), we sought to evaluate the incremental impact of left ventricular global longitudinal strain (LV-GLS) on exercise capacity., Methods: A total of 660 asymptomatic patients with ≥3+ primary MR, non-dilated LV and LVEF ≥60% (mean age, 57±14 years, 66% men, body mass index or BMI 25±4 kg/m 2 ) who underwent RSE at our center between 2001 and 2013 were included. Standard RSE data were obtained. Average resting LV-GLS was measured using Velocity Vector Imaging., Results: Mean mitral effective regurgitant orifice, resting right ventricular systolic pressure (RVSP) and LV-GLS were 0.45±0.2 cm 2 , 31±12 mmHg and -21.7%±2%, respectively; 28% had flail mitral leaflet. Mean metabolic equivalents (METs) and post-stress RVSP were 9.9±3, and 46±15 mmHg; 28% achieved <100% age-gender predicted METs. No patient had ischemia or significant arrhythmias. On logistic regression, resting LV-GLS [odds ratio (OR), 1.40, 95% confidence interval (CI): 1.21-1.55, BMI (OR, 1.11, 95% CI: 1.06-1.17)] and resting RVSP 1.22 (1.02-1.49) were independent predictors of exercise capacity. Area under the curve for association between 100% age-gender predicted METs and various factors were as follows: (I) BMI (0.60, 95% CI: 0.55-0.65, P<0.001); (II) resting RVSP (0.57, 95% CI: 0.52-0.62, P=0.006) and LV-GLS (0.66, 95% CI: 0.61-0.70, P<0.001)., Conclusions: In asymptomatic patients with ≥3+ primary MR, non-dilated LV and preserved LVEF, LV-GLS is independently associated with exercise capacity, beyond known predictors., Competing Interests: Conflicts of Interest: MA Gillinov reports following financial conflicts of interest—speakers’ bureau for Atricure, Edwards, Medtronic and St. Jude’s Medical. He also reports equity stake in Pleuraflow. The other authors have no conflicts of interest to declare.