Naser, Jwan A.

Abstract Background and Aims Incidence and types of secondary tricuspid regurgitation (TR) are not well defined in atrial fibrillation (AFib) and sinus rhythm (SR). Atrial secondary TR (A-STR) is associated with pre-existing AFib; however, close to 50% of patients with A-STR do not have AFib. The aim of this study was to assess incidence, types, and outcomes of ≥ moderate TR in AFib vs. SR. Methods Adults with and without new-onset AFib without structural heart disease or ≥ moderate TR at baseline were followed for the development of ≥ moderate TR. Tricuspid regurgitation types were pacemaker, left-sided valve disease, left ventricular (LV) dysfunction, pulmonary hypertension (PH), isolated ventricular, and A-STR. Results Among 1359 patients with AFib and 20 438 in SR, 109 and 378 patients developed ≥ moderate TR, respectively. The individual types of TR occurred more frequently in AFib related to the higher pacemaker implantation rates (1.12 vs. 0.19 per 100 person-years, P < .001), larger right atrial size (median 78 vs. 53 mL, P < .001), and higher pulmonary pressures (median 30 vs. 28 mmHg, P < .001). The most common TR types irrespective of rhythm were LV dysfunction-TR and A-STR. Among patients in SR, those with A-STR were older, predominantly women with more diastolic abnormalities and higher pulmonary pressures. All types of secondary TR were associated with all-cause mortality, highest in PH-TR and LV dysfunction-TR. Conclusions New-onset AFib vs. SR conferred a higher risk of the individual TR types related to sequelae of AFib and higher pacemaker implantation rates, although the distribution of TR types was similar. Secondary TR was universally associated with increased mortality.

Introduction: Graves’ disease (GD) is known to cause menstrual changes and has been linked to infertility in women. This can be due to the increased production of androstenedione and testosterone induced by the excess thyroid hormone. However, screening for sex hormone levels in all women diagnosed with GD to identify patients more likely to have infertility is not cost-effective and not clinically indicated. AI-augmented ECG can identify the signature of female with an AUC of 0.97. In this model, AI- ECG gives a probability number between 0-100% of being female, with higher numbers indicating high probability of being female. Hypothesis: We hypothesized that this inexpensive AI ECG tool is reflective of hormonal levels and the resulting changes in menstrual cycle and fertility associated with GD in women. Methods: Patients diagnosed with GD (2009-2019) at our institution were included. Three intervals were identified: 1) Before GD: >2 months before GD diagnosis 2) During GD: within 2 months before GD and before attaining euthyroidism 3) After GD: after attaining euthyroidism. The matched-pairs t-test was used to compare probability of being a female; patients acted as their own controls. A subset analysis was done on females <45 years at GD diagnosis to study the association between female probability during GD and menstrual changes (defined as shorter, lighter, or newly irregular menstrual cycles). Results: 154 females with ECGs both before and during GD were identified. AI ECG female probability decreased with GD diagnosis (85% vs. 77%, P<.001). On the other hand, 137 females with GD had ECGs during and after GD. Female probability increased after attaining euthyroidism (76% vs 83%, P=.005). Moreover, 72 females <45 years were identified; 30 (42%) developed menstrual changes and had significantly lower AI-ECG female probability [77 (58-94)% vs 98 (92-100)%, P<.001] Conclusions: AI-enabled ECG signature of female gender in women was altered by GD and was associated with menstrual changes, likely reflecting changes in sex hormone levels. Future studies to confirm its role in screening for infertility in women with GD are needed.

Introduction: Acceleration time (AT) and Doppler Velocity Index (DVI) are critical parameters in assessment of bioprosthetic aortic valve dysfunction (BAVD). AT>100ms and DVI < 0.25 are highly suggestive of prosthetic obstruction and are incorporated in all triage algorithms. Little is known about their relationship with flow. Hypothesis: AT and DVI may be affected by transprosthetic flow in patients with significant bioprosthetic aortic valve dysfunction (BAVD). Methods: Single center, retrospective analysis of patients with severe BAVD (confirmed at surgical re-replacement, transcatheter valve-in-valve implantation, or successfully treated bioprosthetic valve thrombosis). Standard prosthetic parameters (AT, DVI, effective orifice area (EOA), mean gradient, presence and degree of regurgitation), flow (stroke volume index, Svi; transprosthetic flow rate, FR) were measured on the transthoracic echocardiogram at the time of diagnosis with severe BAVD. Results: A total of 266 patients with severe BAVD underwent treatment between 2008-2020. BAVD mechanism was obstruction (N=192, 72%), regurgitation (n=48, 18%) mixed stenosis and regurgitation (n=15, 6%) and patient-prosthetic mismatch (n=12, 5%). DVI was moderately correlated with FR (r=0.50, p<0.001); AT had a weak but significant inverse correlation with FR (r=0.24, p<0.001). SVi had a weaker influence on DVI (r=0.35, p<0.001) and no correlation with AT. Conclusions: AT and DVI are influenced by transprosthetic flow rate. This flow-dependency should be considered when assessing the mechanism of BAVD.

Abstract Aims Augmenting echocardiography with artificial intelligence would allow for automated assessment of routine parameters and identification of disease patterns not easily recognized otherwise. View classification is an essential first step before deep learning can be applied to the echocardiogram. Methods and results We trained two- and three-dimensional convolutional neural networks (CNNs) using transthoracic echocardiographic (TTE) studies obtained from 909 patients to classify nine view categories (10 269 videos). Transthoracic echocardiographic studies from 229 patients were used in internal validation (2582 videos). Convolutional neural networks were tested on 100 patients with comprehensive TTE studies (where the two examples chosen by CNNs as most likely to represent a view were evaluated) and 408 patients with five view categories obtained via point-of-care ultrasound (POCUS). The overall accuracy of the two-dimensional CNN was 96.8%, and the averaged area under the curve (AUC) was 0.997 on the comprehensive TTE testing set; these numbers were 98.4% and 0.998, respectively, on the POCUS set. For the three-dimensional CNN, the accuracy and AUC were 96.3% and 0.998 for full TTE studies and 95.0% and 0.996 on POCUS videos, respectively. The positive predictive value, which defined correctly identified predicted views, was higher with two-dimensional rather than three-dimensional networks, exceeding 93% in apical, short-axis aortic valve, and parasternal long-axis left ventricle views. Conclusion An automated view classifier utilizing CNNs was able to classify cardiac views obtained using TTE and POCUS with high accuracy. The view classifier will facilitate the application of deep learning to echocardiography.

Abstract Aims We aimed to clarify the extent to which cardiac and peripheral impairments to oxygen delivery and utilization contribute to exercise intolerance and risk for adverse events, and how this relates to diversity and multiplicity in pathophysiologic traits. Methods and results Individuals with heart failure with preserved ejection fraction (HFpEF) and non-cardiac dyspnoea (controls) underwent invasive cardiopulmonary exercise testing and clinical follow-up. Haemodynamics and oxygen transport responses were compared. HFpEF patients were then categorized a priori into previously-proposed, non-exclusive descriptive clinical trait phenogroups, including cardiometabolic, pulmonary vascular disease, left atrial myopathy, and vascular stiffening phenogroups based on clinical and haemodynamic profiles to contrast pathophysiology and clinical risk. Overall, patients with HFpEF (n = 643) had impaired cardiac output reserve with exercise (2.3 vs. 2.8 L/min, p = 0.025) and greater reliance on peripheral oxygen extraction augmentation (4.5 vs. 3.8 ml/dl, p &lt; 0.001) compared to dyspnoeic controls (n = 219). Most (94%) patients with HFpEF met criteria for at least one clinical phenogroup, and 67% fulfilled criteria for multiple overlapping phenogroups. There was greater impairment in peripheral limitations in the cardiometabolic group and greater cardiac output limitations and higher pulmonary vascular resistance during exertion in the other phenogroups. Increasing trait multiplicity within a given patient was associated with worse exercise haemodynamics, poorer exercise capacity, lower cardiac output reserve, and greater risk for heart failure hospitalization or death (hazard ratio 1.74, 95% confidence interval 1.08–2.79 for 0–1 vs. ≥2 phenogroup traits present). Conclusions Though cardiac output response to exercise is limited in patients with HFpEF compared to those with non-cardiac dyspnoea, the relative contributions of cardiac and peripheral limitations vary with differing numbers and types of clinical phenotypic traits present. Patients fulfilling criteria for greater multiplicity and diversity of HFpEF phenogroup traits have poorer exercise capacity, worsening haemodynamic perturbations, and greater risk for adverse outcome.

Background: Our team has previously developed convolutional neural networks (CNNs) to estimate age and sex from a 10-second, 12-lead ECG as indicators of patient wellness. Here, we develop an additional wellness network to estimate body mass index (BMI) from the ECG input signal. Aims: To evaluate the performance of neural networks trained to classify and directly estimate BMI. Methods: We identified all adult patients in the Mayo Clinic system who had a 10-second, 12-lead ECG within 15 days of a BMI measurement. Of the&nbsp;174,863 patients in the cohort, 2,895 were underweight (BMI < 18.5), 42,387 were of normal weight (BMI 18.5-24.9), and 57,939 were overweight only (BMI 25.0-29.9). Among patients categorized as obese, 39,788 had class I obesity (BMI 30.0-34.9), 18,978 had class II obesity (BMI 35.0-39.9), and 12,876 had class III obesity (BMI ≥ 40.0). The cohort was split into training, validation, and testing datasets in an 8:1:1 ratio to develop networks for BMI classification and a direct&nbsp;estimation of BMI. Results: In detecting overweight in the testing set, the model achieved an AUC of 0.86 (95% CI 0.85, 0.87). In detecting class I, class II, and class III obesity, the model yielded AUCs of 0.84 (95% CI 0.83, 0.85), 0.84 (95% CI 0.83, 0.85), and 0.85 (95% CI 0.85, 0.86), respectively. The model to estimate BMI achieved a mean absolute error of 3.70. Conclusions: AI-enabled ECGs demonstrate efficient detection of overweight patients and all classes of obesity. Further research is needed to determine if discrepancies between the traditionally measured BMI and the AI-ECG BMI reflect the presence or lack of metabolic abnormalities.

Background: Enlargement or dilation of the left ventricle (LV) is associated with increased cardiovascular morbidity and mortality, and is an established precursor of ventricular dysfunction and heart failure in almost all major cardiovascular conditions. Although there are existing artificial intelligence frameworks that make use of deep learning to detect LV enlargement, many entail LV segmentation, which can be error-prone especially for point-of-care ultrasound (POCUS) images. Goal: To develop a fully end-to-end deep learning framework that takes as input either transthoracic echocardiographic (TTE) or POCUS 2D image clips, and classifies directly (i.e., without first segmenting the LV) whether a patient has an enlarged LV. Methods: Using a sample of retrospective TTE data enriched for LV enlargement (out of 13,868 patients total, 3,351 had an enlarged LV defined using sex-specific guideline-directed criteria), we trained a classifier that takes as input apical 2-chamber and parasternal long axis videos, and provides as output a binary label denoting whether a patient has an enlarged LV. Then, we evaluated the trained model using a holdout testing dataset of retrospective TTE data (2,087 patients) as well as data from a prospective cohort of 428 patients from whom TTE and POCUS were collected simultaneously. Results: For the retrospective TTE testing dataset, the AUC was 0.851 (95% confidence interval (CI) of 0.830-871). For the prospective cohort, the model performed comparably for both TTE and POCUS video clips (TTE: AUC of 0.865, 95% CI of 0.823-0.906; POCUS: AUC of 0.840, 95% CI of 0.797-0.883). Conclusions: The current deep learning model performed comparably well for both TTE and POCUS echocardiograms. We regard these results as a proof-of-concept of the potential utility of this model in helping point-of-care clinicians make decisions regarding whether patients should receive further evaluation for LV enlargement and associated conditions.

Background: Recently, our team developed an end-to-end artificial intelligence (AI) framework that can successfully estimate left ventricular ejection fraction (LVEF) from echocardiogram videos. However, this framework requires specific views (A2C and PLAX), and cannot estimate LVEF for studies from which these views are missing. Goal: To develop a deep learning model to estimate LVEF from any type and number of echocardiogram video(s). Methods: We built a deep learning model for B-mode echocardiogram videos of the left ventricle (A2C, A3C, A4C, or PLAX views), using retrospective transthoracic echocardiography (TTE) data. The model was evaluated with two test datasets: one with random valid (i.e., showing the left ventricle) views, and one with preselected views (both having 433 patients). Later, the model was tested with a prospective patient cohort that had both TTE and point-of-care ultrasound (POCUS) data collected simultaneously (393 patients). Results: On the retrospective TTE test set, we observed excellent performance (five random valid views per study: RMSE of 6.87% and Pearson correlation (PC) of 0.82 on LVEF estimation, and an AUC of 0.96 (95% CI: of 0.94-0.98) on low LVEF classification (<=40%); preselected views: RMSE of 6.75%, PC of 0.83, and an AUC of 0.96 (95% CI: 0.94-0.98)). For both TTE and POCUS data from the prospective cohort, we also observed strong performance (TTE: RMSE of 5.62%, PC of 0.87, and an AUC of 0.97; POCUS: RMSE of 6.12%, PC of 0.84, and an AUC of 0.96). For the prospective test sets as well as for the retrospective TTE test set with preselected views, results reflect the estimated LVEF average of A2C, A3C, A4C, and PLAX videos from each patient. Conclusions: For both TTE and POCUS data, the current model showed strong performance which was comparable to that of our previously established view-specific model. Alongside this comparable performance, the current model surpasses our previous model in terms of flexibility, as it isn't dependent on view classification and allows a broader selection in terms of the number and types of videos per study or patient. This innovative approach could prove particularly beneficial in point-of-care situations in which it may not be feasible to obtain high-quality, view-specific videos.

Background: Clinically significant tricuspid regurgitation (TR) is an important contributor to morbidity and mortality. The Tricuspid Regurgitation Impact on Outcomes (TRIO) score based on 8 parameters (age, gender, chronic heart failure, chronic lung disease, heart rate, creatinine, aspartate transaminase level, and TR severity) is used for risk stratification. The performance of the TRIO score in patients with TR secondary to severe pulmonary hypertension (PH) is unknown. Research Questions: To evaluate whether the TRIO score can be applied in patients with TR and pulmonary hypertension (PH), and to identify other predictors of mortality. Methods: Retrospective study including patients with TR ≥ moderate. Severe PH was defined as right ventricular systolic pressure (RVSP) ≥50 mmHg. The impact of clinical, echocardiographic and laboratory variables on mortality was evaluated with the Cox proportional hazard method. Results: A total of 13,482 patients from the TRIO population were included; severe PH was present in 6,530 (48.4%); mean age 73.4 + 13.3, 55.7% females. TRIO score performance was modest (AUC 0.62) in risk stratification (Figure A). PH patients had significantly higher mortality than those without PH at all levels of TRIO score (Figure B); differences were progressively lower with increasing TRIO scores. At multivariate analysis, only TRIO score [HR = 1.54 (1.35-1.77) per unit change], presence of an elevated NT-proBNP [HR = 1.02 (1.02-1.03) per 1000 pg/ml change], RVSP (HR = 1.08 (1.02-1.16) per 10 mmHg change], presence of RV dysfunction [HR = 1.15, (1.05-1.26)] and pulmonary artery pulsatility index (PAPi) ≤ 4 [HR = 1.34 (1.07-1.67)] were significant predictors of mortality (model AUC 0.65). Conclusions: TRIO score allows risk stratification in PH patients, but performance is modest; incorporating additional factors marginally increased model performance. PH patients have higher mortality at all levels of TRIO score, more so at low scores.

Introduction: Isolated functional tricuspid regurgitation (FTR) in the absence of structural heart disease may develop secondary to elevation in left heart pressures caused by left ventricular diastolic dysfunction (DD), as in heart failure with preserved ejection fraction (HFpEF). FTR may also herald the presence of occult DD or indicate greater risk for new-onset DD. However, prevalence and incidence of diastolic dysfunction (DD) in isolated FTR have not been studied. Aims: Investigate the prevalence and incidence of DD across the different severities of FTR of unclear cause. Methods: Adults with normal EF and without structural heart disease undergoing echocardiography at a tertiary medical center with a repeat study ≥6 months from baseline were identified. Severe DD was defined by ≥3/4 abnormal DD parameters (medial e’, medial E/e’, tricuspid regurgitation velocity, left atrial volume index) and moderate DD by ≥2. Results: Among 30,428 patients (median age 60; 50% women), FTR was absent in 73%, mild in 22%, moderate in 4%, and severe in 0.4%. Severe DD was present in 2%, 5%, 11%, and 13% patients, respectively, (p<.001) and moderate DD in 9%, 19%, 31%, and 48%, respectively (p<.001). The association of DD with increasing FTR was independent of all clinical characteristics. Over median 3.0 (IQR 1.5-5.4) years, incident severe and moderate DD developed more frequently with increasing FTR severity ( Figure) . Findings were confirmed in a propensity-matched subset where patients were matched for age, sex, BMI, atrial fibrillation, and comorbidities and in patients without lung disease or ≥moderate right ventricular enlargement. Conclusions: Patients with FTR of undefined etiology commonly have DD suggesting the presence of underlying left heart disease. Patients without apparent DD or HF at baseline who have FTR remain at significantly increased risk of developing DD over follow-up and may be a target population for therapies to prevent progression to clinical HFpEF.