Citations

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.

Case: A 65-year-old patient presented 4 months after mitral valve repair (annuloplasty band, commissuroplasty, neo-chordae to the anterior mitral leaflet (AL)). At follow up, he developed significant hemolytic anemia (Hemoglobin 10.1g/dL, LDH 650U/L, Bilirubin 1.4mg/dL and haptoglobin<14 mg/dL). TEE revealed the substrate for hemolysis: moderate-severe mitral regurgitation (MR) starting at the coaptation of A2 and P2 scallops due to AL overriding the posterior (PL). The jet was then directed posteriorly to a small (3 x 6 mm confirmed by cardiac CT) area of annuloplasty band dehiscence at the base of the P2 scallop, with the band in direct path of the flow. The unique feature was that AL retained coaptation with the annuloplasty band, thus forming a small tunnel from the A2-P2 coaptation defect to the dehisced area. Given that redo surgery early after initial intervention would be challenging, the Heart Team decided to pursue percutaneous closure at the level of the dehisced annuloplasty to reduce regurgitation and mitigate hemolysis. Under 3D TEE guidance closure with a 12mm AVP plug was attempted but resulted in interference with PL motion with a new jet of regurgitation laterally. Using two 8mm AVP plugs was successful in closing the dehiscence with minimal impact on PL. Overall regurgitation was reduced to mild. Within a week hemolysis labs improved: LDH337 U/L, Bilirubin0.7 mg/dL. Discussion: Annuloplasty band dehiscence is a known cause of recurrent MR after repair. As regurgitation occurs at the leaflet coaptation zone, percutaneous closure of these defects is usually futile. In this patient we took advantage of the maintained coaptation between the AL and annuloplasty band.

Abstract Funding Acknowledgements None Background/Purpose: Anemia is a common condition in patients with cardiovascular disease. The effect of anemia in exercise stress testing has not been evaluated in a large cohort of patients. Methods We performed a retrospective review of patients undergoing exercise stress echocardiography over a 10 year period who had blood hemoglobin assessed within 48 hours of stress testing. World Health Organization cutoffs were used in defining the degree of anemia. The effect of anemia on exercise capacity and stress testing safety outcomes was assessed. Results In 15,652 patients, 90.3% were non-anemic, 7.6% mildly anemic, 2.1% moderately anemic, and 0.1% severely anemic. The incidence of serious supraventricular or ventricular arrhythmias and need to transfer to the hospital was low (0.1%) and did not correlate to anemia severity (p = 0.25). There was an increased incidence of non-sustained SVT or atrial fibrillation with moderate anemia and increased incidence of hypotension with stress testing with mild and moderate anemia. There was a significant reduction in duration of exercise (p &lt; 0.001) and METS (p &lt; 0.001) achieved with worsening anemia. Conclusion Our study demonstrates that exercise stress testing is safe in patients with mild and moderate anemia, albeit with reduced exercise capacity. Low numbers of severely anemic patients despite a large study population limit generalizability of these results in this group. Exercise Capacity Severe Anemia (N = 9) Moderate Anemia (N = 332) Mild Anemia (N = 1183) Non-anemic (N = 14128) P value Resting heart rate 81 ± 13 # 79 ± 14 # 72 ± 14 74 ± 13 &lt;0.001* Peak heart rate 148 ± 22 139 ± 24 # 138 ± 23 # 147 ±22 &lt;0.001* Exercise duration 5.3 ± 2.5 # 6.3 ± 2.1 # 7.2 ± 2.5 # 8.6 ± 2.7 &lt;0.001* METS 6.3 ± 2.5 # 7.1 ± 2.2 # 8.0 ± 2.5 # 9.5 ± 2.7 &lt;0.001* Safety Hypotension with stress 0 (0%) 31 (9.4%) # 94 (8.0%) # 437 (3.1%) &lt;0.001* Mild supraventricular arrhythmia 1 (11.1%) 15 (4.5%) # 20 (1.7%) # 237 (2.3%) 0.04* Severe supraventricular arrhythmia 0 (0%) 0 (0%) 0 (0%) 7 (0.1%) 0.99 Mild ventricular arrhythmia 0 (0%) 20 (6.0%) 74 (6.3%) 756 (5.4%) 0.94 Moderate ventricular arrhythmia 0 (0%) 3 (0.9%) 13 (1.1%) 182 (1.3%) 0.78 Severe ventricular arrhythmia 0 (0%) 1 (0.3%) 2 (0.2%) 1 (0.0%) 0.02 Transfer to hospital 0 (0%) 2 (0.6%) 1 (0.1%) 9 (0.1%) 0.29 Any severe arrhythmia or hospitalization 0 (0%) 2 (0.6%) 3 (0.3%) 17 (0.1%) 0.25 # Multi-comparison adjusted P-value &lt;0.05 compared to non-anemic when group P &lt; 0.05