Pellikka, PA

Abstract Funding Acknowledgements Type of funding sources: None. Background Aortic regurgitation (AR) is a common valvular lesion associated with left ventricular (LV) enlargement and dysfunction and increased risk of death with onset of symptoms, significant LV enlargement, or systolic dysfunction. The current guidelines are based on symptoms, LV size and LV function to evaluate the candidacy for surgery in patients with severe AR. Purpose We propose a machine-learning (ML) based algorithm to identify patients at risk for death from AR independent of aortic valve replacement (AVR) surgery (patients were censored at AVR event unless the operation occurred within 100 days of the first visit, in which case baseline AVR is included as a predictor). Methods The model was trained with 5-fold cross-validation on a dataset of 1035 patients. Model performance is reported on an independent dataset of 207 patients. Optimal predictive performance was observed with a Conditional Random Survival Forest model. A subset of 19 / 41 variables were selected for inclusion in the final model. Variable selection was performed with 10-fold cross validation using Random Survival Forest model; variables that were on an average selected by less than 6/10 splits were excluded. Results The mean age of patients was 60 ± 17 years and 198 (18%) were females. The variables included in the models were age, end-diastolic volume, end systolic dimension, body surface area, ejection fraction, NYHA class, body mass index, diastolic blood pressure, bicuspid valve, regurgitant volume, filling pressure, mid ascending aorta diameter, mitral tissue early relaxation velocity, pulmonary artery systolic pressure, and comorbidities, and the relative variable importances averaged across five splits of cross validation in each repeat were evaluated (Figure). The predictive AUC for predicting survival of the best-performing model was 0.84 at 1 year, 0.86 at 2 years, and 0.87 overall, respectively. Conclusion Using commonly observed echo parameters and patient characteristics we successfully trained a ML model to predict survival in patients with severe AR. This technique could identify high-risk patients who would benefit from early intervention, thereby improving patient outcomes. Abstract Figure.

Abstract Funding Acknowledgements Type of funding sources: None. OnBehalf Stress Echo 2020 study group of the Italian Society of Echocardiography and Cardiovascular Imaging (SIECVI) Background The classical pathophysiological cornerstone of stress cardiac imaging is the ischemic cascade: alterations in coronary flow velocity reserve (CFVR) come first, regional wall motion abnormalities (RWMA) second, and global alterations of left ventricular diastolic and systolic function last. Aim To evaluate the ischemic cascade with stress echo (SE) assessed with ABCDE protocol. Methods We analyzed 1616 patients (age 66 ± 10 years, 1156, 72% males) with chest pain and angiographically documented coronary artery disease (CAD): 928 (57.4%) with previous myocardial infarction and 1451 (89.8%) with previous coronary revascularization. CAD patients were studied with ABCDE-SE protocol by 13 certified laboratories of 5 countries. Stress modality was exercise in 427 (26.4%) dobutamine in 173 (10.7%) and vasodilator stress in 1016 (62.9%) patients (dipyridamole in 982 – 60.8% and adenosine in 34 – 2.1% patients). Step A assessed RWMA; step B, B-lines (diastolic function); step C, left ventricular contractile reserve based on force; step D, CFVR in left anterior descending artery; step E, heart rate reserve. SE response ranged from score 0 (all steps normal) to score 5 (all steps abnormal). Follow up was obtained in all patients. Results Results were abnormal in 350 (21.7) % for step A, 572 (35.4%) for B, 666 (41.2%) for C, 546 (33.5%) for D and 643 (39.8%) for E. The distribution of positivity is shown in figure. The dominant "solitary phenotype" (only one biomarker abnormal during the test) was step A in 10 (0.6%), step B in 124 (7.7%) step C in 100 (6.2%), step D in 64 (4.0%) and step E in 157 (9.7%) patients. The overall sensitivity for CAD detection including each step of ABCDE score (≥1) was 95.7%. During median follow-up of 22 months [IQ range: 13-37 months] 279 coronary revascularizations occurred. ABCDE score predicted coronary revascularization with score 1: HR: 3.182, 95% CI: 1.699-6.067; score 2: HR: 4.921, 95% CI: 2.571-9.420; score 3: HR: 7.234, 95% CI: 3.743-13.979; score 4: HR: 24.570, 95% CI: 13.390-45.086; and score 5: HR: 38.720, 95% CI: 20.975-71.478 Conclusion Inducible ischemia with CAD is characterized by the complexity and multiplicity of diagnostic phenotypes. The overall ABCDE score predicts clinically driven revascularization. Abstract Figure. The 5 circles of ischemia biomarkers

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): Danish Heart Association. Background First-phase ejection fraction (EF1), the ejection fraction (EF) until the time of peak systolic flow may be a sensitive marker of subclinical left ventricular (LV) dysfunction. This study investigated the prognostic value of EF1 in asymptomatic and symptomatic severe aortic stenosis (AS). Methods This study included 94 asymptomatic and 108 symptomatic patients with severe AS. The prognostic value of EF1 was compared with other echocardiographic markers and magnetic resonance imaging (MRI) measured end-systolic wall stress and late gadolinium enhancement fibrosis (LGE). Asymptomatic patients were followed up for 3.0 years (primary outcome death or aortic valve replacement). Symptomatic patients were followed up for 4.3 years (primary outcome death). Results In multivariate regression analysis wall stress (p < 0.001) and LGE (p = 0.03) were associated with EF1. In the asymptomatic cohort EF1 was significantly associated with the end-point, especially among the subgroup of patients with a mean gradient < 40 mmHg (HR 0.91, p = 0.005), while global longitudinal strain was not. In the surgical cohort, EF1 was borderline associated with death (p = 0.08) which was significant after correction for LGE (HR 0.90, p = 0.02). Conclusion EF1 is a predictor of death or AVR in asymptomatic AS, especially among discordantly graded patients with low area and low gradient. Univariate β (95% CI) p-value Multivariate β (95% CI) p-value Age (years) .03 (-.09 to .16) 0.58 .03 (-.12 to .18) 0.69 Sex (male) -.12 (-2.26 to 2.03) 0.91 .24 (-2.31 to 2.80) 0.85 Hypertension -.08 (-2.22 to 2.07) 0.95 Aortic valve area (0.01 cm2) .09 (.04 to .14) 0.001 Aortic mean gradient (mmHg) -.02 (-.08 to .04) 0.56 LV end-diastolic volume (ml) -.07 (-.11 to -.03) <0.001 LV end-systolic volume (ml) -.17 (-.22 to -.12) <0.001 -.07 (-.15 to .02) 0.11 LV ejection fraction (%) .28 (.19 to .38) <0.001 LV peak ejection time (ms) -.07 (-.16 to .02) 0.13 LGE fibrosis -3.04 (-5.67 to -.42) 0.02 -2.64 (-4.99 to -.30) 0.03 Wall stress (kdynes/m3) -.10 (-.13 to -.08) <0.001 -.08 (-.12 to -.04) <0.001 Multivariate linear regression analysis for associations with first phase ejection fraction. LV is left ventricular, LGE late gadolinium enhancement Abstract Figure. AVR-free survival according to EF1