Publications by Year: 2026

BACKGROUND: Whether the 2025 American Society of Echocardiography (ASE) diastolic dysfunction (DD) algorithm (DD25) improves mortality prognostication compared to the 2016 algorithm (DD16) in real-world practice is uncertain.

METHODS: We applied the DD25 algorithm to adult transthoracic echocardiography reports across a large academic multisite echocardiography laboratory from 2018 2024, linked to state mortality data, and determined reclassification of DD and mortality risk.

RESULTS: Of 12,948 included (age 62.8 ± 18.1, 51.4% women, 55.8% outpatient), 10,205 (78.8%) had diastology quantified by the 2016 and 2025 guidelines. Of these, 2,601 (25.5%) were reclassified by DD25 with increases in DD grade in 1,428 (54.9%) and decreases in 1,173 (45.1%). Among those reclassified, 2,391 (91.9%) had a single grade change in DD severity. A larger proportion of female patients (52% vs 48.7%) were classified as having DD by DD25 versus DD16. The rate of indeterminate DD was lower by DD25 (1,358 [10.4%]) vs DD16 (1,983 [15.3%]). The DD25 algorithm improved discrimination of mortality risk compared to DD16 (difference in areas under the curve = 0.02; 95% CI, 0.001-0.04; P = .03), although the magnitude of the associated risk across the follow-up period was similar after multivariable adjustment (P value for comparison of adjusted hazard ratios = .67).

CONCLUSIONS: In a large academic health system, one-quarter of patients had reclassification of diastolic function by the 2025 ASE diastology guidelines. As compared to the 2016 algorithm, the 2025 algorithm resulted in a larger proportion with DD, less indeterminate diastolic function, and improved discrimination of mortality.

BACKGROUND: Sepsis remains a leading cause of mortality, and optimizing treatment is challenging due to patient heterogeneity. Identification of cardiac phenotypes may inform precision medicine approaches and guide resuscitation. We performed a clustering analysis of patients with sepsis using echocardiographic data without using any a priori definitions of cardiac dysfunction or outcomes to establish the subgroups.

METHODS: This was a retrospective cohort study of patients admitted to the hospital with sepsis at a single academic center. Patients were identified using sepsis-related ICD codes, and those who had echocardiogram performed within 14 days of admission underwent chart review to ensure sepsis-3 criteria were met. Those with preexisting heart disease were excluded. Clustering by echocardiographic variables was performed using latent profile analysis. Clinical features such as patient characteristics, laboratory studies, sepsis source, and outcomes were compared across the clusters.

RESULTS: There were 2,071 patients included in the analysis. Our cluster analysis yielded five phenotypes: cluster 1, elevated mean E/e' 24.5 (SD 9.6); cluster 2, reduced ejection fraction, mean 33.1% (SD 10.6), and cardiac index 2.6 L/min/m2 (SD 0.9); cluster 3, right ventricular dilation with right ventricular basal diameter 4.5 cm (SD 0.9) and elevated tricuspid regurgitation gradient 60.0 mmHg (SD 13.5); cluster 4, hyperdynamic with mean left ventricular ejection fraction 75% (SD10.9) and mean cardiac index 6.6 L/min/m2 (SD 2.6); and lastly cluster 5, normal echocardiographic parameters. Group 3 had the highest mortality compared to the normal group (36.9% vs. 19.6%, p = 0.002), with an odds ratio of 2.3 (95%CI 1.4-3.9).

CONCLUSIONS: Using an unsupervised clustering analysis, we identified five phenotypes of cardiac function in sepsis based on commonly recorded echocardiographic data: diastolic dysfunction, left ventricular systolic dysfunction with low cardiac index, right ventricular dilation with elevated tricuspid regurgitation gradient, hyperdynamic cardiac function, and normal. The right ventricular dilation group had the highest mortality. Future research should explore mechanisms and potential treatment implications for these groups.

BACKGROUND: Low left ventricular ejection fraction (LEF) can progress undiagnosed. Artificial intelligence-based electrocardiogram (ECG-AI) screening may provide a scalable means to detect LEF.

OBJECTIVES: The purpose of this study was to validate a complete ECG-AI software as a medical device for LEF detection.

METHODS: Four geographically diverse sites in the United States identified patients with both ECGs and transthoracic echocardiograms performed within 30 days of each other in clinical practice. Data were electronically extracted to specific guidelines and transmitted to the coordinating center for analysis.

RESULTS: Records of 16,000 subjects were extracted, resulting in an evaluable set of 13,960 subjects (mean age 66 years; 52% male). The device demonstrated excellent discrimination (AUROC: 0.92 [95% CI: 0.91-0.93]) and was 84.5% (95% CI: 82.2%-86.6%) sensitive and 83.6% (95% CI: 82.9%-84.2%) specific for LEF. The overall prevalence of LEF in the study data set was 7.9%, with LEF among 1.6% of the ECG-AI negative and 30.5% of ECG-AI positive subjects, contributing to positive and negative predictive values of 30.5% (95% CI: 28.8%-32.1%) and 98.4% (95% CI: 98.2%-98.7%), respectively.

CONCLUSIONS: External validation studies such as this one provide a rigorous framework to validate an algorithm's performance. This study demonstrated the algorithm's strong diagnostic accuracy over a geographically diverse, independent set of patients. In this generally unselected population, the algorithm produced a test negative result in 78% of the cases, suggesting potential utility as a rule-out strategy to defer echocardiography when other clinical findings are absent.

• It is uncertain if the 2017 ASCO recommendations on postanthracycline surveillance echocardiography impacted practice. • Release of the ASCO guidelines was not associated with changes in echocardiogram frequency with under half of patients receiving an echocardiogram within 18 months following chemotherapy initiation.