Key result
Echocardiographic nomogram predicts post-discharge heart failure readmission with ~0.76 AUC at 24 months.
Why the study?
Conventional heart failure risk tools anchored in clinical and biochemical indices often miss early, subclinical signals of recurrence. Speckle-tracking echocardiography provides noninvasive markers of residual myocardial and atrial dysfunction that may refine prognostication for post-discharge readmission.
Does an echocardiography-derived prediction model integrating Age, LVEF, CI, LVGLS, LAVI, and LASr accurately predict the composite of heart failure-related hospital readmission or emergency department visit within 24 months in patients hospitalized for acute heart failure with predominantly preserved or mildly reduced LVEF?
Cohort (n=266)
Blinded outcome assessment (imaging analysts were blinded to clinical outcomes)
No
Does an echocardiography-derived prediction model integrating Age, LVEF, CI, LVGLS, LAVI, and LASr accurately predict the composite of heart failure-related hospital readmission or emergency department visit within 24 months in patients hospitalized for acute heart failure with predominantly preserved or mildly reduced LVEF?
Effect estimate: AUC 0.761 at 24 months (0.667 at 6 months, 0.696 at 12 months)
p-value: p=<0.05
An echocardiography-derived nomogram integrating Age, LVEF, cardiac index, LVGLS, LAVI, and LASr provides modest short-term and acceptable 24-month discrimination (AUC 0.761) for predicting heart failure readmission in patients with preserved or mildly reduced LVEF.
Echocardiographic nomogram shows moderate 24-month HF readmission discrimination; leaves open clinical utility pending external validation.
Heart failure is characterized by high morbidity and a substantial burden of post-discharge events. Conventional risk tools anchored in clinical and biochemical indices often miss early, subclinical signals of recurrence. Speckle-tracking echocardiography provides noninvasive markers of residual myocardial and atrial dysfunction that may refine prognostication and enable individualized follow-up. To derive and internally validate an echocardiography-derived prediction model that integrates structural and functional parameters to estimate the risk of heart failure-related readmission within 24 months after discharge. In a retrospective cohort of 266 consecutive acute heart failure admissions (2023–2024), standardized transthoracic echocardiography quantified left ventricular global longitudinal strain, left atrial indices, and cardiac index. Candidate variables were screened with LASSO, Boruta, and univariable Cox regression, and overlapping predictors entered a multivariable Cox model to construct a nomogram estimating 6-, 12-, and 24-month risk. Model performance was assessed by time-dependent ROC, 1,000-bootstrap calibration, and decision-curve analysis. Over 24 months, 72 patients (27.1%) experienced the composite endpoint of heart failure-related readmission or emergency department visit. Six variables—Age, LVEF, CI, LVGLS, LAVI, and LASr—were independently associated with risk (all p < 0.05). The nomogram achieved time-dependent AUCs of 0.667, 0.696, and 0.761 at 6, 12, and 24 months, respectively, with close agreement between predicted and observed probabilities on calibration analysis and consistent net benefit across broad threshold ranges on DCA. Variable contributions emphasized LVGLS and LASr, underscoring the central role of impaired LV longitudinal myocardial deformation and reduced LA compliance; elevations in LAVI and reductions in CI added complementary information on chronic filling pressure load and perfusion status. Model discrimination increased across the 6-, 12-, and 24-month time points. An imaging-driven, multidimensional nomogram based on echocardiographic mechanics demonstrates modest discrimination at 6 and 12 months and acceptable discrimination at 24 months for post-discharge HF readmission risk in a predominantly preserved or mildly reduced LVEF cohort. Integrating LVGLS, LASr, LAVI, and CI may support exploratory risk stratification during the vulnerable post-discharge phase and warrants further validation before broader clinical implementation. Not applicable. Not applicable.
No takes yet. Share an insight, caveat, or question.
Wenjun et al. (2026) conducted a cohort in Acute or decompensated heart failure with predominantly preserved or mildly reduced ejection fraction (n=266). Echocardiography-based multivariable prediction model (incorporating Age, LVEF, CI, LVGLS, LAVI, and LASr) vs. Individual echocardiographic and clinical predictors was evaluated on Composite of heart failure-related hospital readmission or emergency department visit within 24 months after discharge (AUC 0.761 at 24 months (0.667 at 6 months, 0.696 at 12 months), p=<0.05). An echocardiographic nomogram integrating age, LVEF, CI, LVGLS, LAVI, and LASr predicted post-discharge heart failure readmission with time-dependent AUCs of 0.667, 0.696, and 0.761 at 6, 12, and 24 months.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: