Key result
EHMRG and MEESSI-AHF risk scores demonstrate the strongest external validation for predicting short-term mortality.
Why the study?
Several risk scores exist to predict early mortality and guide disposition for emergency department patients with acute heart failure, but their target outcomes, predictors, and validation settings differ.
Do multivariable risk scores accurately predict short-term mortality in adults presenting to the emergency department with acute heart failure?
Systematic Review (n=19)
Do multivariable risk scores accurately predict short-term mortality in adults presenting to the emergency department with acute heart failure?
EHMRG and MEESSI-AHF are the most robustly validated risk scores for predicting short-term mortality in acute heart failure patients in the emergency department, though local calibration is required before clinical implementation.
Supports ED risk stratification with EHMRG and MEESSI-AHF in acute HF; confirms discrimination across cohorts but local calibration remains essential.
Background: Acute heart failure (AHF) necessitate emergency department (ED) evaluation, where early mortality risk affect monitoring, admission, discharge, and follow-up decisions. Several scores have been proposed and their target outcomes, predictors, and validation settings differ. Objective: To synthesize the discrimination, calibration, and utility of risk scores predicting short-term mortality in patients presenting to EDs with AHF. Methods: Records were identified through PubMed/MEDLINE, Scopus and WOS and citation tracking, with strategies prepared for Scopus, Web of Science, and Cochrane Library through Jun 20, 2026. Eligible studies validated or compared multivariable scores for mortality within 30 days of emergency presentation. Study characteristics, performance and calibration findings, and risk-of-bias domains were narratively synthesized because outcome windows and populations were heterogeneous. Results: We include 19 reports. EHMRG was the most validated score for 7-day mortality, with statistics ranging from 0.73 to 0.81 in Canadian, Spanish, and United States cohorts. MEESSI-AHF showed a strong discrimination for 30-day mortality, between 0.80 and 0.84, although external cohorts identified overprediction in higher-risk groups before recalibration. EAHFE-3D show moderate discrimination for 3-day mortality, with imprecision in smaller external samples. Broader disposition scores incorporated mortality into composite outcomes and offered less direct evidence for death prediction. Conclusions: EHMRG and MEESSI-AHF have the strongest external-validation evidence. Local calibration, predictor availability, and prospective impact evaluation remain essential before disposition use.
No takes yet. Share an insight, caveat, or question.
Alanazi et al. (2026) conducted a systematic review in Acute heart failure (n=19). Multivariable risk scores (EHMRG, MEESSI-AHF, EAHFE-3D) was evaluated on Discrimination and calibration for short-term mortality (3, 7, or 30 days). EHMRG and MEESSI-AHF demonstrated the strongest external validation for predicting short-term mortality, with c-statistics of 0.73-0.81 for 7-day and 0.80-0.84 for 30-day mortality, respectively.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: