Identifying clinical subtypes in acute dyspnoea patients admitted to the emergency department: a secondary model-based clustering of prospective cohorts

Q: What are the key findings of this study?

Assignment to the 'hypoxemic' acute dyspnea subtype was independently associated with a significantly higher 3-month mortality risk (adjusted HR 3.87) compared to the 'non-inflammatory' subtype.

Q: What is the clinical evidence from this study?

Study design: Cohort. Population: Acute dyspnea (n=3278). Intervention: Hypoxemic subtype classification vs. Non-inflammatory subtype classification. Primary outcome: 3-month all-cause mortality (HR 3.87, 95% CI 2.17-6.87, p=<0.0001).

Q: What does this research mean for the field?

Assignment to a 'hypoxemic' clinical subtype among emergency department patients presenting with acute dyspnoea is independently associated with a significantly increased 3-month mortality risk. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

Q: What question did this study set out to answer?

To identify distinct clinical subtypes in patients presenting to the emergency department with acute dyspnoea using latent class analysis and evaluate their prognostic relevance.

Kamilė Čerlinskaitė‐Bajorė; Nikola Kozhuharov; Mojtaba Ahmadiankalati; Desiree Nadine Wussler; Maria Belkin; Christian Eugen Mueller; Jelena Čelutkienė; Alexandre Mebazaa; Sabri Soussi

doi:10.1093/eschf/xvag225

August 22, 2026ESC Heart FailureOpen Access

Identifying clinical subtypes in acute dyspnoea patients admitted to the emergency department: a secondary model-based clustering of prospective cohorts

View Full Paper

Ask AI

Bookmark

Share

Key result

Hypoxemic acute dyspnea subtype is linked to a ~4-fold higher 3-month mortality risk.

HR 3.87
95% CI 2.17-6.87
P<0.0001
n=3,278

Figures

Graphical abstract published by ESC Heart Failure — Graphical abstract · Published by ESC Heart Failure · CC BY

Why the study?

Does a model-based clustering approach classifying emergency department patients with acute dyspnoea into subtypes improve early risk stratification for 3-month mortality?

Key Points

To identify distinct clinical subtypes in patients presenting to the emergency department with acute dyspnoea using latent class analysis and evaluate their prognostic relevance.
Analyzed two prospective emergency department cohorts of adults with acute dyspnoea: the LEDA derivation cohort (N=1392) and the BASEL V external validation cohort (N=1886).
Applied latent class analysis to nine admission clinical and biological variables in LEDA and validated a simplified decision tree classifier in BASEL V.
Latent class analysis identified four reproducible subtypes: non-inflammatory (A), tachycardic (B), anaemic (C), and hypoxemic (D), with inflammatory and cardiovascular biomarker levels rising progressively from A to D (all P < .001).
Subtype assignment independently predicted 3-month mortality, with subtype D showing the highest risk compared to subtype A [adjusted HR (95% CI): LEDA 3.87 (2.17-6.87), P < .0001; BASEL V 5.15 (2.71-9.76), P < .001].
Adding subtype membership improved the Harrell C-index for 3-month mortality over linear combinations of class-defining variables in both LEDA (0.73 to 0.77, P = .04) and BASEL V (0.76 to 0.79, P = .03).

Study Design

Type

Cohort (n=3,278)

Multicenter

Yes

Structured PICO

Does a model-based clustering approach classifying emergency department patients with acute dyspnoea into subtypes improve early risk stratification for 3-month mortality?

P

Population

3,278 adult patients presenting to the emergency department with acute dyspnea across two prospective cohorts (LEDA and BASEL V), followed for 3 months to assess prognostic subtypes.

E

Exposure

Model-based clustering analysis (Latent Class Analysis) classifying patients into four subtypes based on 9 admission variables (SBP, HR, SpO2, sodium, potassium, creatinine, CRP, haemoglobin, glucose): 'non-inflammatory' (A), 'tachycardic' (B), 'anaemic' (C), and 'hypoxemic' (D).

O

Outcome

3-month all-cause mortalityhard clinical

Main Result

Hazard Ratio: 3.87 (95% CI 2.17–6.87)

Absolute Event Rate: 22.4% vs 4.8%

p-value: p=<0.0001

A model-based clustering approach identified four reproducible acute dyspnoea subtypes that transcend conventional diagnoses and independently predict 3-month mortality, improving early risk stratification in the emergency department.

Limitations

Patients were enrolled in European emergency departments, which may limit generalizability to other regions with different demographics or triage systems.
Some data were missing due to the observational nature of the cohorts.
Patients were enrolled prior to the COVID-19 pandemic, which has since altered the case-mix of acute dyspnea presentations.
Analysis of treatment effect heterogeneity across the four subtypes was not possible as no randomized controlled trials were included.
Requirement of written informed consent may have introduced a small selection bias.

Implication

May support subtype-specific monitoring in acute dyspnea; hypothesis-generating for ML phenotyping in ED risk stratification.

Abstract

INTRODUCTION: Acute dyspnoea is a frequent reason for emergency department (ED) presentation, with heterogeneous causes and prognosis often insufficiently captured by traditional diagnostic categories. We applied model-based clustering (i.e. latent class analysis [LCA]) to identify distinct subtypes of acute dyspnoea and assess their prognostic relevance. METHODS: We analysed two prospective ED acute dyspnoea adult cohorts (LEDA and BASEL V). LCA was performed in the LEDA derivation cohort using nine admission clinical/biological variables. A simplified decision tree was derived in the LEDA cohort and externally validated to classify BASEL V patients. Subtypes were compared regarding clinical characteristics, biomarker profiles, and 90-day mortality. RESULTS: Among 1392 LEDA and 1886 BASEL V patients, four reproducible subtypes were identified and labelled as 'non-inflammatory' (A), 'tachycardic' (B), 'anaemic' (C) and 'hypoxemic' (D). Subtypes transcended conventional diagnostic categories. Inflammatory and cardiovascular biomarker levels increased significantly from A to D (all P < .001). After adjusting for traditional risk factors, assignment remained independently associated with the 3-month mortality risk, with the worst outcome for patients assigned to the 'hypoxemic' subtype [adjusted HR (95% CI) for subtype D: LEDA 3.87 (2.17-6.87), P < .0001; BASEL V 5.15 (2.71-9.76), P < .001]. Adding subtype membership improved the Harrell C-index for 3-month mortality beyond a linear combination of the three main class-defining variables (LEDA 0.73 to 0.77, P = .04; BASEL V 0.76 to 0.79, P = .03). CONCLUSIONS: A model-based clustering approach identified four reproducible acute dyspnoea subtypes with distinct clinical, biomarker, and prognostic profiles. This framework may improve early risk stratification and personalized ED management beyond nosologically classified diagnoses.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Čerlinskaitė‐Bajorė et al. (2026) conducted a cohort in Acute dyspnea (n=3,278). Hypoxemic subtype classification vs. Non-inflammatory subtype classification was evaluated on 3-month all-cause mortality (HR 3.87, 95% CI 2.17-6.87, p=<0.0001). Assignment to the 'hypoxemic' acute dyspnea subtype was independently associated with a significantly higher 3-month mortality risk (adjusted HR 3.87) compared to the 'non-inflammatory' subtype.

synapsesocial.com/papers/6ac5918c20d7088ccb11a787 https://doi.org/10.1093/eschf/xvag225

View Full Paper

Ask AI

Bookmark

Share

Identifying clinical subtypes in acute dyspnoea patients admitted to the emergency department: a secondary model-based clustering of prospective cohorts

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Identifying clinical subtypes in acute dyspnoea patients admitted to the emergency department: a secondary model-based clustering of prospective cohorts

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading