Characterizing High Risk Patients in Heart Failure

Frederique J. Hafkamp; Lukas Dekker; R.A. Tio; Dennis van Veghel; Tineke Vinck; Geert Smits; Hareld Kemps; Ramon R. T. van de Ven

doi:10.1097/jcn.0000000000001208

April 8, 2025The Journal of Cardiovascular NursingOpen Access

Characterizing High Risk Patients in Heart Failure

View Full Paper

Ask AI

Bookmark

Share

Why the study?

Heart failure classification based solely on LVEF may not optimally reflect distinct clinical groups or predict risk, highlighting the need for alternative classification methods.

Does latent class analysis based on sociodemographic and clinical data offer better risk stratification for mortality and rehospitalization in patients with heart failure compared to LVEF or NT-proBNP alone?

Structured PICO

Does latent class analysis based on sociodemographic and clinical data offer better risk stratification for mortality and rehospitalization in patients with heart failure compared to LVEF or NT-proBNP alone?

P

Population

1045 patients with heart failure

I

Intervention

Classification using latent class analysis based on sociodemographic and clinical data

C

Comparator

Classification based on LVEF or NT-proBNP alone

O

Outcome

Mortality and rehospitalization riskshard clinical

Latent class analysis successfully categorizes heterogeneous heart failure patients into distinct clinical phenotypes that effectively predict mortality and rehospitalization risks.

Implication

May support HF phenotype-based risk stratification; leaves open added value over LVEF or NT-proBNP.

Abstract

BACKGROUND: Patients with heart failure (HF) are typically classified based on left ventricular ejection fraction (LVEF). However, this may not optimally reflect distinct groups or predict risk. Therefore, alternative classification methods are needed. OBJECTIVE: Our aim was to classify patients with HF based on sociodemographic and clinical data using latent class analysis, assess if latent classes pose varying risks of mortality and rehospitalization, and explore if these classes offer better risk stratification than LVEF or N-terminal prohormone of brain natriuretic peptide (NT-proBNP) alone. METHODS: Data from 1045 patients were analyzed using latent class analysis to identify classes and assess mortality and rehospitalization risks. Kaplan-Meier curves with log-rank tests were used to compare mortality across latent classes, LVEF-based groups, and NT-proBNP-based groups. RESULTS: Three latent classes were identified. Class 1 (48.8%) comprised middle-aged males with HF with reduced ejection fraction (HFrEF) and high NT-proBNP levels. Class 2 (26.1%) consisted of mainly older females with HF with preserved ejection fraction, high body mass index, hypertension, atrial fibrillation, and anemia. Class 3 (25.2%) included younger patients with HFrEF, high body mass index, and a high smoking rate. Class 1 and 2 had higher mortality risks ( P < .001) and longer rehospitalization durations ( P = .011) than class 3. CONCLUSION: Latent class analysis categorized a heterogeneous group of patients with HF into homogeneous classes. These classes provide a close approximation of what could be observed in clinical practice and provide insight into patients at higher risk of mortality and rehospitalization.