Related Experiment Video
Updated: Nov 7, 2025
Cutoff Value of Phase Angle by Bioelectrical Impedance Analysis at Admission as a Prognostic Factor in Patients with Acute Heart Failure
Published on: June 10, 2025
Clinical phenogroups are more effective than left ventricular ejection fraction categories in stratifying heart
Andreas B Gevaert1,2, Semra Tibebu3, Mamas A Mamas4
1Research Group Cardiovascular Diseases, GENCOR Department, University of Antwerp, Antwerp, Belgium.
Insights
Clinical phenogroups, identified by machine learning, offer better heart failure (HF) prognostication than left ventricular ejection fraction (LVEF) categories. These phenotypes provide more accurate risk stratification for patient outcomes.
Area of Science:
- Cardiology
- Medical Informatics
- Clinical Research
Background:
- Current heart failure (HF) guidelines classify patients using left ventricular ejection fraction (LVEF) into reduced, mid-range, or preserved categories.
- This LVEF-based classification may not fully capture the heterogeneity of HF and its prognostic implications.
Purpose of the Study:
- To investigate whether distinct clinical phenogroups, identified through unsupervised machine learning, provide superior prognostic information compared to traditional LVEF categories in hospitalized HF patients.
- To determine the association between identified phenogroups and clinical outcomes, including all-cause death and rehospitalization.
Main Methods:
- A sub-study of the Patient-Centered Care Transitions in HF trial analyzed baseline characteristics of 1693 hospitalized patients with recorded LVEF.
- Unsupervised machine learning (cluster analysis) was employed to identify clinical phenogroups based on predominant comorbidities.
- Associations between phenogroups and primary (all-cause death/rehospitalization) and secondary (cardiovascular death/HF rehospitalization) composite outcomes at 6 and 12 months were assessed.
Main Results:
- Six distinct phenogroups were identified, each characterized by a primary comorbidity (e.g., coronary heart disease, atrial fibrillation, COPD) or few comorbidities.
- These phenogroups were independent of LVEF, encompassing a wide range of LVEF values.
- Phenogroups demonstrated significant associations with primary and secondary outcomes at 6 and 12 months (log-rank P < 0.001 and P < 0.002, respectively), with hazard ratios varying significantly between groups.
- LVEF-based classifications failed to differentiate risk categories for the primary outcomes at both 6 and 12 months (P = 0.69 and P = 0.30, respectively).
Conclusions:
- Clinical phenogroups derived from machine learning offer enhanced prognostic value in hospitalized heart failure patients compared to LVEF-based classifications.
- These phenogroups provide a more nuanced understanding of patient risk, potentially leading to more personalized treatment strategies.
- The findings suggest a shift towards phenogroup-based risk stratification for improved management of heart failure.
Aims:
Heart failure (HF) guidelines place patients into 3 discrete groups according to left ventricular ejection fraction (LVEF): reduced (<40%), mid-range (40-49%), and preserved LVEF (≥50%). We assessed whether clinical phenogroups offer better prognostication than LVEF.
Methods And Results:
This was a sub-study of the Patient-Centered Care Transitions in HF trial. We analysed baseline characteristics of hospitalized patients in whom LVEF was recorded. We used unsupervised machine learning to identify clinical phenogroups and, thereafter, determined associations between phenogroups and outcomes. Primary outcome was the composite of all-cause death or rehospitalization at 6 and 12 months. Secondary outcome was the composite cardiovascular death or HF rehospitalization at 6 and 12 months. Cluster analysis of 1693 patients revealed six discrete phenogroups, each characterized by a predominant comorbidity: coronary heart disease, valvular heart disease, atrial fibrillation (AF), sleep apnoea, chronic obstructive pulmonary disease (COPD), or few comorbidities. Phenogroups were LVEF independent, with each phenogroup encompassing a wide range of LVEFs. For the primary composite outcome at 6 months, the hazard ratios (HRs) for phenogroups ranged from 1.25 [95% confidence interval (CI) 1.00-1.58 for AF] to 2.04 (95% CI 1.62-2.57 for COPD) (log-rank P < 0.001); and at 12 months, the HRs for phenogroups ranged from 1.15 (95% CI 0.94-1.41 for AF) to 1.87 (95% 1.52-3.20 for COPD) (P < 0.002). LVEF-based classifications did not separate patients into different risk categories for the primary outcomes at 6 months (P = 0.69) and 12 months (P = 0.30). Phenogroups also stratified risk of the secondary composite outcome at 6 and 12 months more effectively than LVEF.
Conclusion:
Among patients hospitalized for HF, clinical phenotypes generated by unsupervised machine learning provided greater prognostic information for a composite of clinical endpoints at 6 and 12 months compared with LVEF-based categories.
Trial Registration:
ClinicalTrials.gov Identifier: NCT02112227.
More Related Videos
Related Concept Videos
01:30Heart Failure IV: Classification and Diagnostic Evaluation
01:17Pathophysiology of Heart Failure
01:29Cardiomyopathy V: Interprofessional Care
01:22Heart Failure Drugs: β-Blockers
01:30Heart Failure V: Medical Management
01:29Heart Failure II: Pathophysiology