- Research Article
- 10.37897/rmj.2025.4.15
When the failing heart is not alone: redefining chronic heart failure through phenotypes and comorbidities
- Dec 31, 2025
- Romanian Medical Journal
- Dmitri Savca + 3 more +3
Background and objectives. Heart failure poses a major challenge for healthcare systems. Although left ventricular ejection fraction (LVEF)-based classification guides current therapeutic decisions, it fails to capture the complexity of the syndrome. This narrative review aimed to analyze alternative approaches that integrate comorbidity profiles, biomarker data, and machine learning-based subgroup identification to achieve a more accurate and clinically relevant framework. We focused on adult patients with chronic heart failure and literature published between 2015 and 2025, identified primarily through PubMed, Web of Science, and other indexed scientific databases. Materials and methods. We conducted an extensive narrative review of recent literature on heart failure phenotyping strategies, focusing on comorbidity profiles, biomarker integration, and machine learning-based subgroup identification. As this is a narrative review, the selection of sources may be subject to selection bias. Results. New classification models provide more nuanced stratification of heart failure patients by identifying clinically meaningful subgroups. Comorbidity clusters can reveal distinct trajectories and therapeutic responses. In addition, phenotype-based models, often derived using unsupervised machine learning, reveal latent structures associated with prognosis and risk prediction. Conclusions. A dual-axis model – comorbidity-based subgroups crossed with biologically and clinically defined phenotypes – enhances LVEF-based classification for risk stratification and may guide individualized therapy.
Read more