Disease Network-Based Approaches to Study Comorbidity in Heart Failure: Current State and Future Perspectives

Abstract

Purpose of review: Heart failure (HF) is often accompanied by a constellation of comorbidities, leading to diverse patient presentations and clinical trajectories. While traditional methods have provided valuable insights into our understanding of HF, network medicine approaches seek to leverage these complex relationships by analyzing disease at a systems level. This review introduces the concepts of network medicine and explores the use of comorbidity networks to study HF and heart disease.

Recent findings: Comorbidity networks are used to understand disease trajectories, predict outcomes, and uncover potential molecular mechanisms through identification of genes and pathways relevant to comorbidity. These networks have shown the importance of non-cardiovascular comorbidities to the clinical journey of patients with HF. However, the community should be aware of important limitations in developing and implementing these methods. Network approaches hold promise for unraveling the impact of comorbidities in the complex presentation and genetics of HF. Methods that consider comorbidity presence and timing have the potential to help optimize management strategies and identify pathophysiological mechanisms.

Keywords: Comorbidity; Heart failure; Networks.

Fig. 1

Overview of methods of network construction. Networks can be constructed (A) from data where information on each disease is present for each patient. In these networks, diseases are represented by nodes and the connections between diseases by edges. After network construction they can be used for a variety of study types or can be (B) integrated with data or prior knowledge resources that allow the linking of different levels of information about medicine and biology

Fig. 2

Comparison of two networks built using the Morbinet shiny browser [104] demonstrates how small processing changes can affect networks. The network built using an odds ratio for association of at least 1.8 with (A) International Classification of Primary Care, 2nd edition (ICPC2) codes contains more nodes and connections than that built with (B) simplified ICPC2 codes. Comparison of (C) number of shared nodes and (D) edges are shown in Venn Diagrams with ICPC2 data in blue and simplified ICPC2 data in purple

Fig. 3

Schematic overview of potential of comorbidity network representations to explore the complexity and implications of comorbidities in heart failure. The central panel highlights the complex interplay between heart failure and its most common comorbidities, beyond the mere coexistence of diseases to clinical impact. The upper right panel highlights the commonality of risk associations derived from epidemiological studies between the different conditions surrounding HF. The network in the upper left panel represents the complex interplay between therapeutic classes and different outcomes across HF and related comorbidities, highlighting the differential impact of each therapy concerning the assessed outcome. Finally, the lower panel summarizes the pathophysiological interplay between HF, chronic kidney disease (CKD), and type 2 diabetes mellitus from a molecular perspective, highlighting bidirectional effects related to multisystemic pathways