Knowledge Graph for Cardiovascular Drug Safety and Pharmacovigilance: A Scoping Review

Cardiovascular diseases (CVDs) remain a major cause of death worldwide, and adverse drug reactions (ADRs) are a key concern in their management. Knowledge graphs (KGs) offer new ways to integrate heterogeneous biomedical data and enhance pharmacovigilance (PV). This scoping review maps current KG applications for cardiovascular drug safety, highlighting methodologies, technologies, and evidence gaps. Following PRISMA-ScR guidelines, we systematically searched six databases for studies applying KGs or semantic technologies to CVD-related PV tasks. Six eligible studies were identified, addressing signal detection, mechanistic interpretation, and class-effect analysis. KGs enabled integration of clinical, molecular, and pharmacological data using ontologies such as MedDRA, RxNorm, HPO, and GO, improving mechanistic reasoning and hypothesis generation. However, validation against standard PV methods, data completeness and clinical generalizability remain limited. Overall, KGs show strong potential for cardiovascular PV but require large-scale validation and integration with EHR data to achieve clinical impact.