Background: Cardiovascular-Kidney-Metabolic (CKM) syndrome aggregates obesity, diabetes, hypertension, and renal dysfunction. Clinical outcomes depend on complex systemic interactions, yet current guidelines often treat these conditions in isolation. Emerging evidence links central nervous system (CNS) structures-specifically hypothalamic regions-to metabolic control (the “brain-heart axis”), but population-scale validation is lacking. Furthermore, traditional linear models are insufficient for capturing the non-linear risks of this multi-system syndrome.
Aims: We aim to use UK Biobank data to:
Epidemiology: Characterize CKM prevalence and longitudinal trajectories using linked electronic health records (EHR) and primary care data.
Brain-Heart Axis: Investigate associations between brain MRI phenotypes (focusing on hypothalamic and autonomic structures) and CKM severity to explore neuro-metabolic mechanisms.
Network Analysis: Apply network medicine approaches to map comorbidity clusters and disease progression patterns.
AI Prediction: Build machine learning models (e.g., XGBoost, deep learning) integrating polygenic risk scores, biochemistry, and wearable-derived activity data to predict adverse cardiovascular outcomes.
Methods: We will analyse participants with relevant phenotypic data. Statistical methods will include Cox proportional hazards regression for survival analysis and Mendelian Randomisation (MR) to infer causality between neuro-metabolic traits and CKM outcomes. Network analysis and ML frameworks will be used for risk modelling.
Public Health Impact: This research intends to clarify biological crosstalk between the CNS and systemic metabolism and develop precision, data-driven tools for identifying high-risk CKM patients.Furthermore, the integration of multi-omics data will identify specific molecular signatures, providing a robust evidence base to guide downstream basic experimental validation.
