Metabolic subgroups and cardiometabolic multimorbidity in the UK Biobank

Background: Ischemic heart disease (IHD), diabetes, cancer and dementia share features of age-associated metabolic dysfunction. We hypothesized that metabolic diversity explains the diversity of morbidity later in life. Methods: We analyzed data from the UK Biobank (N = 329,908). A self-organizing map (SOM, an artificial neural network) was trained with 51 metabolic traits adjusted for age and sex. The SOM analyses produced six subgroups that summarized the multi-variable metabolic diversity. The subgroup with the lowest adiposity and disease burden was chosen as the reference. Hazard ratios (HR) were modeled by Cox regression (P < 0.0001 unless otherwise indicated). Enrichment of multi-morbidity over random expectation was tested by permutation analysis. Results: The subgroup with the highest sex hormones was not associated with IHD (HR = 1.04, P = 0.14). The subgroup with high urinary excretion without kidney stress (HR = 1.24) and the subgroup with the highest apolipoprotein B and blood pressure (HR = 1.52) were associated with IHD. The subgroup with high adiposity, inflammation and kidney stress was associated with IHD (HR = 2.11), cancer (HR= 1.29), dementia (HR = 1.70) and mortality (HR = 2.12). The subgroup with high triglycerides and liver enzymes was at risk of diabetes (HR = 15.6). Paradoxical enrichment of multimorbidity in young individuals and in favorable subgroups was observed. Conclusions: These results support metabolic diversity as an explanation to diverging morbidity and demonstrate the potential value of population-based metabolic subgroups as public health targets for reducing aggregate burden of chronic diseases in ageing populations.