Answers are grounded in this paper and cite their sources.
Inflammatory bowel disease (IBD) is associated with systemic metabolic perturbations and cardiovascular burden. We used UK Biobank proton nuclear magnetic resonance data to identify exploratory metabolic profiles and evaluate associations with cardiovascular disease (CVD) and mortality. CatBoost-SHAP feature ranking followed by standardized-feature K-means clustering identified three arbitrary, dataset-specific labels: MQ (relatively lower GlycA and less overall perturbation), IM (an intermediate multivariable pattern), and IC (higher GlycA with some lower lipid measures). In polygenic-risk-score-adjusted outcome models using non-IBD controls as the reference, IC was associated with composite CVD in ulcerative colitis (UC; HR 1.57, 95% CI 1.27–1.95; 85/348 events/participants) and Crohn’s disease (CD; 1.60, 1.14–2.25; 33/149), heart failure (UC 2.26, 1.71–2.98; 50/396; CD 3.24, 2.21–4.77; 26/169), and death (UC 1.45, 1.18–1.79; 87/400; CD 2.40, 1.85–3.10; 58/173). Internal split-sample ARI was 0.919 for UC and 0.754 for CD. However, cluster separation was modest and method-dependent, feature-selection stability was limited, and incremental discrimination changes were small. These single-cohort observational findings are hypothesis-generating and do not establish causality, external validity, or clinical utility.
No takes yet. Share an insight, caveat, or question.
Yuan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: