Metabolites Associated With Long-Term Risk of Peripheral Artery Disease: The ARIC Study
In brief
Thirteen blood metabolites boost PAD risk prediction by about two percent
In a 27-year follow-up of 3,759 middle-aged adults, higher levels of 13 metabolites-including mannose and low 1,5-anhydroglucitol-were linked to new peripheral artery disease and raised the C-statistic for predicting PAD from 0.770 to 0.787. The same metabolomic panel improved detection of severe disease (critical limb ischemia) even more, suggesting a potential early-risk marker, though its incremental benefit over standard factors is modest.
- Journal
- Arteriosclerosis, thrombosis, and vascular biology (Q1)
- Published
- 27 August 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Eyram Cyril Bansah, Xiao Hu, Shoshana H Ballew, Morgan E Grams, Vijay Nambi, Elizabeth Selvin, et al.
- PMID
- 42657467
- DOI
- 10.1161/ATVBAHA.126.324549
Why clinicians should know about it
- Picked for Biochemistry (medical) (paper of the day, 28 August 2026): Metabolomics predicts long‑term PAD risk
Abstract
BACKGROUND: Lower extremity peripheral artery disease (PAD) is a major contributor to cardiovascular morbidity and mortality; however, the metabolic pathways underlying disease development and progression are poorly understood. METHODS: We conducted a prospective cohort study of 3759 participants from the ARIC study (Atherosclerosis Risk in Communities; mean age, 53.5 years; 60% women; 62% Black participants) with untargeted metabolomic profiling at baseline (1987-1989). Incident PAD (458 cases) and its severe form with rest pain or tissue loss, critical limb ischemia (CLI; 135 cases), were identified through hospitalization codes over a median follow-up of 27 years. We used Cox regression with adjustment for traditional risk factors and accounted for multiple testing using a false discovery rate correction at a threshold of 0.05. Risk prediction improvement was evaluated using Harrell C statistics and the net reclassification improvement. RESULTS: Thirteen metabolites were significantly associated with incident PAD and primarily reflected dysregulated glycemic control, impaired redox balance, and altered lipid remodeling (eg, higher levels of mannose and lower levels of 1,5-anhydroglucitol, gamma-glutamyl dipeptides, and lysophospholipids). For CLI, 18 metabolites were identified, 11 of which were unique to CLI and reflected intensified oxidative and bioenergetic stress (eg, homocitrulline, arabonate, and 1-linoleoylglycerol). Metabolites significantly improved risk prediction for PAD beyond traditional risk factors (∆C statistic, 0.017 [95% CI, 0.008-0.025] from a base C statistic of 0.770; net reclassification improvement, 0.053 [95% CI, 0.018-0.083]) and particularly CLI (∆C statistic, 0.041 [95% CI, 0.018-0.064] from a base C statistic of 0.823; net reclassification improvement, 0.176 [95% CI, 0.085-0.268]). CONCLUSIONS: Distinct metabolomic profiles were associated with incident PAD and CLI years before clinical onset, highlighting dysregulated glycemic control, impaired redox balance, and altered lipid remodeling as key underlying pathways. These findings suggest that circulating metabolites may serve as early indicators of metabolic health relevant to PAD risk identification and progression to advanced disease.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.