Nuclear Magnetic Resonance-Based Metabolomics and Risk of CKD.

RATIONALE &

OBJECTIVE

Chronic kidney disease (CKD) leads to lipid and metabolic abnormalities, but a comprehensive investigation of lipids, lipoprotein particles, and circulating metabolites associated with the risk of CKD has been lacking. We examined the associations of nuclear magnetic resonance (NMR)-based metabolomics data with CKD risk in the UK Biobank study.

STUDY

DESIGN

Observational cohort study.

SETTING &

PARTICIPANTS

A total of 91,532 participants in the UK Biobank Study without CKD and not receiving lipid-lowering therapy.

EXPOSURE: Levels of metabolites including lipid concentration and composition within 14 lipoprotein subclasses, as well as other metabolic biomarkers were quantified via NMR spectroscopy.

OUTCOME: Incident CKD identified using ICD codes in any primary care data, hospital admission records, or death register records.

ANALYTICAL APPROACH: Cox proportional hazards regression models were used to estimate hazard ratios and 95% confidence intervals.

RESULTS

We identified 2,269 CKD cases over a median follow-up period of 13.1 years via linkage with the electronic health records. After adjusting for covariates and correcting for multiple testing, 90 of 142 biomarkers were significantly associated with incident CKD. In general, higher concentrations of very-low-density lipoprotein (VLDL) particles were associated with a higher risk of CKD whereas higher concentrations of high-density lipoprotein (HDL) particles were associated with a lower risk of CKD. Higher concentrations of cholesterol, phospholipids, and total lipids within VLDL were associated with a higher risk of CKD, whereas within HDL they were associated with a lower risk of CKD. Further, higher triglyceride levels within all lipoprotein subclasses, including all HDL particles, were associated with greater risk of CKD. We also identified that several amino acids, fatty acids, and inflammatory biomarkers were associated with risk of CKD.

LIMITATIONS: Potential underreporting of CKD cases because of case identification via electronic health records.

CONCLUSIONS

Our findings highlight multiple known and novel pathways linking circulating metabolites to the risk of CKD.

PLAIN-LANGUAGE SUMMARY: The relationship between individual lipoprotein particle subclasses and lipid-related traits and risk of chronic kidney disease (CKD) in general population is unclear. Using data from 91,532 participants in the UK Biobank, we evaluated the associations of metabolites measured using nuclear magnetic resonance testing with the risk of CKD. We identified that 90 out of 142 lipid biomarkers were significantly associated with incident CKD. We found that very-low-density lipoproteins, high-density lipoproteins, the lipid concentration and composition within these lipoproteins, triglycerides within all the lipoprotein subclasses, fatty acids, amino acids, and inflammation biomarkers were associated with CKD risk. These findings advance our knowledge about mechanistic pathways that may contribute to the development of CKD.