A multiple-metabolites model to predict preliminary renal injury induced by iodixanol based on UHPLC/Q-Orbitrap-MS and 1H-NMR

Liying Cheng,Liming Wang,Biying Chen,Chenxi Wang,Mengxi Wang,Jie Li,Xiumei Gao,Zhu Zhang,Lifeng Han
DOI: https://doi.org/10.1007/s11306-022-01942-3
2022-10-28
Metabolomics
Abstract:Background & aims: There are some problems, such as unclear pathological mechanism, delayed diagnosis, and inaccurate therapeutic target of Contrast-induced acute kidney injury (CI-AKI). It is significantly important to find biomarkers and therapeutic targets that can indicate renal injury in the early stage of CI-AKI. This study aims to establish a multiple-metabolites model to predict preliminary renal injury induced by iodixanol and explore its pathogenesis. Methods: Both UHPLC/Q-Orbitrap-MS and 1H-NMR methods were applied for urine metabolomics studies on two independent cohorts who suffered from a preliminary renal injury caused by iodixanol, and the multivariate statistical analysis and random forest (RF) algorithm were used to process the related date. Results: In the discovery cohort (n = 169), 6 metabolic markers (leucine, indole, 5-hydroxy-L-tryptophan, N-acetylvaline, hydroxyhexanoycarnine, and kynurenic acid) were obtained by the cross-validation between the RF and liquid chromatography-mass spectrometry (LC-MS). Secondly, the 6 differential metabolites were confirmed by comparison of standard substance and structural identification of 1H-NMR. Subsequently, the multiple-metabolites model composed of the 6 biomarkers was validated in a validation cohort (n = 165). Conclusions: The concentrations of leucine, indole, N-acetylvaline, 5-hydroxy-L-tryptophan, hydroxyhexanoycarnitine and kynurenic acid in urine were proven to be positively correlated with the degree of renal injury induced by iodixanol. The multiple-metabolites model based on these 6 biomarkers has a good predictive ability to predict early renal injury caused by iodixanol, provides treatment direction for injury intervention and a reference for reducing the incidence of clinical CI-AKI further.
What problem does this paper attempt to address?