Journal
METABOLOMICS
Volume 9, Issue 3, Pages 697-707Publisher
SPRINGER
DOI: 10.1007/s11306-012-0479-4
Keywords
Acute kidney injury (AKI); Metabolomics; NMR spectroscopy; Classification
Categories
Funding
- Bavarian Genomic Network (BayGene)
- German Federal Ministry of Education and Research (BMBF) [01 ER 0821]
- Regensburg School of Medicine
Ask authors/readers for more resources
Acute kidney injury (AKI) is a frequent complication of sepsis, major surgery or nephrotoxic medication use. It is associated with high morbidity and mortality. In an effort to identify novel biomarkers capable of predicting the development of AKI after cardiac surgery with cardiopulmonary bypass use, urine specimens were collected before and at 4 and 24 h after surgery from 106 patients and analyzed by means of nuclear magnetic resonance spectroscopy. Postoperative AKI of stage 1-3 as defined by the Acute Kidney Injury Network (AKIN) developed in 34 cases. Employing Quantile Normalization and support vector machine based classification, spectra of the 24-hour postoperative urine specimens were found to predict AKI across all stages with an average accuracy of 76.0 % and a corresponding area under the receiver operating characteristic curve of 0.83. Considering only AKIN-stage 2 and 3 patients, prediction accuracy increased to 81.7 % and 100 %, respectively. Among the small set of predictive biomarkers identified was carnitine, the urinary concentration of which was elevated significantly in AKI-free patients only, and tranexamic acid, which is routinely applied as an antifibrinolytic agent at the end of surgery, and whose renal excretion was delayed in AKI patients. The study underscores the power of NMR and bioinformatics in identifying novel biomarkers of disease and in gaining new insights into pathomechanisms.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available