Purpose: Septic acute kidney injury (AKI) is associated with poor outcome. This can partly be attributed to delayed diagnosis and incomplete understanding of the underlying pathophysiology. Our aim was to develop an early predictive test for AKI based on the analysis of urinary peptide biomarkers by MALDI-MS. Experimental design: Urine samples from 95 patients with sepsis were analyzed by MALDI-MS. Marker search and multimarker model establishment were performed using the peptide profiles from 17 patients with existing or within the next 5 days developing AKI and 17 with no change in renal function. Replicates of urine sample pools from the AKI and non-AKI patient groups and normal controls were also included to select the analytically most robust AKI markers. Results: Thirty-nine urinary peptides were selected by cross-validated variable selection to generate a support vector machine multidimensional AKI classifier. Prognostic performance of the AKI classifier on an independent validation set including the remaining 61 patients of the study population (17 controls and 44 cases) was good with an area under the receiver operating characteristics curve of 0.82 and a sensitivity and specificity of 86% and 76 respectively. Conclusion and clinical relevance: A urinary peptide marker model detects onset of AKI with acceptable accuracy in septic patients. Such a platform can eventually be transferred to the clinic as fast MALDI-MS test format.
|Number of pages||11|
|Journal||Proteomics Clinical Applications|
|Publication status||Published - 1 Jul 2016|
FingerprintDive into the research topics of 'Development of a MALDI MS-based platform for early detection of acute kidney injury'. Together they form a unique fingerprint.
- Division of Biomedical Sciences - Reader in Bioinformatics
Person: Academic Research Active