Abstract: FR-PO118
Measurements of Volatile Organic Compounds by Proton-Transfer-Reaction Mass Spectrometry for the Diagnosis of AKI in the Intensive Care Unit
Session Information
- AKI Clinical: Outcomes and Biomarkers
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Acute Kidney Injury
- 003 AKI: Clinical and Translational
Authors
- Janech, Michael G., Medical University of South Carolina, Charleston, South Carolina, United States
- Mohialdeen, Mohammed Zuhair, Medical University of South Carolina, Charleston, South Carolina, United States
- Achanti, Anand, Medical University of South Carolina, Charleston, South Carolina, United States
- Budisavljevic, Milos N., Medical University of South Carolina, Charleston, South Carolina, United States
- Velez, Juan Carlos Q., Ochsner Clinic Foundation, New Orleans, Louisiana, United States
- Karakala, Nithin, University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
- Arthur, John M., University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
- Lee, Peter A., College of Charleston, Charleston, South Carolina, United States
Background
Current biomarkers for acute kidney injury (AKI) in the intensive care unit (ICU) do not offer real-time detection capability, are based on serum creatinine or urine proteins, and diagnostic levels significantly lag behind the injury. A lesser explored area of AKI biomarker research is the volatile organic compound (VOC) space. The goal of this pilot study was to assess whether detection of VOCs using a mass spectrometer conducive to near real-time detection capability could detect urine VOCs and classify or characterize patients with AKI in the ICU.
Methods
Urine specimens (40 mL) from 32 ICU subjects with or without AKI (AKIN criteria) were collected at bedside and transferred to an adjacent laboratory for Proton-Transfer-Reaction Mass Spectrometry (PTR-MS)-based analysis. VOCs were detected in positive hydronium ion mode and spectra were background subtracted and aligned. Individual VOCs were assessed for ability to classify AKI or no AKI using area under ROC (AUC) curves. Parametric statistics were utilized for individual VOC comparisons between AKI and non-AKI groups.
Results
105 VOC masses were detected in urine from 32 ICU patients. Mean serum creatinine values were 3.3 ± 3.0 mg/dL for the AKI group (n=14) and 0.9 ± 0.3 mg/dL for the non-AKI (n=18) group. Ten masses comprised 98% of the signal intensity and were less than 63 m/z suggesting that more volatile components were better detected. Eighteen masses were significantly lower in AKI subjects (1.2 to 13 fold lower, P<0.05). A single VOC, putatively identified by molecular weight as formic acid or ethanol (47.2 m/z), was elevated 2-fold in AKI patients compared to non-AKI patients (AUC = 0.70; p=0.03). When the ratio of 47.2 and 70.9 m/z was calculated, this resulted in a better classifier (AUC = 0.78) than 47.2 m/z alone.
Conclusion
PTR-MS has the capability to detect urine VOCs. Elevated urinary formic acid/ethanol combined with a reduction in several VOCs may provide value for early real-time detection of AKI in the ICU. This platform may provide diagnostic and prognostic information during AKI.
Funding
- Other NIH Support