Abstract: FR-PO1138
Identification of COVID-19 Disease and Impact on Kidney Function Using Urine Raman Spectroscopy and Chemometrics
Session Information
- COVID-19 - II
November 03, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Coronavirus (COVID-19)
- 000 Coronavirus (COVID-19)
Authors
- Senger, Ryan S., Virginia Polytechnic Institute and State University College of Engineering, Blacksburg, Virginia, United States
- Gomez de la Espriella, Mariana, Carilion Clinic, Roanoke, Virginia, United States
- Jackson-Akers, Jasmine Y., Carilion Clinic, Roanoke, Virginia, United States
- Sayed Issa, Amr, Virginia Polytechnic Institute and State University College of Engineering, Blacksburg, Virginia, United States
- Robertson, John L., Virginia Polytechnic Institute and State University College of Engineering, Blacksburg, Virginia, United States
- Fazili, Tasaduq, Carilion Clinic, Roanoke, Virginia, United States
Background
COVID19 can have a deleterious effect on renal function. Timely prediction of renal dysfunction is not achievable with available laboratory armamentarium, and scores currently used for severity of illness are highly provider dependent and subjective in nature.
Methods
We developed a Raman spectroscopic technology (Rametrix® molecular urinalysis) to detect COVID19 systemic/renal effects by analysis of patient urine. It is based on chemometric analysis of the Raman spectrum of urine and detects metabolomic differences. The technology is not designed to detect virus/viral components.
We hypothesized that COVID19 disease would alter urine composition and that Rametrix® analysis could detect renal dysfunction via urine molecular ‘fingerprinting’.
We applied Rametrix® analysis on 85 urine specimens from 64 patients hospitalized for COVID19 disease (Omicron variant prevalent). Patients were 30-92 years of age (median age 67), and ca. 50%/50% M/F. Collections were done at admission (n=66), discharge (n=7), and follow-up (n=12). Medical record and laboratory data were correlated with Rametrix® results; 66 patients had GFR values at both admission and discharge. Disease severity scores were collected repeatedly during hospitalization.
Using chemometric analysis, we compared hospitalized patient COVID19 urine spectra with urine spectra from healthy controls (pre-COVID19), patients with CKD (pre-COVID), asymptomatic-mildly symptomatic outpatients (2020/mid-2021 variants), bladder cancer patients (pre-COVID19) and Lyme disease patients (pre-COVID19).
Results
Rametrix® molecular urinalysis distinguished COVID19-associated changes in urine composition with predictive metrics (accuracy, sensitivity, specificity, PPV, and NPV) ranging between 93-96.5%.
A correlation between changes in urine Raman spectra and physician assessment of disease severity was also found through computational analysis.
Spectra from patient urine collected at admission and stratified based on eGFR, demonstrated molecular ‘fingerprints’ predictive of future renal dysfunction in 91.7% of high-risk patients.
Conclusion
Raman molecular urinalysis can be a useful tool in detecting and predicting future renal dysfunction in patients hospitalized with COVID19 disease.
Funding
- Clinical Revenue Support