Quantifying creatinine and urea in human urine through Raman spectroscopy aiming at diagnosis of kidney disease

Abstract

Due to their importance in the regulation of metabolites, the kidneys need continuous monitoring to check for correct functioning, mainly by urea and creatinine urinalysis. This study aimed to develop a model to estimate the concentrations of urea and creatinine in urine by means of Raman spectroscopy (RS) that could be used to diagnose kidney disease. Midstream urine samples were obtained from 54 volunteers with no kidney complaints. Samples were subjected to a standard colorimetric assay of urea and creatinine and submitted to spectroscopic analysis by means of a dispersive Raman spectrometer (830 nm, 350 mW, 30 s). The Raman spectra of urine showed peaks related mainly to urea and creatinine. Partial least squares models were developed using selected Raman bands related to urea and creatinine and the biochemical concentrations in urine measured by the colorimetric method, resulting in r = 0.90 and 0.91 for urea and creatinine, respectively, with root mean square error of cross-validation (RMSEcv) of 312 and 25.2 mg/dL, respectively. RS may become a technique for rapid urinalysis, with concentration errors suitable for population screening aimed at the prevention of renal diseases.

Raman spectroscopy: techniques and applications in the life sciences

“…PLS has been used by Bechtel et al to quantify various components in tissue phantoms [498], by Mekisiarun et al to measure the level of saturation of fat in milk [499], by Saatkamp et al to quantify levels of creatinine and urea in urine [500], and by Shih et al to extract blood glucose concentration [501]. PLS can also be trained to predict non-chemical properties of the sample.…”

Multicomponent analysis using a confocal Raman microscope

“…The Root Mean Square Error (RMSE) (the error between the reference concentrations and the predicted concentration) and the coefficient of determination (R 2 ) (the correlation between the predicted and reference values), are two important metrics that are commonly employed to determine the appropriate number of components to use and to estimate the error of the model in terms of predicting the concentrations of an unknown mixture [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. In the experiment presented here, RMSE and R 2 are used to evaluate the performance of the model based on leave-one-out cross-validation.…”

“…In order to overcome these drawbacks, Raman spectroscopy has been proposed to quantify multiple components simultaneously and in realtime with the advantage of small volume sampling and less sample contact. [1][2][3][4][5][6][7][8][9][10][11][12] A key advantage of Raman spectroscopy is that it is nondestructive; the sample can be reused for further analysis following inspection with Raman spectroscopy. Multivariate statistical analysis of the recorded spectra is central to the approach; most commonly, Partial Least Square (PLS) regression [13] is used to provide a predictive model that can estimate the relationship between a set of independent variables (peak areas in the Raman Spectra) and dependent variables (chemical concentrations).…”

Dual-triggered nanoaggregates of cucurbit[7]uril and gold nanoparticles for multi-spectroscopic quantification of creatinine in urinalysis

“…Urine is one of the major body fluids from which information on metabolism of the body and renal function can be extracted. 1,2 It is readily available and can be obtained in a non-invasive way as opposed to blood serum. Creatinine (CRN), which is the end product of muscle metabolism, is a very important biomarker in clinical diagnosis, for example, it can be used to monitor the kidney filtration function in renal clearance tests.…”