Comparison of SERS Substrates in Direct Analysis of Urine Samples for Detection of Albuminuria: Relatively Large Gold Nano-Spheroids Outperform Gold Nanorods on Gold Film.

Proteins found in human urine can be used as a source of information on kidney function and as possible biomarkers for diseases such as chronic kidney disease (CKD). Focusing on differentiating between high and low albumin concentrations in urine samples, this study examines the roles of various substrates in detecting albuminuria as a specific form of proteinuria using surface-enhanced Raman spectroscopy (SERS). Using gold nanorods (AuNRs) and gold spheroidal nanoparticles (AuNPs) on various substrates to optimize the identification of albuminuria, we studied urine samples from 77 and 124 patients with albumin concentrations above and below 0.3 g/L. SERS spectra were analyzed using principal component analysis-linear discriminant analysis (PCA-LDA), partial least squares with linear discriminant analysis (PLS-LDA), and support vector machine (PLS-SVM)-to differentiate between high- and low-urinary-albumin groups. PLS-LDA was the best-performing method among the three methods. The largest of all tested nanoparticles, 200 nm diameter (D200 nm) AuNPs on gold substrates, performed best in distinguishing between high and low concentrations of albumin in urine, with a maximum average area under the curve (AUC) of 0.97, 100% sensitivity, 92% specificity, and 96% accuracy and a higher enhancement factor of 5.4 × 10