This report is dedicated to development of surface-enhanced Raman spectroscopy (SERS) based analysis protocol for detection of antibiotics in urine. The key step of the protocol is the pretreatment of urine before the detection to minimize background signal. The pretreatment includes extraction of intrinsic urine components using aluminum hydroxide gel (AHG) and further pH adjusting of the purified sample. The protocol was tested by detection of a single antibiotic in artificially spiked samples of real urine. Five antibiotics of cephalosporin class (cefazolin, cefoperazone, cefotaxime, ceftriaxone, and cefuroxime) were used for testing. SERS measurements were performed using a portable Raman spectrometer with 638 nm excitation wavelength and silver nanoparticles as SERS substrate. The calibration curves of four antibiotics (cefuroxime is the exception) cover the concentrations required for detection in patient's urine during therapy (25/100‒500 μg/mL). Random error of the analysis (RSD < 20%) and limits of quantification (20‒90 μg/mL) for these antibiotics demonstrate the applicability of the protocol for reliable quantitative detection during therapeutic drug monitoring. The detection of cefuroxime using the protocol is not sensitive enough, allowing only for qualitative detection. Additionally, time stability and batch-to-batch reproducibility of AHG were studied and negative influence of the pretreatment protocol and its limitations were estimated and discussed.
Keywords: cefazolin; cefoperazone; cefotaxime; ceftriaxone; cefuroxime; sample pretreatment; silver nanoparticles; surface-enhanced Raman spectroscopy; therapeutic drug monitoring.