Programmable flow-based dynamic sorptive microextraction exploiting an octadecyl chemically modified rotating disk extraction system for the determination of acidic drugs in urine

Abstract

A novel automatic sorptive microextraction approach combining sequential injection-based programmable flow with rotating disk sorptive extraction (RDSE) is proposed for the clean-up and concentration of low polarity organic species in urine samples. Non-steroidal anti-inflammatory drugs (NSAIDs), namely, ketoprofen, naproxen, diclofenac and ibuprofen, were selected as model analytes in a proof-of-concept design, and they were further determined by liquid chromatographic (LC) assays. The extracting phase consisted of octadecyl (C18) chemically bonded silica embedded in a polytetrafluoroethylene (PTFE) substrate. The thin film was immobilized onto the surface of an in-house prepared rotating PTFE disk in a dedicated flow-through chamber. The programmable flow-based microextraction method operates under kinetic principles and features software-controlled sample loading and dynamic sorptive unidirectional-flow microextraction for as little as 10min, followed by matrix clean-up and in-line elution with methanol. The hydrophobic thin-film extracting phase was demonstrated to be reusable for at least 15 consecutive extractions in urine without removing or changing the disk. The relative recoveries of the NSAIDs in urine ranged from 101 to 106% using a matrix-matched calibration curve, with extraction efficiencies of 30–38% using a dynamic regime, an enrichment factor of approximately 17 for 10mL sample and relative standard deviations (RSD) between 3 and 6%. The detection limits (3×S/N ratio) of the in-line sample preparation method coupled to LC-UV detection ranged from 0.022 to 0.044mgL−1. Using NSAID monitored in urine from individuals who received oral administration of ibuprofen and diclofenac, the automatic sample handling method was proven to be efficient for urine clean-up and the determination of acidic drugs at biologically relevant levels.