固相萃取净化-超高效液相色谱-高分辨质谱法测定尿液中百草枯和敌草快残留

Determining the presence of paraquat (PQ) and diquat (DQ) in urine samples through physical and chemical testing is challenging. As PQ and DQ have characteristics such as high molecular polarity and good water solubility, they are difficult to be retained by conventional reversed-phase columns. Most of the methods in the literature use hydrophilic interaction chromatography (HILIC) for the retention of PQ and DQ, but they often require high concentrations of buffer salts as the mobile phase, which increase the contamination of the mass spectrometer. In view of the above problems, a rapid and accurate analysis method was developed for the determination of PQ and DQ residuals in urine samples based on weak cation exchange (WCX) solid-phase extraction (SPE) and ultra performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS) in this study. Urine samples were first diluted with phosphate buffer (pH=6.86) and pretreated using the WCX SPE method. Chromatographic separation was performed on a Syncronis HILIC column (100 mm×2.1 mm, 1.7 μm). An electrospray ion source in the positive (ESI(+)) mode and full mass-data dependent MS(2) (full mass-ddMS(2)) mode was used for quantification by matrix-matched external standard method. In this study, the concentration of ammonium formate in the mobile phase in the HILIC mode was effectively reduced to 10 mmol/L by the continuous optimization of the chromatographic conditions. MS optimization results indicated that the molecular ion (M(+·)) of PQ and DQ had the strongest response. In addition, sample pretreatment conditions were also optimized. The obtained results indicated that the hydrophobic polytetrafluoroethylene (PTFE) filter membrane, acetonitrile-water (1∶1, v/v) as a fixing solution, and polypropylene vials were suitable for PQ and DQ analysis. Under the optimal conditions, the linearity of PQ and DQ was good with correlation coefficients (r(2)) greater than 0.998. The limits of detection (LODs, S/N≥3) and limits of quantification (LOQs, S/N≥10) were 0.2 μg/L and 0.6 μg/L, respectively. Mean spiked recoveries of PQ and DQ at the four spiked levels (1.0, 20.0, 100.0, and 200.0 μg/L) were in the range of 85.8%-101% and 80.3%-86.9%, with the RSDs of 0.8%-5.1% and 0.9%-4.2%. The established method was employed for the analysis and confirmation of PQ and DQ for clinical poisoning cases. In one case, a 23-year-old male who had taken approximately 20 mL of pesticide orally was confirmed as DQ poisoning by the developed method. DQ concentration monitoring of the urine samples was conducted for this case during the clinical treatment process. The patient was successfully discharged from the hospital after five times of blood perfusion and other treatments until the DQ concentration was low in the urine samples. In conclusion, the method developed in this study based on WCX SPE-UPLC-HRMS can be used for the confirmation of poisoning cases and concentration monitoring during clinical treatment, providing strong technical support for clinical precision treatment. The method is rapid, simple, sensitive, and accurate, and it is suitable for the detection of PQ and DQ in urine samples.