Selective solid-phase extraction using molecularly imprinted polymers for analysis of venlafaxine, O-desmethylvenlafaxine, and N-desmethylvenlafaxine in plasma samples by liquid chromatography–tandem mass spectrometry

• Novel molecularly imprinted polymer (MIP) for selective extraction of venlafaxine and metabolites from plasma samples.
• The microdisc packed with MIP (8 mg) was reused for more than 60 times.
• The MISPE/LC–MS/MS method presented LLOQs values at traces levels.
• The method was successfully applied to determine VEN, ODV, and NDV in plasma samples from depressive patients undergoing therapy with VEN.
• This method was suitable for analysis of VEN in human plasma samples for pharmacokinetic study in therapeutic drug monitoring.

This paper focuses on the development of a novel miniaturized molecularly imprinted solid-phase extraction (MISPE) and ultra high performance liquid chromatography-tandem mass spectrometry (UHPLC–MS/MS) method to determine venlafaxine (VEN), O-desmethylvenlafaxine (ODV), and N-desmethylvenlafaxine (NDV) in plasma samples. The molecularly imprinted polymer (MIP) was prepared by the precipitation polymerization approach; VEN, metacrylic acid, ethylene glycol dimethacrylate, 2,2-azobisisobutyronitrile, and toluene were used as template, monomer, crosslinker, initiator, and porogen solvent, respectively. MIP and of the non-imprinted control polymer (NIP) sorbents were characterized by Fourier transform infrared spectroscopy and scanning electron microscopy. MIP phase presented higher extraction efficiency (MISPE, using plasma samples spiked with VEN) than the NIP phase (84 and 49% recovery rates, respectively). Analysis of other antidepressants with different chemical structures by MISPE-UHPLC–MS/MS attested to the selectivity of the developed MIP. The developed method presented precision assays with coefficients of variation (CV) smaller than 15%; accuracy assays with relative standard error (RSE%) values ranging from −12 to 16%, and linear ranges from 3 to 700 ng mL−1 for VEN, from 5 to 700 ng mL−1 for ODV, and from 3 to 500 ng mL−1 for NDV. The coefficients of determination (r2) were higher than 0.995. The lack-of-fit test also attested to the linearity of this method. This method was successfully applied to determine VEN, NDV, and ODV in plasma samples from depressed patients undergoing therapy with VEN.