Liquid chromatographic analysis of local anesthetics in human plasma after sample preparation by on-line dialysis. Optimization by use of experimental design

A fully automated method involving dialysis, clean-up and enrichment of the dialysate on a pre-column packed with a strong cation-exchange phase, and subsequent liquid chromatographic (LC) analysis with UV detection at 220 nm has been developed for the determination of local anesthetics (prilocaine, mepivacaine, and bupivacaine) in human plasma. All sample-handling operations were executed automatically by means of an Asted XI system. To identify the most important conditions affecting analyte recovery from the dialysis and trace-enrichment processes, a seven-factor D-optimal design with 16 experimental points was elaborated as a screening test A four-factor D-optimal design with 24 experimental points was then used to predict and optimize analyte recovery. Derringer's desirability function was also used to deduce optimum conditions for analyte recovery and dialysis time within the experimental domain. Finally, the method developed was validated. Mean recoveries were approximately 72% for bupivacaine and approximately 67% for mepivacaine and prilocaine. The limits of quantification were 28 ng mL(-1) for bupivacaine and 25 ng mL(-1) for mepivacaine and prilocaine.