Evaluation of Pre-Analytical Processes on Lipemic Whole Blood Samples Used in Forensic Toxicology

Sammanfattning: Introduction: Post-mortem whole blood samples differ greatly in quality, lipemia is one cause of concern in toxicological analyses. Around 4 % of all samples sent to RMV are given a notation of lipemic content. The aim of the thesis was to study the effects of lipemia on the quantification of 14 benzodiazepines and 5 similar sedative and antianxiety drugs as well as evaluate the pre-analytical process aiming to reduce the effects of lipemia.  Methods: Blood samples were simulated with bovine blood, analyte spiking, and lipid spiking with either the nutrition emulsion Intralipid or with a mixture of post-mortem lipids from authentic samples. The outset was the by RMV currently used LLE method followed by UPLC- MS/MS and the extraction method was altered and evaluated. Matrix effects were also studied.  Results: Lipemia were found to be a great interference when quantifying benzodiazepines. For most analytes, internal standard could compensate for the loss of analyte but there was a problem with analytes not having their own IS. The 7-amino-compounds were greatly affected by lipemia and propiomazine and dihydropropiomazine showed extreme losses. Equilibration of IS did not result in similar loss as analyte. Dilution of sample reduced losses caused by lipemic content. SPE resulted in extracts free from lipids and high yields but there were analyte losses similar to LLE. No matrix effects from the lipids were found. Samples spiked with Intralipid gave poorer analyte yields than those spiked with post-mortem lipids.  Conclusion: Dilution is the most successful method to reduce pre-analytical matrix effects as long as the concentration is not so low that it risks getting lower than the analytical limits when doing so. Not homogenising samples before sampling is giving incorrect results. SPE could, if optimised for the analyte retention and elution, remove lipids from samples and obtain accurate analyte concentrations. Pooling lipids from post-mortem samples is a possible method for simulating lipemic whole blood. Intralipid and the PM-mix gave the same indications, but to different extents. Further studies where the ability to mimic authentic lipids are needed for both Intralipid and PM-mix.