Page 118 - Advances in Forensic Applications of Mass Spectrometry - Jehuda Yinon

1522_C02.fm Page 103 Wednesday, November 12, 2003 9:36 AM

noaffinity extraction columns and determined with LC-ESI-MS, with a LOD
of 2.5 mg/l.

2.6 Screening LC/MS Procedures in Forensic Toxicology

The application of HPLC as a tool of general unknown analysis is relatively
new and is still in the developmental stage. Generally, two approaches were
taken: the use of HPLC with diode array detection (HPLC/DAD) or LC/MS.
HPLC/DAD has a high identification potential due to the combination of
retention parameter and UV spectrum as identification parameters. Several
databases comprising more than one thousand substances were established.
5
The review of these databases was done by Bogusz. The establishing of a
universally applicable HPLC database was hindered by poor interlaboratory
reproducibility of retention time values for the same drugs analyzed in nom-
inally identical conditions in different laboratories. This problem was solved
183
by introduction of retention index scale, using alkyl aryl ketones or
184
1-nitroalkanes. In a further development, 1-nitroalkanes were replaced by
acidic and basic standard drugs with known retention index values. This
approach assured very good interlaboratory comparison of results. 185,186
Beside retention index scaling, relative retention times are used for
HPLC/DAD libraries. Such a system, based on two reference compounds, is
commercially available as REMEDi (Bio-Rad Labs, Hercules, CA). This sys-
tem comprises automatic extraction of a urine sample, on-line separation,
detection, and identification with a library comprising over 800 drugs. The
database can be expanded according to need.
The use of LC/MS for toxicological screening was from the beginning an
extremely attractive possibility since this technique possesses a much broader
detection spectrum than GC-MS and is much more sensitive than HPLC-
DAD. However, the establishing of a generally applicable LC/MS library,
similar to GC/MS databases, is hindered by large interlaboratory variability
in mass spectra. In the study done in three laboratories, mass spectra of
identical substances, analyzed on the same instruments in nominally identical
187
conditions, showed large differences in the degree of fragmentation. On
the other hand, controlled changes in composition of the mobile phase, i.e.,
in the percentage of organic modifier or in the molarity of ammonium
formate buffer, did not have any relevant influence on mass spectra. Existing
LC/MS screening procedures may be divided into two groups: those com-
prising substances belonging to similar pharmacological classes or used for
similar applications (e.g., pesticides) and those applicable for an undirected
search, i.e., for a “general unknown” screening. In the first case, the task is
much easier, since the number of substances is limited, and the sample

© 2004 by CRC Press LLC