Biomedical Engineering Reference
In-Depth Information
Nevertheless, isotope-dilution mass spectrometry (IDMS), and in particular IDMS-
GC-MS are looked upon as “matrix independent” reference methods of extremely
high accuracy, because all matrix-related effects are assumed to affect target ana-
lyte and labelled internal standard the similar manner [
47
]. However, it must be
noted that substantial isotope effects in the ionization efficacy may be observed
with LC-MS/MS, e.g. as previously described for the quantification of carvedilol
[
48
] or piperaquine in plasma [
49
]. In the first case the internal standard did not
completely co-elute with the analyte, which led to significant differences in ion
suppression effects in some specimens. In the later case residual amounts of trieth-
ylamine remaining in the sample solution after solid-phase extraction were found
to suppress the signals piperaquine and its six-fold deuterated internal standard
differently, leading to significant errors in the analyte concentrations.
Furthermore, if high ionization temperatures are employed (particularly in APCI),
hydrogen-deuterium exchange of deuterated internal standard compounds may occur
during the ionization process [
50
]. For that reason, but also due to its typical location
in the molecular “backbone” of an analyte (minimizing the influence on the electro-
negativity distribution of the scaffold),
13
C atoms are considered the more reliable
label for isotope dilution internal standardization compared to deuterium.
Unfortunately, for the majority of small molecule LC-MS/MS analyses, stable
isotope labelled internal standards are not available so far. In such cases, com-
pounds with a very similar molecular structure typically serve as internal standard
(“homologues” or “analogues”). Since the ionization properties are substantially
determined by functional groups of a molecule, ionization behaviour may differ
significantly—even between compounds with very similar over-all molecular
structure. Differential clustering, e.g. with sodium, ammonium or formate ions
often present in mobile phases may as well impact the parity of ionization yield
between analyte and internal standard. Hence the availability of an appropriate
homologue is crucial and critical for the development of reliable LC-MS/MS meth-
ods in TDM [
51
] .
A fundamental requirement for LC-MS/MS calibration materials is that matrix
effects exerted by these materials are most similar to the matrix effects exerted by
actual patients' sample materials. Lyophilisation, virus inactivation and other pro-
cedures applied during the industrial production of calibration and control materi-
als, may notably impact the ionization behaviour of extracts from such samples
and can result in differential matrix effects in calibrators and actual patients' sam-
ples. If the internal standard peak areas found for calibration samples systemati-
cally differ from those found in patients samples, inappropriateness of the
calibration materials should be suspected. However, we have previously observed
that calibration materials from different commercial sources lead to inaccurate tac-
rolimus results in an instrument specific manner, without showing deviations in the
internal standard peak area. This effect was most likely related to ionization
enhancement affecting the target analyte but not the homologue internal standard
(ascomycin) ionization
and
being restricted to calibrator samples. This resulted in
systematically low tacrolimus results of clinical samples in one instrument for one
speci fi c calibrator lot [
52
] .
