| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Clinical Chemistry, George-August University Goettingen, 37075 Goettingen, Germany
aaddress correspondence to this author at: Abteilung Klinische Chemie, Zentrum Innere Medizin, Georg-August-Universität Göttingen, Robert-Koch-Strasse 40, 37075 Göttingen, Germany; fax 49-551-398551, e-mail varmstro{at}med.uni-goettingen.de
A large body of evidence currently documents the utility of the
MEGX test for real-time assessment of liver function in
transplantation, critical care medicine, and various experimental
models (1). The test is based on the conversion of lidocaine
to its deethylated metabolite monoethylglycinexylidide (MEGX),
primarily through the hepatic cytochrome P450 system. In the standard
MEGX test, an intravenous bolus of a small lidocaine dose (1 mg/kg) is
administered over 2 min. Blood specimens are collected for serum MEGX
determination both before and at 15 and/or 30 min after lidocaine
administration. The most commonly used method to measure serum MEGX has
been an automated fluorescence polarization immunoassay (Abbott
Laboratories) with a detection limit of 3 µg/L. This test, however,
is no longer commercially available. HPLC methods with ultraviolet
detection (2)(3) and gas chromatographic (GC)
procedures with ionization or nitrogen-phosphorous detection
(4)(5) were originally reported, but all of
these techniques had a limit of quantification >10 µg/L. Several
studies have shown that transplant candidates with MEGX test results
<10 µg/L have a particularly poor 1-year survival rate
(6)(7)(8). An improved HPLC method with fluorescence detection
(9) and a capillary GC method with nitrogen-phosphorus
detection (10) were therefore developed that achieved an
adequate analytical sensitivity with limits of detection of 
1–2
µg/L. The disadvantage of the HPLC method with fluorescence detection
is the necessity to derivatize MEGX.
Because of its flexibility and high specificity, liquid chromatography–tandem mass spectrometry (LC-MS-MS) is finding increasing application for the quantification of numerous analytes. We now describe a reliable, simple, sensitive, and rapid procedure for determining MEGX in serum by LC-MS-MS. This procedure also allows the simultaneous measurement of serum lidocaine concentrations in the same sample.
MEGX hydrochloride and lidocaine hydrochloride were kind gifts from Astra (Stockholm, Sweden). The internal standard monopropylglycinexylidide (MPGX) was synthesized as described previously (9). HPLC-grade methanol and ammonium acetate were obtained from Merck. The calibrator and in-house controls were prepared in drug-free serum (Bio-Rad) from stock solutions of MEGX in deionized water. The final concentration of MEGX in the single-point calibrator was 75 µg/L, and the final MEGX concentrations in the four in-house controls were 5, 25, 50, and 125 µg/L. To quantify lidocaine, we used a single-point calibrator with a final concentration of 1.1 mg/L and commercial controls (Abbott Laboratories) with final concentrations of 1.5, 3.0, and 7.5 mg/L.
For sample preparation, 100 µL of calibrator, quality-control sample, or patient sample was vortex-mixed with 200 µL of methanol containing the internal standard MPGX (50 µg/L) for 30 s in 1.5-mL polypropylene tubes. After centrifugation for 10 min at 4000g, the supernatants were decanted and, after recentrifugation for 1 min, were placed in a Series 200 autosampler (Perkin-Elmer).
The column was an Oasis® HLB extraction column
(2.1 x 20 mm; Waters) maintained at 50 °C with a DuPont column
oven. The LC-MS-MS system consisted further of a Series 200 binary pump
from Perkin-Elmer, an M480 pump (Dionex) and a six-port Rheodyne valve.
Sample injection (20 µL) was by a Series 200 autoinjector fitted with
a 200-µL sample loop. The column was washed for 1 min (flow rate, 800
µL/min) with methanol–30 mmol/L ammonium acetate (20:80 by volume),
followed by a 2.5-min elution step (flow rate, 1000 µL/min) with
methanol–30 mmol/L ammonium acetate (75:25 by volume). The column was
then reequilibrated for 0.5 min (flow rate, 800 µL/min) with
methanol–30 mmol/L ammonium acetate (20:80 by volume) in preparation
for the next injection. In our experience, a single Oasis column can be
used for 300 injections.
For detection a Sciex API 2000 triple quadrupole mass spectrometer with a turbo-ion spray (heated electrospray) interface (PE Applied Biosystems) was used. The analytes that eluted from the HPLC were introduced into the turbo-ion spray source (heated to 450 °C) at a split of 1:10. High-purity argon was used as the collision gas. Ionization was achieved in the positive-ion mode with an ionization voltage of 2200 V, an orifice voltage and collision energy of 19 eV, and a heater probe temperature of 450 °C. The first quadrupole was set to select the protonated ions [M + H+] of MEGX (m/z 207.0), MPGX (m/z 221.2), and lidocaine (m/z 235.2). The second quadrupole was used as collision chamber. The third quadrupole was used to select the characteristic product ions of MEGX (m/z 58.0), MPGX (m/z 72.0), and lidocaine (m/z 86.0). The elution times for MEGX, MPGX, and lidocaine were 1.7, 1.7 and 1.8 min, respectively.
A PowerMac personal computer running PE Sciex Sample Control (Ver. 1.4) software was used to control the LC-MS-MS and to record the output signals from the detector. Integration of peak areas, calculation of peak-area ratios, calculation of the calibration line, and calculation of the MEGX concentrations were performed with the PE Sciex TurboQuanTM (Ver. 1.0) software.
As a result of the tandem mass spectrometric approach, interference from 3-hydroxy-MEGX (m/z 223.2) or other commonly administered drugs is excluded. The metabolite 3-hydroxy-MEGX, which cross-reacts in the fluorescence polarization immunoassay, is a minor metabolite in humans (11), but it can make a major contribution to lidocaine metabolites in some animal models (12). The lower limit of quantification was set at 1 µg/L (CV <15%; n = 20). The assay was linear over the working range (1–500 µg/L; r >0.9999). To confirm linearity over the working range, drug-free serum was enriched with MEGX from a weighed-in stock solution to final concentrations of 1, 250, and 500 µg/L. These samples were then measured in duplicate on 5 separate days, using the single-point calibration. Measured mean (SD) MEGX concentrations were 1.15 (0.1) µg/L, 253.1 (12.5) µg/L, and 524.7 (25.6) µg/L, respectively.
We evaluated the performance characteristics, using the in-house
controls prepared by enriching drug-free serum with MEGX to the desired
concentration from a stock solution. The four control samples covered
the range of MEGX values that are usually encountered in clinical
practice (Table 1
). For the within-run imprecision, each control sample was
extracted 20 times in one batch. For between-run imprecision, the same
four controls were extracted and measured over 8 working days. The
analytical recovery was also calculated from the same four controls.
The measured values for the in-house control samples were within 5% of
the nominal values. The between- and within-run CVs were comparable to
those observed for the HPLC-fluorometry method (9). In the
case of lidocaine, the within-run imprecision at 1.5, 3.0, and 7.5 mg/L
was 4.5%, 3.6%, and 5.4%, respectively. The corresponding recoveries
were 98%, 95%, and 91%, respectively. To test the stability of the
internal standard (MPGX) and MEGX in the extraction solvent, four
samples were extracted with methanol, and the methanolic extracts were
stored at ambient temperature (20 °C) in closed vials for 3 days.
The differences between the stored samples and the samples that were
measured immediately averaged 1%. The effect of storage at 4 °C on
the stability of MEGX in serum was tested with 10 patient samples (MEGX
concentration, 21–75 µg/L). When we reanalyzed the samples after 9
days, there was good correlation between the initial and stored sample
values (r2 = 0.99; Passing–Bablok
regression line, y = 1.02x - 1.06).
The differences between the stored samples and the samples that were
measured immediately averaged 0.08 µg/L (95% confidence interval,
-1.11 to 1.28 µg/L).
|
To investigate ion suppression, we performed the following experiments. Five MEGX-free serum samples from five patients with hyperbilirubinemia (bilirubin >342 µmol/L) and five patients with bilirubin values <20.5 µmol/L were extracted with methanol as described above. The methanolic extracts were then enriched with MEGX (final nominal concentration, 30 µg/L) and internal standard (MPGX) from a stock solution. Reference methanolic solutions containing the same nominal concentrations of MEGX and MPGX were also prepared. The extracts and reference solutions were injected onto the analytical column, and the peak areas for MEGX and MPGX were compared. The mean (SD) ratio of the serum extract peak area to the reference peak area was 0.41 (0.05) for the internal standard and 0.44 (0.06) for MEGX. Thus, although ion suppression occurred, the extent of this suppression was similar for the internal standard and MEGX. No differences were observed between the hyper- and normobilirubinemic samples.
We analyzed 106 serum samples (from sampling time points at 15 and 30
min) from 53 patients who were routinely undergoing the MEGX test by
the LC-MS-MS method and by the HPLC-fluorometry procedure. A method
difference plot (Fig. 1
) revealed good agreement between both methods over the
clinically relevant range of MEGX concentrations. The mean absolute
difference between the two methods was -0.6 µg/L. The equation for
the Passing–Bablok regression line was: y =
0.9845x + 1.6784 (r2 =
0.96; 95% median distance of the residuals of the Passing–Bablok
regression, 6.02 µg/L). The relative difference between the two
methods tended to be greater at MEGX concentrations <20 µg/L, which
probably reflects the greater between-run imprecision of the two
methods in the low MEGX range. However, these differences were not of
major clinical relevance. The LC-MS-MS concentrations for the four
samples whose values were outside the 95% confidence interval were
confirmed by repeat analysis. There was insufficient material to repeat
the measurements with the HPLC-fluorometry procedure.
|
In conclusion, the LC-MS-MS method described here for measurement of
MEGX in serum samples is rapid, reproducible, specific, and requires
only a small sample volume (100 µL). The procedure offers substantial
advantages over available methods with a comparable limit of
quantification. The HPLC-fluorometry (9) and capillary GC
(10) methods require sample volumes of 0.5 and 1.0 mL,
respectively, to achieve a similar analytical sensitivity. The analysis
time for the LC-MS-MS is also considerably shorter (3 min) than those
(10–15 min) of the HPLC-fluorometry and capillary GC methods.
Furthermore, sample preparation for LC-MS-MS is much simpler than that
required for the other two methods. The technician time to process 30
samples is 0.5 h. The LC-MS-MS procedure also has the advantage that
serum lidocaine concentrations can be measured concomitantly in the
same sample to confirm that an appropriate lidocaine dose was used for
the MEGX test. In our experience, serum lidocaine concentrations 15 min
after a lidocaine dose of 1 mg/kg range from 0.4 to 3.5 mg/L in
patients undergoing the MEGX test. Thus, despite the high initial
capital cost, this LC-MS-MS procedure is an attractive and
cost-effective alternative to existing methods.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
