| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
a author for correspondence: fax 49/551/398551
The hepatic conversion of lidocaine to its metabolite monoethylglycinexylidide (MEGX) is catalyzed by the cytochrome P450 enzyme CYP3A4 in humans. This metabolic capacity of the liver led to the development of the MEGX test, in which serum MEGX concentrations are determined, as a relatively simple real-time test of liver function [1]. A large body of evidence currently documents the clinical utility of this test [2], demonstrating that MEGX concentrations <20 µg/L reflect a compromised hepatic function. Furthermore, several studies have shown that test results <10 µg/L are associated with an extremely high risk for pretransplant nonsurvival (3)(4). The assessment of serum MEGX concentrations <5 µg/L might be particularly relevant in pediatric patients with end-stage liver disease [4].
The automated fluorescence polarization immunoassay (FPIA; Abbott Diagnostics, N. Chicago, IL) of MEGX in serum is almost universally available and widely used. It is sensitive (detection limit: 3 µg/L) and easy to perform. However, the test is subject to interferences, most importantly from high bilirubin concentrations (5)(6). One approach to deal with this interference was proposed by Zoppi and Fumagalli, who precipitated the protein-bound bilirubin with the precipitation reagent from the Digoxin II assay supplied by Abbott (6). Given the lack of sensitive HPLC methods capable of covering the low MEGX values usually observed in patients with severe hyperbilirubinemia, this modification of the FPIA procedure has been neither compared with more-definitive methods, nor validated with patients' samples. In addition, lidocaine concentrations >10 mg/L cross-react in the FPIA (1), which can present problems in experiments with cell cultures or microsomal preparations and in assaying samples from patients whose blood still contains lidocaine from the test dose. Furthermore, OH-MEGX, which is formed from lidocaine in rats, also cross-reacts in the FPIA, thus precluding investigations with this species (7). To achieve analytical sensitivity comparable with that of the FPIA but with a superior specificity, many different HPLC protocols involving liquid–liquid or solid-phase extraction have been used (8). Using UV or electrochemical detection, however, investigators have not been able to accurately measure MEGX concentrations <10 µg/L (8).
We here describe a reliable and simple isocratic reversed-phase HPLC procedure for determining MEGX in serum. Based on the combination of HPLC separation and fluorescence detection—after derivatization of MEGX with 7-fluoro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-F; Aldrich Chemical Co., Steinheim, Germany)—this method demonstrates superior specificity and sensitivity, the lower limit of the working range being 2.5 µg/L.
For sample preparation we used a modification of the extraction procedure of O'Neal et al. (9). Briefly, a 500-µL aliquot of specimen and 250 µL of acetonitrile containing the internal standard monopropylglycinexylidide (MPGX), 200 µg/L, were mixed for 15 s in a 5-mL polypropylene tube. MPGX was synthesized according to a two reaction-step method, with use of 2,6-dimethylaniline, chloroacetylchloride, and propylamine (10). Saturated borate buffer (500 µL, 0.75 mol/L, pH 9.5) was added to the tube and mixed by vortex-mixing for 15 s. Dichloromethane (2 mL) was added, and the sample was centrifuged for 10 min at 4000g after 1 min of vortex-mixing. The supernatant was carefully aspirated and discarded. The lower organic layer was transferred into a clean polypropylene tube and evaporated in a vacuum centrifuge at 38 °C. The residue of the extraction was redissolved in 100 µL of a solution of saturated sodium tetraborate/methanol (2/1 by vol). We then added 10 µL of NBD-F solution [10 g/L NBD-F in ethanol/acetonitrile (3:1 by vol)], vortex-mixed for 3 s, and heated to 60 °C for 10 min. The reaction was stopped by adding 10 µL of HCl (5 mol/L) and cooling in ice water. Any precipitates formed were removed by centrifugation (5 min, 10 000g), and 50 µL of the clear supernatant was used for chromatography.
The HPLC system consisted of a chromatographic pump, an automated injector, a spectrofluorometric detector, a system controller linked to a PC (Shimadzu System LC-10A; Shimadzu, Kyoto, Japan), and a 250 mm x 4.6 mm (i.d.) reversed-phase column packed with Ultrasphere ODS (5-µm particle size; Beckman Instruments, Fullerton, CA). The column was maintained at 47 °C to improve separation. The mobile phase (flow rate 1.5 mL/min) was prepared by mixing 370 mL of acetonitrile, 40 mL of tetrahydrofuran, and 590 mL of potassium dihydrogen phosphate buffer (50 mmol/L, pH 5.0) and then degassing. Fluorescence was detected at 340 nm excitation and 520 nm emission. For calculation, we used the internal standard mode, with peak-height ratios. One-point calibration was routinely performed in each run by assaying an in-house-prepared MEGX calibrator (MEGX added to drug-free serum to a final concentration of 50 µg/L). Quality control was assessed by assaying commercially available serum specimens from Abbott, either undiluted (25 and 125 µg/L) or diluted with drug-free serum to 5 µg/L.
A chromatogram of a blank serum sample and two chromatograms of serum
samples from patients collected 15 min after a bolus intravenous
injection of lidocaine, 1 mg/kg of body wt., are shown in Fig. 1
. Retention times of MEGX and MPGX were 8.2 and 12.3 min,
respectively. Both substances were rapidly eluted as symmetrical peaks
and were fully separated at the baseline (Fig. 1
). The majority of the
front peaks and the changing pattern were caused by impurities in the
fluorescence reagent.
|
Because lidocaine cannot be derivatized, no interference from it is observed in this assay. 3-OH-MEGX was well separated from MEGX because of a considerably shorter retention time. The presence of endogenous interferences was evaluated by separate analysis of 92 MEGX-free patients' specimens. Interference by commonly administered drugs (acetaminophen, N-acetylprocainamide, amikacin, amitriptyline, amoxicilline, amphotericin B, bupivacaine, caffeine, carbamazepine, cephazolin, chloramphenicol, cimetidine, clemastine, clonazepam, cyclosporin A, desipramine, diazepam, digoxin, disopyramide, dopamine, ethosuximide, famotidine, gentamicin, imipramine, lamotrigin, lidocaine, mexiletine, mycophenolate mofetil, netilmicin, phenobarbital, phenytoin, physostigmine, prilocaine, primidone, procainamide, quinidine, rapamycin, salicylate, tacrolimus, theophylline, tobramicin, valproic acid, vancomycin) was evaluated by analyzing (a) patients' specimens received for routine therapeutic drug monitoring (TDM), (b) human TDM quality-control sera (Chiron Diagnostics, Fernwald, Germany), or (c) drug-containing methanol calibrators. None of the drugs interfered: They either did not react with NBD-F, were insufficiently retained by the column, or were not detected during a chromatographic run of 20 min.
The detection limit (signal-to-noise ratio = 3) was 1.7 µg/L.
The assay was linear over the working range, between 2.5 and 250 µg/L
(r >0.999). Performance characteristics were tested at
several concentrations of MEGX added to drug-free serum (Table 1
). Within-run imprecision was adequate (CV 11.5%) at the lower
limit of the working range (2.5 µg/L), better at higher
concentrations. [In contrast, the FPIA CV was 15.6% at 4 µg/L for
MEGX (n = 20).] The between-run CVs were 
12.7%. The deviation
from target values (Abbott controls: 25, 50, 100, and 200 µg/L) was
-4.8%, 2.6%, 4.2%, and 5.1%, respectively (n = 5). Analytical
recovery was calculated from the ratio of the peak heights for the
MEGX-supplemented serum samples and for sodium tetraborate/methanol
(2/1 by vol) solutions that contained the same amounts of MEGX, were
derivatized, and were directly injected onto the column. The
extraction efficiency was >88% for MEGX and 90.6% for the internal
standard MPGX (Table 1
).
|
We compared the present HPLC-fluorometric method and the TDx FPIA (Abbott) by assaying with both methods 96 serum samples from 51 patients routinely undergoing the MEGX test; samples that could not be directly analyzed with the FPIA because of high bilirubin concentrations (TDx reading: MX BKG) were pretreated with the Digoxin II precipitation reagent according to Zoppi and Fumagalli (6). According to regression analysis by Passing and Bablock (11), the HPLC (x) method correlated well with the FPIA (y) for analysis of the 96 patients' samples, of which 21% had MEGX values <10 µg/L: y = 0.974x + 1.308 (r = 0.987).
The purpose of the present study was to develop a highly sensitive and specific method for MEGX determination that would cover the complete clinically relevant decision range, including very low MEGX concentrations. Accordingly, we developed a derivatization procedure that substantially improved the sensitivity and specificity of MEGX determination, by using NBD-F, which is highly reactive with primary as well as secondary amino groups (12) (the latter being found in MEGX). The reaction is simple, short (10 min), and does not require special conditions or equipment. The reaction product is stable for at least 2 days at 4 °C in the dark, allowing repeat analyses within this period. The combination of a simple liquid–liquid extraction procedure with isocratic chromatography makes the method easy to perform in almost every laboratory. The internal standard MPGX can be readily synthesized, but as an alternative one can use the commercially available butanilicaine citrate (Hoechst AG, Frankfurt, Germany; data not shown). Use of the latter compound, however, increases the run time because of its later elution (~18 min) from the column. If required, an even lower detection limit can be achieved by increasing the sample volume to 1 mL.
We have now had experience with >800 chromatographic runs in one C18 column without any deterioration of the separation performance. The technician time required to process 30 samples is ~1.5 h, and each chromatographic run takes 15 min. This seems to be adequate even in laboratories with a high workload, a same-day turnaround being sufficient for most clinical indications. We conclude that this HPLC method is an attractive and cost-effective alternative to FPIA.
Acknowledgments
M. A. was supported by a grant from the Boehringer Ingelheim Fonds.
Footnotes
Abt. Klin. Chem., Zentrum Innere Med., Georg-August-Universität Göttingen, Robert Koch Str. 40, D-37075 Göttingen, Germany
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. Streit, P.-D. Niedmann, M. Shipkova, V. W. Armstrong, and M. Oellerich Rapid and Sensitive Liquid Chromatography-Tandem Mass Spectrometry Method for Determination of Monoethylglycinexylidide Clin. Chem., October 1, 2001; 47(10): 1853 - 1856. [Full Text] [PDF]
|
|
|
|
|
|
E. Schutz, M. Shipkova, P. D. Niedmann, E. Wieland, V. W. Armstrong, and M. Oellerich Determination of monoethylglycinexylidide by fluorescence polarization immunoassay in highly icteric serum samples: modified precipitation procedure and HPLC compared Clin. Chem., June 1, 1998; 44(6): 1269 - 1274. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
