| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Université Claude Bernard Lyon 1 (U.C.B.L.), Institut des Sciences Pharmaceutiques et Biologiques (I.S.P.B.), Département de Pharmacie Clinique, de Pharmacocinétique et d’Evaluation du Médicament, 8 Avenue Rockefeller, 69373 Lyon cedex 08, France
2
Hôpital Neuro-Cardiologique, Laboratoire de Dosage des Médicaments, 59 Boulevard Pinel, 69394 Lyon cedex, France
a author for
correspondence: fax 334-72-35-73-31, e-mail
roselyne.boulieu{at}chu-lyon.fr
Sufentanil (SF) is a highly potent synthetic opioid (1) used extensively in anesthesia (2)(3). Most of the techniques proposed for SF analysis in plasma samples are based on the RIA method described by Michiels et al. (4). Some methods using gas chromatography–mass spectrometry (GC-MS) have been reported (5)(6)(7). Plasma concentrations of SF in the analgesic range are extremely low, and pharmacokinetic studies require the development of sensitive GC-MS assay methods.
Reliable and reproducible isolation of SF from plasma is the most important and most critical step in pharmacokinetic analysis. Among different extraction methods currently used in pharmaceutics, the use of solid-phase extraction (SPE) has grown dramatically. However, according to the literature, liquid-liquid extraction (LLE) is the technique used most frequently for isolating morphinomimetic compounds (4)(5)(6)(7)(8), whereas only one method using SPE with RIA has been reported (9).
We performed comparative studies on the isolation of SF and alfentanil (AF; used as an internal standard) from plasma using LLE and SPE methods, and extracts were assessed using GC-MS. The method was applied to drug monitoring in intensive care patients.
SF citrate and AF hydrochloride were purchased from Janssen Pharmaceutica. Lyophilized plasma (Lyphochek®) was purchased from Bio-Rad Laboratories. Deionized water was prepared using a MilliQ® Water System (Millipore). Methanol (Uvasol® grade), dichloromethane (Uvasol grade), isoamyl alcohol (analytical grade), propan-2-ol (analytical grade), 13.2 mol/L ammonia, and sodium hydroxide (Normapur®) were purchased from Merck. n-Heptane (analytical grade), 18 mol/L sulfuric acid, and ammonium phosphate monobasic were purchased from Carlo Erba. Acetic acid and Sigmacote® were purchased from Sigma Aldrich. Oasis® MCX 60-mg SPE columns (Waters Corporation) were used, and the vacuum manifold for semiautomatic processing of the columns was purchased from Supelco. The Oasis MCX is a mixed-mode, polymeric sorbent (30-µm particle size) with strong cation-exchange sulfonic acid groups located on the surface of a poly(divinylbenzene-co-N-vinylpyrrolidone) copolymer.
Analyses were carried out using a Hewlett Packard (HP) 5890 Series II gas chromatograph (Agilent Technology) interfaced to a HP 5989A MS Engine quadrupole mass selective spectrometer (Agilent Technology). The gas chromatograph was equipped with a HP 6890 autosampler (Agilent Technology) and a split-splitless capillary inlet system. The injector, containing a quartz-deactivated liner, was operated in splitless mode. The chromatographic separations were achieved on a PTA-5 base-deactivated, fused-silica capillary column [30 m x 0.25 mm (i.d.); 0.50-µm film thickness; Sigma Aldrich]. Data acquisition was performed using a HP Chemstation 59940 (HP-UX series).
Injections of 3 µL were made in splitless mode with GC+-grade helium (Air Product) as the carrier gas. The column-head pressure was 16.3 psi at 200 °C (flow rate, 1 mL/min; linear velocity, 38.3 cm/s). The optimum inlet purge time was 2 min, and the inlet vent and septum purge were set at 50 and 1 mL/min, respectively. The oven temperature was held at 200 °C for 2.20 min, increased to 280 °C at 25 °C/min, maintained at this temperature for 17 min, and then increased to 300 °C at 25 °C/min. The optimum injector and transfer line temperatures for SF, based on maximum peak areas, were 280 and 260 °C, respectively.
The mass spectrometer was operated in the electron-impact mode (ionizing energy, 70 eV; trap current, 300 µA) with selected-ion monitoring. The ion source and quadrupole temperatures were set at 220 and 110 °C, respectively, and the source pressure was held at >6.6 x 10-4 Pa. The mass spectrometer detector was tuned to optimum sensitivity for m/z 264, 288, and 314 of the perfluorotributylamine certified standard and programmed to monitor the m/z 289 main fragments for SF and AF, as well as the m/z 140 and 268 characteristic fragments of SF and AF, respectively (dwell time, 100 ms). For quantification, the m/z 289 fragment was used for SF and AF, and the characteristic fragment of each compound (m/z 140 for SF; m/z 268 for AF) was used as the qualifier.
Stock solutions of SF (pKa 8.0) and AF (pKa 6.5) were prepared at a concentration of 100 mg/L in water. Weights were normalized to the free base, and the stock solutions were aliquoted into silanized tubes and kept protected from light at -25 °C until use. Five working solutions (25, 50, 250, 500, and 1000 µg/L) of SF containing the same amount of internal standard (500 µg/L) were freshly made from the frozen stock solutions by appropriate dilutions in acetonitrile. Calibration curves were constructed by supplementing plasma samples with SF (calibrator concentrations were 0.5, 1, 5, 10, and 20 µg/L).
Before all LLE or SPE extractions, a 1-mL plasma sample containing 100 µL of internal standard (AF) at a concentration of 100 µg/L was mixed for 2 min. LLE was performed according to the procedure reported by Woestenborghs and co-workers (5)(7). Plasma was alkalinized with 1 mL of 0.1 mol/L sodium hydroxide (pH 13) and extracted twice with 3.5 mL of heptane–isoamyl alcohol (95:5 by volume). The combined organic layers were back-extracted with 4 mL of 0.05 mol/L sulfuric acid and, after alkalinization of the latter phase with 0.15 mL of 13.2 mol/L ammonia, reextracted twice with 2.5-mL aliquots of the extraction solvent.
MCX 60-mg SPE columns (Oasis) were cleaned by rinsing twice with 1 mL of dichloromethane, conditioned by rinsing twice with 1 mL of methanol, and equilibrated by rinsing twice with 1 mL of water and then once with 1 mL of 100 mmol/L ammonium phosphate buffer. Plasma was acidified by the addition of 1 mL of 100 mmol/L ammonium phosphate buffer and then applied to the SPE unit. The columns were washed three times with 1 mL of water, twice with 1 mL of 1.0 mol/L acetic acid, twice with 1 mL of methanol, twice with 1 mL of propan-2-ol, twice with 1 mL of dichloromethane–propan-2-ol (80:20 by volume), and twice with 1 mL of H2O–CH3OH–NH4OH (58:40:2 by volume). The compounds of interest were eluted with four 0.5-mL volumes of CH2Cl2–propan-2-ol–NH4OH (78:20:2 by volume).
The eluates were collected in silanized conic tubes and evaporated to dryness under a gentle stream of purified nitrogen at <40 °C before analysis. The dry eluates were then redissolved in a total volume of 20 µL of ethyl acetate, and 3 µL was applied to the GC column. It should be noted that all glassware used for LLE and SPE were treated with a silanized reagent (Sigmacote) to avoid adsorption to the glass.
When an extraction technique is chosen for routine analyses, attention
must be paid to many elements to develop a technique that provides
reproducible results and clean extracts. SPE represents a rapid
and reliable sample treatment procedure for the determination of SF and
AF in plasma samples compared with conventional LLE, which involves
laborious and time-consuming extraction steps, including
back-extractions, which take 
8 h compared with 3 h
for SPE. Moreover, 36 samples, including calibrators and
quality-control and patient samples, can be analyzed in a
single SPE run, compared with only 18 samples with LLE. Some
difficulties also appear during the use of the LLE method, such as
emulsions breaking up incompletely and a risk of drug degradation
because of the need for a temperature of 60 °C for complete solvent
evaporation compared with 40 °C for SPE. Moreover, the extraction
efficiency of the SPE procedure was better than LLE for the analysis of
SF in plasma, with mean analytical recoveries of 95% ± 4% (n =
6) and 94% ± 3% (n = 6) for SPE at concentrations of 1 and 10
µg/L, respectively, compared with 86% ± 14% (n = 6) and 89%
± 13% (n = 6) for LLE. The SPE method also produced cleaner
chromatograms compared with LLE and prevented faster degradation of the
column and rapid contamination of the spectrometer ion source. The SPE
stationary phase presented several advantages: MCX columns, consisting
of polymeric sorbent, exhibited both polar-apolar and cationic
interactions and could be used from pH 0 to pH 14. With regard to the
economic aspects, the cost of extraction columns must be reconsidered
in view of the time savings and the efficiency of the SPE procedure,
such as the advantage of being easily automated. Although LLE
is widely used for morphinomimetics extraction, the steps involved in
LLE are inherently more complicated than those in SPE. The only
difficulty in the SPE procedure was optimization of the method,
particularly the choice of suitable eluents for the conditioning,
washing, and elution steps. However, when optimized, this method was
more convenient, rapid, sensitive, efficient, and reliable than
conventional LLE in obtaining clean plasma extracts with optimum
recoveries.
A chromatogram of a plasma supplemented with SF and AF at
concentrations of 1 and 10 µg/L, respectively, and extracted using
the SPE method is shown in Fig. 1A
. The separation of SF and its internal standard was achieved
in 20 min, and a total run of 42 min was necessary for complete
elution of interferents found in plasma. Concentrations were
calculated electronically using peak areas and internal
standardization. The calibration curves for 0.5–5 and 5–20 µg/L
(1.5–60 pg injected) of SF free base were linear (r =
0.999). The intra- and interday accuracy and precision of the method
were assessed by replicate analysis of plasma at three different
concentrations. The results are shown in Table 1
. The minimum detectable amount, defined as a signal-to-noise
ratio of 3, was 0.18 pg for SF. The quantification limit was 0.3 µg/L
with an relative standard deviation of 13.0% (relative error,
5.0%) for a 1-mL sample.
|
|
The method described was used for the determination of SF concentration
in plasma samples from intensive care patients. Chromatograms of a
blank plasma sample and a plasma sample from a patient who received a
SF infusion of 0.2
µg·kg-1·h-1 are
presented in Fig. 1, B and C
, respectively. The concentration of SF
recovered in the plasma sample was 0.80 µg/L.
In conclusion, SPE before GC-MS represents a convenient method for quickly obtaining clean extracts with optimum recoveries and could be an accurate analytical tool for the monitoring of SF concentrations in plasma samples during pharmacokinetic studies.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
