Alginate and Chitosan Foam Combined with Electromembrane

Sep 19, 2012 - the punched out dried blood spot and the foam was dissolved in. 300 μL of 1 mM HCl. With alginate foam as sampling medium, the analyte...
0 downloads 0 Views 282KB Size
Download PDF
Article pubs.acs.org/ac

Alginate and Chitosan Foam Combined with Electromembrane Extraction for Dried Blood Spot Analysis Lars Erik Eng Eibak,† Anne Bee Hegge,† Knut Einar Rasmussen,† Stig Pedersen-Bjergaard,†,‡ and Astrid Gjelstad*,† †

School of Pharmacy, University of Oslo, P.O. Box 1068 Blindern, 0316 Oslo, Norway Department of Pharmaceutics and Analytical Chemistry, Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen, Denmark

‡

ABSTRACT: Samples of 10 μL of whole blood containing citalopram, loperamide, methadone, and sertraline as model substances were spotted on alginate and chitosan foams as sampling media. After drying and storage at room temperature, the punched out dried blood spot and the foam was dissolved in 300 μL of 1 mM HCl. With alginate foam as sampling medium, the analytes dissolved completely after 3 min. Enrichment and cleanup was performed with electromembrane extraction for 10 min. The analytes were collected in 21 μL of 10 mM formic acid as the acceptor phase, and the extracts were analyzed by liquid chromatography−mass spectrometry (LC−MS). Sample preparation of blood spots on commercial cards was also performed (Whatman FTA DMPK and Agilent Bond Elut DMS) using elution procedures recommended by the manufacturers. The recoveries obtained with the commercial cards were lower for most of the model analytes compared to the recoveries obtained with alginate and chitosan foams as sampling media. The procedure used for Agilent Bond Elut DMS showed higher recoveries than the procedure used for Whatman FTA DMPK-A, but the time needed for sample preparation was significantly longer (nearly 2 h). The stability of the model substances on the alginate foam was acceptable within 50 days of storage. The limit of quantification (LOQ) defined as S/N = 10, was 1.2, 5.5, 2.0, and 5.3 ng/ mL for citalopram, loperamide, methadone, and sertraline, respectively. Linear calibration graphs were obtained in the range 17.5−560 ng/mL with r2 values 0.983−0.995, and the relative standard deviations were below 20%.

T

DBS. DBS cards are now available from several manufacturers (e.g Agilent, Whatman). The majority of the DBS cards are made of pure cellulose and can contain substances that lyse cells and denature proteins on contact.9 Recently noncellulosebased cards like Agilent Bond Elut DMS have become available as an alternative. Commercial cards have circles marked for blood spotting. Handling and analysis is simple and involves only a few steps. One method of sample collection is a finger or heel prick followed by blood flow into a glass capillary or microvessel. Typically 10−20 μL of whole blood is applied for each spot by the capillary or by a micropipette, and the blood spots are allowed to dry for a period of 2 h at room temperature.1 For analysis, a circular portion of each dried blood spot or the whole blood spot is punched out and transferred into a vial. The analyte is eluted by a few hundred microliters of an organic solvent typically methanol or methanol−water mixtures containing an internal standard. The sensitivity and selectivity offered by liquid chromatography−tandem mass spectrometry (LC−MS/MS) has con-

he analysis of small drug substances in whole blood samples collected and stored on filter paper, known as dried blood spots (DBS), has increased rapidly in the last year.1 The major interest in switching from traditional wet plasma to DBS is due to the ethical, economical, and practical advantages.1,2 Traditional methods are based on isolation of plasma from whole blood and centrifugation. Thereafter, plasma samples are prepared for analysis using protein precipitation, solid-phase extraction, or liquid−liquid extraction. This time-consuming process limits the number of samples that can be tested. The small blood volumes required for DBS (less than 100 μL) makes this approach particularly suitable for quantification of drug substances in drug metabolism (DM), pharmacokinetic (PK), and toxicokinetic (TK) studies. In addition, it offers the advantage of less invasive sampling such as finger or heel prick, instead of sampling with the conventional cannula, which is a great advantage in pediatric studies and in therapeutic drug monitoring (TDM).3,4 In addition, easy storage at room temperature and easy shipment to analytical laboratories offer further advantages.5 The stability of DBS under different storage conditions have been evaluated and considered as satisfactory.6−8 Patients can administer the finger prick at home without the need of trained phlebotomists. © 2012 American Chemical Society

Received: July 30, 2012 Accepted: September 19, 2012 Published: September 19, 2012 8783

dx.doi.org/10.1021/ac301996n | Anal. Chem. 2012, 84, 8783−8789

Analytical Chemistry

Article

spots. The results are compared with dried blood spot analysis on commercial cards using standard sample preparation procedures recommended by the manufacturers.

tributed to the success of the technique, and LC−MS/MS is the “gold standard” for DBS analysis.1 Electromembrane Extraction (EME). Despite the many advantages offered by DBS analysis on commercial cards, there are still some drawbacks. Elution of analytes is time-consuming and elution times of 1 h are common.1 Elution volumes of a few hundreds of microliters lead to a great reduction of assay sensitivity, and enrichment by evaporation and reconstitution of the sample in a smaller volume may be necessary to improve sensitivity. Some constituents may contribute to ion suppression in MS, and additional cleanup by an extraction method may be necessary. To address these disadvantages, this study focuses on new materials for DBS spotting combined with electromembrane extraction (EME) for enrichment and cleanup. Biopolymers such as alginate and chitosan can be manufactured as fibers, foams, and gels and are used commercially in areas such as wound management, tissue engineering, and controlled drug release. Alginate and chitosan have gel forming abilities, and alginate has previously been described as a main component in gelled alginate foams.10,11 In the present work, alginate and chitosan foams were used as new materials for blood spotting. These foams are porous, swollen hydrophilic matrices that can absorb water or blood. Chitosan foams are soluble in dilute acidic solutions below pH 6.0,12 where the amino groups are protonated and thus become positively charged. The alginate foam in the present study is formulated to disintegrate in water upon hydration; however, the foam structure is stable after blood sampling at physiological pH. When alginate and chitosan foams are used for dried blood spotting, a punched out dried blood spot can be rapidly dissolved in an acidic aqueous solution. These foams were tested as possible sampling medium for dried blood spots containing alkaline drugs, which are soluble in acidic solutions. As the solution contains the dissolved biopolymer and other foam components in addition to blood components and analytes, extraction of analytes or precipitation of the biopolymer is necessary prior to analysis. Since the introduction of EME as a new extraction technique,13 several papers have demonstrated that charged basic substances migrated electrokinetically through a supported liquid membrane (SLM) immobilized in a hollow fiber and toward the cathode located in the acceptor solution inside the hollow fiber. Further papers have reported selective extraction of basic or acidic substances from water, plasma, urine, whole blood, breast milk, and tap water.14−17 EME extractions from 500 to 1000 μL sample are completed within 10 min of extraction time at 10−300 V.18 In order to simplify the EME setup and reduce the required amounts of chemicals and sample, some recent papers reported the ability of replacing the adjustable power supply with a 9 V battery and successfully extracted analytes from 70 μL of untreated whole blood and human plasma. 19 Another substantial benefit is the enrichment factor obtained without any need for evaporation or reconstitution. The aqueous extracts provided with EME are directly compatible with LC− MS/MS. In a recent paper, exhaustive electromembrane extraction of basic drugs was demonstrated from undiluted human plasma followed by LC−MS.20 This approach is used in the present investigation to extract the basic model substances from the acidic aqueous solution of the dissolved blood spot and the biopolymer. The aim of this study was to examine the properties of biopolymers as sampling media for dried blood

■

MATERIAL AND METHODS Chemicals. Methadone hydrochloride and loperamide hydrochloride were from Sigma-Aldrich (St. Louis, MO). These substances were dissolved at 1 mg/mL to obtain appropriate standard solutions of methadone and loperamide. In order to obtain standard solutions at 1 mg/mL of sertraline and citalopram, a sertraline hydrochloride 50 mg tablet from Pfizer Italiana (Latina, Italy) was extracted with 50 mL of ethanol and a citalopram hydrobromide 20 mg tablet from H. Lundbeck (Copenhagen, Denmark) was extracted with 20 mL of ethanol. The tablet extracts were not standardized. The internal standard Lu 10-202, a citalopram analogue (fluorine is replaced by chlorine in the para position to the aromatic ring), was obtained from H. Lundbeck. 2-Nitrophenyl octylether was from Sigma-Aldrich (St. Louis, MO). The chemical structures of the model substances are presented in Figure 1. Water was

Figure 1. Chemical structure of the model substances.

obtained with a Milli-Q water purification system (Molsheim, France). Formic acid, methanol, and acetonitrile were all from Merck (Darmstadt, Germany). Drug-free human whole blood was obtained from a healthy volunteer at the School of Pharmacy (University of Oslo, Norway) and stored at −32 °C. Standard Solutions. Whole blood was evacuated and stored in BD Vacutainer sodium heparin/lithium heparin tubes (Franklin Lakes, NJ). A single stock solution of the model substances with a concentration of 1 mg/mL was prepared in ethanol. Calibration standards with the following concentrations, 17.5, 35, 70, 140, 280, and 560 ng/mL, were prepared by diluting the stock solution with whole blood. Subsequently, aliquots were spotted on the sampling medium of interest. Sampling Medium. Alginate foam was provided in house, and chitosan foam was prepared by FMC BioPolymer AS/ 8784

dx.doi.org/10.1021/ac301996n | Anal. Chem. 2012, 84, 8783−8789

Analytical Chemistry

Article

drying, the DBS was punched out (3 mm diameter) with a puncher and subsequently eluted in 300 μL of 0.1% formic acid in 80% methanol. The elution process was conducted in a 1500 μL vial with internal diameter of 10 mm and height of 32 mm. After 60 min of elution at 3000 rpm, the solution was centrifuged, evaporated to dryness, and reconstituted in 200 μL of mobile phase A, and 20 μL was injected into the LC−MS. Electromembrane Extraction (EME) Procedure. A 1500 μL vial with an internal diameter of 10 mm and height of 32 mm was used as a sample compartment for the dissolved blood spot as described in the section Polymeric Foams. The EME setup is illustrated in Figure 2. Three pieces of PP Q3/2

NovaMatrix (Sandvika, Norway). The alginate foam formulation was as follows: sodium alginate (PROTANAL LF 200S, lot number S20202) with a specified viscosity of 200−400 mPa s (1%, 20 °C) and a guluronic acid content of 65−75% was provided by FMC BioPolymer/NovaMatrix (Sandvika, Norway). The foaming agent, hydroxypropyl methylcellulose (HPMC, Pharmacoat 603, substitution type 2910, viscosity 3 mPa s), was kindly donated by Shin-Etsu Chemicals Co (Niigata, Japan). Calcium carbonate (CaCO3, Hubercal Elite 500) with a average particle size of 6 μm was a gift from J.M. Huber Corporation (Atlanta, GA). Delta-(+)-gluconic acid δlactone (GDL, purum; ≥99%) was delivered by Fluka Chemie Gmb, Sigma-Aldrich Chemie GmbH (Steinheim, Germany). Glycerol (85%, w/v) (Ph. Eur) and sorbitol (70%, w/v) (Ph. Eur.) were delivered by Norsk Medisinaldepot AS, Norway. The preparation of alginate foam was based on previously described methods.11 HPMC, Ca2+ (from CaCO3), GDL, sorbitol, and glycerol were selected as the foaming agent, gelling ion, pH modifier, and plasticizers, respectively. Alginate foams are formed through cross-linking between Ca2+-ions and guluronan blocks in the alginate and can be formulated to disintegrate or remain as an insoluble gel upon hydration, depending on the selected degree of cross-linking. A molar amount of Ca2+, equivalent to a selected low cross-linking degree of 30% of the guluronan units was used. Such foam will rapidly disintegrate in water. In the calculations, it was assumed that one Ca2+ ion interacts with two guluronan units and that sodium alginate contains 66% guluronan. The commercial cards examined were FTA DMPK-A cards produced by Whatman (Kent, United Kingdom) and Bond Elut DMS produced by Agilent (Santa Clara, CA). Blood Sampling. Polymeric Foams. A volume of 10 μL of whole blood spiked with model substances was applied to the sampling medium and dried at room temperature for 3 h. After drying, the entire DBS was punched out (8 mm diameter) with a puncher, and 2 μL of internal standard was added and dried for another 30 min. Subsequently, the dried blood drop and the biopolymer was dissolved in 300 μL of 1 mM hydrochloric acid in a 1500 μL vial (screw cap vial, Agilent Technologies, Waldbronn, Germany) with an internal diameter of 10 mm and height of 32 mm. The analytes were extracted and isolated by EME as described in the section Electromembrane Extraction (EME) Procedure. Precipitation of the biopolymer was investigated in an alternative procedure. A volume of 300 μL of ice-cold acetonitrile was added to the solution in order to precipitate the biopolymer. The solution was mixed (3000 rpm) for 5 min and centrifuged (10 000 rpm) for 5 min. The supernatant was evaporated to dryness, and the residue was dissolved in 100 μL of mobile phase A and thereafter 20 μL was injected into the LC−MS. Whatman FTA DMPK-A. A volume of 20 μL of whole blood spiked with model substances was spotted onto the sampling medium and dried under room temperature for 3 h. After drying, the DBS was punched out (3 mm diameter) with a puncher and subsequently eluted with 100 μL of 0.1% formic acid in 80% methanol for 10 min at 3000 rpm. The elution process was conducted in a 1500 μL vial with internal diameter of 10 mm and height of 32 mm. After elution the solution was centrifuged and 20 μL was injected into the LC−MS. Agilent Bond Elut DMS. A volume of 20 μL of whole blood spiked with model substances was applied to the sampling medium and dried under room temperature for 3 h. After

Figure 2. Illustration of electromembrane extraction setup.

polypropylene hollow fiber (Membrana, Wuppertal, Germany) with a pore size of 0.2 μm, a wall thickness of 200 μm, an inner diameter of 0.6 mm, and a length of 30 mm were closed in the lower end by mechanical pressure. The supported liquid membrane was made by impregnating the three porous hollow fibers with 2-nitrophenyl octyl ether (NPOE) for 5 s. The excess NPOE was gently removed with a medical wipe. A volume of 7 μL of 10 mM formic acid was filled with an airtight syringe into the lumen of each of the three porous hollow fibers. Platinum wires with a diameter of 0.2 mm (K.A Rasmussen, Hamar, Norway) were connected to the power supply and utilized as electrodes. The anode was placed in the sample compartment, and the three cathodes were placed in the lumens of the three porous hollow fibers as illustrated in Figure 2, one in each of the hollow fibers. A power supply ES 03000.45 from Delta Power Supplies (Delta Electronika, Zierikzee, The Netherlands) was operated at 100 V. The sample compartment was agitated at 3000 rpm during extraction for 10 min with an IKA MS 3 digital vortex mixer (Staufen, Germany). The extracts obtained with EME were diluted 1:1 with mobile phase A, and 20 μL was injected into the LC−MS. Liquid Chromatography−Mass Spectrometry (LC− MS). The chromatographic separation was conducted with a Biobasic-C8 reversed-phase column (50 mm × 1 mm) from Thermo Fisher Scientific (Waltham, MA) with an average pore size of 300 Å and particle diameter of 5 μm. The chromatographic system consisted of a Shimadzu SIL10ADvp auto injector, two Shimadzu LC-10ADvp gradient pumps, a Shimadzu DGU-14A degasser, a Shimadzu SCL10Avp system controller, and a Shimadzu LCMS-2010A singlequadrupole MS detector (all Shimadzu Scientific Instruments, 8785

dx.doi.org/10.1021/ac301996n | Anal. Chem. 2012, 84, 8783−8789

Analytical Chemistry

Article

Kyoto, Japan). Data acquisition and processing were carried out using Shimadzu LCMS Solution software, version 2.04-H3. The compositions of the mobile phases were as follows: mobile phase A, 20 mM formic acid and methanol (95:5, v/v); and mobile phase B, 20 mM formic acid and methanol (5:95, v/v). A linear gradient was run from 20% mobile phase B up to 100% mobile phase B after 15 min; subsequently, the mobile phase composition was kept constant for 3 min. Thereafter, the column was reconditioned with 20% mobile phase B for 6.1 min. The injection volume was set to 20 μL, and the mobile phase flow rate was 50 μL/min. The MS was operated with an electrospray ionization (ESI) source operated in the positive ionization mode and used to interface the HPLC and the MS. Analyses were performed with selected ion monitoring (SIM), with the following m/z values, 325, 341, 477, 310, and 306 for citalopram, Lu 10-202 (internal standard), loperamide, methadone, and sertraline, respectively. The MS operating conditions were as follows: flow rate of drying gas, 15 L/min; flow rate of nebulizing gas, 1.5 L/min; temperature of the curved desolvation line, 200 °C; block temperature, 200 °C; and probe voltage, +4.5 kV. Postcolumn Infusion. A standard solution of citalopram, Lu 10-202, loperamide, methadone, and sertraline with a concentration of 200 ng/mL in 10 mM formic acid was infused directly with a syringe pump (Gilson 402 Dilutor dispenser, Middleton, WI) into the ESI source with a T-piece. The Tpiece connected the mobile phase flow from the chromatographic separation with the standard solution delivered from the syringe pump. A volume of 20 μL of an EME extract from a blank blood spot sample was injected onto the chromatographic column in order to examine potential ion suppression.

■

RESULTS AND DISCUSSION Polymeric Foams As Sampling Media. The study set out to investigate the properties of alginate and chitosan foams as sampling media for DBS. The focus was on dissolution and extraction of analytes from the foams and the whole blood. Alginate, which is a main component in gelled alginate foams, is the salt of alginic acid. Alginic acid is widely distributed in the cell walls of brown algae and is a linear copolymer with homopolymeric blocks of (1−4)-linked β-D-mannuronate and the C-5 epimer α-L-guluronate. Gelled alginate foams have high water absorption capacities11 and can therefore easily absorb and contain blood spots. The selected alginate foam is insoluble in alcohol but soluble in aqueous solutions at certain pH-values. However, the solubility in water decreases below pH 3 due to protonation of the mannuronic and guluronic acid groups (pKa values are 3.38 and 3.65 for mannuronic- and guluronic acid, respectively).21 Chitosan is a cationic polysaccharide containing more than 5000 glucosamine units and is obtained commercially from shrimp and crab shell chitin. The pKa value of chitosan is 6.3, and the reported pH values for complete solubility are in the lower pH range.22,23 Sheets of alginate and chitosan foam were prepared for this study. The properties of these polymeric foams as spotting cards in terms of dissolution and extraction recoveries were compared with the properties of commercial cellulose spotting cards (Whatman FTA DMPK-A) and noncellulose spotting cards (Agilent Bond Elut DMS). In dried matrix spotting, typically a small portion of the spot is punched out for analysis. This procedure works well when the blood spot is consistent and when the dispersion of analyte in the spot is homogeneous. Figure 3a shows the distribution of 10 μL of whole blood spotted on

Figure 3. A volume of 10 μL of whole blood spotted on and dried at room temperature for 3 h on alginate foam (a) and solution obtained 3 min after dissolution of an alginate DBS in 300 μL of 1 mM HCl and 3000 rpm (b).

alginate foam. The dried blood spot occupied a diameter of 5 mm and left the foam intact. In the present investigation, the focus was on dissolution and extraction of analytes from the biopolymers and the whole blood spot was punched out for analysis using a homemade 8 mm puncher. Dissolution of Analytes. In a series of initial experiments, the time needed to dissolve a punched out DBS was investigated. A volume of 10 μL of whole blood samples containing 280 ng/mL of citalopram, methadone, loperamide, and sertraline as model substances was accurately spotted with a microsyringe. After drying for 2 h at room temperature, the whole blood spot was punched out and dissolved in various acidic solutions. A DBS spotted on alginate foam was completely dissolved after 3 min in 1 mM HCl. Figure 3b shows a photograph of a punched out DBS dissolved in 300 μL of 1 mM HCl. As shown in the photograph, a translucent reddish solution was obtained which indicated complete dissolution of the polymer, the foam components, and the DBS at pH 3 (1 mM HCl). Chitosan dissolved more slowly, and after 5 min a small portion of undissolved matter was still observed in the bottom of the tube; nevertheless, it did not seem to reduce recoveries. Prior to the experiment there was 8786

dx.doi.org/10.1021/ac301996n | Anal. Chem. 2012, 84, 8783−8789

Analytical Chemistry

Article

for dried blood spots can therefore offer a considerable saving of time. Figure 4a−d shows chromatograms of citalopram from

concern about possible ionic interactions between the basic analytes and negatively charged groups on alginate, which could lead to reduced recoveries. Because the recoveries from alginate and chitosan foams were similar and close to 100%, there were no indications of interactions at the conditions used. Extraction of Analytes. Several methods can be used to extract basic drugs from complex aqueous acidic sample solutions, such as solid-phase extraction or liquid−liquid extraction. A liquid-phase microextraction technique such as three phase hollow-fiber liquid-phase microextraction (HFLPME) has advantages as it allows enrichment and cleanup of basic analytes in an acidic acceptor solutions that can be injected directly into the LC−MS.24 One drawback of HFLPME is the long extraction times of 30−60 min needed to reach equilibrium. In this investigation, EME was selected because of the speed of extraction. In the three-fiber setup, extraction times were 10 min and the analytes were collected in 21 μL of 10 mM formic acid. The extract was diluted 1:1 with mobile phase A, and 20 μL was injected into the LC−MS. The recoveries obtained after extraction of the model analytes from solutions of DBS on alginate foam, chitosan foam, Bond Elut DMS, and FTA DMPK-A are shown in Table 1. The recoveries Table 1. Recoveries Obtained by Combining Alginate- and Chitosan Foam with EME, Agilent Bond Elut DMS, and Whatman FTA DMPK-A, Respectively extraction recovery, (RSD, %)a citalopram methadone loperamide sertraline a

alginate

chitosan

100% (4) 105% (6) 90% (6) 44% (4)

103% (2) 115% (4) 91% (2) 57% (2)

Bond Elut DMS 69% 59% 59% 66%

(6) (7) (4) (5)

FTA DMPK-A 17% 33% 24% 11%

(17) (12) (8) (3)

Figure 4. Chromatograms obtained for citalopram (m/z 325) with procedures outlined in the section Blood Sampling for FTA DMPK-A (a), Bond Elut DMS (b), alginate-EME (c), and alginate-precipitation (d), respectively.

n = 3.

were in the range of 44−115% by combining storage on alginate- or chitosan foam with EME and were similar to the ones reported for extraction of the same model analytes from wet plasma.20 EME of sertraline has been challenging in terms of extraction recovery, probably due to physical-chemical properties and compatibilty with the proposed EME-setup. In Table 1, some of the experiments provided recoveries above 100%; however, this is probably ascribed to operational errors and not to interferences from the sample solution. Elution of Analytes from DBS on Commercial Cards. Elution of analytes from DBS on commercial cards was performed with the standard procedures recommended by Agilent and Whatman as stated in the section Blood Sampling. The recoveries obtained by using the standard procedures are shown in Table 1. Elution of analytes from Agilent Bond Elut DMS with the procedure recommended by Agilent reported recoveries in the range 59−69%. This procedure involved elution for 1 h, centrifugation, evaporation to dryness, and reconstitution in the mobile phase. The recoveries from Whatman FTA DMPK-A cards using elution for 10 min were considerable lower (11−33%) most probably because the recommended elution time of 10 min is far too short for elution of the model substances studied in this investigation. Comparison of Procedures for Commercial Cards and Polymeric Foams. High recoveries are obtained with the procedure used for Bond Elut DMS, but the time needed is significantly higher compared to the procedure used for biopolymers. The use of polymeric foams as sampling medium

DBS on FTA DMPK-A (a), Bond Elut DMS (b), alginate-EME (c), and alginate followed by precipitation with ice-cold acetonitrile (d). Small additional peaks are in the chromatograms in parts a, b, and d. In the chromatogram in part c from alginate-EME, the baseline noise is almost eliminated by the extraction. The supported liquid membrane utilized in the case of alginate-EME acts as a selective cleanup barrier and excludes proteins and reduces the coextraction of matrix components. This experiment emphasized that EME provides clean extracts from complex biological matrices that are directly compatible with LC−MS. The time needed for sample preparation is 15 min including dissolution and extraction. Validation. The proposed method has been evaluated with respect to FDA’s guidance for industry concerning bioanalytical method validation, and the results are summarized in Table 2. Limit of Detection (LOD) and Limit of Quantification (LOQ). Both the limit of detection (LOD) defined as a signalto-noise ratio of 3 and limit of quantification (LOQ) defined as a signal-to-noise ratio of 10 were examined with respect to citalopram, loperamide, methadone, and sertraline. The LOQ was below the lower therapeutic level for all model substances.25 The combination of sensitive LC−MS analysis with the high recoveries obtained with EME contributed to LOQ below the 8787

dx.doi.org/10.1021/ac301996n | Anal. Chem. 2012, 84, 8783−8789

Analytical Chemistry

Article

Table 2. Validation Results with Alginate Combined with EME-LC−MS from Dried Blood Spot Spiked with Model Substances, FDA Requirements in Parenthesesa citalopram

loperamide

LOD (ng/mL) LOQ (ng/mL)

0.4 1.2

1.6 5.5

range (ng/mL) R2

17.5−560 0.983

1120 ng/mL (n = 5) 140 ng/mL (n = 5) 17.5 ng/mL (n = 5)

4 17 5

methadone

0.6 2.0 Linearity (>0.990) 17.5−560 17.5−560 0.986 0.992 Repeatability (%)b (