Journal of Chromatography A, 1345 (2014) 207–211

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Large volume sample stacking for rapid and sensitive determination of antidiabetic drug metformin in human urine and serum by capillary electrophoresis with contactless conductivity detection ∗ ˚ Petr Tuma Institute of Biochemistry, Cell and Molecular Biology, Third Faculty of Medicine, Charles University, Ruská 87, 100 00 Prague 10, Czech Republic

a r t i c l e

i n f o

Article history: Received 19 February 2014 Received in revised form 3 April 2014 Accepted 7 April 2014 Available online 13 April 2014 I would like dedicate this paper to my teacher Prof. Frantiˇsek Opekar. Keywords: Antidiabetic drug Capillary electrophoresis Contactless conductivity detection Large volume sample stacking Rapid separation

a b s t r a c t Two CE methods with contactless conductivity detection have been developed for determining the oral antidiabetic drug metformin in human urine and blood. The determination of metformin is performed on a separation capillary with an effective length of 14 cm, using a maximum voltage of 30 kV and with a small injection of 50-fold diluted urine into the capillary. Under these conditions, the migration time of metformin is 35 s and the LOD is 0.3 ␮M. Large-volume sample stacking was used to determine low metformin levels in serum. The injection of a sample of serum deproteinized with acetonitrile was 10 times greater compared to the injected amount of urine. This enabled reduction of the LOD to 0.03 ␮M and the metformin migration time equalled 86 s. The undesirable solvent from sample zone was forced out of the capillary to ensure rapidity and good repeatability of the determination. The RSD values for the migration time are 0.1% for urine and 0.7% for serum; RSD for the peak areas equalled 1.4% for urine and 2.6% for serum. The developed CE technique was tested on performance of routine analyses of metformin in the urine and serum of patients suffering from type II diabetes mellitus. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Metformin (Fig. 1) is an oral antidiabetic drug of the biguanide group. At the present time, metformin is the drug of choice for treating type II diabetes and is prescribed primarily for overweight people (www.idf.org). Clinical studies have demonstrated that ten-year administration of metformin to obese diabetics reduces morbidity and mortality by 30% compared to treatment with insulin and sulphonyl urea [1]. In addition, administration of metformin to obese and overweight diabetics does not contribute to a further increase in their weight [2]. Metformin is further characterized by a low risk of lactic acidosis compared to other biguanides [3]. At the present time, metformin is the most extensively used antidiabetic drug and 48 million packages were prescribed in 2010 in the U.S.A. alone [4]. In addition to treatment of advanced type II diabetes, metformin is also used to treat gestational diabetes [5] and prediabetes

Abbreviations: ACN, acetonitrile; BGE, background electrolyte; C4 D, capacitively coupled contactless conductivity detection; ECE, electrochemiluminescence; EOF, electroosmotic flow; FASS, field amplified sample stacking; HAc, acetic acid; INST, coating solution for fused silica capillary; LOD, limit of detection; LOQ, limit of quantification. ∗ Tel.: +420 267 102 585; fax: +420 267 102 460. E-mail address: [email protected] http://dx.doi.org/10.1016/j.chroma.2014.04.016 0021-9673/© 2014 Elsevier B.V. All rights reserved.

in high-risk groups. Metformin is also used to treat polycystic ovary syndrome [6] and successful use in cancer prevention has also been described [7]. The plant Galega officinalis [8] is a natural source of drugs in the biguanide group. However, metformin is produced synthetically for use in contemporary medicine [9]. A large number of analytical methods have been developed for controlling the purity of drugs, determining metformin in plant products and in clinical samples: HPLC with MS [10–13] or UV detection [14–17]; hollow fibre extraction with consequent HPLC determination [18,19]; high-performance thin-layer chromatography [20]; H–NMR in combination with HPLC–MS [21]; GC–MS [22] and direct chemiluminescence determination [23]. The capillary electrophoresis method can be successfully used to determine ionogenic metformin, which is also highly soluble in water. In the analysis of biological and clinical samples, CE is characterized by low requirements on pretreatment of a complicated sample matrix compared to HPLC and GC. Only a few references in the literature describe the CE determination of metformin in pharmaceutical products [24–28], blood plasma [29–32] and urine [33]. This work is concerned with the development of a very rapid CE method for determination of metformin in human urine and blood [34]. Treatment of urine and serum for CE analysis is very simple and consists only in diluting the body fluid and, as appropriate,

208

P. T˚ uma / J. Chromatogr. A 1345 (2014) 207–211

Fig. 1. Structure of metformin (N,N-dimethylimidodicarbonimidic diamide).

deproteinization of the sample. The high sensitivity of the determination when using universal capacitively coupled contactless detection (C4 D) [35–39] is based on separation in a short capillary, where excessive broadening of the sample is eliminated, and also on on-line preconcentration of the sample using large-volume sample stacking [40–44]. 2. Experimental 2.1. CE experiments and capillary arrangement Electrophoretic measurements were carried out using the HP3D CE system (Agilent Technologies, Waldbronn Germany) equipped with a C4 D detector [45], which is placed in the electrophoretic cassette thermostated at constant temperature of 25 ◦ C. All CE separations of urine and serum samples were performed in a fused-silica capillary (Composite Metal Services, UK), 50 ␮m I.D., 363 ␮m O.D., 31.7 cm total length. This is the minimum capillary length for HP3D CE systems. The inner surface of the capillary was covered using INST coating solution (Biotaq, U.S.A.) to prevent electro-osmotic flow (EOF) [46]. A new capillary was washed stepwise with 0.1 M NaOH (5 min), water (5 min), INST coating solution (2 min), water (5 min) and BGE (5 min). 0.5 min washing with the BGE was used between individual analyses. All the separations were carried out in the optimized background electrolyte (BGE) of 2.0 M acetic acid (pH 2.15). The C4 D detector is placed in cassette asymmetrically to the electrophoretic capillary, 17.0 cm from the one end (called the long end) and 14.7 cm from the other end (called the short end). Two different sets of experimental conditions were used for determination of metformin in urine (i) and serum samples (ii): i) Urine samples, treated by dilution with water, were injected into the short end of capillary in a small amount (pressure 50 mbar for 2 s, corresponding to a sample zone length of 2.3 mm and a sample zone volume of 4.5 nL). Then the maximal voltage of +30 kV (in relation to the short end of capillary, current 36.8 ␮A) was switched on and simultaneously the sample zone was forced out of the capillary by application of negative pressure. The hydrodynamic impulse used to force the sample out of the capillary was the same as the injection impulse, 100 mbar s. After turning on the separation voltage, the metformin cations and other ions rapidly leave the sample zone and subsequently the remaining solvent in the sample zone is forced out of the capillary. When the sample zone was not forced out of the capillary, the electric current was interrupted during the separation. The explanation is, that after the ions have left, the sample zone presents a large resistance for passage of the electric current during the separation. ii) Serum samples, deproteinized by addition of acetonitrile, were injected into the long end of the capillary in a large volume (pressure 50 mbar for 20 s, corresponding to a sample zone with a length of 22.9 mm and a volume of 45 nL). Then a separation voltage of +20 kV was turned on (current 21.5 ␮A) and simultaneously the acetonitrile zone was forced out of the capillary by application of a negative pressure of −50 mbar for 20 s. Metformin again leaves the acetonitrile zone and enters the surrounding BGE which, if the acetonitrile zone were not removed, would constitute an obstacle for passage of the electric current.

Fig. 2. Electropherogram of diabetic urine diluted 50-fold with BGE (A) and diluted 50-fold with water (B). Peak identification: inorganic cations (1), metformin (2), creatinine (3).

2.2. Chemicals All the chemicals used were of p.a. purity: acetic acid (HAc, Sigma), creatinine (Fluka), metformin hydrochloride (Fluka), NaOH (Fluka), HCl (Sigma), acetonitrile (ACN, Fluka). The stock solutions of metformin and creatinine were prepared at concentrations of 10 mM. Deionized Milli-Q water (18.2 M cm, Millipore, Molsheim, France) was used for preparation of the BGE and the stock solutions of the standards, which were stored in a refrigerator at 4 ◦ C. 2.3. Pretreatment of urine and serum samples Samples of morning urine and venous blood were collected from patients suffering from type II diabetes mellitus, undergoing treatment at the 2nd Department of Internal Medicine, Královské Vinohrady Faculty Hospital in Prague. Metformini hydrochloridum was administered to them in an amount of 1000 mg per day. The control urine and blood samples were obtained from healthy adult volunteers. The collected urine samples were stored in a freezer at −20 ◦ C and maintained at this temperature until the analysis. Prior to the CE measurement, unfrozen urine samples were only diluted 50-fold with 0.01 M HCl acid (20 ␮L of urine were added to 980 ␮L of 0.01 M HCl). Dilution of the urine with acidified water is very important for sharpening the sample zone; water suppresses the conductivity of the sample and causes it to become shorter. Acidification of the sample with HCl maintains the metformin in the protonated form, which rapidly leaves the sample zone. The importance of water for sharpening the sample zone is clearly demonstrated in Fig. 2. Simultaneously, Fig. 2 depicts a recording of 50-fold dilution of the urine with the BGE, which does not lead to sample sharpening and the peaks of the analyte are lost in the detector noise. Serum samples obtained from coagulated venous blood were stored in a freezer at −20 ◦ C until the analysis. Before analysis, the unfrozen serum samples were deproteinized by mixing 100 ␮L of serum with 300 ␮L of acetonitrile acidified with HCl at a concentration of 0.01 M. Deproteinization was performed in an Eppendorf tube after 30 s shaking. Then the serum samples were filtered through a Durapore polyvinyl difluoride membrane (pore size 0.45 ␮m, centrifugal filter devices, Millipore, Bedford, USA) and

P. T˚ uma / J. Chromatogr. A 1345 (2014) 207–211

Fig. 3. Enlarged electropherogram of diabetic urine (A) and diabetic serum (B). Peak identification: inorganic cations (1), metformin (2), creatinine (3) and basic amino acids (4).

100 ␮L of the obtained solution were taken for CE analysis. Acidification of the sample with HCl is important for complete focusing of metformin from the long sample zone. In the opposite case, metformin is lost during forcing the sample out of the capillary, causing bending of the calibration dependence towards the x-axis for higher metformin concentrations in the sample. 2.4. Treatment and evaluation of the results All the CE analyses of the model samples were carried out in five consecutive runs and the plots represent the average values ± the standard deviations. The Origin 8.0 programme (OriginLab Corporation, Northampton, MA, U.S.A.) was used to evaluate and statistically treat the experimental data. The number of theoretical plates was calculated from the formula, N = 5.54 (tM /w1/2 )2 , where tM is the migration time and w1/2 is the peak width at half-height. 3. Results and discussion 3.1. CE separation of metformin in human urine Metformin is a divalent base with acid dissociation constants (pKa ) 2.8 and 11.5 and migrates as a cation in acidic BGE [47]. Consequently, solutions of acetic acid (HAc) were tested for the CE separation of metformin. HAc solutions were formerly used successfully for the separation of amino acids [48,49], peptides [50,51], neurotransmitters [52] and biogenic amines [53], which have basic natures similar to that of metformin. In addition, C4 D detectors exhibit low noise levels and high baseline stabilities in HAc-based BGEs. The BGE composition was optimized directly in separation of diabetic urine samples. Of the tested concentration range, 0.1–4.0 M HAc, the best results were obtained in 2.0 M HAc. In this BGE, metformin is completely separated from the inorganic cations present in urine (Fig. 3). The metformin peak does not overlap with that of any micro-component present in the urine, which was verified in separation of various samples of physiological urine, where no other peak was found in the same position as metformin (Fig. 4). The peaks of the inorganic cations, metformin and creatinine (waste product of skeletal muscle metabolism) can be seen in the complex electropherogram of urine. In the doses that are used to treat diabetic patients, metformin represents the dominant component of the urine. Consequently, the development of the CE determination of metformin in urine

209

Fig. 4. Detailed electropherogram of physiological urine (A) and urine of a diabetic patient with determined metformin concentration of 3.2 mM (B). Inset (C): blank physiological urine (C1) and physiological urine spiked with 10 ␮M metformin (C2). Peak identification: metformin (1) and creatinine (2).

concentrated on rapid routine analyses. The separation is performed on a short effective separation path of 14 cm, at a high separation voltage value of 30 kV and low sample injection (100 mbar s). Under these conditions, the migration time of metformin is 35 s and the separation efficiency value equals 116,000 theoretical plates m−1 for a diabetic urine sample containing 3.2 mM metformin. This method has the further advantage that creatinine can be determined simultaneously, with a migration time of 45 s; the concentrations of the analytes in the urine are recalculated to this value to correct for different diuresis. 3.2. CE separation of metformin in human serum The determination of metformin in serum is a complicated task, because the concentration of metformin in the serum of diabetics is approx. 1000 times lower than in the urine. Consequently, development of the method was directed towards achieving low LOD values using large-volume sample stacking. 10-fold higher serum sample injection compared to the urine sample leads to a value of 0.03 ␮M. Under these injection conditions, the metformin peak is separated well from the other components of the serum, especially inorganic ions and basic amino acids (Fig. 3). It is apparent from the electropherogram of a physiological serum sample spiked with 1.0 ␮M of metformin (Fig. 5) that metformin once again does not overlap with any other peak of a substance present in the serum. The use of large-volume sample stacking is associated with suppression of the serum conductivity by the addition of acetonitrile, the metformin zone rapidly migrates in this environment and becomes sharper as it enters the surrounding BGE with high conductivity [54]. The acetonitrile zone must then be removed from the capillary to prevent interruption of the separation or undesirable prolonging of the migration time. Under these conditions, metformin was successfully separated from the other serum components over a short effective separation pathway of 17 cm at a voltage of 20 kV, with a separation time of 86 s. The separation efficiency of metformin analysis in serum is very high, 490,000 theoretical plates m−1 . This is connected with good sharpness of the large injected volume of sample into a narrow zone and also with the low metformin concentration in the serum (approx. 4.0 ␮M). The determination is also highly sensitive, as is reflected by the addition of 1.0 ␮M metformin to the serum, which can be clearly distinguished from the noise level C4 D.

210

P. T˚ uma / J. Chromatogr. A 1345 (2014) 207–211 Table 2 Comparison of the CE method used for determination of metformin in clinical and pharmaceutical samples. Sample

Detection

LOD (␮g/L)

Plasma Serum Plasma (spiked) Plasma Urine Tablets Plasma Tablets Tablets Tablets Urine Herbal formulations

MS C4 D + LVSS UV + SPE UV C4 D UV UV + FASS UV UV UV ECEa C4 D

2.1 4 12 30 40 60 250 800 1000 2000 2300 3200

a

Fig. 5. Detailed electropherogram of a physiological serum (A) and the serum of a diabetic patient with determined metformin concentration of 4.4 ␮M (B). (A1): physiological serum spiked with 1.0 ␮M metformin (1). Table 1 Migration time, separation efficiency, calibration parameters, LOD, LOQ, RSD for CE determination of metformin in the urine and serum of a diabetic patient. Urine

Serum

Migration time (s) Number of theoretical plates (m−1 ) Spiked concentration range (␮M) Slope – peak area (␮V s ␮M−1 ) R – peak area Slope – peak height (␮V.␮M−1 ) R – peak height LOD (␮M/␮g/L) LOQ (␮M/␮g/L)

35.4 116,000 (4000) 5–100 14.9 (0.1)a 0.9996 24.3 (0.3) 0.9996 0.3/40 1.0/130

86.6 490,000 (20,000) 0.2–10 198.1 (1.6) 0.9999 248.5 (0.7) 0.9985 0.03/3.9 0.1/13.0

Intra-day repeatability RSD – migration time (%) RSD – peak area (%)

0.1 1.4

0.7 2.6

Inter-day repeatability RSD – migration time (%) RSD – peak area (%)

0.4 2.9

1.0 4.1

a

Standard deviations are in parenthesis.

3.3. Calibration dependence and repeatability of the determination The calibration dependences calculated from the area and height of the peak for urine and serum in the tested concentration intervals are linear with correlation coefficient (R) values close to 1.000 (Table 1). Acidification of the sample by addition of HCl is important for linearity of the calibration dependence; metformin rapidly leaves the acidified zone to enter the BGE and it is not lost during forcing of the unwanted solvent out of the capillary. Metformin can be quantified using the peak height and area, yielding equivalent values: urine – 3.19 mM from the area and 3.26 mM from the height; serum – 4.4 ␮M from both the area and the height. The value of the slope of the calibration curve is identical for various samples of physiological and diabetic urine; this is also true for the serum samples. The obtained LOD values are adequate for performance of routine analyses of urine and serum: 0.3 ␮M for urine and 0.03 ␮M for serum. The 10-fold reduction of LOD in the serum compared with urine is directly connected with the 10-fold greater injection volume of the serum sample into the capillary. Creatinine was also determined in urine; the slope of the calibration curve is 14.8 ␮V s ␮M−1 (peak area, tested interval 0–100 ␮M) and LOD is 0.3 ␮M.

tM (min) 8 1.4 3 6 0.6 8 4 10 9 4 3 4

Ref. [29] Present [31] [30] Present [27] [32] [24] [28] [26] [33] [25]

Electrochemiluminescence (ECE).

The intra-day and inter-day repeatability of the determination were tested for samples of diabetic urine and serum (Table 1). The intra-day repeatability was determined from 20 consecutive determinations of one of the treated urine and serum samples. The inter-day repeatability was determined for the determination of one sample of diabetic urine and serum, which was determined on three consecutive days; each day the sample was newly treated and determined 20 times consecutively. The RSD values are fully comparable with the normal electrophoretic determinations. The introduction of a step forcing excess solvent from the capillary is an additional factor that could make a negative contribution in reducing the repeatability of the CE determination. It follows from the obtained results that the double hydrodynamic impulse (injection and forcing out the sample) does not contribute to reducing the repeatability. 3.4. Evaluation of the results Table 2 summarizes the electrophoretic determinations of metformin in clinical and pharmaceutical samples. The methods are listed in order of worsening LOD and the table also contains the migration times. The first half of the table lists methods with low LOD values, which can be employed for determining metformin on blood plasma. High sensitivity is attained in these determinations by using MS detection, sample preconcentration using SPE or field amplified sample stacking (FASS). On the other hand, methods with high LOD, usually combined with UV detection (2nd half of the table), are sufficient for checking the purity of pharmaceutical products. Both newly developed CE-C4 D methods have the advantage of faster separation than that achieved to date with other methods. It is also shown that, even the use of a universal detection technique like C4 D enables achieving LOD values for metformin at the level of 10−8 M (Table 1), which was formerly attained only when the separation was combined with MS detection [29]. High sensitivity is achieved in this work: (a) by performing rapid separation over a short pathway, which reduces broadening of the sample through diffusion and adsorption; (b) introduction of effective large-volume sample stacking, as the sensitivity of the determination increases with increasing amount of sample injected into the capillary; (c) forcing undesirable solvent out of the capillary, accelerating the separation (elimination of low-conductivity zones) and thus contributing to improved repeatability of the determination [54]. The determined concentrations of metformin in urine and serum correspond to pharmacokinetics and are in accordance with the published results. It follows from the determined metformin concentrations in urine (3.2 mM, 413 mg/L) and the measured 24h urine production (1.8 L) that the amount of metformin excreted by the kidneys is approx. 745 mg (the patient is administered one

P. T˚ uma / J. Chromatogr. A 1345 (2014) 207–211

coated tablet daily, containing 1000 mg of metformini hydrochloridum, corresponding to 780 mg of pure metformin). The entire dose of metformin is excreted from the body by the kidneys in unaltered form; metformin is not biotransformed and is not accumulated in the fat tissue [55]. The measured metformin concentration in the serum of 4.4 ␮M (568 ␮g/L) is several times less than in the urine (approx. 730 times). The metformin concentration in the blood increases rapidly from zero to a maximum during 3–5 h (following administration of 1000 mg metformin hydrochloride, the maximum concentration equalled approx. 2000 ␮g/L) and then this value decreased to zero after approx. 24 h. Blood was taken for this study 10 h after administration of a 1000 mg tablet and the determined concentration of 568 ␮g/L corresponds to the determined pharmacokinetics published in previous studies [10,32]. 4. Conclusions Electrophoretic separation performed over a short separation pathway permits migration times of under 1 min to be attained even when unspecialized commercial CE instruments are used. These rapid separations are useful for routine determinations of pharmaceuticals in a wide range of clinical samples. Universal contactless conductivity detection can be employed for a wide range of biogenic analytes instead of using more expensive MS. The sensitivity of CE-C4 D determination can be increased several fold by preconcentration of the sample directly in the capillary. The introduction of subsequent forcing of undesirable solvent out of the sample zone contributes to achieving very rapid separation and simultaneously improves the repeatability of the determination. CE-C4 D is a very useful alternative to widely employed HPLC–MS and offers a number of advantages for biomedical applications. Acknowledgements Financial support from the Charles University in Prague, the Projects PRVOUK P31 and UNCE 204015/2012, is gratefully acknowledged. References [1] UK Prospective Diabetes Study Group, Lancet 352 (1998) 854. [2] E. Selvin, S. Bolen, H. Yeh, et al., Arch. Intern. Med 168 (2008) 2070. [3] S.R. Salpeter, E. Greyber, G.A. Pasternak, E.E. Salpeter, Arch. Intern. Med. 163 (2003) 2594. [4] The Use of Medicines in the United States: Review of 2010, IMS Institute for Healthcare Informatics, New York, 2011. [5] J.A. Rowan, W.M. Hague, W. Gao, M.R. Battin, M.P. Moore, N. Engl. J. Med. 358 (2008) 2003. [6] J.M. Lord, I.H.K. Flight, R.J. Norman, BMJ 327 (2003) 951. [7] D. Li, S.C.J. Yeung, M.M. Hassan, M. Konopleva, J.L. Abbruzzese, Gastroenterology 137 (2009) 482. [8] C.J. Bailey, C. Day, Pract. Diab. Int. 21 (2004) 115. [9] E.A. Werner, J. Bell, J. Chem. Soc. Trans. 121 (1922) 1790. [10] M.A.S. Marques, A.D. Soares, O.W. Pinto, P.T.W. Barroso, D.P. Pinto, M. Ferreira, E. Werneck-Barroso, J. Chromatogr. B 852 (2007) 308.

211

[11] A. Liu, S.P. Coleman, J. Chromatogr. B 877 (2009) 3695. [12] H.N. Mistri, A.G. Jangid, P.S. Shrivastav, J. Pharm. Biomed. Anal. 45 (2007) 97. [13] N. Koseki, H. Kawashita, M. Niina, Y. Nagae, N. Masuda, J. Pharm. Biomed. Anal. 36 (2005) 1063. [14] F.A. Siddiqui, N. Sher, A. Zubair, H. Shamshad, N. Shafi, A.Z. Mirza, Anal. Methods 5 (2013) 5096. [15] M.S. Arayne, N. Sultana, M.H. Zuberi, F.A. Siddiqui, J. Chromatogr. Sci. 48 (2010) 721. [16] K.M. Huttunen, J. Rautio, J. Leppanen, J. Vepsalainen, P. Keski-Rahkonen, J. Pharm. Biomed. Anal. 50 (2009) 469. [17] C. Yardimci, N. Ozaltin, A. Gurlek, Talanta 72 (2007) 1416. [18] L. Zhang, Z.Q. Zhang, W.C. Dong, S.J. Jing, J.F. Zhang, Y. Jiang, J. Chromatogr. A 1318 (2013) 265. [19] G.M. Ben-Hander, A. Makahleh, B. Saad, M.I. Saleh, J. Chromatogr. B 941 (2013) 123. [20] D. Kale, R. Kakde, JPC-J. Planar Chromatogr. Mod. TLC 24 (2011) 331. [21] T.G. Huo, S. Cai, X.M. Lu, Y. Sha, M.Y. Yu, F.M. Li, J. Pharm. Biomed. Anal. 49 (2009) 976. [22] E. Ucakturk, Anal. Methods 5 (2013) 4723. [23] C. He, Z.J. Zhang, D.Y. He, Y. Xiong, Anal. Bioanal. Chem. 385 (2006) 128. [24] A.T. Barden, B.L. Piccoli, N.M. Volpato, M. Steppe, Anal. Methods 5 (2013) 5701. [25] C. Viana, M. Ferreira, C.S. Romero, M.R. Bortoluzzi, F.O. Lima, C.M.B. Rolim, L.M. de Carvalho, Anal. Methods 5 (2013) 2126. [26] I.I. Hamdan, A.K.B. Jaber, A.M. Abushoffa, J. Pharm. Biomed. Anal. 53 (2010) 1254. [27] I. Ali, H.Y. Aboul-Enein, V.K. Gupta, Comb. Chem. High Throughput Screen 10 (2007) 611. [28] C. Yardimci, N. Ozaltin, Anal. Chim. Acta 549 (2005) 88. [29] J. Znaleziona, V. Maier, V. Ranc, J. Sevcik, J. Sep. Sci. 34 (2011) 1167. [30] S.Y. Wel, H.H. Yeh, F.F. Liao, S.H. Chen, J. Sep. Sci. 32 (2009) 413. [31] E.P.C. Lai, S.Y. Feng, J. Chromatogr. B 843 (2006) 94. [32] J.Z. Song, H.F. Chen, S.J. Tian, Z.P. Sun, J. Chromatogr. B 708 (1998) 277. [33] B.Y. Deng, A.H. Shi, Y.H. Kang, L.Q. Li, Luminescence 26 (2011) 592. [34] F. Opekar, P. Coufal, K. Stulik, Chem. Rev. 109 (2009) 4487. [35] P. Kuban, P.C. Hauser, Electrophoresis 34 (2013) 55. [36] P. Kuban, P.C. Hauser, Electrophoresis 32 (2011) 30. [37] P. Kuban, P.C. Hauser, Electrophoresis 30 (2009) 176. [38] M. Pumera, I. Jelinek, J. Jindrich, Fresenius J. Anal. Chem. 369 (2001) 666. [39] J. Wang, M. Pumera, Anal. Chem. 75 (2003) 341. [40] T. Kawai, M. Ueda, Y. Fukushima, K. Sueyoshi, F. Kitagawa, K. Otsuka, Electrophoresis 34 (2013) 2303. [41] T. Kawai, J. Ito, K. Sueyoshi, F. Kitagawa, K. Otsuka, J. Chromatogr. A 1267 (2012) 65. [42] M.C. Breadmore, A.I. Shallan, H.R. Rabanes, D. Gstoettenmayr, A.S.A. Keyon, A. Gaspar, M. Dawod, J.P. Quirino, Electrophoresis 34 (2013) 29. [43] A. Slampova, Z. Mala, P. Pantuckova, P. Gebauer, P. Bocek, Electrophoresis 34 (2013) 3. [44] Z. Mala, P. Gebauer, P. Bocek, Electrophoresis 32 (2011) 116. [45] B. Gas, J. Zuska, P. Coufal, T. van de Goor, Electrophoresis 23 (2002) 3520. [46] P. Tuma, J. Sep. Sci. 37 (2014) 1026. [47] G.G. Graham, J. Punt, M. Arora, R.O. Day, M.P. Doogue, J.K. Duong, T.J. Furlong, J.R. Greenfield, L.C. Greenup, C.M. Kirkpatrick, J.E. Ray, P. Timmins, K.M. Williams, Clin. Pharmacokinet. 50 (2011) 81. [48] P. Coufal, J. Zuska, T. van de Goor, V. Smith, B. Gas, Electrophoresis 24 (2003) 671. [49] P. Tuma, K. Malkova, E. Samcova, K. Stulik, J. Sep. Sci. 33 (2010) 2394. [50] V. Solinova, V. Kasicka, P. Sazelova, T. Barth, I. Miksik, J. Chromatogr. A 1155 (2007) 146. [51] V. Solinova, V. Kasicka, D. Koval, T. Barth, A. Ciencialova, L. Zakova, J. Chromatogr. B 808 (2004) 75. [52] P. Tuma, M. Soukupova, E. Samcova, K. Stulik, Electrophoresis 30 (2009) 3436. [53] X.Y. Gong, P.C. Hauser, Electrophoresis 27 (2006) 4375. [54] P. Tuma, M. Sustkova-Fiserova, F. Opekar, V. Pavlicek, K. Malkova, J. Chromatogr. A 16 (2013) 94. [55] F. Robert, S. Fendri, L. Hary, C. Lacroix, M. Andrejak, J.D. Lalau, Diabet. Metab. 29 (2003) 279.

Large volume sample stacking for rapid and sensitive determination of antidiabetic drug metformin in human urine and serum by capillary electrophoresis with contactless conductivity detection.

Two CE methods with contactless conductivity detection have been developed for determining the oral antidiabetic drug metformin in human urine and blo...
543KB Sizes 0 Downloads 4 Views