- Email: [email protected]
Comparison of UV and charged aerosol detection approach in pharmaceutical analysis of statins
Talanta 78 (2009) 834–839
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Comparison of UV and charged aerosol detection approach in pharmaceutical analysis of statins ˇ Lucie Nováková a,∗ , Sofía Arnal Lopéz b , Dagmar Solichová c , Dalibor Satínsk y´ a , d d,e a Bohumila Kulichová , Aleˇs Horna , Petr Solich a
Department of Analytical Chemistry, Faculty of Pharmacy, Charles University, Heyrovského 1203, 500 05 Hradec Králové, Czech Republic Faculty of Chemistry, University of Valencia, Valencia, Spain c Department of Metabolic Care and Gerontology, Charles University, Faculty of Medicine and University Hospital in Hradec Králové, Sokolská 581, 500 05 Hradec Králové, Czech Republic d Radanal Ltd., Okruˇzní 613, 530 03 Pardubice, Czech Republic e Tomas Bata University in Zlín, University Institute, T.G. Masaryka 5555, 760 01 Zlín, Czech Republic b
a r t i c l e
i n f o
Article history: Received 23 July 2008 Received in revised form 17 December 2008 Accepted 22 December 2008 Available online 15 January 2009 Keywords: Atorvastatin Lovastatin Simvastatin HPLC CAD Pharmaceutical analysis
a b s t r a c t CAD (charged aerosol detector) has recently become a new alternative detection system in HPLC. This detection approach was applied in a new HPLC method for the determination of three of the major statins used in clinical treatment—simvastatin, lovastatin and atorvastatin. The method was optimized and the influence of individual parameters on CAD response and sensitivity was carefully studied. Chromatography was performed on a Zorbax Eclipse XDB C18 (4.6 mm × 75 mm, 3.5 m), using acetonitrile and formic acid 0.1% as mobile phase. The detection was performed using both CAD (20 pA range) and DAD (diode array detector—238 nm) simultaneously connected in series. In terms of linearity, precision and accuracy, the method was validated using tablets containing atorvastatin and simvastatin. The CAD is designated to be a non-linear detector in a wide dynamic range, however, in this application and in the tested concentration range its response was found to be perfectly linear. The limits of quantitation (0.1 g/ml) were found to be two times lower than those of UV detection. © 2009 Elsevier B.V. All rights reserved.
1. Introduction The CAD (charged aerosol detector) belongs among universal detectors and it operates independent of the physiochemical and spectral properties of non-volatile analytes. The eluent of the chromatographic system is first nebulized using a flow of nitrogen. Created droplets are dried in a drift tube to remove mobile phase, producing analyte particles. A secondary stream of nitrogen becomes positively charged as it passes through a high-voltage, platinum corona wire. The charged nitrogen is then mixed with the stream of analyte particles where the charge migrates to analyte. Charged analyte particles proceed to a collector region where a highly sensitive electrometer produces a signal that is proportional to the weight of sample present, independent of chemical structure [1,2]. Recently CAD has been used in the analysis of triacylglycerols [3], synthetic polymers [4] in analysis of squalene, cholesterol and ceramide [5] and in analysis of lipids [6]. Phar-
∗ Corresponding author. Tel.: +420 495067345; fax: +420 495067164. E-mail address: [email protected] (L. Nováková). 0039-9140/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.12.057
maceutical applications of CAD are very rare, it appears only two papers have been published [7,8]. Brunelli et al. coupled CAD to supercritical fluid chromatography in the analysis of a model standard mixture of theophylline, testosterone, cortisone, naproxen, sulfadimidine, sulfamerazine, sulfamethoxazole, sulfaquinoxaline and sulfamethizole [7]. The CAD has been found to achieve a higher level of sensitivity than other universal detectors, such as ELSD or RI. The second paper compared the performance of ELSD and CAD in pharmaceutical analysis, also using standard mixtures of azole derivatives and some other compounds [8]. Practical applications in the field of pharmacy, using CAD on drug formulations, are still missing. Statins drugs are commonly used for the treatment of several forms of hypercholesterolemia. They possess high effectiveness in reducing total cholesterol and low-density lipoprotein (LDL) cholesterol levels in human body. Statins are able to significantly reduce the morbidity and mortality associated with coronary heart disease, as demonstrated in numerous clinical trials [9–12]. Statins include natural (lovastatin), semi-synthetic (simvastatin, and pravastatin) as well as synthetic compounds (fluvastatin, atorvastatin, cerivastatin, rosuvastatin and pitavastatin). They are potent specific and competitive inhibitors of 3-hydroxy-3-methlyglutaryl coenzyme A
L. Nováková et al. / Talanta 78 (2009) 834–839
835
Fig. 1. Chemical structures of simvastatin, lovastatin and atorvastatin.
(HMG-CoA) reductase, which is a key enzyme that catalyzes the conversion of HMG-CoA to mevalonate. This is an early rate-limiting step in the cholesterol biosynthetic pathway [9,13,14]. Simvastatin and atorvastatin are two of the most commercially common drugs available as pharmaceutical formulations used for the clinical treatment of hypercholesterolemia. Structures can be seen in Fig. 1. The determination of drugs is a multi-disciplinary task. During the drug manufacturing process there is a need for analytical methods of quality control to discover any impurities. Bio-analytical methods are necessary for the clinical trials, for therapeutic drug monitoring and individual dosage scheme adjustment. Recently there has been a lot interest in monitoring pharmaceutical residuals in the environment, outlining a need for environmentally focused analytical methods. It is important to note, that higher sensitivity and selectivity is needed for the majority of bio-analytical or environmental methods. High performance liquid chromatography (HPLC) together with various types of detection – UV (ultraviolet), FD (fluorescence detection) and MS (mass spectrometry) – is the technique of choice during pharmaceutical QC method development. Analytical methods, used for the determination of simvastatin and atorvastatin, were recently reviewed by our group [15]. In pharmaceutical applications UV detection was most commonly used. Identification of impurities in simvastatin the bulk drug and tablets was done by Vuletic et al. [16] using LC–MS/MS approach. The simvastatin assay in the tablets was determined by HPLC–UV by Malenovic et al. using microemulsion eluent [17]. The analytical method for HPLC–UV determination of group of five statins (atorvastatin, lovastatin, pravastatin, rosuvastatin and simvastatin) was developed by Pasha et al. [18]. Erturk et al. used a simple HPLC–UV method to determine atorvastatin and impurities in both the bulk drug and tablets forms [19]. A similar method to determine atorvastatin without impurities was achieved by Altuntas et al. [20]. Atorvastatin together with amlodipine, in combined commercial tablets was also determined by HPLC–UV by Mohammadi et al. [21]. CAD coupled to
HPLC however, has not yet been used in pharmaceutical applications with drug formulations. The aim of this study was to develop and validate a new method for the determination of atorvastatin and simvastatin in real tablet samples using CAD. The results generated would then be compared with those obtained using UV detection. 2. Experimental 2.1. Chemicals and reagents Working standards of simvastatin (97%, HPLC), lovastatin (98%, HPLC) and atorvastatin were used during this study. The first two were obtained from Sigma–Aldrich (Prague, Czech Republic), while atorvastatin was obtained from Zentiva (Prague, Czech Republic). The composition of drug formulations was as follows: simvastatin (excipients: butylhydroxyanisol, ascorbic acid, citric acid, microcrystallic cellulose, magnesium stearate, lactose, hypromelose, titanium oxide, talc and ferric oxide). Atorvastatin drug formulation contained excipients as follows: calcium carbonate, microcrystalic cellulose, lactose, polysorbate, hyprolose and magnesium stearate. Both the formic acid, reagent grade, and the acetonitrile, HPLC gradient grade, were purchased from Sigma–Aldrich. HPLC grade water was prepared by Milli-Q reverse osmosis Millipore (Bedford, MA, USA) and it meets European Pharmacopoeia requirements. 2.2. Chromatography A Shimadzu Prominence LC 20 system (Shimadzu, Kyoto, Japan) was used to perform all of the analyses. The instrument was equipped with a column oven SIL-20 AC enabling temperature control. The built-in auto-sampler CTO-20 AC also enabled cooling. Chromatographic software Lab Solution was used for data collection and processing. Detection of statins was accomplished using a
836
L. Nováková et al. / Talanta 78 (2009) 834–839
Fig. 2. The influence of various additives to CAD response.
diode array detector SPD-M20A. A Corona CAD detector (ESA, USA) was connected in series. A Zorbax Eclipse XDB C18 (75 mm × 4.6 mm, 3.5 m) analytical column (Agilent, Czech Republic), was used for the HPLC separation of the three statins. The column oven temperature was kept at 30 ◦ C. The binary mobile phase, which composed of acetonitrile and 0.1% formic acid (70:30), was pumped at flow-rate 1.0 ml/min. DAD detection was performed at 238 nm. The injection volume was 10 l and the autosampler was cooled to 4 ◦ C. Optimal CAD detection was performed using a nitrogen pressure of 35 psi and range of 20 pA.
precision, accuracy and the assay of statins was established. For precision, six samples of tablets were tested for each preparation at 100% level of simvastatin (10 and 20 mg) and atorvastatin (20 mg) content, which corresponds to ICH (International Conference on Harmonization) requirements. Accuracy was determined by spiking samples by known amount of statin (10 g/ml).
2.3. Preparation of standard solutions and samples
The separation of the three statins, atorvastatin, simvastatin and lovastatin was not a difficult analytical task. The percentage at least of 30 of water part in the mobile phase had to be kept to get a good resolution of all compounds within reasonable time period. Volatile additives are recommended for the coupling with CAD, thus this approach was respected and volatile additives were used during the method development and further experiments. DAD detection was achieved at 238 nm, which was determined from the absorbance spectra of the individual compounds. The individual parameters which could influence the response of CAD were tested as follows: the range on CAD, the ratio of water/organic part of mobile phase, the flow-rate changes and various additives including formic acid, acetic acid and ammonium acetate buffers at pH 4.0, 5.0 and 7.0 at different concentrations—see Figs. 2 and 3. The range set-up is a key parameter which influences the sensitivity of the CAD. It was tested as follows: 1, 2, 5, 10, 20, 50, 100, 200, and 500 pA. A difference could be seen in the response according to range as well as comparing to the response of UV detection. The best responses were obtained within the 1 and 2 pA range, however these values are practically impossible to use due to the very high signal to noise ratio. The value of 20 pA was chosen for further experiments because of the best S/N ratio. The influence of mobile phase additives including formic acid, acetic acid and ammonium acetate buffer at various concentrations and pH values were tested—see Fig. 2. Neither formic nor acetic acid in mobile phase and their changing concentrations (0.01–1%) was found to have a significant influence on CAD response. The ammonium acetate buffer pH 4.0, at all concentrations, showed a small amount of influence. However, increasing the buffer’s pH to 5.0 demonstrated a much stronger influence on the CAD response. Higher concentrations of buffer (5 mM) resulted in a much lower
The stock solutions of standards were prepared by dissolving 1.0 mg of each statin standard into 1.0 ml of either acetonitrile, for simvastatin and lovastatin, or a mixture of acetonitrile and water (50:50) for atorvastatin due to solubility reasons. Stock solutions were further diluted by acetonitrile to achieve a concentration 10 g/ml for SST (system suitability test) measurements, and to get individual points of calibration curve in the range 50–0.1 g/ml, using nine calibration points (50, 25, 10, 5, 2.5, 1.0, 0.5, 0.25 and 0.1 g/ml). Tablet samples were prepared by dissolution into 5.0 ml of 5 mM of an ammonium acetate buffer pH 4.0 (stability reasons) using an ultra-sonic bath. Subsequent dilutions of the simvastatin preparations used acetonitrile. A mixture of water and acetonitrile (50:50) was used for the atorvastatin tablets. The samples were filtrated through 0.45 m PTFE membrane prior to injection into HPLC system. 2.4. System suitability test and validation An important part of method validation is the SST, details of which are usually given in pharmacopoeias [22,23]. The SST was performed under optimized chromatographic conditions, using both UV and CAD detection approaches. Theoretical plates, peak asymmetry, resolution of individual compounds and the repeatability of reference standard solution injections have been established (retentions times and peak areas were checked). Calibration curves of all statins in the concentration range 50–0.1 g/ml were measured. The applicability of the method was verified on real samples of pharmaceutical tablets containing simvastatin and atorvastatin at 10 and/or 20 mg levels. Method
3. Results and discussion 3.1. Chromatographic conditions—UV detection and charged aerosol detection
L. Nováková et al. / Talanta 78 (2009) 834–839
837
Table 1 System suitability results—analysis of three statins using UV and CAD detection approach. Atorvastatin SST Theoretical platesa UV 944 CAD 900 HETPa UV CAD
Lovastatin
Simvastatin
Limits
1534 1595
1623 1717
N > 900
158.81 166.80
97.77 94.05
92.41 87.35
Not given
Asymmetrya UV CAD
1.31 1.18
1.31 1.27
1.42 1.37
As < 1.5
Resolutiona UV CAD
– –
7.08 6.85
2.72 2.71
Rij > 1.5
Repeatability-tr a [% R.S.D.] UV 0.08 CAD 0.04
0.05 0.07
0.05 0.08
R.S.D. < 1%
Repeatability-Aa [% R.S.D.] UV 0.17 CAD 0.88
0.32 0.45
0.19 0.73
R.S.D. < 1%
UV = UV detection; CAD = charged aerosol detection; HETP = height equivalent of theoretical plates. a Made in 10 replicates.
Fig. 3. The influence of flow-rate to CAD response. The influence of composition of mobile phase to CAD response.
response of CAD. At pH 7, a lower response and lower signal to noise ratio were observed as well as very low quality data. The influence of the mobile phase was tested by changing the ratio of organic/water content. The range of 10–40% of formic acid 0.1% in the mixture with acetonitrile was tested. The best results were obtained using 20% formic acid in the mobile phase. The difference comparing to 10% of formic acid was about 50% of detector response, the difference comparing to the content of 40% of formic acid was about 30% of detector response—see Fig. 3. A mobile phase containing 30% formic acid 0.1% was found to demonstrate a higher
degree of resolution between peaks of the simvastatin and lovastatin. The response of the CAD was found to have a slightly dependence on the mobile phase flow-rate. It was higher with lower flow-rates, such as 0.6 ml/min. The difference was about 25% of peak area comparing to 1.0 ml/min flow-rate or almost 40% of peak area comparing to the flow-rate 1.4 ml/min—see Fig. 3. 3.2. System suitability test and validation The SST was performed by 10 subsequent injections of mixed solutions of all statins which were analyzed under optimum conditions. A typical chromatogram could be seen in Fig. 4. Parameters such as number of theoretical plates, peak asymmetry, resolution of individual compounds and the repeatability of reference standard solution injection (retentions times and peak areas were checked,
Fig. 4. Typical chromatogram of the separation of standard mixture of statins—CAD.
838
L. Nováková et al. / Talanta 78 (2009) 834–839
Fig. 5. Analysis of drug formulations: atorvastatin and simvastatin tablets—CAD (top chromatogram the analysis of simvastatin tablets 10 mg, bottom chromatogram the analysis of atorvastatin tablets 20 mg).
the repeatability was expressed as R.S.D. in %) were established. SST results were compared for UV and CAD detection approach—Table 1. Both the UV and CAD measurements produced results which met the requirements of the appropriate authorities (see the last column in Table 1) concerning all SST parameters. Excellent repeatability of injection, noted by the peak retention time and peak area, was observed for both detectors (R.S.D. < 1%). 3.2.1. Linearity–calibration range Calibration curves for the three statins in the concentration range 0.1–50 g/ml were measured. Results can be seen in Table 2. Nine calibration points were obtained for simvastatin and lovastatin (0.50 g/ml) and seven calibration points were obtained for atorvastatin (0.1–50 g/ml). The calibration curves were linear in the defined range, so it can be concluded that the method was appropriate for quantitative purposes for both UV and CAD detection approach in spite of statement, that CAD is not universally linear detector. 3.2.2. Accuracy and precision Method accuracy and precision were tested using tablet samples of atorvastatin and simvastatin containing 10 or 20 mg of
active substance. Six samples of tablets were prepared for each experiment. Precision was expressed as % R.S.D. of six determinations. Accuracy was established by spiking of statin tablets with a known amount of statin standard and was expressed as % of recovery—see Table 2. All the results were in correspondence with the requirements for method validation in pharmaceutical application. 3.2.3. Assay The assay of the active substance, atorvastatin and simvastatin, in atorvastatin and simvastatin tablets was determined (Fig. 5). The amounts of 100 ± 5% of declared assay were found by CAD, the amounts 100 ± 10% of declared assay were found by UV detector, which corresponds to the requirements (up to 100 ± 10% of declared assay) [23], see Table 2. 3.2.4. Limits of detection and quantitation Limits of detection and quantitation were established for both detection approaches. They were found to be two times higher than the UV detection approach, see Table 2, which is advantageous comparing to ELSD or RI detectors.
L. Nováková et al. / Talanta 78 (2009) 834–839
839
Table 2 Method validation results for simvastatin and atorvastatin pharmaceutical formulations using UV and CAD detection approach. Validation
Atorvastatin
Simvastatin
Limits
Level
20 mg
Level
Assaya [% of declared content]
UV
90.2%
20 mg 10 mg
100.5 94.5
Declared amount ±10%
Assaya [% of declared content]
CAD
95.7%
20 mg 10 mg
97.4 97.8
Declared amount ±10%
Accuracya [%]
UV
97.9
Recovery = 100 ± 5%
Accuracya [%]
CAD
105.1
Recovery = 100 ± 5%
a
Precision [% R.S.D.]
UV
3.47
20 mg 10 mg
3.80 3.80
R.S.D. < 5%
Precisiona [% R.S.D.]
CAD
2.11
20 mg 10 mg
4.43 4.79
R.S.D. < 5%
Linearityb (correlation coefficient) Linearityb (equation)
UV UV
0.9991 y = 2E+07x−9905.9
0.9999 y = 3E+07x−6090.5
R > 0.9990 –
Linearityb (correlation coefficient) Linearityb (equation)
CAD CAD
0.9996 y = 1E+08x−28415
0.9995 y = 3E+07x−12809
Non linear detector! –
LOQ LOD
UV UV
0.50 g/ml 0.17 g/ml
0.25 g/ml 0.08 g/ml
– –
LOQ LOD
CAD CAD
0.25 g/ml 0.08 g/ml
0.10 g/ml 0.03 g/ml
– –
a b
Made in six replicates. Nine calibration levels for simvastatin, seven calibration levels for atorvastatin, each injected in three replicates.
4. Conclusions
References [1] [2] [3] [4]
The new analytical methods were developed for the determination of three statins—atorvastatin, lovastatin and simvastatin using UV and CAD detection approaches, which were compared in this study. Using a Zorbax Eclipse XDB C18 stationary phase, the separation was completed in 4.5 min with a simple volatile mobile phase, composed of acetonitrile and formic acid 0.1% (70:30) at the flow-rate 1.0 ml/min. DAD detection was performed at 238 nm, CAD worked in the 20 pA range. The results obtained in method optimization showed an influence of the mobile phase flow-rate about 25–40% decrease of detector response, the influence of the composition of mobile phase about 30–40% of the decrease of the detector response and a strong negative impact was demonstrated when using higher concentration of buffers at pH > 4. The SST and validation results were in good agreement with validation requirements for both detectors. The method repeatability in the frame of SST showed a R.S.D. lower than 1%. Both detectors gave linear response in the tested range, CAD in spite of belonging among non-linear detectors. The sensitivity of CAD detection was two times greater than the UV detection when applied to simvastatin, atorvastatin and lovastatin analysis.
[20] [21]
Acknowledgements
[22]
The authors gratefully acknowledge the financial support of IGA MZ CR No. 1A/8689-4 and MSM 0021620822.
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
[23]
R.W. Dixon, D.S. Peterson, Anal. Chem. 74 (2002) 2930. I. Sinclair, R. Gallagher, Chromatogr. Today 1 (2008) 5. M. Lísa, F. Lynen, M. Holˇcapek, P. Sandra, J. Chromatogr. A 1176 (2007) 135. K. Takahashi, S. Kinugasa, M. Senda, K. Kimizuka, K. Fukushima, T. Matsumoto, Y. Shibata, J. Christensen, J. Chromatogr. A 1193 (2008) 151. A. Hazzotte, D. Libong, M. Matoga, P. Chaminade, J. Chromatogr. A 1170 (2007) 52. R.A. Moreau, Lipids 41 (2006) 727. C. Brunelli, T. Gorecki, Y. Zao, P. Sandra, Anal. Chem. 79 (2007) 2472. N. Vervoort, D. Daemen, G. Torok, J. Chromatogr. A 1189 (2008) 92. Y. Shitara, Y. Sugiyama, Pharmacol. Ther. 112 (2006) 71–105. World Health Organization, World Health Report, Report of the DirectorGeneral, Geneva, WHO, 1998. F.M. Sacks, Am. J. Cardiol. 88 (Suppl.) (2001) 14N. S. Bertolini, G.B. Bon, L.M. Campbell, M. Farnier, J. Lagan, G. Mahla, Atherosclerosis 130 (1997) 191–197. L.Y. Ye, P.S. Firby, M.J. Moore, Ther. Drug Monit. 22 (2000) 737. K.J. Williams, I. Tabas, Atheroscler. Thromb. Vasc. Biol. 15 (1995) 551. ˇ ´ P. Solich, Trends Anal. Chem. 27 (2008) 352. L. Nováková, D. Satínsk y, M. Vuletic, M. Cindric, J.D. Koruˇznjak, J. Pharm. Biomed. Anal. 37 (2005) 715. A. Malenovic, M. Medenica, D. Ivanovic, B. Jancic, Chromatography (Suppl. 63) (2006) S95. M.K. Pasha, S. Muzeeb, S.J.S. Basha, D. Shashikumar, R. Mullangi, N.R. Srinivas, Biomed. Chromatogr. 20 (2006) 282. S. Erturk, E.S. Aktas, L. Ersoy, S. Ficicioglu, J. Pharm. Biomed. Anal. 33 (2003) 1017. T.G. Altuntas, N. Erk, J. Liq. Chromatogr. Relat. Technol. 27 (2004) 83. A. Mohammadi, N. Rezanour, M.A. Dogahen, F.G. Bidkorbeh, M. Hashem, R.B. Walker, J. Chromatogr. B 846 (2007) 215. European Pharmacopoeia 5 edition (Ph. Eur. 5), Council of Europe, Strasbourg, 2004. United States Pharmacopoeia 30, United States Pharmacopoeial Convention, Rockville, MD 20852, United States, 2007.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Comparison of UV and charged aerosol detection approach in pharmaceutical analysis of statins ˇ Lucie Nováková a,∗ , Sofía Arnal Lopéz b , Dagmar Solichová c , Dalibor Satínsk y´ a , d d,e a Bohumila Kulichová , Aleˇs Horna , Petr Solich a
Department of Analytical Chemistry, Faculty of Pharmacy, Charles University, Heyrovského 1203, 500 05 Hradec Králové, Czech Republic Faculty of Chemistry, University of Valencia, Valencia, Spain c Department of Metabolic Care and Gerontology, Charles University, Faculty of Medicine and University Hospital in Hradec Králové, Sokolská 581, 500 05 Hradec Králové, Czech Republic d Radanal Ltd., Okruˇzní 613, 530 03 Pardubice, Czech Republic e Tomas Bata University in Zlín, University Institute, T.G. Masaryka 5555, 760 01 Zlín, Czech Republic b
a r t i c l e
i n f o
Article history: Received 23 July 2008 Received in revised form 17 December 2008 Accepted 22 December 2008 Available online 15 January 2009 Keywords: Atorvastatin Lovastatin Simvastatin HPLC CAD Pharmaceutical analysis
a b s t r a c t CAD (charged aerosol detector) has recently become a new alternative detection system in HPLC. This detection approach was applied in a new HPLC method for the determination of three of the major statins used in clinical treatment—simvastatin, lovastatin and atorvastatin. The method was optimized and the influence of individual parameters on CAD response and sensitivity was carefully studied. Chromatography was performed on a Zorbax Eclipse XDB C18 (4.6 mm × 75 mm, 3.5 m), using acetonitrile and formic acid 0.1% as mobile phase. The detection was performed using both CAD (20 pA range) and DAD (diode array detector—238 nm) simultaneously connected in series. In terms of linearity, precision and accuracy, the method was validated using tablets containing atorvastatin and simvastatin. The CAD is designated to be a non-linear detector in a wide dynamic range, however, in this application and in the tested concentration range its response was found to be perfectly linear. The limits of quantitation (0.1 g/ml) were found to be two times lower than those of UV detection. © 2009 Elsevier B.V. All rights reserved.
1. Introduction The CAD (charged aerosol detector) belongs among universal detectors and it operates independent of the physiochemical and spectral properties of non-volatile analytes. The eluent of the chromatographic system is first nebulized using a flow of nitrogen. Created droplets are dried in a drift tube to remove mobile phase, producing analyte particles. A secondary stream of nitrogen becomes positively charged as it passes through a high-voltage, platinum corona wire. The charged nitrogen is then mixed with the stream of analyte particles where the charge migrates to analyte. Charged analyte particles proceed to a collector region where a highly sensitive electrometer produces a signal that is proportional to the weight of sample present, independent of chemical structure [1,2]. Recently CAD has been used in the analysis of triacylglycerols [3], synthetic polymers [4] in analysis of squalene, cholesterol and ceramide [5] and in analysis of lipids [6]. Phar-
∗ Corresponding author. Tel.: +420 495067345; fax: +420 495067164. E-mail address: [email protected] (L. Nováková). 0039-9140/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.12.057
maceutical applications of CAD are very rare, it appears only two papers have been published [7,8]. Brunelli et al. coupled CAD to supercritical fluid chromatography in the analysis of a model standard mixture of theophylline, testosterone, cortisone, naproxen, sulfadimidine, sulfamerazine, sulfamethoxazole, sulfaquinoxaline and sulfamethizole [7]. The CAD has been found to achieve a higher level of sensitivity than other universal detectors, such as ELSD or RI. The second paper compared the performance of ELSD and CAD in pharmaceutical analysis, also using standard mixtures of azole derivatives and some other compounds [8]. Practical applications in the field of pharmacy, using CAD on drug formulations, are still missing. Statins drugs are commonly used for the treatment of several forms of hypercholesterolemia. They possess high effectiveness in reducing total cholesterol and low-density lipoprotein (LDL) cholesterol levels in human body. Statins are able to significantly reduce the morbidity and mortality associated with coronary heart disease, as demonstrated in numerous clinical trials [9–12]. Statins include natural (lovastatin), semi-synthetic (simvastatin, and pravastatin) as well as synthetic compounds (fluvastatin, atorvastatin, cerivastatin, rosuvastatin and pitavastatin). They are potent specific and competitive inhibitors of 3-hydroxy-3-methlyglutaryl coenzyme A
L. Nováková et al. / Talanta 78 (2009) 834–839
835
Fig. 1. Chemical structures of simvastatin, lovastatin and atorvastatin.
(HMG-CoA) reductase, which is a key enzyme that catalyzes the conversion of HMG-CoA to mevalonate. This is an early rate-limiting step in the cholesterol biosynthetic pathway [9,13,14]. Simvastatin and atorvastatin are two of the most commercially common drugs available as pharmaceutical formulations used for the clinical treatment of hypercholesterolemia. Structures can be seen in Fig. 1. The determination of drugs is a multi-disciplinary task. During the drug manufacturing process there is a need for analytical methods of quality control to discover any impurities. Bio-analytical methods are necessary for the clinical trials, for therapeutic drug monitoring and individual dosage scheme adjustment. Recently there has been a lot interest in monitoring pharmaceutical residuals in the environment, outlining a need for environmentally focused analytical methods. It is important to note, that higher sensitivity and selectivity is needed for the majority of bio-analytical or environmental methods. High performance liquid chromatography (HPLC) together with various types of detection – UV (ultraviolet), FD (fluorescence detection) and MS (mass spectrometry) – is the technique of choice during pharmaceutical QC method development. Analytical methods, used for the determination of simvastatin and atorvastatin, were recently reviewed by our group [15]. In pharmaceutical applications UV detection was most commonly used. Identification of impurities in simvastatin the bulk drug and tablets was done by Vuletic et al. [16] using LC–MS/MS approach. The simvastatin assay in the tablets was determined by HPLC–UV by Malenovic et al. using microemulsion eluent [17]. The analytical method for HPLC–UV determination of group of five statins (atorvastatin, lovastatin, pravastatin, rosuvastatin and simvastatin) was developed by Pasha et al. [18]. Erturk et al. used a simple HPLC–UV method to determine atorvastatin and impurities in both the bulk drug and tablets forms [19]. A similar method to determine atorvastatin without impurities was achieved by Altuntas et al. [20]. Atorvastatin together with amlodipine, in combined commercial tablets was also determined by HPLC–UV by Mohammadi et al. [21]. CAD coupled to
HPLC however, has not yet been used in pharmaceutical applications with drug formulations. The aim of this study was to develop and validate a new method for the determination of atorvastatin and simvastatin in real tablet samples using CAD. The results generated would then be compared with those obtained using UV detection. 2. Experimental 2.1. Chemicals and reagents Working standards of simvastatin (97%, HPLC), lovastatin (98%, HPLC) and atorvastatin were used during this study. The first two were obtained from Sigma–Aldrich (Prague, Czech Republic), while atorvastatin was obtained from Zentiva (Prague, Czech Republic). The composition of drug formulations was as follows: simvastatin (excipients: butylhydroxyanisol, ascorbic acid, citric acid, microcrystallic cellulose, magnesium stearate, lactose, hypromelose, titanium oxide, talc and ferric oxide). Atorvastatin drug formulation contained excipients as follows: calcium carbonate, microcrystalic cellulose, lactose, polysorbate, hyprolose and magnesium stearate. Both the formic acid, reagent grade, and the acetonitrile, HPLC gradient grade, were purchased from Sigma–Aldrich. HPLC grade water was prepared by Milli-Q reverse osmosis Millipore (Bedford, MA, USA) and it meets European Pharmacopoeia requirements. 2.2. Chromatography A Shimadzu Prominence LC 20 system (Shimadzu, Kyoto, Japan) was used to perform all of the analyses. The instrument was equipped with a column oven SIL-20 AC enabling temperature control. The built-in auto-sampler CTO-20 AC also enabled cooling. Chromatographic software Lab Solution was used for data collection and processing. Detection of statins was accomplished using a
836
L. Nováková et al. / Talanta 78 (2009) 834–839
Fig. 2. The influence of various additives to CAD response.
diode array detector SPD-M20A. A Corona CAD detector (ESA, USA) was connected in series. A Zorbax Eclipse XDB C18 (75 mm × 4.6 mm, 3.5 m) analytical column (Agilent, Czech Republic), was used for the HPLC separation of the three statins. The column oven temperature was kept at 30 ◦ C. The binary mobile phase, which composed of acetonitrile and 0.1% formic acid (70:30), was pumped at flow-rate 1.0 ml/min. DAD detection was performed at 238 nm. The injection volume was 10 l and the autosampler was cooled to 4 ◦ C. Optimal CAD detection was performed using a nitrogen pressure of 35 psi and range of 20 pA.
precision, accuracy and the assay of statins was established. For precision, six samples of tablets were tested for each preparation at 100% level of simvastatin (10 and 20 mg) and atorvastatin (20 mg) content, which corresponds to ICH (International Conference on Harmonization) requirements. Accuracy was determined by spiking samples by known amount of statin (10 g/ml).
2.3. Preparation of standard solutions and samples
The separation of the three statins, atorvastatin, simvastatin and lovastatin was not a difficult analytical task. The percentage at least of 30 of water part in the mobile phase had to be kept to get a good resolution of all compounds within reasonable time period. Volatile additives are recommended for the coupling with CAD, thus this approach was respected and volatile additives were used during the method development and further experiments. DAD detection was achieved at 238 nm, which was determined from the absorbance spectra of the individual compounds. The individual parameters which could influence the response of CAD were tested as follows: the range on CAD, the ratio of water/organic part of mobile phase, the flow-rate changes and various additives including formic acid, acetic acid and ammonium acetate buffers at pH 4.0, 5.0 and 7.0 at different concentrations—see Figs. 2 and 3. The range set-up is a key parameter which influences the sensitivity of the CAD. It was tested as follows: 1, 2, 5, 10, 20, 50, 100, 200, and 500 pA. A difference could be seen in the response according to range as well as comparing to the response of UV detection. The best responses were obtained within the 1 and 2 pA range, however these values are practically impossible to use due to the very high signal to noise ratio. The value of 20 pA was chosen for further experiments because of the best S/N ratio. The influence of mobile phase additives including formic acid, acetic acid and ammonium acetate buffer at various concentrations and pH values were tested—see Fig. 2. Neither formic nor acetic acid in mobile phase and their changing concentrations (0.01–1%) was found to have a significant influence on CAD response. The ammonium acetate buffer pH 4.0, at all concentrations, showed a small amount of influence. However, increasing the buffer’s pH to 5.0 demonstrated a much stronger influence on the CAD response. Higher concentrations of buffer (5 mM) resulted in a much lower
The stock solutions of standards were prepared by dissolving 1.0 mg of each statin standard into 1.0 ml of either acetonitrile, for simvastatin and lovastatin, or a mixture of acetonitrile and water (50:50) for atorvastatin due to solubility reasons. Stock solutions were further diluted by acetonitrile to achieve a concentration 10 g/ml for SST (system suitability test) measurements, and to get individual points of calibration curve in the range 50–0.1 g/ml, using nine calibration points (50, 25, 10, 5, 2.5, 1.0, 0.5, 0.25 and 0.1 g/ml). Tablet samples were prepared by dissolution into 5.0 ml of 5 mM of an ammonium acetate buffer pH 4.0 (stability reasons) using an ultra-sonic bath. Subsequent dilutions of the simvastatin preparations used acetonitrile. A mixture of water and acetonitrile (50:50) was used for the atorvastatin tablets. The samples were filtrated through 0.45 m PTFE membrane prior to injection into HPLC system. 2.4. System suitability test and validation An important part of method validation is the SST, details of which are usually given in pharmacopoeias [22,23]. The SST was performed under optimized chromatographic conditions, using both UV and CAD detection approaches. Theoretical plates, peak asymmetry, resolution of individual compounds and the repeatability of reference standard solution injections have been established (retentions times and peak areas were checked). Calibration curves of all statins in the concentration range 50–0.1 g/ml were measured. The applicability of the method was verified on real samples of pharmaceutical tablets containing simvastatin and atorvastatin at 10 and/or 20 mg levels. Method
3. Results and discussion 3.1. Chromatographic conditions—UV detection and charged aerosol detection
L. Nováková et al. / Talanta 78 (2009) 834–839
837
Table 1 System suitability results—analysis of three statins using UV and CAD detection approach. Atorvastatin SST Theoretical platesa UV 944 CAD 900 HETPa UV CAD
Lovastatin
Simvastatin
Limits
1534 1595
1623 1717
N > 900
158.81 166.80
97.77 94.05
92.41 87.35
Not given
Asymmetrya UV CAD
1.31 1.18
1.31 1.27
1.42 1.37
As < 1.5
Resolutiona UV CAD
– –
7.08 6.85
2.72 2.71
Rij > 1.5
Repeatability-tr a [% R.S.D.] UV 0.08 CAD 0.04
0.05 0.07
0.05 0.08
R.S.D. < 1%
Repeatability-Aa [% R.S.D.] UV 0.17 CAD 0.88
0.32 0.45
0.19 0.73
R.S.D. < 1%
UV = UV detection; CAD = charged aerosol detection; HETP = height equivalent of theoretical plates. a Made in 10 replicates.
Fig. 3. The influence of flow-rate to CAD response. The influence of composition of mobile phase to CAD response.
response of CAD. At pH 7, a lower response and lower signal to noise ratio were observed as well as very low quality data. The influence of the mobile phase was tested by changing the ratio of organic/water content. The range of 10–40% of formic acid 0.1% in the mixture with acetonitrile was tested. The best results were obtained using 20% formic acid in the mobile phase. The difference comparing to 10% of formic acid was about 50% of detector response, the difference comparing to the content of 40% of formic acid was about 30% of detector response—see Fig. 3. A mobile phase containing 30% formic acid 0.1% was found to demonstrate a higher
degree of resolution between peaks of the simvastatin and lovastatin. The response of the CAD was found to have a slightly dependence on the mobile phase flow-rate. It was higher with lower flow-rates, such as 0.6 ml/min. The difference was about 25% of peak area comparing to 1.0 ml/min flow-rate or almost 40% of peak area comparing to the flow-rate 1.4 ml/min—see Fig. 3. 3.2. System suitability test and validation The SST was performed by 10 subsequent injections of mixed solutions of all statins which were analyzed under optimum conditions. A typical chromatogram could be seen in Fig. 4. Parameters such as number of theoretical plates, peak asymmetry, resolution of individual compounds and the repeatability of reference standard solution injection (retentions times and peak areas were checked,
Fig. 4. Typical chromatogram of the separation of standard mixture of statins—CAD.
838
L. Nováková et al. / Talanta 78 (2009) 834–839
Fig. 5. Analysis of drug formulations: atorvastatin and simvastatin tablets—CAD (top chromatogram the analysis of simvastatin tablets 10 mg, bottom chromatogram the analysis of atorvastatin tablets 20 mg).
the repeatability was expressed as R.S.D. in %) were established. SST results were compared for UV and CAD detection approach—Table 1. Both the UV and CAD measurements produced results which met the requirements of the appropriate authorities (see the last column in Table 1) concerning all SST parameters. Excellent repeatability of injection, noted by the peak retention time and peak area, was observed for both detectors (R.S.D. < 1%). 3.2.1. Linearity–calibration range Calibration curves for the three statins in the concentration range 0.1–50 g/ml were measured. Results can be seen in Table 2. Nine calibration points were obtained for simvastatin and lovastatin (0.50 g/ml) and seven calibration points were obtained for atorvastatin (0.1–50 g/ml). The calibration curves were linear in the defined range, so it can be concluded that the method was appropriate for quantitative purposes for both UV and CAD detection approach in spite of statement, that CAD is not universally linear detector. 3.2.2. Accuracy and precision Method accuracy and precision were tested using tablet samples of atorvastatin and simvastatin containing 10 or 20 mg of
active substance. Six samples of tablets were prepared for each experiment. Precision was expressed as % R.S.D. of six determinations. Accuracy was established by spiking of statin tablets with a known amount of statin standard and was expressed as % of recovery—see Table 2. All the results were in correspondence with the requirements for method validation in pharmaceutical application. 3.2.3. Assay The assay of the active substance, atorvastatin and simvastatin, in atorvastatin and simvastatin tablets was determined (Fig. 5). The amounts of 100 ± 5% of declared assay were found by CAD, the amounts 100 ± 10% of declared assay were found by UV detector, which corresponds to the requirements (up to 100 ± 10% of declared assay) [23], see Table 2. 3.2.4. Limits of detection and quantitation Limits of detection and quantitation were established for both detection approaches. They were found to be two times higher than the UV detection approach, see Table 2, which is advantageous comparing to ELSD or RI detectors.
L. Nováková et al. / Talanta 78 (2009) 834–839
839
Table 2 Method validation results for simvastatin and atorvastatin pharmaceutical formulations using UV and CAD detection approach. Validation
Atorvastatin
Simvastatin
Limits
Level
20 mg
Level
Assaya [% of declared content]
UV
90.2%
20 mg 10 mg
100.5 94.5
Declared amount ±10%
Assaya [% of declared content]
CAD
95.7%
20 mg 10 mg
97.4 97.8
Declared amount ±10%
Accuracya [%]
UV
97.9
Recovery = 100 ± 5%
Accuracya [%]
CAD
105.1
Recovery = 100 ± 5%
a
Precision [% R.S.D.]
UV
3.47
20 mg 10 mg
3.80 3.80
R.S.D. < 5%
Precisiona [% R.S.D.]
CAD
2.11
20 mg 10 mg
4.43 4.79
R.S.D. < 5%
Linearityb (correlation coefficient) Linearityb (equation)
UV UV
0.9991 y = 2E+07x−9905.9
0.9999 y = 3E+07x−6090.5
R > 0.9990 –
Linearityb (correlation coefficient) Linearityb (equation)
CAD CAD
0.9996 y = 1E+08x−28415
0.9995 y = 3E+07x−12809
Non linear detector! –
LOQ LOD
UV UV
0.50 g/ml 0.17 g/ml
0.25 g/ml 0.08 g/ml
– –
LOQ LOD
CAD CAD
0.25 g/ml 0.08 g/ml
0.10 g/ml 0.03 g/ml
– –
a b
Made in six replicates. Nine calibration levels for simvastatin, seven calibration levels for atorvastatin, each injected in three replicates.
4. Conclusions
References [1] [2] [3] [4]
The new analytical methods were developed for the determination of three statins—atorvastatin, lovastatin and simvastatin using UV and CAD detection approaches, which were compared in this study. Using a Zorbax Eclipse XDB C18 stationary phase, the separation was completed in 4.5 min with a simple volatile mobile phase, composed of acetonitrile and formic acid 0.1% (70:30) at the flow-rate 1.0 ml/min. DAD detection was performed at 238 nm, CAD worked in the 20 pA range. The results obtained in method optimization showed an influence of the mobile phase flow-rate about 25–40% decrease of detector response, the influence of the composition of mobile phase about 30–40% of the decrease of the detector response and a strong negative impact was demonstrated when using higher concentration of buffers at pH > 4. The SST and validation results were in good agreement with validation requirements for both detectors. The method repeatability in the frame of SST showed a R.S.D. lower than 1%. Both detectors gave linear response in the tested range, CAD in spite of belonging among non-linear detectors. The sensitivity of CAD detection was two times greater than the UV detection when applied to simvastatin, atorvastatin and lovastatin analysis.
[20] [21]
Acknowledgements
[22]
The authors gratefully acknowledge the financial support of IGA MZ CR No. 1A/8689-4 and MSM 0021620822.
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
[23]
R.W. Dixon, D.S. Peterson, Anal. Chem. 74 (2002) 2930. I. Sinclair, R. Gallagher, Chromatogr. Today 1 (2008) 5. M. Lísa, F. Lynen, M. Holˇcapek, P. Sandra, J. Chromatogr. A 1176 (2007) 135. K. Takahashi, S. Kinugasa, M. Senda, K. Kimizuka, K. Fukushima, T. Matsumoto, Y. Shibata, J. Christensen, J. Chromatogr. A 1193 (2008) 151. A. Hazzotte, D. Libong, M. Matoga, P. Chaminade, J. Chromatogr. A 1170 (2007) 52. R.A. Moreau, Lipids 41 (2006) 727. C. Brunelli, T. Gorecki, Y. Zao, P. Sandra, Anal. Chem. 79 (2007) 2472. N. Vervoort, D. Daemen, G. Torok, J. Chromatogr. A 1189 (2008) 92. Y. Shitara, Y. Sugiyama, Pharmacol. Ther. 112 (2006) 71–105. World Health Organization, World Health Report, Report of the DirectorGeneral, Geneva, WHO, 1998. F.M. Sacks, Am. J. Cardiol. 88 (Suppl.) (2001) 14N. S. Bertolini, G.B. Bon, L.M. Campbell, M. Farnier, J. Lagan, G. Mahla, Atherosclerosis 130 (1997) 191–197. L.Y. Ye, P.S. Firby, M.J. Moore, Ther. Drug Monit. 22 (2000) 737. K.J. Williams, I. Tabas, Atheroscler. Thromb. Vasc. Biol. 15 (1995) 551. ˇ ´ P. Solich, Trends Anal. Chem. 27 (2008) 352. L. Nováková, D. Satínsk y, M. Vuletic, M. Cindric, J.D. Koruˇznjak, J. Pharm. Biomed. Anal. 37 (2005) 715. A. Malenovic, M. Medenica, D. Ivanovic, B. Jancic, Chromatography (Suppl. 63) (2006) S95. M.K. Pasha, S. Muzeeb, S.J.S. Basha, D. Shashikumar, R. Mullangi, N.R. Srinivas, Biomed. Chromatogr. 20 (2006) 282. S. Erturk, E.S. Aktas, L. Ersoy, S. Ficicioglu, J. Pharm. Biomed. Anal. 33 (2003) 1017. T.G. Altuntas, N. Erk, J. Liq. Chromatogr. Relat. Technol. 27 (2004) 83. A. Mohammadi, N. Rezanour, M.A. Dogahen, F.G. Bidkorbeh, M. Hashem, R.B. Walker, J. Chromatogr. B 846 (2007) 215. European Pharmacopoeia 5 edition (Ph. Eur. 5), Council of Europe, Strasbourg, 2004. United States Pharmacopoeia 30, United States Pharmacopoeial Convention, Rockville, MD 20852, United States, 2007.