Simultaneous Densitometric Determination of Indomethacin and Its ...

3 downloads 0 Views 41KB Size Report
than that of phenylbutazone. Despite its high toxicity, indomethacin is a primary medicine used for treatment of rheumatoid arthritis, gout, and collagen disease ...
KRZEK & STAREK: JOURNAL OF AOAC INTERNATIONAL VOL. 84, NO. 6, 2001 1703 DRUGS, COSMETICS, FORENSIC SCIENCES

Simultaneous Densitometric Determination of Indomethacin and Its Degradation Products, 4-Chlorobenzoic Acid and 5-Methoxy-2-Methyl-3-Indoleacetic Acid, in Pharmaceutical Preparations JAN KRZEK and MAGORZATA STAREK Jagiellonian University, Collegium Medicum, Department of Inorganic and Analytical Chemistry, 9 Medyczna St, 30-699, Krak\w, Poland

A densitometric method was developed for the identification and determination of indomethacin and its degradation products, 4-chlorobenzoic acid and 5-methoxy-2-methyl-3-indoleacetic acid, in pharmaceuticals. To separate these compounds, silica gel-coated thin-layer chromatography plates and the following mobile phase were used: 2-propanol–25% ammonia–water (8 + 1 + 1, v/v). UV densitometric measurements were made by comparing the absorption spectra and Rf values of appropriate standards with the pharmaceutical preparations examined. The conditions for separation were established and a low detection limit was obtained. Average recoveries were 100.69, 90.09, and 91.17% for indomethacin, 4-chlorobeznzoic acid, and 5-methoxy-2-methyl-3-indoleacetic acid, respectively.

indomethacin and the accompanying impurities, i.e., 4-chlorobenzoic acid and 5-methoxy-2-methyl-3-indoleacetic acid. The research presented in this paper has been justified by previous successful applications of densitometric quantitation to the purity assessment of pharmaceuticals. It has been shown that this technique has many advantages, provides accurate, quick, and simple determinations, and can be an alternative to the commonly used LC method (16–19). In contrast with LC, the densitometric method presented in this paper allows analysis of several samples at the same time, which shortens the time of analysis. Moreover, because large amounts of organic solvents are not needed, the method does not produce much pollution in the environment. The concentration levels of the determined constituents used in the analysis are similar to those used in LC. Experimental

Apparatus ndomethacin is a derivative of indoleacetic acid with strong anti-inflammatory action, about 20 times stronger than that of phenylbutazone. Despite its high toxicity, indomethacin is a primary medicine used for treatment of rheumatoid arthritis, gout, and collagen disease (1). It has been reported that indomethacin relatively easily decomposes by hydrolysis to 5-methoxy-2-methyl-3-indoleacetic acid and 4-chlorobenzoic acid (Figure 1; 2). According to the U.S. Pharmacopeia (3), a liquid chromatographic (LC) method is recommended for the purity assessment and quantitative analysis of indomethacin preparations. The same method has also been used by many researchers to determine indomethacin and 4-chlorobenzoic acid both in medicines (4–6) and in body fluids (7–9). Indomethacin preparations have also been analyzed by gas chromatographic (6), thin-layer chromatographic (TLC; 10), spectrophotometric (11, 12), and electrochemical methods (13–15). The aim of the work described in this paper was to establish conditions for the simultaneous identification and determination of

I

Received July 5, 2000. Accepted by JM December 9, 2000. Corresponding author’s e-mail: [email protected].

(a) Densitometer.—Camag (Muttenz, Switzerland) TLC Scanner 3 with CAT3 software. (b) Sample applicator.—Linomat IV (Camag). (c) Computer.—PC Pentium MMX, 16 MB RAM, Hewlett-Packard LaserJet 6L printer and software (Microsoft Office 97, Statistica 5.1 edition ‘97). (d) TLC plates.—7 × 12 cm (cut from 20 × 20 cm precoated TLC sheets of silica gel on aluminum; Art. 1.05553; E. Merck, Darmstadt, Germany). (e) TLC chamber.—18 × 9 × 18 cm (Sigma Aldrich, Cat. No. Z20, 415-3).

Reagents (a) Standard solutions.—Indomethacin (I; CEM Co., Italy), 4-chlorobenzoic acid (C; Avocado, Cat. No. 15135) and 5-methoxy-2-methyl-3-indoleacetic acid (M; Sigma, Cat. No. M-7639) were dissolved in methanol so that solutions containing the standard at 20 µg/mL were obtained. (b) Sample solutions.—The following pharmaceutical preparations were used: Metindol injections (Polfa Cracow, Poland) and Metindol tablets (Polfa Warsaw, Poland), of different series, that were randomly sampled from current stock. In the case of injections, solutions were prepared in methanol at

1704 KRZEK & STAREK: JOURNAL OF AOAC INTERNATIONAL VOL. 84, NO. 6, 2001

Figure 1. Decomposition of (a) indomethacin to (b) 5-methoxy-2-methyl-3-indoleacetic acid and (c) 4-chlorobenzoic acid.

concentrations of up to 20 µg/mL, and various aliquots of solution were taken, depending on the analyte under determination (the volume of solution was 5 times greater for impurities than for indomethacin). The solutions of tablets were prepared by grinding 10 tablets and weighing (with accuracy of ±0.1 mg) ca 0.1 g, which was dissolved in methanol as above. (c) Mobile phase.—2-Propanol–25% ammonia–water (8 + 1 + 1, v/v). (d) Solvents.—2-Propanol (propan-2-ol), methanol, and 25% ammonia (POCH, Gliwice, Poland).

Preliminary Analysis To establish the optimum conditions for the method, including preliminary validation parameters (20), the research was performed by 2 independent analytical chemists. Effect of parameter changes on measurements.—This effect was considered in the preparation for analysis by taking into account the method of determination, sample preparation, time and height of chromatographic separation, method of detection, and changes in mobile phase. During densitometric

Figure 2. Absorption spectra recorded directly from chromatograms for constituents under examination: a = indomethacin; b = 4-chlorobenzoic acid; and c = 5-methoxy-2-methoyl-3-indoleacetic acid.

Figure 3. Densitogram (top) and chromatogram (bottom) for indomethacin (3), 4-chlorobenzoic acid (2), and 5-methoxy-2-methoyl-3-indoleacetic acid (1). The location of the spots in the chromatogram was established densitometrically in the UV region.

analysis, absorption spectra, peak areas, Rf values, and changes in absorbance were measured. Aliquots of 1–15 µL of standard and sample solutions were applied in the form of bands on chromatographic plates, 7 × 12 cm in size. Chromatograms were developed in the following mobile phase: 2-propanol–25% ammonia–water (8 + 1 + 1, v/v), which was established experimentally by comparisons with many other mobile phases such as the following: n-hexane–chloroform–glacial acetic acid (4 + 1 + 1, v/v), Rf (I) = 0.87, Rf (C) = 0.98, and Rf (M) = 0.51; propanol–25% ammonia (8 + 2, v/v), Rf (I) = 0.69, Rf (C) = 0.57, and Rf (M) = 0.56; and ethyl acetate–methanol–25% ammonia (17 + 2 + 1, v/v), Rf (I) = 0.17, Rf (C) = 0.16, and Rf (M) = 0.13, which gave no satisfactory development of indomethacin and its impurities with results similar to those obtained with the mobile phases described in ref. 2. Consequently, 2-propanol–25% ammonia–water (8 + 1 + 1, v/v) was chosen as the mobile phase and used for further investigation. Chromatograms were developed for ca 2.5 h. Absorption spectra were recorded by densitometry to identify chromatogram spots and to determine peak areas and Rf values for appropriate standard and sample solutions. In quantitative analyses, the recorded absorption spectra for all constituents with their characteristic maxima at λ (I) = 268 nm, λ (C) = 236 nm, and λ (M) = 292 nm were analyzed. Two analytical wavelengths (λ1 = 284 nm, λ2 = 230 nm), where the analytes have absorbance of >70%, were chosen, so that the simultaneous determination of both impurities was possible. The results are presented in Figures 2 and 3 and Table 1.

KRZEK & STAREK: JOURNAL OF AOAC INTERNATIONAL VOL. 84, NO. 6, 2001 1705 Table 1. Validation parameters and results obtained for the densitometric method Method parameter

Indomethacin

4-Chlorobenzoic acid

5-Methoxy-2-methyl-3-indoleacetic acid

λ, nm

284

230

230

Rf

0.71

0.65

0.57

20

20

100

100.69

90.09

91.17

8–48

4–40

12–24

a = 0.07

a = 0.10

a = 0.06

b = 0.16

b = 0.10

b = –0.48

Quantitation and detection limits, ng a

Recovery, %

Linearity range, µg/mL b

Regression coefficients P = a c + b c

Precision

y = × ± 2 SD a b c

xmin = 0.0542

xmin = 0.9000

xmin = 3.9500

xmax = 0.0630

xmax = 1.0000

xmax = 4.2000

y = 0.0592 ± 0.0071

y = 0.9417 ± 0.7237

y = 4.1000 ± 0.2000

n = 9. P = peak area; c = concentration; a and b = regression coefficients. xmin = lowest result; xmax = highest result; x = arithmetic mean; SD = standard deviation; y = x ± 2 SD (95% confidence level).

Limits of detection and quantitation.—The limits of detection and quantitation (Table 1) were considered together, because well-developed peaks at low noise level were obtained by densitometry under the specified chromatographic conditions. The signals of samples to which an appropriate amount of analyte was added were ≥4 times greater than that of the unspiked sample. Linearity.—The linearity (Table 1) was examined by using 6 solutions of different concentrations, ranging from 50 to 150% of expected results. The results were analyzed by the linear regression method and are presented in Table 1 and Figure 4. Accuracy.—The accuracy was expressed as percent recovery of an analyte added to relevant substances, in amounts of 80 to 120% of the levels in the preparations under examination. Nine determinations for each of 3 concentration ranges were performed (Table 1). Precision.—The consistency of the results was checked for 3 concentrations, with 8 results for each analyte. The tests were performed for model solutions obtained by dissolving the standards in methanol. The precision of the method was estimated by calculating the standard deviation (SD) while assuming that the individual determined values (×) should lie within the range y = × ± 2 SD (95% confidence level). As a result of these experiments, the parameters for the determination of indomethacin and its impurities in pharmaceutical preparations were established.

peaks were recorded by densitometry at λ = 284 nm for indomethacin and at λ = 230 nm for the impurities. The concentrations of the analytes in the pharmaceutical preparations under investigation were calculated from comparisons of peak areas for appropriate solutions of standards and pharmaceutical preparations by using computer software (Microsoft Office 97, Statistica 5.1 edition ‘97). Three mea-

Quantitative Analysis Standard solutions of indomethacin (5 µL), 4-chlorobenzoic acid (5 µL), and 5-methoxy-2-methyl-3-indoleacetic acid (10 µL) and 10 µL aliquots of relevant solutions of the pharmaceutical preparation under investigation were applied to the 7 × 12 cm TLC plates in the form of a 1 cm wide band. Chromatograms were developed to a height of 10 cm by using the mobile phase 2-propanol–25% ammonia–water (8 + 1 + 1, v/v), and the plates were dried at room temperature for 15 min. Absorption spectra and

Figure 4. Standard curve for indomethacin (Ž), 4-chlorobenzoic acid (), and 5-methoxy-2methyl3-indoleacetic acid (—).

1706 KRZEK & STAREK: JOURNAL OF AOAC INTERNATIONAL VOL. 84, NO. 6, 2001 Table 2.

Resultsa for the determination of indomethacinb and impuritiesc in pharmaceutical preparationsd Metindol injections No. 21093e

Parameter

Metindol tablets

No. 20993f

No. 30299g

No. 01LM0899h

No. 01LM0299i

Indomethacin Mean, %

0.0587

0.0692

0.0609

0.0243

0.0229

SD

0.0045

0.0013

0.0036

0.0018

0.0026

SDxk

0.0016

0.0005

0.0013

0.0007

0.0009

0.0587 ± 0.0038

0.0692 ± 0.0012

0.0609 ± 0.0030

0.0243 ± 0.0015

0.0229 ± 0.0022

j

95% confidence interval

4-Chlorobenzoic acid Mean, %

4.1063

2.8313

SD

0.1700

0.6210

0.0601

0.2196

4.1060 ± 0.1421

2.8313 ± 0.5193

SDx 95% confidence interval

5-Methoxy-2-methyl-3-indoleacetic acid Mean, %

0.9500

1.1313

SD

0.0756

0.1999

SDx

0.0267

0.0707

0.9500 ± 0.0632

1.1313 ± 0.1671

95% confidence interval a b c d e f g h i j k

n = 8. Declared indomethacin content: Metindol injections, 60 mg/2 mL; Metindol tablets, 25 mg/tablet. There are no declared content data for impurities; results for impurities are given as percentages of the indomethacin content. Lot numbers are given for the Metindol injections and tablets. Expiration date: 10/15/96. Expiration date: 9/15/96. Expiration date: 2/02. Expiration date: 8/02. Expiration date: 3/02. SD = standard deviation for individual points. SDx = standard deviation for arithmetic mean.

surements were made for each determination. The final results represent the mean values (Table 2). Results and Discussion The aim of this paper was to establish a simple and accurate method of quantitative and qualitative analysis for the identification and determination of the active substance and impurities in indomethacin preparations. It was found that individual constituents can be identified on the basis of absorption spectra recorded directly from chromatograms that differ from each other (Figure 2) and Rf values from densitograms of standards and samples examined (Table 1). The mobile phase that was established experimentally provides well-developed peaks for identification of the individual constituents, even for relatively small differences in Rf values, because the spots are permanent and compact (Figure 3). This mobile phase is the best among all that were used in experiments and among all described in ref. 2.

Although the developing time for the chromatogram is not very short (2.5 h), the method can be considered as relatively fast, because identification, quantitative analysis, and purity assessment can be performed at the same time. Furthermore, because it is possible to analyze several samples simultaneously, the total time for these analyses is shorter than with other methods. Under established conditions, the method leads to average recoveries of 100.69, 90.09, and 91.17% for indomethacin and its impurities, i.e., 4-chlorobenzoic acid and 5-methoxy-2-methyl3-indoleacetic acid, respectively (Table 1). Unfortunately, it was not possible to obtain higher recoveries of the impurities by using different wavelengths for the densitometric measurements. The recoveries in Table 1 are high enough when the amounts of the impurities are considered, ranging from 1 to 5 mg, in a sample containing 60 mg indomethacin. The mobile phase used in the analysis probably did not have any influence on the results obtained because all constituents were well separated and the recovery of indomethacin under these conditions was 100.69%. Lower recoveries of impurities could be associated with the preparation of solutions for analysis or with the determination of low concentra-

KRZEK & STAREK: JOURNAL OF AOAC INTERNATIONAL VOL. 84, NO. 6, 2001 1707

tions. However, the results obtained are within the tolerance limits of 85 to 110% of the determined concentrations (3, 21). The developed method is sufficiently sensitive. The limits of detection and quantitation are both sufficiently low for the determination of low concentrations such as those determined by the LC method recommended by the U.S. Pharmacopeia (3). Linear standard curves [P = f(c)] were obtained for the concentration range under investigation (Figure 4). Regression coefficients defining the linear standard curves, presented in Table 1, show a small value of the “b” coefficient. The accuracy of the method shows that repeatable results can be obtained. The scatter of results (Table 1) for all determined constituents at 95% probability is lower than twice the SD. The reliability and suitability of the method were confirmed by the results presented in Table 2, in particular those for indomethacin; its mean concentration is consistent with the values specified by the manufacturer. In the case of pharmaceutical preparations, the medicines in the form of injections and tablets commonly used for treatment were analyzed. The preparations, mainly injections, came from different batches manufactured at different times. Our results suggest that 4-chlorobenzoic acid and 5-methoxy-2-methyl-3-indoleacetic acid are present as impurities in outdated products only, whereas no traces of such substances are found in tablets and injections that have not passed the expiration date given by the manufacturer. Thus, these impurities can indicate a slow decomposition of indomethacin during storage. This process probably is more rapid in a wet environment. It appears that the method presented in this paper is suitable for purity assessment related to the determination of active substances and impurities. Conclusions A relatively fast, simple, and accurate method has been developed for the determination of indomethacin and its impurities, i.e., 4-chlorobenzoic acid and 5-methoxy-2-methyl-3-indoleacetic acid. The method allows identification, quantitative analysis, and purity assessment to be performed simultaneously both for active substances and for impurities and appears to be suitable for indomethacin quality control.

References (1) Zejc, A., & Gorczyca, M. (1998) Chemia Leków, PZWL, Warsaw, Poland (in Polish) (2) Krasowska, H., Krówczy½ski, L., & Bogdanik, Z. (1973) Pol. J. Pharmacol. Pharm. (in Polish) 25, 417–421 (3) U.S. Pharmacopeia (2000) 24th Ed., U.S. Pharmacopeial Convention, Inc., Rockville, MD (4) Kwong, E., Pillai, G.K., & McErlane, K.M. (1982) J. Pharm. Sci. 71, 828–830 (5) Plakogiannis, F.M., Ali, A., & Kazmi, S. (1981) Drug Dev. Ind. Pharm. 7, 215–221 (6) Vajda, M., Otta-Horvath, K., & Toth-Forgacs, E. (1994) Acta Pharm. Hung. 64, 131–134 (7) Flores-Picazo, Y., Perez-Krizar, S., & Castaneda-Hernandez, G. (1998) Rev. Mex. Cienc. Farm. 29, 22–25 (8) Vree, T.B., Van-Den-Biggelaar-Martea, M., & Verwey Van Wissen, C.P. (1993) J. Chromatogr. B. 127, 271–282 (9) Caturla, M.C., & Cusido, E. (1992) J. Chromatogr. B. 119, 101–107 (10) Sinsheimer, J.E., & Breault, G.O. (1971) J. Pharm. Sci. 60, 255–257 (11) Carlucci, G., Mazzeo, P., Quaglia, M.G., & Vetuschi C. (1989) Int. J. Pharm. 49, 79–82 (12) Sastry, C.S., Mangala, D.S., & Rao, K.E. (1986) Analyst 111, 323–325 (13) Poctova, M., & Kaka“, B. (1979) Cesk. Farm. 28, 344–346 (14) Larsen, C. (1989) Int. J. Pharm. 52, 55–61 (15) Ali, A.M. (1998) J. Pharm. Biomed. Anal. 18, 1005–1020 (16) Krzek, J. (1998) Acta Pol. Pharm. 55, 99–103 (17) Krzek, J., & Starek, M. (1999) J. Planar Chromatogr. 12, 356–360 (18) Krzek, J., & Kwiecie½, A. (1999) J. Pharm. Biomed. Anal. 21, 451–457 (19) Krzek, J., & Maœlanka, A. (2000) Acta Pol. Pharm. 57, 23–26 (20) World Health Organization (WHO) (1989) The Validation of Analytical Procedures Used in the Examination of Pharmaceutical Materials, 85.541/Rev. 2, Geneva, Switzerland (21) Farmakopea Polska (1996) P.T. Farm., Warszawa, Poland