TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005 1173

DRUGS, COSMETICS, FORENSIC SCIENCES

Simultaneous Determination of N-Butylscopolamine and Oxazepam in Pharmaceutical Formulations by First-Order Digital Derivative Spectrophotometry

MARIA INÉS TORAL,MARCELO A. MUÑOZ,andSANDRA L. ORELLANA University of Chile, Faculty of Sciences, Department of Chemistry, PO Box 653, Santiago, Chile Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021

A simple method has been developed for the In the case of oxazepam, various analytical methods have simultaneous determination of N-butylscopolamine been developed, which include the use of liquid bromide and oxazepam in pharmaceutical chromatography (LC; 2, 3), LC with mass spectrometry formulations using first-order digital derivative (LC/MS; 4), gas chromatograpy/MS (GC/MS; 5, 6), GC/ion spectrophotometry. Acetonitrile was selected as trap tandem mass spectrometry (TMS; 7), micellar-LC (8), the solvent in which both compounds showed CE-electrospray mass spectrometry (EMS; 9), derivative UV well-defined bands. Both analytes showed good spectrophotometry (10), and other techniques. In contrast, stability in this solvent when solutions of the N-butylscopolamine has been determinated only by using LC analytes were exposed to light and temperatures (11). Nevertheless, no methodologies have been reported for between 20° and 80°C. The simultaneous the simultaneous determination of oxazepam and determination of both drugs was performed by the N-butylscopolamine. zero-crossing method at 226.0 and 257.0 nm for Because derivative spectrophotometry has been N-butylscopolamine and oxazepam, respectively. successfully used in the determination of mixtures of drugs in The linear range of determination was found to be pharmaceutical formulations (12–14), we developed a simple, –7 –5 2.5 ´ 10 to 8.0 ´ 10 mol/L for rapid, accurate, and inexpensive method using first-order –8 N-butylscopolamine and 7.1 ´ 10 to 8.0 ´ digital derivative spectrophotometry for the simultaneous –5 10 mol/L for oxazepam. A very good level of determination of oxazepam and N-butylscopolamine. Our repeatability (relative standard deviation) of 0.2% work included selection of the best solvent for the analysis, was observed for N-butylscopolamine and optimization of the chemical and spectral variables, and a oxazepam. The ingredients commonly found in stability study of the drugs when exposed to different pharmaceutical formulations do not interfere. The conditions of light and temperature. The proposed method proposed method was applied to the determination was successfully applied to simulated and commercial of these drugs in pharmaceutical formulations pharmaceutical formulations (capsules). (capsules). Experimental

-butylscopolamine bromide (I) is a quaternary form of Instrumentation the tropane alkaloid (-)-scopolamine. Scopolamine N A Shimadzu (Shimadzu Co., Kyoto, Japan) UV-1603 and its derivatives are anticholinergic drugs with spectrophotometer with 10 mm quartz cells was used for antispasmodic properties that are frequently used as measurement of the absorbance and derivative absorption endoscopic premedications to inhibit digestive tract spectra. For all the tested solutions, the first-derivative spectra motility (1). were recorded over a range of 190–400 nm versus solvent by Oxazepam (II) is part of a well-known group of using slit width values and sampling intervals of 2.0 and 0.2 compounds with the 1,4-benzodiazepine-2-one basic nm, respectively. A scan speed of 480 nm/min was also used. structure. Benzodiazepines are widely used to treat symptoms The spectral data were processed by Shimadzu kit version 3.7 of emotional stress symptoms like anxiety, insomnia, and (P/N 206-60570-04) software. All solid samples were irritability (1). In combination, N-butylscopolamine and weighed to within ± 0.01 mg by using a Sartorius R 200D oxazepam exhibit a tranquilizing antispasmodic effect balance. suitable for the treatment of abdominal and gynecological pathologies. Reagents

All reagents were analytical grade. Received November 3, 2004. Accepted by JM December 22, 2004. (a) N-butylscopolamine bromide (I).—99.9%. Purchased Corresponding author's e-mail: [email protected]. from Sigma (St. Louis, MO). 1174 TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005

the presence of I and II at 3.0 ´ 10–5 mol/L. In all cases, the corresponding absolute values of the first-derivative spectra at 226.0 nm for I and 257.0 nm for II were obtained and plotted versus the corresponding concentrations.

Procedure for Simultaneous Determination of I and II in Simulated Mixtures with Different Mass Ratios

Stock solutions of each compound were prepared in acetonitrile, and aliquots were appropriately diluted to obtain solutions with mass ratios of the analytes between 1:4 and 4:1. For each solution 5 replicates were made. Then, the first-derivative spectra were evaluated, and the concentration Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021 of each compound was calculated, allowing the determination Figure 1. Structures of N-butylscopolamine (I) and of the corresponding values for recovery and relative standard oxazepam (II). deviation (RSD).

Procedure for Stability Studies of I and II When (b) Oxazepam (II).—99.6%. Laboratorios Andrómaco Exposed to Changes in Light and Temperature (Santiago, Chile). (c) Stock solutions of I and II at 1.0 ´ 10–3 Stock solutions of a mixture of I and II with a concentration –5 mol/L.—Prepared by dissolving 22 and 14 mg compound, of 4.0 ´ 10 mol/L for each compound in acetonitrile were respectively, in acetonitrile and diluting to 50 mL. used in stability to changes in light and temperature. In the (d) Other ranges of concentrations.—Prepared by case of the light studies, the solutions were exposed to direct appropriate dilution with the same solvent. The light, indirect light, and darkness, and the first-derivative pharmaceutical product Novalona (Laboratorios Andrómaco) spectra were evaluated 9 times during a period of 24 h. For the containing both compounds was also dissolved in the same temperature studies, the solutions were kept in a water bath at solvent. Furthermore, to conduct a study of solvent effects on 20°,30°,40°,50°,60°,70°, and 80°C and the first-derivative spectral behavior, stock solutions of I and II at 1.0 ´ 10–3 spectra were evaluated after 10 min. mol/L were prepared by dissolving the same amounts described above in different solvents. Procedure for Simultaneous Determination of I and (e) Other concentration ranges.—Prepared by II in Pharmaceutical Formulations (Capsules) appropriate dilution with the respective solvent. The contents of various capsules of Novalona were Calibration Procedure for Determination of I and II in weighed and powdered. A quantity of 369 mg powder was Mixtures weighed, dissolved in acetonitrile, and diluted to 50 mL with Aliquots of the stock solutions of I and II were acetonitrile. The solution was shaken for 20 min and simultaneously diluted with acetonitrile over the centrifuged. A 200 mL aliquot of the supernatant was diluted concentration range 1.0 ´ 10–5 to 5.0 ´ 10–5 mol/L. The with acetonitrile to 10 mL, and the solution was evaluated by calibration procedure was performed for each compound in first-derivative spectrophotometry.

Figure 2. Zero-order spectra of (a) N-butylscopolamine and (b) oxazepam in (A) acetonitrile, (B) methanol, and (C) ethanol. TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005 1175 Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021

Figure 3. (a) First,- (b) second-, (c) third-, and (d) fourth-derivative spectra of (I) N-butylscopolamine and (II) oxazepam.

Results and Discussion complex systems, allowing simultaneous determination. The spectral variables were optimized to improve the analytical The structures of I and II are quite different (Figure 1). I has methodology and obtain the best analytical features. an a unsubstituted phenyl ring and an ester carbonyl and (a) Selection of derivative order.—The first-, second-, shows a strong absorption between 190 and 250 nm in the third-, and fourth-derivative spectra were obtained from the classical UV spectrum as expected. In contrast, II has a much zero-order spectra by using digital differentiation (Figure 3). more complicated p-system consisting of a chloro-substituted Figure 3, a and b, shows that the first and second derivatives diazepine ring and a phenyl ring, both conjugated with a could be used for the simultaneous determination of I and II, double bond in the main structure. This more complicated because in both cases the derivative presents characteristic system is evident in the classical UV spectrum, where 3 zones for each compound, which can be used for analytical different bands are found between 190 and 350 nm. purposes. When the derivative order increases, the sensitivity decreases. In this context, when the first derivative is used, the Solvent Effect on the Spectra simultaneous determination can be achieved easily, because To improve the quality of the analytical method, the the spectra present well-defined zones for determination of spectral behavior of both compounds at a concentration of 4.0 ´ 10–5 mol/L was studied in the following solvents: chloroform, 1-2-dichloroethane, methanol, ethanol, and acetonitrile. Chlorinated solvents were eliminated because of the high amount of overlap from analyte and solvent bands. On the other hand, well-defined bands were obtained for both compounds in ethanol, methanol, and acetonitrile (Figure 2). Acetonitrile was selected over the alcohols because the bands had better resolution, especially in the case of I. Also, acetonitrile demonstrated better analytical features like low volatility. Spectral Behavior and Selection of Spectral Variables The spectral behavior of I and II does not permit the simultaneous determination of both drugs by using classical Figure 4. First-derivative spectra of (I) spectrophotometry because of overlapping bands. Derivative N-butylscopolamine and (II) oxazepam with –5 –5 spectrophotometry has shown to be very helpful in resolving concentrations between 1.0 ´ 10 and 4.0 ´ 10 mol/L. 1176 TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005 Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021

Figure 5. Effect of changes in (a) light and (b) temperature on the spectra of (I) N-butylscopolamine and (II) oxazepam. each analyte, and the sensitivities are greater. Thus, the first (c) Selection of scale factor.—Although the scale factor derivative was selected. The second, third, and fourth does not improve the sensibility, because an increase in the derivatives were discarded because they do not present any analytical signal is accompanied by an increase in the blank 8 analytical advantages. response, this factor was studied between 1 and 10 . A scale (b) Selection of smoothing factor.—By using the first factor of 10 000 was selected because it allows the direct derivative, the smoothing factor was varied, and the following reading of the values for the analytical measurements. values were used: 2, 4, 8, and 16. These values are defined by (d) Selection of analytical wavelengths.—The analytical wavelengths were selected by using the zero-crossing default and are related to the wavelength range over which the approximation and the spectral parameters previously spectra are scanned. When the smoothing factor is increased, optimized. This selection is based on sensitivity and the S/N the heights of the derivative signal decrease but the noise ratio of the signals in the first-order derivative spectra decreases faster, and thus in this way the signal-to-noise (S/N) (Figure 3a). The first-derivative spectrum of I in acetonitrile ratio increases. A value of 16 (16 000 experimental points) was evaluated directly versus the solvent and showed a no was selected in order to have the best S/N ratio. On the other absorption zone from 254.0 nm through greater wavelengths. hand, the shapes of the spectral bands were not altered in the The point at 257.0 nm was selected for the quantitation of concentration range for which the simultaneous determination oxazepam, based on the high sensitivity and good S/N ratio of was proposed. This means that no distortion effects were the analytical signal. induced with this value of the smoothing factor. As expected by the more complicated p-system of II, its In this work, sensibility was sacrificed to ensure the first-derivative spectrum shows 4 principal zero-crossing accuracy and precision of the results. points at 213.0, 226.0, 311.0, and 361.0 nm. The wavelengths

Figure 6. First-derivative spectra of mixtures of N-butylscopolamine and oxazepam: (a) I at 3.0 ´ 10–5 mol/L and II between 1.0 ´ 10–5 and 5.0 ´ 10–5 mol/L; (b) II at 3.0 ´ 10–5 mol/L and I between 1.0 ´ 10–5 and 5.0 ´ 10–5 mol/L. TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005 1177

Table 1. Analytical features of the method

Analytical parameters N-Butylscopolamine Oxazepam

Detection limit, mol/L 8.3 ´ 10–8 2.4 ´ 10–8 Determination limit, mol/L 2.5 ´ 10–7 7.1 ´ 10–8 Determination range, mol/L 2.5 ´ 10–7–8.0 ´ 10–5 7.1 ´ 10–8–8.0 ´ 10–5 Repeatability (RSD, %) 2 1.6 a 6 6 Regression lines h1 =7.3´ 10 C+6.6 h2 =5.5´ 10 +0.9 Correlation coefficient 0.999 0.999 a Where h is in derivative units, and C is the analyte concentration in mol/L. Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021

of 311.0 and 361.0 nm were discarded because of the The repeatability, expressed as the RSD, was obtained by extremely weak analytical signal. The wavelength of using 9 standard samples containing each drug at 1 ´ 10–5,2´ 226.0 nm was selected over 213.0 nm because the analytical 10–5,and3´ 10–5 mol/L, and 3 replicates at each signal is highly sensitive, it is located in the center of the peak, concentration. The results are shown in Table 1. and it has better definition. To confirm that the wavelengths The determination range was defined as between the previously selected have all the features necessary, quantitation limit and the loss of linearity. In this context, it is first-derivative spectra of one analyte at different possible to obtain an acceptable degree of linearity degree concentrations were recorded (Figure 4). As expected, no with accuracy and precision in the results when the proposed changes in these wavelengths were observed with the method is applied. All analytical features are shown in variation in concentration. According to this analysis, the Table 1. simultaneous determinations of I and II can be performed at Application 226.0 and 257.0 nm, respectively. To establish the proportions at which one analyte can be Stability Studies measured accurately in the presence of the other, recovery values were recorded for mixed standard solutions with mass To establish the photostability of I and II in acetonitrile, a ratios between 1:4 and 4:1 (Table 2). The concentration of solution of both compounds was exposed to direct light, each compound can be determined if the concentration ratio is indirect light, and darkness. Also, the thermostability was between 1:4 and 4:1. According to the results, it is possible to evaluated by exposing the same standard solutions to conclude that this method has a wide range of application and ° ° temperatures between 20 and 80 C. In all cases, permits the simultaneous determination of both drugs in real photochemical and thermal degradations were not found pharmaceutical formulations. because the analytical signals were not altered (Figure 5). An interferents study using the most common excipients (magnesium stearate–gelatin, 3–5%, and lactose–starch, Analytical Features

The first-derivative spectra of one analyte at different concentrations in the presence of the other were recorded Table 2. Determinations of N-butylscopolamine (I) and (Figure 6). The results confirm that the wavelengths oxazepam (II) in mixtures with different mass ratios previously selected can be used for analytical purposes, a because when the concentration of one analyte remains Mass added, mg Mass found, mg (recovery, %) constant, the analytical signal at the selected wavelength also Ratio I:II IIII II remains constant. Additionally, when the concentration of one analyte is varied, the analytical signal increases. 1:4 17.6 4.4 17 ( 96.6) 4.2 (96.4) Calibration curves (95% confidence limit) of the 1:3 13.2 4.4 13 (98.5) 4.4 (100.1) first-derivative values obtained versus the respective analyte concentrations were prepared for both compounds. 1:2 8.8 4.4 8.7 (99.0) 4.5 (101.6) The analytical features were obtained according to the 1:1 4.4 4.4 4.3 (98.0) 4.5 (102.2) criteria of the International Conference on 2:1 4.4 8.8 4.5 (102.1) 9.2 (102.9) Harmonization (15) in order to calculate the detection limits 3:1 4.4 13.3 4.3 (96.5) 13.6 (102.2) and quantitation limits. The following definitions were 4:1 4.4 17.6 4.5 (102.2) 18 (102.2) adopted: 3.3 s/S and 10 s/S, respectively, where S is the slope of the calibration curve and sigma is the standard deviation a Each value is the average of 5 determinations; in all cases the corresponding to the response of 11 blanks. RSD was <2%. 1178 TORAL ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 88, NO. 4, 2005

95–97%) was also performed. Recoveries obtained were (4) Walles, M., Mullett, W.M., & Pawliszyn, J. (2004) J. 100.6 ± 1.7 and 97.4 ± 2.0% for I and II, respectively. These Chromatogr. A 1025, 85–92 results demonstrate that common excipients normally found (5) Borrey,D.,Meyer,E.,Lanbert,W.,VanPeteghem,C.,&De in pharmaceutical formulations do not interfere in the Leenheer, A.P. (2001) J. Chromatogr. B 765, 187–197 proposed method. (6) Reubsaet, K.J., Norli, H.R., Hemmersbach, P., & Rasmussen, Finally, the concentrations of I and II in Novalona were K.E. (1998) J. Pharm. Biomed. Anal. 18, 667–680 determined. This pharmaceutical formulation contains the (7) Pichini, S., Pacifici, R., Altieri, I., Palmeri, A., Pellegrini, M., following nominal amounts per capsule: 10 mg of I and 10 mg & Zuccaro, P. (1999) J. Chromatogr. B 732, 509–514 of II. With the proposed method, the average levels found per (8) Gil-Agusti, M., Carda-Broch, S., Garcia-Alvarez-Coque, capsule were 9.84 ± 0.2 mg for I and 9.86 ± 0.2 mg for II. M.C., & Esteve-Romero, J. (2000) J. Liq. Chromatogr. 23, 1387–1401 Acknowledgments (9) McClean, S., O'Kane, E.J., & Smyth, W.F. (2000) Electrophoresis 21, 1381–1389 Downloaded from https://academic.oup.com/jaoac/article/88/4/1173/5657450 by guest on 24 September 2021 We are grateful to the National Fund for Development of (10) Tan, J.Y., Jiang, Z.L., & Wu, Y.H. (1999) Chin.J.Anal. Sciences and Technology (FONDECYT), project 1020692 Chem. 27, 1317–1319 for financial support, and Laboratorios Andrómaco for help in (11) Huller, G., Barthel, W., Klatt, H., & Haustein, K.O. (1993) providing oxazepam. Marcelo MuZoz thanks the National Pharmazie 48, 548–549 Commission of Scientific and Technological Investigation (CONICYT) for his doctoral fellowship. (12) Toral,M.I.,Lara,N.,Richter,P.,Tassara,A.,Tapia,A.E.,& Rodríguez, C. (2001) J. AOAC Int. 84, 37–42 References (13) Toral, M.I., Pope, S., Quintanilla, S., & Richter, P. (2002) Int. J. Pharm. 249, 117–126 (1) Gringauz, A. (1997) Introduction to Medicinal Chemistry: (14) Toral, M.I., Tassara, A., Soto, C., & Richter, P. (2003) J. How Drugs Act and Why,Wiley-VCH,Inc.,NewYork,NY AOAC Int. 86, 241–245 (2) Bugey, A., & Staub, C. (2004) J. Pharm. Biomed. Anal. 35, (15) Validation of Analytical Procedures: Methodology (1997) 555–562 International Conference on Harmonization of Technical (3) Pistos, C., & Stewart, J.T. (2003) J. Pharm. Biomed. Anal. Requirements for Registration of Pharmaceuticals for Human 33, 1135–1142 Use, ICH Q2B, Geneva, Switzerland