ANALYST, JANUARY 1986, VOL. 111 45 Spectrophotometric Determination of Some Anti-inflammatory Agents Using A/-Bromosuccinimide Sonia T. Hassib, Hany M. Safwat and Ramzeia 1. El-Bagry Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Cairo University, Cairo, Egypt Four non-steroidal anti-inflammatory agents have been determined titrimetrically using N- b ro m 0s u cc i n i m id e . F I u f e n a m i c a n d m ef e n a m i c acids react q u a n t i t a t ive I y with N- b ro m o s u cc i n i m id e i n a n acidic medium whereas allopurinol and indomethacin can be determined with reproducible results in an aqueous pyridine solution. The method has been successfully applied to the determination of flufenamic acid, mefenamic acid and allopurinol in their pharmaceutical formulations.In addition, a sensitive spectropho- tometric method, based on the same reaction, has been used to determine selectively indomethacin in admixture with its hydrolysis products, and for the determination of indomethacin in capsules. Keywords: Anti-inflammatory agents determination; N -bromosuccinimide; spectrophotometry; titrimetry; pharmaceutical formulations N-Phenylanthranilic acid derivatives, namely flufenamic acid (I) and mefenamic acid (11), have been determined in blood and urine using diverse chromatographic techniques. 1-4 The native fluorescence shown by these anti-inflammatory agents I II CHZCOOH OH I @CO IV H c‘r Ill in organic solvents has been used for their detection and determination .5,6 Cyclisation reactions of N-phenylanthranilic acid derivatives using formaldehyde or sulphuric acid to give benzoxazines or a mixture of acridones, respectively, have been found to be useful for sensitive fluorimetric determina- ti0ns.~.8 A spectrophotometric method for the determination of flufenamic acid, based on ion-pair formation, has also been adopted .9 Indomethacin (111) has been determined by a number of techniques: spectrophotometric,l0-’2 titrimetric,l3J4 polaro- graphic15 and chromatographic.16>17 Allopurinol (IV), which is used in the treatment of hyperuricaemia associated with gout, has been determined by quantitative chromatographic methods,l”2* in addition to electrochemical oxidation, used for its determination in the presence of uric acid.23 The British Pharmacopoeia describes a non-aqueous titra- tion for the determination of allopurinol,24 and a spectro- photometric assay following ion-exchange separation has been reported for its determination in the presence of its decompo- sition products.25 In this paper, an accurate titrimetric determination of the four mentioned anti-inflamatory agents using N-bromosucci- nimide is presented.The reaction conditions have been optimised and the stoicheiometries of the reactions ascer- tained. The method can be applied to the analysis of their pharmaceutical preparations, except indomethacin capsules, for which the titrimetric procedure cannot differentiate between indomethacin and its alkaline hydrolysis products and in this instance a spectrophotometric assay with N-bromo- succinimide was used. This method proved successful for determining indomethacin in capsules and in samples pre- pared by mixing intact and hydrolysed indomethacin.Experimental Materials and Reagents All chemicals and reagents used were of analytical-reagent or pharmaceutical grade and distilled water was used through- out. Solvents were of spectroscopic grade. Flufenamic and mefenamic acids. Flufenamic acid, Arlef capsules of Lot No. 5000, nominally containing 100 mg of flufenamic acid, mefenamic acid and Ponstan capsules of Lot No. 82545, nominally containing 250 mg of mefenamic acid, were supplied by the Nile Co. for Pharmaceutical Industries, Cairo (under licence from Parke-Davis, Detroit, MI). Zndomethacin BP. Indomethacin BP was supplied by John Bell & Croyden, London. Indocid capsules of Lot No.822667, nominally containing 25 mg of indomethacin, were supplied by Kahira Pharmaceuticals & Chemical Industries Co., Cairo (under licence from Merck Sharp & Dohme, Rahway, NJ). Allopurinol BP. Allopurinol BP and Zyloric tablets of Lot No. 14099, nominally containing 100 mg of allopurinol, were supplied by Burroughs Wellcome & Co., London. Glacial acetic acid. N- Bromosuccinimide solutions, 0.05 and 0.005 M . These were prepared by dissolving 4.45 and 445 mg, respectively, of N-bromosuccinimide (BDH Chemicals, Poole) in 500 ml of water. Sodium thiosulphate solution, 0.01 N. This solution was standardised by the dichromate method. Potassium iodide solution, 10% mlV. Apparatus A Shimadzu Graphtcord UV 240 UV - visible spectropho- tometer was used for recording the UV spectra.46 ANALYST, JANUARY 1986, VOL.111 Methods Determintion of flufenamic acid Flufenamic acid (50 mg) was placed in a 250-ml calibrated flask and dissolved in 150 ml of acetic acid, then the solution was diluted to volume with water. A volume containing 0.92-4 mg of flufenamic acid was transferred into a stoppered conical flask. Acetic acid was added, if necessary, in such amounts that the final volume before addition of N-bromosuccinimide (NBS) was 10 ml. The solution was allowed to react with 40 ml of 0.005 M NBS solution for 15 min in the dark (temperature = 25 k 2 "C). Potassium iodide solution (10 ml) was added, and the solution was titrated with 0.01 N sodium thiosulphate solution to the starch end-point. A blank determination was carried out.For the determination of flufenamic acid in capsules, the ,contents of 20 Arlef capsules were mixed thoroughly and an accurately weighed portion of the mixed powder , nominally containing 25 mg of flufenamic acid, was extracted with diethyl ether (3 X 10 ml). The residue remaining after the evaporation of diethyl ether was dissolved in acetic acid and treated as described for the determination of flufenamic acid. Determination of mefenamic acid A solution of mefenamic acid in acetic acid (0.02% mlV) was prepared. A volume of this solution containing 1.5-3.5 mg of mefenamic acid was allowed to react with 40 ml of 0.005 M NBS solution for 10 rnin in the dark (temperature = 25 k 2 "C). The procedure was followed as described for the determina- tion of flufenamic acid.For the determination of mefenamic acid in capsules the contents of 20 Ponstan capsules were mixed thoroughly and an accurately weighed portion of the mixed powder, nominally containing 25 mg of mefenamic acid, was extracted with diethyl ether (3 x 10 ml). The residue remaining after evaporation of diethyl ether was dissolved in acetic acid and the procedure was continued as described for the determination of mefenamic acid. Determination of indomethacin ( a ) Titrimetric procedure. Indomethacin (100 mg) was placed in a 250-ml calibrated flask and dissolved in 10 ml of pyridine. The solution was diluted to volume with water. An aliquot containing 5-10 mg of indomethacin was allowed to react with 40 ml of 0.005 M NBS solution for 45 rnin in the dark (temperature = 25 k 2 "C).Acetic acid (10 ml) followed by )potassium iodide solution (10 ml) were then added. The solution was titrated with 0.01 N sodium thiosulphate solution to the starch end-point. A blank determination was carried out. For the determination of indomethacin in capsules, the contents of 20 Indocid capsules were mixed thoroughly and an accurately weighed portion of the powdered capsules, nomi- nally containing 50 mg of indomethacin, was extracted with chloroform (3 x 10 ml and 1 x 5 ml). The residue left after the chloroform had been evaporated was dissolved in 10 ml of pyridine and the procedure was continued as described for the determination of indomethacin. (6) Spectrophotometric procedure. Indomethacin (25-50 mg) was dissolved in a mixture of pyridine and water (5 ml of 1 + 4 V/V); next, 40 ml of 0.05 M NBS solution were added and the reaction was allowed to proceed in the dark for 45 rnin (temperature = 25 k 2 "C).The precipitate formed was filtered and washed with water (5 x 10 ml). After decanting the water the precipitate was dissolved in hot ethanol (50 ml), cooled and transferred into a 100-ml calibrated flask. It was then diluted to the mark with ethanol. An aliquot equivalent to 0.02-0.016 mg of indomethacin was placed in a 10-ml calibrated flask and completed to volume with ethanol. The absorbance of the solution was measured at 227 nm. The indomethacin content was determined from a calibration graph prepared according to the procedure described above. For the determination of indomethacin in Indocid capsules, the contents of 20 capsules were mixed thoroughly and an accurately weighed portion of the mixed powder, nominally containing 25 mg of indomethacin, was extracted with chloroform (3 X 10 ml and 1 x 5 ml).The residue remaining after evaporation of the chloroform was dissolved in 5 ml of a mixture of pyridine and water (1 + 4) and the solution was allowed to react with 40 ml of 0.05 M NBS solution for 45 min in the dark (temperature = 25 k 2 "C). The procedure was continued as described for the spectrophotometric determina- tion of indomethacin. Determination of allopurinol Allopurinol (100 mg) was placed in a 250-ml calibrated flask. Pyridine (10 ml) was added followed by the dropwise addition of water to complete solubility.The solution was then diluted to volume with water. An aliquot containing 2-8 mg of allopurinol was reacted with 40 ml of 0.005 M NBS solution for 60 min (temperature = 25 k 2 "C). Acetic acid (10 ml) and potassium iodide solution (10 ml) were added and the solution was titrated with 0.01 N sodium thiosulphate solution to the starch end-point. A blank determination was also carried out. For the determination of allopurinol in tablets, 20 Zyloric tablets were mixed thoroughly and an accurately weighed portion of the powdered tablets, containing 50 mg of allopurinol, was extracted with a mixture of pyridine and water (1 + 4) (5 x 10 ml). The extract was diluted with water to 250 ml in a calibrated flask and treated as described for the determination of allopurinol. Results and Discussion An accurate titrimetric method for the determination of flufenamic acid, mefenamic acid and allopurinol using NBS has been developed. A study of the optimum conditions has been carried out and the stoicheiometry of the reaction of each drug with NBS has been ascertained.The conditions opti- mised were the choice of medium (for a quantitative reaction), the time required for the reaction to be completed and the amount of reagent. Under optimum conditions each of the N-phenylanthranilic acid derivatives studied was found to consume 4 mol of NBS per mole of acid in the presence of acetic acid. For flufenamic acid constant results were obtained within 10-30 min. Mefenamic acid consumed 4 mol of NBS per mole within 10-15 rnin but during the following 30 rnin a slower consumption of NBS was recorded (Fig.1). The amount of reagent used for both acids was at least three times more than the amount of NBS consumed in the reaction. The method was successfully applied to the determination of flufenamic and mefenamic acids in Arlef and Ponstan cap- sules, respectively. Both acids were extracted from their capsules using diethyl ether, which eliminates any interference from the capsules' contents other than the active ingredient. 8 1 I Timeimi n Fig. 1. and (B) mefenamic acid using NBS Effect of time on the determination of (A) flufenamic acidANALYST, JANUARY 1986, VOL. 111 47 Table 1. Titrimetric determination of anti-inflammatory agents Method of Anti-inflammatory agent determination* Flufenamic acid .. . . . . . . . . 1 Flufenamic acid . . . . . . . . . . 2 Added authentic . . . . . . . . 1 Flufenamic acid . . . . . . . . . . 4 Mefenamic acid . . . . . . . . . . 1 Added authentic . . . . . . . . 1 Mefenamic acid . . . . . . . . . . 4 Indome t hacin . . . . . . . . . . 3 Added authentic . . . . . . . . 3 Indomethacin . . . . . . . . . . 4 Arlef capsules . . . . . . . . . . 1 Ponstan capsules . . . . . . . . 1 Indocid capsules . . . , . . . . . . 3 Allopurinol . . . . . . . . . . 3 Zyloric tablets . . . . . . . . . . 3 Allopurinol . . . . . . . . . . 4 Added authentic . . . . . . . . 3 No. of experiments 9 9 10 10 5 10 9 9 5 10 14 14 4 12 8 8 6 Mass usedlmg 0.92-4.00 1.00-4.00 1.00-2.40 0.60-2.40 1 SO-3 S O 3 1.25-1 87.5 1.25-2.00 1.25-2.00 30.00-150.0 5.00-10.00 3.00-5 .OO 3.00-5.00 150.00-240 .OO 2.00-8.00 1.00-4.00 0.80-3.20 20.00-120.00 Recovery, 70 97.34-101.78 97.18-103.03 100.09-1 05.36 96.58-102.40 98.70-99.58 97.39-101.81 96.38-100.40 96.38-99.40 97.33-101.43 96.28-101.40 97.10-101.27 94.55-98.06 96.63-100.37 99.17-100.99 97.72-101.40 97.85-100.65 96.3 1-101.71 Mean recovery 99.62 f 1.18 100.40 f 2.14 103.54 f 1.38 99.09 f 1.23 99.20 f 0.39 99.77 f 0.86 97.92 f 0.90 98.25 f 0.74 100.00 f 0.80 92.12 f 1.00 100.09 _+ 0.71 99.39 f 0.95 99.76 f 2.07 99.10 f 0.81 96.99 f 0.99 99.85 f 1.44 98.54 f 1.31 ( P = 0.05), % Moles of re agent consumed per mole of anti- inflammatory agent 4 2 4 1 4 4 1 2 2 * Methods: 1, N-bromosuccinimide in an acid medium; 2, N-bromosuccinimide in hydrogen carbonate solution; 3, N-bromosuccinimide in pyridine solution; and 4, BP method.24 15 30 45 60 75 90 Tirnelmin Effect of time on the determination of (A) indomethacin and Fi .2.(B? allopurinol using NBS The results obtained for the determination of these two acids in the pure form and in capsules are summarised in Table 1, which also shows the results obtained by applying the BP method. The NBS method was found to be more sensitive than the BP method. It is noteworthy that 1 mol of flufenamic acid consumed 2 mol of NBS in the presence of potassium hydrogen carbonate; however, this determination was not as accurate as when carried out in the presence of acetic acid. A quantitative determination of allopurinol was achieved when the reaction with NBS was conducted in an aqueous solution containing a small amount of pyridine. The time required for a constant consumption of NBS was found to be within 45-90 min (Fig.2). Under optimum conditions 1 mol of allopurinol consumed 1 mol of NBS. The solubility of allopurinol in water was enhanced by pyridine, although it is, itself, almost insoluble in water. Allopurinol was extracted from Zyloric tablets using a 20% V/V aqueous solution of pyridine. (An aqueous solution was used because of the poor solubility of allopurinol in organic solvents.) TLC of the extract using methanol - chloroform (1 + 9) revealed that only allopurinol was extracted from the tablets when a 20% V/V aqueous solution of pyridine was used, The results obtained for the determination of allopurinol and Zyloric tablets are given in Table 1 and show good agreement with those obtained by the BP method.The reaction of NBS with indomethacin was studied by a titrimetric determination using a pure sample of indo- methacin. The time required for a stable consumption of NBS was between 45 and 60 min where 2 mol of NBS were Table 2. Spectrophotometric determination of indomethacin (25-50 mg) in the presence of its alkaline hydrolysis products (6 mg) using N-bromosuccinimide Concentration of indomethacin used/ mg per 10 ml 0.040 0.048 0.072 0.080 0.096 0.120 0.140 Amount of indomethacin Absorbance recoveredl Recovery, 0.273 0.0405 101.25 0.307 0.0470 97.92 0.442 0.0710 98.61 0.500 0.0805 100.63 0.578 0.0950 98.96 0.703 0.1170 97.50 0.827 0.138 98.57 Mean 99.06 Standard deviation 1.38 Standard error 0.52 Confidence limit ( P 0.05) f1.27 at 227 nm mg per 10 ml Y O consumed per mole of indomethacin.The following 15 min showed an increase in consumption to over 2 mol (Fig. 2). Good recoveries were obtained with amounts varying from 2 to 10 mg (Table 1). The titrimetric determination of indomethacin with NBS could not be applied to the determination of indomethacin when this was in admixture with its hydrolysis products, because the latter have been found to consume at least 9 mol of NBS per mole of hydrolysed indomethacin. This consump- tion is thought to be due to the indole component as, in our experience, p-chlorobenzoic acid does not consume any NBS under the applied conditions. However, it was observed that indomethacin reacted with NBS to give a pale yellow precipitate in contrast to its hydrolysis products, which did not give any precipitate under the same conditions.Making use of this behaviour led to a method that proved capable of determining the content of the intact drug in samples prepared by mixing intact and hydro- lysed indomethacin. TLC of the precipitate (after filtration and washing with water) showed an absence of indomethacin, and proved that the precipitate is a single component with an RF of 0.71 in chloroform.48 ANALYST, JANUARY 1986, VOL. 111 Table 3. Spectrophotometric determination of indomethacin in capsules (Indocid) using N-bromosuccinimide Amount of Amount of Amount of as labelled added to found in indomethacin in capsules/ capsules/ capsules/ recovered/ Recovery mg per 10 ml mg per 10 ml mg per 10 ml mg per 10 ml in capsules, YO indomethacin indomethacin indomethacin Amount of 0.020 0.030 0.040 0.050 0.060 0.032 0.048 0.064 0.080 0.020 0.030 0.040 0.050 0.060 0.016 0.024 0.032 0.040 0.0200 0.0195 0.0300 0,0300 0.3800 0.0400 0.0485 0.0500 0.0585 0.0585 0.0310 0.0160 0.0475 0.0235 0.0620 0.0320 0.0780 0.0395 Mean Standard deviation Standard error Confidence limit (P = 0.05) 100.00 100.00 95.00 97.00 97.50 96.88 98.96 96.88 97.50 97.75 1.63 0.54 1.26 Recovery of added authentic, % 97.50 100.00 100.00 100.00 97.50 100.00 97.92 100.00 98.75 99.07 1.16 0.39 0.89 Its absorption spectrum exhibited a maximum at 227 nm and a plateau around 367 nm.The absence of indomethacin in such a product justified measuring the absorbance at 227 nm. The calibration graph used for the spectrophotometric determination of indomethacin showed that Beer’s law is obeyed in the range 0.02-0.16 mg per 10 ml.A mean recovery of 99.06 k 1.27% was obtained in the determination of indomethacin in the presence of its hydroly- sis products (Table 2). The analysis of Indocid capsules gave recoveries of 97.75 k 1.26% for indomethacin in capsules and 99.07 k 0.89% when a standard additions method was applied (Table 3). Chloroform used in the extraction of Indocid capsules did not extract any excipients from the capsules, but extracted the hydrolysis products of indomethacin that were present. The BP method24 for the capsules showed negligible absorbance at 318 nm (the prescribed h) when applied to indomethacin hydrolysis products. However, the proposed spectrophotometric method is more selective.It is also more sensitive, as shown by the value, which was 595 in this method compared with 180 in the BP method. References 1. Dusci, L. J., and Hackett, L. P., J. Chromatogr., 1978, 161, 340. 2. Cotellessa, L., Riva, R., Salva, P., Marcucci, F., and Mussini, E . , J. Chromatogr., 1980, 192, 441. 3. Demetriou, B., and Osborne, B. G., J. Chromatogr., 1974,90, 405. 4. Bland, S. A., Blake, J. W., and Ray, R. S . , J. Chromatogr. Sci., 1976, 14, 201. 5. Hattori, Y., Arai, T., Mori, T., and Fujihira, E., Chem. Pharm. Bull., 1970, 18, 1063. 6. Mehta, A. C., and Schulman, S . G., Talanta, 1973, 20, 702. 7. Dell, H. D., and Kamp, R., Arch. Pharm., 1970, 303, 785. 8. 9. Schmollack, W., and Wenzel, U., Pharmazie, 1974, 29, 583. Beltagy, Y . A., Zentralbl. Pharm., Pharmakother. Labora- toriumsdiagn., 1977, 116, 925; Chem. Abstr., 1978, 88, 158545m. 10. Van der Meer, M. J., and Hundt, H. K. L., J . Chromatogr., 1980, 181,282. 11. Peterkova, M., Kakac, B., and Matousova, O., Cesk. Farm., 1980, 29, 73. 12. Krasowska, H., Krowczynski, L., and Bogdanik, Z . , Pol. J. Pharmacol. Pharm., 1973, 25, 417; Chem. Abstr., 1974, 80, 115885m. 13. “British Pharmacopeia 1980,” HM Stationery Office, London, 1980, p. 239. 14. Tajne, M. R., Kasture, A. V., and Wadodkar, S . G., Indian J. Pharm. Sci., 1978, 40, 196. 15. Nino, N., and Boneva, A., Zzv. Durzh. Znst. Kontrol Lek. Sredstva, 1979, 12, 18; Chem. Abstr., 1980, 92, 47263~. 16. Plazonnet, B., and Vandenheuvel, W. J. A., J. Chromatogr., 1977, 142,587. 17. Terweij-Groen, C . P., Heemstra, S., and Kraak, J. C., J. Chromatogr., 1980, 181, 385. 18. Endele, R., and Lettenbauer, G., J . Chromatogr., 1975, 115, 228. 19. Brown, M., and Bye, A . , J. Chrornatogr., 1977, 143, 195. 20. Kramer, W. G., and Feldman, S . , J . Chromatogr., 1979, 162, 94. 21. Putterman, G. J., Shaikh, B., Hallmark, M. R., Sawyer, C . G., Hixson, C. V., and Perini, F., Anal. Biochem., 1979, 98, 18. 22. Chang, S. L., and Kramer, W. G., J. Chromatogr., 1980, 181, 286. 23. Dryhurst, G., and De, P. K., Anal. Chim. Acta, 1972,58, 183. 24. “British Pharmacopeia 1980,” HM Stationery Office, London, 1980, p. 19. 25. Gressel, P. D., and Gallelli, J. F., J. Pharm. Sci., 1968,57,335. Paper A4/403 Received November 15th, 1984 Accepted July Ist, 1985