A Simple High-Throughput Method for Determination of Antiepileptic Analogues of γ-Aminobutyric Acid in Pharmaceutical Dosage Forms Using Microplate Fluorescence Reader

Abstract

Pregabalin (PGB), gabapentin (GBP), and vigabatrin (VGB) are structural analogues of γ-aminobutyric acid used for the treatment of different forms of epilepsy. Their analytical determination is challenging since these molecules have no significant UV or visible absorption. Several derivatization methods have been developed and used for their determination in bulk or pharmaceutical dosage forms. We aimed to develop a high- throughput method using a microplate reader with fluorescence detection and simple derivatization with fluorescamine. Obtained method involves derivatization step of only 5 min at room temperature and simultaneous measurements of 96 samples (λ_ex 395, λ_em 476 nm) thus rendering excellent high-throughput analysis. The method was found to be linear with r²>0.998 across investigated analytical ranges of 0.75 to 30.0 µg/mL for PGB, 2.00 to 80.0 µg/mL for GBP, and 1.50 to 60.0 µg/mL for VGB. Intraday and interday precision values did not exceed 4.93%. The accuracy was ranging between 96.6 to 103.5%. The method was also found to be specific since used excipients did not interfere with the method. The robustness study showed that derivatization procedure is more robust than spectrofluorimetric conditions. The developed high-throughput method was successfully applied for determination of drug content and dissolution profiles in pharmaceutical dosage forms of studied antiepileptic drugs.

Pregabalin ((3S)-3-(aminomethyl)-5-methylhexanoic acid) (PGB), gabapentin (1-(aminomethyl)-cyclohexaneacetic acid) (GBP), and vigabatrin (4-amino-5-hexenoic acid) (VGB) are three relatively recent drugs used for the treatment of different forms of epilepsy (Fig. 1a). All three drugs are structural analogues of inhibitory neurotransmitter γ-aminobutyric acid (GABA), originally designed to target the GABAergic system.¹⁾

Fig. 1. (a) Chemical Structures of Pregabalin (PGB), Gabapentin (GBP), and Vigabatrin (VGB) and (b) Derivatization Scheme of Amines with Fluorescamine

The analytical method development for PGB, GBP and VGB determination is challenging since these molecules have no significant UV or visible absorption. Several derivatization methods have been proposed^2–19) and used for their determination in bulk or pharmaceutical dosage forms, plasma, serum, and urine. These methods employed derivatization reagents such as fluorescamine,^{2–4,17–19)} 7-chloro-4-nitrobenzofurazan,^5–7) 1-fluoro-2,4-dinitrobenzene,⁸⁾ o-phtaldialdehyde,^9,10) 2,4,6-trinitrobenzene sulfonic acid,^11,12) and 9-fluorenylmethylchloroformate.¹³⁾ The resulted derivatives were measured using non-separation spectrophotometric^5,6,15–17) and spectrofluorimetric methods.^{2,5–7,17,19)} Moreover, HPLC^3,8–14,18) and capillary electrophoresis (CE) methods⁴⁾ using precolumn derivatization with either UV or fluorescent detector were also used. Limited non-derivatization methods were developed for GBP determination in pharmaceutical dosage forms using HPLC with direct UV detection,^20,21) CE with indirect UV detection,²²⁾ cyclic voltammetry,²³⁾ and flow analyses method with chemiluminometric detection.²⁴⁾ Indirect UV detection using CE was also used for VGB determination in pharmaceutical dosage forms.²⁵⁾ Sophisticated LC MS/MS methods have also been used for quantification of PGB, GBP and VGB in plasma or serum.^26–28)

The analytical method for drug determination in formulation development or routine quality control purposes has to fulfil several criteria. Ideally it requires relative low sample volume and simple sample preparation procedure. In addition method should meet validation criteria in terms of precision, accuracy, robustness etc., and most of all it has to offer high-throughput. So far developed methods for determination of PGB,^3,5,17,19) GBP,^{2,7,8,15,16,18,20–24)} and VGB^2,6,7,18,25) use from a few µL to approximately 1 mL of sample. The sample preparation is the time-consuming step in derivatization methods and in case of PGB, GBP, and VGB derivatization can take from few seconds to 50 min. Non-separation methods employing derivatization with 7-chloro-4-nitrobenzofurazan require also extraction of the drug derivative.^5,6) Non-derivatization methods have simple sample preparation, using only dilution of sample. However, HPLC and CE methods with UV detection can have inadequate sensitivity, especially in dissolution studies where formulations with low drug amount are tested.^20,21) Inadequate sensitivity can also be problematic in CE with indirect UV detection, moreover between runs rinses of the capillary prolong duration of analysis.^22,25) Cyclic voltammetry²³⁾ and flow analysis with chemiluminometric detection²⁴⁾ are rarely used for drug determination in pharmaceutical dosage forms and therefore less available for routine quality control analysis. All the above mentioned methods have been partially validated and proved to be suitable for quantification of PGB, GBP, and VGB in pharmaceutical dosage forms, however only a few methods were exposed to robustness test^{3,17,19–22)} which is crucial for routine quality control analysis. Developed HPLC or CE methods have run times longer than 4 to 5 min, which does not allow a high-throughput analysis. High-throughput analyses are also not possible in spectrophotometric or spectrofluorimetric methods with classical measurements in cuvette.

To overcome drawbacks of the literature methods, we decided to couple a simple derivatization method (Fig. 1b) developed by our group and described elsewhere³⁾ together with microplate reader with fluorescence detection. The method involves 5 min derivatization step at room temperature, with no subsequent need of extraction, and simultaneous measurements of 96 samples rendering possible high throughput analysis. Additionally, the method developed for PGB analysis was successfully applied to GBP and VGB and finally used for drug content and dissolution profile determination in a pharmaceutical dosage forms.

Experimental

Chemicals and Materials

Water for all applications was obtained from a Q-POD Milli-Q water purification system, (Millipore Corp., Billerica, MA, U.S.A.) with resistivity 18.2 MΩ cm. All chemicals used were at least of analytical grade and were purchased from: suprapur boric acid and sodium hydroxide (Merck, Darmstadt or Hohenbrunn, Germany); acetonitrile (J.T. Baker, Deventer, Holland); fluorescamine (Fluka, Buchs, Switzerland); pregabalin and gabapentin (Sequoia Research Products Ltd., Oxford, U.K.); vigabatrin (Sigma-Aldrich, Steinheim, Deutschland); Lyrica® capsules (Pfizer, Freiburg, Deutschland) containing 25 mg of PGB; Neurontin® capsules (Pfizer, Freiburg, Deutschland) containing 100 mg of GBP; Sabril® tablets (Sanofi-Aventis, Bourgoin-Jallieu, France) containing 500 mg of VGB. The pharmaceutical preparations were purchased from a local drugstore. For measurements the 96-well black flat bottom polystyrene microplates (Techno Plastic Products AG, Trasadingen, Switzerland) were used.

Instrumentation

The fluorescence signal was measured using a Tecan SafireII™ (Tecan Group Ltd., Männedorf, Switzerland) microplate reader. Excitation and emission wavelengths were set at 395 nm and 476 nm, respectively.

Standard Solutions

The stock solutions of PGB, GBP, and VGB were prepared in water to obtain 1 mg/mL. Calibration standards and quality control (QC) samples obtained from the stock were as follows: PGB from 0.75 to 30.0 µg/mL and QC samples (1.00, 12.5, 27.5 µg/mL); GBP from 2.00 to 80.0 µg/mL and QC samples (3.00, 30.0, 70.0 µg/mL); and VGB from 1.50 to 60.0 µg/mL and QC samples (2.00, 15.0, 50.0 µg/mL). All standards were prepared fresh daily. The stock solutions were stable at least 2 weeks if stored at 5°C.

Sample Preparation

Standards used for calibration curve, QC samples, or samples from analyte content in dosage form assays were prepared in duplicates by adding 20 µL of the water standard solution of PGB, GBP, or VGB, or 20 µL of the water sample obtained in dosage form assay, 150 µL borate buffer (pH 9.5; 0.25 M), 100 µL hydrochloric acid (0.06 M) and 30 µL fluorescamine solution (2.0 mg/mL in acetonitrile) into the microplate wells. Samples from dissolution studies were prepared equally just that this time 100 µL of the 0.06 M hydrochloric acid represented the sample and 20 µL of water standard solution was replaced by water. The obtained reaction mixture was gently mixed in microplate at room temperature for 5 min before analyzed.

The borate buffer was stable for at least 2 weeks and the fluorescamine solution for at least 2 d if stored at 5°C.

Optimization of Reaction Conditions

The derivatization with fluorescamine was optimized on PGB, using standard of 2.5 µg/mL as described elsewhere.³⁾

Method Validation

The method was validated according to International Conference on Harmonization (ICH) guidelines.²⁹⁾ Validation was performed on five separate days, each day including a minimum of ten calibration standards and three replicates of QC samples for all three drugs. A nonweighted linear regression was applied to calculate slopes and intercepts of the calibration curves constructed as signal intensity versus the analyte concentration. The determined validation parameters were: linearity, intraday and interday precision, accuracy, range, and limit of quantification.

The selectivity of the method was demonstrated by comparison of studied pharmaceutical dosage forms and placebo mixtures, which included the same excipients in maximal recommended amounts.³⁰⁾ The PGB and GBP capsules were completely emptied and filled with the same excipients as used in respective final products. For VGB tablets, only the mixture of its excipients was prepared. The average weight of the capsules content or tablet was 100, 133, and 680 mg for Lyrica®, Neurontin®, and Sabril®, respectively. The following mixtures equivalent to one pharmaceutical dosage form were prepared: Lyrica® (lactose monohydrate 75 mg, maize starch 7.5 mg, talc 1.5 mg), Neurontin® (lactose monohydrate 33 mg, maize starch 3.3 mg, talc 0.66 mg), and Sabril® (microcrystalline cellulose 180 mg, povidone 18 mg, sodium starch glycollate 9 mg, magnesium stearate 1.8 mg).

The stability of derivatized QC samples of PGB was investigated at room temperature for 2.5 h after derivatization.

The robustness of the method was determined using the one-variable-at-a-time approach. The robustness was investigated at three different levels: nominal zero level using optimal conditions, and by varying the parameter 1 unit up or down around the nominal zero level, +1 and −1 levels were obtained, respectively. The robustness study was divided into derivatization and spectrofluorimetric parts. In the derivatization test, pH of the derivatization buffer (9.0, 9.5, 10.0); concentration of fluorescamine (1.8, 2.0, 2.2 mg/mL); and time of derivatization (4, 5, 6 min) were modified. In the spectrofluorimetry excitation wavelength (391, 395, and 399 nm) and emission wavelength (471, 476, 481 nm) were modified. The robustness study was carried out by analyzing PGB QC samples. The obtained PGB response at level −1 and level +1 were compared with the nominal level and expressed as relative error in percent.

Applications of the Method

a) Assay of Active Moiety in Dosage Form

Accurate weights of mixed contents of 20 capsules (PGB, GBP) or powdered 20 tablets (VGB) equivalent to declared dose of the individual drug were transferred into a 100 mL volumetric flask, and the flask was filled with water. After 15 min of mixing on a magnetic stirrer, an aliquot of solution was filtered using a 0.45 µm nylon filter and appropriately diluted with water. Twenty microliter samples were analysed as described above. The nominal contents of the tablets or capsules were calculated using the calibration curves.

b) Capsules and Tablets Dissolution Studies

Drug release was measured using a United States Pharmacopeia (USP) 29 Apparatus 2 (Vankel VK 7000; Vankel Industries Inc., Cary, NC. U.S.A.) following U.S. Food and Drug Administration (FDA) recommendations for dissolution testing of PGB and GBP capsules. For dissolution testing of VGB the same procedure was used. The apparatus consisted of eight vessels in a water bath at 37°C and was equipped with an autosampler. The dissolution medium was 900 mL of 0.06 M hydrochloric acid. The rotation speed of the paddles was 50 rpm. For each dissolution profile, one pharmaceutical formulation was added to the medium, and 1 mL samples were withdrawn at 0, 5, 10, 15, 20, 30 and 45 min. Two additional sampling times were used for VGB, i.e., 60 and 120 min. The solutions were passed through the flow filter of the autosampler and treated according to the above procedure.

Results and Discussion

Method Optimization

Fluorescamine reacts with primary amino groups, forming fluorogenic product. The reaction is fast, proceeds at room temperature and resulted derivatives are stable.³¹⁾ In order to obtain optimal measurement conditions of the PGB derivative, its excitation and emission spectra were recorded. The spectra and chosen wavelengths (excitation maxima of 395 nm and emission maxima of 476 nm) are the same as those reported in our previous work.³⁾ For measurements, the 96-well flat bottom black polystyrene microplates were used with 300 µL working volume. The optimal reaction conditions were obtained by selecting appropriate organic and water solvent, their ratio, buffer pH and concentration, fluorescamine to PGB ratio, derivatization time, and derivatization temperature. To determine drug in dissolution samples, the concentration range of the analytical method should cover concentrations corresponding to 5–150% of drug present in dosage forms. This concentration range could be obtained if the volume of dissolution samples was at least 100 µL. The hydrochloric acid present in dissolution samples can influence the pH in reaction mixture. To enable constant pH of the reaction mixture at least 150 µL of 250 mM borate buffer was required. Using smaller volumes of borate buffer with higher concentrations resulted in slight precipitate formation. Considering the required volumes of assay samples (100 µL), standards (20 µL), and buffer (150 µL), only 30 µL of the reaction mixture remained for fluorescamine addition. The chosen volumes of reaction mixture solvents minimized the study of organic-to-water ratio influence on the reaction. Previously determined organic-to-water ratio have shown that the fluorescence responses of acetone to water of 1 : 9 were about 70% smaller than at the optimal ratio of 1 : 4.³⁾ Nevertheless, high concentration of organic solvent has to be avoided because organic solvent can also damage the polystyrene microplates. Among studied organic solvents (methanol, ethanol, acetone, and acetonitrile) acetonitrile did not noticeably damage the polystyrene, moreover it enables the highest yields of PGB derivative. The effect of pH on the reaction was examined using borate buffer solutions. Since fluorescamine react with a primary amine group under alkaline conditions, pH values between 8.5 and 10.5 were tested. The maximum fluorescence response was observed when the reaction was carried out at pH 9.5. Comparison of the influence of molar ratio of fluorescamine to PGB, time of derivatization and temperature of the derivatization reaction mixture showed the same trends as mentioned in our previous work.³⁾

Method Validation

The following method validation parameters were addressed for PGB according to the ICH guidelines: linearity, precision, accuracy, stability, specificity, and robustness. Parameters were based on the following acceptance criteria:

1. Linearity (5–150% of the drug content in pharmaceutical dosage forms): correlation coefficient (r²) of standards should be higher than 0.995.

2. Intraday precision: the RSD derived from repetitive measurements of QC samples should not exceed 5.0%.

3. Interday precision: the RSD derived from repetitive measurements of QC samples obtained within five consecutive days should not exceed 5.0%.

4. Accuracy: the determined concentrations of QC samples compared to their nominal concentrations should be within the range 95.0–105.0%.

5. Specificity: analyzing excipients in active moiety free preparation mixture should reveal no signal response.

6. Stability: the response of the QC samples at specified time compared to their response at time zero should be within the range 95.0–105.0%.

7. Robustness: deliberate variations in derivatization reaction conditions and spectrofluorimetric parameters 1 unit up or down around the nominal level should cause relative error of less than ±5.0% in analysis of QC samples.

Linearity, Limit of Detection (LOD) and Limit of Quantification (LOQ)

The nominal concentration of PGB after the dissolution experiments is 27.8 µg/mL (25 mg PGB per capsule in 900 mL of dissolution medium). In order to accurately determine the PGB dissolution profile, linearity was validated in the range of 0.75–30.0 µg/mL, covering more than 5–150% of the target value, considering the appropriate sample dilution, using twelve calibration points (n=12). A linear regression equation was obtained with a regression coefficient, r²=0.998.

The LOQ for PGB was 0.38 µg/mL. It was determined based on the standard deviation of y-intercepts (σ) slopes (S) of the analyte calibration curve s (n=3) using the following equation: LOQ=10σ/S.²⁹⁾ The same procedure was performed for GBP and VGB, and the results are summarized in Table 1.

Table 1. Method Validation Parameters of PGB, GBP, and VGB Standards

	PGB			GBP			VGB
Concentration range	0.75–30.0 µg/mL			2.0–80.0 µg/mL			1.5–60.0 µg/mL
Intraday precision (n=3)
Level	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)
QCl	1.0	1.0	3.5	3.0	2.9	1.1	2.0	2.1	1.7
QCm	12.5	12.7	4.4	30.0	30.7	2.6	15.0	15.2	1.5
QCh	27.5	28.1	3.4	70.0	70.4	0.4	50.0	50.5	2.1
Interday precision (n=5)
Level	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)	Nominal (µg/mL)	Mean (µg/mL)	RSD (%)
QCl	1.0	1.0	4.8	3.0	2.9	4.9	2.0	2.0	4.0
QCm	12.5	12.8	1.9	30.0	30.8	3.3	15.0	15.4	3.1
QCh	27.5	28.7	2.9	70.0	73.1	4.8	50.0	50.6	1.7
Accuracy (n=3)
Level	%			%			%
QCl	97.6			96.6			103.5
QCm	101.8			102.2			101.5
QCh	102.2			100.6			101.1
Linearity	y=740±6.5x−616±28.4			y=395±4.7x−1279±54.9			y=235±3.1x−833±15.6
	r2=0.998			r2=0.998			r2=0.999
Limit of quantification	0.38 µg/mL			1.39 µg/mL			0.66 µg/mL

QC_l: quality control samples at low concentration; QC_m: quality control samples at medium concentration; QC_h: quality control samples at high concentration.

The present method for PBG determination has higher LOQ than method developed by Onal and Sagirli.⁵⁾ However, the main advantage of our method is wider range, and high-throughput.

Furthermore, in the case of GBP and VGB determination, the proposed method offers similar sensitivity as method published by Belal et al.²⁾ Again, their method does not cover the measurements in the entire expected range and is not high-throughput.

Precision (Intraday and Interday)

The intraday precision was evaluated by replicate analysis of PGB, GBP and VGB QC samples during the same day, while the interday precision included analysis of QC samples for five consecutive days. The obtained intraday and interday precision of the method for PGB, GBP, and VGB were well within target limits with values not exceeding 4.44, 2.60, and 2.12% for intraday and 4.83, 4.93, and 4.02% for interday precision, respectively (Table 1).

Accuracy

The results of accuracy are presented in the Table 1. The accuracy met the acceptance criteria in all cases ranging between 97.6 and 102.2% for PGB, 96.6 and 100.6% for GBP and 101.1 and 103.5% for VGB, respectively.

Specificity

The derivatization reaction is specific for primary amino groups. Excipients in tested pharmaceutical dosage forms do not possess amino groups in their structure. Nevertheless, active moiety-free pharmaceutical preparation mixtures were prepared using all declared excipients in each pharmaceutical preparation. No interference of excipients was observed.

Stability

The response of the QC samples of PGB (n=3) was investigated for 2.5 h. The QC samples at low and medium level met the acceptance criteria within 2 h with response change ranging from 95.2 to 100.8% and RSD not exceeding 4.6%. The QC samples at high level were stable only 1.5 h with response changes ranging from 95.1 to 100.0% and RSD not exceeding 1.5%. Therefore, the derivatized samples should be measured at least 1.5 h after the end of derivatization procedure.

Robustness

The goal of robustness study was to determine the most critical parameters that affected our analytical procedure. Different parameters that affect derivatization procedure and spectrofluorimetry were investigated using commonly employed one-variable-at-a-time approach.³²⁾ The response of QC samples was most affected by fluorescamine concentration and spectrofluorimetric conditions. The later were shown to be most critical (Table 2). Therefore, precise and constant spectrofluorimetric conditions have to be assured to obtain reliable results.

Table 2. Robustness Study of Derivatization and Spectrofluorimetric Conditions

Derivatization	pH		Fluorescamine (mg/mL)		Time (min)		Spectrofluorimetry	Ex (nm)		Em (nm)
	9	10	1.8	2.2	4	6		391	399	471	481
QCl	−5.13	0.44	9.81	10.5	−1.55	−2.76	QCl	−11.8	−5.07	−12.3	−6.90
QCm	−2.78	−6.40	1.47	0.87	−6.16	−2.42	QCm	−10.7	−1.99	−11.4	−4.30
QCh	−3.46	−0.19	1.83	1.16	−1.28	−2.02	QCh	−11.3	−2.45	−12.8	−6.83

The results represent the relative error (%) for response of PGB QC samples. n=3. Ex, excitation wavelength; Em, emission wavelength.

Application of the Method

Assay of PGB, GBP and VGB in Pharmaceutical Dosage Form

The applicability of the optimized method was tested by determining the PGB, GBP and VGB in pharmaceutical dosage forms. The content of PGB in the 25 mg capsules, GBP in 100 mg capsule and VGB in 500 mg tablets was 26±0.4, 103±1.9 and 488±16 mg, respectively. The results obtained are in a good agreement with declared amount of PGB, GBP and VGB by manufacturers with the relative error not exceeding 4.8, 4.8 and 3.4%, respectively. Moreover, the content of PGB in the 25 mg capsules was determined also with our previously published method³⁾ employing HPLC. The results from HPLC and microplate fluorescence reader were statistically compared. Student’s t-test (t_cal=0.58; t_crit=2.12) and the variance ratio F-test (F_cal=0.247; F_crit=0.291) reveal no significant difference between determined PGB content in capsules at 5% significance level using both methods. The Bland–Altman analysis showed that the average difference in concentration between the two methods was 0.1857 mg; 95% interval of confidence (±1.96 SD, SD=0.8481 mg) ranged from −1.477 to 1.848 (Fig. 2). Therefore, we can conclude that investigated methods give comparable results.

Fig. 2. Bland–Altman Plot for PGB Content Determination in 25 mg Capsules Using Proposed Method (Spectrofluorimetric) and Previously Published HPLC Method³⁾

Capsules Dissolution Studies

The validated method was also used in dissolution studies of PGB, GBP and VGB pharmaceutical dosage forms at the lowest active moiety registered on the market. In accordance with the FDA recommendations for dissolution testing of PGB and GBP capsules, all dissolution profiles were determined in 0.06 M hydrochloric acid. PGB and GBP studied capsules showed similar dissolution profile with >90% release of the active moiety within 20 min and >97% in 30 min. VGB tablets showed a bit slower release, with >90% release of the active moiety within 45 min and >97% in above 120 min (Fig. 3). The dissolution results of GBP are in accordance with its monograph in the USP as more than 80% of the drug amount is released in 20 min.³³⁾

Fig. 3. Dissolution Profiles of PGB and Gabapentin GBP Capsules and VGB Tablets (n=3) in 0.06 M Hydrochloric Acid Using the USP 29 Apparatus 2

Conclusion

A simple, fast and sensitive spectrofluorimetric method for determination of PGB, GBP, and VGB was developed using microplate reader and fluorescamine as derivatization agent. The method offers high-throughput analysis which is its main advantage. The method was found to be linear, precise, and accurate across investigated analytical ranges. Moreover, it is also specific and robust. The application of the method demonstrated that it is suitable for drug content and release rate determination in pharmaceutical dosage forms. From the industrial perspective the developed method enables evaluation of considerable amounts of samples generated in pharmaceutical research and routine quality control using a limited amount of materials within a short turn around. Moreover, by optimization of the developed derivatization procedure the method could also be used for the determination of drug content and dissolution studies of pharmaceutical dosage forms containing drugs with amino groups especially if they have no significant UV or visible absorption.

Acknowledgment

This work was financially supported by the Slovenian Research Agency (ARRS Grant P1-0189).

References

• 1)  Kwan P., Sills G. J., Brodie M. J., Pharmacol. Ther., 90, 21–34 (2001).
• 2)  Belal F., Abdine H., Al-Majed A., Khalil N. Y., J. Pharm. Biomed. Anal., 27, 253–260 (2002).
• 3)  Martinc B., Grabnar I., Mrhar A., Vovk T., J. AOAC Int., 93, 1069–1076 (2010).
• 4)  Garcia L. L., Shihabi Z. K., Oles K., J. Chromatogr. B Biomed. Appl., 669, 157–162 (1995).
• 5)  Onal A., Sagirli O., Spectrochim. Acta A Mol. Biomol. Spectrosc., 72, 68–71 (2009).
• 6)  Olgun N., Erturk S., Atmaca S., J. Pharm. Biomed. Anal., 29, 1–5 (2002).
• 7)  Hassan E. M., Belal F., Al-Deeb O. A., Khalil N. Y., J. AOAC Int., 84, 1017–1024 (2001).
• 8)  Souri E., Jalalizadeh H., Shafiee A., Chem. Pharm. Bull., 55, 1427–1430 (2007).
• 9)  Chollet D. F., Goumaz L., Juliano C., Anderegg G., J. Chromatogr. B Biomed. Sci. Appl., 746, 311–314 (2000).
• 10)  Vermeij T. A., Edelbroek P. M., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 810, 297–303 (2004).
• 11)  Wad N., Kramer G., J. Chromatogr. B Biomed. Sci. Appl., 705, 154–158 (1998).
• 12)  Berry D., Millington C., Ther. Drug Monit., 27, 451–456 (2005).
• 13)  Bahrami G., Kiani A., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 835, 123–126 (2006).
• 14)  Mercolini L., Mandrioli R., Amore M., Raggi M. A., J. Pharm. Biomed. Anal., 53, 62–67 (2010).
• 15)  Abdellatef H. E., Khalil H. M., J. Pharm. Biomed. Anal., 31, 209–214 (2003).
• 16)  Siddiqui F. A., Arayne M. S., Sultana N., Qureshi F., Mirza A. Z., Zuberi M. H., Bahadur S. S., Afridi N. S., Shamshad H., Rehman N., Eur. J. Med. Chem., 45, 2761–2767 (2010).
• 17)  Shaalan R. A.-A., Int. J. Biomed. Sci., 6, 260–267 (2010).
• 18)  Al-Majed A. A., J. Liq. Chromatogr. Relat. Technol., 28, 3119–3129 (2005).
• 19)  Walash M. I., Belal F., El-Enany N., El-Maghrabey M. H., Luminescence, 26, 342–348 (2011).
• 20)  Ciavarella A. B., Gupta A., Sayeed V. A., Khan M. A., Faustino P. J., J. Pharm. Biomed. Anal., 43, 1647–1653 (2007).
• 21)  Gupta A., Ciavarella A. B., Sayeed V. A., Khan M. A., Faustino P. J., J. Pharm. Biomed. Anal., 46, 181–186 (2008).
• 22)  Sekar R., Azhaguvel S., J. Pharm. Biomed. Anal., 36, 663–667 (2004).
• 23)  Hegde R. N., Swamy B. E. K., Shetti N. P., Nandibewoor S. T., J. Electroanal. Chem., 635, 51–57 (2009).
• 24)  Manera M., Miró M., Ribeiro M. F., Estelaa J. M., Cerdà V., Santos J. L., Lima J. L., Luminescence, 24, 10–14 (2009).
• 25)  Shafaati A., Lucy C., J. Pharm. Pharm. Sci., 8, 190–198 (2005).
• 26)  Carlsson K. C., Reubsaet J. L., J. Pharm. Biomed. Anal., 34, 415–423 (2004).
• 27)  Oertel R., Arenz N., Pietsch J., Kirch W., J. Sep. Sci., 32, 238–243 (2009).
• 28)  Nirogi R., Kandikere V., Mudigonda K., Komarneni P., Aleti R., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 877, 3899–3906 (2009).
• 29)  International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (I.C.H.), “Q2/R1Validation of analytical procedures. Text and methodology”: ‹http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q2_R1/Step4/Q2_R1__Guideline.pdf›, cited November 2005.
• 30)  Alderborn G., “Aulton’s Pharmaceutics: The Design and Manufacture of Medicines,” 3rd ed., Chap. 31, ed. by Aulton M. E., Churchill Livingstone, Edinburgh, New York, 2007, pp. 441–482.
• 31)  Zhu R., Kok W. T., J. Chromatogr. A, 814, 213–221 (1998).
• 32)  Injac R., Boskovic M., Kocevar N., Vovk T., Anal. Chim. Acta, 620, 150–161 (2008).
• 33)  “United States Pharmacopeia-National Formulary,” USP 35-NF 30, Rockville, 2012, pp. 3296–3300.