Stability-Indicating Spectrophotometric Methods for Determination of Milnacipran HCl and Duloxetine HCl in Bulk Drug and Pharmaceutical Formulations

The present study describes accurate and sensitive stability-indicating spectrophotometric methods for the determination of Milnacipran HCl in presence of its acid, base degradates, and Duloxetine HCl in presence of its base degradates; including Dual-wavelength, Ratio difference, and Ratio derivative techniques. The developed methods were validated according to the International Conference on Harmonisation (ICH) guidelines. The recovery percentage of Milnacipran HCl and Duloxetine HCl were found to be in the ranges 99.42-100.42% and 100.17-100.3%, respectively. The low relative standard deviation of precision results confirms the suitability of the proposed methods for the estimation of the studied drugs in pure form, laboratory-prepared mixtures, and pharmaceutical formulations. Statistical comparison of the proposed methods with the reported methods revealed that there were no significant differences concerning the accuracy and precision of the adopted techniques. Validation studies demonstrated that the proposed methods are easy, specific, and rapid for the determination of Milnacipran HCl and Duloxetine HCl.

Fibromyalgia is an idiopathic, chronic, nonarticular pain syndrome, which commonly begins after a physical trauma, surgery, infection or significant psychological stress [2].
The widespread occurrence of fibromyalgia is common in the female gender and increasing age. The pathophysiology of fibromyalgia is not fully understood, however, exercise, relaxation, and stress-reduction were found useful in decreasing fibromyalgia symptoms. In addition, drugs increasing serotonin and norepinephrine activity have an important role in treating fibromyalgia [3]. Antidepressants as Milnacipran HCl (Fig. 1) and Duloxetine HCl (Fig. 2) can be used for the treatment of fibromyalgia. Duloxetine is a mixed serotonin/norepinephrine-reuptake inhibitor. Both the antidepressant activities and pain inhibitory properties of duloxetine are believed to be related to its potentiation of serotonergic and noradrenergic activity in the CNS, although the exact mechanisms in humans are unknown [4].
Milnacipran is a serotonin-norepinephrine reuptake inhibitor in a ratio of 1:3. Treatment of both depression and fibromyalgia is achieved by increasing both neurotransmitters concentration simultaneously [5].
Major Pharmacopeias like USP, EP, and BP have not reported spectrophotometric methods for the determination of Milnacipran HCl and Duloxetine HCl. Some sophisticated analytical methods for estimation of Milnacipran HCl [6][7][8][9][10][11][12][13] and Duloxetine HCl [14 -18] were stated in the literature. However, there hasn't been a stability-indicating spectrophotometric method yet developed for their determination in bulk and pharmaceutical formulations.
The proposed methods are considered the first spectrophotometric methods for the determination of Milnacipran and Duloxetine in presence of their degradation products by utilizing dual-wavelength [19], ratio difference [20], and ratio derivative [21,22] techniques. The scientific novelty of the developed methods is that they were found to be easier, more rapid, less expensive and less time-consuming than the sophisticated HPLC methods for the determination of the studied drugs.

Solutions preparation
ICH guidelines Q2(R1) [23] and Q1A(R2) [24] didn't mention specified conditions or reagents for stress testing of the drug substances. Studying the acid and base hydrolysis of a drug can be carried out by refluxing the drug in 0.1N HCl or NaOH for 8 h. If acceptable degradation was achieved, the trial can be stopped. Nevertheless, if inacceptable degradation occurs under these conditions, the drug should be refluxed in acid or base of higher concentrations and for longer periods. So, mild conditions (0.1N HCl or 0.1N NaOH) were tried at first but didn't give complete degradation. Therefore, drastic conditions were used to achieve complete degradation.

Milnacipran HCl
(1) Stock Standard Solution (1 mg/mL) In a 100-mL volumetric flask, 100 mg of the intact drug was dissolved in methanol A.R., sonicated and completed to volume with the same solvent.
(2) Working Solution (100 µg/mL) In a 100-mL volumetric flask, 10 mL of the stock standard solution was diluted to 100 mL with methanol A. R.
(3) Acid-induced Forced Degradation of Milnacipran HCl (20 µg/mL) In a conical flask, 2 mL stock standard solution (1 mg/mL) were added on 12 mL concentrated HCl, heated for 13 h on a hot plate (adjusted at 180 ο C), cooled, then excess acid was neutralized with 10M NaOH using a pH meter, and completed to 100 mL with methanol A.R.
(4) Base-induced Forced Degradation of Milnacipran HCl (40 µg/mL) In a conical flask, 4 mL stock standard solution (1 mg/mL) was added on 4 mL 10 M NaOH, heated for 120 min. on a hot plate (adjusted at 150 ο C), cooled, then the excess base was neutralized with 6 M HCl using a pH meter, and completed to 100 mL with methanol A. R.
The contents of five (Myodonia ® ) 25 mg and 50 mg tablets were crushed and mixed separately. In a 100-mL volumetric flask, an accurately weighed amount (equivalent to 100 mg of Milnacipran HCl) was diluted to the mark with methanol A. R, and then the flask was sonicated for 15 min and filtered.

Duloxetine HCl
(1) Stock Standard Solution (1 mg/mL) In a 100-mL volumetric flask, 100 mg of the intact drug was dissolved in methanol A.R, sonicated and completed to volume with the same solvent.
(2) Working Solution (100 µg/mL) In a 100-mL volumetric flask, 10 mL of the stock standard solution was diluted to 100 mL with methanol A. R.
(3) Acid-induced Forced Degradation of Duloxetine HCl (200 µg/mL) In a conical flask, 10 mL stock standard solution (1 mg/mL) were added on 10 mL 1 M HCl, heated for 40 min on a hot plate (adjusted at 90 ο C), cooled, then excess acid was neutralized with 10 mL 1 M NaOH using a pH meter, and completed to 50 mL with methanol A. R.
(4) Sample Preparation (100 µg/mL) The contents of five Cymbatex ® 30 mg and 60 mg capsules were emptied, crushed and mixed separately. In a 100-mL volumetric flask, an accurately weighed amount (equivalent to 100 mg of Duloxetine HCl) was diluted to the mark with methanol A. R., and then the flask was sonicated for 15 min and filtered. Then, 10 mL was further diluted to 100 mL with methanol A. R.

Spectral Characteristics
(1) Milnacipran HCl and its acid-induced and base-induced degradates The zero-order absorption spectra of Milnacipran HCl (10 µg/mL) and (20 µg/mL) were measured in the range 240 nm to 300 nm against methanol A. R. The zero-order absorption spectra of acid-induced degradates (10 µg/mL) and base-induced degradates (20 µg/mL) were measured at the same wavelength range, where the absorbance difference at 252.8 nm and 259.7 nm, at 255 nm and 261.4 nm was found to be zero for acid-induced degradates and baseinduced degradates; respectively.
(2) Duloxetine HCl and its acid-induced degradates The zero-order absorption spectrum of Duloxetine HCl (4 µg/mL) was recorded over the spectral wavelength range 220 nm to 340 nm against methanol A. R. The zero-order absorption spectrum of acid-induced degradates (4 µg/mL) was measured at the same wavelength range where the absorbance difference at 323 nm and 274.5 nm was found to be zero. Into a series of 100-mL volumetric flasks, accurately measured aliquots from the working solution (100 µg/mL) were separately transferred to produce 6-14 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in triplicate in the wavelength range from 240 nm to 300 nm; then the absorbance was determined at 252.8 nm and 259.7 nm, where the difference in absorbance of acid-induced degradation was found to be zero. A calibration curve was constructed by plotting the difference in absorbance at the selected wavelengths against Milnacipran concentrations.

Method
Similarly, another set of 100-mL volumetric flasks were prepared, accurately measured aliquots of working solution (100 µg/mL) were transferred to produce 4-30 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in the same wavelength range. The absorbance was determined at 255 nm and 261.4 nm, where the difference in absorbance of base-induced degradation was found to be zero. A calibration curve was constructed by plotting the difference in absorbance at the selected wavelengths against Milnacipran concentrations.

(2) Ratio Difference Spectrophotometric Method
Into a series of 100-mL volumetric flasks, accurately measured aliquots from the working solution (100 µg/mL) were separately transferred to produce 2-18 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in triplicate in the wavelength range from 240 nm to 300 nm. Each of these spectra was divided by acid-induced degradation of Milnacipran HCl (10 µg/mL) as a divisor to get the corresponding ratio spectra. Similarly, another set of 100-mL volumetric flasks were prepared, accurately measured aliquots of working solution (100 µg/mL) were transferred to produce 4-30 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in the same wavelength range.
Each of these spectra was divided by baseinduced degradates of Milnacipran HCl (20 µg/mL) as a divisor to get the corresponding ratio spectra. In the case of acid-induced degradates, the absorbance was determined at 249.6 nm and 270.6 nm. In the case of base-induced degradates, the absorbance was determined at 255 nm and 270.9 nm. Two calibration curves were constructed relating the absorbance difference at the selected wavelengths versus the corresponding concentrations of Milnacipran in each case.

(3) Ratio Derivative Spectrophotometric Method
Into a series of 100-mL volumetric flasks, accurately measured aliquots from the working solution (100 µg/mL) were separately transferred to produce 2-15 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in triplicate in the wavelength range from 240 nm to 300 nm. Each of these spectra was divided by acid-induced degradation of Milnacipran HCl (10 µg/mL) as a divisor to get the corresponding ratio spectra. The first derivative of each of the obtained ratio spectra was computed using a scaling factor 10 and ∆ʎ= 4 nm. Peak amplitudes were measured at 272.7 nm and plotted against Milnacipran HCl concentrations in the prepared solutions to construct a calibration curve.
The same procedure was applied to the ratio spectra of the solutions 4-30 µg/mL prepared into a series of 100-mL volumetric flasks, from measured aliquots of working solution (100 µg/mL), measured in triplicate in the same wavelength range mentioned above and divided by base-induced degradates of Milnacipran HCl (20 µg/mL) as a divisor. Peak amplitudes were measured at 273.6 nm and plotted against Milnacipran HCl concentrations in the prepared solutions to construct a calibration curve.

Accuracy
Accuracy of the mentioned spectrophotometric methods under Linearity (2.4.2.1.1) was calculated by determining the %recoveries of 8 -12 µg/mL standard solutions, prepared from working solution (100 µg/mL) (in case of intact drug with acid-induced degradates) and 14-30 µg/mL standard solutions, prepared from working solution (100 µg/mL) (in case of intact drug with base-induced degradates), measured in triplicate using the corresponding regression equations.

(a) The intra-day precision
Three replicates of Milnacipran HCl at three different concentration levels of 8, 10, and 12 µg/mL (in case of intact drug with acid-induced degradates) and 16, 20, 24 µg/mL (in case of intact drug with base-induced degradates) were analyzed on the same day by applying the same procedures under Linearity (2.4.2.1.1), and the relative standard deviation was calculated.

(b) The inter-day precision
The above-mentioned concentrations of Milnacipran HCl were analyzed on three successive days by applying the same procedures under Linearity (2.4.2.1.1) and the relative standard deviation was calculated.

Specificity
For acid degradation, into three separate 50-mL volumetric flasks, accurately measured aliquots of working solution (100 µg/mL) were transferred to produce 8, 10, 12 µg/mL each, accurately measured aliquots of acid-induced degradates were added to produce 10, 8, 6 µg/mL, each on a flask, volume was completed to the mark with methanol A. R. Each of these solutions was measured in triplicate using the procedures mentioned under Linearity (2.4.2.1.1).
For base-induced degradation, into three separate 50-mL volumetric flasks, accurately measured aliquots of working solution (100 µg/mL) were transferred to produce 16, 20, 24 µg/mL each, accurately measured aliquots of base-induced degradates were added to produce 20, 16, 12 µg/mL, each on a flask, volume was completed to the mark with methanol A. R. Each of these solutions was measured in triplicate using the procedures mentioned under Linearity (2.4.2.1.1).

Limit of Detection (LOD) and Limit of Quantitation (LOQ)
Approaches based on the standard deviation of the intercept and the slope were used for determining the LOD and LOQ, where Into a series of 100-mL volumetric flasks, accurately measured aliquots from the working solution (100 µg/mL), were separately transferred, to produce 1-8 µg/mL and then completed to the mark with methanol A. R. Each of these solutions was measured in triplicate in the wavelength range from 220 nm to 340 nm. The absorbance was determined at 323 nm and 274.5 nm, where the difference in absorbance of acid-induced degradation was found to be zero. A calibration curve was constructed between the difference in absorbance at the selected wavelengths and Duloxetine concentrations.

(2) Ratio Difference Spectrophotometric Method
Each of the spectra of the solutions (1-8 µg/mL) obtained in the previously mentioned method (IV.B.1.6.2.1.1.) was divided by acid-induced degradation of Duloxetine HCl (4 µg/mL) as a divisor to get the corresponding ratio spectra. The absorbance was measured at 320 nm and 257 nm. A calibration curve was constructed by plotting the absorbance difference at the selected wavelengths versus the concentrations of Duloxetine HCl.

(3) Ratio Derivative Spectrophotometric Method
The first derivative of each of the ratio spectra of the solutions (1-8 µg/mL) obtained in the previously mentioned method (IV.B.1.6.2.1.2.) was computed using a scaling factor 10 and ∆ʎ= 4 nm. Peak amplitudes were measured at 322 nm and plotted against Duloxetine HCl concentrations in the prepared solutions to construct a calibration curve.

Precision
(a) The intra-day precision: Three replicates of Duloxetine HCl at three different concentration levels of 3, 4, and 5 µg/mL were analyzed on the same day by applying the procedures under Linearity (2.4.2.2.1), and the relative standard deviation was calculated.

(b) The inter-day precision
The above-mentioned concentrations of Duloxetine HCl were analyzed on three successive days by applying the procedures under Linearity (2.4.2.2.1) and the relative standard deviation was calculated.

Specificity
Into three separate 100-mL volumetric flasks, accurately measured aliquots of working solution (100 µg/mL) were transferred to produce 3, 4, 5 µg/mL each, accurately measured aliquots of acid-induced degradates were added to produce 5, 4, 3 µg/mL, each on a flask, volume was completed to the mark with methanol A. R. Each of these solutions was measured in triplicate using the procedures mentioned under Linearity (2.4.2.2.1).

Limit of Detection (LOD) and Limit of Quantitation (LOQ)
Approaches based on the standard deviation of the intercept and the slope were used for determining the LOD and LOQ, where Further, the standard addition technique was followed: Into three separate 50-mL volumetric flasks, accurately measured aliquots of sample solution prepared using 25 mg tablets as under (2.3.1.5) were transferred separately to produce 6 µg/mL and 12 µg/mL each, accurately measured aliquots of working solution of Milnacipran HCl as prepared under (2.3.1.2) were added to produce 2, 4, 6 µg/mL and 4, 8, 12 µg/mL each on a flask; respectively, volume was completed to the mark with methanol A. R. Each of these solutions was measured in triplicate using the procedures mentioned under Linearity (2.4.2.1.1).
The same steps were repeated with solutions prepared using 50 mg tablets as under (2.3.1.5).

Srinivas et al [12] and Hussain et al [13]
developed colorimetric methods for the determination of Milnacipran HCl in bulk and pharmaceutical formulations. Samal and Prusty [16] developed a zero-order UV spectrophotometric method for the determination of Duloxetine HCl in bulk and pharmaceutical formulations. The advantage of the methods discussed in this paper over the methods in literature is that they are specific stabilityindicating methods used to determine Milnacipran in presence of its acid and base degradation products; and Duloxetine in presence of its acid degradation products.
The absorption spectra of Milnacipran HCl and its degradates are strongly overlapped, so the application of the traditional spectral techniques failed to resolve this problem. However, the resolving power of the proposed methods could be demonstrated by this spectral overlapping. Fig. 3 shows both the overlap between acid, base-induced degradation spectra and Milnacipran HCl spectrum; and the overlap between acid-induced degradation spectrum and Duloxetine HCl spectrum. Fig. 3. Zero order absorption spectra of intact Milnacipran HCl (10 µg/mL) and its (a) acid degradation product (10 µg/mL) in methanol (b) base degradation product (20 µg/mL) in methanol (c) intact Duloxetine HCl (4 µg/mL) and its acid degradation product (4 µg/mL) in methanol Dry heat degradation and photodegradation were tried but no significant change in the peak area appeared, indicating the stability of Milnacipran HCl and Duloxetine HCl to thermal and photodegradation.
Optimization of the influence of different variables as the choice of the divisor and the working wavelengths on the developed spectrophotometric methods was studied, consequently, in case of ratio difference method, different concentrations of acid-induced degradates (6, 8, 10, 12 μg/mL) of Milnacipran HCl were tried as divisors and the concentration of 10 μg/mL was selected as the one giving better selectivity and minimum noise. Similarly, different concentrations of base-induced degradates (12, 16, 20, 24 μg/mL) of Milnacipran HCl were tried as divisors and the concentration of 20 μg/mL was selected. Finally, different concentrations of acid-induced degradates (2, 3, 4, 5 μg/mL) of Duloxetine HCl were tried as divisors and the concentration of 4 μg/mL was selected as the one giving minimum noise and better selectivity.
Concerning wavelength selection, in case of dual-wavelength, 252.8 nm and 259.7 nm were selected, where the difference in Milnacipran HCl acid-induced degradation absorbance was found to be zero; 255 nm and 261.4 nm were selected, where the difference in Milnacipran HCl base-induced degradate absorbance was found to be zero and 323 nm and 274.5 nm were selected, where the difference in Duloxetine HCl acid-induced degradation absorbance was found to be zero allowing determination of Milnacipran HCl and Duloxetine HCl without interference from their acid and base degradates. While ratio difference, the absorbance was measured at 249.6 nm and 270.6 nm in case of Milnacipran HCl acid-induced degradates; at 255 nm and 270.9 nm in case of Milnacipran HCl base-induced degradates and 320 nm and 257 nm in case of Duloxetine HCl acid-induced degradates. For the ratio derivative method, different trials were performed to optimize the scaling factor and ∆ʎ where 10 and 4 were found to give the best spectrum shapes respectively, and absorbances were determined at peak amplitude in each case.

Method Validation
Method validation was performed according to the ICH guidelines [23] for the suggested spectrophotometric methods. Linearity was assessed by analyzing different concentrations of Milnacipran HCl in the ranges of 6-14 μg/mL, 2-18 μg/mL and 2-15 μg/mL for dual-wavelength, ratio difference and ratio derivative; respectively (in case of acid-induced degradates) and 4-30 μg/mL for the three methods (in case of base-induced degradates). Similarly, the proposed methods were found to be linear in the range of 1-8 μg/mL for Duloxetine HCl. Calibration curves were constructed between either the average absorbance difference -in dual-wavelength and ratio difference methods -or average peak amplitude in ratio derivative method, and the corresponding concentrations of Milnacipran or Duloxetine.
The low values of %RSD obtained in both cases by the proposed methods suggest excellent precision. The results of precision are shown in Tables 1 and 2.
The LOD and LOQ were calculated using the calibration curves of the proposed methods and the obtained values are shown in Tables 1 and 2. Fig. 6. First derivative of ratio spectra for (a) Milnacipran HCl (2-15 µg/mL) with its acid degradation product using (10 µg/mL) acid degradation product as the divisor (b) Milnacipran HCl (4-30 µg/mL) with its base degradation product using (20 µg/mL) base degradation product as the divisor Fig. 7. (a) Zero-order absorption spectra of intact Duloxetine HCl (1-8 µg/mL) with its acid degradation product (4 µg/mL) using the dual-wavelength method (b) Ratio absorption spectra of intact Duloxetine HCl (1-8 µg/mL) after a resolution from its acid degradation product (4 µg/mL) using the ratio difference method at the specified wavelengths (c) The first derivative of ratio spectra for solutions of Duloxetine HCl (1-8 µg/mL) with its acid degradation product using (4 µg/mL) acid degradation product as the divisor Fig. 8. Linearity of the absorbance difference corresponding to the concentration of (a) Milnacipran HCl (6-14 µg/mL) with its acid degradation and (b) Milnacipran HCl (4-30 µg/mL) with its base degradation; respectively using dual-wavelength method Fig. 9. Linearity of the absorbance difference corresponding to the concentration of (a) Milnacipran HCl (2-18 µg/mL) with its acid degradation and (b) Milnacipran HCl (4-30 µg/mL) with its base degradation; respectively using the ratio difference method Fig. 10. Linearity of the peak amplitude corresponding to the concentration of (a) Milnacipran HCl (2-15 µg/mL) with its acid degradation and (b) Milnacipran HCl (4-30 µg/mL) with its base degradation; respectively using the ratio derivative method Fig. 11. (a) Linearity of the absorbance difference corresponding to the concentration of Duloxetine HCl (1-8 µg/mL) with its acid degradation using the dual-wavelength method (b) The linearity of the absorbance difference corresponding to the concentration of Duloxetine HCl (1-8 µg/mL) with its acid degradation using the ratio difference method (c) The linearity of the peak amplitude corresponding to the concentration of Duloxetine HCl (1-8 µg/mL) with its acid degradation using ratio derivative method       Specificity was checked in laboratory prepared mixtures to ensure the ability of each proposed method to measure the intact drug response without interference from its degradation products. Recovery and standard deviations were calculated. Satisfactory results were obtained as shown in Tables 5 and 6.

Application to Pharmaceutical Formulations
The developed methods were successfully used to determine Milnacipran HCl and Duloxetine HCl in their pharmaceutical formulations. Moreover, the standard addition technique was applied to assess the validity of the proposed methods as shown in Tables 7 and 8.
A statistical comparison of the results obtained by the proposed methods and the reported methods [6,14] for the determination of Milnacipran HCl and Duloxetine HCl was done. Statistical comparison between the results was performed using a student's t-test and F-test at a 95% confidence level where there was no significant difference concerning accuracy and precision between the proposed methods and the reported methods [6, 14] as shown in Tables 9 and 10; respectively.

CONCLUSION
This paper describes simple, rapid, and costeffective methods for the determination of Milnacipran and Duloxetine. The proposed methods are sensitive and depend on the use of simple and cost-effective chemicals and techniques; however, they provide a sensitivity equivalent to that obtained by sophisticated and expensive techniques like HPLC. In addition, they are considered stability-indicating assay methods as they can separate each intact drug from its degradation products.

Ethics approval and consent to participate
Not applicable

Consent to publish
Not applicable