Simultaneous quantitative determination of paracetamol and tramadol in tablet formulation using UV spectrophotometry and chemometric methods

    Simultaneous quantitative determination of paracetamol and tramadol in tablet formulation using UV spectrophotometry and chemometric methods Siniˇsa Glavanovi´c, Marija Glavanovi´c, Vladislav Tomiˇsi´c PII: DOI: Reference:

S1386-1425(15)30343-7 doi: 10.1016/j.saa.2015.12.020 SAA 14198

To appear in: Received date: Revised date: Accepted date:

31 July 2015 20 November 2015 17 December 2015

Please cite this article as: Siniˇsa Glavanovi´c, Marija Glavanovi´c, Vladislav Tomiˇsi´c, Simultaneous quantitative determination of paracetamol and tramadol in tablet formulation using UV spectrophotometry and chemometric methods, (2015), doi: 10.1016/j.saa.2015.12.020

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Simultaneous quantitative determination of paracetamol and tramadol in tablet formulation using UV spectrophotometry and chemometric methods

T

Siniša Glavanović a,*, Marija Glavanović a, and Vladislav Tomišić b,* a

IP

Belupo Inc., Research and Development, Koprivnica, Croatia, E-mail: [email protected] Division of Physical Chemistry, Department of Chemistry, Faculty of Science, University of Zagreb, Zagreb, Croatia, E-mail: [email protected]

SC R

b

NU

Abstract

The UV spectrophotometric methods for simultaneous quantitative determination of paracetamol and tramadol in paracetamol-tramadol tablets were developed. The

MA

spectrophotometric data obtained were processed by means of partial least squares (PLS) and genetic algorithm coupled with PLS (GA-PLS) methods in order to determine the content of

D

active substances in the tablets. The results gained by chemometric processing of the

TE

spectroscopic data were statistically compared with those obtained by means of validated ultra-high performance liquid chromatographic (UHPLC) method. The accuracy and

CE P

precision of data obtained by the developed chemometric models were verified by analysing the synthetic mixture of drugs, and by calculating recovery as well as relative standard error (RSE). A statistically good agreement was found between the amounts of paracetamol determined using PLS and GA-PLS algorithms, and that obtained by UHPLC analysis,

AC

whereas for tramadol GA-PLS results were proven to be more reliable compared to those of PLS. The simplest and the most accurate and precise models were constructed by using the PLS method for paracetamol (mean recovery 99.5 %, RSE 0.89 %) and the GA-PLS method for tramadol (mean recovery 99.4 %, RSE 1.69 %).

Keywords: UV spectrometry; Chemometric methods; Tramadol; Paracetamol; Tablets; Quantitative determination

ACCEPTED MANUSCRIPT 1

Introduction

Paracetamol (N-(4-hydroxyphenyl)acetamide is one of the most popular and most commonly used analgesic and antipyretic drug with no anti-inflammatory activity. Its mechanism of

T

action is complex and to date has not been completely elucidated. Due to a lack of significant

IP

peripheral action on prostaglandins, paracetamol is better tolerated than non-steroidal anti-

SC R

inflammatory drugs (NSAIDs) and has no gastrointestinal side effects [1,2]. Tramadol, (1RS,2RS)-2-[(dimethylamino)methyl]-1-(3-methoxyphenyl)cyclohexanol hydrochloride, is a centrally acting atypical opioid analgesic consisting of two enantiomers,

NU

both of which contribute to its activity by different mechanisms. It inhibits noradrenaline (norepinephrine) and serotonin reuptake by binding to their neuronal reuptake sites resulting in the simultaneous reduction of afferent pain signalling and the amplification of efferent

MA

inhibitory signalling [3,4].

Orally administered fixed-dose combination of tramadol and paracetamol in the form

D

of tablets is indicated for symptomatic treatment of moderate to severe pain. It is effective in providing pain relief in adult patients with postoperative pain after minor surgery,

TE

musculoskeletal pain, painful diabetic peripheral neuropathy, or migraine pain. As these two drugs have complementary modes of action and target multiple sites, their combination

CE P

provides better analgetic action against several types and sources of pain [5]. Chemometric calibration techniques in spectral analysis are widely used in quality control of drugs, especially in the analysis of mixtures and multicomponent pharmaceutical

AC

formulations with overlapping spectra. The advantage of such techniques lie in the fact that separation procedures of the drugs are not required. Partial least squares (PLS) is a regression method long employed for the quantitative processing of spectroscopic data in order to reduce dimensionality of the variables and to extract only relevant information [6,7]. This method, coupled with a genetic algorithm (GA), is a useful tool for wavelength selection in spectral data analysis. The wavelength reduction in PLS calibration using genetic algorithm can provide valuable information about the system studied, improve accuracy and precision of the results obtained by applying mathematical models, and reduce their complexity [8,9]. Numerous quantitative analytical methods have been reported for the separate determination of tramadol and paracetamol in pharmaceutical products and biological materials. Descriptions of flow injection–Fourier transform infrared spectrometric method [10], Fourier transform infrared [11], near infrared [12–15], UV-Vis [16,17], and fluorescence spectroscopy [18], flow-injection spectrophotometry [19], HPLC [20], HPLC2

ACCEPTED MANUSCRIPT MS [21], HPLC-MS-MS [22], GC [23,24], GC-MS [25,26], capillary electrophoresis with electrochemiluminescence detection [27,28], voltammetry [29,30], flow injection analysis system with square-wave voltammetric and amperometric detection [31], titrimetry [32], and

T

flow injection with chemiluminescence detection [33] can be found in the literature.

IP

Derivative spectrophotometry [34,35] and spectrophotometry coupled with multivariate calibration techniques [36–38] have been used for the determination of paracetamol in the

SC R

multicomponent mixture of drugs. Despite a large number of published papers describing individual determination of two analytes, there is a couple of papers on simultaneous quantification of two APIs in drug products and human plasma. A reversed phase high

NU

performance liquid chromatography [39], a second derivative spectrophotometry [40], and a differential pulse voltammetry with a glassy carbon electrode as a sensor [41] have been used

whereas

a

high

performance

MA

for the quantitative analysis of tramadol and paracetamol in pharmaceutical products, liquid

chromatography–electrospray

ionization-mass

spectrometric method has been reported for the determination of both drugs in human plasma

D

[42]. Due to a significant difference in amounts of the APIs in the tablet formulation (mass

TE

fraction of tramadol is 8.8 % and that of paracetamol is 76.5 %) as well as to their significant spectral overlapping, application of spectrophotometry for the quantitative analysis of tablets

CE P

is very challenging. To overcome these issues, in the present work PLS and GA-PLS

2 2.1

AC

chemometric methods have been applied for the UV spectrophotometric data processing.

Experimental Materials

The paracetamol-tramadol tablets (Paracetamol + Tramadol 325 mg + 37.5 mg tablets) were produced in Belupo’s pilot production plant. The quality of all raw materials used for tablet production complied with Ph. Eur. quality. The tramadol was bought from Chemagis Ltd. (Israel) and paracetamol from Mallinckrodt Inc. (USA). Both substances were standardized according to the corresponding analytical procedures described in Ph. Eur. monographs and afterwards used as the working standards. All the solvents and chemicals used were of analytical grade (Merck, Germany).

3

ACCEPTED MANUSCRIPT 2.2

Instruments

The UV spectra were recorded by means of a Varian Cary 100 UV-Vis spectrophotometer using 1 cm quartz cells from 200 to 320 nm with a sampling interval 1 nm and integration

T

time 0.5 s. The spectral data were processed by PLS Toolbox Solo (software version 7.1,

IP

demo). The statistical comparison of the results was carried out using OriginPro (software version 7.5).

SC R

The UHPLC analyses were performed by an Agilent 1290 Infinity LC system equipped with a quaternary pump, diode array detector, oven and automatic injector using a Zorbax SB C18, 50  2.1 mm, 1.8 µm (Agilent) column. The chromatographic conditions

NU

were: isocratic flow rate: 0.4 mL min-1; detection: 274 nm for tramadol and 244 nm for paracetamol; run time: 4 min; injection volume: 3 µL, and column temperature: 35 ºC.

MA

Acetonitrile sodium dihydrogen phosphate (c = 0.025 mol L-1, pH = 2.5) solution, with 3.5 mL of trimethylamine (15 : 85) was used as a mobile phase. The data acquisition, chromatogram processing and all calculations were performed by

D

means of an Empower Enterprise (Waters, USA) chromatographic software. The method was

Procedures

CE P

2.3

TE

validated according to the current ICH guidelines on validation of analytical procedures [43].

2.3.1 Preparation of standard and sample solutions

AC

The standard stock solutions for spectrophotometric analyses were prepared by dissolving 50 mg and 20 mg of paracetamol and tramadol in 50 mL and 200 mL mixture of 96 % ethanol and 0.1 mol L-1 HCl (1:1, volume ratio). The combined working standard solutions of both drugs were prepared by transferring appropriate volume of each standard stock solution into a 20 mL volumetric flask and diluting to volume with the previously mentioned mixture of solvents. The prepared concentrations were in the range of 15–37 µg mL-1 for paracetamol and 1.7–4.3 µg mL-1 for tramadol. Two sets of standard solutions were prepared, the calibration set containing 25 solutions and the validation set containing 19 solutions. The standard solutions for UHPLC analysis were prepared by dissolving appropriate amounts of tramadol and paracetamol in a mixture of 96 % ethanol and hydrochloric acid (c = 0.1 mol L-1) (1:4, volume ratio) to obtain the concentration of 18.7 µg mL-1 of tramadol and 162.5 µg mL-1 of paracetamol. 4

ACCEPTED MANUSCRIPT For the sample solution, one tablet was placed into a 100 mL volumetric flask and 1.0 mL of water was added to disintegrate the tablet. The sample was diluted to volume with 96 % ethanol / 0.1 mol L-1 HCl (1:1, volume ratio), sonicated to dissolve APIs and then filtered

T

through a 0.45 µm regenerated cellulose filter (Whatman, UK). For the spectrophotometric

IP

measurements the prepared solution was diluted with the same mixture of solvents 1 to 100 mL and for UHPLC analysis with 96 % ethanol / 0.1 mol L-1 HCl (1:4, volume ratio) 1 to 20

SC R

mL.

The sample and standard solutions were filtered using a 0.2 µm regenerated cellulose

Results and discussion

MA

3

NU

filter (Whatman, UK) before injection into the chromatographic system.

The UV spectra of standard solutions of paracetamol, tramadol and a mixture of both drugs are shown in Figure 1. Significant overlapping of the spectra of the drug substances can be

D

observed. In addition, the absorbances in the tramadol spectrum are significantly lower than

TE

those corresponding to paracetamol, and therefore the spectra of combined solutions are strongly dominated by the paracetamol spectrum.

CE P

The compositions of binary mixtures of standard solutions were chosen randomly to avoid correlation between the concentrations of two APIs. From the total number of prepared solutions, twenty-five were randomly selected and used for calibration, whereas nineteen solutions were used to validate the developed chemometric models. The compositions of the

AC

solutions are given in Tables 1 and 2. The calibration models were constructed by means of partial least squares method [6,7] and genetic algorithm coupled with PLS [9,44,45] in order to determine simultaneously concentrations of tramadol and paracetamol in the solutions.

5

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

CE P

TE

Figure 1. UV spectra of (a) tramadol (γ = 10 mg mL-1); (b) paracetamol (γ = 10 mg mL-1), and (c) mixture of tramadol (γ = 2 mg mL-1) and paracetamol (γ = 20 mg mL-1) in 96 % ethanol / 0.1 mol L-1 HCl (1:1, volume ratio); l = 1 cm.

The validated calibration models were then used to calculate the amounts of drugs

AC

under study in sample solutions of tablets. The same samples of paracetamol-tramadol tablets were simultaneously analysed using UV spectrophotometric and UHPLC methods, and the obtained results were statistically compared by means of one-way analysis of variance (ANOVA) [8].

6

ACCEPTED MANUSCRIPT

Table 1 Composition of calibration solutions.

IP

SC R

NU

MA

CE P

TE

Paracetamol 16.0 18.0 24.0 20.0 31.0 35.0 21.0 35.0 32.0 36.0 28.0 25.0 18.0 33.0 18.0 32.0 17.0 15.0 34.0 26.0 37.0 35.0 32.0 28.0 37.0

Chemometric methods

AC

3.1

Tramadol 3.10 1.70 3.90 2.10 2.90 3.70 4.20 3.80 1.70 4.30 4.00 1.90 2.20 1.90 4.00 2.50 1.90 2.50 2.60 3.20 4.00 4.00 1.70 4.00 3.80

T

 / g mL-1

D

Calibration solution 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

3.1.1 Partial least squares As already stated, the partial least squares method is frequently used for quantitative analysis of spectroscopic data to determine the amount of drugs in multicomponent pharmaceutical products [36,46–49]. The PLS regression is a full-spectrum method where calibration model is based on the latent variable decomposition. The algorithm takes into account the absorption UV spectra and the concentration of drugs in the calibration set of solutions simultaneously. There are two different algorithms to build the models. With of PLS1 regression the model is built for each analyte using its concentration vector whereas with PLS2 regression all analyte concentrations are used simultaneously [6,8]. In this paper the PLS1 algorithm was applied for building the calibration model, and leave one out sample at the time cross validation method was used to select of the optimum number of latent variables [8]. 7

ACCEPTED MANUSCRIPT The predicted tramadol and paracetamol concentrations in each sample, not included in the construction of the model, were compared with the actual concentrations in the calibration set, and the prediction residual error sum of squares (PRESS) as well as the root-

T

mean-square error of cross-validation (RMSECV) were calculated. To find the optimal

IP

model, the one with the lowest RMSECV and the fewest number of latent variables was located. After the addition of each new factor, RMSECV was recalculated and the variable

SC R

was retained if the value was improved for at least 5 %. The procedure was repeated for each drug separately. The optimal number of factors for paracetamol was 2 and for tramadol 6. The nineteen synthetic mixtures used to validate calibration model were analysed by

NU

the proposed method, and for each drug component the recovery and prediction error were calculated. The error was expressed as the relative standard error (RSE, eq. 1) of the

MA

predicted concentrations. The results are listed in Table 2.

TE

D

 N 2    ( predicted)i   ( added)i  RSE   i 1 N  ( ( added)i ) 2   i 1 

     

(1)

CE P

For tramadol the mean recovery of 99.0 % and RSE 2.58 % were obtained, whereas for paracetamol the respective values were 99.5 % and 0.85 %. It can be concluded that the

AC

calibration model developed with the PLS method yielded reliable and accurate results.

3.1.2 Genetic algorithm coupled with PLS Genetic algorithm is the method used for solving complex optimization problems. The appropriate selection of optimal wavelengths for construction of calibration model improves PLS predictive ability. In addition, GA can help understand which spectral regions correlate with specific physico-chemical properties of the system under study, especially when visual analysis is very difficult or even impossible to carry out. That is often the case in multicomponent systems with significant spectral overlapping of the analytes, or when chemical and physical interactions are present [50]. In GA, a subset of selected features (wavelengths) is called a chromosome and it is presented as one genetic vector. Each chromosome represents a possible or candidate solution of the problem. The chromosomes consist of genes representing the individual wavelength

8

ACCEPTED MANUSCRIPT values, and are coded into a binary digit system. The PLS model is generated using only the selected wavelengths in the genetic vectors.

TE

Recovery / % (PLS) Tramadol Paracetamol 99.2 98.0 99.7 99.5 98.3 99.9 101.5 99.5 100.8 99.1 99.8 101.2 105.1 100.2 95.8 99.5 105.2 99.4 96.9 99.9 96.5 99.1 95.9 99.3 98.6 100.1 98.6 98.8 94.9 100.4 96.2 100.2 101.0 100.1 99.3 97.1 97.9 98.8 99.0 99.5 2.58 0.85

NU

MA

D

Paracetamol 17.0 30.0 36.0 37.0 20.0 33.0 36.0 25.0 35.0 28.0 16.0 34.0 16.0 19.0 15.0 29.0 31.0 20.0 23.0

SC R

added / g mL-1 Tramadol 2.50 3.30 3.00 2.40 3.10 4.10 2.70 3.20 1.70 4.00 2.60 2.30 2.30 3.10 1.80 2.70 3.50 3.90 2.00

Recovery / % (GA-PLS) Tramadol Paracetamol 100.9 97.0 99.4 99.3 97.0 99.7 99.5 99.6 101.9 98.8 101.9 100.8 102.0 100.5 99.2 99.5 100.6 99.5 100.6 99.4 99.3 98.2 97.1 99.5 98.4 99.6 97.4 99.0 99.0 100.0 96.8 99.9 100.0 99.9 99.5 97.1 96.9 98.7 99.4 99.3 1.69 0.90

CE P

Validation solution 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Mean recovery RSE / %

IP

T

Table 2 Composition of validation samples and their recoveries calculated using PLS and GA-PLS methods.

The optimal number of factors is determined by cross validation on the model containing all the variables. For each chromosome the PRESS value is calculated and the

AC

members of the population are ranked according to their fitness values. Best solutions are kept to breed whereas other solutions are discarded. After that, the best chromosomes are recombined and mutated to generate a new generation of chromosomes termed offspring population. The parent population is replaced by a new set of generated chromosomes. Again, for each chromosome in the new population PRESS value is calculated and the whole process is iteratively repeated until the chromosomes with the highest fitness values are found. The creation of the new population of chromosomes represents one iteration cycle of genetic algorithm and is called a generation. The GA can be terminated after a fixed number of generations or after formation of the chromosome with a predefined fitness value [9,50–52]. As GA is a stochastic algorithm, GA-PLS calculations were repeated 10 times for 121 variables (200 – 320 nm) and the maximum number of latent variables used for building the PLS model was 10. The GA was terminated after 100 generations or when 50 % of the population was identical at convergence. The selected wavelengths were 221, 222, 223, 224, 9

ACCEPTED MANUSCRIPT 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 285, 286, 287, 288, 289 and 290 nm for tramadol and 205, 206, 207, 302, 303, 304, 305, 307 and 308 nm for paracetamol (Figure 2). For the construction of the calibration model for tramadol determination, only 2 latent

T

variables were found to be optimal, compared to the six when whole spectra were used. In the

IP

case of paracetamol, the optimal number of factors determined by means of the GA-PLS algorithm was three, whereas for PLS it was two. With the developed models excellent

SC R

accuracy and precision of the results was again obtained (Table 2).

Table 3 UHPLC and UV spectrophotometric analyses of paracetamol-tramadol tablets. UHPLC mg / tbl. % 324.7 99.9 328.4 101.1 324.3 99.8 330.1 101.6 328.9 101.2 324.0 99.7 322.4 99.2 329.2 101.3 328.8 101.2 317.5 97.7 100.3

Paracetamol PLS mg / tbl. % 325.9 100.3 328.3 101.0 322.5 99.2 329.8 101.5 327.5 100.8 323.8 99.6 323.5 99.5 329.6 101.4 329.4 101.3 319.7 98.4 100.3

GA-PLS mg / tbl. % 322.4 99.2 327.5 100.8 320.5 98.6 326.8 100.6 324.9 100.0 321.5 98.9 320.2 98.5 325.6 100.2 327.5 100.8 316.4 97.3 99.5

1.75

1.20

1.23

0.69

1.15

MA

D

CE P

1.47

NU

GA-PLS mg / tbl. % 37.7 100.6 38.1 101.5 37.9 101.1 37.7 100.6 38.2 101.8 37.4 99.8 37.6 100.2 37.8 100.8 38.3 102.2 36.7 97.9 100.7

AC

SD / %

Tramadol PLS mg / tbl. % 37.6 99.9 37.5 99.7 37.5 99.7 36.3 96.4 36.9 98.1 36.5 97.2 36.1 96.0 36.4 96.9 37.0 98.5 35.7 94.8 97.7

TE

UHPLC Tablet mg / tbl. % 1 37.1 98.8 2 37.9 101.0 3 37.9 101.2 4 37.8 100.7 5 38.2 101.8 6 37.9 101.1 7 37.6 100.3 8 37.7 100.7 9 37.9 101.1 10 36.3 96.8 Mean 100.4

10

ACCEPTED MANUSCRIPT Figure 2. Spectral regions of paracetamol (left) and tramadol (right) selected by GA-PLS method. UV spectra of mixture of paracetamol and tramadol (a), paracetamol (b), and

IP

T

tramadol (c).

SC R

3.1.3 Application of developed methods

To verify additionally the applicability of the proposed methods, ten individual tablets were simultaneously analysed using newly developed UV spectrophotometric and validated

NU

UHPLC methods.

SSa 5.89 8.45 14.34

Tramadol MSb F 2.94 9.41 0.31

D

DF 2 27 29

P - value 0.0008

SSa 44.69 376.58 421.27

Paracetamol MSb F 22.34 1.60 13.95

TE

Source of variation Between methods Within methods Total a SS – Sum of Squares b MS – Mean Square

MA

Table 4 Statistical comparison of tramadol and paracetamol amounts obtained by proposed chemometric and referent UHPLC methods by means of one-way ANOVA.

CE P

The amounts of both drugs in each tablet were determined by UHPLC against standard solutions of tramadol and paracetamol. Typical chromatograms recorded during the UHPLC analyses of tablets are shown in Figure 3. The results obtained by two multivariate chemometric and referent UHPLC methods are listed in Table 3. Statistical comparison of the

AC

data was carried out by applying one-way ANOVA (Table 4). In the case of paracetamol no significant difference between the methods was found, whereas in the case of tramadol the difference was significant. Hence, Bonferroni, Scheffe, Tukey and Fischer LSD tests were performed in order to determine which of two methods, PLS or GA-PLS, gave similar results compared to the data obtained by chromatographic analysis [8]. According to all tests, the amount of tramadol determined by applying GA-PLS for processing the UV spectrophotometric data was statistically in accordance with the value gained by the referent UHPLC method. A good agreement of the drug amounts obtained by different techniques indicates that the excipients in the pharmaceutical preparation did not interfere in the measurements of tramadol and paracetamol contents.

11

P - value 0.220

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

4

Conclusion

CE P

Figure 3. Typical chromatograms of sample solution of paracetamol-tramadol tablet. Tramadol is detected at 274 nm (left) and paracetamol at 244 nm (right).

AC

New UV spectrophotometric methods based on the multivariate chemometric techniques for quantitative data analysis were successfully applied for the simultaneous determination of active pharmaceutical ingredients in paracetamol-tramadol tablets. For both drugs a very good agreement between the results obtained by the new and referent UHPLC methods was found, whereby the GA-PLS based approach was proven to be somewhat superior to the ordinary PLS. In the case of tramadol, the former method significantly reduced the complexity of the calibration model and improved the accuracy and RSE of the predicted results. The proposed methods are simple, accurate, and do not require time-consuming sample preparation. Therefore, they can be easily used for a reliable simultaneous determination of tramadol and paracetamol in tablet formulation.

12

ACCEPTED MANUSCRIPT

Acknowledgements This work was supported by the Ministry of Science, Education and Sports of the Republic of

AC

CE P

TE

D

MA

NU

SC R

IP

T

Croatia, Croatian Science Foundation (project IP-2014-09-7309), and Belupo Inc.

13

ACCEPTED MANUSCRIPT 5

References B.J. Anderson, Paracetamol (Acetaminophen): Mechanisms of action, Paediatr. Anaesth. 18 (2008) 915–921. doi:10.1111/j.1460-9592.2008.02764.x.

[2]

M. Jozwiak-Bebenista, J.Z. Nowak, Paracetamol: Mechanism of action, applications and safety concern, Acta Pol. Pharm. - Drug Res. 71 (2014) 11–23.

[3]

T.A. Bamigbade, C. Davidson, R.M. Langford, J.A. Stamford, Actions of tramadol, its enantiomers and principal metabolite, O-desmethyltramadol, on serotonin (5-HT) efflux and uptake in the rat dorsal raphe nucleus., Br. J. Anaesth. 79 (1997) 352–356. doi:10.1093/bja/79.3.352.

[4]

S. Grond, A. Soblotzki, Clinical Pharmacology of Tramadol, Clin. Pharmacokinet. 46 (2004) 879–923.

[5]

S. Dhillon, Tramadol / Paracetamol Fixed-Dose Combination A Review of its Use in the Management of Moderate to Severe Pain, Clin. Drug Investig. 30 (2010) 711–733.

[6]

D.M. Haaland, E. V. Thomas, Partial least-squares methods for spectral analyses. 1. Relation to other quantitative calibration methods and the extraction of qualitative information, Anal. Chem. 60 (1988) 1193–1202. doi:10.1021/ac00162a020.

[7]

H. Abdi, Partial Least Square Regression PLS-Regression, in: N.J. Salkind (Ed.), Encyclopedia of Measurement and Statistics, Thousands Oaks, CA, 2007, pp. 1–13.

[8]

D.L. Massart, B.G.M. Vandeginste, L.M.C. Buydens, S. De Jong, P.J. Lewi, J. Smeyers-Verbeke, Handbook of Chemometrics and Qualimetrics: Part A, Elsevier, Amsterdam, 1997.

[9]

R. Leardi, Application of genetic algorithm–PLS for feature selection in spectral data sets, J. Chemom. 14 (2000) 643–655.

AC

CE P

TE

D

MA

NU

SC R

IP

T

[1]

[10] Z. Bouhsain, S. Garrigues, M. de la Guardia, Flow injection–Fourier transform infrared spectrometric determination of paracetamol in pharmaceuticals, Analyst. 121 (1996) 635 – 639. [11] M.A. Mallah, Simultaneous Quantification of Ibuprofen and Paracetamol in Tablet Formulations Using Transmission Fourier Transform Infrared Spectroscopy, Am. J. Anal. Chem. 03 (2012) 503–511. doi:10.4236/ajac.2012.38067. [12] E. Ziémons, J. Mantanus, P. Lebrun, E. Rozet, B. Evrard, P. Hubert, Acetaminophen determination in low-dose pharmaceutical syrup by NIR spectroscopy, J. Pharm. Biomed. Anal. 53 (2010) 510–516. doi:10.1016/j.jpba.2010.06.003. 14

ACCEPTED MANUSCRIPT

T

[13] Z. Wang, B. Xiang, Application of artificial neural network to determination of active principle ingredient in pharmaceutical quality control based on near infrared spectroscopy, Microchem. J. 89 (2008) 52–57. doi:10.1016/j.microc.2007.11.016.

SC R

IP

[14] M.M. Said, S. Gibbons, A.C. Moffat, M. Zloh, Near-infrared spectroscopy (NIRS) and chemometric analysis of Malaysian and UK paracetamol tablets: A spectral database study, Int. J. Pharm. 415 (2011) 102–109. doi:10.1016/j.ijpharm.2011.05.057.

NU

[15] M. Baptistao, W.F.D.C. Rocha, R.J. Poppi, Quality control of the paracetamol drug by chemometrics and imaging spectroscopy in the near infrared region, J. Mol. Struct. 1002 (2011) 167–171. doi:10.1016/j.molstruc.2011.07.019.

MA

[16] F.A. Mohamed, M.A. AbdAllah, S.M. Shammat, Selective spectrophotometric determination of p-aminophenol and acetaminophen, Talanta. 44 (1997) 61–68. doi:10.1016/S0039-9140(96)02013-9.

TE

D

[17] O.A. Ismaiel, M.M. Hosny, Development and Validation of a Spectrophotometric Method for the Determination of Tramadol in Human Urine Using Liquid-Liquid Extraction and Ion Pair Formation, Int. J. Instrum. Sci. 1 (2012) 34–40. doi:10.5923/j.instrument.20120103.02.

AC

CE P

[18] J.C.L. Alves, R.J. Poppi, Pharmaceutical analysis in solids using front face fluorescence spectroscopy and multivariate calibration with matrix correction by piecewise direct standardization, Spectrochim. Acta - Part A Mol. Biomol. Spectrosc. 103 (2013) 311–318. doi:10.1016/j.saa.2012.10.074. [19] M.S. Bloomfield, A sensitive and rapid assay for 4-aminophenol in paracetamol drug and tablet formulation, by flow injection analysis with spectrophotometric detection, Talanta. 58 (2002) 1301–1310. doi:10.1016/S0039-9140(02)00421-6. [20] A.P. Dewani, B.B. Barik, V.D. Chipade, R.L. Bakal, A.V. Chandewar, S.K. Kanungo, RP-HPLC-DAD method for the determination of phenylepherine, paracetamol, caffeine and chlorpheniramine in bulk and marketed formulation, Arab. J. Chem. 7 (2012) 811–816. doi:10.1016/j.arabjc.2012.07.010. [21] F. Musshoff, B. Madea, F. Stuber, U.M. Stamer, Enantiomeric determination of tramadol and O-desmethyltramadol by liquid chromatography- mass spectrometry and application to postoperative patients receiving tramadol., J. Anal. Toxicol. 30 (2006) 463–467. [22] B.N. Patel, N. Sharma, M. Sanyal, P.S. Shrivastav, An accurate, rapid and sensitive determination of tramadol and its active metabolite O-desmethyltramadol in human 15

ACCEPTED MANUSCRIPT plasma by LC-MS/MS, J. Pharm. doi:10.1016/j.jpba.2008.10.030.

Biomed.

Anal.

49

(2009)

354–366.

SC R

IP

T

[23] S.T. Ho, J.J. Wan, W.J. Liaw, C.M. Ho, J.H. Li, Determination of tramadol by capillary gas chromatography with flame ionization detection. Application to human and rabbit pharmacokinetic studies., J. Chromatogr. B. Biomed. Sci. Appl. 736 (1999) 89–96.

NU

[24] Q. Tao, D.J. Stone, M.R. Borenstein, V. Jean-Bart, E.E. Codd, T.P. Coogan, et al., Gas chromatographic method using nitrogen-phosphorus detection for the measurement of tramadol and its O-desmethyl metabolite in plasma and brain tissue of mice and rats., J. Chromatogr. B. Biomed. Sci. Appl. 763 (2001) 165–171. doi:10.1016/S03784347(01)00388-7.

MA

[25] Y.F. Sha, S. Shen, G.L. Duan, Rapid determination of tramadol in human plasma by headspace solid-phase microextraction and capillary gas chromatography-mass spectrometry, J. Pharm. Biomed. Anal. 37 (2005) 143–147. doi:10.1016/j.jpba.2004.09.050.

CE P

TE

D

[26] H.J. Leis, G. Fauler, W. Windischhofer, Synthesis of d1-N-ethyltramadol as an internal standard for the quantitative determination of tramadol in human plasma by gas chromatography-mass spectrometry, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 804 (2004) 369–374. doi:10.1016/j.jchromb.2004.01.045.

AC

[27] J. Li, H. Ju, Simultaneous determination of ethamsylate, tramadol and lidocaine in human urine by capillary electrophoresis with electrochemiluminescence detection, Electrophoresis. 27 (2006) 3467 – 3474. [28] S. Zhao, W. Bai, H. Yuan, D. Xiao, Detection of paracetamol by capillary electrophoresis with chemiluminescence detection, Anal. Chim. Acta. 559 (2006) 195– 199. doi:10.1016/j.aca.2005.11.071. [29] C. Engin, S. Yilmaz, G. Saglikoglu, S. Yagmur, M. Sadikoglu, Electroanalytical Investigation of Paracetamol on Glassy Carbon Electrode by Voltammetry, 10 (2015) 1916–1925. [30] C. Radovan, C. Cofan, D. Cinghita, Simultaneous Determination of Acetaminophen and Ascorbic Acid at an Unmodified Boron-Doped Diamond Electrode by Differential Pulse Voltammetry in Buffered Media, Electroanalysis. 20 (2008) 1346–1353. [31] E.M.P.J. Garrido, J.M.P.J. Garrido, F. Borges, C. Delerue-Matos, Development of electrochemical methods for determination of tramadol--analytical application to pharmaceutical dosage forms., J. Pharm. Biomed. Anal. 32 (2003) 975–981. doi:10.1016/S0731-7085(03)00199-7. 16

ACCEPTED MANUSCRIPT

T

[32] G. Burgot, F. Auffret, J.L. Burgot, Determination of acetaminophen by thermometric titrimetry, Anal. Chim. Acta. 343 (1997) 125–128. doi:10.1016/S00032670(96)00613-7.

SC R

IP

[33] D. Easwaramoorthy, Y.C. Yu, H.J. Huang, Chemiluminescence detection of paracetamol by a luminol-permanganate based reaction, Anal. Chim. Acta. 439 (2001) 95–100. doi:10.1016/S0003-2670(01)00968-0.

NU

[34] E. Dinç, A. Ozdemir, D. Baleanu, An application of derivative and continuous wavelet transforms to the overlapping ratio spectra for the quantitative multiresolution of a ternary mixture of paracetamol, acetylsalicylic acid and caffeine in tablets., Talanta. 65 (2005) 36–47. doi:10.1016/j.talanta.2004.05.011.

D

MA

[35] E. Dinç, The spectrophotometric multicomponent analysis of a ternary mixture of ascorbic acid, acetylsalicylic acid and paracetamol by the double divisor-ratio spectra derivative and ratio spectra-zero crossing methods, Talanta. 48 (1999) 1145–1157. doi:10.1016/S0039-9140(98)00337-3.

CE P

TE

[36] E. Dinç, D. Baleanu, G. Ioele, M. De Luca, G. Ragno, Multivariate analysis of paracetamol, propiphenazone, caffeine and thiamine in quaternary mixtures by PCR, PLS and ANN calibrations applied on wavelet transform data, J. Pharm. Biomed. Anal. 48 (2008) 1471–1475. doi:10.1016/j.jpba.2008.09.035.

AC

[37] M.R. Khoshayand, H. Abdollahi, M. Shariatpanahi, A. Saadatfard, A. Mohammadi, Simultaneous spectrophotometric determination of paracetamol, ibuprofen and caffeine in pharmaceuticals by chemometric methods, Spectrochim. Acta - Part A Mol. Biomol. Spectrosc. 70 (2008) 491–499. doi:10.1016/j.saa.2007.07.033. [38] M.R. Khoshayand, H. Abdollahi, A. Ghaffari, M. Shariatpanahi, H. Farzanegan, Simultaneous spectrophotometric determination of paracetamol, phenylephrine and chlropheniramine in pharmaceuticals using chemometric approaches., Daru. 18 (2010) 292–7. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3304346&tool=pmcentrez &rendertype=abstract. [39] R.M. Kamble, S.G. Singh, Stability-indicating RP-HPLC method for analysis of paracetamol and tramadol in a pharmaceutical dosage form, E-Journal Chem. 9 (2012) 1347–1356. doi:10.1155/2012/732506. [40] M. Inés Toral, J. Rivas, M. Saldías, C. Soto, S. Orellana, Simultaneous determination of acetaminophen and tramadol by second derivative spectrophotometry, J. Chil. Chem. Soc. 53 (2008) 1543–1548. doi:10.4067/S0717-97072008000200022. 17

ACCEPTED MANUSCRIPT

T

[41] F. Ghorbani-Bidkorbeh, S. Shahrokhian, A. Mohammadi, R. Dinarvand, Simultaneous voltammetric determination of tramadol and acetaminophen using carbon nanoparticles modified glassy carbon electrode, Electrochim. Acta. 55 (2010) 2752– 2759. doi:10.1016/j.electacta.2009.12.052.

SC R

IP

[42] T. Zhu, L. Ding, X. Guo, L. Yang, A. Wen, Simultaneous Determination of Tramadol and Acetaminophen in Human Plasma by LC–ESI–MS, Chromatographia. 66 (2007) 171–178. doi:10.1365/s10337-007-0268-9.

NU

[43] ICH, Validation of a analytical Procedures: text and methodology Q2(R1), Guidance, 2005, pp. 1–17.

MA

[44] J. Ghasemi, A. Niazi, R. Leardi, Genetic-algorithm-based wavelength selection in multicomponent spectrophotometric determination by PLS: Application on copper and zinc mixture, Talanta. 59 (2003) 311–317. doi:10.1016/S0039-9140(02)00505-2.

TE

D

[45] U. Depczynski, V.J. Frost, K. Molt, Genetic algorithms applied to the selection of factors in principal component regression, Anal. Chim. Acta. 420 (2000) 217–227. doi:10.1016/S0003-2670(00)00893-X.

CE P

[46] M. De Luca, F. Oliverio, G. Ioele, G. Ragno, Multivariate calibration techniques applied to derivative spectroscopy data for the analysis of pharmaceutical mixtures, Chemom. Intell. Lab. Syst. 96 (2009) 14–21. doi:10.1016/j.chemolab.2008.10.009.

AC

[47] S.M. Tawakkol, M. Farouk, O.A. Elaziz, A. Hemdan, M.A. Shehata, Comparative study between univariate spectrophotometry and multivariate calibration as analytical tools for simultaneous quantitation of Moexipril and Hydrochlorothiazide, Spectrochim. Acta - Part A Mol. Biomol. Spectrosc. 133 (2014) 300–306. doi:10.1016/j.saa.2014.05.061. [48] H.W. Darwish, S.A. Hassan, M.Y. Salem, B.A. El-Zeany, Different approaches in Partial Least Squares and Artificial Neural Network models applied for the analysis of a ternary mixture of Amlodipine, Valsartan and Hydrochlorothiazide, Spectrochim. Acta - Part A Mol. Biomol. Spectrosc. 122 (2014) 744–750. doi:10.1016/j.saa.2013.11.045. [49] H.C. Goicoechea, A.C. Olivieri, Chemometric assisted simultaneous spectrophotometric determination of four-component nasal solutions with a reduced number of calibration samples, Anal. Chim. Acta. 453 (2002) 289–300. doi:10.1016/S0003-2670(01)01232-6. [50] R. Leari, Genetic algorithm-PLS as a tool for wavelength selection in spectral data 18

ACCEPTED MANUSCRIPT sets, in: R. Leardi (Ed.), Nature-insprired Methods in Chemometrics: Genetic Algorithms and Artificial Neural Networks, 1st ed., Elsevier, Amsterdam, Boston, 2003, pp. 169–194.

SC R

IP

T

[51] M.M. Elkhoudary, R.A.A. Salam, G.M. Hadad, Comparative artificial neural network and partial least squares models for analysis of Metronidazole , Diloxanide , Spiramycin and Cliquinol in pharmaceutical preparations, Spectrochim. Acta Part a Mol. Biomol. Spectrosc. 130 (2014) 222–229. doi:10.1016/j.saa.2014.04.002.

AC

CE P

TE

D

MA

NU

[52] L. Sratthaphut, N. Ruangwises, Genetic algorithms-based approach for wavelength selection in spectrophotometric determination of vitamin B12 in pharmaceutical tablets by partial least-squares, Procedia Eng. 32 (2012) 225–231. doi:10.1016/j.proeng.2012.01.1261.

19

ACCEPTED MANUSCRIPT Highlights: Contents of paracetamol and tramadol in tablets were simultaneously determined. UV spectrometric data were analysed by PLS and GA-PLS methods.

AC

CE P

TE

D

MA

NU

IP

SC R

The proposed methods are simple, accurate, and reliable.

T

The obtained results were in agreement with those of validated UHPLC method.

20