Quantification of Acetaminophen, Caffeine and Ibuprofen in Solid Dosage Forms by UV Spectroscopy coupled with Multivariate analysis

Nguyen Anh Thu*

Faculty of Pharmacy, Hoa Binh University, No. 8 Bui Xuan Phai, My Dinh II, Nam Tu Liem, Hanoi, Vietnam.

*Corresponding Author E-mail: ntathu@daihochoabinh.edu.vn

ABSTRACT:

Quantification of acetaminophen, caffeine and ibuprofen in ternary mixtures was studied by using UV spectroscopy coupled with multivariate analysis [tri-linear regression-calibration (TLRC), classical least squares (CLS) and multi linear regression calibration (MLRC)]. To validate UV spectroscopic methods, the wavelength range of 215 – 290nm and a matrix (composed of 21 standard solutions containing acetaminophen 20-32.5mg/L, caffeine 1-3.5mg/L and ibuprofen 12-32mg/L) were selected. With the exception of TLRC, CLS and MLRS algorithms proved to be relevant for obtaining UV spectroscopic assay results and dissolution profile of acetaminophen, caffeine and ibuprofen in their combined capsules and tablets, with precision (RSD < 3%) and accuracy (98.3 – 101.9% recovery). It is also suggested that these UV spectroscopic methods could replace UHPLC analysis in the routine quality control of these compounds in their solid pharmaceutical dosage forms (p > 0.05).

KEYWORDS: Acetaminophen, Caffeine, Ibuprofen, Tablets, Capsules, Multivariate analysis, TLRC, CLS, MLRC.

1. INTRODUCTION:

The combination of two or more active ingredients in a single-dosage formulation has been more and more clinically exploited to enhance efficacy and minimize adverse effects. Acetaminophen is a popular non-opioid antipyretic and analgesic drug for pain and fever treatment, but it has limited antiflammatory effect¹. To overcome this drawback, acetaminophen could be concomitantly used with a non-steroidal anti-inflammatory (NSAID) drug such as ibuprofen².

Caffeine is a xanthine alkaloid found in non-alcoholic beverages (e.g. tea, coffee, and cocoa)³, which may boost mood, metabolism, mental and physical performance. When being orally taken, this compound can reduce pain sensation (via central blockage of adenosine receptors) and increase cerebral arterial vasoconstriction leading to a reduction in brain blood volume and cerebral blood inflow. As a result, the combination of acetaminophen and caffeine can strengthen markedly the pain relief of acetaminophen as well as help patients stay much more vigilant.

In the literature, chemometrics (especially, multivariate analysis) has been numerously applied in pharmaceutical analysis since the 1990s^4,5. It has been shown that HPLC and chemometrics-based UV spectrophotometry were numerously studied for assay of principle ingredients in binary and ternary mixtures^6-15.

This study aimed at proposing a quantitation method for acetaminophen, caffeine and ibuprofen in combined solid dosage forms by coupling UV spectroscopy and multivariate analysis. The applicability of this method was assessed by content determination and dissolution test.

2. EXPERIMENTAL:

Three Vietnamsese solid dosage forms (containing acetaminophen, caffeine and ibuprofen) under investigation were purchased at local pharmacies i.e. Glotasic tablet (Glomed Pharmaceutical Co., Inc), Ibu-acetalvic capsule (Vidipha), Bidi–ipalvic capsule (Bidiphar). Vietnamese Pharmacopoeia chemical reference substances were used: acetaminophen (AC, 99.7%), caffeine (CA, 99.8%) and ibuprofen (IB, 99.9%). The spectrophotometric solvent was water. Mixture standard solutions were prepared by appropriate dilution of stock solutions of AC 500mg/L, CA 200 mg/L and IB 500mg/L.

For assay, the content of 20 capsules (or 20 tablets) was finely crushed and thoroughly mingled in a mortar. A mass (equivalent to a theoretical mixture containing 325 (300)mg AC, 25 (20)mg CA and 200mg IB) was accurately weighed and ultrasonic-dissolved in about 60 mL of water in a 100-mL graduated flask for 20 min before adding water to the mark. This resultant solution was filtered and the first 20-30mL of the filtrate discarded. 10mL of the filtrate was accurately taken for dilution to 100mL with water to obtain a solution for measurement.

Dissolution test was realized at 37°C by using 900mL water as the dissolution medium on an Erweka DT 626 dissolution tester (Germany) with a paddle apparatus set at 150rpm. At pre-determined time intervals (within 0 – 60 min), 5mL dissolution samples were appropriately diluted (2 – 10 times) with water and filtered before measurement.

Spectroscopic measurement in a stardard 1-cm cuvette was done with a double-beam UV-1800 spectrophometer Shimadzu (Japan) with the optimum slit width fixed at 1.0nm. The absorption spectra were recorded between 200 and 300nm with average scanning speed and 0.05-nm resolution.

Multivariate analysis was done by using Microsoft EXCEL with Add – in XLSTAT (Addinsoft, USA).

For comparison, the assay of AC, CA and IB in tablets and capsules was aslo chromatographically done with an Ascentis Express C18 (50×2.1mm, 2.7μm) set at 35°C¹⁶ on a UHPLC, Agilent 1290 Infinity II LC System (USA). The mobile phase for gradient UHPLC analysis composed of [A] (5mM ammonium phosphate monobasic, pH 2.0 adjusted with phosphoric acid: acetonitrile (98:2, v/v) and [B] (acetonitrile). The gradient was a linear change from 0 to 68% B in 3 min and subsequently held at 68% B for 2 min when the mobile phase delivered at 0.4mL/min. The solvent for sample preparation was a mixture (95:5, water: methanol). The injection volume was 1µL and detector set at 210nm.

3. THEORETICAL BACKGROUND:

In this study, multivariate analysis applied to UV spectroscopic data include (i) tri-linear regression-calibration (TLRC), (ii) classical least squares (CLS) and (iii) multi linear regression calibration (MLRC)^8,17,18.

Basically, the linear relationship between concentration and absorbance of an analyte can be mathematically described by the following equation

A_Xi = b_Xi C_X + a_Xi                                                                (1)

A_Xi: absorbance of X at a wavelength λ_i

C_X: concentration of X

b_Xi: slope of the regression line

a_Xi: intercept of the regression line

(i) Tri-Linear Regression Calibration – TLRC:

The absorbances of a ternary mixture (X, Y, Z) measured at 3 different wavelengths (λ_i = 1, 2 và 3), can be described by the three following equations:

Where:

A_mix1, A_mix2, and A_mix3 are the absorbances of a ternary mixture (X, Y and Z) measured at 3 different wavelengths;

b_{X1,2 và 3}, b_Y1,2 and ₃ b_{Z1,2 và 3} are the slopes of the regression lines;

a_XYZ1, a_XYZ2, and a_XYZ3 are the sums of the intercepts of the regression lines (a_XYZ1 = a_X1+ a_Y1 + a_Z1, a_XYZ2 = a_X2 + a_Y2 + a_Z2 and a_XYZ3 = a_X3 + a_Y3 + a_Z3).

Equation (2) can be expressed in the form of a matrix:

If absorbance matrices A_{mix1, 2 and 3} and intercept matrices a_{XYZ1, 2 and 3} have the same size, Equation (3) can be changed to

Or simply:

Matrix b (corresponding to the slopes of the linear regression lines) is also known as matrix K:

Thus, the concentration of X, Y and Z in a ternary mixture can be determined as follows:

(ii) Multi-Linear Regression Calibration - MLRC:

If the absorption of a ternary mixture (X, Y and Z) is recorded at n different wavelengths (λ_i = 1, 2, ..., n), Equations (3 – 5) will become Equations (8 – 10)

Or simply:

In the compact form:

As above-mentioned, matrix K is:

Herein, the determination of the concentration of X, Y and Z in a ternary mixture is as follows:

(iii) Classical Least Squares - CLS:

The absorbance of a ternary mixture (X, Y and Z) is:

A_j = a_1jc₁ + a_2jc₂ + a_3jc₃                                                     (13)

Where:

A_j is the absorbance at the jth wavelength

a_nj is the absorptivity of the nth pure material at wavelength j

c_n is the concentration of the nth pure material

If the spectra of those pure components are known, then a least squares computation can be set up this way.

Let’s define define E_j as the error in the determination of the value of A_j for the jth wavelength:

The sum-squares error is:

Take the derivative with respect to concentration c_n, we get:

Set all these derivatives to zero and divide each equation by 2 gives us:

Or we can have the following equation after rearrangement:

Alternatively, all Equations (14-18) can be converted into a matrix expression as follows:

[c] [aa^T] = [a] [A^T]

which, when solved for the concentration [c] is:

[c] = [A] [a]^T [aa^T] ^–1

4. RESULTS AND DISCUSSION:

UV spectroscopic data for standards, mixture of standards and samples:

According to UV spectroscopic data, it is obvious that the spectral additivity was obeyed for these three compounds over the wavelength range of 215 – 290nm (e.g Figure 1). It is impossible to quantify any compound by classical UV spectroscopy due to severe spectral overlapping. Thus, in our study multivare analysis [Tri-Linear Regression Calibration – TLRC, Multi-Linear Regression Calibration – MLRC, and Classical Least Squares - CLS] was employed for the UV-spectroscopic quantification of AC, CA and IB in tablets and capsules.

Figure 1: UV spectral profile for AC, IB, CA, their corresponding standard mixture at working concentrations and additivity check

Investigation on multivariate analysis of UV spectra of ternary mixtures:

For TLRC, three working wavelengths were chosen as 215, 222.3 và 236.5 nm. At these wavelengths, the absolute value of K matrix (equivalent to the slope of the linear regression line) obtained maximized (Table 1). In this case, the content of each compound in ternary mixtures could be determined by Cramer’s rule¹⁹.

Table 1: Linear regression equations for AC, IB and CA at three working wavelengths

Wavelength	Linear regression equation
	AC	IB	CA
215 nm	y = 0.0310x + 0.0149 R^2 = 0.9998	y = 0.0317x + 0.0687 R^2 = 0.9985	y = 0.0753x + 0.0011 R^2 = 0.9991
222.3 nm	y = 0.0359x + 0.0253 R^2 = 0.9991	y = 0.0329x + 0.1250 R^2 = 0.9977	y = 0.0385x – 0.0010 R^2 = 0.9989
236.5 nm	y = 0.0501x + 0.1772 R^2 = 0.9994	y = 0.0051x + 0.0102 R^2 = 0.9979	y = 0.0229x – 0.0013 R^2 = 0.9981

For CLS- and MLRC-assisted UV spectroscopic assay, the concentration matrix X consists of 21 standard solutions (centered at the solution containing AC 28 mg/L, CA 2 mg/L and IB 20 mg/L). In the wavelength range of 215 - 290 nm, 76 wavelengths (∆λ = 1 nm) and 16 wavelengths (∆λ = 5 nm) were respectively selected for CLS and MLRC algorithms.

Table 2 presents the assay results for standard mixtures by UV spectroscopy coupled with multivariate analysis, clearly indicating that the TLRC method was not suitable for spectroscopic analysis of AC, IB and CA in ternary mixtures (% recovery in the range 85 – 115%). For the CLS method, the accuracy of the assay could be better by using a wider working range of wavelengths. In contrast, for MLRC method the number of working wavelengths must be smaller than that of standard solutions in the concentration matrix under study.

Table 2: Assay results of standard mixtures by UV spectroscopy coupled with multivariate analysis

Mixtures		TLRC		CLS			MLRC
	AC	IB	CA	AC	IB	CA	AC	IB	CA
CA20 IB24 CA2	119.0	101.6	83.6	99.9	99.5	100.1	99.9	100.1	100.1
CA20 IB240 CA2.5	115.7	100.7	89.6	99.4	99.4	99.4	99.4	99.1	99.7
CA20 IB16 CA3	116.4	99.9	99.9	99.3	100	99.9	100.2	99.5	99.1
CA20 IB 16 CA2.5	109.7	99.4	77.4	100	99.9	100.5	99.5	100.1	100.1
CA20 IB 12 CA3.5	118.8	100.1	63.7	100	99.8	99.3	100.2	100.5	99.5
CA20 IB28 CA1.5	114.4	100.7	77.6	98.6	101.4	101.6	99.9	100.8	101.9
CA24 IB28 CA1	126.8	100.3	89.9	99.9	100.4	100.7	99.9	99.8	99.2
CA24 IB24 CA1.5	122.3	99.7	75.8	100	100.5	98.9	100.8	101.1	101.1
CA24 IB20 CA2	121.1	101.2	94.1	100	99.6	100	100.7	99.6	101.1
CA24 IB12 CA3	115.9	99.5	84.7	101	101.0	101.7	100.4	100.6	101
CA20 IB32 CA1	115.3	100.2	83.8	101	99.0	98.3	100.4	99.1	98.6
CA28 IB12 CA2.5	118.3	99.6	77.6	100	100.7	99.9	100.2	100.5	101.1
CA32.5 IB16 CA1.5	105.7	100	94.6	100	99.3	100	99.6	99.6	101.1
CA32.5 IB12 CA2	120.4	99.8	91.1	99.5	99.6	98.7	99.6	99.1	99.2
CA28 IB24 CA1	117.2	100.1	83.5	99.9	99.8	100	100.6	99.9	98.3
CA28 IB20 CA1.5	112.4	100.2	77.5	100	100.3	99.4	100.2	100.4	98.7
CA28 IB16 CA3	117.0	99.9	93.7	100	99.6	99.8	99.8	99.6	99.7
CA32.5 IB20 CA1	111.6	100.1	92.7	99.5	99.8	101.8	99.6	100.1	99.6
CA36 IB16 CA1	109.7	99.7	87.8	99.8	99.8	100.5	100	99.1	101.9
CA36 IB12 CA1.5	105.7	99.2	73.9	99.6	100.4	99.1	99.5	100.9	99.3
CA40 IB12 CA1	110.4	100.3	95.7	100	100.5	101.1	100.5	100.5	99.9
Mean	115.4	100.1	85.2	99.9	100	100	100	100	100

Assay and dissolution test of solid dosage forms by UV spectroscopy coupled with multivariate analysis:

Table 3 presents the assay results for some solid dosage forms commercially available by using UV spectroscopy coupled with multivariate analysis (CLS and MLRC), showing that all spectrophotometric methods were precise (RSD < 3%). The accuracy of these methods were assessed by standard addition technique (i.e. a standard amount equivalent to 20% of the nomial content of each compound). It was shown that all UV spectroscopic methods were accurate (% recovery in the range 98.3 – 101.9%). Statistical evaluation apparently indicated that there is no significant difference between these UV spectroscopic methods and UHPLC method reported with reference to precision (Bartlett test) and accuracy (one-way ANOVA test) at the confidence level 95%.

In this study, all the spectrophotometric methods were also applied for dissolution test using water as the dissolution medium. The dissolution data (e.g. Figures 2 and 3) implied that more than 80% of the content of each compound (compared to their label claim) in vitro dissolved after 30 and 45 min for capsules and tablets, respectively.

Table 3:  Assay results (% of lable claim) of CA, CA and IB in solid dosage forms by UV spectroscopy coupled with multivaritate analysis (CLS and MLRC) and UHPLC (mean ± SD, n = 3)

Sample	Bidi - Ipalvic			Ibu Acetalvic			Glotasic
	AC	IB	CA	AC	IB	CA	AC	IB	CA
MLRC	100.8±2.3	101.1±2.7	102.0±1.9	100.4±2.0	100.7±2.0	100.6±2.5	100.7±1.9	100.8±1.9	100.7±2.2
CLS	100.5±2.5	100.9±2.8	101.8±2.1	100.3±2.2	100.9±2.2	100.6±2.4	100.8±1.8	100.5±2.0	100.5±2.5
UHPLCC	100.4±2.6	101.0±2.7	101.9±2.5	100.3±2.3	100.8±2.1	100.7±2.4	100.7±1.9	100.7±2.1	100.6±2.4

Figure 2: Dissolution profiles for Glotasic tablet by UV spectroscopy coupled with CLS (n = 6, RSD < 5%)

Figure 3: Dissolution profiles for Bidi – ipalvic capsule by UV spectroscopy coupled with CLS (n = 6, RSD < 5%)

5. CONCLUSION:

It is concluded that the quantification of CA, IB and CA in pharmaceutical solid dosage forms (tablets and capsules) could be precisely and accurately done by using UV spectroscopy coupled with multivariate analysis (CLS and MLRC). All the UV spectrophotometric methods only require simple sample treatment, time-saving process and use enviromental preferable solvent (water). Consequently, they could be applicable to routine drug quality control (both assay and dissolution test), especially when high-performance liquid chromatography (HPLC) is not ready for a large number of samples.

6. REFERENCES:

1. Aronson JK. Meyler's Side Effects of Drugs (Sixteenth Edition). Paracetamol (acetaminophen) and combinations. The International Encyclopedia of Adverse Drug Reactions and Interactions. 2016; 474-493.

2. Rabia B and Nousheen A. An overview of clinical pharmacology of ibuprofen. Oman Medical Journal. 2010; 25(3): 155–161.

3. Hiroshi A, Kouichi M, Takao Y and Alan Crozier. Xanthine alkaloids: Occurrence, biosynthesis, and function in plants. Progress in the Chemistry of Organic Natural Products. 2017; 105: 1-88.

4. Forina M, Casolino MC and de la Pezuela Martinez C. Multivariate calibration: applications to pharmaceutical analysis. Journal of Pharmaceutical and Biomedical Analysis. 1998; 18: 21-33.

5. Saeed AM, Sultana N and Sher BS. Multivariate calibrations in UV spectrophotometric analysis. Pakistan Journal of Pharmaceutical Sciences. 2007; 20(2): 163-174.

6. Bhadresh VS, Ashutosh KP and Hashumati AR. Simultaneous determination of propranolol hydrochloride and hydrochlorothiazide in tablets formulation using spectrophotometric technique (simultaneous equation method). Asian Journal of Pharmaceutical Analysis. 2014; 5(1): 36-40.

7. Sathish KK, Adinarayana P, Vinod KM and Govinda RK. Development and validation of RP-HPLC method for simultaneous estimation of paracetamol and flupirtine maleate. Asian Journal of Pharmaceutical Analysis. 2015; 5(2): 105-111.

8. Bhadresh VS, Ashutosh KP and Hashumati AR. Simultaneous determination of propranolol hydrochloride and hydrochlorothiazide in tablets formulation using spectrophotometric technique (first derivative method). Asian Journal of Pharmaceutical Research. 2015; 5(1): 31-36.

9. Zeeshan HMA, Kumaraswamy G, Gandla L and Suthakaran R. Development and validation of RP-HPLC for simultaneous estimation of cefpodoxime proxetil and dicloxacillin sodium tablets. Asian Journal of Research in Pharmaceutical Sciences. 2014; 4(4): 155-159.

10. Daharwal SJ and Veena DS. Development of chemometric assisted methods for simultaneous estimation of ternary mixture of telmisartan hydrochloride, amlodipine besylate and hydrochlorothiazide. Asian Journal of Pharmacy and Technology. 2015; 5(2): 122-126.

11. Sayali SM, Sandeep SS, Santosh SC and Sanjay JK. development and validation of RP-HPLC method for simultaneous estimation of saxagliptin and dapagliflozin in tablets. Asian Journal of Pharmacy and Technology. 2018; 8(3): 145-148.

12. Prasanna RB and Reddy MS. RP-HPLC method for simultaneous estimation of paracetamol and ibuprofen in tablets. Asian Journal of Research in Chemistry 2009; 2(1): 70-72.

13. Sohan SC, Ranjana S, Sagar BW and Amol AK. Spectrophotometric methods for simultaneous estimation of dexibuprofen and paracetamol. Asian Journal of Research in Chemistry. 2009; 2(1): 30-33.

14. Godge RK, Damle MC, Pattan SR, Kendre PN, Lateef SN and Burange PJ. RP- HPLC method for simultaneous estimation of pseudoephidrine sulphate and desloratidine from bulk and tablets. Asian Journal of Research in Chemistry. 2009; 2(2): 139-142.

15. Duong TTA and Vu DH. Simultaneous determination of paracetamol and codeine phosphate in combined tablets by first-order derivative and ratio spectra first-order derivative UV spectrophotometry. Asian Journal of Research in Chemistry. 2009; 2(2): 143-147.

16. HPLC analysis of acetaminophen, caffeine, chlorpheniramine, ibuprofen, naproxen, phenylephrine, andpseudoephedrine on Ascentis® Express C18. Available from URL: https:// www.sigmaaldrich.com/technical-documents/articles/analytical-applications/hplc/hplc-analysis-of-acetaminophen-caffeine-chlorpheniramine-ibuprofen-naproxen-phenylephrine-and-pseudoephedrine-g005716.html

17. Jerome W Jr. Classical Least Squares, Part I: Mathematical Theory. 2010. Available from URL: https:// www.spectroscopyonline.com/view/classical-least-squares-part-i-mathematical-theory

18. O’Haver T. Curve fitting B: Multicomponent Spectroscopy. Available from URL: https://terpconnect.umd.edu/~toh/spectrum/ CurveFittingB.html

19. Kramer R. “Chemometric techniques in quantitative analysis”, Marcel Dekker Inc., New York. 1998, pp. 51-130.

20. Bhadresh VS, Ashutosh KP and Hashumati AR. Simultaneous determination of propranolol hydrochloride and hydrochlorothiazide in tablets formulation using spectrophotometric technique (simultaneous equation method). Asian Journal of Pharmaceutical Analysis. 2014; 5(1): 36-40.

21. Sathish KK, Adinarayana P, Vinod KM and Govinda RK. Development and validation of RP-HPLC method for simultaneous estimation of paracetamol and flupirtine maleate. Asian Journal of Pharmaceutical Analysis. 2015; 5(2): 105-111.

22. Bhadresh VS, Ashutosh KP and Hashumati AR. Simultaneous determination of propranolol hydrochloride and hydrochlorothiazide in tablets formulation using spectrophotometric technique (first derivative method). Asian Journal of Pharmaceutical Research. 2015; 5(1): 31-36.

23. Zeeshan HMA, Kumaraswamy G, Gandla L and Suthakaran R. Development and validation of RP-HPLC for simultaneous estimation of cefpodoxime proxetil and dicloxacillin sodium tablets. Asian Journal of Research in Pharmaceutical Sciences. 2014; 4(4): 155-159.

24.  Daharwal SJ and Veena DS. Development of chemometric assisted methods for simultaneous estimation of ternary mixture of telmisartan hydrochloride, amlodipine besylate and hydrochlorothiazide. Asian Journal of Pharmacy and Technology. 2015; 5(2): 122-126.

25. Sayali SM, Sandeep SS, Santosh SC and Sanjay JK. development and validation of RP-HPLC method for simultaneous estimation of saxagliptin and dapagliflozin in tablets. Asian Journal of Pharmacy and Technology. 2018; 8(3): 145-148.

26. Prasanna RB and Reddy MS. RP-HPLC method for simultaneous estimation of paracetamol and ibuprofen in tablets. Asian Journal of Research in Chemistry 2009; 2(1): 70-72.

27. Sohan SC, Ranjana S, Sagar BW and Amol AK. Spectrophotometric methods for simultaneous estimation of dexibuprofen and paracetamol. Asian Journal of Research in Chemistry. 2009; 2(1): 30-33.

28. Godge RK, Damle MC, Pattan SR, Kendre PN, Lateef SN and Burange PJ. RP- HPLC method for simultaneous estimation of pseudoephidrine sulphate and desloratidine from bulk and tablets. Asian Journal of Research in Chemistry. 2009; 2(2): 139-142.

29. Duong TTA and Vu DH. Simultaneous determination of paracetamol and codeine phosphate in combined tablets by first-order derivative and ratio spectra first-order derivative UV spectrophotometry. Asian Journal of Research in Chemistry. 2009; 2(2): 143-147.

Received on 25.12.2020        Revised on 09.02.2021

Accepted on 19.03.2021     ©Asian Pharma Press All Right Reserved

Asian Journal of Pharmaceutical Analysis. 2021; 11(2):127-132.

DOI: 10.52711/2231-5675.2021.00022