The quality of over-the-counter (OTC) pain relievers is important to ensure the safety of the marketed products in order to maintain the overall health care of patients. In this study, the multivariate curve resolution-alternating least squares (MCR-ALS) chemometric method was developed and validated for the resolution and quantification of the most commonly consumed OTC pain relievers (acetaminophen, acetylsalicylic acid, ibuprofen, naproxen, and caffeine) in commercial drug formulations. The analytical performance of the developed chemometric methods such as root mean square error of prediction, bias, standard error of prediction, relative error of prediction, and coefficients of determination was calculated for the developed model. The obtained results are linear with concentration in the range of 0.5–7
Pain relievers are among the most widely consumed medications worldwide. These medications are used for relieving inflammatory pain conditions [
Ibuprofen is a nonsteroidal anti-inflammatory (NSAID) drug. Its chemical name is 2-(4-isobutylphenyl)-propionic acid. Ibuprofen is commonly used for treating pain and inflammation in musculoskeletal disorders, especially in rheumatoid arthritis [
Because of the growing consumption of OTC pain relievers, controlling the quality of these medications is highly required in order to ensure their efficiency. Therefore, developing new reliable, fast, and cost-effective methods for analyzing such drugs is of paramount importance.
Implementation of green procedures in analytical laboratories is of paramount interest in order to minimize the negative environmental impacts [
To the best of our knowledge, no studies have been reported for simultaneous determination of paracetamol, aspirin, caffeine, naproxen, and ibuprofen in pharmaceutical formulations using multivariate calibration methods. The aim of this work is to develop a green, fast, low-cost analytical method utilizing MCR-ALS multivariate calibration for simultaneous determination of the cited OTC pain relievers in their single and coformulated products without any separation step with the aid of easily accessible instruments (e.g., UV spectrophotometer). The proposed method is economic, fast, eco-friendly, and amenable for routine analysis.
A UV-1800 Shimadzu double-beam spectrophotometer (Shimadzu, Kyoto, Japan) with 1 cm quartz cell was used. The scanning speed was maintained at 2800 nm·min−1, and the wavelength range was set from 200 to 400 nm with a bandwidth of 1 nm. Spectra were automatically obtained by Shimadzu UV-Probe 2.62 software. For performing MCR-ALS calculations, MCR-ALS GUI 2.0 software was used with Matlab 2015a [
The standards used in this study including paracetamol (PAR), acetylsalicylic acid (ASA), ibuprofen (IPF), naproxen (NPX), and caffeine (CAF) were purchased from Sigma-Aldrich (Steinheim, Germany) and certified to contain ≥98%. Methanol (CAS No. 67-56-1) was obtained from Merck (Darmstadt, Germany). Ultrapure water (18.2 MΩ) purified using the PURELAB Ultra water system (ELGA, High Wycombe, UK) was used for sample preparation.
Stock solutions of the studied drugs were prepared separately by dissolving 10 mg of each drug in 10 ml methanol to obtain a concentration of 1 mg·mL−1. All solutions were kept in dark at 4°C. Working standard solutions were freshly prepared by appropriate dilution in ultrapure water.
The calibration model (a set of twenty-five calibration solutions) was built using a five-factor, five-level design [
Concentration matrix (
Mixture | PAR | ASA | CAF | NPX | IPF |
---|---|---|---|---|---|
1 | 3.8 | 8.0 | 1.8 | 2.0 | 8.0 |
2 | 3.8 | 1.0 | 0.5 | 3.5 | 4.5 |
3 | 0.5 | 1.0 | 3.0 | 1.3 | 15.0 |
4 | 0.5 | 15.0 | 1.1 | 3.5 | 8.0 |
5 | 7.0 | 4.5 | 3.0 | 2.0 | 4.5 |
6 | 2.1 | 15.0 | 1.8 | 1.3 | 4.5 |
7 | 7.0 | 8.0 | 1.1 | 1.3 | 11.5 |
8 | 3.5 | 4.5 | 1.1 | 2.8 | 15.0 |
9 | 2.1 | 4.5 | 2.4 | 3.5 | 11.5 |
10 | 2.1 | 11.5 | 3.0 | 2.8 | 8.0 |
11 | 5.4 | 15.0 | 2.4 | 2.0 | 15.0 |
12 | 7.0 | 11.5 | 1.8 | 3.5 | 15.0 |
13 | 5.4 | 8.0 | 3.0 | 3.5 | 1.0 |
14 | 3.8 | 15.0 | 3.0 | 0.5 | 11.5 |
15 | 7.0 | 15.0 | 0.5 | 2.8 | 1.0 |
16 | 7.0 | 1.0 | 2.4 | 0.5 | 8.0 |
17 | 0.5 | 11.5 | 0.5 | 2.0 | 11.5 |
18 | 5.4 | 1.0 | 1.8 | 2.8 | 11.5 |
19 | 0.5 | 8.0 | 2.4 | 2.8 | 4.5 |
20 | 3.8 | 11.5 | 2.4 | 1.3 | 1.0 |
21 | 5.4 | 11.5 | 1.1 | 0.5 | 4.5 |
22 | 5.4 | 4.5 | 0.5 | 1.3 | 8.0 |
23 | 2.1 | 1.0 | 1.1 | 2.0 | 1.0 |
24 | 0.5 | 4.5 | 1.8 | 0.5 | 1.0 |
25 | 2.1 | 8.0 | 0.5 | 0.5 | 15.0 |
Concentration matrix (
Mixture | PAR | ASA | CAF | NPX | IPF |
---|---|---|---|---|---|
1 | 3.5 | 0.5 | 2.0 | 2.0 | 8.5 |
2 | 3.5 | 2.0 | 1.0 | 2.0 | 5.0 |
3 | 1.0 | 2.0 | 3.0 | 1.5 | 15.0 |
4 | 1.0 | 15.0 | 1.5 | 3.0 | 8.5 |
5 | 6.0 | 5.0 | 3.0 | 2.0 | 5.0 |
6 | 2.0 | 15.0 | 2.0 | 1.5 | 5.0 |
7 | 6.0 | 8.5 | 1.5 | 1.5 | 12.0 |
8 | 3.5 | 5.0 | 1.5 | 2.5 | 15.0 |
9 | 2.0 | 5.0 | 2.5 | 3.0 | 12.0 |
10 | 2.0 | 12.0 | 3.0 | 2.5 | 8.5 |
11 | 5.0 | 15.0 | 2.5 | 2.0 | 15.0 |
12 | 6.0 | 12.0 | 2.0 | 3.0 | 15.0 |
13 | 5.0 | 8.5 | 3.0 | 3.0 | 2.0 |
14 | 3.5 | 15.0 | 3.0 | 1.0 | 12.0 |
15 | 6.0 | 15.0 | 1.0 | 2.5 | 2.0 |
A real challenge of the presented work would be to quantify the studied drugs in commercial products presenting different compositions and interferences. In this study, ten commercial pharmaceutical tablets were collected from the local pharmacies (Eastern Province, Saudi Arabia) with highly variable compositions regarding the excipients and some containing other active ingredients such as codeine phosphate. The specified concentration level was described on the label of all the analyzed samples. The composition of the analyzed samples is as follows: (1) Fevadol Plus® tablets (SPIMACO) labeled to contain 500 mg of PAR, 30 mg of CAF, and 8 mg codeine per tablet, (2) Panadol Extra® tablets (SPIMACO) labeled to contain 500 mg of PAR and 65 mg of CAF per tablet, (3) Panadol® tablets (GSK) labeled to contain 500 mg of PAR per tablet, (4) Adol® tablets (Julphar) labeled to contain 500 mg of PAR per tablet, (5) Proxen® tablets (Grunenthal) labeled to contain 500 mg of NPX per tablet, (6) Omarfen® tablets (National Pharmaceutical Industries) labeled to contain 400 mg of IPF per tablet, (7) Disprin® tablets (Riyadh Pharma) labeled to contain 81 mg of ASA per tablet. (8) Fevadol® tablets (SPIMACO) labeled to contain 500 mg of PAR per tablet, (9) Riaproxe® tablets (Riyadh pharma) labeled to contain 500 mg of NPX per tablet, and (10) Brufen® tablets (Abbott) labeled to contain 500 mg of IPF per tablet.
Ten tablets of each pharmaceutical product were separately weighed and finely powdered. A portion of the powder equivalent to the average tablet weight of each product was separately dissolved in 50 mL methanol using ultrasonication for 15 min. Then, the solution was cooled and filtrated in a 100 mL volumetric flask using Whatman® filter papers. Finally, the volume was completed to 100 mL with methanol and suitable volumes of the stock solutions were mixed and diluted with ultrapure water to obtain different concentrations of the studied drugs at the specified linearity range mentioned above. The sample spectra were recorded using the same procedures described for the calibration and test sample sets.
MCR-ALS is a chemometric method that provides relevant information about the pure components by decomposing the bilinear data matrix. In MCR-ALS, the matrix of mixed signals (
For optimizing the MCR-ALS model and achieving solutions for
An external validation data set of 15 laboratory-prepared mixtures was used to evaluate the quantitative performance of the developed method by predicting the concentration of the studied drugs in a validation set which was not used for the development of the model (Table
The accuracy of the proposed chemometric method for the measurement of the studied drugs was tested by using the standard addition technique at 80%, 100%, and 120% of the test concentration. The study was performed by addition of known amounts of the studied drugs into a known concentration of the commercial pharmaceutical tablets. The resulting mixtures were analyzed, and the results obtained were compared with the expected results. The recovery of the exogenous amount added was calculated by the following equation:
Intraday precision (repeatability) and interday precision (reproducibility) of the developed method were evaluated by measuring three concentration levels that cover lower, middle, and upper limits of calibration curves.
Figure
UV absorption spectra of 1
The appropriate selection of the wavelength range is an important factor that strongly affects the quality of multivariate analysis [
Optimization of the MCR-ALS model is a very important step in order to achieve optimum performance of the model. Because of the variability in ratios of the studied drugs, each drug was calibrated individually to provide good results. Proper selection of the latent variables (LVs) number is important to achieve correct quantitation. The optimum number of LVs should be the smallest number that results in no significant difference between RMSECV of that factor and the next one [
For MCR-ALS, different constraints were optimized and satisfactory results were obtained when applying nonnegativity constraints for spectral and concentration matrices, in addition to a correlation constraint. The convergence criterion was set at 0.1%, and the maximum number of iterations was 50; however, no more than 8 iterations were required to achieve convergence in all tested samples. Table
Figures of merits of the calibration set for the developed MCR-ALS model.
Parameters | PAR | ASA | CAF | NPX | IPF |
---|---|---|---|---|---|
Calibration range | 0.5–7 | 1–15 | 0.5–3 | 0.5–3.5 | 1–15 |
Slope | 1.0000 | 0.9999 | 1.0000 | 1.0000 | 0.9999 |
Intercept | 8.75 × 10−15 | −1.96 × 10−14 | −2.02 × 10−13 | 2.51 × 10−3 | 2.13 × 10−3 |
RMSECV | 0.136 | 0.045 | 0.194 | 0.152 | 0.064 |
SEP | 0.153 | 0.032 | 0.170 | 0.149 | 0.063 |
Bias | 2.64 × 10−2 | 4.12 × 10−3 | −2.58 × 10−2 | 3.22 × 10−2 | 3.11 × 10−3 |
RE (%) | 1.32 | 0.94 | 1.65 | 1.28 | 1.15 |
Coefficient of determination ( |
0.9998 | 0.9995 | 0.9993 | 0.9998 | 0.9996 |
Plots of actual concentrations versus predicted values of the studied drugs for the calibration and validation sets using the MCR-ALS model. (a) IPF. (b) CAF. (c) ASA. (d) PAR. (e) NPX.
The developed model was validated by predicting the concentrations of the studied drugs in an external validation set (fifteen mixtures) which are not included in the model development. Different parameters such as RMSEP, SEP, RE (%), and
Validation parameters of the developed MCR-ALS model.
Parameter | PAR | ASA | CAF | NPX | IPF |
---|---|---|---|---|---|
Accuracya | 98.1 ± 1.28 | 99.5 ± 0.99 | 98.0 ± 1.12 | 98.9 ± 1.02 | 99.7 ± 1.62 |
Intraday precisionb | 0.93 | 0.85 | 1.23 | 1.02 | 1.23 |
Interday precisionc | 1.02 | 0.94 | 1.31 | 1.34 | 1.27 |
RMSEP | 0.251 | 0.099 | 0.284 | 0.232 | 0.160 |
SEP | 0.753 | 0.092 | 0.250 | 0.219 | 0.103 |
Bias | 3.54 × 10−2 | 4.99 × 10−3 | −3.98 × 10−2 | 4.29 × 10−2 | 4.11 × 10−3 |
RE (%) | 1.52 | 1.59 | 1.35 | 1.49 | 1.43 |
Correlation coefficient ( |
0.9993 | 0.9994 | 0.9991 | 0.9996 | 0.9992 |
aMean
Satisfactory validation results were obtained and showed the high predictive ability of the model. The plots of the predicted concentrations versus the actual concentrations for the validation set are also illustrated in Figure
The accuracy of the proposed method was evaluated by the standard addition method. The percent recoveries ranged from 98.0% to 99.7% with% RSDs not higher than 1.62% (Table
The proposed chemometric method was used to analyze the studied drugs in commercial tablets. Five replicate determinations were performed. Table
Determination of the studied drugs in commercial tablets by the MCR-ALS model by the proposed and the reported HPLC method.
MCR-ALS | HPLC | ||
---|---|---|---|
Sample 1 | PAR | ||
(Mean + SD) | 99.22 ± 1.11 | 99.86 ± 0.86 | |
|
0.86 | — | |
|
1.69 | — | |
CAF | |||
(Mean + SD) | 99.74 ± 0.72 | 99.8 ± 0.70 | |
|
0.50 | — | |
|
1.08 | — | |
|
|||
Sample 2 | PAR | ||
(Mean + SD) | 99.06 ± 0.96 | 99.6 ± 0.83 | |
|
1.61 | — | |
|
1.35 | — | |
CAF | |||
(Mean + SD) | 99.14 ± 1.28 | 99.0 ± 1.44 | |
|
0.09 | ||
|
0.79 | ||
|
|||
Sample 3 | PAR | ||
(Mean + SD) | 100.1 ± 1.11 | 100.5 ± 1.27 | |
|
0.80 | — | |
|
0.76 | — | |
|
|||
Sample 4 | PAR | ||
(Mean + SD) | 99.5 ± 1.45 | 100.5 ± 1.47 | |
|
1.07 | - | |
|
0.97 | - | |
|
|||
Sample 5 | NPX | ||
(Mean + SD) | 98.7 ± 1.50 | 99.9 ± 1.25 | |
|
2.67 | — | |
|
1.44 | — | |
|
|||
Sample 6 | IPF | ||
(Mean + SD) | 99.6 ± 1.77 | 99.1 ± 1.49 | |
|
0.59 | — | |
|
1.40 | — | |
|
|||
Sample 7 | ASA | ||
(Mean + SD) | 100.2 ± 1.85 | 99.9 ± 1.58 | |
|
0.34 | — | |
|
1.37 | — | |
|
|||
Sample 8 | PAR | ||
(Mean + SD) | 100.82 ± 1.34 | 100.4 ± 1.45 | |
|
0.35 | — | |
|
0.85 | — | |
|
|||
Sample 9 | NPX | ||
(Mean + SD) | 99.7 ± 1.55 | 100.4 ± 1.45 | |
|
0.93 | — | |
|
1.14 | — | |
|
|||
Sample 10 | IPF | ||
(Mean + SD) | 100.2 ± 1.68 | 99.5 ± 1.54 | |
|
0.56 | — | |
|
1.19 | — |
The reference HPLC published method used C18 (250 × 4.6 mm, 5.0
An eco-friendly, fast, and accurate MCR-ALS chemometric method for the quantification of the most widely consumed OTC pain relievers has been developed and validated. The proposed model proved to be a green, nondestructive, and low-cost alternative to chromatographic techniques for the determination of the studied drugs in pure and commercial pharmaceutical formulations. The developed MCR-ALS method can be used as a green alternative for the analysis of the studied drugs without sample preparation steps and expensive solvents, especially in developing countries where resources are limited.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that they have no conflicts of interest to disclose.
The authors gratefully acknowledge the College of Clinical Pharmacy, Immam Abdel Rahman Bin Faisal University, for providing research facilities.