Journal of Chromatographic Science, Vol.

47, November/December 2009

Determination of Paracetamol and Tramadol Hydrochloride

in Pharmaceutical Mixture Using HPLC and GC–MS
Tarek Belal1, Tamer Awad2, and C. Randall Clark2*
1Department of Pharmaceutical Analytical Chemistry, Faculty of Pharmacy, Alexandria University, Alexandria 21521, Egypt and 2Department
of Pharmacal Sciences, Harrison School of Pharmacy, Auburn University, Auburn, AL 36849, USA

activity and does not produce gastrointestinal damage (1,2).

Abstract
Unlike opiates it is almost ineffective in intense pain and has no
depressant effect on respiration (1,2). The major advantage of PR
Two simple, rapid, and selective analytical procedures were
developed for the simultaneous determination of paracetamol (PR)
lies in its relative lack of serious side effects (1,2). Although PR has
and tramadol hydrochloride (TR) in a binary mixture using high- been used clinically for more than a century, its mode of action has
performance liquid chromatography with UV detection been a mystery until recently when it was discovered that the anal-
(HPLC–UV) and gas chromatography with mass spectrometry gesic effect of PR is due to the indirect activation on cannabinoid
(GC–MS) techniques. HPLC resolved the two compounds on a CB1 receptors (1,2). Tramadol hydrochloride (TR), (1RS,2RS)-2-
Hypurity Advance column using a mobile phase consisting of [(Dimethyl-amino)methyl]-1-(3-methoxy-phenyl) cyclohexanol
phosphate buffer pH 6.3 and acetonitrile (90:10, v/v). PR and TR hydrochloride, is a centrally acting analgesic consisting of two
were detected by their UV absorption at 220 nm. GC–MS involved enantiomers, both of which contribute to the analgesic activity via
separation of the two compounds using 100% different mechanisms (3). (+)-Tramadol is an agonist of the µ-
dimethylpolysiloxane (Rtx-1) column with temperature opioid receptor, and it inhibits serotonin reuptake whereas (–)-tra-
programming. The EI mass spectrum of PR was characterized by
madol inhibits norepinephrine reuptake, enhancing inhibitory
[M]+ at 151 and a base peak at m/z 109 while TR mass spectrum
was characterized by [M]+ at 263 and a base peak at m/z 58.
effects on pain transmission in the spinal cord (3). TR is an effec-
Quantification of the analytes in both methods was based on tive and well-tolerated agent that reduces pain resulting from
measuring the peak areas. The reliability and analytical trauma, renal or biliary colic, and labor, and also for the manage-
performance of the proposed methods including linearity, ranges, ment of chronic pain of malignant or nonmalignant origin, par-
precision, accuracy, detection, and quantification limits were ticularly neuropathic pain (3). TR appears to produce less
statistically validated. Calibration curves were linear over the constipation and dependence than equi-analgesic doses of strong
range 10–400 µg/mL for both PR and TR using the HPLC method opioids (3). The analgesic efficacy of tramadol can further be
and over the ranges of 75–500 and 25–350 µg/mL for PR and TR, improved by combination with a non-opioid analgesic (3).
respectively, using the GC–MS method. The proposed methods Structures of PR and TR are shown in Figure 1.
were successfully applied for the determination of the two The United States Pharmacopeia suggests a spectrophoto-
compounds in laboratory-prepared mixtures and in commercially
metric Amax procedure for the analysis of PR powder and high-
available tablet formulation. No interference peaks were observed
from common pharmaceutical adjuvants. The results compared
performance liquid chromatography (HPLC) for all its
favorably with those obtained by a derivative spectrophotometric preparations (4). The British Pharmacopoeia shows various ana-
method. lytical procedures for the assay of PR in bulk powder and dosage

Introduction

Paracetamol (PR) or acetaminophen, N-(4-hydroxyphenyl)-

acetamide, is one of the most popular and widely used drugs for
the treatment of pain and fever. It occupies a unique position
among analgesic drugs. According to a recent update of the
American College of Rheumatology (ACR) guidelines for
osteoarthrosis, PR remains a first-line therapy because of its cost,
efficacy, and safety profiles (1). Unlike non-steroidal anti-inflam-
matory drugs, it is considered to have no anti-inflammatory
Figure 1. Structures of paracetamol (A) and tramadol (B).
* Author to whom correspondence should be sent: Email: [email protected].

Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission. 849
 Journal of Chromatographic Science, Vol. 47, November/December 2009

forms including cerium sulphate titrimetry, Amax spectropho- their review article demonstrated over 300 reports of different
tometry, and HPLC. For TR, potentiometric non-aqueous titra- optical, electrochemical, and chromatographic techniques used
tion is applied for the assay of the bulk powder whereas HPLC is for quantification of PR in pharmaceutical formulations and bio-
used for the capsules (5). logical samples in the last four decades. TR was determined in
Several analytical procedures have been reported for the deter- different matrices using a variety of analytical techniques
mination of the two compounds. Espinosa Bosch et al. (6) in including HPLC (7,8), gas chromatography with mass spectrom-
etry (GC–MS) (9,10), thin layer chromatography (TLC)–densito-
metry (11), capillary electrophoresis (12,13), adsorptive
stripping voltammetry (14), square-wave voltammetry and flow
injection analysis system with amperometric detection (15),
selective PVC membrane electrodes (16), spectrofluorimetry
(17), and spectrophotometry (17,18).
The simultaneous determination of the two drugs has been
mV

reported in a few publications. They were estimated in human

plasma samples using liquid chromatography (LC)–MS (19,20).
In tablets, they were determined using spectrophotometric
(21,22) and reverse-phase HPLC methods using C18 columns
(23–25). No attempts have yet been made to determine this drug
mixture by GC. In the present work, two chromatographic pro-
cedures are proposed and validated for the simultaneous deter-
Figure 2. HPLC chromatogram of 20-µL injection of mixture of 50 µg/mL PR mination of PR and TR in their bulk form and in tablet form. The
(A) and 100 µg/mL TR (B). HPLC method involved the use of different stationary phase
other than the traditional C18 column and a different mobile
phase composition. The second method involved
the application of GC–MS.

Experimental

Instrumentation
The HPLC system comprised of an LC-10AS
Shimadzu liquid chromatograph (Kyoto, Japan)
with SIL-10A auto-injector and SPD-10AV UV-vis-
ible detector. EZstart 7.4 chromatography software
Figure 3. GC chromatogram of 1-µL injection of mixture of 400 µg/mL PR (A) and 50 µg/mL TR (B). (Shimadzu) was used for processing of data and
peaks integration. The column used was Hypurity
Advance column (5 µm, 150 × 4.6 mm i.d., Thermo-
Hypersil Keystone, Bellefonte, PA). The flow rate was 1.0
mL/min, and the injection volume was 20 µL. The detector was
set at λ = 220 nm, and all determinations were performed at
room temperature.
The GC–MS study was conducted using an Agilent Technolo-
gies (Santa Clara, CA) 7890A gas chromatograph and an Agilent
7683B auto injector coupled with a 5975C VL Agilent mass selec-
tive detector. The injection volume was 1 µL, and the mass spec-
tral scan rate was 2.86 scans per second. The GC was operated in
splitless mode with a carrier gas (helium grade 5), flow rate at 0.7
mL/min, and a column head pressure of 10 psi. The mass spec-
trometer was operated on the electron impact (EI) mode using
an ionization voltage of 70 eV and a source temperature of
230°C. The GC injector was maintained at 250°C and the transfer
line at 280°C. The temperature program used consisted of an ini-
tial temperature hold at 70°C for 1 min, ramped up to 250°C at a
rate of 30°C/min followed by a hold at 250°C for 20 min. The
mass spectra reported were obtained by background subtraction
and are the average of at least five scans. The chromatographic
Figure 4. Mass spectra of paracetamol (A) and tramadol (B).
separations (and collection of retention data) were carried out on

850
Journal of Chromatographic Science, Vol. 47, November/December 2009

a 30 m × 0.25 mm-i.d. column coated with 0.25 µm 100% Lawn, NJ), HPLC-grade o-phosphoric acid 85% (Fisher
dimethyl polysiloxane (Rtx-1) purchased from Restek Scientific, Fair Lawn, NJ), and high purity distilled water were
Corporation (Bellefonte, PA). used in the study.

Drugs and reagents General procedures

PR (minimum 99.0%) (Sigma Aldrich, St. Louis, MO) and TR HPLC
hydrochloride (≥ 99.0%, HPLC-grade) (Fluka BioChemika, Phosphate buffer was prepared by mixing aqueous 0.05 M
Buchs, Switzerland) were used in the study. Tramol-Plus tablets phosphoric acid solution with 0.2 M sodium hydroxide solution
(Zypher, Laborate Pharmaceuticals India Ltd, India, B.N. ZTPT to reach pH 6.3. The mobile phase was prepared by mixing phos-
702) labeled to contain 50 mg TR and 325 mg PR per tablets were phate buffer pH 6.3 and acetonitrile in the ratio of 90:10 (v/v),
purchased from a local commercial source. HPLC-grade solvents then it was filtered and degassed.
including methanol and acetonitrile (Fisher Scientific, Fair PR and TR stock solutions (500 µg/mL) were prepared in
methanol. The working solutions were prepared by dilution of
the stock solutions with the mobile phase to reach a concentra-
tion range 10–400 µg/mL for both PR and TR. Injections were
made for each concentration and chromatographed under the
previously described LC conditions.

GC–MS
PR and TR stock solutions (1000 µg/mL) were prepared in
methanol. The working solutions were prepared by dilution of
the stock solutions with methanol to reach concentration ranges
75–500 and 25–350 µg/mL for PR and TR, respectively.
Injections were made for each concentration and chro-
Figure 5. Fragmentation pattern of paracetamol. matographed under the previously described GC conditions.
The peak areas obtained from both HPLC and
GC methods were plotted against the corre-
sponding concentrations to obtain the calibration
graphs.

Assay of tablets
A total of 10 tablets were weighed and finely
powdered. Methanol (60 mL) was added to a quan-
tity of the powdered tablets equivalent to 260 mg
PR and 40 mg TR, stirred for 10 min, then filtered
into a 100-mL calibrated flask. The residue was
washed with 2 × 10 mL methanol, and washings
were added to the filtrate and diluted to final
volume with methanol. Aliquots of the tablet solu-
tion (prepared in methanol) were diluted with
Figure 6. Mass spectrum of d3-paracetamol.
either the HPLC mobile phase (for HPLC mea-
surement) or methanol (for GC measurement) to
obtain final concentrations within the previously mentioned
ranges and then treated as under the procedures for HPLC and
GC–MS methods.

Results and Discussion

HPLC
A liquid chromatography method coupled with UV detection
was developed to provide a suitable procedure for the rapid and
reliable quality control analysis of PR and TR in their combined
pharmaceutical preparation. Several reverse-phase stationary
phases were tried including C18, C8, and phenylhexyl columns.
Figure 7. Fragmentation pattern of tramadol. Although these columns gave satisfactory resolution of the two
analytes, PR eluted at the void volume, regardless of the composi-

851
 Journal of Chromatographic Science, Vol. 47, November/December 2009

tion of the mobile phases. This can be attributed to the weak reten- polar capillary GC columns were evaluated in an effort to find the
tion of the relatively polar PR on these non-polar stationary appropriate stationary phase providing the optimum separation.
phases. The Hypurity advance column with its embedded polar The 100% dimethyl polysiloxane (Rtx-1) column gave better
character, which contains a polar amide group embedded within a peak shapes and resolution of the two analytes within shorter
C8 chain, gave better retention for PR with good resolution of the analysis time compared to other non-polar columns, such as the
two analytes. Consequently, it became the column of choice for 5% diphenyl–95% dimethyl polysiloxane (Rtx-5) and the 50%
this study. phenyl–50% methyl polysiloxane (Rxi-50) columns. Also, several
Several mobile phases were tried using various proportions of temperature programs were evaluated, and a program showing
several solvents and buffers at different pH values. The best reso- the best resolution in a reasonable analysis time was selected.
lution and analysis time was obtained through isocratic elution Programs with higher initial temperatures or higher ramp rates
using a mobile phase consisting of 10% (by volume) acetonitrile led to poor resolution whereas lower temperature ramps
in phosphate buffer pH 6.3 (Figure 2). Methanol produced broad resulted in longer retention times and excessive peak tailing. PR
asymmetric peak with TR, hence it was not used as a component eluted first (tR = 10.19 ± 0.077 min) followed by TR (tR = 12.27 ±
of the mobile phase. Buffer pH was evaluated in the range from 0.022 min), and the resolution (Rs) was found to be 7.12. An
2.0 to 7.0, and best resolution and peak shapes were achieved at example chromatogram is shown in Figure 3.
pH ranging from 6.0 to 6.5. Lower acidic pH values led to weak Mass spectrum of PR (Figure 4A) is characterized by a base
resolution whereas increasing pH to more than 6.5 resulted in peak at m/z 109 formed by hydrogen transfer from the methyl
tailing and decrease in the sharpness of the peaks. Quantification group of the acetyl moiety to the ionized nitrogen followed by
was made with UV detection based on measuring the peak area. alpha cleavage. Subsequent rearrangement followed by the loss
The UV detector was set at 220 nm, which was found to be of the formaldehyde radical results in the formation of the more
optimum in measuring the two analytes. stable conjugated cyclopentadienylidene ammonium cation at
The previously described chromatographic conditions showed m/z 80. Ionization of the amide oxygen followed by alpha
symmetric peaks and adequate resolution (Rs = 4.67) between PR cleavage gave the CH3CO cation at m/z 43. Structures of the PR
(tR = 3.65 ± 0.026 min) and TR (tR = 5.81 ± 0.043 min) as shown fragment ions are shown in Figure 5. Further proof of the sug-
in Figure 2. The capacity factors (k’) were found to be equal to gested pathways was possible after the preparation of d3-PR from
0.68 and 1.70 for PR and TR, respectively, and the selectivity (α) condensation of 4-aminophenol and d6-acetic anhydride. The
was 2.50. mass spectrum in Figure 6 shows the analogous peaks at m/z
110, 81, and 46, respectively.
GC–MS The mass spectrum of TR (Figure 4B) is characterized by ions
The second part of this study involved the determination of PR at m/z 58 (base peak), m/z 188, and other ions of low relative
and TR in a binary mixture using GC–MS. Previous studies have abundance. Typical alpha cleavage between the tertiary amine
shown the GC–MS determination of PR and TR individually in and the cyclohexanol ring results in the imine ion at m/z 58 (27).
different matrices (9,10,26). However, the simultaneous deter- Ionization of the π-electron in the benzene ring followed by rear-
mination of these two analytes in pharmaceutical binary mixture rangements by the loss of both the hydroxyl group and the ter-
using GC–MS has not been previously reported. The primary tiary amine radicals results in the formation of the methoxy
goal of this part of the study is to provide a direct, fast, and reli- phenyl cyclohexene at m/z 188. Structures of the fragment ions
able method for such determinations. In this regard, some non- of TR are shown in Figure 7.

Table II. Precision and Accuracy for the Determination of PR

Table I. Analytical Parameters for the Determination of PR and and TR Using the Proposed Chromatographic Methods
TR Using the Proposed Chromatographic Methods
Nominal Found ± SD*
HPLC GC–MS
Analyte value (µg/mL) (µg/mL) RSD(%)† Er(%)‡
Parameter PR Tramadol PR Tramadol
HPLC:
Concentration 10–400 10–400 75–500 25–350 PR 50 50.78 ± 0.70 1.38 1.56
range (µg/mL) 100 99.62 ± 1.12 1.12 –0.38
Intercept (a) 63.0 × 103 36.5 × 103 –17.4 × 104 –10.9 × 104 200 201.46 ± 1.78 0.88 0.73
Sa* 98.1 × 103 41.2 × 103 16.6 × 104 22.4 × 104 Tramadol 50 50.05 ± 0.60 1.20 0.10
Slope (b) 36132 24096 25189 50843 100 100.12 ± 1.59 1.59 0.12
Sb† 454 191 415 900 200 199.58 ± 2.30 1.15 –0.21
RSD% of the slope 1.26 0.79 1.65 1.77
Correlation 0.99960 0.99984 0.99851 0.99900 GC–MS:
coefficient (r) PR 100 101.25 ± 1.54 1.52 1.25
F‡ 6314 15845 1677 2581 200 196.38 ± 3.12 1.59 –1.81
Significance F 59.8 × 10-8 60.0 × 10-9 16.4 × 10-6 55.9 × 10-7 300 297.57 ± 4.38 1.47 –0.81
LOD§ (µg/mL) 0.64 1.36 20.00 6.00 Tramadol 100 99.56 ± 1.56 1.57 –0.44
LOQ** (µg/mL) 2.13 4.53 66.70 20.00 200 197.02 ± 2.58 1.31 –1.49
300 294.75 ± 4.38 1.49 –1.75
* Standard deviation of the intercept. † Standard deviation of the slope.
‡ F equals the mean of squares due to regression divided by the mean of squares about
* Mean ± standard deviation for five determinations.
regression (due to residuals). § Limit of detection. ** Limit of quantification. † % Relative standard deviation. ‡ % Relative error.

852
Journal of Chromatographic Science, Vol. 47, November/December 2009

Analytical performance of the Detection and quantification limits

proposed methods According to the U.S. Pharmacopeia recommendations (4),
Concentration ranges and calibration graphs limit of detection is defined as the concentration that has a
For both methods, the linearity of the detector response for signal-to-noise ratio of 3:1 whereas for limit of quantification the
the determination of PR and TR was evaluated by analyzing a ratio considered is 10:1. These values were calculated and pre-
series of different concentrations of each compound. Seven con- sented in Table I.
centrations were chosen with triplicate injections for each con-
centration; this approach provided information on the variation Precision and accuracy
in peak areas between samples of the same concentration. The The precision and accuracy for each method were examined at
linear regression equations were generated by least squares three concentration levels for the analyte by five replicate deter-
treatment of the calibration data. Table I presents the perfor- minations for each concentration. The percentage relative stan-
mance data and statistical parameters for the proposed methods dard deviation (RSD %) and the percentage relative error (Er %)
including linear regression equations, concentration ranges, did not exceed 2%, proving the high repeatability and accuracy of
correlation coefficients, standard deviations of the intercept (Sa), the developed methods (Table II).
and the slope (Sb). The analysis of variance (ANOVA) test for the
regression lines reveals that for equal degrees of freedom, an Selectivity
increase in the variance ratio (F-values) means an increase in the The selectivity of the proposed methods was tested by
mean of squares due to regression and decrease in the mean of preparing different mixtures of PR and TR within the concentra-
squares due to residuals. The greater the mean of squares due to tion ranges mentioned in Table I. These mixtures were of dif-
regression, the steeper the regression line. The smaller the mean ferent ratios both above and below the normal ratio expected in
of squares due to residuals, the less the scatter of experimental the tablets. The laboratory-prepared mixtures were analyzed
points around the regression line. Consequently, regression lines according to the previously mentioned HPLC and GC proce-
with high F values (low significance F) are much better than dures. The recovery values, RSD %, and Er % shown in Tables III
those with lower ones. Good regression lines show high values and IV were satisfactory, thus validating the selectivity, precision,
for both (r) and (F) statistical parameters (28). and accuracy of the developed methods.
Robustness
Table III. Determination of PR and TR Laboratory-made Mixtures Using the Robustness was examined by evaluating the
Proposed HPLC Method influence of small variations in different experi-
mental conditions such as working wavelength
Nominal
value (µg/mL) Found ± SD* (µg/mL) RSD(%)† Er(%)‡
(± 2 nm), mobile phase pH (± 0.2 pH units), and
organic strength (± 5%) for the HPLC method and
PR TR PR TR PR TR PR TR
temperature program ramp (± 2°C) for the
200 20 197.52 ± 1.98 20.55 ± 0.23 1.00 1.12 –1.24 2.75 GC–MS method. These variations did not have
200 25 198.40 ± 2.46 24.73 ± 0.27 1.24 1.09 –0.80 –1.08 significant effect on the measured responses or
200 40 198.54 ± 2.00 40.88 ± 0.55 1.01 1.35 –0.73 2.20
150 50 151.79 ± 1.91 49.81 ± 0.51 1.26 1.02 1.19 –0.38
the chromatographic resolution.
100 50 98.42 ± 1.17 50.16 ± 0.74 1.19 1.48 –1.58 0.32
100 100 100.59 ± 1.55 101.29 ± 0.92 1.54 0.91 0.59 1.29 Stability
50 100 48.94 ± 0.50 100.98 ± 1.22 1.02 1.21 –2.12 0.98 The stability of the PR and TR methanolic
50 150 48.33 ± 0.60 152.24 ± 1.44 1.24 0.95 –3.34 1.49 working solutions (for GC) was tested, and they
* Mean ± SD for 5 determinations. † % Relative standard deviation. ‡ % Relative error.
were found to be stable for at least four days at
room temperature. Also, the stability of the
working solutions in the mobile phase (for HPLC)
was verified, and no chromatographic changes
Table IV. Determination of PR and TR Laboratory-made Mixtures Using the were observed within 24 h at room temperature.
Proposed GC–MS Method
Analysis of pharmaceutical preparation
Nominal
The developed chromatographic methods were
value (µg/mL) Found ± SD* (µg/mL) RSD(%)† Er(%)‡
applied for the assay of the two drugs in their
PR TR PR TR PR TR PR TR combined pharmaceutical formulation (Tramol-
400 40 391.32 ± 3.52 40.61 ± 0.76 0.90 1.87 –2.17 1.53 Plus tablets). Table V shows the results obtained
400 50 403.04 ± 4.60 51.08 ± 1.09 1.14 2.13 0.76 2.16 for the proposed methods as well as the reference
250 50 255.10 ± 3.18 50.18 ± 0.90 1.25 1.79 2.04 0.36 derivative spectrophotometric method (21). The
300 100 292.41 ± 2.91 101.66 ± 0.97 1.00 0.95 –2.53 1.66
200 100 196.12 ± 3.12 97.89 ± 1.12 1.59 1.14 –1.94 –2.11 assay results showed good precision and accuracy
200 200 199.04 ± 3.54 201.06 ± 3.28 1.78 1.63 –0.48 0.53 as indicated from % recovery, SD, and RSD (%)
100 200 102.00 ± 1.87 195.60 ± 1.94 1.83 0.99 2.00 –2.20 values. No interfering peaks were observed in the
100 300 101.12 ± 1.68 304.65 ± 3.90 1.66 1.28 1.12 1.55
HPLC or GC chromatograms of the tablets.
* Mean ± SD for 5 determinations. † % Relative standard deviation. ‡ % Relative error. Results obtained by the developed methods were
statistically compared with those of the previously

853
 Journal of Chromatographic Science, Vol. 47, November/December 2009

Pharmacokinet. 43(13): 879–923 (2004).

Table V. Application of the Proposed Chromatographic 4. The United States Pharmacopeia 30th edition, The National Formulary, 25th edi-
Methods in the Analysis of PR and TR Commercial Tablets tion “The Official Compendia of Standards”, United States Pharmacopeial
Convention, Inc., Asian Edition, Washington, D.C. (2007).
5. The British Pharmacopoeia, Volumes II and III, Her Majesty’s Stationery Office,
PR HPLC GC–MS Reference method London, UK (2007).
6. M. Espinosa Bosch, A.J. Ruiz Sánchez, F. Sánchez Rojas, and C. Bosch Ojeda.
% Recovery ± SD* 96.27 ± 1.02 96.59 ± 1.21 95.15 ± 1.17 Determination of paracetamol: Historical evolution. J. Pharm. Biomed. Anal. 42:
RSD(%) 1.06 1.25 1.23 291–321 (2006).
7. W.F. Kartinasari, T. Palupi, and G. Indrayanto. HPLC determination and validation
ANOVA of tramadol hydrochloride in capsules. J. Liq. Chromatogr. Relat. Technol. 27(4):
Source of Sum of Degrees of Mean of 737–744 (2004).
Variation Squares Freedom Squares F† 8. M.R. Rouini, Y.H. Ardakani, F. Soltani, H.Y. Aboul-Enein, and A. Foroumadi.
Development and validation of a rapid HPLC method for simultaneous determi-
Between Groups 5.682 2 2.841 2.198 nation of tramadol, and its two main metabolites in human plasma.
Within Groups 15.510 12 1.293 J. Chromatogr. B 830(2): 207–211 (2006).
Total Variation 21.193 14 9. K.A. Hadidi, J.K. Almasad, T. Al-Nsour, and S. Abu-Ragheib. Determination of tra-
madol in hair using solid phase extraction and GC–MS. Forensic Sci. Int. 135:
TR HPLC GC–MS Reference method 129–136 (2003).
10. Y.F. Sha, S. Shen, and G.L. Duan. Rapid determination of tramadol in human
% Recovery ± SD* 95.73 ± 0.95 95.37 ± 1.09 94.84 ± 1.01 plasma by headspace solid-phase microextraction and capillary gas chromatog-
RSD(%) 0.99 1.14 1.07 raphy–mass spectrometry. J. Pharm. Biomed. Anal. 37: 143–147 (2005).
11. J. Krzek and M. Starek. Quality assessment for tramadol in pharmaceutical prepa-
ANOVA rations with thin layer chromatography and densitometry. Biomed. Chromatogr.
Source of Sum of Degrees of Mean of 18(8): 589–599 (2004).
Variation Squares Freedom Squares F† 12. U.B. Soetebeer, M.O. Schierenberg, H. Schulz, G. Gruenefeld, P. Andresen, and
G. Blaschke. Assay of tramadol in urine by capillary electrophoresis using laser-
Between Groups 1.998 2 0.999 0.965 induced native fluorescence detection. J. Chromatogr. B 745(2): 271–278 (2000).
Within Groups 12.428 12 1.036 13. J.G. Li and H.X. Ju. Simultaneous determination of ethamsylate, tramadol and
Total 14.427 14 lidocaine in human urine by capillary electrophoresis with electrochemilumines-
cence detection. Electrophoresis 27(17): 3467–3474 (2006).
* Mean ± standard deviation for five determinations. 14. P. Norouzi, R. Dinarvand, M.R. Ganjali, and A.S.E. Meibodi. Application of
‡ The theoretical value for F equals 3.885 at P = 0.05. Adsorptive Stripping Voltammetry for the Nano-Level Detection of Tramadol in
Biological Fluids and Tablets Using Fast Fourier Transform Continuous Cyclic
Voltammetry at an Au Microelectrode in a Flowing System. Anal. Lett. 40(11):
published derivative spectrophotometric method using single 2252–2270 (2007).
15. E.M.P.J. Garrido, J.M.P.J. Garrido, F. Borges, and C. Delerue-Matos. Development of
factor analysis of variance (ANOVA) test (29), which is considered electrochemical methods for determination of tramadol–analytical application to
a useful statistical tool for comparison of recovery data obtained pharmaceutical dosage forms. J. Pharm. Biomed. Anal 32(4–5): 975–981 (2003).
16. H. Hopkala, G. Misztal, and A. Wieczorek. Tramadol selective PVC membrane
from more than two methods. The calculated F-values did not electrodes and their analytical application. Die Pharmazie 53(12): 869–871 (1998).
exceed the critical value for either of the two drugs, indicating no 17. H.E. Abdellatef, M.M. El-Henawee, H.M. El-Sayed, and M.M. Ayad.
Spectrophotometric and spectrofluorimetric methods for analysis of tramadol,
significant differences between the proposed methods together acebutolol and dothiepin in pharmaceutical preparations. Spectrochim. Acta A
with the reference method. 65(5): 1087–1092 (2006).
18. H.E. Abdellatef. Kinetic spectrophotometric determination of tramadol
hydrochloride in pharmaceutical formulation. J. Pharm. Biomed. Anal. 29(5):
835–842 (2002).
Conclusion 19. T. Zhu, L. Ding, X. Guo, L. Yang, and A. Wen. Simultaneous Determination of
Tramadol and Acetaminophen in Human Plasma by LC–ESI–MS.
Chromatographia 66: 171–178 (2007).
In this study, two simple, rapid, and selective chromato- 20. Z. Tan, D. Ouyang, G. Zhou, L. Wang, Z. Li, D. Wang, G. Chen, S. Huang, Y. Liu,
D. Hu, and H. Zhou. Determination of tramadol and paracetamol in human
graphic procedures were established for the simultaneous deter- plasma by liquid chromatography ion trap mass spectrum. Yaowu Fenxi Zazhi
mination of PR and TR in laboratory-prepared mixtures as well 25(7): 795–798 (2005).
21. K.K. Srinivasan, J. Alex, A.A. Shirwaikar, S. Jacob, M.R. Sunil Kumar, and S.L.
as in commercially available tablets. The HPLC method made the Prabu. Simultaneous derivative spectrophotometric estimation of aceclofenac
use of an embedded amide reverse C8 stationary-phase with a low and tramadol with paracetamol in combination solid dosage forms. Indian J.
Pharm. Sci. 69(4): 540–545 (2007).
organic modifier content (10% acetonitrile) mobile phase. 22. Y. Li, X. Wan, G. Wang, and J. Cui. Simultaneous determination of tramadol
Additionally, the liquid chromatographic method reported here hydrochloride and paracetamol in their compound sustained release tablets by
principal component regression spectrophotometry. Zhongguo Yaoxue Zazhi
showed wider linearity ranges and better sensitivity compared to (Beijing, China) 41(20): 1594–1595 (2006).
the previously published HPLC methods. The GC–MS method is 23. X. Liu, J. Shi, Y. Liu, and Z. He. Simultaneous determination of contents of tra-
madol hydrochloride and paracetamol in their compound tablets by HPLC.
direct, requiring minimal sample preparation and made use of Shenyang Yaoke Daxue Xuebao 21(2): 111–113 (2004).
the total ion current detection. Both methods need no derivati- 24. L. Hong and J. Li. Determination of compound tramadol hydrochloride tablets
zation or pretreatment of the target compounds. The described and its related substances by HPLC. Zhongguo Yiyao Gongye Zazhi 36(5):
293–95 (2005).
validated chromatographic methods offer selectivity advantage 25. D. Peng, K. Huang, Y. Liu, and M. Li. Simultaneous determination of two con-
over the spectrophotometric-based non-separation methods. stituents in compound tramadol tablets by HPLC. Zhongguo Xinyao Zazhi 15(2):
120–122 (2006).
Finally, the proposed methods were found accurate and precise, 26. D.J. Speed, S.J. Dickson, E.R. Cairns, and N.D. Kim. Analysis of paracetamol
thus making them convenient for quality control purposes. using solid-phase extraction, deuterated internal standards, and gas chromatog-
raphy-mass spectrometry. J. Anal. Toxicol. 25(3): 198–202 (2001).
27. F.W. McLafferty, and F. Turecek. Interpretation of Mass Spectra, Fourth edition
References University Science Books, Sausalito, CA, (1993).
28. P. Armitage, and G. Berry. Statistical Methods in Medical Research, Third Edition,
Blackwell, Oxford, pp. 283–285 (1994).
1. B. Hinz and K. Brune. Antipyretic Analgesics: Nonsteroidal Antiinflammatory 29. J.N. Miller, and J.C. Miller. Statistics and Chemometrics for Analytical Chemistry,
Drugs, Selective COX-2 Inhibitors, Paracetamol and Pyrazolinones. Handbook of Fourth Edition, p. 57–64,77,78 (2000).
Experimental Pharmacology. 177(Analgesia): 65–93 (2006).
2. A. Bertolini, A. Ferrari, A. Ottani, S. Guerzoni, R. Tacchi, and S. Leone. Paracetamol:
New Vistas of an Old Drug. CNS Drug Reviews 12(3–4): 250–275 (2006). Manuscript received May 1, 2008;
3. S. Grond and A. Sablotzki. Clinical Pharmacology of Tramadol. Clin. Revision received June 24, 2008.

854