research article determination of cefoperazone sodium in...
Post on 22-Jul-2020
1 Views
Preview:
TRANSCRIPT
Research ArticleDetermination of Cefoperazone Sodium in Presence ofRelated Impurities by Improved Classical Least SquaresChemometric Methods A Comparative Study
Ibrahim A Naguib12 and Hany W Darwish34
1Pharmaceutical Chemistry Department Faculty of Pharmacy University of Tabuk Tabuk 71491 Saudi Arabia2Pharmaceutical Analytical Chemistry Department Faculty of Pharmacy Beni-Suef UniversityAlshaheed Shehata Ahmad Hegazy Street Beni-Suef 62514 Egypt3Department of Pharmaceutical Chemistry College of Pharmacy King Saud University PO Box 2457 Riyadh 11451 Saudi Arabia4Analytical Chemistry Department Faculty of Pharmacy Cairo University Kasr El-Aini Street Cairo 11562 Egypt
Correspondence should be addressed to Hany W Darwish hdarwish75yahoocom
Received 7 January 2016 Revised 29 March 2016 Accepted 21 April 2016
Academic Editor Pranav S Shrivastav
Copyright copy 2016 I A Naguib and H W Darwish This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited
A comparative study is established among 4 chemometric models depending on classical least squares (CLS) approach namelyspectral residual augmented CLS (SRACLS) net analyte processing CLS (NAP-CLS) orthogonal signal correction CLS (OSC-CLS)and direct orthogonal signal correction CLS (DOSC-CLS)The comparison is expressed through analysis of a case study dataset ofUV spectral data of Cefoperazone Sodium (CEF) and its two related impurities in pure powder form and in pharmaceutical dosageform Four-level three-factor experimental design was established for optimum analysis The adopted experimental design gaverise to a training set consisting of 16 mixtures (containing different ratios of interfering species) To test the prediction power of thesuggested models an independent test set consisting of 9 mixtures was usedThe presented results show the ability of the proposedmodels to quantify CEF in presence of two related impurities with high accuracy and selectivity (10376 plusmn 103 10207 plusmn 09110161 plusmn 072 and 10160 plusmn 072 for SRACLS NAP-CLS OSC-CLS and DOSC-CLS resp) Dosage form analysis results werecompared statistically to a published HPLC methodology showing insignificant difference in terms of precision and accuracyindicating the suggested models reliability and their suitability for quality control analysis of drug product Compared to othermodels OSC-CLS and DOSC-CLS models gave more accurate results with lower prediction error for test set samples
1 Introduction
Cefoperazone Sodium (CEF) (Figure 1(a)) [1] belongs to thethird-generation cephalosporin group that works via pre-venting the biosynthesis of bacterial cell wall [2] Accord-ing to British Pharmacopeia [3] 7-amino-cephalospora-nic acid (7-ACA) (6R7R)-3-[(acetyloxy)methyl]-7-amino-8-oxo-5-thia-1-azabicyclo[420]oct-2-ene-2-carboxylic acid(Figure 1(b)) and 5-mercapto-1-methyl-tetrazole (5-MER)and 1-methyl-1H-tetrazole-5-thiol (Figure 1(c)) are deemed tobe specific impurities for CEF Pharmacological significanceof both impurities was mentioned in literature [4]
A literature review showed a group of analytical meth-ods for CEF quantification in its drug products includingspectrophotometry [5 6] NIR [7] and derivative UV-spec-trophotometry for determination of CEF in binary mixtureswith sulbactam [8] Chromatographic methods were utilizedfor assay of CEF [9] CEF and sulbactam [10 11] besidesHPLC methodology using 120573-cyclodextrin stationary phasefor assay of ternarymixture of CEF ampicillin and sulbactamwas described [12] Furthermore LC-MSMS method wasdescribed for assay of CEF and sulbactam binary mixturein plasma [13] Moreover electrochemical and voltammetricassay of CEF were stated [14 15]
Hindawi Publishing CorporationJournal of ChemistryVolume 2016 Article ID 7570643 8 pageshttpdxdoiorg10115520167570643
2 Journal of Chemistry
HO
HN
O
H NH
N
O
N
O
O
N
S
O
ONaO
S
N
N N
NH H
CH3
CH3
(a)
N
SH
O
H
OHO
O
O
CH3
NH2
(b)
N
NN
N
SH
CH3
(c)
Figure 1 Chemical structures of CEF (a) and its reported impurities 7-ACA (b) and 5-MER (c)
CEF and its two mentioned impurities were analyzedby HPLC and HPTLC chromatographic methods [16] andchemometric methods [17]
There are twomain aims for the presented work First weaim to establish a comparison among 4 chemometric modelsdepending on classical least squares (CLS) approach namelyspectral residual augmented CLS (SRACLS) net analyteprocessing CLS (NAP-CLS) orthogonal signal correctionCLS (OSC-CLS) and direct orthogonal signal correctionCLS(DOSC-CLS)The last 3models are preprocessing techniquesimplemented to increase the predictive capabilities of CLSmodel The proposed improvement could offer CLS modelwider applications for quantitative analysis yet keeping theadvantage of its built-in qualitative properties The compar-ison shows the underlying algorithm for each model andcompares the analysis results of different mixtures of CEF 7-ACA and 5-MER as a case study to indicate which of the4 models is best to improve CLS prediction ability Secondthe presented models show the ability of chemometricsto analyze selectively CEF in ternary mixtures with itstwo reported impurities utilizing inexpensive and availableinstruments like UV-spectrophotometer with suggested rou-tine application of the models for quality control analysis ofthe pharmaceutical dosage form The selected models offercomparable accuracy and precision for the quantitation ofCEF in pharmaceutical formulation compared to the officialHPLC method [9]
2 Materials and Methods
Full description of instrument materials chemical reagentspharmaceutical formulations stock and working prepara-tions applied in the presented study is mentioned in ourprevious published work [17]
21 Linearity UV spectra for different samples of CEF rang-ing from 1 to 70 120583gmLminus1 were recorded from 210 to 300 nmCEF showed linearity between 5 and 50120583gmLminus1 at its 120582maxat 229 nm The superimposed spectra of 10 120583gmLminus1 of CEF7-ACA and 5-MER are shown in Figure 2
Abso
rban
ce
000
020
040
060
080
100
2000 2500 3000
Wavelength (nm)
Figure 2 Zero-order absorption spectra of 10 120583g mLminus1 of CEF (mdash)7-ACA (- - -) and 5-MER ( ) using methanol as blank
22 Experimental Design
221 Calibration Set Four-level three-factor experimentaldesign was accomplished utilizing four concentration levelscoded as +2 +1 minus1 and minus2 where minus1 is the central levelfor CEF and its impurities (7-ACA and 5-MER) The aim ofthe current design is to cover the mixture space in a properway ending up with a training set composed of 16 mixtures[18] Twenty 120583gmLminus1 of CEF was chosen as a central levelfor the design and the proposed concentrations for each levelfor CEF were dependent on its calibration range Concerningimpurities their concentration levels were based on the factthat we include them in up to 3 of CEF calculated on molarbasis The concentration design matrix was represented inTable 1The two-dimensional (2D) scores plot for the first twoPCs of the concentrationmatrix was constructed to check theorthogonality symmetry and rotatability for the mixtures ofthe training set (presented as circles) as shown in Figure 3The best preprocessing procedure that gives the optimumresults was mean centering one
222 Test Set To assess the validity and the predictabilityof the compared chemometric models the test set mixtures
Journal of Chemistry 3
Table 1 Four-level three-factor experimental design of the trainingset (16 mixtures) and test set (9 mixtures) shown as concentrationsof the mixture components in 120583gmLminus1
Training set Test setCEF 7-ACA 5-MER CEF 7-ACA 5-MER18 013 0065 25 017 00918 015 007 19 014 00720 015 009 22 018 007520 02 007 21 014 00826 015 0065 23 015 006520 013 0085 25 02 008518 018 0085 19 013 00824 018 007 22 016 00724 015 0085 21 016 00820 018 006524 013 00918 02 00926 02 008526 018 00924 02 006526 013 007
minus5 minus4 minus3 minus2 minus1 0 1 2 3 4
minus003
minus002
minus001
0
001
002
003
004
Scores PC1
Scor
es P
C2
Test setTraining set
Figure 3 Two-dimensional scores plot for the mean centered 16training set samples (circles) and the 9 test set samples (stars) ofconcentration matrices of the 4-level 3-factor experimental designwith variance of 997 for first PC and 028 for second PC
were attained by preparing nine independent mixtures otherthan the training set mixtures but within the concentrationspace of the adopted design as indicated in Table 1 The wellpositioning of the mixtures of both training set (circles) andtest set mixtures (stars) is indicated in Figure 3
223 Analysis of Cefobid Vial Working solution of CEF(100 120583gmLminus1) was prepared usingmethanol as solvent Lastlytwo mL of this working solution was completed to ten mL
withmethanolThemean of three spectra wasmeasuredThisexperiment was repeated six times and the attained spectrawere analyzed by the suggested models
23 Software Codes for SRCALS were written with Mat-lab 710246 (R14) NAP-CLS OSC-CLS and DOSC-CLSmethods were executed in Matlab 710246 (R14) utilizingMVC1 toolboxes [19] Statistical tests (119905-test and 119865-test) wereaccomplished using Microsoft Excel
3 Chemometric Methods
31 Spectral Residual Augmented Classical Least Squares(SRACLS) Full mathematical description of SRACLS modelcan be found in literature [20ndash22]
The new predicted concentration will be calculatedaccording to the following equation
Cnew = XK1015840
new (KnewK1015840
new)minus1
asymp XK+
new (1)
where Cnew is the new predicted concentration after neces-sary augmentation of K and Knew is the augmented pure com-ponent contributionwhich takes into account residual errorscontribution of impurities and contribution of interferingpure component spectra The optimum number of loadingsused for augmentation of K matrix was selected throughrunning leave-one-out cross-validation (LOO-CV) [22]
32 Improved Signal-CLSModels CLS as a direct calibrationmodel is the simplest developed chemometric techniqueCLS method necessitates that all the components in thetraining set should be well recognized Contrasting CLSPCR (principal component regression) and PLS (partialleast squares) methods can be applied for determination ofthe components under inspection even in the presence ofunknown interfering component that offered the two modelsan advantage over CLS [23] Predictive power of CLS modelcan be improved greatly by using preprocessing techniquessuch as the methods proposed in this paper like net analytepreprocessing (NAP) orthogonal signal correction (OSC)and direct orthogonal signal correction (DOSC) Preprocess-ing of the data prior to calibration step may be applied tominimize systematic variations influence that is unrelated tothe interesting parameters [24]
321 Net Analyte Preprocessing (NAP) Net analyte signal(NAS) calculation is the basis of NAP NAS gave rise tonumerous new calibration models based on the same con-cept extracting portion of the signal which is directly corre-lated to the concentration of analyte and thus beneficial forprediction purposes [25] Additionally NAS was applied forcalculation of analytical figures of merit and for developingsensor selection technique [25] The fundamental principleof NAS-based calibration models is to distinguish betweentwo kinds of contributions in the training data matrix Xone originates from the analyte of interest while the secondoriginates from other sources of variability The full details ofNAP are mentioned by Goicoechea and Olivieri [25]
4 Journal of Chemistry
322 Orthogonal Signal Correction (OSC) and Direct Orthog-onal Signal Correction (DOSC) Numerous algorithms forapplyingOSC as a filteringmechanism to datamatrixX weredescribed in detail in literature [24 26 27] In the presentedwork we refer to the OSC algorithm discussed by Fearn [27]as it is the most easily interpreted and can be compared toNAP methodology
These algorithms are employed to get rid of parts ofthe spectra that are orthogonal to the concentration Fearnrsquosalgorithm spans the same space as PLS or PCRmodels on theunprocessed data and extracts components that are strictlyorthogonal to the concentrationThe first step is a projectionof 119883 onto the subspace that is orthogonal to 119883119879119884
119896
to obtainthe variation in 119883 that is correlated to 119884
119896
(where 119884119896
is con-centration of 119896 analyte) Fearnrsquos algorithm removed almostthe drawbacks that were seen in the other OSC algorithmssuch as lack of orthogonality leading to removal of usefulinformation or adding useless data in the corrected matrix
DOSC approach introduced by Westerhuis et al [28] isbased only on least squares steps It finds components whichare orthogonal to 119884 which label the main variation of 119883 tobe removed from corrected119883 For implementation of DOSCfirstly119884 is decomposed into two orthogonal parts (the pro-jection of 119884 onto119883) and 119865 (the residual part ie orthogonalto119883) Secondly119883 is decomposed into two orthogonal partsone part that is of the same range as and the other partthat is orthogonal to it Finally principal component analysis(PCA) is applied on the part of119883 orthogonal to 119884 to removeit from119883 giving rise to corrected119883
Mathematically DOSC can be described by the followingsteps
Step 1 Step 1 is as follows
119884 = 119875119883
119884 + 119860119883
119884 = + 119865 (2)
Step 2 Step 2 is as follows
119883 = 119875
119884
119883 + 119860
119884
119883 (3)
Step 3 Step 3 is as follows
119883DOSC = 119883corrected = 119883 minus 119905119875119879
(4)
where 119905 is score vector and 119875 is loading vector
The threemodels (NAP-CLSOSC-CLS andDOSC-CLS)are simpler when compared with other methods such as PLSThis simplicity originates from the utilization of orthogonalprojections concepts followed by the well-established classi-cal least squares fitting
(1) Optimization of Number of Factors for the NAP-CLS OSC-CLS and DOSC-CLSModels Leave-one-out cross-validation(LOO-CV) is adopted in the current work for optimization offactors number for construction of the proposedmodels [23]by building the model using 119868 minus 1 samplesrsquo set (15 mixturesfrom training or calibration set) to predict the one sample
left (validation sample) The root mean square error of cross-validation (RMSECV) is computed as follows
RMSECV = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894 CV)2
(5)
where 119868 is the number of objects in the calibration set 119888119894
is the known concentration for sample 119894 and 119860119894 CV is the
predicted concentration of sample 119894 using 119860 componentsMean centering was applied to the calibration set each timesuccessive samples were left out
4 Results and Discussion
The current work was designed to accomplish a number ofgoals First goal is to develop simple selective and precisechemometric models for assaying CEF in presence of itsimpurities in pure form and dosage form Second goal isto demonstrate the quantitative power as well as qualitativepower of the suggested models and compare their inherentcharacteristics using the analyzed mixtures as a case studyAdditionally we aim to display the influence of variouspreprocessing steps such as NAP OSC and DOSC on theperformance of CLS in quantitative analysis Optimizationof methodsrsquo parameters was the first step to run modelsproperly For the SRACLS model the optimum number ofloadings119875new used for augmentation of Kmatrix was selectedthrough running LOO-CV and found to be 4
For appropriate building of NAP-CLS OSC-CLS andDOSC-CLS methods number of projection matrix factors(for NAP-CLS) and number of extracted factors (for OSC-CLS and DOSC-CLS) were adjusted For this purpose LOO-CV was applied where log PRESS (predicted residual errorsumof squares) valueswere computedTheoptimumnumberof factors was chosen in accordance with Haaland andThomas approach [29] In all improved CLS models twofactors were essential for constructing the models except inDOSC-CLS model where three factors were required Thisinformation demonstrates that DOSC as a preprocessingtechnique is more complex than NAP and OSC procedures
After parametersrsquo optimization and training procedureall proposed methods together with ordinary CLS wereapplied successfully for estimation of CEF in training setand test set as indicated in Tables 2 and 3 respectivelyRecovery percentages mean recoveries standard deviationand RMSEC and RMSEP values are anticipated in Tables 2and 3
RMSEC (for training set) and RMSEP (for test set) werecomputed in a similar way according to the following equa-tion
RMSEC (119875) = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894
)2
(6)
where 119868 is the number of samples in the test set (in case ofRMSEP) and 119868 minus 1 is for training set (in case of RMSEC)119888119894
is the known concentration for sample 119894 and 119860119894
is theestimated concentration of sample 119894 using 119860 components
Journal of Chemistry 5
Table2Analysis
results
forthe
predictio
nof
training
set(autopredictio
n)of
CEFby
thep
ropo
sedchem
ometric
mod
els
Training
set
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
181878
10433
1829
1016
11855
10306
1858
10322
1858
10322
181786
9921
1799
9994
1799
9994
1801
10006
1801
10006
20197
9852
1985
9925
1988
9940
1984
9920
1984
9920
201969
9845
1977
9885
1984
9920
1989
9945
1988
9940
262572
9894
2617
10065
2621
10081
2614
10054
2614
10054
201978
9891
1992
9960
1990
9950
1989
9945
1989
9945
181785
9919
1794
9967
1789
9939
1788
9933
1788
9933
242357
9821
2384
9933
2382
9925
2374
9892
2374
9892
242393
9971
2416
10067
2409
10038
2400
10000
2401
10004
201974
9872
1981
9905
1972
9860
1972
9860
1972
9860
242403
10011
2414
10058
2410
10042
2401
10004
2401
10004
181835
1019
71816
10089
1808
10044
1807
10039
1807
10039
262637
1014
42645
1017
32637
1014
22625
10096
2625
10096
26261
10039
2519
9688
2591
9965
2593
9973
2593
9973
242433
1013
62426
1010
82393
9971
2413
10054
2413
10054
262618
10069
2604
10015
2573
9896
2594
9977
2594
9977
Mean
10001
9999
10010
10001
10001
SD16
412
110
710
610
7RM
SEC
032
028
022
020
020
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
2 Journal of Chemistry
HO
HN
O
H NH
N
O
N
O
O
N
S
O
ONaO
S
N
N N
NH H
CH3
CH3
(a)
N
SH
O
H
OHO
O
O
CH3
NH2
(b)
N
NN
N
SH
CH3
(c)
Figure 1 Chemical structures of CEF (a) and its reported impurities 7-ACA (b) and 5-MER (c)
CEF and its two mentioned impurities were analyzedby HPLC and HPTLC chromatographic methods [16] andchemometric methods [17]
There are twomain aims for the presented work First weaim to establish a comparison among 4 chemometric modelsdepending on classical least squares (CLS) approach namelyspectral residual augmented CLS (SRACLS) net analyteprocessing CLS (NAP-CLS) orthogonal signal correctionCLS (OSC-CLS) and direct orthogonal signal correctionCLS(DOSC-CLS)The last 3models are preprocessing techniquesimplemented to increase the predictive capabilities of CLSmodel The proposed improvement could offer CLS modelwider applications for quantitative analysis yet keeping theadvantage of its built-in qualitative properties The compar-ison shows the underlying algorithm for each model andcompares the analysis results of different mixtures of CEF 7-ACA and 5-MER as a case study to indicate which of the4 models is best to improve CLS prediction ability Secondthe presented models show the ability of chemometricsto analyze selectively CEF in ternary mixtures with itstwo reported impurities utilizing inexpensive and availableinstruments like UV-spectrophotometer with suggested rou-tine application of the models for quality control analysis ofthe pharmaceutical dosage form The selected models offercomparable accuracy and precision for the quantitation ofCEF in pharmaceutical formulation compared to the officialHPLC method [9]
2 Materials and Methods
Full description of instrument materials chemical reagentspharmaceutical formulations stock and working prepara-tions applied in the presented study is mentioned in ourprevious published work [17]
21 Linearity UV spectra for different samples of CEF rang-ing from 1 to 70 120583gmLminus1 were recorded from 210 to 300 nmCEF showed linearity between 5 and 50120583gmLminus1 at its 120582maxat 229 nm The superimposed spectra of 10 120583gmLminus1 of CEF7-ACA and 5-MER are shown in Figure 2
Abso
rban
ce
000
020
040
060
080
100
2000 2500 3000
Wavelength (nm)
Figure 2 Zero-order absorption spectra of 10 120583g mLminus1 of CEF (mdash)7-ACA (- - -) and 5-MER ( ) using methanol as blank
22 Experimental Design
221 Calibration Set Four-level three-factor experimentaldesign was accomplished utilizing four concentration levelscoded as +2 +1 minus1 and minus2 where minus1 is the central levelfor CEF and its impurities (7-ACA and 5-MER) The aim ofthe current design is to cover the mixture space in a properway ending up with a training set composed of 16 mixtures[18] Twenty 120583gmLminus1 of CEF was chosen as a central levelfor the design and the proposed concentrations for each levelfor CEF were dependent on its calibration range Concerningimpurities their concentration levels were based on the factthat we include them in up to 3 of CEF calculated on molarbasis The concentration design matrix was represented inTable 1The two-dimensional (2D) scores plot for the first twoPCs of the concentrationmatrix was constructed to check theorthogonality symmetry and rotatability for the mixtures ofthe training set (presented as circles) as shown in Figure 3The best preprocessing procedure that gives the optimumresults was mean centering one
222 Test Set To assess the validity and the predictabilityof the compared chemometric models the test set mixtures
Journal of Chemistry 3
Table 1 Four-level three-factor experimental design of the trainingset (16 mixtures) and test set (9 mixtures) shown as concentrationsof the mixture components in 120583gmLminus1
Training set Test setCEF 7-ACA 5-MER CEF 7-ACA 5-MER18 013 0065 25 017 00918 015 007 19 014 00720 015 009 22 018 007520 02 007 21 014 00826 015 0065 23 015 006520 013 0085 25 02 008518 018 0085 19 013 00824 018 007 22 016 00724 015 0085 21 016 00820 018 006524 013 00918 02 00926 02 008526 018 00924 02 006526 013 007
minus5 minus4 minus3 minus2 minus1 0 1 2 3 4
minus003
minus002
minus001
0
001
002
003
004
Scores PC1
Scor
es P
C2
Test setTraining set
Figure 3 Two-dimensional scores plot for the mean centered 16training set samples (circles) and the 9 test set samples (stars) ofconcentration matrices of the 4-level 3-factor experimental designwith variance of 997 for first PC and 028 for second PC
were attained by preparing nine independent mixtures otherthan the training set mixtures but within the concentrationspace of the adopted design as indicated in Table 1 The wellpositioning of the mixtures of both training set (circles) andtest set mixtures (stars) is indicated in Figure 3
223 Analysis of Cefobid Vial Working solution of CEF(100 120583gmLminus1) was prepared usingmethanol as solvent Lastlytwo mL of this working solution was completed to ten mL
withmethanolThemean of three spectra wasmeasuredThisexperiment was repeated six times and the attained spectrawere analyzed by the suggested models
23 Software Codes for SRCALS were written with Mat-lab 710246 (R14) NAP-CLS OSC-CLS and DOSC-CLSmethods were executed in Matlab 710246 (R14) utilizingMVC1 toolboxes [19] Statistical tests (119905-test and 119865-test) wereaccomplished using Microsoft Excel
3 Chemometric Methods
31 Spectral Residual Augmented Classical Least Squares(SRACLS) Full mathematical description of SRACLS modelcan be found in literature [20ndash22]
The new predicted concentration will be calculatedaccording to the following equation
Cnew = XK1015840
new (KnewK1015840
new)minus1
asymp XK+
new (1)
where Cnew is the new predicted concentration after neces-sary augmentation of K and Knew is the augmented pure com-ponent contributionwhich takes into account residual errorscontribution of impurities and contribution of interferingpure component spectra The optimum number of loadingsused for augmentation of K matrix was selected throughrunning leave-one-out cross-validation (LOO-CV) [22]
32 Improved Signal-CLSModels CLS as a direct calibrationmodel is the simplest developed chemometric techniqueCLS method necessitates that all the components in thetraining set should be well recognized Contrasting CLSPCR (principal component regression) and PLS (partialleast squares) methods can be applied for determination ofthe components under inspection even in the presence ofunknown interfering component that offered the two modelsan advantage over CLS [23] Predictive power of CLS modelcan be improved greatly by using preprocessing techniquessuch as the methods proposed in this paper like net analytepreprocessing (NAP) orthogonal signal correction (OSC)and direct orthogonal signal correction (DOSC) Preprocess-ing of the data prior to calibration step may be applied tominimize systematic variations influence that is unrelated tothe interesting parameters [24]
321 Net Analyte Preprocessing (NAP) Net analyte signal(NAS) calculation is the basis of NAP NAS gave rise tonumerous new calibration models based on the same con-cept extracting portion of the signal which is directly corre-lated to the concentration of analyte and thus beneficial forprediction purposes [25] Additionally NAS was applied forcalculation of analytical figures of merit and for developingsensor selection technique [25] The fundamental principleof NAS-based calibration models is to distinguish betweentwo kinds of contributions in the training data matrix Xone originates from the analyte of interest while the secondoriginates from other sources of variability The full details ofNAP are mentioned by Goicoechea and Olivieri [25]
4 Journal of Chemistry
322 Orthogonal Signal Correction (OSC) and Direct Orthog-onal Signal Correction (DOSC) Numerous algorithms forapplyingOSC as a filteringmechanism to datamatrixX weredescribed in detail in literature [24 26 27] In the presentedwork we refer to the OSC algorithm discussed by Fearn [27]as it is the most easily interpreted and can be compared toNAP methodology
These algorithms are employed to get rid of parts ofthe spectra that are orthogonal to the concentration Fearnrsquosalgorithm spans the same space as PLS or PCRmodels on theunprocessed data and extracts components that are strictlyorthogonal to the concentrationThe first step is a projectionof 119883 onto the subspace that is orthogonal to 119883119879119884
119896
to obtainthe variation in 119883 that is correlated to 119884
119896
(where 119884119896
is con-centration of 119896 analyte) Fearnrsquos algorithm removed almostthe drawbacks that were seen in the other OSC algorithmssuch as lack of orthogonality leading to removal of usefulinformation or adding useless data in the corrected matrix
DOSC approach introduced by Westerhuis et al [28] isbased only on least squares steps It finds components whichare orthogonal to 119884 which label the main variation of 119883 tobe removed from corrected119883 For implementation of DOSCfirstly119884 is decomposed into two orthogonal parts (the pro-jection of 119884 onto119883) and 119865 (the residual part ie orthogonalto119883) Secondly119883 is decomposed into two orthogonal partsone part that is of the same range as and the other partthat is orthogonal to it Finally principal component analysis(PCA) is applied on the part of119883 orthogonal to 119884 to removeit from119883 giving rise to corrected119883
Mathematically DOSC can be described by the followingsteps
Step 1 Step 1 is as follows
119884 = 119875119883
119884 + 119860119883
119884 = + 119865 (2)
Step 2 Step 2 is as follows
119883 = 119875
119884
119883 + 119860
119884
119883 (3)
Step 3 Step 3 is as follows
119883DOSC = 119883corrected = 119883 minus 119905119875119879
(4)
where 119905 is score vector and 119875 is loading vector
The threemodels (NAP-CLSOSC-CLS andDOSC-CLS)are simpler when compared with other methods such as PLSThis simplicity originates from the utilization of orthogonalprojections concepts followed by the well-established classi-cal least squares fitting
(1) Optimization of Number of Factors for the NAP-CLS OSC-CLS and DOSC-CLSModels Leave-one-out cross-validation(LOO-CV) is adopted in the current work for optimization offactors number for construction of the proposedmodels [23]by building the model using 119868 minus 1 samplesrsquo set (15 mixturesfrom training or calibration set) to predict the one sample
left (validation sample) The root mean square error of cross-validation (RMSECV) is computed as follows
RMSECV = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894 CV)2
(5)
where 119868 is the number of objects in the calibration set 119888119894
is the known concentration for sample 119894 and 119860119894 CV is the
predicted concentration of sample 119894 using 119860 componentsMean centering was applied to the calibration set each timesuccessive samples were left out
4 Results and Discussion
The current work was designed to accomplish a number ofgoals First goal is to develop simple selective and precisechemometric models for assaying CEF in presence of itsimpurities in pure form and dosage form Second goal isto demonstrate the quantitative power as well as qualitativepower of the suggested models and compare their inherentcharacteristics using the analyzed mixtures as a case studyAdditionally we aim to display the influence of variouspreprocessing steps such as NAP OSC and DOSC on theperformance of CLS in quantitative analysis Optimizationof methodsrsquo parameters was the first step to run modelsproperly For the SRACLS model the optimum number ofloadings119875new used for augmentation of Kmatrix was selectedthrough running LOO-CV and found to be 4
For appropriate building of NAP-CLS OSC-CLS andDOSC-CLS methods number of projection matrix factors(for NAP-CLS) and number of extracted factors (for OSC-CLS and DOSC-CLS) were adjusted For this purpose LOO-CV was applied where log PRESS (predicted residual errorsumof squares) valueswere computedTheoptimumnumberof factors was chosen in accordance with Haaland andThomas approach [29] In all improved CLS models twofactors were essential for constructing the models except inDOSC-CLS model where three factors were required Thisinformation demonstrates that DOSC as a preprocessingtechnique is more complex than NAP and OSC procedures
After parametersrsquo optimization and training procedureall proposed methods together with ordinary CLS wereapplied successfully for estimation of CEF in training setand test set as indicated in Tables 2 and 3 respectivelyRecovery percentages mean recoveries standard deviationand RMSEC and RMSEP values are anticipated in Tables 2and 3
RMSEC (for training set) and RMSEP (for test set) werecomputed in a similar way according to the following equa-tion
RMSEC (119875) = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894
)2
(6)
where 119868 is the number of samples in the test set (in case ofRMSEP) and 119868 minus 1 is for training set (in case of RMSEC)119888119894
is the known concentration for sample 119894 and 119860119894
is theestimated concentration of sample 119894 using 119860 components
Journal of Chemistry 5
Table2Analysis
results
forthe
predictio
nof
training
set(autopredictio
n)of
CEFby
thep
ropo
sedchem
ometric
mod
els
Training
set
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
181878
10433
1829
1016
11855
10306
1858
10322
1858
10322
181786
9921
1799
9994
1799
9994
1801
10006
1801
10006
20197
9852
1985
9925
1988
9940
1984
9920
1984
9920
201969
9845
1977
9885
1984
9920
1989
9945
1988
9940
262572
9894
2617
10065
2621
10081
2614
10054
2614
10054
201978
9891
1992
9960
1990
9950
1989
9945
1989
9945
181785
9919
1794
9967
1789
9939
1788
9933
1788
9933
242357
9821
2384
9933
2382
9925
2374
9892
2374
9892
242393
9971
2416
10067
2409
10038
2400
10000
2401
10004
201974
9872
1981
9905
1972
9860
1972
9860
1972
9860
242403
10011
2414
10058
2410
10042
2401
10004
2401
10004
181835
1019
71816
10089
1808
10044
1807
10039
1807
10039
262637
1014
42645
1017
32637
1014
22625
10096
2625
10096
26261
10039
2519
9688
2591
9965
2593
9973
2593
9973
242433
1013
62426
1010
82393
9971
2413
10054
2413
10054
262618
10069
2604
10015
2573
9896
2594
9977
2594
9977
Mean
10001
9999
10010
10001
10001
SD16
412
110
710
610
7RM
SEC
032
028
022
020
020
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
Journal of Chemistry 3
Table 1 Four-level three-factor experimental design of the trainingset (16 mixtures) and test set (9 mixtures) shown as concentrationsof the mixture components in 120583gmLminus1
Training set Test setCEF 7-ACA 5-MER CEF 7-ACA 5-MER18 013 0065 25 017 00918 015 007 19 014 00720 015 009 22 018 007520 02 007 21 014 00826 015 0065 23 015 006520 013 0085 25 02 008518 018 0085 19 013 00824 018 007 22 016 00724 015 0085 21 016 00820 018 006524 013 00918 02 00926 02 008526 018 00924 02 006526 013 007
minus5 minus4 minus3 minus2 minus1 0 1 2 3 4
minus003
minus002
minus001
0
001
002
003
004
Scores PC1
Scor
es P
C2
Test setTraining set
Figure 3 Two-dimensional scores plot for the mean centered 16training set samples (circles) and the 9 test set samples (stars) ofconcentration matrices of the 4-level 3-factor experimental designwith variance of 997 for first PC and 028 for second PC
were attained by preparing nine independent mixtures otherthan the training set mixtures but within the concentrationspace of the adopted design as indicated in Table 1 The wellpositioning of the mixtures of both training set (circles) andtest set mixtures (stars) is indicated in Figure 3
223 Analysis of Cefobid Vial Working solution of CEF(100 120583gmLminus1) was prepared usingmethanol as solvent Lastlytwo mL of this working solution was completed to ten mL
withmethanolThemean of three spectra wasmeasuredThisexperiment was repeated six times and the attained spectrawere analyzed by the suggested models
23 Software Codes for SRCALS were written with Mat-lab 710246 (R14) NAP-CLS OSC-CLS and DOSC-CLSmethods were executed in Matlab 710246 (R14) utilizingMVC1 toolboxes [19] Statistical tests (119905-test and 119865-test) wereaccomplished using Microsoft Excel
3 Chemometric Methods
31 Spectral Residual Augmented Classical Least Squares(SRACLS) Full mathematical description of SRACLS modelcan be found in literature [20ndash22]
The new predicted concentration will be calculatedaccording to the following equation
Cnew = XK1015840
new (KnewK1015840
new)minus1
asymp XK+
new (1)
where Cnew is the new predicted concentration after neces-sary augmentation of K and Knew is the augmented pure com-ponent contributionwhich takes into account residual errorscontribution of impurities and contribution of interferingpure component spectra The optimum number of loadingsused for augmentation of K matrix was selected throughrunning leave-one-out cross-validation (LOO-CV) [22]
32 Improved Signal-CLSModels CLS as a direct calibrationmodel is the simplest developed chemometric techniqueCLS method necessitates that all the components in thetraining set should be well recognized Contrasting CLSPCR (principal component regression) and PLS (partialleast squares) methods can be applied for determination ofthe components under inspection even in the presence ofunknown interfering component that offered the two modelsan advantage over CLS [23] Predictive power of CLS modelcan be improved greatly by using preprocessing techniquessuch as the methods proposed in this paper like net analytepreprocessing (NAP) orthogonal signal correction (OSC)and direct orthogonal signal correction (DOSC) Preprocess-ing of the data prior to calibration step may be applied tominimize systematic variations influence that is unrelated tothe interesting parameters [24]
321 Net Analyte Preprocessing (NAP) Net analyte signal(NAS) calculation is the basis of NAP NAS gave rise tonumerous new calibration models based on the same con-cept extracting portion of the signal which is directly corre-lated to the concentration of analyte and thus beneficial forprediction purposes [25] Additionally NAS was applied forcalculation of analytical figures of merit and for developingsensor selection technique [25] The fundamental principleof NAS-based calibration models is to distinguish betweentwo kinds of contributions in the training data matrix Xone originates from the analyte of interest while the secondoriginates from other sources of variability The full details ofNAP are mentioned by Goicoechea and Olivieri [25]
4 Journal of Chemistry
322 Orthogonal Signal Correction (OSC) and Direct Orthog-onal Signal Correction (DOSC) Numerous algorithms forapplyingOSC as a filteringmechanism to datamatrixX weredescribed in detail in literature [24 26 27] In the presentedwork we refer to the OSC algorithm discussed by Fearn [27]as it is the most easily interpreted and can be compared toNAP methodology
These algorithms are employed to get rid of parts ofthe spectra that are orthogonal to the concentration Fearnrsquosalgorithm spans the same space as PLS or PCRmodels on theunprocessed data and extracts components that are strictlyorthogonal to the concentrationThe first step is a projectionof 119883 onto the subspace that is orthogonal to 119883119879119884
119896
to obtainthe variation in 119883 that is correlated to 119884
119896
(where 119884119896
is con-centration of 119896 analyte) Fearnrsquos algorithm removed almostthe drawbacks that were seen in the other OSC algorithmssuch as lack of orthogonality leading to removal of usefulinformation or adding useless data in the corrected matrix
DOSC approach introduced by Westerhuis et al [28] isbased only on least squares steps It finds components whichare orthogonal to 119884 which label the main variation of 119883 tobe removed from corrected119883 For implementation of DOSCfirstly119884 is decomposed into two orthogonal parts (the pro-jection of 119884 onto119883) and 119865 (the residual part ie orthogonalto119883) Secondly119883 is decomposed into two orthogonal partsone part that is of the same range as and the other partthat is orthogonal to it Finally principal component analysis(PCA) is applied on the part of119883 orthogonal to 119884 to removeit from119883 giving rise to corrected119883
Mathematically DOSC can be described by the followingsteps
Step 1 Step 1 is as follows
119884 = 119875119883
119884 + 119860119883
119884 = + 119865 (2)
Step 2 Step 2 is as follows
119883 = 119875
119884
119883 + 119860
119884
119883 (3)
Step 3 Step 3 is as follows
119883DOSC = 119883corrected = 119883 minus 119905119875119879
(4)
where 119905 is score vector and 119875 is loading vector
The threemodels (NAP-CLSOSC-CLS andDOSC-CLS)are simpler when compared with other methods such as PLSThis simplicity originates from the utilization of orthogonalprojections concepts followed by the well-established classi-cal least squares fitting
(1) Optimization of Number of Factors for the NAP-CLS OSC-CLS and DOSC-CLSModels Leave-one-out cross-validation(LOO-CV) is adopted in the current work for optimization offactors number for construction of the proposedmodels [23]by building the model using 119868 minus 1 samplesrsquo set (15 mixturesfrom training or calibration set) to predict the one sample
left (validation sample) The root mean square error of cross-validation (RMSECV) is computed as follows
RMSECV = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894 CV)2
(5)
where 119868 is the number of objects in the calibration set 119888119894
is the known concentration for sample 119894 and 119860119894 CV is the
predicted concentration of sample 119894 using 119860 componentsMean centering was applied to the calibration set each timesuccessive samples were left out
4 Results and Discussion
The current work was designed to accomplish a number ofgoals First goal is to develop simple selective and precisechemometric models for assaying CEF in presence of itsimpurities in pure form and dosage form Second goal isto demonstrate the quantitative power as well as qualitativepower of the suggested models and compare their inherentcharacteristics using the analyzed mixtures as a case studyAdditionally we aim to display the influence of variouspreprocessing steps such as NAP OSC and DOSC on theperformance of CLS in quantitative analysis Optimizationof methodsrsquo parameters was the first step to run modelsproperly For the SRACLS model the optimum number ofloadings119875new used for augmentation of Kmatrix was selectedthrough running LOO-CV and found to be 4
For appropriate building of NAP-CLS OSC-CLS andDOSC-CLS methods number of projection matrix factors(for NAP-CLS) and number of extracted factors (for OSC-CLS and DOSC-CLS) were adjusted For this purpose LOO-CV was applied where log PRESS (predicted residual errorsumof squares) valueswere computedTheoptimumnumberof factors was chosen in accordance with Haaland andThomas approach [29] In all improved CLS models twofactors were essential for constructing the models except inDOSC-CLS model where three factors were required Thisinformation demonstrates that DOSC as a preprocessingtechnique is more complex than NAP and OSC procedures
After parametersrsquo optimization and training procedureall proposed methods together with ordinary CLS wereapplied successfully for estimation of CEF in training setand test set as indicated in Tables 2 and 3 respectivelyRecovery percentages mean recoveries standard deviationand RMSEC and RMSEP values are anticipated in Tables 2and 3
RMSEC (for training set) and RMSEP (for test set) werecomputed in a similar way according to the following equa-tion
RMSEC (119875) = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894
)2
(6)
where 119868 is the number of samples in the test set (in case ofRMSEP) and 119868 minus 1 is for training set (in case of RMSEC)119888119894
is the known concentration for sample 119894 and 119860119894
is theestimated concentration of sample 119894 using 119860 components
Journal of Chemistry 5
Table2Analysis
results
forthe
predictio
nof
training
set(autopredictio
n)of
CEFby
thep
ropo
sedchem
ometric
mod
els
Training
set
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
181878
10433
1829
1016
11855
10306
1858
10322
1858
10322
181786
9921
1799
9994
1799
9994
1801
10006
1801
10006
20197
9852
1985
9925
1988
9940
1984
9920
1984
9920
201969
9845
1977
9885
1984
9920
1989
9945
1988
9940
262572
9894
2617
10065
2621
10081
2614
10054
2614
10054
201978
9891
1992
9960
1990
9950
1989
9945
1989
9945
181785
9919
1794
9967
1789
9939
1788
9933
1788
9933
242357
9821
2384
9933
2382
9925
2374
9892
2374
9892
242393
9971
2416
10067
2409
10038
2400
10000
2401
10004
201974
9872
1981
9905
1972
9860
1972
9860
1972
9860
242403
10011
2414
10058
2410
10042
2401
10004
2401
10004
181835
1019
71816
10089
1808
10044
1807
10039
1807
10039
262637
1014
42645
1017
32637
1014
22625
10096
2625
10096
26261
10039
2519
9688
2591
9965
2593
9973
2593
9973
242433
1013
62426
1010
82393
9971
2413
10054
2413
10054
262618
10069
2604
10015
2573
9896
2594
9977
2594
9977
Mean
10001
9999
10010
10001
10001
SD16
412
110
710
610
7RM
SEC
032
028
022
020
020
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
4 Journal of Chemistry
322 Orthogonal Signal Correction (OSC) and Direct Orthog-onal Signal Correction (DOSC) Numerous algorithms forapplyingOSC as a filteringmechanism to datamatrixX weredescribed in detail in literature [24 26 27] In the presentedwork we refer to the OSC algorithm discussed by Fearn [27]as it is the most easily interpreted and can be compared toNAP methodology
These algorithms are employed to get rid of parts ofthe spectra that are orthogonal to the concentration Fearnrsquosalgorithm spans the same space as PLS or PCRmodels on theunprocessed data and extracts components that are strictlyorthogonal to the concentrationThe first step is a projectionof 119883 onto the subspace that is orthogonal to 119883119879119884
119896
to obtainthe variation in 119883 that is correlated to 119884
119896
(where 119884119896
is con-centration of 119896 analyte) Fearnrsquos algorithm removed almostthe drawbacks that were seen in the other OSC algorithmssuch as lack of orthogonality leading to removal of usefulinformation or adding useless data in the corrected matrix
DOSC approach introduced by Westerhuis et al [28] isbased only on least squares steps It finds components whichare orthogonal to 119884 which label the main variation of 119883 tobe removed from corrected119883 For implementation of DOSCfirstly119884 is decomposed into two orthogonal parts (the pro-jection of 119884 onto119883) and 119865 (the residual part ie orthogonalto119883) Secondly119883 is decomposed into two orthogonal partsone part that is of the same range as and the other partthat is orthogonal to it Finally principal component analysis(PCA) is applied on the part of119883 orthogonal to 119884 to removeit from119883 giving rise to corrected119883
Mathematically DOSC can be described by the followingsteps
Step 1 Step 1 is as follows
119884 = 119875119883
119884 + 119860119883
119884 = + 119865 (2)
Step 2 Step 2 is as follows
119883 = 119875
119884
119883 + 119860
119884
119883 (3)
Step 3 Step 3 is as follows
119883DOSC = 119883corrected = 119883 minus 119905119875119879
(4)
where 119905 is score vector and 119875 is loading vector
The threemodels (NAP-CLSOSC-CLS andDOSC-CLS)are simpler when compared with other methods such as PLSThis simplicity originates from the utilization of orthogonalprojections concepts followed by the well-established classi-cal least squares fitting
(1) Optimization of Number of Factors for the NAP-CLS OSC-CLS and DOSC-CLSModels Leave-one-out cross-validation(LOO-CV) is adopted in the current work for optimization offactors number for construction of the proposedmodels [23]by building the model using 119868 minus 1 samplesrsquo set (15 mixturesfrom training or calibration set) to predict the one sample
left (validation sample) The root mean square error of cross-validation (RMSECV) is computed as follows
RMSECV = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894 CV)2
(5)
where 119868 is the number of objects in the calibration set 119888119894
is the known concentration for sample 119894 and 119860119894 CV is the
predicted concentration of sample 119894 using 119860 componentsMean centering was applied to the calibration set each timesuccessive samples were left out
4 Results and Discussion
The current work was designed to accomplish a number ofgoals First goal is to develop simple selective and precisechemometric models for assaying CEF in presence of itsimpurities in pure form and dosage form Second goal isto demonstrate the quantitative power as well as qualitativepower of the suggested models and compare their inherentcharacteristics using the analyzed mixtures as a case studyAdditionally we aim to display the influence of variouspreprocessing steps such as NAP OSC and DOSC on theperformance of CLS in quantitative analysis Optimizationof methodsrsquo parameters was the first step to run modelsproperly For the SRACLS model the optimum number ofloadings119875new used for augmentation of Kmatrix was selectedthrough running LOO-CV and found to be 4
For appropriate building of NAP-CLS OSC-CLS andDOSC-CLS methods number of projection matrix factors(for NAP-CLS) and number of extracted factors (for OSC-CLS and DOSC-CLS) were adjusted For this purpose LOO-CV was applied where log PRESS (predicted residual errorsumof squares) valueswere computedTheoptimumnumberof factors was chosen in accordance with Haaland andThomas approach [29] In all improved CLS models twofactors were essential for constructing the models except inDOSC-CLS model where three factors were required Thisinformation demonstrates that DOSC as a preprocessingtechnique is more complex than NAP and OSC procedures
After parametersrsquo optimization and training procedureall proposed methods together with ordinary CLS wereapplied successfully for estimation of CEF in training setand test set as indicated in Tables 2 and 3 respectivelyRecovery percentages mean recoveries standard deviationand RMSEC and RMSEP values are anticipated in Tables 2and 3
RMSEC (for training set) and RMSEP (for test set) werecomputed in a similar way according to the following equa-tion
RMSEC (119875) = radic 1119868
119868
sum
119894=1
(119888119894
minus 119860
119894
)2
(6)
where 119868 is the number of samples in the test set (in case ofRMSEP) and 119868 minus 1 is for training set (in case of RMSEC)119888119894
is the known concentration for sample 119894 and 119860119894
is theestimated concentration of sample 119894 using 119860 components
Journal of Chemistry 5
Table2Analysis
results
forthe
predictio
nof
training
set(autopredictio
n)of
CEFby
thep
ropo
sedchem
ometric
mod
els
Training
set
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
181878
10433
1829
1016
11855
10306
1858
10322
1858
10322
181786
9921
1799
9994
1799
9994
1801
10006
1801
10006
20197
9852
1985
9925
1988
9940
1984
9920
1984
9920
201969
9845
1977
9885
1984
9920
1989
9945
1988
9940
262572
9894
2617
10065
2621
10081
2614
10054
2614
10054
201978
9891
1992
9960
1990
9950
1989
9945
1989
9945
181785
9919
1794
9967
1789
9939
1788
9933
1788
9933
242357
9821
2384
9933
2382
9925
2374
9892
2374
9892
242393
9971
2416
10067
2409
10038
2400
10000
2401
10004
201974
9872
1981
9905
1972
9860
1972
9860
1972
9860
242403
10011
2414
10058
2410
10042
2401
10004
2401
10004
181835
1019
71816
10089
1808
10044
1807
10039
1807
10039
262637
1014
42645
1017
32637
1014
22625
10096
2625
10096
26261
10039
2519
9688
2591
9965
2593
9973
2593
9973
242433
1013
62426
1010
82393
9971
2413
10054
2413
10054
262618
10069
2604
10015
2573
9896
2594
9977
2594
9977
Mean
10001
9999
10010
10001
10001
SD16
412
110
710
610
7RM
SEC
032
028
022
020
020
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
Journal of Chemistry 5
Table2Analysis
results
forthe
predictio
nof
training
set(autopredictio
n)of
CEFby
thep
ropo
sedchem
ometric
mod
els
Training
set
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
181878
10433
1829
1016
11855
10306
1858
10322
1858
10322
181786
9921
1799
9994
1799
9994
1801
10006
1801
10006
20197
9852
1985
9925
1988
9940
1984
9920
1984
9920
201969
9845
1977
9885
1984
9920
1989
9945
1988
9940
262572
9894
2617
10065
2621
10081
2614
10054
2614
10054
201978
9891
1992
9960
1990
9950
1989
9945
1989
9945
181785
9919
1794
9967
1789
9939
1788
9933
1788
9933
242357
9821
2384
9933
2382
9925
2374
9892
2374
9892
242393
9971
2416
10067
2409
10038
2400
10000
2401
10004
201974
9872
1981
9905
1972
9860
1972
9860
1972
9860
242403
10011
2414
10058
2410
10042
2401
10004
2401
10004
181835
1019
71816
10089
1808
10044
1807
10039
1807
10039
262637
1014
42645
1017
32637
1014
22625
10096
2625
10096
26261
10039
2519
9688
2591
9965
2593
9973
2593
9973
242433
1013
62426
1010
82393
9971
2413
10054
2413
10054
262618
10069
2604
10015
2573
9896
2594
9977
2594
9977
Mean
10001
9999
10010
10001
10001
SD16
412
110
710
610
7RM
SEC
032
028
022
020
020
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
6 Journal of Chemistry
Table3Analysis
results
forthe
predictio
nof
testseto
fCEF
andits
impu
ritiesb
ythep
ropo
sedchem
ometric
mod
els
Testset
CLS
SRAC
LSNAP-CL
SOSC
-CLS
DOSC
-CLS
Taken(120583gm
Lminus1
)Fo
und(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
Foun
d(120583gm
Lminus1
)119877
252723
10894
2594
10376
2587
10216
2540
1016
02540
1016
019
2058
10829
1933
10174
1972
10016
1901
10005
1901
10005
222416
1098
32277
10350
2288
1018
22230
1013
62230
1013
621
2318
1104
2176
10362
2181
10200
2135
1016
72135
1016
723
2561
11133
2405
10457
2394
10261
2345
1019
62345
1019
625
2792
1116
62625
10500
2564
10308
2558
10232
2557
10228
192108
11095
1955
10289
1959
1014
21924
1012
61924
1012
622
2454
1115
42284
10382
2273
10228
2237
1016
82237
1016
821
2376
11313
2204
10495
2193
10310
2154
10257
2154
10257
Mean
11067
10376
10207
1016
11016
0SD
149
103
091
072
072
RMSE
P237
103
051
039
039
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
Journal of Chemistry 7
Table 4 Statistical comparison of the results obtained by the proposed methods and the reported HPLC method for determination of CEFin pharmaceutical formulation (Cefobid Vial)
Parameter SRACLS NAP-CLS OSC-CLS DOSC-CLS Reported HPLClowastlowast
Mean 10296 10418 10493 10497 10285SD 174 074 196 196 142Variance 301 055 383 382 203119899 6 6 6 6 6Studentrsquos 119905-test (2228) 0119 2017 2097 2145 mdash119865-test (5050) 1485 3673 1889 1889 mdashlowastlowastReference official method is HPLC [9]
Table 5 One-way ANOVA parameters for the different proposed models used for the determination of CEF in Cefobid Vial
Analyte Source of variation DF Sum of squares Mean square 119865-value
CEF Between exp 3 15829 5276 1882Within exp 20 56069 2803
There was no significant difference among the models using one-way ANOVA (119865-test) where 119865 tabulated is 3098 at 119901 lt 005
0
05
1
15
2
25
SRACLSNAP-CLSOSC-CLSDOSC-CLSCLS
RMSEP comparative plot
Figure 4 RMSEP comparative plot for the prediction of test setsamples 1-SRACLS 2-NAP-CLS 3-OSC-CLS 4-DOSC-CLS and 5-CLS models
The small values of the RMSEP indicate the minor error ofprediction and the high predictive ability of the developedmodels Figure 4 shows the RMSEP comparative plot for theprediction of test set samples by the proposed models where allof the four proposed models are performing better than CLSand best results are given by OSC-CLS and DOSC-CLSmodels
The described models were then applied with a greatsuccess for the analysis of the available dosage form (Table 4)This fact was further confirmed by the statistical comparisonof the suggested models to the official HPLC method [9](Table 4) where the calculated 119905 and 119865 values are less than thetabulated ones indicating that there is no significant differ-ence between ourmodels and the referencemethod regardingboth accuracy and precision All themodels were additionallycompared by one-way ANOVA (Table 5) where the calcu-lated 119865-value was less than the tabulated one as well showingthat there is no significant difference among all modelsregarding precision The obtained results suggest the validityof proposed models to be used for routine quality controlanalysis of CEF in pure form and pharmaceutical product
5 Conclusion
In the presented paper different chemometric models wereapplied for the analysis of CEF in presence of its relatedimpurities The proposed models are CLS SRACLS NAP-CLS OSC-CLS and DOSC-CLS The developed modelscombine advantages of rapidity and easiness of traditionalspectrometric methods in addition to their quantitativepower (prediction of concentrations of CEF in its mixtures)The traditional CLS model gave the worst results for pre-diction of independent test set samples while among theproposed models OSC-CLS and DOSC-CLS were the mostpowerful ones that increase the quantitative power of CLSmethod based on the analyzed case study All the suggestedmodels were optimized and validated by prediction of testset samples The methods can be applied for routine qualitycontrol analysis of Cefobid Vials without prior separation orinterference from commonly encountered additives
Competing Interests
No conflict of interests exists in the present paper
Acknowledgments
The authors would like to extend their sincere appreciation tothe Deanship of Scientific Research at King Saud Universityfor its funding of this research through the Research GroupProject no RGP-322 Additionally the authors would like toextend their sincere appreciation to Dr Essraa A Hussein forhelping in preparation of the chemical data
References
[1] M J OrsquoNeil The Merck Index An Encyclopedia of ChemicalsDrugs and Biologicals Merck and Co Whitehouse Station NJUSA 14th edition 2013
[2] J E F Reynolds Martindale The Extra Pharmacopoeia RoyalPharmaceutica Society London UK 31st edition 1996
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
8 Journal of Chemistry
[3] British Pharmacopeia Her Majestyrsquos Stationary Office LondonUK 1980
[4] Q Tan Q Song andDWei ldquoSingle-pot conversion of cephalo-sporin C to 7-aminocephalosporanic acid using cell-bound andsupport-bound enzymesrdquo Enzyme and Microbial Technologyvol 39 no 5 pp 1166ndash1172 2006
[5] M Senthilraja and P Sanjaypai ldquoSpectrophotometric methodfor the determination of cefoperazone sodium in pharmaceu-tical formulationsrdquo Indian Journal of Pharmaceutical Sciencesvol 68 no 3 pp 384ndash385 2006
[6] L Chen O ShaW-J Zhai and J-H Yang ldquoFading spectropho-tometric determination of cefoperazone sodium in injections byFe3+-sulfosalicylic acid systemrdquoAsian Journal of Chemistry vol25 no 14 pp 7918ndash7920 2013
[7] H-H Pang Y-C Feng C-Q Hu and B-R Xiang ldquoConstruc-tion of universal quantitative models for determination of cef-operazone sodium for injection from different manufacturersusing near infrared reflectance spectroscopyrdquoGuang Pu Xue YuGuang Pu Fen Xi vol 26 no 12 pp 2214ndash2218 2006
[8] A Parra J Garcia-Villanova V Rodenas and M D GomezldquoFirst and second derivative spectrophotometric determinationof cefoperazone and sulbactam in injectionsrdquo Journal of Phar-maceutical and Biomedical Analysis vol 12 no 5 pp 653ndash6571994
[9] The United States Pharmacopoeia (USP30) National Formu-lary (NF25) The United States Pharmacopeial ConventionRockville Md USA 2007
[10] F S Li Z X Xu H B Xiao and X M Liang ldquoSimultaneousdetermination of sulbactam sodium and cefoperazone sodiumin sulperazon by high performance liquid chromatographyrdquo SePu vol 18 no 6 pp 525ndash526 2000
[11] K V Kumar J Dharuman and A K Sree ldquoRP-HPLC methoddevelopment and validation for simultaneous estimation ofsulbactam and cefoperazone in dosage form and in plasmardquoInternational Journal of Pharma and Bio Sciences vol 1 no 4article 12 pp 88ndash96 2010
[12] T-L Tsou Y-C Huang C-W Lee A-R Lee H-J Wang andS-H Chen ldquoSimultaneous determination of ampicillin cef-operazone and sulbactam in pharmaceutical formulations byHPLCwith 120573-cyclodextrin stationary phaserdquo Journal of Separa-tion Science vol 30 no 15 pp 2407ndash2413 2007
[13] Y Zhou J Zhang B Guo et al ldquoLiquid chromatographytan-dem mass spectrometry assay for the simultaneous determina-tion of cefoperazone and sulbactam in plasma and its applica-tion to a pharmacokinetic studyrdquo Journal of Chromatography Bvol 878 no 30 pp 3119ndash3124 2010
[14] E Hammam M A El-Attar and A M Beltagi ldquoVoltammetricstudies on the antibiotic drug cefoperazone quantificationand pharmacokinetic studiesrdquo Journal of Pharmaceutical andBiomedical Analysis vol 42 no 4 pp 523ndash527 2006
[15] B Dogan A Golcu M Dolaz and S A Ozkan ldquoElectrochem-ical behaviour of the bactericidal cefoperazone and its selectivevoltammetric determination in pharmaceutical dosage formsand human serumrdquo Current Pharmaceutical Analysis vol 5 no2 pp 179ndash189 2009
[16] E A Abdelaleem I A Naguib H E Zaazaa and E A HusseinldquoDevelopment and validation of HPLC and HPTLC methodsfor determination of cefoperazone and its related impuritiesrdquoJournal of Chromatographic Science vol 54 no 2 pp 179ndash1862015
[17] I A Naguib E A Abdelaleem H E Zaazaa and E A HusseinldquoDetermination of cefoperazone sodium in presence of relatedimpurities by linear support vector regression and partial leastsquares chemometric modelsrdquo Journal of Analytical Methods inChemistry vol 2015 Article ID 593892 8 pages 2015
[18] R G Brereton ldquoMultilevel multifactor designs for multivariatecalibrationrdquo Analyst vol 122 no 12 pp 1521ndash1529 1997
[19] ACOlivieriHCGoicoechea and FA Inon ldquoMVC1 an inte-grated MatLab toolbox for first-order multivariate calibrationrdquoChemometrics and Intelligent Laboratory Systems vol 73 no 2pp 189ndash197 2004
[20] D M Haaland and D K Melgaard ldquoNew prediction-aug-mented classical least-squares (PACLS) methods application tounmodeled interferentsrdquoApplied Spectroscopy vol 54 no 9 pp1303ndash1312 2000
[21] W Saeys K Beullens J Lammertyn H Ramon and T NaesldquoIncreasing robustness against changes in the interferent struc-ture by incorporating prior information in the augmented clas-sical least-squares frameworkrdquoAnalytical Chemistry vol 80 no13 pp 4951ndash4959 2008
[22] I ANaguib ldquoStability indicating analysis of bisacodyl by partialleast squares regression spectral residual augmented classicalleast squares and support vector regression chemometric mod-els a comparative studyrdquo Bulletin of Faculty of Pharmacy CairoUniversity vol 49 no 2 pp 91ndash100 2011
[23] R Kramer Chemometric Techniques for Quantitative AnalysisCRC Press 1998
[24] S Wold H Antti F Lindgren and J Ohman ldquoOrthogonal sig-nal correction of near-infrared spectrardquo Chemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 175ndash185 1998
[25] H C Goicoechea and A C Olivieri ldquoA comparison of orthog-onal signal correction and net analyte preprocessing methodsTheoretical and experimental studyrdquo Chemometrics and Intelli-gent Laboratory Systems vol 56 no 2 pp 73ndash81 2001
[26] J Sjoblom O Svensson M Josefson H Kullberg and S WoldldquoAn evaluation of orthogonal signal correction applied tocalibration transfer of near infrared spectrardquoChemometrics andIntelligent Laboratory Systems vol 44 no 1-2 pp 229ndash244 1998
[27] T Fearn ldquoOn orthogonal signal correctionrdquo Chemometrics andIntelligent Laboratory Systems vol 50 no 1 pp 47ndash52 2000
[28] J A Westerhuis S de Jong and A K Smilde ldquoDirect orthogo-nal signal correctionrdquo Chemometrics and Intelligent LaboratorySystems vol 56 no 1 pp 13ndash25 2001
[29] DMHaaland andEVThomas ldquoPartial least-squaresmethodsfor spectral analyses 1 Relation to other quantitative calibrationmethods and the extraction of qualitative informationrdquoAnalyt-ical Chemistry vol 60 no 11 pp 1193ndash1202 1988
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Inorganic ChemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
International Journal ofPhotoenergy
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Carbohydrate Chemistry
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Physical Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom
Analytical Methods in Chemistry
Journal of
Volume 2014
Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
SpectroscopyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Chromatography Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Theoretical ChemistryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Spectroscopy
Analytical ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Quantum Chemistry
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Organic Chemistry International
ElectrochemistryInternational Journal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
CatalystsJournal of
top related