j. Soc.Cosmet.

Chem.,
47, 325-336 (September/October
1996)

Tretinoinassayin cosmeticsand pharmaceuticals

by carbon
phaseextraction
G. RAGNO, M. VERONICO, R. MADDALENA,andC. VETUSCHI,Pharmaco
Chemistry
Department
of University,
Via E. Orabona
4, 70126 Bari, Italy.
Accepted
forpublication
September
30, 1996. Presented
at IV CRSAAE
Bologna,
Italy, October
1991.
Synopsis

A tretinoinassay,
usefulto verifyits absence
in cosmetics,
wherethe drug is forbidden,wasproposed.
The
methodwasbasedon a carbonphaseextractionthat allowsoneto obtainthe analytefreefrom interfering
matrix componentsand in an enrichedconcentrationin sucha way as to attain a higher sensitivity.When
the analytewaspresent,a quantitationwasperformedby UV-derivativespectrophotometry.
In the presence
of substances
with a high absorptivity,
like sunfilters,besidesthe carbonextractionstep,a furtherseparation by HPTLC wasrequired.The drugabsence
wasvalidatedover0.1 mg/100g.Sincethe TRT amountin
pharmaceuticals
for topicaluseis 10 to 100 mg/100g,this revelationlimit assures
a sufficientwarranty.A
simplifiedprocedure,
by directderivativespectrophotometry,
maybe appliedfor tretinoindeterminationin
pharmaceuticals.

INTRODUCTION

Tretinoin (TRT), all-trans retinoic acid, is usedfor the treatment of severecystic acne
(1-2) and other skin diseases
(3), for its positiveaction on cell proliferationand keratinizationof the skin, aswell asfor decreasing
sebumsecretionand inflammation(4).
Researchwith retinoidsand clinical observations
in humanssometimesgivesapparently
contradictoryresults:while stimulating the proliferation of normal epidermal cells,
retinoidsmay checkthe growthof psoriaticcellsand neoplasms,
but the sideeffectsof
thesecompoundsare fairly unpleasant(3). Sun exposureduring treatmentmust be

stringentlyavoided.Teratogenicity
for isotretinoinis well documented
(5,6), and it is
suspectedas a potential effectof tretinoin. For thesereasons,the use of tretinoin in
pregnancyis forbidden.

On 27 July 1976 theEuropean

Communitypromulgated
Law76/768, actuatedin Italy
with Law 11 October 1986 no. 713, which forbids the use of TRT and several other

chemicalsin the cosmeticfield. Furthermore,in Italy the Ministry of Health, with Circular 18 October 1990 no. 27, hascommitted the public laboratoriesto checkfor the
absence of TRT

in cosmetics.

Analyticalmethodsfor TRT havebeendevelopedby usingHPLC on biologicalsamples

(7) and on anti-aging cosmetics(8,9); the U.S. Pharmacopeia23rd Rev. (10) reportsa
325

326

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

chromatographic
assayon creamsand gels calculatingthe analyteamount in comparisonwith a standardpreparation.
In this paper a procedureis describedfor routine control of skin-carecosmetics,like
creams,gels,lotions,andointments,ableto revealthe possiblepresence
of TRT in very
low concentration.Due to the interferencesof the complexcosmeticformulations,a
treatment was required to simplify the matrix and to enrich the TRT concentration
level in the sample.For this purpose,the cosmetic,dispersedin tetrahydrofuran,was
treatedwith a carboncolumn. After removingexcipientsamongthe carbongranules
and washingwith repeatedamountsof chloroform,pyridinewasfinally usedas analyte
eluent. TRT determinationwas performedon the pyridinic solutionby UV-derivative
spectrophotometry.

In the caseof cosmeticscontainingUV filters highly overlappingthe UV curve of

TRT, a full fractionation with an alternative method was needed. Therefore, a further

procedureby HPTLC densitometrywasdeveloped,performingthe acidificationof the

pyridinic eluate with hydrochloricacid and the analyte extraction with methylene
chloride.

It is plain that, in any case,when the UV signalsare not clearlyascribedto TRT, due to
whateverinterference,
the analytepresence
canbe assuredby the HPTLC method.Several attempts to apply an HPTLC procedurewithout a prior fractionationand analyte
enrichmentof the samplesprovedto be unsuccessful.

The derivativespectrophotometric
methodwasalsosuitablefor TRT assayin pharmaceutics,directly on a tetrahydrofuransuspension
of the sampleswithout any removalof
the other components.

EXPERIMENTAL
MATERIALS

Tretinoinwaspurchased
from AldrichChemicals(USA);HPTLC plates(Kieselgel60, 10
)< 10 cm) fromMerck (Germany);andnylonmembranefilters,0.45-1m-poresize,from
Whatman(England).All solvents,
suppliedby C. Erba(Italy),wereof analyticalgrade.
Active carbontypesstudiedare reportedin TableI. Ointment and creambases,usedfor
recoverystudies,weremarketedby Resriva(Italy) and Schering-Plough
(Italy).

Table I

CarbonTypesInvestigated
Proprietaryname

Particlesize

Supplyingfirms

Granular

1.5 mm

Merck (Germany)

Granular

2.5 mm

Merck

Darco

4-12 mesh

Aldrich (USA)

Darco

12-20

mesh

Aldrich

Darco

20-40

mesh

Aldrich

Norit RB1
Norit R0

0.6 pellets
0.8 pellets

Aldrich
Aldrich

TRETINOIN

ASSAY BY CARBON

PHASE EXTRACTION

327

Pharmaceuticalforms assayedwere: Retin-A cream 0.010%, 0.025%, 0.05%, gel

0.025%, and lotion 0.05 mg/ml (Cilag, Switzerland);Airol cream 0.05% and lotion
0.05% (Roche,Switzerland);and Apsorointment 0.1% (IDI Farm., Italy).

The following commonlyusedcreamand ointment excipientswere used:3-butyl-4hydroxyanisole,

diethyleneglycolmonoethylether,dimethyl polysiloxane,glycerylbehenate,glyceril monostearate,
glycerol,isopropylmyristate,lanolin, lanolin isopropyl
esters,oleic acid esters,paraffin,perfluoropoliether,
polyoxyethylen40 stearate,perhydrosqualene,propyleneglycol, saturatedfatty acids, triglycerides,saturatedpolyglycolyzedglycerides,sorbitol70%, xanthangum, and white wax. Severalformulations
with varying concentrations
of thesesubstances
were preparedwith the addition of
known amounts of TRT.

The following products,variouslyemployedin commercialanti-aging creams,were

addedto the basecreamsto studythe potentialinterference:
allantoin,p-aminobenzoic
acid, benzoicacid, camphor,N-dimethyaminobenzoic
acid, retinol, dl-o-tocopherol,
salicylicacid, salicylicesters,sorbicacid,and stearicacid.

The UV filters investigated

were 2,2'-dihydroxhy-4,4'-dimethoxy-benzophenone,
2ethylhexyl-p-methoxycinnamate,
1-(4-methoxyphenyl)-3-tert-buylphenyl)
propan-1,3dione,3-(4-methylbenzylidene)-camphor,
and octyl-dimethylp-aminobenzoicacid.
APPARATUS

Spectrophotometry.
Spectrawere recordedover the wavelengthrangeof 450-250 nm in
10-mm silicaquartzcellsusinga Perkin-ElmerLambda16 spectrophotometer;
scan
speed2 nm/s; response
(time constant)1 s for zero-orderand 5 s for second-and third-

orderderivative
spectra;
spectral
bandwidth1 nm; A) 8 for bothderivativeorders.The
spectrawereelaboratedwith PECSS4.0 softwareby PerkinElmer.
Densitometry.
Measurements
wereperformedwith a Shimadzu(USA) modelCS930. Experimentalconditionswere:absorptionreadingmodeat 370 nm; scanningspeed(linear)1 mm/s;recorderbaseline
200 mV; beamsize0.4 X 0.4 mm. The development
was
performedin a lineardevelopment
chamber(Camag,Switzerland).
Carboncolumns.
Glasstubesof 30 X 0.5 cm (i.d.) were used,with a capillary end of
0.05 cm (i.d.) to optimize the solventflow. Three to five grams of carbonwere
washed,refluxing in Soxheletwith 200 ml of chloroformfor 12 hours, then vacuumdried at room temperatureuntil a constantweight. One gram of dry carbon,accuratelyweighed,wastransferredandgently packedin onecolumn,obtaininga granular phaseheight of 14 __+0.5 cm. For this column the intra- and inter-particles
volumesprovedto be 0.2 and 0.8 ml with THF, and 0.7 and 1.5 ml with pyridine,
respectively.
LABORATORY

PRECAUTIONS

All the assayprocedureswere carriedout in a dark room providedwith a red lamp

of 60 W kept at a distanceof 2 metersto avoidphotodegradation
of the retinoicacid
(11,12). The carboncolumnswereprotectedwith tin foil.

328

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

STANDARD

SOLUTIONS

A stocksolutionof TRT wasprepared,dissolving100 mg, accuratelyweighed,in a

250-ml volumetricflaskwith tetrahydrofuran.
Twentyworkingstandardsolutionswere
preparedby diluting this stocksolutionwith THF to obtainan analyteconcentration
rangingbetween0.1 and 60 g/ml. Analogous
standardsolutionsin pyridinewereprepared.
For HPTLC analysis,ethanolicsolutionswerepreparedin sucha way that, loadingon
the plate volumesof 100 nl, the analyteamountwasbetween0.01 and 30 g/deposit.
Leastsquareslinear regression
wasappliedto fit plots of signalvaluesversustheoretical
concentration.

METHODS
COSMETICS

Carbonextraction.
A dispersionof 3 g of the sample,accuratelyweighed,in 20 ml of
THF, wasloadedon the carboncolumnand flushedabout 0.5 ml per minute. The column was washedwith portionsof 2 ml of chloroformfor a total of 50 ml, then discarded.The analytewas eluted with portionsof 1 ml of pyridine for a total of 10 mi.
The pyridineflow wasnot to be higherthan0.1 ml per minute.
UV procedure.
If the cosmeticwasa dermoprotective
formulationfor generaluse,TRT
determinationwasdirectlyperformedon the pyridiniceluateby derivativespectrophotometrythroughthe 389-nm and 360-410-nm signals,in secondand third derivative
spectra,respectively.
In the presence
of sunfilters,a furtherseparation
with HPTLC was
needed.

HPTLC procedure.
The solutionwasfiltered througha 0.45-ptm-pore-size
membrane,
acidifiedwith hydrochloricacid 37% (sample/HC11:1.25 v/v) and then extractedwith
methylenechloride5 X mi. The extractswereevaporatedundera gentlestreamof nitrogenand the residuedissolvedin 100 p,l of ethanol.Volumesof 100 nl were spotted
1 cm from the bottomof the plate and 0.5 cm apartand developed
with a hexane/acetonemixture(6:4 v/v). In theseconditionsTRT presenteda Rf valueof 0.72.
PHARMACEUTICALS

An amountof samplecontaininga declaredamountof 0.5 mg of TRT wasaccurately

weighedand transferred
into a 25-ml volumetricflaskcontaining10 ml of tetrahydrofuran. The flask was vigorouslyshaken,and the suspension,
diluted to volume, was
spectrophotometrically
analyzedby usingthe peak-trough363-413 nm in the thirdorderderivativespectrum.
LABORATORY

SAMPLE

Syntheticpreparationswere made by spiking ointment and creambaseswith TRT to

simulatesampleswith analytelevelswithin the rangeof 0.05 to 200 mg/100g. These
sampleswere usedto establishthe accuracyof the method. Interferencestudieswere
performed,adding the above-mentioned
excipientsand UV filters in variousmixtures
and concentrations.

TRETINOIN
RESULTS

AND

ASSAY BY CARBON

PHASE EXTRACTION

329

DISCUSSION

The commercialformsof skin-carecosmetics

containexcipientsof very differentpolarities that make it very difficult for the preparationof an analyticalsample.Among the
high numberof solventstested,tetrahydrofurangavesolutionsor cleardispersedphases.
Unfortunately,the absorbance
spectrain THF, for the commercialsamplesinvestigated,
werenot usefulfor TRT assay,
due to othercomponents
totally or partially overlapping
the analytesignal(353 nm) or the absorbance
increase
by turbidity background.
In thesecases,a derivativespectrophotometric
methodappearedvery helpful,sincethis
techniqueallows one to obtain a resolutionenhancementof the spectralcurve and,
without prior separation
of insolubleexcipients,to eliminatebroadabsorptionbandsresultingfrom matrix turbidity.

Forpharmaceutical
samples,
the peak-trough363413 nm in the third-orderderivative
spectra(Figure 1) wasfoundto be very highly correlatedwith the analyteconcentration
and not influencedby the othercomponents.The relativeregression
equation,obtained
from data of twenty THF standardsolutions,is reportedin TableII.

Severalattemptsto apply this procedureto the analysisof cosmetics

wereunsuccessful,
dueto the fact that in all casesthe TRT signalwascompletelyoverlappedby interfering
excipients.
Therefore,a treatmentof the THF samples
wasnecessary.
Since there are few referencesfor TRT extraction from whole creams(13,14), different
procedureswere investigated.When the THF suspensionsampleswere loaded on
columnspackedwith differentphases,the activecarbonturned out to be a very selective materialfor TRT, beingverystronglyadsorbed
on this phase.

Preliminaryexperimentsto choosethe right carbonweremade,discardingat first those

with a granulometryover 50 mesh to prevent the column occlusion.Therefore,for
sevencarbontypes,reportedin Table I, the adsorbability(15) to TRT was evaluated.
Ten milliliters of a THF solutionof TRT 10 g/ml were loadedon columnspacked
with 1 g of eachdry carbontype, the eluatecollectedin a 10-ml volumetricflask and
broughtto volumewith THE The amountof adsorbed
TRT wascalculatedasthe differencebetweenthe originalsampleandthe eluateconcentrations.
In a secondstep,varioussolventsweretestedto selectthe mostefficientoneto recover
TRT from the column.Pyridinewasfoundto performthe highestrecoveryfor all carbon
types,with valuesrangingfrom 30 to 95%. On the contrary,chloroform,which presentednull extractionof TRT, waschosento washthe columnbeforethe analyteelution.
The obtainedresults,shownin the graph of Figure 2, led to selectingthe carbontype
Darco 20/40 mesh (Aldrich), showingadsorbabilityand recoveryvaluesof 97% and
90%, respectively.
Afterwards,sinceTRT recoverywas demonstratedto be dependent
on the elution speed,a recoveryvalue of 98% was achieved,optimizing the pyridine
elutionspeedat about6 ml/hour.

The pyridineeluatesobtainedby applyingthis extractionprocedureon cosmeticformulationsshowtwo signals,the 389-nm maximumin the second-order
andthe 360-410nm peak-troughin the third-orderderivativespectra(Figure3), due only to the analyte
concentrationand not influencedby other components.These signal valuesand the
drug concentrations
were correlatedthroughthe regression
equationreportedin Table
II. The determinationlimit for this spectrophotometric
method was calculatedto be
0.3 g/ml.

330

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

,'/

V,,

', ,./
250

300

350

400

450

daA/dL3

I/ \,

250

300

350

400

450

Wavelengths
Figure 1. Absorbance
andthird-orderderivativespectraof pureTRT (

andApsor
ointment
(.......

) intetrahydrofurane.

), Retin-A
cream
(....

),

TRETINOIN

ASSAY BY CARBON

PHASE EXTRACTION

331

Table II

CalibrationGraphsfor TRT Determination

Method

3D363413
(THF)

2D89'(Pyridine)
3D360,410
(Pyridine)

HPTLC (areaRf7.2)

Slope(_+ SD)

Intercept (+_ SD)

331.37+ 23.45

--0.0151--0.011

0.541 0.23

0.9993

416.66+ 34.90

-0.217 + 0.15

0.9991

67.11 5.87

1752.81 -+ 86.65

81.243 + 7.39

0.9999

0.9990

TRT concentrationis expressed

as g/ml in UV-derivativemethodand g in HPTLC method.

In orderto studythe optimalratio sample/carbon,

amountsfrom 0.1 to 6 g of a cream
with TRT 0.01% were suspended
in 10 ml of THF and loadedon columnscontaining
1 g of activatedcarbon.The first resultshowedthat the samples
over3 g wereto bediscardedbecauseof seriousproblemsof column occlusion.The packingswere washed
with 20 ml of chloroformand then eluted with increasingpyridine volumes.The least
volumeable to elute the adsorbedanalyte(96-98%) from all the columnswasverified
not to be less then 10 ml.

Therefore,in orderto detecta TRT concentrationas low aspossible,a standardanalyticalprocedurewasdefined,fixing a ratio sample/carbon
of 3:1 (w/w) anda pyridineelution volume of 10 mi. In theseconditions,TRT can be determinedto a percent of
0.0001, revealingthe analyteconcentration
of the relativecollectedsolutionto be 0.3
Ig/ml (UV determinationlimit).

Recovery
80-

Figure 2. Adsorbabilityto TRT andrecovery

of TRT by pyridineelutionfor variouscarbontypes.

332

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

389

d'A/d)

250

350

400

450

[-3
360
-!-

I
I

daA/d.a

I
I

250

I ,I

300

350

400

450

Wavelengths
Figure
3.
Secondand
third-order
derivative
sectra
ofpyridinic
eluates
from
carbon
column
forpure
TRT
(
) and a cosmeticformulation(Resriva baseointment)(....
), spikedwith TRT.

TRETINOIN

ASSAY BY CARBON

PHASE EXTRACTION

333

24O
I

_I

160 -

120 -

80-

40-

0
0

60

120

180

240

I
300

360

Eluate fractions (mL)

Figure 4. TRT recovery
of sequential10-ml eluatefractionsfrom 1-g carboncolumn,loadedcontinuously
with a TRT 208 xg/ml solution.

The adsorbance
capacityof the chosencarbonto TRT wascalculatedeluting a 1-g carboncolumnwith a TRT 208-g/ml solutionin THF andcollectingsequential
portions
of 10 ml. Figure4, whereTRT recovery
of the eluatefractionsversusvolumesareplotted, showsthe completeadsorptionfor aboutthe first 80 ml, corresponding
to --16.5
mg, and the carbonsaturationafter --270 mi. Consideringthat pharmaceuticals
contain at most0.5 mg/ml, this capacityvalueassures
completeTRT adsorptionwhen applying the proposedprocedure.
For some sun-protectiveformulationsthe carbonextractionwas not able to eliminate
completelythe signal interferences.
In fact, in thesepreparationsthe sun filters are

presentin high amounts(5-10%), and theyare not at all adsorbed

on the carbonphase.
Amongthesunfiltersexamined,
the 2,2'-dihydroxhy-4,4'-dimethoxy-benzophenone
and
the 1-(4-methoxyphenyl)-3-tert-buthylphenyl)
propan-l,3-dionepresentedUV curves
overlapping
TRT absorbance
signals,whereas
the otheronesdid not absorbover350 nm.
For thesecases,a furtherseparation
by HPTLC wasdefined.To removeany suspended
particles,the pyridinic eluate was filtered through a 0.45-m-pore-size membrane,
acidifiedwith HC1, and then extractedwith methylenechloride.The extractswere

evaporated
underN2, andtheresidue,
dissolved
in ethanol,wasloadedon theplate.A
good separationof TRT (Rf 0.72) was obtainedwith an eluent acetone/hexane
mixture

(6:4),andits quantitative
evaluation
wascarriedoutby correlating
theanalyteconcentrationwith the spotareathroughthe regression
equationreportedin TableII.

334

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

In any case,the presence

of the TRT signalsin the second-and third-orderderivative
spectraallowsone to quantifythe analytethroughthe relativeequationsreportedin
TableII. When thesesignalsareabsentor they areinfluencedby whateverinterference,
the analytepresence
canbe assured
by applyingthe HPTLC method.

VALIDATION

The accuracyof the methodswasdeterminedby applyingrecoverystudieson cream

basesto which known amountsof TRT over the range of 0.05 to 200 mg/100g had
beenadded.The precisionof the methodswas determinedby assayingfive replicated
analyses
of the samples,obtainingRSD% valuesfrom 1.0 to 5.3%. The results,shown
in Table III, demonstrategoodrecovery,
with meanvaluesof 99.3 +- 2.3% for the diTable III

TRT AssayResultsin CreamBases

andCommercial
Cosmetics,
SpikedWith VaryingAmountsof Analyte,
by UV-Derivative and HPTLC Methods
Added

(mg/100g)

UV-Derivative

Founds

HPTLC

RSD%

Found

RSD%

Cream bases

Restiva

Resriva

cream base

ointment

base

Schering-Plough
Essexbase

2.03

1.95

4.87

1.90

3.23

20.32

19.71

3.10

19.31

5.09

203.24

201.21

1.56

192.05

4.63

0.51
5.08
50.82

0.49
4.89
48.39

5.23
2.89
3.67

0.46
4.72
47.77

5.10
4.23
4.84

0.10
1.02
10.24

0.10
0.99
10.04

3.52
4.56
2.89

0.09
0.95
9.62

5.48
4.69
6.23

Commercial
samples
Korff "Antia 45 "

N.E b
0.55

Vichy "Day cream"

N.F.
1.06

Medestea"S.Angelica"

-10.65

Restiva"Angstrom"

-50.10

Phas"Exigence"

ROC "Myosphere
"

5.24

N.E
10.33

-3.25

-4.52

-3.69

-2.89

N.E
0.98

N.E
4.99

N.E
10.12

-5.23

-4.98

-5.64

-3.89

--

N.E

--

2.57

47.30

4.28

N.F.

--

97.20

3.74

--

N. E

--

0.51

N.E

--

20.50

N.E

48.01

100.20
2.73
L' O real "Plenitude "

1.02

N.E
5.46

Rydelle"RicercaDerm"

0.53

2.62
N.F.

19.65

-5.65
--

4.12

N.E
95.23

N.E
2.51
N. E

19.35

-4.34

-5.82
--

5.71

SAverage
of five determinations.
bN.E: not found.

Manufacturers:
Resriva(Milan, Italy); Schering-Plough
(Milan, Italy); Korff (Vicenza,Italy); Vichy (CosmetiqueActiveFrance,Levallois-Perret,
France);Medestea(Turin,Italy); Rydelle(Johnson
Wax, Racine,WI);
Phas(PhasC.A.I., Paris, France);ROC (Colombes,France).

TRETINOIN

ASSAY BY CARBON

PHASE EXTRACTION

335

Table IV

TRT AssayResultsin CommercialPharmaceuticals

by UV-Derivative Method
Sample

Nominal (mg/100g)

Founda

RSD%

Airol cream

50.00

49.20

2.03

Airol lotion

50.00

51.02

2.41

Apsorointment

10.00

9.68

1.82

Retin-A

10.00

9.70

2.25

Retin-A cream
Retin-A cream

25.00
50.00

24.53

1.96

48.45

3.15

Retin-A gel

25.00

25.58

2.63

Retin-A lotion

50.00

49.30

1.56

cream

aAverageof five determinations.

rect spectrophotometric
analysis,97.8 --- 4.2% for the carbonextractionprocedure,and
95.4 + 5.2% for the HPTLC assay.
Severalsubstances,
reportedabove,wereaddedwith
varying concentrations
to the cream basesand to severalcommercialformulations,
showingin all casesno interference
with the TRT assay.

The linearityfor UV analysiswascarriedout by analysisof twentyTRT standardsolutions in THF and pyridine over the rangeof 0.3 to 60 pg/ml. The correlationcoefficientswere not lessthan 0.998. Analogousresultswereobtainedby analyzingtwenty
creamsamples,spikedwith TRT between0.1% and 200 mg/100 g, by direct spectrophotometricanalysisandcarbonphaseextraction.
Assumingthat the signal-to-noise
ratioshouldbe at least3, the determinationlimit for
the spectrophotometric
methodboth in THF and pyridine solutionswas calculatedto
be 0.1 mg/100 g. For the HPTLC method,detectionand determinationlimits proved
to be 0.05 and 0.1 mg/100 g, respectively.
The UV and HPTLC methodswere applied to severalcosmeticsand to commercially
availablepharmaceutics.
All the cosmetics
tested,reportedin TableIII, did not present
an appreciableTRT amount.For pharmaceutics,
the concentration
valuesfound were in
goodagreementwith the declaredamounts(TableIV).

CONCLUSION

The proposedcarbonphaseextractioncombinedwith the UV-derivativespectrophotometricanalysiswasfoundto be suitableto separate

and determinetretinoinaccurately
and selectivelyin pharmaceuticaland cosmeticpreparations.The appropriatechoiceof
the analytic conditionsallowedfor an accuratedeterminationof the analyteat a very
low limit (0.1 mg/100 g).

ACKNOWLEDGMENTS

This work was carriedout with grantsfrom CNR (ConsiglioNazionaledelle Ricerche)

andMURST (Ministerodell' Universite dellaRicercaScientifica

e Tecnologica)
of Italy.

336

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

REFERENCES

(1) G. L. Peck,T. G. Olsen,F. W. Yoder,J. S. Strauss,D. T. Downing,M. Pandya,D. Butkus,andJ.

Arnaud-Battanier,
Prolonged
remission
of cystisacheandconglobate
achewith retinoicacid,N. Engl.
J. Med., 300, 329-333 (1979).
(2) W. Bollag,C. E. Orfanos,O. Braun-Falco,
E. M. Farber,Ch. Grupper,M. K. Polano,andR. Schuppli,
Retinoids:
Advances
in BasicResearch
andTherapy
(Springer,Berlin, ! 98 !), pp. 7-11.
(3) W. Bollag,Ed., Retinoids:
NewTrends
in Research
andTherapy
0. H. Saurat,Basle,1985),pp. 274-277.
(4) R. Hermitre,Agedskin,retinoidsandalphahydroxyacids,Cosmet.
Toilets,107, 63-67 (1992).
(5) E.J. Hixson,J. A. Burdershaw,E. P. Denine, and S. D. Harrison,Comparativesubchronictoxicity of
all-transand 13-cisretinoicacid in Sprague-Dawley
rats, ToxicoLAppl. PharmacoL,
47, 359-365
(1979).

(6) D. M. Kochar,J. D. Penner,andC. I. Tellone,Comparative

teratogenic
activitiesof two retinoids:Effectsonpalateandlimb development,
Teratogen.
Carcinogen.
Mutagen.,
4, 377-387 (1984).
(7) R. WyssandF. Bucheli,Determination
of highlyproteinbounddrugsin plasmausinghigh-performanceliquid chromatography
and columnswitching,exemplifiedby the retinoids.
J. Chromatog,
456,
33-43 (1988).

(8) L. Gagliardi,A. Amaro, L. Turchetto,D. Tonelli, and F. Chimenti, Identificazionee dosaggiodella

tretinoinanei prodotticosmeticimedianteHPLC, Costa.
DermatoL,5, 19-24 (1989).
(9) M. B. Kril, K. A. Burke,J. E. Dinunzio, and R. R. Gadde,Determinationof tretinoin in creamsby
high-performance
liquid chromatography.
J. Chromatog,
522,227-234 (1990).
(10) United StatesPharmacopeia,23rd ed. (United StatesPharmacopeiaConvention,Rockville, MD,
1995).

(11) G. M. LandersandJ. A. Olson,Absenceof isomerization

of retinyl palmirate,retinoland retinalin
chlorinated
andnonchlorinated
solvents
undergoldlight,J. Assoc.
Off Anal Chem.,
69, 50-54 (1986).
(12) L. Bonhomme,B. Duleba, T. Beugre,and G. Fredj, HPLC determinationof the stabilityof retinoic
acidin gel formulation,Int.J. Pharm.,65, R9-R10 (1990).
(13) S.C. Kundu, A.D. Cameron,N.M. Meltzer, and T. W. Quick, Developmentand validation of
methodfor determination
of in vitro release
of retinoicacidfromcreams,DrugsDev.Ind. Pharm.,19,
425-438 (1993).

(14) M. J. Lucero,J. Vigo, andM. J. Leon,Stabilityof hydrophilicgelsof tretinoin,Int.J. Pharm.,110,

241-248 (1994).

(15) I. Abe, H. Kamaya,and I. Ueda,Adsorptionof localanesthetics

on activatedcarbon:Freundlichadsorptionisotherms,J.Pharm.Sci.,79, 354-358 (1990).