Effect of topically applied lipids on surfactant-irritated skin

Hritish Journal of Dermatology 1996: 1 3 4 : 2 1 5 - 2 2 0 . Effect of topically applied lipids on surfactant-irritated skin M.LODEN* AND A.-C.ANDERSSONf *ACO HUD AB. Research and Development, and Department of Dermatology, University Hospital, Uppsala, Sweden -fKarlshamns AB. Research and Development, S~i74 82 Karlshamii Sweden Accepted for publication 2 i March 1995 Summary Moisturizers are used daily by many people to alleviate symptoms of dry skin. All of them contain lipids. It has heen suggested that topically applied lipids may interfere with the structure and ftmction of thepertiicability barrier. The ititluence of a single applictitloti of nine different lipids on normal skin and skin irritated by sodium lauryl sulphate {SLS) was studied in 21 healthy subjects. Parameters assessed were visible signs of irritation, and objectively measured cutaneous hlood How and transepidermal water loss (TEWL). The substances tested were hydrocortisone. petrolatum, tish oil. borage oil. sunflower seed oil, canola oil, shea butter, and fractions of unsaponifiable lipids from canola oil and shea butter. Water was included as a control. On normal skin, no signilicant differences iti the eftecls of the test substances were found, whereas signilicant differences were observed when they were applied to SLS-irritated skin. The visible signs of SLS-induced irritation were significantly less pronounced after treatment with the sterol-enriched fraction from canola oil than yfter treatment with water. This fraction, and hydrocortisone. reduced cutaneous blood flow. Furthermore, application of hydrocortisone. canola oil, and its sterol-enriched fraction, resulted in significantly lower TEWL than with water. The other lipids had no effect on the degree of irritation. In conclusion. lipids commonly used in moisturizers may reduce skin reactions to irritants. Previous studies have shown that, in harrier perturbed skin, the synthesis of sterols is Increased. The observed effects of canola oil and its fraction of unsaponitiablc iipids on SLS-induced irritation suggest the possibility that they assisted the skin in supplying the damaged barrier with adequate lipids. One of the key functions of the epidermis is to form a barrier between the organism and the outside world. Lipids in the stratum corneum play a significant role in mainUiining the permeability barrier of the skin. " The lipids are a mixture of sphingolipids, cholesterol and free fatty acids, which form intercellular membrane bilayers.' ''^ Changes in the lipid composition are associated with different skin symptoms. Deprivation of essential falty acids causes a scaly eczematous skin.^ Likewise, exposure of normal skin to organic solvents, stich as ether and acetone."''"' or detergents^** will remove lipids and affect the barrier integrity."^ Along wilh the depletion of lipids, the multiple lamellar structures hetween the corneocytes are disrupted.'"'' ' Topical application of lipids might restore the barrier function of damaged skin. The syndrome of essential fatty-acid deficiency (EEAD) in humans cati be readily reversed by topical application of sunllower seed oil, rich in linoleic acid.^^ The correction of the barrier in Correspondence: Dr M.Loden. ACO HDD AB. Research and Development, PO Box 542, S-1K2 1 5 Danderyd. Sweden. 1996 British Association ofDermatologists EFAD could be due either to reversal of the underlying deficiency state, or to structural effects of a specitic lipid or its analogues.'"^ Petrolatum has also been shown to be absorbed into delipidized stratum corneum and to decrease transepidermal waler loss (TEWL) by forming a separate non-lamellar phase.^^ Skin diseases such as psoriasis, atopic eczema and infantile sehorrhoeic dermatitis, have been treated more or less successfully with lipids. Some clinical studies of psoriasis have shown improvement after topical treatment with fish oils rich iti uj-3 fatty acids.'^ Recently, it was aiso reported that fish oil.' ^ and puriiied ethyl ester of eicosapenlaenoic acid (20:5, n-3) from fish oil."' have anti-intlammatory effects on ultraviolet B (IJVB)induced acute inflammation. Similarly, reports have emerged on the clinical improvement of patients with alopic dermatitis after oral administration of evening primrose oil, a vegetable oil rich in 7~Iinoienic acid (GLA). a fatty acid of the uj-h family ( 1 8 : 3 , n-f>).^' GLA is also claimed to be effective against infantile seborrhoeic dermatitis. 215 216 M.LODEN AND A.-C.ANDERSSON Other lipid components, such as the unsaponifiable fractions of oils, are also claimed to have anli-inilammatory activities'''"" and the unsaponitiables might also influence the barrier function of damaged skin. The sterols in the unsaponitiable fraction are related to steroid hormones and vitamin D in the skin, and they may also participate as structural lipids in the bilayer membrane of the stratum corneum.^"^ Generally, vegetable oils have a low content (<1%) of unsaponiiiables."' However, there are a few oils that contain more than average amounts, e.g. shea butter with 5-9% unsaponifiables."" "^ Not only the content but also the composition of unsaponitiables varies considerably between different oils.^'^* Prolonged exposure to surfactants oflen induces irritant contact dermatitis. Sodium lauryl sulphate (SLS) is a commonly used surfactant which penetrates human skin." induces significant changes in cytokine levels in the atlerent skin lymph."^ and removes skin lipids. such as cholesterol, cholesterol ester, free fatty acids and sphingolipids.^** The sometimes promising effects of treatment of other skin disorders with lipids prompted us lo study the influence of different lipids on SLSirritated skin. Oils rich in uj-b and u;-3 fatty acids (borage oil and Ush oil. respectively) were tested. along with sunflower seed oil (rich in linoleic acid. 18:2, n-6). shea butter (mainly oleic aTid stearic acid) and partially hydrogenated canola oil (rich in oleic acid. 1 8 : 1 , n-9). Moreover, fractions of unsaponifiable lipids from shea bulter and canola oil were prepared and included in the study. The unsaponiiiable fraction in shea butter contains a significant amount of cinnaniic aeid esteriiied with triterpene alcohols.'^""' whereas the unsaponiiiables in canola oil constitutes mainly of stert)Is,'^ In addition, petrolatum (long-chain aliphatic hydrocarbons), water and one hydrocortisone cream were included in the study. Normal skin was exposed to SLS and then treated with the different lipids. The influence of the treatments was evaluated visually and by use of non-invasive biophysical measurements. The instruments used were an evaporimeter for measurement of TEWL and a laser Doppler velocimeter for measurement of superficial skin biood flow. Materials and methods Test substances Hydrocortisone cream (Hydrokortison 1% ACO. Pharmacia, Sweden), petrolatum (Petroleum jelly, Wilburine. Witco Corporation, Sweden), borage oil (Ropufa Table 1, Perceiiiayc laity acid composition of the tested oils. Fatty acids below 1% are not specified Fatty acid 14:0 16:0 16:1 (n-7) 16:1 (n-9) 16:4 fn-3) 18:0 18:1 (n-7) 18:1 (n-9) 18:2 (n-6) 18:3(n-6) 18:4 (n-3) 20:0 20:1 ln-9) 20:1 (n-11) 20:4 (n-6) 20:5 (n-3) 22:1 (n-ll) 22:1 (n-9) 22:5 (n-3) 22: 6 (n-3) 24:1 (n-9) Others Borage oil* (Ropufa 25 Omega-6 oU) Canoia oil (Akorex L) 11 3 4 6 1 15 37 23 m6 .. .~ Fish oil (CPL Fish oil 30) Shea butter (Upex 205) Sunflower seed oil 7 26 7 « 4 7 2 28 4 3 8 1 56 9 22 64 (SunOower oil) 3 1 4 1 1 1 2 18 1 1 2 14 2 2 1 4 2 3 * Values given in the spccifitation by Rot. he. 1996 British Association of Dermatologists, British Journal of Dermatology. 134, 21S'22 TOPICAL LIPIDS AND IRRITATED SKIN I-y Omega-f) oil. Roche, Sweden) and fish oil (CPL Fish oil iO. Karlshamns Lipid Teknik, Sweden) were studied, together with the following purified oils from Karlshatnns AB (Karlshamn, Sweden): partially hydrogenated canola oil (Akorex L), winterized sunflower seed oil (Sunflower oil), shea butter liquid (Lipex 205) atid unsaponiiiable fractions froru canola oil and shea butter. The unsaponifiable fractions were obtained by low temperature crystallization. All substances were used well in advance of their date of expiry and they were kept sealed in a cool and dark place until used. The fatty acid composition of borage oil was given by Roche in their product specification, while the compositioti of the other oils was analysed as tiiethylesters by gas-liquid chromatography (GLC). The eompositions are shown in Tahle 1. The vegetable oils used varied in their contents of unsaponitiables. The tested shea butter (Lipex 205) is a liquid traction of shea butter containing ^% unsaponiiiable lipids. mainly triterpene esters and sterol esters with einnamie acid, as determined by high pressure liquid chromatography (HPLC) using cholesteryl cinnamate as the internal standard. The content of sterols is only about i-S% of the unsaponifiables and consists mainly of (v-spinasterol and A'-stigmastenol. ^' The tested fraction of unsaponifiables contained 22% unsaponifiable lipids. out of which 40% consisted of triterpene and sterol esters with einnamie acid. The tested canola oil (Akorex L) is a partially hydrogenated vegetable oil from the seeds of Jirassica napus and ii. canipestris. Our GLC analysis showed that it contained 1 % unsaponifiables and ()• 7% free sterols. which roughly Figure 1. The efTett ol tbe test substances on surfiictant-irritatcd skin. The results arc expressed as the nH-dian tIiiTe^ena^ in tninscpidermal water loss (TFWL) and skin blood Ilow compared with exposure to WHkT ill : 21). The bars denote 25/75 pi'rcentill's. The asterisks denote values signilicantly different from tbose after treatment witb water. AU, arbitrary unit.s; [ISF, unsaponiiiable lipids. 217 corresponds to the finditigs of others." The unsaponiiiable fraction from canola oil contained 7% free sterols. the main ones of which are /i-sitosterol. campesterol and brassicasterol in descending order, as detennined by GLC equipped with a high temperature capillary column. The samples were silylated before analysis and 5a-eholestane was used as internal standtird. Sunflower seed oi! is reported to contain 0-7% unsaponifiable lipids and 0*4% sterols.^' and the tested fish oil was found to contain 0 2'!^i unsaponifiables. Volunteers Twenty-one healthy individuals, aged 2 2 - 5 7 years, seven men and 14 women, with no visible signs of skin disease, participated iti the study. The study was approved by the local ethics committee. Treatment The effect on normal skin was studied on the right volar forearm using a single patch test exposure along with visual and instrumental evaluation of the degree of irritation. Fifty microlitres of each substance was pipetted into large aluminium chatubers (12 mm, Finn chambers, Epitest Oy. Finland). One layer of biter paper was used in the chambers to keep all substances but petrolatuin atid the hydroccirtisone creatu in place. The chambers were attached to the skin with adhesive tape (Scanpore, Norgeplaster. Norway). After 48 h exposure, the tesl substances were removed. The left forearm was exposed to an aqueous solution (50/d) of 14% SLS (Ph. Eur.) in 10 large aluminium Hydfocxirtisone Canola oil Fish Oil Shea bulter Sunflower Borage oil Canola USF Petrolatum Shea butter USF © 1996 British Association of Dermatologists. British ]ourna! of Dermatology. 134, 2 t 5 - 2 2 0 218 M.LODEN AND A.-C.ANDERSSON spots, which reduces variation due to spotty erythema. The probe was attached to the skin with a standard probe holder without pressure and using double adhesive tape. The output signals were recorded on a chart strip recorder (Servogor 120. flBC) until equilibrium was reached, usually within 1-2 tnin. The value at equilibrium was used for the calculations. chambers for 7h. Upon removal of the patches, the skin was gently rinsed with water and allowed to dry. The test substances were then randomly applied to the SLStreated areas as above and held in place for 17h. After removal of the substances, the left and right arm were gently cleaned with a mild soap solution (ACO Mild Tval. ACO Hud. Sweden) and dried with soft paper. Twenty-four hours later the areas were examined. Statistics Evaluation To test the hypothesis that there is no difference in the effect of the 10 substances with respect to the three variables (visual signs of irritation. TEWL and skin blood fiow), a non-parametric test as described by iYiedman was used. To compare each treatment with the control, water, the non-parametric two-sided Wilcoxon signed rank test was used. Correction for multiplicity was made according to Holm"** to get the overall significance level of <0-05. Median values and 25/7S percentiles were calculated for each substance, as well as for the intraindividual difTerence between the treatments and the control. Upon removal of the patches the skin was examined visually. Twenty-four hours later, the degree of irritation was evaluated by visual scoring. The evaluation was made without access to the application charts according to the following scale: 0 — no reaction 0-5 = barely perceptible very weak spotty erythema 1 = slight erythema, either spotty or diffuse 2 — moderate erythema 3 — intense erythema, infiltration, possible vesicles. After the visual evaluation, the skin reactions were quantified objectively using an Evaporimeter EPl (Servonied. Kinna. Sweden) which measures TEWL,"'' and a laser Doppler flowmeter (Periflux Pfl. Perimed, Stockholm. Sweden) which measures superficial blood flow.''' The probe for the tiieasurement of TEWL was equipped with a screen and grid to reduce air conveetion.^^ The laser Doppler fiowmeter was equipped with a special multifibre probe (l'F 113 integrating probe, Perimed) which has seven fibre triplets Instead of one. one being in the middle and six around it, forming a circle 8 mm in diameter, in the probe head. Thus, the blood flow value is the mean of the seven Results Oti normal untreated skin, six of the applied stibstances caused no irritation and four of them (water, petrolatum, shea butter and sunflower seed oil) induced barely perceptible erythema in a small number of subjects 24 h after patch removal. No sigtiiticant difierences atnong the treatments were detected. Hydrocortisone induced bfanching in several subjects, which was noted at the time of patch removal but not 24 h later when the areas were examined. Table 2. Transepidermal water loss (TEWL) (g/m^/h) and cutaneous blood flow (arbitrary units. Ati) after 7h of sodium lauryl sulphate (Sl-Sl exposure and treatment with the diiTcrent tesl substances for 17 b. Tbe areas were examined 24 h after removal of ihe test substances. F.acb treatment was compared statistically witb water, and correction for multiplicity was made to gel the overall signiiicance level of <0-05. For TKWL and blood flow are shown median values, with the 25/75 percentiles in parentheses (« = 2 i) Substance Water Hydrocortisone Borage oil Canola oil Canola tJSF Fish oil Petrolatum Shea butter Shea butter tJSF Sunflower oil 'ITiVVL (g/m' per h) i'-value Signiticance 13-8(8-6-21-5) 10-5(8-6-13-6) 13-7(9 9-17-0) 9-2(8-3-16-3) 8-0(6-9-12-1) 12-3(9-7-16-2) 12-2(10-1-18-7) 11-3(8-7-18-9) 8-7(7-4-19-6) 1 i-8 (7-8-18-3) 0-0003 0-8212 0-0054 0-0003 0-3603 0-9307 0-2813 0-4445 0-0476 S NS S s NS NS NS NS NS Blood Uow (AU) J'-value 30-1 (16-3-50-0) 14-2(11-6-18-7) 2O-2(15-9-3tl-2) 21-8(7-1-34 91 15-0(11-7-194) 21-1 (17'0-32-0) 25-1 (17-0-42-7) 19-9(14'7-36-3) 16-7(15-2-34-3) 19-0(16-0-38-4) 0-0031 0-6142 0-0560 0-0004 0-0403 0-8620 0-2250 0'2374 0-0250 — s NS NS S NS NS NS NS NS NS. not signiflcant: S, signiflcant: USF, unsaponiflable lipids. © 1996 British Association of Dermatoioests, British Journal of Dermatology. 134, 215-220 TOPICAL LIPIDS AND IRRITATED SKIN Tiihle 3. Clinical evaluation of patch test reactions. Total number of rfci{lings scorcil as 0. 0-5. \. 2 and 3 afler treatment of surfactantirriliilcd skin with different suhstanocs. The rcaclinys were made 24 h after removal of the patcht's Product/score 0 0-5 1 2 3 Water Hydrocortisone 5 12 7 6 9 3 6 10 « 10 6 6 S 9 4 3 1 2 0 0 1 3 1 2 2 0 0 0 0 HofHRC nil Canola oii Canola USF* Fish oil IVtroIalum 11 8 6 SIK'H butler 7 SlitM Inittcr USF Siintlower seed oil 9 6 5 6 2 2 6 7 2 4 0 0 0 0 0 0 Signiiicantl • y different from water. USF. unsaponiflable lipids. When the suhstances were applied to the surfactantirriUitcd skin, significant differences were found (Fig. I atid Tables 2 and 5). Treatment with hydrocorlisone signilicanlly reduced the TEWL and hlood flow compared with the control area treated wilh waler (Fig. 1 and Table 2). 'I'he sterol-enriched fraction of canola oil also significantly reduced TEWL and skin blood tlow compared with waler (Fig. 1 and Table 2). Canola oil in itself reduced TEWL significantly, whereas the decreased in blood tlow did not differ signiticantly from that in the water-treated area. The visual assessments olthe effects on the surfactantirritated skin corresponded well to the instrumental readings (cf. Tables 2 and 5). The variation among Ihc elTects induced by the substances was signiticantly larger than could be expected to be due to chance, and the sterol-enriched fraction of canola oil gave significantly lower valties than water. Discussion Moisturizers are used daily by many people to alleviate symptoms of dry skin. All of them contain lipids. Lipids are also found in vehicles for pharmacologically active substances in the treatment of skin diseases. Traditionally, lipids in moisturizers and vehicles have been considered as inactive and incorporated into formulations on the basis of their technical and sensory properties rather than on their possible impact on the epidermis. However, moisturizers or ointment bases which do not contain any known pharmacologically active ingredients, often assist in the improvement of skin diseases, such as irritant contact dermatitis. ** The mechanisms for the beneficial effects are nol yet fully understood. 219 In the present study, neither the tested lipids nor the hydrocortisone cream induced any significant visible changes of normal skin. Neither did the lest substances significantly alter TEWL or superficial skin blood flow. However, on the surfactant-irritated skin, a significantly lower degree of irritation was found on the sites treated with hydrocortisone. We also noted skin blanching after exposure to hydrocortisone. Antiinflammatory activity of hydrocortisone has also been observed in other irritation experiments in humans.^" and blanching by corticosteroids is a pharmacological effect that correlates well with clinical efficacy.'^ No eftect was observed from the treatment with lish oil (rich in eicosapentaenoic acid) or borage oil (rich in GLA and linoleic acid), although they have been linked to antiintlammatory properties in other studies.^"^ '^ Neither did we tind any eflects from the trealment with sunflower seed oil. shea butter and petrolatum. The only lipids that significantly reduced the degree of irritation compared with waler were canola oil and its sterol-enriched fraction. The effect of the latter fraction was especially pronounced and tended to be greater than that of the hydrocortisone cream. The visible signs of irritation were less marked and TEWL and skin blood flow were lower on the sites treated with this sterolenriched fraction than on the control sites treated with water. Canola oil has a relatively high content of sterols and. besides having anti-inflammatory activities,'**'^" it is possible that sterois might influence the structure of the epidennal membratie lipids. A mixture of sterols. free fatty acids and sphingolipids forms the intercellular membrane bilayers of the stratum corneum. and this mixture is presumed to regulate the epidermal barrier function.^"* Topical treatment with SLS removes stratum corneum lipids progressively over a long time period**'^ and the disruption of the barrier function by the lipid extraction is followed by a burst of epidermal lipid synthesis."^''** The synthetic activity includes unsaponifiable Iipids,''''^'"' fatty acids'* and sphingolipids.*'^ Sterols and fatty acids are synthesized immediately after barrier disruption, whereas the increase in sphingolipid synthesis is somewhat delayed. Over time, the contents of lipids in the stratum corneum are restored to normal levels in parallel with the return of barrier function.^''• ^'"''^ ^ Oitr findings that canola oil and its sterol-enriched fraction had beneficial effects on the irritation, suggest the possibility that they assisted the skin in supplying the SLS-damaged barrier with adequate lipids. Recent studies have shown that topically applied lipids may interfere with the lipid synthetic (j 1996 Britisii Association ofDermatologists. British Jourmd of Dermatology. 134. 2 1 5 - 2 2 0 220 M.LOUEN AND A.~C.ANDERSSON activity of the skin." However, applications of ceramides. linoleic acid and a variety of other fatty acids alone have been reported to actually delay barrier recovery in acetone-treated murine skin, despite the fact that these iipids are required for barrier homeostasis.'^ Two-component mixtures of fatty acid plus ceramide. cholesterol plus iatty acid, or cholesterol plus ceramide also delayed barrier recovery. The only trealments that allowed normal barrier recovery were applications of complete mixtures of ceramide, fatty acid and cholesterol, or pure cholesterol.'^ Our observations that canola oil and its sterolenriched fraction ameliorated the surfactant-induced irritation, emphasize that lipids used in moisturizers do not merely form an inert, epicutaneous. occlusive membrane, but may penetrate and influence the barrier properties of the skin. The mechanism and the timecourse of this effect need to be investigated further. 11 12 1i 14 15 16 17 18 19 Acknowledgments The excellent technical assistance of Mr Per Bostrdm. ACO ilud Ali. is gratefully acknowledged. We also wish to thank Ur Magnus Lindberg. Department of Dermatology, University Hospital. Uppsala, for reviewing this manuscript, and Ms Anna Cironbladh. Pharmacia AB. Stockholm, for statistical discussions. 20 21 22 2J References 24 1 lilias I'M. [-ipids and the epidermal permeability barrier. Arch Dermatot Res 19SI; 270: 95-117. 2 Grubauert;, h'eingnld KK. Harris RM, Elias PM. Lipid content and lipid type as dctenniiiants oJ' Ihe epidennal permeabilily barrier. j Lipid Res 19S9: 30: S9-96. J Yiirdley II|. Summerly R. Lipid i-omposition and nielabolism in normal and diseased epidermis. Phannacol Ther 1981; 13: 357-83. 4 Kliiis I'M. tloerkc J. l-'riund DS. Mammalian epidermal barrier layer lipids: composition and influence on structure. / Invest Dermatol 1977:69: 515-46. 5 Hansen AK. Knott EM. Weise HF et al. Eczema and essential falty acids. Am j Dis Ctiild 1947: 7J: 1-18. 6 Grubauer G. I'eingold KR. Elias PM. Relationship of epidermal Upogenesis to cutiinoous barrier function. / Lipid Res 19K7: 28: 746-52. 7 Imokawii G. Haltori M. A possible function of structural lipids in Ihe water-holding properties of Ihe stratum corneum. / Invest Dermatot 198S; 8 4 : 2 8 2 - 4 . H Imokawa G, Akasaki S, Minematsu Y, Kawai M. Importance of inlerccllular lipids in water-retention properties of the stratum corneum: induction and recovery study of surtactant dry skin. .'\rch Dermalol Res 1989: 281: 4 5 - S l . 9 Menon GK, Keinjiold KK. Moser AH ('( al. De novo stcrologenesis in the skin. II. Regulation by cutaneous barrier requirements. / Lipid Res 1 9 8 5 : 2 6 : 4 1 8 - 2 7 . 10 l-eingotd KR. Mao-Qiang M, Menon GK et al. Cholesterol synthesis 25 2fi 27 28 29 30 31 32 is required for cutaneous barrier function in mice. / Clin Invest 1990: 86: 1738-45. Man M-Q. I'eingold KK. Ellas PM. Exogenous lipids inlluence permcahilily barrier recovery in aeetone-treated murine skin. Arch Dermatol 1993: 129: 728-38. Press M. Harlop l'|. Prottey C. t'orrt'clion of cssenlial Iatty acid deflciency in man by the cutanoous appliciiii<in ol'sunllower-seed oil. Uincet 1974: i: 597-9. Ghadially R. Halkier-Sorensen 1.. Elias PM. Eiffectsoi petrolatum on stratum corneum structure and function. / Am Acad Dermatot 1992:26: J87-96. Escobar SO. Achenbach R, Innantuono R. Torem V. Topical lisli oil in psoriasis—a controlled and blind study. C%i Exp Dermuwt 1992: 17: 159-f>2. Orengo II'. Black HS. VVolf|i;. Inllneitceoflish oil supplementation on the minimal erythemii dose in humans. Arch Dermatol Res 1992: 2 8 4 : 2 1 9 - 2 1 . Danno K. Ikai K. Imamura S. Anli-inllammator>' etTects of eicosapenlaenoic acid on experimental skin inllammalion models. Arch Dermatol Res 1993: 285: 4 J 2 - 5 . Wright S. Burton JL. Oral evening primrose seed oil improves atopic eczema, lancet 1982: ii: 112(1-2. Tollesson A, Frithz A. Borage oil. an eflective new treatment ibr infanlile seborrhoeic dermatitis. Br I Dennatol I 9 9 i : 129: 95. Bouic PJI). Albrocht CVV. Compositions ot I'hytosicrols and i'hjitosterolins as Immunomodidators. European Patent Office. 509656 Al, 1992: 1-28. Saeed SA, Famaz S. Simjee Rll, Malik A. Tdlerpenes and betasitosterol from piper belie: isolation, anliplatelet and antiinflammatory elTects. Biochi-m Soc Trans 199 J: 2 1 : 462S. Itoh T, Tiimura T. Malsumoto T. Sleroi composition of 19 vegetable oils. / .^m Oil Chem Soc I 9 7 i : SO: 122-5. Peers KE. The non-gtyceride saponiftables of shea butter. / Sci l-'omi Agrie 1977: 28: i()(H)-9. Itoh T, Tamura T. Matsumoto 'l\ Stcrots and melhyislerols in some tropical and subtropical vegelable oils. Oleagineii.x 1974: 29:25 J-8. Loden M. The simultaneous penetration of water and »)dium lauryl sulfate through isolated humim skin. / Sw Cosmet Chem 1990:41:227-3J. HunzikerT. Brand t'il. Kapp Ad al. incrt-ased levelsorinllamniiitory eytokities in human skin lytiiph derived Trom smlium !iiur\'l siilphale-inducod contact dermatitis. Br} Dermatoi 1992: 127: 254- 7. Pinnagoda |, Tupker RA. Agner T, Serup |. Guidelines for tnmsepidcrmal water loss ITEWI.) measurement. A report from the standardization group lyi' the European society of contact dermatitis. Cmtact Ui-rnunilis 1990; 22: 164-78. Tenland T. On laser Doppler flowmetry. Methods and microvascuiar applications. Unkoping Medical Dissertations No. t i 6 , Unkopiiig trniversity. 1982. Holm S. A simple sequentially rejective multiple test procedure. Scand j Statist 1979: 6: 65-70, Halkicr-Sorcnscn L, ihestrup-Pedersen K. The efficacy of a moisturizer (Locobase) among cleaners and kitchen assistants during everyday exposure to waler and detergents. Ccmtact Iknnaiitis I 9 9 i : 29: 2 6 6 - 7 1 . Kaidbey KH, Kligman AM. Assay oT topical corticosteroids by suppression oT experimental inilammatton in humans, j Invest Dermalot 1974; 63: 2 9 2 - 7 . McKenzie AW. Stoughlon KB. Method for comparing percutaneous absorption. Arch Dennatol 1962: 86: 608-10, Holleran WM. Feingold KK. Mao-Qiang M cl al. Regulation of epidermal sphingolipid synthesis by permeability barrier function. jLipidRes 1991; J2: U 5 1 - 8 . 1996 British Association ofDermatologists. British journal of Dermatology, 134, 215-220