Coupled exposure to ingredients of cosmetic products: III. Ultraviolet filters

Contact Dermatitis Original Article COD Contact Dermatitis Coupled exposure to ingredients of cosmetic products: III. Ultraviolet filters Wolfgang Uter,, Margarida Gonçalo 2,, Kerem Yazar 3, Eva-Maria Kratz 4, Gerd Mildau 4 and Carola Lidén 3 Department of Medical Informatics, Biometry and Epidemiology, Friedrich-Alexander-University Erlangen/Nürnberg, D-904 Erlangen, Germany, 2 Department of Dermatology, Faculty of Medicine, University Hospital, University of Coimbra, Praceta Mota Pinto, P-300-07 Coimbra, Portugal, 3 Department of Occupational and Environmental Dermatology, Institute of Environmental Medicine, Karolinska Institutet, SE-7 77 Stockholm, Sweden, and 4 Department for Cosmetics and Medicinal Products, Chemical and Veterinary Investigation Office (CVUA Karlsruhe), D-7687 Karlsruhe, Germany doi:0./cod.224 Summary Background. The use of cosmetics exposes consumers to mixtures of ingredients, many of which are potential allergens. Ultraviolet (UV) filters are used not just in sunscreens, but also in other products. Many UV filters are known contact allergens and photoallergens. Objective. To examine the pattern of co-exposure to UV filters in cosmetics. Methods. A survey of products marketed in Germany, conducted in 2006 2009 by the Chemical and Veterinary Investigation Office in Karlsruhe, identified 4447 products (of all 667 cosmetic products examined) (i) that were categorizable according to Annex I to the Cosmetics Directive, and (ii) with information on the presence of UV filters or zinc oxide. The occurrence and co-occurrence of UV filters were analysed and presented in tabular and graphical format. Results. UV filters or zinc oxide were present in 22.% of all 4447 products, ranging from almost 00% in sunscreens to a few per cent in, for example, some hair products; they were absent in two product categories. Frequently, several different UV filters were included in one product, for example in sunscreens (median 4) and in perfumes (median 3). The overall most frequent UV filters were butyl methoxydibenzoylmethane and titanium dioxide, combined mostly with octocrylene in sunscreens and with ethylhexyl methoxycinnamate in creams. Conclusions. The frequent co-occurrence of UV filters in cosmetic products possibly facilitates sensitization, and may explain why patients often react to chemically unrelated UV filters. Key words: cosmetics; exposure analysis; ingredient label; UV filters. Correspondence: Wolfgang Uter, Department of Medical Informatics, Biometry and Epidemiology, Friedrich-Alexander University Erlangen- Nürnberg, Waldstr. 4-6, D-904 Erlangen, Germany. Tel: +49 93 82270; Fax: +49 93 82272. E-mail: wolfgang.uter@imbe.med.unierlangen.de Shared first authorship. Conflict of interests: W.U. has accepted honoraria for presentations or travel reimbursement from cosmetic industry associations for presentations given to these. M.G., C.L., K.Y., E.-M. K. and G.M. have no conflict of interests to disclose. Accepted for publication March 204 Because of the increasing awareness of the harmful effects of ultraviolet (UV) radiation, sunscreens are being more frequently used, both during acute, intermittent UV exposure and in daily life. Apart from UVB (290 320 nm), which causes acute sunburn and directly harms DNA, UVA (320 400 nm) is also responsible for photosensitivity, photo-ageing, photo-immunosuppression, and skin carcinogenesis, including melanoma. Therefore, in 2006 the European Commission recommended that the ratio between UVA protection and the sun protection factor (SPF), defining UVB protection, should be :3 or lower (). To comply with this recommendation, cover the whole 62 Contact Dermatitis, 7, 62 69

UV solar spectrum, including UVA- (340 400 nm), and have higher protection factors, sunscreens include several, partly new, UV filters, often in complex and synergistic mixtures of chemical and physical filters. Currently, and during the study period, 26 substances are permitted for use as UV filters to protect skin and hair, according to Annex VII of the Cosmetics Directive (now corresponding to Annex VI of the Cosmetics Regulation). Moreover, UV filters are partly used as UV absorbers to protect products from the effects of UV light. Cosmetic products and other sources contribute to skin exposure to a complex mixture of allergens, irritants and penetration enhancers in daily life. Recent experimental findings have indicated that the ingredients of such a cocktail of substances may mutually enhance sensitization (2). In the the present study, we addressed this research question by analysing coupled exposure to UV filters in cosmetics, as a number of UV filters are well-known photoallergens or allergens. Following previous similar analyses of co-occurring fragrances (3) and preservatives (4), this article presents results based on a large survey of the German market. The use of titanium dioxide and zinc oxide in cosmetics, particularly in the form of nano materials, has also been an issue of concern in risk assessment (, 6). Therefore, although these physical UV filters have not been reported to cause (photo)allergic dermatitis, they were also analysed. Methods The Chemical and Veterinary Investigation Office (CVUA) in Karlsruhe (Germany) is a state laboratory for risk analysis and assessment. In the years 2006 2009, the CVUA analysed various types of cosmetic product to investigate the conformity of cosmetic products with legal provisions (667 in the 4 years); see (3). In short, INCI labelling was studied with regard to the presence of UV filters permitted for use in cosmetics (Annex VII to the Cosmetics Directive); thus, qualitative information on the presence of substances was collected. Zinc oxide, which is used as a UV absorber, was included in this survey, although it is not listed as a UV filter in Annex VII. All products were categorized according to a hierarchical system of product types with, effectively, 70 items. For the present analysis, these product type categories were aggregated to the items of Annex I of the Cosmetics Directive valid at the time of writing, with effectively 23 different categories ( hair setting products and tinted bases were not applicable; a separate category, not categorizable, has been introduced). The present analysis included those products for which: (i) the above categorization was possible, thus excluding products that were not categorizable (n = 246); and (ii) the presence or absence of UV filters was recorded in terms of an INCI declaration. Thereby, 4447 products in 23 product categories fulfilled the above requirements. Two product categories (deodorants and antiperspirants, n = 4; products for external intimate hygiene, n = 27) did not include any products with UV filters or zinc oxide; hence, 4306 products in 2 categories were further analysed. Data management and analysis were performed with the statistical software r [version 3.0. (7)], in particular with the r-package igraph (http://igraph.sourceforge.net/index.html) for producing so-called social network diagrams, which have previously been used to present similar data (3, 4, 8, 9). Results The frequency of occurrence of the different aggregated product types is shown in Table, ordered by the relative frequency of products containing chemical or physical UV filters. Altogether, among the 4306 cosmetic products analysed, 000 (23.2%) products contained at least one chemical and/or physical UV filter. However, it should be noted that, in some product categories, titanium dioxide or zinc oxide may have been present in some or all of the respective products for purposes other than UV protection, as in toothpastes and related materials. The products most frequently containing UV filters were sunscreens, followed by skin whitening products, hair styling products, cosmetics for the lips, and perfumes. The proportions of products with UV filters and, among products containing filters at all, the proportions with chemical filters and physical filters, respectively, are shown in Table. Whereas the majority of product categories, if having a UV filter, only contained one filter (median of ), sunscreens contained a median of four filters, perfumes with filters a median of three filters, and creams (of those 6.3% of 44 products containing filters) a median of two filters (Table ). Products contained chemical UV filters more often than physical UV filters, except for make-up products. Physical filters were present in a notable proportion of sunscreens (6.7%) and lip cosmetics (40.3%). The distribution of UV filters is shown in Table 2, both overall and across the product categories included in further analyses. Product categories that were not further analysed, owing to small subsample sizes or the presence of physical filters only, are those without a short-hand term shown in Table. Regarding the aggregation of original product categories to hair and make-up, see Table. From the whole spectrum of UV filters listed in Annex VII of the Cosmetics Directive, three camphor Contact Dermatitis, 7, 62 69 63

Table. Types of cosmetic product included in the analysis, aggregated to the categories of Annex I of the Cosmetics Directive, with frequency of occurrence of the respective product type (n), proportion of products with ultraviolet (UV) filters, and proportions of these products containing chemical UV filters and physical UV filters (titanium dioxide or zinc oxide ). Product categories with neither UV filters nor zinc oxide are not shown. The distribution of the number of chemical filters in products containing these is given in terms of first quartile (Q), median, third quartile (Q3), and maximum Category Short n Among products with UV filters: Distribution of no. of chemical filters %withuv filters % chemical % physical Q Median Q3 Maximum Sunbathing products (sunscreens) Sun 464 99.6 90.0 6.7 3 4 4 7 Skin-lightening products 23 2.2 9.7 6.7 2 2 Hairdressing products (lotions, Hair 87 44.8 00.0. lacquers, and brilliantines) Products intended for application to Lip 74 3.6 8. 40.3 2 4 the lips Perfumes, toilet waters, and eau de Perf 2 27.6 00.0 0.0 2 3 3 4 Cologne Products for nail care and make-up Nail 44 22.9 87.9 2. 2 Products for tanning without sun 36 6.7 00.0 0.0.2 2 2.7 3 Creams, emulsions, lotions, and gels Cream 44 6.3 8.8 39.0 2 3 4 and oils for the skin (hands, face, feet, etc.) Make-up powders, after-bath Make-up 44.9 42.9 7. powders, hygienic powders, etc. Products for making up and Make-up 409.7 3.2 70.8 removing make-up from the face and the eyes Anti-wrinkle products 88 9. 62. 37. 2 Hair conditioning products (lotions, Hair 0 9. 00.0 0.0 2 creams, and oils) Depilatories 33 9. 0 00.0 Shaving products (creams, foams, 3 6. 00.0 0.0 lotions, etc.) Face masks (with the exception of 48 6.2 0 00.0 peeling products) Hair cleansing products (lotions, 98. 00.0 0.0.7 2 powders, and shampoos) Toilet soaps, deodorant soaps, etc. 87 4.6 0 00.0 Products for waving, straightening Hair 9 4.2 7.0 2.0 and fixing hair Bath and shower preparations (salts, 363 2.8 90.0 20.0 foams, oils, gels, etc.) Products for care of the teeth and the 20. 0 00 mouth Hair tints and bleaches Hair 0.0 00.0 0.0 Overall 4306 22. 84.3 46.6 3 4 7 Zinc oxide is used as a UV absorber, bulk substance and colorant; it is included here although it is not listed as a UV filter in Annex VII of the Cosmetics Directive. derivatives were not found in the sample. Especially in those products in which UV filters are presumably used for technical purposes, and not to protect the human skin from excessive UV radiation, for example perfume, nail and hair care products, use is limited to relatively few, but commonly used, filters. 4-Aminobenzoic acid (PABA), delisted from Annex VII in October 2009 (i.e. prohibited as a UV filter), was found in four cosmetic products; of these, two were shampoos. The number of different chemical UV filters per product, in the product categories subject to further analysis, is shown as modified violin plot in Fig.. Another focus of the present analysis was to present the combinations of UV filters in certain cosmetic product types. Figure 2 shows the various combinations for four product types with a notable occurrence of one or more filters in a semi-quantitative, graphical manner, including constituents occurring in at least % of products. The diameter of circles (nodes) representing single UV filters is scaled to represent the frequency of occurrence in the 64 Contact Dermatitis, 7, 62 69

Table 2. Overall occurrence of ultraviolet (UV) filters among all products in the sample (n= 4447), and occurrence in those products considered for further analysis (939 of 000 products in the 7 product categories with at least some products containing chemical UV filters) by their INCI name and, in parentheses, the International Nonproprietary Name (INN), if available according to the CosIng database. Multiple occurrences of filters are possible (see Table and Fig. 2); hence, column percentages may add up to > 00%. For aggregation of categories to make-up and hair, see Table. The frequency of positive (photo)patch test reactions among 03 patients tested in a European multicentre study is taken from (0) Product type (no.) Overall Sun Cream Lip Make-up Perf Nail Hair % positive reactions Short CAS no. 4447 462 23 62 42 33 4 PPT PT Benzophenone-3 (INN: oxybenzone) * BE3 3-7-7 8.8 3.2.6 6. 3.6 28.6 78.8 9.3 3.6 0.6 Benzophenone-4 (INN: BE4 406-4-6. 0 2.6.6 0 0 6. 7.4 0.3 0 sulisobenzone) * 3-Benzylidene camphor 3BC 087-24-8 0. 0 0.4 0 0 0 0 0 NT NT Benzylidene camphor sulfonic acid BCSA 6039-8-8 0 0 0 0 0 0 0 0 NT NT Bis-ethylhexyloxyphenol BOPMPT 87393-00-6 6.4 34.4 2.2 0 0 0 0 0 0.3 0 methoxyphenyl triazine (INN: bemotrizinol) * Butyl methoxydibenzoylmethane BMX 7036-09- 48.7 74 38. 33.9 0 64.3 0 0.7 0.3 (INN: avobenzone) * Camphor benzalkonium CBM 2793-97-2 0 0 0 0 0 0 0 0 methosulfate Diethylamino hydroxybenzoyl hexyl DHHB 302776-68-7.8 2.6 0.6 0 4.8 0.9 0.4 0. benzoate * Polysilicone- PS 20774-74- 0. 0 0.4 0 0 0 0 0 0. 0 Diethylhexyl butamido triazone (INN: DBT 4702-- 6.3 0 2.9 0 0 0 0 0 0 iscotrizinol) Disodium phenyl dibenzimidazole DPD 80898-37-7 0.4 0.9 0 0 0 0 0 0 0.3 0. tetrasulfonate (INN: bisdisulizole disodium) Drometrizole trisiloxane * DT 633-4-8 6.. 2.2 3.2 0 0 3 0 0. 0 Ethylhexyl methoxycinnamate (INN: EHMC 466-77-3 38. 36.6. 40.3 8.2 8.7 3 6.7 0.7 0.2 octinoxate) * Ethylhexyl triazone * EHT 8822-99-0 3.7 22.7 7.4 22.6 0 0 0 0 0.3 0 Homosalate HS 8-6-9 0.4 0.9 0 0 0 0 0 0 0. 0 Isoamyl p-methoxycinnamate (INN: IMC 767-0-2 4. 7. 2.2 3.2 0 0 0 0.0 0.2 amiloxate) * 4-Methylbenzylidene camphor (INN: MBC 3686-47-9/3802-62-4.9 0 2.6 8. 0 0 0.9 0.3 0.4 enzacamene) * Methylene bis-benzotriazolyl MBTBP 0397-4-.4 0 3. 0 0 0 0 0 0.. tetramethylbutylphenol (INN: bisoctrizole) * Octocrylene (INN: octocrilene) * OCT 697-30-4 30.7 8.2 6 0.8 0 0 0 4.0 0.7 Ethylhexyl dimethyl PABA (INN: EDPABA 224-02-3 0. 0.2 0.6 0 0 0 0 padimate-o) Ethylhexyl salicylate (octyl salicylate) OSALI 8-60- 2.6.4.7.6 0 6.9 3 0 0.2 0. (INN: octisalate) PABA (INN: aminobenzoic acid) * PABA 0-3-0 0.4 0 0.4 0 0 0 0 0 NT NT PEG-2 PABA P2PABA 6242-27-4 0 0.9 0 0 0 0 4.8 NT NT Phenylbenzimidazole sulfonic acid PBISA 2703-8-7.6.8.6 0 9. 0 3 0 0 0 (INN: ensulizole) Polyacrylamidomethyl benzylidene PAMBC 3783-6-2 0 0 0 0 0 0 0 0 NT NT camphor Terephthalylidene dicamphor sulfonic TPDCSA 9276-26-7/9047-82-2 6. 2. 2.2 0 0 0 0 0 0.4 0.4 acid (INN: ecamsule) * Titanium dioxide (TiO 2 ) Ti.Ox 3463-67-7 40..8 34.6 30.6 40 0 9..9 NT NT Zinc oxide (ZnO) Zn.Ox 34-3-2 0.6 2.8.6.3 34. 0 6. 3.7 NT NT NT, not tested; PPT, photopatch test; PT, patch test. * These are included in the recommended European photopatch baseline series (). Substance also categorized as UV absorber in the Cosmetics Directive. Zinc oxide is used as a UV absorber and colorant; it is included here although it is not listed as a UV filter in Annex VII of the Cosmetics Directive. Contact Dermatitis, 7, 62 69 6

Number of chemical filters 7 6 4 3 2 0 Number of products with chemical or mineral filters in category 462 23 62 42 33 4 Sun Cream Lip Make-up Perfume Nail Hair Product category Fig.. Distribution of the number of chemical ultraviolet (UV) filters in products containing any UV filter as a modified violin plot, accounting for the discrete nature of the count data, corresponding to centred histograms. For instance, the bottom set of rectangles indicates the proportion of products with no chemical filter [but possibly physical filter(s)] listed, and the broadest rectangles in Nail and Hair cosmetics indicate that these mostly contained just one filter. product category (Table 2). The width and darkness of the lines (links) vary with the frequency of bivariate occurrence the more frequently two filters are used together, the wider and darker the line. To give a reference, the highest frequency, corresponding to the widest and darkest line, is given in the figure legend for each plot. Regarding the physical filters, 32 of 462 sunscreens, three of 23 creams and one of 62 lip products contained both titanium dioxide and zinc oxide. Discussion The present analysis extends a previous approach of analysing the co-occurrence of hair dye components (8, 9, 2) and the occurrence of fragrances (3) and preservatives (3, 4), respectively, in products on the market, and, in particular, of the co-occurrence of fragrances(3) and preservatives(4) in cosmetics, the latter two analyses being based on the same data sample. This is, to our knowledge, the first analysis of the (co-)occurrence of UV filters in all categories of cosmetic product. The number of UV filters is higher in sunscreens than in other cosmetic products, certainly with the objective of covering the whole UV solar spectrum and obtaining a high SPF and a high UVA protection factor. Concomitant exposure to many chemical filters, particularly in sunscreens, may explain why patients develop hypersensitivity to more than one UV filter, often from unrelated chemical groups (, 6). The higher number of UV filters in sunscreens may also explain why these cosmetics are the main cause of (photo)allergic contact dermatitis resulting from the use of UV filters (0), although UV filters contained in other cosmetics can also cause (photo)allergic contact dermatitis (, 7). The higher concentrations of UV filters normally found in sunscreens, which were not evaluated in this study, may also explain why they are responsible for most cases of allergic and photoallergic contact dermatitis. The two sunscreen products (0.4%) containing neither chemical nor physical filters, labelled Sun Gel and Sun Cream, respectively, apparently relied on other ingredients for sun protection. Butyl methoxydibenzoylmethane was the most frequently observed UV filter in sunscreens, as in a study from 20 performed in the United Kingdom (8), whereas it ranks second in cosmetics after the UVB filter ethylhexyl methoxycinnamate. In agreement with this finding, in a recent European multicentre study performed between 2009 and 200, butyl methoxydibenzoylmethane was considered to be the UV filter responsible for most positive photopatch test reactions, if positive reactions to octocrylene and benzophenone-3, which have often been associated with photoallergy to ketoprofen, were excluded (0). Cinnamates are also frequent contact and photocontact allergens, but in contrast to its relative frequency in this study, isoamyl p-methoxycinnamate caused more positive patch and photopatch test reactions than ethylhexyl methoxycinnamate in the multicentre photopatch test study cited above (0). Many of the chemical UV filters are well-known allergens and photoallergens (), but the newer photostable filters have hitherto rarely been reported to cause dermatitis, which is also in agreement with their less frequent presence in cosmetics. There is probably one exception, namely methylene bis-benzotriazolyl tetramethylbutylphenol, which induces allergic contact dermatitis because of the surfactant decyl glucoside that is necessary to solubilize methylene bis-benzotriazolyl tetramethylbutylphenol in this complex hybrid UV filter, which is both a chemical and a physical filter (0, 9, 20). Chemical UV filters, such as benzophenone-4, which was not found in sunscreen products, may be used not only to protect the skin or hair of the consumer from UV effects, but also to prevent the photodegradation of other ingredients in cosmetic products (7, ). This may explain the high frequency of certain 66 Contact Dermatitis, 7, 62 69

Sun EHT PBISA BOPMPT 0 73 9 OSALI EHMC 7 69 Zn.Ox 9 TPDCSA BMX Ti..Ox 28 OCT 6 DT 3 DBT 342 269 DHHB 2 BE3 2 MBC IMC 46 MBTBP 33 46 BE3 3 Zn.Ox 3 MBTBP 8 BE4 6 EHT 7 MBC 6 Cream OSALI 27 PBISA 36 OCT 37 Ti..Ox 80 EHMC BMX 8 89 TPDCSA IMC BOPMPT DT Lip Perfume Zn.Ox 7 MBC DBT 8 BE3 0 EHT 4 Ti.Ox 9 EHMC BMX 2 2 EHMC 36 BMX 27 OSALI 26 DT 2 IMC 2 BE4 DHHB OSALI EDPABA BE3 2 DHHB 2 Fig. 2. Co-occurrence of the most commonly used ultraviolet (UV) filters, identified by the INCI name on the label, in different product categories according to Annex I of the Cosmetics Directive. Abbreviations are explained in Table 2. The two vertices connected by the widest edge,and the respective frequency ofco-occurrence,are indicated for reference (see text):sun: Sunscreens (n= 462). Most frequent co-occurrence: butyl methoxydibenzoylmethane with octocrylene: 242. Creams: Creams, emulsions, lotions, gels and oils for the skin (hands, face, feet, etc.) (n = 23). Most frequent co-occurrence: butyl methoxydibenzoylmethane with ethylhexyl methoxycinnamate: 48. Lip: Products intended for application to the lips (n = 62). Most frequent co-occurrence: butyl methoxydibenzoylmethane with ethylhexyl triazone: 3. Perfume: Perfumes, toilet waters and eau de Cologne (n = 42). Most frequent co-occurrence: ethylhexyl methoxycinnamate with ethylhexyl salicylate: 26. chemical UV filters in, for example, perfumes, nail and hair cosmetics that we found in our analysis. Indeed, several cosmetic ingredients may have more than one function. Twenty-three of the 26 permitted UV filters are also used as UV absorbers to protect the cosmetic product, so the aggregated exposure to these ingredients is increased. In all, 9 substances are recorded as UV absorbers in the CosIng database (http://ec.europa.eu/consumers/cosmetics/cosing/, Functions ; last accessed 7 February 204). It is not stated on the label for what purpose substances with more than one function are used in specific products. The use concentrations are generally not known, although each chemical UV filter has a maximum authorized concentration ranging from 2% to 0%, with titanium dioxide being authorized up to 2%. The actual level of skin exposure to UV filters and other ingredients in cosmetics is thus largely unknown in quantitative terms. Titanium dioxide and zinc oxide are used as UV absorbers and colorants (titanium dioxide CI 7789, zinc oxide CI 77947), and zinc oxide is also used as a bulk substance. However, the use of zinc oxide as a UV Contact Dermatitis, 7, 62 69 67

filter is currently not permitted. Owing to concerns about possible health risks from nano materials, the Scientific Committee on Consumer Safety and its predecessors assessed the safety of nanoparticles in cosmetic products, and particularly of titanium dioxide and zinc oxide as UV filters. In 2007, it was concluded that there are large data gaps (2). Further risk assessment concluded that titanium dioxide nanoparticles (size distribution of 30 00 nm), up to 2% as a UV filter in sunscreens, can be considered to not pose any risk of adverse effects after application on healthy, intact or sunburnt skin, but the conclusion would not apply to inhalation exposure, such as from powders or sprayable products (). It was concluded, concerning zinc oxide nanoparticles (size distribution above 30 nm) (6) and larger particles, that up to 2% as a UV filter in sunscreens can be considered to not pose a risk of adverse effects after dermal application, but that products that may result in inhalation are of concern, owing to the risk of lung inflammation. PABA was responsible for a considerable number of cases of allergic and photoallergic contact dermatitis in the 90s and 960s, particularly in the United States (22, 23). It was delisted from Annex VII of the Cosmetics Directive in October 2009, and moved to Annex II/67 of the Cosmetics Directive (and Regulation), that is, the list of substances prohibited in cosmetics. At the end of the sampling period of the present study, it was already a rarely used UV filter. The most frequently used UV filters differ according to the type of cosmetic, which probably depends on the purpose of their use and the lipophilicity of the product. There are few water-soluble UV filters, apart from PABA, benzophenone-4, and the newer UV filter terephthalylidene dicamphor sulfonic acid (24). The more effective UVA chemical filters are frequently found in sunscreens, namely butyl methoxydibenzoylmethane, octocrylene, and bis-ethylhexyloxyphenol methoxyphenyl triazine, whereas UVB filters that protect mainly from acute sunburn are surprisingly more prevalent in cosmetics, in which we would expect more UVA filters to protect from chronic UVA effects. Also, the more photostable UVA filters (bis-ethylhexyloxyphenol methoxyphenyl triazine, drometrizole trisiloxane, methylene bis-benzotriazolyl tetramethylbutylphenol, terephthalylidene dicamphor sulfonic acid, diethylamino hydroxybenzoyl hexyl benzoate, and diethylhexyl butamido triazone) are rarely found in cosmetics other than sunscreens, at least during the sampling period of the present study. The main UV filter combination in sunscreens butyl methoxydibenzoylmethane and octocrylene is used because butyl methoxydibenzoylmethane is easily photodegraded by UV light, losing its UV protection capacity, and octocrylene is able to stabilize this filter, which therefore keeps its effectiveness. There is a synergistic effect on UV protection of some combinations of UV filters (2), as is apparent in the combination observed in sunscreens [terephthalylidene dicamphor sulfonic acid (e.g. Mexoryl SX) and drometrizole trisiloxane], whereas another synergistic combination was not observed in this analysis, namely the combination of terephthalylidene dicamphor sulfonic acid and bis-ethylhexyloxyphenol methoxyphenyl triazine (e.g. Tinosorb S). Although many UV filters are frequently associated in the same sunscreen or cosmetic, eventually enhancing their sensitizing effect, UV filters are not so frequently reported as a cause of (photo)allergic contact dermatitis (6, 26). This may possibly be explained by the antiinflammatory properties of some UV filters, which, according to animal experiments, inhibit phorbol 2- myristate 3-acetate-induced oedema in the mouse ear to a similar extent as hydrocortisone 7-butyrate or the non-steroidal anti-inflammatory drugs ketoprofen, diclofenac, and niflumic acid (27). The consequences of this phenomenon, apart from interfering with SPF measurement, for diagnosis and sensitization risk are unknown. (Photo)patch testing with sunscreens may be performed less often than is warranted by the patient s history, as (i) facilities for UV exposure are not always available, and/or (ii) too few UV filters are, if at all, (photo)patch tested. A recommendation on an updated and extended photopatch test series has recently been published (). Hence, (photo)contact allergy to UV filters may currently be vastly underdiagnosed. References Lodén M, Beitner H, Gonzalez H et al. Sunscreen use: controversies, challenges and regulatory aspects. Br J Dermatol 20: 6: 2 262. 2 Bonefeld C M, Nielsen M M, Rubin I M et al. Enhanced sensitization and elicitation responses caused by mixtures of common fragrance allergens. Contact Dermatitis 20: 6: 336 342. 3 Uter W, Yazar K, Kratz E M, Mildau G, Lidén C. Coupled exposure to ingredients of cosmetic products: I. fragrances. Contact Dermatitis 203: 69: 33 34. 4 Uter W, Yazar K, Kratz E M, Mildau G, Lidén C. Coupled exposure to ingredients of cosmetic products: II. Preservatives. Contact Dermatitis 204: 70: 29 226. SCCS. Scientific Committee on Consumer Safety Opinion on titanium dioxide (nano form), 22 July 203, 203. Available at: http://ec.europa.eu/health/scientific_ committees/consumer_safety/docs/sccs_o_ 36.pdf (last accessed 2 February 204). 6 SCCS. Scientific Committee on Consumer Safety Opinion on ZnO (nano form), 8 68 Contact Dermatitis, 7, 62 69

September 202, 202. Available at: http://ec.europa.eu/health/scientific_ committees/consumer_safety/docs/sccs_o_ 03.pdf (last accessed 2 February 204). 7 R Development Core Team. R: A Language and Environment for Statistical Computing: Wien, The R Project for Statistical Computing, 203. 8 Yazar K, Boman A, Lidén C. p-phenylenediamine and other hair dye sensitizers in Spain. Contact Dermatitis 202: 66: 27 32. 9 Hamann D, Yazar K, Hamann C R, Thyssen J P, Lidén C.p-Phenylenediamine and other allergens in hair dye products in the United States: a consumer exposure study. Contact Dermatitis 204: 70: 23 28. 0 KerrAC,FergusonJ,HaylettAKetal.A European multicentre photopatch test study. Br J Dermatol 202: 66: 002 009. Gonçalo M, Ferguson J, Bonevalle A et al. Photopatch testing: recommendations for a European photopatch test baseline series.contact Dermatitis 203: 68: 239 243. 2 Yazar K, Boman A, Lidén C. Potent skin sensitizers in oxidative hair dye products on the Swedish market. Contact Dermatitis 2009: 6: 269 27. 3 Yazar K, Johnsson S, Lind M L, Boman A, Lidén C. Preservatives and fragrances in selected consumer-available cosmetics and detergents. Contact Dermatitis 20: 64: 26 272. 4 Schnuch A, Mildau G, Kratz E M, Uter W. Risk of sensitization to preservatives estimated on the basis of patch test data and exposure, according to a sample of 34 leave-on products. Contact Dermatitis 20: 6: 67 74. Gonçalo M, Ruas E, Figueiredo A, Gonçalo S. Contact and photocontact sensitivity to sunscreens.contact Dermatitis 99: 33: 278 280. 6 Bryden A M, Moseley H, Ibbotson S H et al. Photopatch testing of patients: results of the UK multicentre photopatch study group. Br J Dermatol 2006: : 737 747. 7 Hughes T M, Stone N M. Benzophenone 4: an emerging allergen in cosmetics and toiletries?contact Dermatitis 2007: 6: 3 6. 8 Kerr A C. A survey of the availability of sunscreen filters in the UK. Clin Exp Dermatol 20: 36: 4 43. 9 Andersen K E, Goossens A. Decyl glucoside contact allergy from a sunscreen product. Contact Dermatitis 2006: 4: 349 30. 20 Pereira N, Coutinho I, Andrade P, Gonçalo M. The UV filter tinosorb M, containing decyl glucoside, is a frequent cause of allergic contact dermatitis. Dermatitis 203: 24: 4 43. 2 SCCP. Scientific Committee on Consumer Products: safety of nanomaterials in cosmetic products, 8 December 2007, 2007. Available at: http://ec.europa. eu/health/archive/ph_risk/committees/04 _sccp/docs/sccp_o_23.pdf (last accessed 2 February 204). 22 Waters A J, Sandhu D R, Lowe G, Ferguson J. Photocontact allergy to PABA in sunscreens: the need for continued vigilance. Contact Dermatitis 2009: 60: 72 73. 23 Dromgoole S H, Maibach H I. Sunscreening agent intolerance: contact and photocontact sensitization and contact urticaria. J Am Acad Dermatol 990: 22: 068 078. 24 Klein K. Sunscreen products: formulation and regulatory considerations. In: Sunscreens Development, Evaluation and Regulatory Aspects, Lowe N L, Shaath N A, Pathak M A (eds): New York, Marcel Dekker, Inc., 997: pp. 28 3. 2 Moyal D. The development of efficient sunscreens. Indian J Dermatol Venereol Leprol 202: 78(Suppl. ): S3 S34. 26 Shaw T, Simpson B, Wilson B, Oostman H, Rainey D, Storrs F. True photoallergy to sunscreens is rare despite popular belief. Dermatitis 200: 2: 8 98. 27 Couteau C, Chauvet C, Paparis E, Coiffard L. UV filters, ingredients with a recognized anti-inflammatory effect. PLoS One 202: 7: e4687. Contact Dermatitis, 7, 62 69 69