Haw-Yueh Thong Mei-Ping Tseng Chee-Ching Sun

A Survey of the Formaldehyde Content in Commercially Available Shampoos and Skin Care Cosmetics, its Regulation; and a Brief Review of Currently Available Methods of Formaldehyde Determination Haw-Yueh Thong Mei-Ping Tseng Chee-Ching Sun Formaldehyde (FA) is a common source of allergic contact dermatitis. The source of FA exposure is often difficult to be elucidated. It is generally agreed that the most frequent FA exposure source is cosmetics. 1 The 2,4-dinitrophenyhydrazine (2,4-DNPH) derivatization method followed by high performance liquid chromatography (HPLC) analysis was used to determine the FA content in some commercial products in the Taiwan market. 57 skin care cosmetics and 35 hair shampoos were examined. It was shown that 24.6% of the skin care cosmetics contained a FA content > 500 ppm. The FA level in shampoos were much less than that in skin care cosmetics. Of the 35 shampoos investigated, only 14.3% contained a FA level >50 ppm, with a maximal level of 149 ppm. Convenient methods for determining FA were also briefly reviewed in this article.(dermatol Sinica 21 : 209-215, 2003) Key words: Allergic contact dermatitis, High performance liquid chromatography (HPLC) method, 2,4-Dinitrophenylhydrazine (2,4-DNPH) (high performance liquid chromatography) 57 35 24.6% > 500 ppm 35 From the Department of Dermatology, National Taiwan University Hospital Accepted for publication: December 24, 2002 Reprint requests: Haw-Yueh Thong, M.D., Department of Dermatology, No. 7, Chung-Shan South Road, Taipei 100, Taiwan, R.O.C. TEL: 886-2-23562141 FAX: 886-2-23934177 209

14.3% 50 ppm 149 ppm ( 21 : 209-215, 2003) INTRODUCTION Formaldehyde (FA) is a well-known contact allergen. FA dermatitis is acquired most commonly from cosmetics and occupational exposure. 1 Since the patients are subjected to a widespread exposure to FA, the prognosis is poor. An intensive search for FA exposure in the environment can markedly improve the prognosis. The reported incidence of FA sensitivity lies within the range of 2.0% to 6.0%, however, known exposure can be found in only less than 50% of cases. 1 Similar problems were encountered in the contact dermatitis clinic of National Taiwan University Hospital (data submitted for publication). 3.8% of 3,403 patients patch tested from 1978 to 2000 showed a positive reaction to FA. Most patients showed dermatitis involving the head and neck region. 19.5% of our FA-sensitive patients were cosmetic-related. They were exclusively women. Only one patient presented as scalp dermatitis and yielded a positive patch test result from both FA and a hair shampoo. In Taiwan, the maximal permitted limit of detectable free FA in cosmetics is 1,000 ppm. Regulation regarding the declaration of FA concentration does not exist. The standard method for the detection of FA is not accessible under most circumstances. It is practically impossible for the clinicians or patients to identify products containing FA or FA releasers. The purpose of the present investigation was to determine the FA content in commercially available skin care cosmetics and shampoos. The methods for FA determination were also introduced. We would also like to call for an execution of the regulation in labeling the FA concentration in cosmetics and other household products as a preventive measure. MATERIAL AND METHODS Skin care cosmetics and hair shampoos 57 samples of creams, lotions, cleansers, tonics, sunscreens and similar products as well as 35 samples of hair shampoos were obtained from Taiwanese retail shops. The skin care cosmetics analyzed comprised 28 brands, whereas the hair shampoo samples comprised 35 brands. The samples represented commonly used cosmetics and toiletries products in Taiwan. Chemicals Tetrahydrofuran (THF)-water solution (THF : water=9 : 1), 0.1% 2,4-dinitrophenylhydrazine (2,4-DNPH) and 37% FA solution were obtained from Merck, Germany. Fresh FA standard solutions were prepared by using THFwater (9 : 1) to dilute the 37% formaldehyde solution into 37 ppm, 27.7 ppm, 18.5 ppm, 9.2 ppm, 0.92 ppm. Samples were mixed with the THF-water in a 1 : 1 dilution prior to the derivatization procedure. Analysis 1. Derivatization procedure : 0.4 ml of 0.1% 2,4-DNPH was added to 1ml of the diluted sample or standard solution and stirred for 60 sec. The solutions were left to react for 2 minutes at room temperature. 0.4 ml of 0.1M PBS (ph=6.8) and 0.7ml 1M NaOH were then added and the FA content of the final solutions was determined using the HPLC system. 2. HPLC analysis: The instruments used in the HPLC analyses were a Waters 745 / 745B Data Module, a Waters 501 HPLC pump, a Waters U-6-K injector, a Waters 484 tunable absorbance detector, and a Waters automated gradient controller. The analyses were performed on a Dermatol Sinica, September 2003 210

Table I. The content of total formaldehyde in the products investigated Total formaldehyde content (ppm) % of investigated (no.) Skin care cosmetics (n = 57) Undetectable 45.6 ( 26) < 1 10.5 (6) 1-499 19.3 (11) 500-999 7.0 ( 4) 1000-1499 7.0 ( 4) 1500-1999 5.3 ( 3) 2000-2499 1.8 ( 1) > 2500 3.5 (2) Hair shampoos (n = 35) Undetectable 11.4 ( 4) < 1 20.0 (7) 1-49 54.3 (19) 50-99 8.6 ( 3) 100-149 5.7 ( 2) LiChrospherR RP-18 (5 mm, 250-4 mm endcapped) column from Merck, Germany. The mobile phase was acetonitrile-water (1 : 1). The detector was set at 345 nm. The injected volume was 6 ml. Each analysis was performed in triplicate. The amount of FA was calculated from the calibration curve using FA standard solutions as reference. RESULTS The contents of total FA in the products analyzed by HPLC method are displayed in table I. Total FA content in 31 (54.4%) of the analyzed skin care cosmetics lied within the range of 0.1 ppm-2720 ppm, 14 of the products investigated (24.6%) were found to contain > 500 ppm total FA. These products were mostly skin creams, lotions and moisturizers. Total FA content in the hair shampoos was detected within the range of 0.008 ppm-149 ppm. 19 (42.8%)of the products investigated were found to contain > 10ppm total FA. The maximum FA content in shampoos was lower than the skin care cosmetics. DISCUSSION FA is a well-known skin sensitizer and is a widely used preservative in cosmetics, metalworking fluid, textile, paper products, etc. FA is present in one of three sources : as free FA particularly in cosmetics, shampoos and hair care products, as FA donated from a FA-releasing preservative, and as FA resins. The incidence of FA sensitivity varies greatly. Most of the reported figures ranged from 0.9% to19.2%. The incidence of FA allergy in patients with cosmetic dermatitis was estimated to be 1.3% - 9.4%. 1 Cosmetic sensitization to FA is considered to be the most frequently encountered etiology for FA allergy, followed by occupational sensitization. 1 However, FA allergy is sometimes unsuspected by the contact dermatitis specialists. The source of sensitization is frequently unidentifiable as history from the patients is often non-informative. The determination of a positive FA patch test reaction is difficult and relies on the identification of FA sources in the patient's environment. The prognosis of a FA-sensitive patient is poor if the sources of FA could not be eliminated. FA tests for the evaluation of products are important in the clinical investigation of FA exposure. Many techniques are available for the determination of free FA (Table II). The simplest qualitative test for the detection of FA is the Schiff's reagent method (pararosaniline method). In this test, a rose-violet color specif- 211 Dermatol Sinica, September 2003

Table II. Common methods of formaldehyde determination Test Methods Detection Schiff's reagent method 3 Samples added to several Qualitative: rose-violet color drops of pararosaniline specific for FA at 560nm combined with sulfite (Schiff's reagent) for aldehyde detection Chromotropic acid test 7 Samples added to several Qualitative: violet color drops of H2SO4 / Chromotropic Qualitative: the FA content acid solution in 1 / 4 mixture ratio is proportional to time to violet color appearance (or compared with standard solutions) Acetylacetone method 5 Samples reacts with reagent Qualitative: yellow color containing acetylacetone, in 60 o C for 10 minutes and read Qualitative: read 412 nm with spectrophotometer, stock solution comparison Derivatization with 2,4- Sample reacts with 2,4-DNPH Qualitative: read with UV DNPH followed by HPLC to form hydrazone which is then detector, the amount of FA method 12 analyzed with HPLC is calculated from the calibration curve HPLC and post-column After HPLC, the sample reacts Qualitative: read with UV derivatization method 6 with derivatization reagent detector, the FA content is FA : formaldehyde containing 3,5-diacetyl- 1,4-dihydrolutidine calculated from the calibration curve ic for FA at 560 nm is produced when FA reacts with the Schiff's reagent (pararosaniline combined with sulfite). 2 In the present study, quantification of FA is obtained after derivatization with 2,4-DNPH using the HPLC method. 3 This is a sensitive method, but the co-existing FA donors are often degraded with this method, making it less suitable for the determination of free FA when FA donors are present in the formula. To compensate for this drawback, Karlberg et al. proposed the "closed container diffusion method", which seemed suitable for the detection of free FA in products preserved with FA donors. 4 The standard method for FA detection in European Union (EU) adopted the HPLC with post-column derivatization method. This method is based on the principle from the "acetylacetone method". 5 In the "acetylacetone method", free FA is condensed with ammonium acetate and acetylacetone (pentane-2,4-dione) to form fluorescent 3,5-diacetyl-1,4-dihydrolutidine. Quantitative analysis of FA is eligible in this rapid method. The HPLC method with post-column derivatization is then used to separate free FA from FA donors. However, this method is too complicated and not clinically applicable. 6 The most frequently used method is the "chromotropic acid method". 7 This is an indirect, simple, qualitative method suitable for use in clinics. The analysis is based on the violet color formed in the reaction of free FA and chromotropic acid. However, false negative results tend to occur when other aldehydes and ketones coexist in the samples. 8 False positive reactions have also been reported, compromising the reliability of this method. 9 These problems can be solved by using a modification or a 10, 11 complementary method. Gryllaki-Berger et al. 12 compared the efficacy of the chromotropic acid test, the acetylacetone test and HPLC method. They found that the acetylacetone test was more efficient in Dermatol Sinica, September 2003 212

Table III. The regulations regarding formaldehyde content and its declaration in cosmetics among various countries @ Country Restrictions of free FA in Restrictions of FA-donor Detection method cosmetics preservatives* Taiwan 0.10 % free FA level 1.DMDM hydantoin: 0.60% HPLC method (released by FA releasing 2.Imidazolidinl urea: 0.60% preservatives) 3.Quarternium 15: 0.20% USA 0.20% for rinse-off products NR NA (not allowed for leave-on products) (labeling FA at 0.1% or more) European 0.20% # 1.DMDM hydantoin: 0.60% HPLC method union 2.Imidazolidinyl urea: 0.60% 3.Quaternium 15: 0.20% China 0.20% # 1.Imidazolidinyl urea: 0.60% NA in cosmetics 2.Glydant/Quarternium-15: 0.20% as free FA India 0.20% # NR NA Japan FA is not allowed to be 1.DMDM hydantoin: 0.30% HPLC method added into cosmetics. 2.Imidazolidinyl urea: 0.30% The generally acceptable for shampoo (cosmetics and level of free FA as "quasi-drug") impurities in cosmetics by the government : 75ppm Korea 0.20% 1.DMDM hydatoin: 0.60% NA (2% for hair care products) 2.Imidazolidinyl urea: 0.60% 3.Diazolidinyl urea: 0.30% 4.Quarternium 15: 0.20% Malaysia 0.20% Imidazolidinyl urea: 0.60% NA Philippines 0.20% NR NA Thailand 0.20% NR NA Australia NR NR NA New Zealand NR NR NA Singapore NR NR NA Vietnam NR NR NA FA : formaldehyde * : FA-donor preservatives include DMDM hydantoin, Imidazolidinyl urea, Quaternium 15, Diazolidinyl urea, 2-bromo-2-nitropropane-1,3-diol) # : must label "containing FA if more than 0.05% NR : not restricted NA: not available @ : 213 Dermatol Sinica, September 2003

determining the FA content in a clinical laboratory. Nevertheless, the ultimate choice of a FA test is dependent on whether a quick laboratory work is given priority (e.g. the chromotropic acid test which is more easily performed but has a delay in the test results for several days) or the test results are needed quickly (e.g. the acetylacetone test which may provide the answers on the first day of testing), as well as the available facilities of the dermatology clinic. The regulations on the declaration of FA and the permitted FA content vary from country to country. Table III is listed for reference because most of the cosmetics in Taiwan are imported. The content of FA as a preservative in cosmetics is regulated in Europe by an EEC directive, 76 / 768 / EEC. The declaration of FA in cosmetics is legally obligatory in Germany since 1979. In Denmark, the acceptable FA content in cosmetic products is 0.2%. Products containing > 0.05% FA must be labeled "contains formaldehyde". 13 In Taiwan, the acceptable limit of free FA detectable in cosmetics is 0.1%, but the declaration of FA concentration is not required. The results of the present survey demonstrated that 59.8% (55 of 92 samples) of the commonly used skin care cosmetics and hair shampoos in Taiwan contained FA. In hair shampoos, FA was present in a much lower concentration. With the low concentration of FA, the short period of application and the low contact dermatitis reactivity of the scalp, the risk of FA dermatitis imposed by shampoos could be lower. 14 The prognosis of a FA-induced allergic contact dermatitis depends on the ability to elucidate and then eliminate the FA exposure. For further clarification of the exposure source, we may refer to the label or the material safety data sheets of the suspicious agents. We may also perform a chemical analysis on the potential FA - containing products. We may also contact the suppliers, manufacturers, information centers on toxic material for further information. To conclude, we demonstrated that some skin care cosmetics and hair shampoos in the Taiwan market contained a certain amount of FA. Our results may help the clinicians to clarify the exposure source of FA in a FA-sensitive individual. The chromotropic acid test or the acetylacetone method seems to be a useful screening tool clinically whereas the HPLC method can be used for further confirmation and precise quantification of the FA level. REFERENCES 1. Fransway AF: The problem of preservation in the 1990s: I. Statement of the problem, solution(s) of the industry, and the current use of formaldehyde and formaldehydereleasing biocides. Am J Contact Dermatitis 2: 6-23, 1991. 2. Guin JD: Practical contact dermatitis. A handbook for the practitioner. New York: McGraw-Hill, 784-786, 1995. 3. Selim S: Separation and quantitative determination of traces of carbonyl compounds as their 2,4-dinitrophenylhydrazones by high-pressure liquid chromatography. J Chromatography 136: 271-277, 1977. 4. Karlberg A-T, Skare L, Lindberg I, et al.: A method for quantification of formaldehyde in the presence of formaldehyde donors in skin-care products. Contact Dermatitis 38: 20-28, 1998. 5. Commission Directive of 4 th April 1990 (90/207 EEC) (IV): Identification and determination of free formaldehyde. Official Journal of European Communities no. L108, 28/04/90, pp. 92-101. 6. Engelhardt H, Klinkner R: Determination of free formaldehyde in the presence of donators in cosmetics by HPLC and post-column derivatization. Chromatographia 20: 559-565, 1985. 7. Blohm S-G: Formaldehyde contact dermatitis (I). A simple method for determination of small amounts of formaldehyde. Acta Derm Venereol 39: 450-453, 1959. 8. Flyvholm M-A, Menne T: Allergic contact dermatitis from formaldehyde. A case study focusing on sources of formaldehyde exposures. Contact Dermatitis 27: 27-36, 1992. Dermatol Sinica, September 2003 214

9. Dahlquist I, Fregert S, Gruvberger B: Reliability of the chromotropic acid method for qualitative formaldehyde determination. Contact Dermatitis 6: 357-358, 1980. 10. Flyvholm M-A, Tiedemann E, Menne T: Comparison of 2 tests for clinical assessment of formaldehyde exposure. Contact Dermatitis 34: 35-38, 1996. 11. Fregert S, Dahlquist I, Gruvberger B: A simple method for the detection of formaldehyde. Contact Dermatitis 10: 132-134, 1984. 12. Gryllaki-Berger M, Mugny C, Perrenoud D, et al.: A comparative study of formaldehyde detection using chromotropic acid, acetylacetone and HPLC in cosmetics and household cleaning products. Contact Dermatitis 26: 149-154, 1992. 13. Rastogi SC: A survey of formaldehyde in shampoos and skin creams on the Danish market. Contact Dermatitis 27: 235-240, 1992. 14. Bork K, Heise D, Rosinus A: Formaldehyde in hair shampoos. Derm Beruf Umwelt 27: 10-12, 1979. 215 Dermatol Sinica, September 2003