II
Illllll
I
l llll
9 II
I
I
Ill
II
I
Allergic contact dermatitis from formaldehyde resins in permanent press clothing: An underdiagnosed cause of generalized dermatitis Joseph F. Fowler, Jr., MD, a Steven M. Skinner, MD, a and Donald V. Belsito, M D b Louisville, Kentucky, and New York, New York
Background and Methods: Formaldehyde resins have been used to impart wrinkle resistance to clothing fabrics since 1926. After several patients with positive patch tests to formaldehyde resins had been examined, a study was undertaken of the records of all patch tests performed at the University of LouisviLle Patch Test Clinic and the Allergy Section of the Skin and Cancer Clinic of New York University Medical Center from January 1988 through April 1990 to determine the prevalence of positive patch-test reactions to formaldehyde-based textile resins and the clinical and demographic patterns associated with textile resin allergy. Results: Seventeen patients were identified at the two centers. Twelve were allergic to formaldehyde as well as to formaldehyde textile resins. Several clinical patterns were found, including accentuation of dermatitis in areas of tight clothing, primary occurrence in clothingcovered areas, and a chronic recalcitrant course. Ethylene urea melamine formaldehyde resin was the best screening agent with 14 definite positive reactions and one equivocal reaction. Conclusion: Formaldehyde textile resin allergy is more common than has been previously recognized. Patch testing with one or more formaldehyde textile resins is indicated in patients with a particular pattern of dermatitis. (J AM ACAD DERMATOL 1992;27:962-8.)
Treatment of fabric with formaldehyde-containhag resins to reduce wrinkling was first undertaken in 1926. In 1959, Marcussen I reported cases of allergic contact dermatitis ( A C D ) caused by formaldehyde in clothing fabrics, Since then, there have been relatively few reports of contact hypersensitivity to formaldehyde or formaldehyde resins in textiles, although formaldehyde is a well-known allergen and is included in standard patch testing trays. At two institutions we identified 17 cases of contact dermatitis to formaldehyde resins in textiles. This report is intended to clarify the clinical presentation of patients with A C D to permanent press clothing finishes, discuss the relation of formaldehyde allergy to formaldehyde resin allergy, and illustrate the patch testing technique to identify the allergens. We believe that allergy to permanent press clothing is underdiagnosed. With a higher inFrom the Departmentof Medicine,Divisionof Dermatology,University of Louisvillea;and the Departmentof Dermatology,New York UniversityMedicalCenter.b Acceptedfor publicationJune 8, 1992. Reprint requests:JosephF. Fowler,Jr., MD, 310 E. Broadway,Suite 100, Louisville,KY40292. 1•/I/40027 962
dex of suspicion and proper patch testing, more patients with this condition can be identified. METHODS This study was performed retrospectively on 678 patients seen at The University of Louisville Patch Test Clinic and 344 patients seen at the Allergy Section of the Skin and Cancer Clinic, New York University Medical Center from January 1988 through April 1990. All patients were referred for patch testing because of an eczematous dermatitis believed to be allergic. These patients were patch tested with commercially prepared textile allergens obtained from either Chemotechnique Diagnostics AB (Malmo, Sweden) or Hermal Kurt Hermann (Reinbek/Hamburg, Germany). In addition, all patients were tested to the standard screening tray of the North American Contact Dermatitis Group (NACDG) and to a series of other allergens selected individually by the investigators. Patch tests were placed in Finn Chambers as recommended by the NACDG and were removed at 48 hours with immediate readings at 48 hours and delayed readings at either 4 days (Kentucky) or 7 days (New York). CASE REPORTS Several case reports illustrate particular features of formaldehyde resin allergy.
Volume 27 Number 6, Part 1 December 1992
ACD from formaldehyde resins 963
T a b l e I. Demographic data of patients
Testing center Louisville Patients All tested Male Female Formaldehyde resin allergy Total Male Female Percent of total
No. of pafien~
678 288 390
I
13 7 6 1.9
Case 2 (formaldehyde negative). A 27-year-old, white male restaurant manager developed a dermatitis of the axillae and antecubital areas that he believed may have been caused by a new shirt he had worn 4 days earlier. He had two similar flares several months later. After the dermatitis cleared, patch tests were placed. Results are given in Table III. He tested positive to three of the formaldehyde resins while testing negative to formaldehyde itself. Case 11 (generalized dermatitis). A 75-year-old housewife had a lifelong history of atopic dermatitis of the eyelids, antecubital and popliteal fossae, and the back of the neck. For 3 years, she had had an extremely pruritic generalized eruption that was most severe on the upper back, lateral chest, medial arms, and posterior calves. This failed to respond to topical therapy but improved with systemic corticosteroids. The dermatitis significantly worsened during the winter months (while wearing more clothing) and improved during the summer months. Patch test results in Table III indicate reactions to ethylene urea melamine formaldehyde resin and quaternium15. In addition, she had a positive reaction to her thermal underwear and a questionable reaction to a flannel nightgown. With proper counseling, the severe dermatitis remitted, although she remained bothered by the underlying atopic diathesis. Case 12 (occupationally related). A 40-year-old white female sewing machine operator handled many fabric and textile materials. A generalized dermatitis developed with severe involvement of the hands and some involvement of the arms and legs. Patch testing revealed positive reactions to formaldehyde and several formaldehyde resins as well as two rubber allergens and a fragrance (oak moss absolute). After discussion of the results of the testing with the patient and her employment counselors, she was given a job unrelated to the textile industry. Along with avoiding permanent press clothing, this has resulted in good improvement in her dermatitis. Case 15 (nummular dermatitis). This 43-year-old woman had a 5-year history of an intermittent patchy
New York Average age (yr)
42.6
45.5
No. of patients 344 98 246 4 1 3 1.2
I L
Average age (yr)
46.6
49.0
dermatitis of the legs. Patch testing revealed reactions to formaldehyde resins as well as formaldehyde-releasing preservatives. Avoidance of permanent press clothing and skin and hair care products containing formaldehyde-releasing preservatives resulted in clearing of the dermatitis. RESULTS As can be seen from Table I, the demographic data on these patients in terms of age and sex do not differ significantly between the two centers. H o w ever, the Louisville patients were somewhat younger than the New York patients. In N e w York, there was a female preponderance of all patch-tested patients as well as the formaldehyde resin-positive patients. In Louisville, however, more women were tested, but more men had positive reactions. Although the percentage of blacks and Hispanies patch-tested at each center was consistent with the local population mix, all patients with textile-resin allergy were white. The clinical features for each of the 17 patients aUergic to formaldehyde resins are summarized in Table II. Of the 17 patients with ACD to formaldehyde resins, five (three garment workers, one printer, one chemical worker) had occupationally related disease. These patients were younger than the nonoccupational patients, with an average age of 40 versus 49 years. They also tended to have more hand involvement than the other patients. Three of the four patients older than 55 years of age had generalized erythroderma, which suggests a more severe involvement in older patients. All patients had dermatitis on areas normally covered with clothing. Six of the 17 had accentuation in areas of tight clothing such as waistband and flexurai areas.
964
Journal of the American Academy of Dermatology
Fowler et aL
Table II. Clinical data on 17 patients allergic to textile resins Patient No.
, Age (yr!,
[ Sex
Work-related
1
66
M
No
2
27
M
No
3
45
M
No
4 5
40 46
F M
6 7 8
50 43 41
M M F
No Yes (Chemical worker) No No No
9
40
F
Yes (garment worker)
10
40
M
No
11 12
75 40
F F
13
38
M
14 15 16
59 43 36
F F F
17
59
F
No Yes (garment worker) Yes (printing worker) No No Yes (garment worker) No
t
Other allergens (nonformaldehyde)
Distribution
Face, V of neck, arms, and hands Axillae, antecubital fossae, waist, legs Trunk, chest, antecubital fossae Diffuse Neck, chest, waist, abdomen, axillae Trunk, arms, hands Trunk, hands Anterior and posterior axillary folds Upper and mid back, waistband, periaxillary areas, dorsal hands, buttocks, and thighs Waistband, posterolaterai thighs and buttocks Generalized erythroderma Generalized, hams and arms worst Covered areas only, except hands Generalized erythroderma Nummular, scattered patches Face, neck, arms Generalized
Fragrances None None None None None Nickel Nickel and thimerosal Colophony, cinnamic alcohol, cinnamic aldehyde, toluene, sulfonamide resin Disperse brown # I None Rubber (carba, thiuram), oak moss
Nofle
None None Nickel, cobalt Kathon CG, rubber (thiuram)
No~: Allpatientswerewhite. The head and neck were rarely involved, except in cases of generalized erythroderma. Nine patients (53%) were patch-test negative to nonformaldehyde allergens. Of the eight with other positive reactions, three (18%) were allergic to nickel, three (18%) to fragrances, and two (12%) to rubber allergens. Several other positive reactions are noted in Table II. Table III lists the full range of positive reactions to formaldehyde-related allergens. Five patients (30%) were not allergic to formaldehyde itself. A mixture of ethylene urea and melamine formaldehyde resins was the best screening agent, with a positive rate of 82.3% (14 of 17 patients). Two of the three patients with negative patch tests had occupationally related ACD, which may reflect a specific work-exposure pattern. Many patients reacted to one or more commonly used formaldehyde-releasing preservatives, as shown in Table III.
DISCUSSION The incidence of formaldehyde-related textile A C D is unknown. However, since 1968 the incidence of formaldehyde sensitivity has been reported to be between 3% and 8%. 2 Recent data f r o m the N A C D G indicated that 132 of 1647 patients tested had positive patch tests to formaldehyde. O f these, at least 83 (51%) were deemed relevant to the presenting problem. In 1982, Andersen and H a m a n n 3 estimated that 8.6% of patients allergic to formaldehyde were sensitive to textiles. Similar studies between 1959 and 1964 placed the number between 8.6% and 65%. 2 Textile-related A C D has been reported to be more common in women than in men; the reported rates vary from 3:14 to as high as 5:1 5 Our data show a more even distribution with nine women and eight men affected. Although sensitivity can occur at any age, there appears to be a peak incidence between the ages of 35 and 60 years. 2 This
Volume 27 Number 6, Part 1 December 1992
ACD from formaldehyde resins 965
T a b l e III. Reactions to formaldehyde-related allergens Patient No. Allergens
Formaldehyde (1% aq) Dimethylol dihydroxyethylene urea formaldehyde resin (Fixapret r (4.5% aq) Dimethylol propylene urea formaldehyde resin (Fixapret PH) (5.0% aq) Tetramethylol acetylenediurea formaldehyde resin (Fixapret 140)
'
I I' I I I'J
1-;
Too,, ,
3+
-
1+ 2+
1+ 1+ 1+ 1+ 1+
1+
1+
-
2+
1+
2+
1+ 2+
1+
-
1+ 1+ 1+
2+ 1+ 2+
1+
-
1+
2+
?
2+
1+
-
1+ 2+ 2+ 1+ 2+ 2+ t +
2+
-
2+
-
1+ 1+ 2+
2+
1+ 2+ 2+
-
1+ 1+
-
1+ 12(70.6)
1+ 1+
?
5(29.4)
1+
1+
-
1+ I+
-
1+
1+ Ii(64.7)
1+
8(47.1)
(5.0%) Ethylene urea melamine formaldehyde resin (Fixapret AC) (10% pet) Urea formaldehyde resin (Kaurit S) (10% pet) Melamine formaldehyde resin (Kaurit M70) (7%) Ethylene urea formaldehyde resin (1.0% pet) Phenol formaldehyde resin (1% pet) Quaternium- 15
NTNTNTNT
-
=
2+ 1+ 1+
-
1+ 14(82.3)
3+ 1+
1+ 1+
-
1+ 10(58.8)
1+
2+ 1+ 2+
-
1+ 11 (64.7)
I+ 2+
NT NT N T 2+
I+
1+ 2+ N T N T N T N T N T N T 2+
3+
1+ 1+
-
1+ 3+
-
1+
1+
-
1+ 1+ 1+
-
(4) -
1 (5.9)
2+ 11(64,7)
(2% pet) Glutaraldehyde (1% aq)
-
Diazolidinylurea (ITo pet or 1% aq) Imidazolidinylurea
I+
(2%) D M D M hydantoin (1% aq) Bioban CS 1135 (1% pet) Bioban CS 1246 (1% pet)
1 (5.9)
1+ 1+
-
1+ 1+
1+
1+
2+
6(35.3)
1+
2 (11.8)
1+
1+ NT NT NT NTNT
-
2 (11.8)
1+ N T N T N T N T N T N T NT N T N T N T NT
N T N T N T N T NT I+ N T N T N T N T N T N T NT N T N T N T N T
Grading systemof NACDG used. aq, Aqueous;NT, not tested;pet, petrolatum. was confirmed in our patient population; only 3 of 17 fell outside this range. In its early stage, textile resin-related ACD appears as a pruritic, paputovesieular, erythematous
eruption that m a y subsequently become chronic and lichenified if not recognized and treated}, 6 T h e dermatitis m a y be most prominent at follicular orifices. 7 Although most reports note a characteristic
Journal of the American Academy of Dermatology
Fowler et al.
966
Table
IV. Cross-linking fabric finishing agents (durable-press)
Resin type
Polycondensation polymers (1930s) Urea formaldehyde (dimethylol urea)
Melamine
Linear reactants Dimethylol carbamates (1960s)
Cyclic ureas (1960s) N-methylol propylene urea derivatives: (dimethylolpropylene urea) (dimethylolethylene urea) Dimethylol dihydroxyethylene urea (1960s)
DMDHEU blended or reacted with glycols Dirnethoxymethyl dihydroxyethylene urea (1980s) (methylated DMDHEU)
Abbreviation
DMU
MF
DMC
Eslimated total market share (1990)
6%
20%
3%
3% DMPU DMEU DMDHEU
16%
34%
DMDHEU
10%
,. Uses and characteristics
Cotton pile fabrics, velvet, corduroy coatings, old-fashioned women's dress goods, flannelettes, upholstery, viscous rayon, used with some water repellents Imparts crease and shrink resistance to cotton and rayon, not chlorine resistant, increases wash fastness, but washes out faster than others, inexpensive "Heavy fabrics," "stiff" finishes, draperies, coatings, some collars, some water and oil repellent and flame retardant finishes; effective on synthetic and natural fibers Used with many other finishes to improve durability to washing and dry cleaning Less than 2% of clothing; some on sheeting, pillow cases, cotton-polyester blends, fabricliners and draperies, white shirts Soft finish, excellent chlorine resistance, not very light fast and incompatible with some dyes, more expensive Colored cotton textiles and white goods, shirts, some upholstery, more popular in France, Italy, and Spain; little or no use in U.S. Soft finish, excellent chlorine resistance, used in paper and red wine DMDHEU and its derivatives are used on 80%-90% of U.S. apparel fabric blends; "wash and wear", colored and white fabrics and shirtings; not used on 100% pure synthetic fabrics; also used on draperies and sheeting; chlorine resistant, stiffer than earbamates, some paper use, relatively inexpensive Performance similar to DMDHEU but slightly slower curing As etherified glyoxal reactant; similar uses and characteristics as DMDHEU with lower formaldehyde release, more expensive than DMDHEU
Relative formaldehyde release
High
High
Medium
Medium
Low
Very low Low
ThistablewasadaptedwithpermissionfromFisherA.A,ed. Contactdermatitis.3rd cd.Philadelphia:Lea & Febiger,1986:288.We alsoacknowledge the generousassistanceof membersof the textileindustry,especiallyJ. C. Winchester,AmericanCynamidCo.,Manager, CharlotteLaboratory, Charlotte, N.C.;M. D. Hurwitz,Professor,Clothing& Textiles,Universityof North Carolina,Greensboro,N.C.; P. S. Pal, ScientificApplications & InnovationsCorp.(textileindustryconsultant),Charlotte,N.C.; and R. R. Stewart,Proctor & GambleCo. (textilescientist),Cincinnati,Ohio.
Volume 27 N u m b e r 6, Part 1 December 1992
ACD from formaldehyde resins 967
Table IV. Cont'd. Estimated total market share Resin type
Abbreviation
(~99o)
Uses and characteristics
Soft, white fabrics, used on collar linings in past durable finishes, some drapery, and upholstery use Not as chlorine resistant as others; little use in U.S. but may be added to MF or mixed with other finishhag agents Children's apparel at present ha U.S. May be used in Japan on polyester/cotton blends or on Marks & Spencer fabrics ha England, expensive; not as effectiveas DMDHEU or its derivativesfor shrinkage control or durable press appearance
Uron formaldehyde (1950s) N,N'-Bis (methoxymethyl) uron
UF
1%
N,N'-dimethyl 4,5 dihydroxyethylene
DHDMI or DMcDHEU
1%
distribution (e.g., the axillary folds in women and the neck in men), our patients tended to have a more diffuse distribution (see Table II). Covered areas tended to be most frequently involved, with the face, legs, and feet often spared. A pattern clinically suggestive of n u m m u l a r eczema was seen in several patients. In addition to formaldehyde resins, t2 of our patients (70%) were allergic to formaldehydereleasing preservatives. Of these, most reacted to quaternium-15. Clinically, these reactions may be significant because these preservatives are commonly used in skin care products and topical corticosteroid creams. This secondary ACD may lead to diagnostic confusion if complete patch testing is not undertaken. In 1934, Marcussen 1 reported the first case of A C D caused by formaldehyde in textiles. However, it was not until 1965 that the first authenticated U.S. cases were reported by O'Quinn and Kennedy. 6 In recent years, there has been some debate as to whether the A C D associated with permanent press clothing is actually caused by formaldehyde, the formaldehyde resins, or both. Storrs, 7 Hatch and Maibach, 2 and Malten 8 have reported cases in which patients were sensitive to formaldehyde or formaldehyde resins, but not necessarily to both, whereas Andersen and Hamann 3 reported, "A negative patch test to formaldehyde virtually excluded allergic contact dermatitis from textile finishes."
Relative formaldehyde release
High
None
(p. 66) Our data clearly support the idea that a patient can be allergic to the formaldehyde resins without being allergic to formaldehyde itself as 5 of the 17 (30%) were allergic to formaldehyde resins but not to formaldehyde. In our experience, ethylene urea-melamine formaldehyde proved to be the best screening allergen and clearly identified 14 of 17 patients. There are two plausible theories as to how formaldehyde sensitivity can result in sensitivity to permanent press clothing. The quality of wrinkle resistance results from the formation of covalent cross finks between adjacent cellulose molecules that form when cellulose and formaldehyde react with a resin. Because this is a heat-driven reaction that occurs in the presence of an acid catalyst (e.g., zinc nitrate, magnesium chloride, or alkanolamine hydrochloride),2, 7 formaldehyde release could occur by one of two means: (1) when there is incomplete cross-linking of some resin, which allows degradation of the resin by chlorine bleach with release of formaldehyde; and (2) ff acid catalyst remains in the fabric, such that formaldehyde may be released by hydrolysis under warm moist conditions.7 Each of the formaldehyde resins used in durable press fabrics can be categorized according to their relative formaldehyde release (Table IV). Those being classified as "high" formaldehyde releasers include dimethylol propylene urea, urea formaldehyde (dimethylol
Journal of the American Academy of Dermatology
968 Fowler et al.
urea), melamine formaldehyde, and uron formaldehyde. "Medium" releasers include dimethylol ethylene urea and dimethylol carbarnates. 7 In contrast, dimethylol dihydroxyethyleneurea and dimethoxymethyl dihydroxyethyleneurea, which represent the major resins used in clothing fabrics manufacturers in the United States, release low levels of formaldehyde. Furthermore, dirnethyl 4,5 dihydroxyethylene urea apparently does not release formaldehyde. This finish is used primarily on children's clothing in the United States, in part because of its expense. 7 In 1974, Schorr et al.9 studied 112 samples of American clothing and discovered that 18 of these (16%) had a free formaldehyde content of more than 750 ppm. Fisher et al. I~ suggested that dermatitis may occur when formaldehyde content is 750 ppm, whereas Berrens et al.ll believe that the critical level may be closer to 500 ppm. In 1974 the textile industry claimed that levels of formaldehyde release had decreased so that 100 to 200 ppm was standard. 12 However, Hatch and Maibach2 noted that the United States has neither a mandatory nor a vohntary formaldehyde level restriction. Our findings that patients were allergic to the resin but not to formaldehyde are consistent with the reports of lowerformaldehyderelease from U.S. textiles. However, during the past decade, there has been a progressive shift in this country to imported textiles, especially from the Far East, where little information can be obtained concerning the resins used in finishing fabrics, the curing process, or the number of afterwashes. In addition to formaldehyde resins, 12 of our pa-
tients (70%) were allergic to formaldehyde-releasing preservatives. Of these, most reacted to quaternium-15. Clinically, these reactions may be significant because these preservatives are commonly.used in skin care products and topical corticosteroid creams. This secondary ACD may lead to diagnostic confusion if complete patch testing is not undertaken. REFERENCES 1. Marcussen PV. Contact dermatitis due to formaldehyde in textiles, 1934-1958. Acts Derm Venereol (Stockh) 1959; 39:348-56. 2. Hatch KL, Maibach HI. Textile chemical finish dermatitis. Contact Dermatitis 1986;14:1-13. 3. Andersen KEo Hamatm K. Cost benefit of patch testing with textile finish resins. Contact Dermatitis 1982;8:64-7. 4. Wereide K. Formaldehyde as a contact allergen in textiles, Aeta Allergol 1964;19:351-63. 5. Cronin E. Formalin textile dermatitis. Br J Dermatol 1963;75:267-73. 6. O'Quinn SE, Kennedy CB. Contact dermatitis due to formaldehydein clothing textiles. J A M A 1965;194(6):58396. 7. Storrs FJ. Dermatitis from clothing and shoes. In: Fisher AA, ed. Contact dermatitis, 3rd ed. Philadelphia: Lea & Febiger, 1986:283-337. 8. Malten KE. Textile finish contact hypersensitivity. Arch Dermatol 1964;100:215-21. 9. Schorr WF, Keran E, Plotka E. Formaldehyde allergy: the quantitative analysis of American clothing for free formaldehyde and its relevance in clinical practice. Arch Dermatol 1974;I 10:73-6. 10. Fisher AA, Kanof NB, Biondi EM. Free formaldehyde in textile and paper. Arch Dermatol 1962;86:753-6. 11. Berrens L, Young E, Jansen LH. Free formaldehyde in textiles in relation to fo/'rnalin contact sensitivity. Br J Dermatol 1964;76:110-5. 12. Vail SL, Reinhardt RM. What do formaldehyde tests measure? Textile Chemist Colorist 1981;13:13-22.
BOUND VOLUMF_SAVAILABLE TO SUBSCRIBERS Bound volumesof the JOURNALOFTHEAMERICANACADEMYOFDERMATOLOGYare availableto subscribers(only) for the 1992 issues from the Publisherat a costof $66.00 for domestic, $88.62 for Canadian, and $84.00 for international for volume26 (January-June) and volume 27 (July-December).Shipping charges are included. Each bound volume contains a subject and author index and all advertising is re. moved. Copiesare shippedwithin 60 days after publication of the last issue in the volume.The binding is durable buckram with the journal name, volumenumber, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Ycar Book, Inc., SubscriptionServices, 11830 W~t|ine Industrial Dr., St. Louis, MO 63146-3318. USA: phone (800) 325-4177, ext. 4351; (314) 453-4351. Subscrlptiottr must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.
Illllll
I
l llll
9 II
I
I
Ill
II
I
Allergic contact dermatitis from formaldehyde resins in permanent press clothing: An underdiagnosed cause of generalized dermatitis Joseph F. Fowler, Jr., MD, a Steven M. Skinner, MD, a and Donald V. Belsito, M D b Louisville, Kentucky, and New York, New York
Background and Methods: Formaldehyde resins have been used to impart wrinkle resistance to clothing fabrics since 1926. After several patients with positive patch tests to formaldehyde resins had been examined, a study was undertaken of the records of all patch tests performed at the University of LouisviLle Patch Test Clinic and the Allergy Section of the Skin and Cancer Clinic of New York University Medical Center from January 1988 through April 1990 to determine the prevalence of positive patch-test reactions to formaldehyde-based textile resins and the clinical and demographic patterns associated with textile resin allergy. Results: Seventeen patients were identified at the two centers. Twelve were allergic to formaldehyde as well as to formaldehyde textile resins. Several clinical patterns were found, including accentuation of dermatitis in areas of tight clothing, primary occurrence in clothingcovered areas, and a chronic recalcitrant course. Ethylene urea melamine formaldehyde resin was the best screening agent with 14 definite positive reactions and one equivocal reaction. Conclusion: Formaldehyde textile resin allergy is more common than has been previously recognized. Patch testing with one or more formaldehyde textile resins is indicated in patients with a particular pattern of dermatitis. (J AM ACAD DERMATOL 1992;27:962-8.)
Treatment of fabric with formaldehyde-containhag resins to reduce wrinkling was first undertaken in 1926. In 1959, Marcussen I reported cases of allergic contact dermatitis ( A C D ) caused by formaldehyde in clothing fabrics, Since then, there have been relatively few reports of contact hypersensitivity to formaldehyde or formaldehyde resins in textiles, although formaldehyde is a well-known allergen and is included in standard patch testing trays. At two institutions we identified 17 cases of contact dermatitis to formaldehyde resins in textiles. This report is intended to clarify the clinical presentation of patients with A C D to permanent press clothing finishes, discuss the relation of formaldehyde allergy to formaldehyde resin allergy, and illustrate the patch testing technique to identify the allergens. We believe that allergy to permanent press clothing is underdiagnosed. With a higher inFrom the Departmentof Medicine,Divisionof Dermatology,University of Louisvillea;and the Departmentof Dermatology,New York UniversityMedicalCenter.b Acceptedfor publicationJune 8, 1992. Reprint requests:JosephF. Fowler,Jr., MD, 310 E. Broadway,Suite 100, Louisville,KY40292. 1•/I/40027 962
dex of suspicion and proper patch testing, more patients with this condition can be identified. METHODS This study was performed retrospectively on 678 patients seen at The University of Louisville Patch Test Clinic and 344 patients seen at the Allergy Section of the Skin and Cancer Clinic, New York University Medical Center from January 1988 through April 1990. All patients were referred for patch testing because of an eczematous dermatitis believed to be allergic. These patients were patch tested with commercially prepared textile allergens obtained from either Chemotechnique Diagnostics AB (Malmo, Sweden) or Hermal Kurt Hermann (Reinbek/Hamburg, Germany). In addition, all patients were tested to the standard screening tray of the North American Contact Dermatitis Group (NACDG) and to a series of other allergens selected individually by the investigators. Patch tests were placed in Finn Chambers as recommended by the NACDG and were removed at 48 hours with immediate readings at 48 hours and delayed readings at either 4 days (Kentucky) or 7 days (New York). CASE REPORTS Several case reports illustrate particular features of formaldehyde resin allergy.
Volume 27 Number 6, Part 1 December 1992
ACD from formaldehyde resins 963
T a b l e I. Demographic data of patients
Testing center Louisville Patients All tested Male Female Formaldehyde resin allergy Total Male Female Percent of total
No. of pafien~
678 288 390
I
13 7 6 1.9
Case 2 (formaldehyde negative). A 27-year-old, white male restaurant manager developed a dermatitis of the axillae and antecubital areas that he believed may have been caused by a new shirt he had worn 4 days earlier. He had two similar flares several months later. After the dermatitis cleared, patch tests were placed. Results are given in Table III. He tested positive to three of the formaldehyde resins while testing negative to formaldehyde itself. Case 11 (generalized dermatitis). A 75-year-old housewife had a lifelong history of atopic dermatitis of the eyelids, antecubital and popliteal fossae, and the back of the neck. For 3 years, she had had an extremely pruritic generalized eruption that was most severe on the upper back, lateral chest, medial arms, and posterior calves. This failed to respond to topical therapy but improved with systemic corticosteroids. The dermatitis significantly worsened during the winter months (while wearing more clothing) and improved during the summer months. Patch test results in Table III indicate reactions to ethylene urea melamine formaldehyde resin and quaternium15. In addition, she had a positive reaction to her thermal underwear and a questionable reaction to a flannel nightgown. With proper counseling, the severe dermatitis remitted, although she remained bothered by the underlying atopic diathesis. Case 12 (occupationally related). A 40-year-old white female sewing machine operator handled many fabric and textile materials. A generalized dermatitis developed with severe involvement of the hands and some involvement of the arms and legs. Patch testing revealed positive reactions to formaldehyde and several formaldehyde resins as well as two rubber allergens and a fragrance (oak moss absolute). After discussion of the results of the testing with the patient and her employment counselors, she was given a job unrelated to the textile industry. Along with avoiding permanent press clothing, this has resulted in good improvement in her dermatitis. Case 15 (nummular dermatitis). This 43-year-old woman had a 5-year history of an intermittent patchy
New York Average age (yr)
42.6
45.5
No. of patients 344 98 246 4 1 3 1.2
I L
Average age (yr)
46.6
49.0
dermatitis of the legs. Patch testing revealed reactions to formaldehyde resins as well as formaldehyde-releasing preservatives. Avoidance of permanent press clothing and skin and hair care products containing formaldehyde-releasing preservatives resulted in clearing of the dermatitis. RESULTS As can be seen from Table I, the demographic data on these patients in terms of age and sex do not differ significantly between the two centers. H o w ever, the Louisville patients were somewhat younger than the New York patients. In N e w York, there was a female preponderance of all patch-tested patients as well as the formaldehyde resin-positive patients. In Louisville, however, more women were tested, but more men had positive reactions. Although the percentage of blacks and Hispanies patch-tested at each center was consistent with the local population mix, all patients with textile-resin allergy were white. The clinical features for each of the 17 patients aUergic to formaldehyde resins are summarized in Table II. Of the 17 patients with ACD to formaldehyde resins, five (three garment workers, one printer, one chemical worker) had occupationally related disease. These patients were younger than the nonoccupational patients, with an average age of 40 versus 49 years. They also tended to have more hand involvement than the other patients. Three of the four patients older than 55 years of age had generalized erythroderma, which suggests a more severe involvement in older patients. All patients had dermatitis on areas normally covered with clothing. Six of the 17 had accentuation in areas of tight clothing such as waistband and flexurai areas.
964
Journal of the American Academy of Dermatology
Fowler et aL
Table II. Clinical data on 17 patients allergic to textile resins Patient No.
, Age (yr!,
[ Sex
Work-related
1
66
M
No
2
27
M
No
3
45
M
No
4 5
40 46
F M
6 7 8
50 43 41
M M F
No Yes (Chemical worker) No No No
9
40
F
Yes (garment worker)
10
40
M
No
11 12
75 40
F F
13
38
M
14 15 16
59 43 36
F F F
17
59
F
No Yes (garment worker) Yes (printing worker) No No Yes (garment worker) No
t
Other allergens (nonformaldehyde)
Distribution
Face, V of neck, arms, and hands Axillae, antecubital fossae, waist, legs Trunk, chest, antecubital fossae Diffuse Neck, chest, waist, abdomen, axillae Trunk, arms, hands Trunk, hands Anterior and posterior axillary folds Upper and mid back, waistband, periaxillary areas, dorsal hands, buttocks, and thighs Waistband, posterolaterai thighs and buttocks Generalized erythroderma Generalized, hams and arms worst Covered areas only, except hands Generalized erythroderma Nummular, scattered patches Face, neck, arms Generalized
Fragrances None None None None None Nickel Nickel and thimerosal Colophony, cinnamic alcohol, cinnamic aldehyde, toluene, sulfonamide resin Disperse brown # I None Rubber (carba, thiuram), oak moss
Nofle
None None Nickel, cobalt Kathon CG, rubber (thiuram)
No~: Allpatientswerewhite. The head and neck were rarely involved, except in cases of generalized erythroderma. Nine patients (53%) were patch-test negative to nonformaldehyde allergens. Of the eight with other positive reactions, three (18%) were allergic to nickel, three (18%) to fragrances, and two (12%) to rubber allergens. Several other positive reactions are noted in Table II. Table III lists the full range of positive reactions to formaldehyde-related allergens. Five patients (30%) were not allergic to formaldehyde itself. A mixture of ethylene urea and melamine formaldehyde resins was the best screening agent, with a positive rate of 82.3% (14 of 17 patients). Two of the three patients with negative patch tests had occupationally related ACD, which may reflect a specific work-exposure pattern. Many patients reacted to one or more commonly used formaldehyde-releasing preservatives, as shown in Table III.
DISCUSSION The incidence of formaldehyde-related textile A C D is unknown. However, since 1968 the incidence of formaldehyde sensitivity has been reported to be between 3% and 8%. 2 Recent data f r o m the N A C D G indicated that 132 of 1647 patients tested had positive patch tests to formaldehyde. O f these, at least 83 (51%) were deemed relevant to the presenting problem. In 1982, Andersen and H a m a n n 3 estimated that 8.6% of patients allergic to formaldehyde were sensitive to textiles. Similar studies between 1959 and 1964 placed the number between 8.6% and 65%. 2 Textile-related A C D has been reported to be more common in women than in men; the reported rates vary from 3:14 to as high as 5:1 5 Our data show a more even distribution with nine women and eight men affected. Although sensitivity can occur at any age, there appears to be a peak incidence between the ages of 35 and 60 years. 2 This
Volume 27 Number 6, Part 1 December 1992
ACD from formaldehyde resins 965
T a b l e III. Reactions to formaldehyde-related allergens Patient No. Allergens
Formaldehyde (1% aq) Dimethylol dihydroxyethylene urea formaldehyde resin (Fixapret r (4.5% aq) Dimethylol propylene urea formaldehyde resin (Fixapret PH) (5.0% aq) Tetramethylol acetylenediurea formaldehyde resin (Fixapret 140)
'
I I' I I I'J
1-;
Too,, ,
3+
-
1+ 2+
1+ 1+ 1+ 1+ 1+
1+
1+
-
2+
1+
2+
1+ 2+
1+
-
1+ 1+ 1+
2+ 1+ 2+
1+
-
1+
2+
?
2+
1+
-
1+ 2+ 2+ 1+ 2+ 2+ t +
2+
-
2+
-
1+ 1+ 2+
2+
1+ 2+ 2+
-
1+ 1+
-
1+ 12(70.6)
1+ 1+
?
5(29.4)
1+
1+
-
1+ I+
-
1+
1+ Ii(64.7)
1+
8(47.1)
(5.0%) Ethylene urea melamine formaldehyde resin (Fixapret AC) (10% pet) Urea formaldehyde resin (Kaurit S) (10% pet) Melamine formaldehyde resin (Kaurit M70) (7%) Ethylene urea formaldehyde resin (1.0% pet) Phenol formaldehyde resin (1% pet) Quaternium- 15
NTNTNTNT
-
=
2+ 1+ 1+
-
1+ 14(82.3)
3+ 1+
1+ 1+
-
1+ 10(58.8)
1+
2+ 1+ 2+
-
1+ 11 (64.7)
I+ 2+
NT NT N T 2+
I+
1+ 2+ N T N T N T N T N T N T 2+
3+
1+ 1+
-
1+ 3+
-
1+
1+
-
1+ 1+ 1+
-
(4) -
1 (5.9)
2+ 11(64,7)
(2% pet) Glutaraldehyde (1% aq)
-
Diazolidinylurea (ITo pet or 1% aq) Imidazolidinylurea
I+
(2%) D M D M hydantoin (1% aq) Bioban CS 1135 (1% pet) Bioban CS 1246 (1% pet)
1 (5.9)
1+ 1+
-
1+ 1+
1+
1+
2+
6(35.3)
1+
2 (11.8)
1+
1+ NT NT NT NTNT
-
2 (11.8)
1+ N T N T N T N T N T N T NT N T N T N T NT
N T N T N T N T NT I+ N T N T N T N T N T N T NT N T N T N T N T
Grading systemof NACDG used. aq, Aqueous;NT, not tested;pet, petrolatum. was confirmed in our patient population; only 3 of 17 fell outside this range. In its early stage, textile resin-related ACD appears as a pruritic, paputovesieular, erythematous
eruption that m a y subsequently become chronic and lichenified if not recognized and treated}, 6 T h e dermatitis m a y be most prominent at follicular orifices. 7 Although most reports note a characteristic
Journal of the American Academy of Dermatology
Fowler et al.
966
Table
IV. Cross-linking fabric finishing agents (durable-press)
Resin type
Polycondensation polymers (1930s) Urea formaldehyde (dimethylol urea)
Melamine
Linear reactants Dimethylol carbamates (1960s)
Cyclic ureas (1960s) N-methylol propylene urea derivatives: (dimethylolpropylene urea) (dimethylolethylene urea) Dimethylol dihydroxyethylene urea (1960s)
DMDHEU blended or reacted with glycols Dirnethoxymethyl dihydroxyethylene urea (1980s) (methylated DMDHEU)
Abbreviation
DMU
MF
DMC
Eslimated total market share (1990)
6%
20%
3%
3% DMPU DMEU DMDHEU
16%
34%
DMDHEU
10%
,. Uses and characteristics
Cotton pile fabrics, velvet, corduroy coatings, old-fashioned women's dress goods, flannelettes, upholstery, viscous rayon, used with some water repellents Imparts crease and shrink resistance to cotton and rayon, not chlorine resistant, increases wash fastness, but washes out faster than others, inexpensive "Heavy fabrics," "stiff" finishes, draperies, coatings, some collars, some water and oil repellent and flame retardant finishes; effective on synthetic and natural fibers Used with many other finishes to improve durability to washing and dry cleaning Less than 2% of clothing; some on sheeting, pillow cases, cotton-polyester blends, fabricliners and draperies, white shirts Soft finish, excellent chlorine resistance, not very light fast and incompatible with some dyes, more expensive Colored cotton textiles and white goods, shirts, some upholstery, more popular in France, Italy, and Spain; little or no use in U.S. Soft finish, excellent chlorine resistance, used in paper and red wine DMDHEU and its derivatives are used on 80%-90% of U.S. apparel fabric blends; "wash and wear", colored and white fabrics and shirtings; not used on 100% pure synthetic fabrics; also used on draperies and sheeting; chlorine resistant, stiffer than earbamates, some paper use, relatively inexpensive Performance similar to DMDHEU but slightly slower curing As etherified glyoxal reactant; similar uses and characteristics as DMDHEU with lower formaldehyde release, more expensive than DMDHEU
Relative formaldehyde release
High
High
Medium
Medium
Low
Very low Low
ThistablewasadaptedwithpermissionfromFisherA.A,ed. Contactdermatitis.3rd cd.Philadelphia:Lea & Febiger,1986:288.We alsoacknowledge the generousassistanceof membersof the textileindustry,especiallyJ. C. Winchester,AmericanCynamidCo.,Manager, CharlotteLaboratory, Charlotte, N.C.;M. D. Hurwitz,Professor,Clothing& Textiles,Universityof North Carolina,Greensboro,N.C.; P. S. Pal, ScientificApplications & InnovationsCorp.(textileindustryconsultant),Charlotte,N.C.; and R. R. Stewart,Proctor & GambleCo. (textilescientist),Cincinnati,Ohio.
Volume 27 N u m b e r 6, Part 1 December 1992
ACD from formaldehyde resins 967
Table IV. Cont'd. Estimated total market share Resin type
Abbreviation
(~99o)
Uses and characteristics
Soft, white fabrics, used on collar linings in past durable finishes, some drapery, and upholstery use Not as chlorine resistant as others; little use in U.S. but may be added to MF or mixed with other finishhag agents Children's apparel at present ha U.S. May be used in Japan on polyester/cotton blends or on Marks & Spencer fabrics ha England, expensive; not as effectiveas DMDHEU or its derivativesfor shrinkage control or durable press appearance
Uron formaldehyde (1950s) N,N'-Bis (methoxymethyl) uron
UF
1%
N,N'-dimethyl 4,5 dihydroxyethylene
DHDMI or DMcDHEU
1%
distribution (e.g., the axillary folds in women and the neck in men), our patients tended to have a more diffuse distribution (see Table II). Covered areas tended to be most frequently involved, with the face, legs, and feet often spared. A pattern clinically suggestive of n u m m u l a r eczema was seen in several patients. In addition to formaldehyde resins, t2 of our patients (70%) were allergic to formaldehydereleasing preservatives. Of these, most reacted to quaternium-15. Clinically, these reactions may be significant because these preservatives are commonly used in skin care products and topical corticosteroid creams. This secondary ACD may lead to diagnostic confusion if complete patch testing is not undertaken. In 1934, Marcussen 1 reported the first case of A C D caused by formaldehyde in textiles. However, it was not until 1965 that the first authenticated U.S. cases were reported by O'Quinn and Kennedy. 6 In recent years, there has been some debate as to whether the A C D associated with permanent press clothing is actually caused by formaldehyde, the formaldehyde resins, or both. Storrs, 7 Hatch and Maibach, 2 and Malten 8 have reported cases in which patients were sensitive to formaldehyde or formaldehyde resins, but not necessarily to both, whereas Andersen and Hamann 3 reported, "A negative patch test to formaldehyde virtually excluded allergic contact dermatitis from textile finishes."
Relative formaldehyde release
High
None
(p. 66) Our data clearly support the idea that a patient can be allergic to the formaldehyde resins without being allergic to formaldehyde itself as 5 of the 17 (30%) were allergic to formaldehyde resins but not to formaldehyde. In our experience, ethylene urea-melamine formaldehyde proved to be the best screening allergen and clearly identified 14 of 17 patients. There are two plausible theories as to how formaldehyde sensitivity can result in sensitivity to permanent press clothing. The quality of wrinkle resistance results from the formation of covalent cross finks between adjacent cellulose molecules that form when cellulose and formaldehyde react with a resin. Because this is a heat-driven reaction that occurs in the presence of an acid catalyst (e.g., zinc nitrate, magnesium chloride, or alkanolamine hydrochloride),2, 7 formaldehyde release could occur by one of two means: (1) when there is incomplete cross-linking of some resin, which allows degradation of the resin by chlorine bleach with release of formaldehyde; and (2) ff acid catalyst remains in the fabric, such that formaldehyde may be released by hydrolysis under warm moist conditions.7 Each of the formaldehyde resins used in durable press fabrics can be categorized according to their relative formaldehyde release (Table IV). Those being classified as "high" formaldehyde releasers include dimethylol propylene urea, urea formaldehyde (dimethylol
Journal of the American Academy of Dermatology
968 Fowler et al.
urea), melamine formaldehyde, and uron formaldehyde. "Medium" releasers include dimethylol ethylene urea and dimethylol carbarnates. 7 In contrast, dimethylol dihydroxyethyleneurea and dimethoxymethyl dihydroxyethyleneurea, which represent the major resins used in clothing fabrics manufacturers in the United States, release low levels of formaldehyde. Furthermore, dirnethyl 4,5 dihydroxyethylene urea apparently does not release formaldehyde. This finish is used primarily on children's clothing in the United States, in part because of its expense. 7 In 1974, Schorr et al.9 studied 112 samples of American clothing and discovered that 18 of these (16%) had a free formaldehyde content of more than 750 ppm. Fisher et al. I~ suggested that dermatitis may occur when formaldehyde content is 750 ppm, whereas Berrens et al.ll believe that the critical level may be closer to 500 ppm. In 1974 the textile industry claimed that levels of formaldehyde release had decreased so that 100 to 200 ppm was standard. 12 However, Hatch and Maibach2 noted that the United States has neither a mandatory nor a vohntary formaldehyde level restriction. Our findings that patients were allergic to the resin but not to formaldehyde are consistent with the reports of lowerformaldehyderelease from U.S. textiles. However, during the past decade, there has been a progressive shift in this country to imported textiles, especially from the Far East, where little information can be obtained concerning the resins used in finishing fabrics, the curing process, or the number of afterwashes. In addition to formaldehyde resins, 12 of our pa-
tients (70%) were allergic to formaldehyde-releasing preservatives. Of these, most reacted to quaternium-15. Clinically, these reactions may be significant because these preservatives are commonly.used in skin care products and topical corticosteroid creams. This secondary ACD may lead to diagnostic confusion if complete patch testing is not undertaken. REFERENCES 1. Marcussen PV. Contact dermatitis due to formaldehyde in textiles, 1934-1958. Acts Derm Venereol (Stockh) 1959; 39:348-56. 2. Hatch KL, Maibach HI. Textile chemical finish dermatitis. Contact Dermatitis 1986;14:1-13. 3. Andersen KEo Hamatm K. Cost benefit of patch testing with textile finish resins. Contact Dermatitis 1982;8:64-7. 4. Wereide K. Formaldehyde as a contact allergen in textiles, Aeta Allergol 1964;19:351-63. 5. Cronin E. Formalin textile dermatitis. Br J Dermatol 1963;75:267-73. 6. O'Quinn SE, Kennedy CB. Contact dermatitis due to formaldehydein clothing textiles. J A M A 1965;194(6):58396. 7. Storrs FJ. Dermatitis from clothing and shoes. In: Fisher AA, ed. Contact dermatitis, 3rd ed. Philadelphia: Lea & Febiger, 1986:283-337. 8. Malten KE. Textile finish contact hypersensitivity. Arch Dermatol 1964;100:215-21. 9. Schorr WF, Keran E, Plotka E. Formaldehyde allergy: the quantitative analysis of American clothing for free formaldehyde and its relevance in clinical practice. Arch Dermatol 1974;I 10:73-6. 10. Fisher AA, Kanof NB, Biondi EM. Free formaldehyde in textile and paper. Arch Dermatol 1962;86:753-6. 11. Berrens L, Young E, Jansen LH. Free formaldehyde in textiles in relation to fo/'rnalin contact sensitivity. Br J Dermatol 1964;76:110-5. 12. Vail SL, Reinhardt RM. What do formaldehyde tests measure? Textile Chemist Colorist 1981;13:13-22.
BOUND VOLUMF_SAVAILABLE TO SUBSCRIBERS Bound volumesof the JOURNALOFTHEAMERICANACADEMYOFDERMATOLOGYare availableto subscribers(only) for the 1992 issues from the Publisherat a costof $66.00 for domestic, $88.62 for Canadian, and $84.00 for international for volume26 (January-June) and volume 27 (July-December).Shipping charges are included. Each bound volume contains a subject and author index and all advertising is re. moved. Copiesare shippedwithin 60 days after publication of the last issue in the volume.The binding is durable buckram with the journal name, volumenumber, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Ycar Book, Inc., SubscriptionServices, 11830 W~t|ine Industrial Dr., St. Louis, MO 63146-3318. USA: phone (800) 325-4177, ext. 4351; (314) 453-4351. Subscrlptiottr must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.
