Immediate and delayed allergy from epoxy resins based on diglycidyl ether of bisphenol A.

Immediate and delayed allergy from epoxy resins based on diglycidyl ether of bisphenol A. Scand J Work Environ Health 1991:17:208-215. This case report presents two patients with imme diate and delayed allergy to epoxy resins based on diglycidyl ether of bisphenol A (DGEBA). In patch testing, the epoxy resin (DGEBA-bascd) of the standard series gave allergic reactions. Both patients had a prick test reaction of histamine size or larger to the human serum albumin (HSA) conjugate of DGEBA based epoxyresins. One had been occupationally exposedto methyl tetrahydrophthalic anhydride (MTHPA) and had a histamine-size prick test reaction to the HSA conjugate of MTHPA; the other did not react to the conjugate. Determinations of specific immunoglobulin E were carried out with HSA-DGEBA con jugates, two DGEBA-based epoxy resins, and phthalic anhydrides. The first patient had positive tests to DGEBA, the DGEBA-based epoxy resins, and two phthalic anhydrides, and the second to DGEBA and the DGEBA-based epoxy resins, but not to the phthalic anhydrides.

Very few cases of immediate allergy from epoxy resins have been reported. In an older report a patch test with an epoxy resin induced generalized urticaria and an asthmatic reaction (12). Romaguera et al (13) described a case having hand eczema, rhinitis, and conjunctivitis caused by DGEBA-based epoxy resins. In 1983 Suhonen (14) reported two patients who had immediate urticarial patch test reaction s, probably from an impure DGEBA-based epoxy resin used in the patients' workplace. The epoxy resin in the patch test standard series did not give an immediate reaction, and the causative agent was not further clarified . 208 During the past 15 years we have had 1622 cases of occupational dermatoses (4,5). Only two cases of occupational contact urticaria were caused by chemicals of the epoxy resin system (3,5). Skin tests indicated that in both cases the causative agent was methyl hexahydrophthalic anhydride (MHHPA). Recently, we have reported additional cases of occupational asthma and rhiniti s caused by exposure to MHHPA (15).
In the present report, finding s of two patients who had occupational immediate allergy (asthma) and delayed allergy (allergic contact dermat itis) from exposure to DGEBA-based epoxy resins are described. Furthermore, the same skin testings and specific immunoglobulin (lg) E determinations that were performed for these two patients were also performed for five other patients who had occupationally been exposed to chemicals in the epoxy resin system and who had developed respiratory symptoms suspected to be related with th at work. However, no specific immediate allergy to epoxy resin chemicals was detected.
Patients and methods Patients. Two patients who had immediate and delayed allergy to DGEBA-based epoxy resins are presented . These two patients have earlier been described as patient s 32 and 34 in the report of Jolanki et al (5). Furthermore, data on investigation s used to detect immediate allergy to epoxy resins and phthalic anh ydrides in five patients with suspicion of work-related respiratory symptoms in a factory manu facturing insulators from epoxy resins are given (16) . Patients 5-7 in the present study correspond to the same patient numbers in a previous report (16).
Epikote 827 is the DGEBA-based epoxy resin from which our standard patch test epoxy resin preparation was made. Lekutherm X 287 is a mixture of DGEBA and diglycidyl aniline type epoxy resins, whereas Araldit CY 184 and Lekutherm X 100 are cycloaliphatic epoxy resins based on a monomer of molecular weight 284 [ie, diglycidyl ester of hexahydrophthalic acid (DGEHHPA)] (16).
For more-detailed information on the purity of the epoxy resins, the amounts of the allergens DGEBA, DGEHHPA, and nine reactive diluents (17) were determined with gas chromatography and high-performance liquid chromatography (18)(19)(20). The results are given in table I.
Preparation oj hapten conjugates of human serum albumin. Stock solutions of epoxy resin conjugates and anhydride conjugates of HSA were prepared as fol-lows. The conjugation between haptens (ie, epoxy resins and phthalic anhydrides) and HSA was carried out essentially as described by Howe et al in 1983 (21 Briefly, an aliquot of a I % solution of the epoxy resin in acetone was spread as a line on a silica thin-layer plate (DC-Alufolien Kieselgel60, F 254, Merck, Germany) 2 cm from the bottom. The plate was eluted with a mixture of chloroform and acetonitrile 90/10 [volume/volume (vI v )], visualized under ultraviolet light , and the DGEBA band was scraped with a spatula. The scraped materia! was dissolved in acetone spread on another thin-layer plate, eluted, visualized: scraped, and dissolved as already described. All the dissolved material (less than 400 mg) was used as the 209 hapten for the preparation of the HSA conjugates of DGEBA. A control sample was scraped from the plate above the DGEBA band.
Skin tests. Skin tests to 20 common environmental allergens, including epithelia, pollens, and dust mite (Allergologisk Laboratorium A/S, Denmark) were performed with the standard prick test technique of Aas & Belin (23). Histamine hydrochloride (10 mg/m!) was used as a positive control, and a test solution prepared as antigen solution without an antigen served as a negative control (23). The positivity of the prick test results was evaluated according to commonly accepted rules (23,24). The same technique and corresponding control solutions were used when the prick tests were performed with hapten-HSA conjugates.
The patch tests were done and scored as previously described (5,25) with an occlusion time of 48 h. The European standard series (Chemotechnique Diagnostics AB, Sweden), a series of plastics and glues (Chemotechnique Diagnostics AB and Trolab, Hermal-Chemie Kurt Herrmann, Germany) (26), the (meth)acrylate series (Chemotechnique Diagnostics AB), and a series of nine epoxy reactive diluents (17) were used for the patch testing. Epoxy resins and phthalic anhydrides were tested at 1 070 [weight/weight (w/w)] in petrolatum; other substances brought in by the patients ("own substances") were tested at 1-10 070 (w/w) in petrolatum.
Specific immunoglobulin E determinations to epoxy resinsand phthalic anhydrides. Hapten-HSA conjugate solutions and control solutions at a concentration of 2 % HSA were coupled to filter paper discs (Munktell OOH) activated with cyanogen bromide, 100 !-II of conjugate being used per disc, as described by Ceska & Lundkvist (27). The discs and Phadezym RAST reagents (Pharmacia Diagnostics, Sweden) were used in the specific IgE determinations and the inhibition experiments. The quantity of specific IgE bound to a disc is given as the ratio of absorbances at 420 nm of the hapten-HSA disc to the HSA-control disc. A ratio of three or more was defined as positive (table 2) provided that the absorbance of the hapten-HSA disc was half that of the reference (RAST reference D; 0.35 PRU/ml, Pharmacia Diagnostics).  Inhibition tests. Inhibition of th e specific IgE determinat ion s was carried out as described by Yma n et al (28). Briefly, a dilut ion series (I. I v/v, I: 10, I ; 100, etc) was prepared from the stoc k hapten-HS A conjugate solut ion with 0.1 M potassium phosph ate bu ffer , pH 7.0, containing sodium azide (3 mmolll

Results and case reports
The main characteri stics and skin test result s of the two pati en ts who had asthma are summa rized in table 2. Twent y unexpo sed controls and five oth er exposed patients (ie, workers from a n insulator facto ry) ( Patient 1. Patient 1 had been working as a cleaner since 1980 in a factory manufacturing ski poles and other sports articles made of DGEBA-based epoxy resins and polyester resins. The epoxy hardener was earlier a polyamine compound, and since 1985it has been MTHPA. The patient had no own or famil y history of atopy. In 1982 she experienced the first symptoms of dermatitis on the ba ck of her hands and lower ar ms . Since 1984 she had had dyspnea that was aggravated by exercise; since 1987 she had also had asthmatic symptom s at wor k. Concomitantly her dermatiti s had spread to her forehead and neck. She had had epo xy resin exposure via bo th respiratory and skin contact because her duty was to clean rooms containing much plastic dust. She worked without protective gloves until she started to develop skin symptoms. When examined in February 1988 her physical status was normal except for the wheezing rales from her lungs. The diurnal variation in peak expir atory flow was 8 0J0. After prop er asthma treatment the symptom s disappeared, a nd no rales were heard. She had a mild ob stru ction in her spiro metry {FE V 070 [(lOOx FE Vl.o)/ forced vita l capacity] 70-78], but the bronchodilata tion test was negati ve. In February of 1988 she had mode rate bronchial hyperreactivity which had disappeared in April after she had been away from work and used asthma medi cation. She had no strainindu ced asthma, and her chest radiog raph was normal . The bron chial pro vocation tests with the placebo, poly- Table 3. Characteristics and test resul ts of fi ve persons exposed to epoxy resin and phthalic anhydride in an insulator factory and suffering from resp iratory symptoms. (DGEBA =diglycidyl ethe r of bisphenol A, Ig =immunoglobulin, MHHPA =methyl hexahyd rophthalic anhydride, HHPA =hexahydrophthalic anhydride , MTHPA =methyl tetrahydrophthal ic anhyd ride) The diagnosis of probable occupational asthma was based on history, the sensitization to DGEBA-based epoxy resins and MTHPA verified by prick tests and specific IgE determinations and the disappearance of hyperreactivity after an unexposed period. The reaction with short MTHPA challenge was not significant. When the patient was seen after six months without exposure, she used asthma medication, but had fewer symptoms, and her spirometry was normal. Patient 2. Patient 2 worked as a connector in a factory manufacturing electronic devices. She was admitted to examination because of asthmatic symptoms. She had worked in the electrical industry in [1962][1963][1963][1964][1965][1966], and since 1969 in her present job. She had had mild atopic dermatitis but had not suffered from any respiratory symptoms in her youth. During 1962-1963 she developed severe dermatitis on her face, hands, and lower arms when she was exposed to epoxy resins while working as an insulator. She was asked to change her job, but no patch tests were performed. It seems evident that at that time she had been sensitized to epoxy resins , to which she was commonly exposed in that factory . Since then she has experienced pruritic hand and face dermatitis after being exposed to epoxy glues. In her present work she was handling electrical components containing epoxy glue which was still partly uncured. The epoxy glue contained DGEBA-based epoxy resins and aliphatic polyamines as hardeners. She had also probably been exposed to airborne epoxy resin because plenty of uncured epoxy resin was handled close to her worksite. Since the late 1970s she had been on sick leave several times because of her severe hand and face dermatitis. In 1986 she began to develop asthmatic symptoms , which worsened during the work shifts. She was also exposed to cyanoacrylate and colophony in her work.
The prick test with Epikote 827-HSA was positive, but with PA-HSA it was negative (table 2). The specific IgE determinations to DGEBA and the two epoxy resins containing DGEBA were strongly positive, but to phthalic anhydrides the specific IgE determinations were negative . In the standard patch test series cobalt chloride, 10/0 in petrolatum, gave a 2+ reaction, and the standard epoxy resin gave a strong vesicular ( (25). The series of plastics and glues, including cyanoacrylate, was negative.
The diurnal variation in peak expiratory flow was 7 % (365-340 l/min). The spirometric values were normal. The patient had mild bronchial hyperreactivity. The bronchial provocation tests with the placebo were negative, as well as the challenge with colophony and a cyanoacrylate glue. Challenge tests with epoxy compounds or surveillance of peak expiratory flow at the workplace was not carried out because of the severe allergic contact dermatitis from exposure to DGEBA-based epoxy resins. After six months without exposure, the asthmatic symptoms had decreased, and the patient had no nonspecific bronchial hyperreactivity in the histamine challenge test.
The diagnosis of probable occupational asthma was based on the history, the sensitization to DGEBAbased epoxy resins verified by the prick tests and specific IgE determinations, and the disappearance of hyperreactivity after a period of nonexposure.
Inhibition tests. Experiments to inhibit specific IgE determinations were carried out with serum from both patients to prove the specificity of the determinations for the DGEBA-based epoxy resin Epikote-827. The results showed that the specific IgE determinations could be inhibited nearly completely with HSA conjugate of Epikote 827 (figure I). When the MTHPA-HSA conjugate was used as the inhibitor, no specific inhibition could be demonstrated.

Patients from an insulator factory. Eighty patients
were working in a department manufacturing insulators from the epoxy resins Araldit CY 184, Lekutherm X 100, and Lekutherm X 287 using phthalic anhydrides (PA, HHPA, MHHPA) as hardeners. Twenty-one of the 80 workers had skin and/or respiratory symptoms, 14 having only skin symptoms, three only respiratory symptoms, and four both types of symptoms. Twelve workers (of 18 with skin symptoms) were found to have allergic contact dermatitis from cycloaliphatic epoxy resins. The patch test results have been published elsewhere (16).  Five patients who complained of respiratory symptom s which could have been due to the epoxy resins or their hardeners were studied with the same prick tests and specific IgE determinations as the two patients that ha ve already been presented (ie, DGEBA, four other epoxy resins, and three phthalic anhydrides). Four of these patients had skin symptoms and were patch tested . Three of them (patien ts 5-7) had allergic contact dermatitis from the epoxy resins used in the factory.
The characteristics of the patients and test results are given in tabl e 3. The prick tests and specific 19E determinations to the epoxy resins and phth alic anhydrides were negative in all cases and therefore indicated that none of the patients had a specific immediate allergy to the chemicals.

Discussion
Low-molecular weight chemicals act as hapten s in type IV (delayed) allergic reaction s causing allergic contact dermatitis. Proteins, on the other hand, have been regarded as the inducers of IgE production and type I (immediate) allergic reactions. More recent clinical obser vations indicate that chemicals with a molecular weight usually below 1000 may also induce immediate allergic symptoms (31,32). Thi s phenomenon has largely been neglected in the dermatologic literature because type I allergic reactions often affect the respiratory tract, and in many countries dermatologists do not regularly perform type I allergy investigations (23). With the increase in awareness of contact urticaria (33), dermatologists are now paying more attention to type I allergy. Immediate reactions should be remembered as it has become evident that type I and type IV allergic reactions to the same compounds coexist more often than was previously believed (24). The present report indicates that DGEBA-based epoxy resins can be added to the list of substances able to cause both of these types of allergies.
Hapten-HSA conjugates for the pr ick test and specific 19E determin ations have to be adapted to each individual hapten (3 I). The allergen formed in vivo is not known, and therefore the procedure for preparing hapten-protein conjugates is currently largely on the " trial and error" level. Several variables (such as substance concentration and pH) may influence the preparation of conjugates (3I). By using the method of Howe et al (21), we were able to get reproducible hapten-HSA conjugates with DGEBA-based epoxy resins and phthalic anhydrides. Pr ick tests, specific IgE determinations , and the appropriate controls indicated that our patients had Igfi-mediated allergy from the chemicals in the epoxy resin system. Earlier, specific IgE antibodies have been demonstrated for , for example, the following low-molecular weight chemicals: isocyanates, reactive textile dyes, acid anh ydrides, platinum group of metal salts, chloramine-T, and ethylene oxide (31).
According to the history, exposure, and sensitization to DGEBA-based epoxy resins (which was verified with prick tests and specific 19E determinations) and because the bronch ial hyperreacti vity disappeared after the unexpo sed period, it was feasible to conclude that both patients had occupational asthma due to exposure to DGEBA-based epoxy resins . Unfortunate-Iy, we were unable to perform bronchial pro vocation tests with DGEBA-based epoxy resins for ethical reasons. Once earlier, an epoxy resin bronchial pro vocation test was performed in the exposure chamber with an asthmatic patient who had allergic contact dermatitis from exposure to DGEBA-based epoxy resins, and the patient developed extensive erythrodermic dermatitis. This reaction was probably due to airborne contact with the resin but may also have reflected systemic contact dermatitis (34). From the point of insurance 5 213 medicine the diagnostic tests were considered sufficient, and both patients were compensated for both occupational asthma and occupational allergic contact dermatitis. Neither of the patients could continue their previous work and had to retire. The first patient experienced dyspnea in 1984 before MTHPA was taken into use as a hardener. Therefore the patient had probably been sensitized first to DGEBA-based epoxy resins and then to MTHPA. Bronchial asthma (12), rhinitis (13), and urticaria (12,14) from exposure to epoxy resins have earlier been reported. To the best of our knowledge, tests to demonstrate that these symptoms could be IgE -mediated have not been performed earlier. The type I allergy of our patients appeared as respiratory symptoms and pruritus. Both patients complained of immediate pruritus from epoxy resin. It is evident that this occurrence was a sign of contact urticaria although hives were not present. IgE-mediated pruritus without hives has been reported to result from the use of latex gloves (35). It is possible that sensitization had occurred through skin contact. Our observations of the patient group from an insulator factory indicate that type I allergy from epoxy resins is possibly rare although the patients have strong exposure to epoxy resins. It remains to be seen whether more cases with type I allergy from epox y resins of the DGEBA or non-DGEBA type (eg, cycloaliphatic epoxy resins) will be detected when specific IgE determinations and prick tests for epoxy resins are used to investigate patients who are exposed to epoxy resins and have respiratory or (contact) urticaria symptoms. Using HSA conjugate of cycloaliphatic epoxy resins, we have had no positive prick tests or specific IgE determinations so far, and therefore we cannot be certain whether HSA conjugates of cycloaliphatic epoxy -resins are usable in the verification of immediate allergy to cycloaliphatic epoxy resins. No cross-reactivity was observed between the specific IgE determinations for DGEBA-based epoxy resins and for cycloaliphatic epoxy resins, as observed between phthalic anhydrides. Positive specific IgE determinations to DGEBA (ie, purified DGEBAbased epoxy resin of molecular weight 340) indicated that the DGEBA was the specific allergen in the two asthma patients. Absorbance at 420 nm of the DGEBA-HSA disc was less than that of the DGEBAbased epoxy resin (Epikote 827)-HSA disc, probably because a smaller amount of hapten was used for the HSA conjugates of DGBBA than for the HSA conjugates of the DOBBA-based epoxy resin. Specific IgE determinations to Lekutherm X 287 were positive due to the large amount of DGEBA (table 1).
The patch tests and clinical data confirmed that both patients had allergic contact dermatitis from exposure to DGEBA-based epoxy resins. Patient 2 also had positive patch test reactions to epoxy acrylates (BIS-GMA, BIS-EMA, and BIS-GA). She was the only one of six patients (with allergic contact dermatitis due to DOBBA-based epoxy resins and tested with BIS-OMA) 214 who had concomitant allergic reactions to DGEBAbased epoxy resins and BIS-GMA (5). This finding may indicate cross-reactivity between the test substances. Other possibilities for the concomitant positive reactions with the standard epoxy resin and the epoxy aerylates in patch testing reactions ha ve recently been discussed elsewhere (25,36).
Patient 2 was also positive to cobalt chloride in the patch test. In the 1960s she had been exposed to polyester resins in which cobalt was used as an accelerator. Thus she had probably become occupationally sensitized to cobalt. Earlier we also diagnosed occupational contact allergy to cobalt in a patient from the same workplace. Recently we had a patient who had become sensitized to cobalt from exposure to polyester laminating resin in boat building (unpublished) .