Registration Dossier

Toxicological information

Sensitisation data (human)

Currently viewing:

Administrative data

sensitisation data (humans)
Type of information:
other: Chemical Assessment Report
Adequacy of study:
key study
Study period:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
other: publicly available data
Reason / purpose:
reference to same study

Data source

Reference Type:
other: Priority Existing Chemical Assessment Report
Report Date:

Materials and methods

Type of sensitisation studied:
Test guideline
no guideline required
GLP compliance:
not specified

Test material

Details on test material:


Clinical history:

Results and discussion

Applicant's summary and conclusion

This section includes the known health effects of ammonium, potassium and sodium persulfate on humans. While systematic studies of the health effects of these compounds are generally lacking, a substantial body of evidence is available from case reports and epidemiological studies relating to dermal and respiratory pathologies and are discussed below.
Executive summary:

Human Health Effects

This section includes the known health effects of ammonium, potassium and sodium persulfate on humans. While systematic studies of the health effects of these compounds are generally lacking, a substantial body of evidence is available from case reports and epidemiological studies relating to dermal and respiratory pathologies and are discussed below.

Irritant contact dermatitis

Standard 24/48-h patch tests with 2.5% or 5% persulfate are used to differentiate between irritant and allergic contact dermatitis. These tests have shown that 5% ammonium persulfate is irritant (Calnan & Shuster, 1963; Cronin, 1980) although a separate study found 1/20 people exhibited an equivocal response when tested with 5% to 10% persulfate (Forck, 1968).

Two cases of irritant reactions to hair bleach containing ammonium persulfate have been described (Cronin, 1980; Fisher & Dooms-Goossens, 1976) involving erythema of the scalp and forehead developing over several hours followed by weeping and crusting.

Skin rashes appearing within one month of beginning work in a factory manufacturing ammonium and potassium persulfates occurred in 20% to 70% of new employees (White et al., 1982). The early onset of the rashes was interpreted as evidence of an irritant response. Workers engaged in the manufacture of the potassium salt were the most likely to exhibit symptoms.

Allergic contact dermatitis and respiratory effects

Immediate inflammatory reactions are normally mediated by specific IgE and/or IgG4 binding to mast cells or basophils (Garssen et al., 1996). After first contact with the specific antigen, antibody producing cells will produce specific IgE and/or IgG4. Subsequent contact with the allergen will result in release of inflammatory mediators. No specific IgE antibodies have been found in subjects exhibiting immediate skin or pulmonary reactions to persulfates. The reactions, therefore, have been ascribed to a non-allergic mechanism such as chemical irritation (Mensing et al., 1994 cited in Yawalkar et al., 1999) or pseudoallergic mechanisms through the release of histamine or leukotriene B4 by direct action on mast cells or neutrophils, respectively (Calnan & Shuster, 1963; Koller et al., 1996; Schwaiblmair et al., 1990). However, it was found that the radio-allergosorbent (RAST) test used to detect specific antibodies can give negative results in individuals positive in skin prick tests, in this case individuals who exhibited respiratory hypersensitivity to acid anhydrides (Drexler et al., 1993 cited in Kimber, 1996). The possibility that all three mechanisms contribute to immediate hypersensitivity has been raised.

Delayed contact hypersensitivity and late onset asthma appear to be T-cell-mediated responses but specific IgE antibodies may be important for triggering late-phase reactions and the chronic bronchial inflammation associated with asthma (Kimber, 1996).

Experimental Studies

Two unpublished studies of the sensitisation potential of persulfates in humans have been described. In the first study, two males and 44 females were treated with a lightener/developer mixture of ammonium, potassium and sodium persulfate at a concentration of 17.5% under occlusive patch for four hours. Patches were applied three times per week for three weeks. The first two applications were for four hours and the remaining seven for one hour as a result of irritant responses from the vehicle. For the last seven applications, skin sites were rotated on the same forearm. Following a 2-week recovery period, two challenge applications consisting of 0.2 mL of 2% of each persulfate under occlusive dressing were applied 48 h apart. The dressings were applied for 24 h to the right inner forearm and the sites examined at one and 48 h. The results were negative (CIR, 1998).

In a second study, subjects were treated with 0.001, 0.01 and 0.5% sodium persulfate (26 subjects/concentration) under occlusive patch four times per week for three weeks. After a 1-week recovery period, challenge was performed at each of the induction concentrations. No positive responses were observed when the two lower concentrations were used for challenge but challenge with 0.5% sodium persulfate resulted in 5/26 (19%) grade 4 skin reactions including: redness, induration, swelling, papules and vesicles. Following rechallenge with either 0.001% or 0.25% sodium persulfate for 24 or 48 h, one subject had reactions at the 0.001% site and two subjects had grade 4 reactions at the 0.25% site. The grade 4 reactions at the naive sites were considered by the authors of the study to be indicative of sensitisation to sodium persulfate (E. I. Dupont de Nemours and Company, 1992).

Case Reports

A number of case studies of women exposed to persulfate-containing hair bleaches either self-administered or administered by a hairdresser were summarised in CIR (1998) as follows:

A 54-year-old woman developed itchiness of the face and became red and swollen on the upper part of her body within 30 minutes of having a hair bleach applied. She went into shock and had generalised erythema and urticaria. Patch tests with 2% ammonium persulfate were negative. However, direct application of a supersaturated solution of ammonium persulfate caused a 1+ response, and when the solution was rubbed in the skin a 4+ response. A freshly prepared hair bleach caused a response of 1+ when applied to the skin and a 2+ urticarial wheal when rubbed into the skin (Brubaker, 1972).

The face of a 49-year-old woman became red and oedematous immediately following exposure to a hair bleach containing a persulfate-peroxide mixture. This condition lasted for several hours. Generalised Urticaria persisted for 24 h. Patch tests with 2% and 5% aq. ammonium persulfate were negative, but tests with 5% aq. ammonium sulfate were positive (Fisher & Dooms-Goossens, 1976).

A 46-year-old woman developed redness and slight crustiness on the anterior portion of the scalp and forehead one day after treatment with a hair bleach containing ammonium persulfate. Erythema and crusting were apparent on day 3 Patch tests with 2% and 5% ammonium persulfate were negative. The authors believed that the reaction was due to "... excessive concentrations of ammonium persulfate producing a strongly irritating alkaline effect." (Fisher & Dooms-Goossens, 1976)

The face of a 49-year-old woman became red and oedematous, her eyelids could not be opened, and generalised urticaria developed immediately upon her first-time application of a persulfate-peroxide hair bleach. Oedema lasted for several hours and generalised urticaria persisted for 24 h. An open patch test with 2% aq. ammonium persulfate applied to the forearm produced a large urticarial wheal within 7 min for the woman but not in three controls. The author believed this was a severe histamine reaction since it was a first time exposure and that ammonium persulfate is not primarily urticariogenic because the controls did not have a reaction (Fisher, 1977).

A 45-year-old woman stated that on several occasions immediately upon application of ammonium persulfate hair bleach, a burning sensation and diffuse erythema developed on the forehead, back of the neck and upper back, followed by a mild crusted dermatitis of the scalp and back of the neck the next day. The use of prednisone and the antihistamine Chlor-Trimeton, prior to bleaching resulted in minimal symptoms (Fisher, 1977).

A 72-year old woman developed erythema and oedema of the face one hour following exposure to hair bleach containing 5% aqueous ammonium persulfate. The following day, her cheeks and forehead were sharply demarcated and she had marked oedematous urticaria on her face and forehead. When she was tested with 5% ammonium persulfate an immediate wheal was produced. However, a 48 h patch test with 2% aq. ammonium persulfate was negative (Fisher, 1985a).

A 70-year-old woman developed pruritic oedema on her cheeks and forehead three hours after the application of a hair bleach containing ammonium persulfate (Fisher, 1985a).

A 69-year-old woman experienced facial flushing following exposure to a hair bleaching formulation containing 2% ammonium persulfate. She reported a stinging and burning sensation of the scalp and her forehead and face were erythematous with no itching. This condition persisted for 48 h. Patch tests with 2% ammonium persulfate were negative (Fisher, 1993).

Case studies of people exhibiting reactions to persulfates following occupational exposure were also summarised by CIR et al. (1998) as follows:

A 29-year-old woman acquired rhinitis and asthma while working in a beauty salon. A scratch test performed using 1% aq. ammonium persulfate immediately produced a wheal, followed by a mild asthma attack (Fisher & Dooms-Goossens, 1976).

A 21-year-old hairdresser had a delayed asthmatic reaction to hair bleach containing persulfates. This type of reaction was reproduced by exposure to the bleach and was blocked by inhalation of beclomethasone dipropionate but not by sodium cromoglycate. Patch tests with potassium persulfate and the bleach were negative. The investigators noted that at the time of these tests, the subject had changed Jobs and was no longer being exposed to the bleach (Pepys et al., 1976).

A 21-year-old hairdresser suffered from rhinitis and wheezing dyspnoea during 5.5 years of employment when she was exposed to hair bleaches and hair dyes containing bleaches. She had elevated total IgE in allergy tests and a provocation test with 10 mg/mL histamine was positive. Exposure tests with a hair bleaching product and 1% ammonium persulfate caused wheezing and dyspnoea three to four hours following exposure. These responses were partially inhibited when disodium cromoglycate was inhaled 15 min prior to exposure and completely inhibited when betamethasone was administered. The investigators concluded that the patient suffered from late onset bronchial asthma due to sensitivity to ammonium persulfate (Gamboa et al., 1989).

A hairdresser developed rhinoconjunctivitis and bronchial asthma associated with hair bleach containing persulfate after 2 years. A prick test was positive for the persulfate (Pankow et al., 1989).

A 21-year-old hairdresser developed rhinitis from exposure to commercial bleaches, had urticarial reactions when she applied the bleach to her own hair, and eventually developed conjunctivitis and oedema of the eyelids. Patch tests were positive for potassium and sodium persulfate, and inhalation tests with the hair bleach produced an immediate asthmatic reaction within 1 min (Pepys et al, 1976).

A 23-year-old hairdresser developed acute pruritus and rashes on her hands and forearms after using hair bleach containing ammonium persulfate. An open test with 5% aq. ammonium persulfate caused slight reddening and pruritus after 20 min. A scratch test with 1% aq. ammonium persulfate caused erythema and wheal formation after 5 min. A closed patch test with 2% aq. ammonium persulfate was positive at 72 h (Widstrom, 1977).

A hairdresser who developed cutaneous and respiratory symptoms after 1 year of employment was tested in clinical and immunological studies. Skin prick tests with 1:5 w/v potassium and sodium persulfate were positive but were negative with 10 control subjects. The hairdresser had no reaction to a 2% concentration of either of the persulfates in an open patch test. Hyper-reactivity was observed in a methacholine-inhalation test. A bronchial provocation test with 1:50 w/v potassium persulfate elicited a delayed asthmatic response, which was followed by a recurrent nocturnal fall in airflow that was resolved after 3 days. Plethysmography indicated air trapping due to increased airway resistance. Histamine release tests were not conclusive and determinations of specific immunoglobins against persulfate salts were negative (Parra et al., 1992).

Two industrial workers developed dermatitis, rhinitis, bronchitis, and asthma following occupational exposure to the dust of persulfate salts. Patch tests induced late cutaneous reactions and occupational exposure at the workplace for eight hours induced a pathological increase in airway resistance (Baur & Fruhmann, 1979).

Other case studies reported in the literature are as follows:

A 24-year-old female hairdresser experienced episodes of dry cough, usually beginning one to two hours following exposure to bleaching powder and aggravated during the night, together with eczematous lesions on both sides of her hands and forearms. Skin prick and patch tests with 1% and 5% ammonium persulfate were negative at 15 minutes but a papule reproducibly developed at the prick site after 24 h in a dose-dependent manner. A lymphocyte transformation analysis using heparinised blood reproducibly demonstrated a significant dose-dependent proliferation of T cells to bleaching powder. It was suggested that the cutaneous and pulmonary responses had a common T-cell-mediated mechanism (Yawalkar et al., 1999).

A 35-year-old female beautician with a 4-year history of increasing difficulty with cough, chest tightness, congestion, and dyspnoea was negative to methacholine bronchoprovocation. Bronchoprovocation testing was performed in a simulated work environment with frosting agents and hair bleach (containing 30% persulfates) with the latter giving a positive response 30 minutes after exposure (Schwartz, 1989).

The U.K. Health and Safety Executive (HSE) have published an assessment of the evidence for persulfates as asthmagens (Health and Safety Executive, 1997). They have pointed out that well-conducted studies have been performed in which hairdressers with work-related asthmatic symptoms have undergone specific bronchial challenge tests, performed blinded. In a number of these studies, controls that were either non-asthmatic or asthmatic for reasons unrelated to work in the hairdressing industry were used. The studies include Blainey et al. (1986), Parra et al. (1992), Pankow et al. (1989) and Schwartz (1989) described elsewhere in this section. The HSE document also cites a further study (Agustin et al., 1992) in which 2 hairdressers developed work-related rhinitis and conjunctivitis (one also had work-related asthma) and were positive following specific bronchial challenge: one showing late asthmatic symptoms, the other showing immediate severe nasal Symptoms. The HSE document states that the well-conducted studies are supported by case reports where the challenges were not blinded and no controls were employed. These case reports include those of Pepys et al. (1976), Gamboa et al. (1989) and Schwaiblmair et al. (1990) described elsewhere in this section. The HSE document further cites two other studies; one with 5 cases of asthma (2 positive, 1 negative, 1 equivocal, 1 unknown at specific challenge) (Therond et al., 1989); another with 3 cases, 1 challenged and positive at specific challenge (Wallenstein et al., 1993). The HSE document stated that there was sufficient evidence to conclude that persulfate salts meet the revised EU criteria for Classification as respiratory sensitisers.

A number of other studies were cited in a recent review of the toxicology of ammonium persulfate (BG Chemie, 1996). These studies reported cases of skin conditions and asthma (Blandin, 1970; Brun, 1975; Brun et al., 1966; Gaultier et al., 1966; Kleinhans & Ranneberg, 1989; Marks & Cronin, 1977; Meindl & Meyer, 1969; Schulz, 1967; Subiza-Martin, 1951; von Krogh & Maibach, 1981). A further study in the same vein (Reiffers et al., 1974) was cited in the review by BIBRA International (BIBRA International, 1997).

Epidemiological and multiple case studies

A number of large scale studies of hairdressers were summarised by CIR (1998) as follows: Twelve of forty-nine hairdressers (24%) patch tested with 2.5% ammonium persulfate in petrolatum had positive reactions, compared to 1/118 non-hairdressers tested (Kellett & Beck, 1985).

Over a 5 yr period, 2320 patients with reactions to one or more allergens in a standard series were also tested with 2.5% pet. ammonium persulfate and 2% aq.potassium persulfate. A total of 22 individuals had positive reactions to these persulfates. Retrospectively, 14 of these patients were hairdressers, of which 11 reacted to both persulfates and 3 reacted to only ammonium persulfate. Of the remaining 8 non-hairdressers, 5 reacted to both persulfates and 3 reacted to only ammonium persulfate. The investigators noted that the hand dermatitis of 4 of these non-hairdressers was exacerbated by their personal use of hair bleaches (Van Joost & Roesyanto, 1991).

A multicenter study was performed in order to evaluate the frequency and source of contact sensitisation in hairdressers. Of the 302 hairdressers studied, 11.3% tested positive to 2.5% pet. ammonium persulfate (Guerra et al., 1992).

Patch test results from 9 European centres were evaluated in order to determine the frequency of sensitisation among European hairdressers. Of the 809 hairdressers tested, 8% had positive patch test results with 2.5% pet. ammonium persulfate. Of 104 clients who were patch tested because of suspected contact sensitisation, none reacted to ammonium persulfate (Frosch et al, 1993).

Over a 5-yr period, 143 atopic and non-atopic hairdressers with hand eczema were patch-tested with a hairdressers and standard series of allergens. The subjects were divided into three groups: 45 were eczematous atopics, 32 were mucous membrane atopics, and 66 were non-atopic. Seven (16%), 4 (13%), and 10 (15%) of the subjects of each group, respectively, were sensitised to ammonium persulfate (Sutthipisal et al., 1993).

One hundred and three hairdressers were patch-tested with a number of allergens over a 4-yr period. Thirty-seven hairdressers reacted to 2.5% ammonium persulfate in petrolatum. One patient had a Type 1 reaction, with airways obstruction, in addition to allergic contact dermatitis (van der Walle & Brunsveld, 1994).

Over a 9-yr period, 106 hairdressers were patch-tested with a hairdressers and standard series of allergens. Nineteen subjects (17.9%) had a positive reaction to 2.5% ammonium persulfate in petrolatum (Katsarou et al, 1995).

Eleven of twenty-three employees of a hair salon complained of upper or lower respiratory tract symptoms. Four of six had asthma which was occupationally related. These subjects developed late type asthmatic reactions after exposure to bleach powder. Bronchial provocation tests with the components of the bleach indicated that potassium persulfate was the cause (Davies & Blainey., 1983).

Four of twenty-three employees of one hairdressing salon had occupational asthma due to inhalation of bleach powders containing persulfate salts. One of the four was positive in a skin prick test to persulfate salts. When specific bronchial provocation tests were conducted on 14 of the employees, as well as 8 other individuals, the investigators reported that only those with a history of work related asthma and bronchial hyper-reactivity had positive reactions. They concluded that the response to the bleach powder was specific. Further studies indicated that the response was caused by changes in airway calibre rather than lung volumes and that mast cells may play a part in the pathogenesis of persulfate induced asthma (Blainey et al., 1986).

In a study of 98 hairdressers in,, only one case of reaction to ammonium persulfate was observed (Guo et al., 1994).

A further study (Estlander, 1990) was described in the review by BIBRA International (BIBRA International, 1997) in which out of 1082 patients (not necessarily hairdressers) with occupational skin disease between 1974 and 1983, one case was attributed to ammonium persulfate.

Other large scale studies reported in the literature are as follows:

A retrospective study of 379 hairdressing workers who attended a dermatology clinic between 1980 and 1993 was reported (Conde-Salazar et al., 1995). Patch tests were performed with a number of chemicals in a hairdressing series with 7.9% (30/379) testing positive to ammonium persulfate (2.5% in petrolatum) and 3.4% (13/379) testing positive to potassium persulfate (2.5% in petrolatum). In a number of other studies cited in this paper, positive responses to persulfates were observed in 12.8% (total cases = 35) (Conde-Salazar et al., 1985), 8.5% (total cases = 70) (Gonzalez et al., 1992) and 6.8% (total cases = 268) (Baz Manchado, 1993) of cases.

Hairdressers in North Bavaria,, with type IV (delayed) contact allergy apparently were separated into those with or without skin atopy. The percentage of those apparently reacting to ammonium persulfate was 22% (of 215) in the atopic patients and 26% (of 312) in the non-atopic patients (Coenraads & Diepgen, 1998).

Fifty-five hairdressers out of 1440 patients with suspected contact dermatitis were patch tested with two sets of hairdressing chemicals. Of thirty-four hairdressers tested with ammonium persulfate, 15% were positive (Holness & Nethercott, 1990).

Four of thirty-two hairdressers classed as having occupational dermatitis were positive when patch tested with ammonium persulfate (Shah et al., 1996).

A random sample of 500 female Finnish hairdressers was drawn from trade union membership registers. Of 355 interviewed, 130/189 who reported work-related skin or respiratory Symptoms undertook a structured computer-aided occupational medical interview and physical examination. Of these 109 were suspected and 17 were clinically diagnosed to have occupational skin or respiratory disease. Ammonium persulfate was the causative agent in 10 cases: 3 of asthma (2 of these had allergic rhinitis), 6 of allergic rhinitis (1 of these had irritant contact dermatitis), 1 of irritant contact dermatitis, 1 of laryngitis, 1 of allergic contact dermatitis and 1 of contact Urticaria. Ammonium persulfate caused 55% of the diagnosed skin and respiratory diseases and 90% of the respiratory diseases (Leino et al., 1998).

In a study of 405 hairdressers, 203 were diagnosed with dermatitis by a medical practitioner and 98 of these were stated to be allergic to one or more substances, including hair care products (38%). It was concluded that there was a marginally significant association (0.1 < p < 0.15) between the use of hair bleach and the occurrence of dermatitis when allergy and sex were controlled for. There was an increase, then a levelling, in the odds of dermatitis with increased use in both females and males. It was suggested this reflected susceptible individuals leaving hairdressing for other occupations (Stovall et al., 1983).

Large scale studies of workers other than hairdressers were also reported by CIR (1998) as follows:

Of 106 workers in a hydrogen-peroxide factory, 34% had eczematous dermatitis and 15% had asthmatic bronchitis thought to be occupational in nature. Patch tests with ammonium persulfate were positive in 32/46 workers. None of the workers had positive responses to potassium persulfate, sulphuric acid, or hydrogen peroxide. It was noted that inhalation tests with aerosolised ammonium persulfate exacerbated the Symptoms. The investigators concluded that the observed reactions were allergic in nature (Barsotti et al., 1951).

A cross-sectional study of 52 employees of a plant that produced persulfates was performed; 12 subjects were directly involved in persulfate production, the remaining 40 subjects had indirect contact. Thirteen persons from the medical profession were used as controls. Questionnaires were administered, skin prick tests were performed with 1 and 5% (w/v) ammonium and potassium persulfate, atopy Screening was done, and lung function was assessed. Three, 2, and 3 test subjects reacted to ammonium, potassium, and both ammonium and potassium persulfate, respectively; of these 8 subjects, only 3 had direct contact with persulfates. Six of the 8 subjects reported workplace-related breathing difficulties; 9/44 non-reactive test subjects also reported breathing difficulties. None of the controls reacted to the prick test. The mean total IgE was increased in 16 subjects; a Phadiatop test (Screening of IgE antibodies against common inhalant allergens) reported positive results in 12 test subjects. Test subjects that had positive results to the prick test had decreased lung function values compared to those subjects that had negative results (Wrbitzky et al., 1995).

A cross-sectional study of 32 employees of a chemical plant that produced persulfates was performed. Eighteen workers at the plant that were not exposed to persulfates were used as the controls. Questionnaires were administered concerning symptoms, history of allergy and persulfate exposure. Skin prick tests were performed with 80 mg/mL buffered ammonium (pH 3.1) and sodium persulfate (pH 3.9), total IgE and specific IgE were measured and lung function and bronchial responsiveness to histamine were assessed. Work-related rhinitis was reported by 1 test subject, and work-related Conjunctivitis and bronchitis were reported by 2 control subjects. Early and/or late skin reactions to persulfates were not observed for test or control subjects. Lung function, total IgE, and response to histamine were similar for test and control subjects. Bronchial hyper responsiveness was present in 4 non-atopic test subjects and in 1 non-atopic and 1 atopic control worker. It was noted that 7/36 ex-workers left because of medical reasons; 6 had work-related contact dermatitis and 1 reported asthma (Merget et al., 1996).

Forty-two of four hundred bakers examined had positive patch test reactions to ammonium persulfate. However, only 1/150 individuals not in the baking industry reacted to this ingredient (Grosfeld, 1951).

Five bakers with occupational eczematous dermatitis were tested with a variety of baking ingredients using on-off and patch tests to determine the cause of their dermatitis. Two of the workers were sensitive to persulfates (Nava et al., 1983).

In summary, immediate and delayed contact hypersensitivity, contact urticaria, rhinitis, bronchitis and asthma (early and late onset) have all been observed in hairdressers as a result of exposure to hair bleaching powders containing persulfates.