Registration Dossier

Administrative data

Description of key information

Key value for chemical safety assessment

Skin sensitisation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

Read-across concept for sulfites, hydrogensulfites, metabisulfites, dithionites and thiosulfates:

A comprehensive read-across concept has been developed for sulfites, hydrogensulfites and metabisulfites, based on the pH-dependant equilibrium in aqueous solutions which is summarised in the following equations:[1],[2]

           SO2+ H2O <->`H2SO3´         H2SO3<->H++ HSO3-<->2H++SO32-    2HSO3-<->H2O +S2O52-

Since the nature of the cation (i.e., sodium, potassium, ammonium…) is not assumed to contribute substantially to differences in toxicity and solubility (all compounds are very soluble in water), only the chemical and biological properties of the anion are considered as relevant determinants. Based on the described equilibrium correlations, unrestricted read-across between the groups of sulfites, hydrogensulfites and metabisulfites is considered justified.

 

Additionally, it is known that sodium dithionite disproportionates in water to form sodium hydrogen sulfite and sodium thiosulfate (equation II)2,[1], so that this substance can also be considered to be covered by the read-across concept described above. Since it can easily be anticipated that the substance is not stable enough under physiological conditions to fulfil the requirements of study guidelines, instead the products of decomposition have to be considered:

 

       2 S2O42-+ H2O→2HSO3-+ S2O32-

 

Not fully covered by this read-across concept is the substance class of thiosulfates: although the thiosulfates are also well known to disproportionate in aqueous solution to form polythionic acids and SO2(HSO3-), this requires somewhat different, more acidic conditions. Therefore, read-across to sulfites is primarily restricted to appropriate physiological conditions, i.e. oral administration where the gastric passage with the strongly acidic conditions in the stomach will facilitate the chemical disproportionation described above:

 

       HS2O3-+ H2S2O3HS3O3- + SO2+ H2O

 

[1]Hollemann Wiberg, Lehrbuch der Anorganischen Chemie, 101.Auflage

[2]Handbook of Chemistry and Physics, Ed. Lide, DR, 88thedition, CRC Press

Skin sensitisation:

One reliable animal study, according to OECD guideline429 (Skin Sensitisation: Local Lymph Node Assay, modified according to Ehlings et al. 2005)and according to GLP has been performed for sodium metabisulfite which shows that the test item has no sensitising properties.

The positive control group caused the expected increases in lymph node cell count (S.I.: 2.427) and lymph node weight (S.I.: 2.160) (statistically significant at p ≤ 0.01). Therefore, the study is regarded as valid. The stimulation index for ear weight was 1.033.

Treatment with sodium metabisulfite at concentrations of 10%, 25% or 50% did not reveal statistical significantly increased values for lymph node cell count, all stimulation indices for the lymph node cell count were beneath the threshold value of 1.4. In addition, the lymph node weight was not increased. Hence, the test item is classified as not sensitising.

One further reliable animal study, according to OECD guideline429 (Skin Sensitisation: Local Lymph Node Assay, modified according to Ehlings et al. 2005) and according to GLP has been performed for sodium sulfite which shows that the test item has no sensitising properties.

The positive control group caused the expected increases in lymph node cell count (S.I.: 2.106) (statistically significant at p ≤ 0.01). Therefore, the study is regarded as valid. The stimulation index for lymph node weight was 2.160 and for ear weight 1.040.

In addition to industrial use with resulting occupational exposure, sulfites are in wide-spread use (among others) in foods/food processing, as well as in cosmetics and pharmaceutical preparations, resulting in extensive consumer exposure. For this obvious reason, the potential for sensitisation has been reviewed extensively by several renown international (SCF, 1997; SCCNFP, 2003; CIR, 2003; EFSA, 2004) and national (MAK, 2014) scientific organisations. In addition, an OECD SIDS (2001) exists, based upon which all OECD Member States have agreed on a mutually agreed dataset (“MAD”). Therefore, despite the extensive data base that is reflected in this registration dossier, an own interpretation and evaluation has not been undertaken; instead, the opinions of these scientific bodies has been adopted, as summarised below:

 

German MAK Commission (2014):

Effects in humans – immediate reactions:

Intolerance reactions which develop within a few minutes after the oral or (less often) pulmonary uptake of sulfites manifest mainly as bronchospasms, asthma and urticaria. Reported less often are also itching, oedema, a runny nose and nasal congestion (Atkinson et al. 1993; Baker et al. 1981; Belchi-Hernandez et al. 1993; Cartier et al. 1995; Halpern et al. 1985; Howland and Simon 1989; Jamieson et al. 1985; Koepke and Seiner 1982; Malo et al. 1995; Peroni and Boner 1995; Prenner and Stevens 1976; Schwartz and Sher 1985 a, b; Schwartz et al. 1989).

The mechanism by which sulfites lead to bronchoconstriction and bronchial asthma has not been completely clarified. It is probably the direct action of sulfur dioxide on nerve receptors in the lungs (Atzori et al. 1992; Bannenberg et al. 1994; Meyers et al. 1986; Snashall and Baldwin 1982). The release of calcitonin gene-related peptide (CGRP) from capsaicin-sensitive sensory nerves is thought to be involved in the reaction (Bannenberg et al. 1994). Although a cholinergic reflex is supposed (Meyers et al. 1986; Snashall and Baldwin 1982), the role of the parasympathic nerve system has not been completely clarified, as the intravenous administration of atropine does not necessarily prevent sodium metabisulfite-induced bronchoconstriction (Lötvall et al. 1990). Also the direct or indirect sulfite-mediated release of histamine from basophilic leukocytes has been discussed (Sokol and Hydick 1990; Twarog 1977). However, antihistamines are ineffective in the therapy of sulfite-induced asthma (Dixon and Ind 1988).

In some, but not all patients with intolerance reactions to ingested sodium metabisulfite, prick testing or intracutaneous testing with sodium metabisulfite or potassium metabisulfite leads to reactions of immediate type (Boxer et al. 1988; Simon and Wasserman 1986; Sokol and Hydick 1990; Wüthrich and Huwyler 1989; Yang et al. 1986). In individual cases, passive transferral of the sulfite intolerance to other persons or to monkeys was possible via unheated serum. Attempts to do this using sera heated to above 56°C were not successful (Boxer et al. 1988; Simon and Wasserman 1986; Yang et al. 1986). These findings were regarded by various authors as clear evidence that sulfite intolerance - at least in individual cases - is a genuine IgE-mediated allergic reaction, although specific IgE against sulfites has never been detected (Boxer et al. 1988; Simon and Wasserman 1986; Sokol and Hydick 1990; Wüthrich and Huwyler 1989; Yang et al. 1986).

In addition, a partial deficiency in the enzyme sulfite oxidase, which is responsible for the oxidation of sulfite to form the inactive sulfate, has been suggested as another possible cause of sulfite intolerance. In individual cases, the administration of vitamin B12was able to prevent sulfite-induced bronchospasms in asthmatic children. The authors attribute this to the possible vitamin B12-induced oxidation of sulfite to sulfate (Añíbarro et al. 1992).

Investigations in patients with suspected sulfite-induced asthma are usually carried out as oral challenge tests. Sodium metabisulfite is ingested in increasing doses from 1 to 200 mg; every 30 minutes the next-higher dose is taken. Sulfite-sensitive persons react even to low doses with airway obstruction (decrease in the FEV1of more than 20%), while non-sensitive persons tolerate even the final dose and the high cumulative total dose without pulmonary reactions (Acosta et al. 1989; Hein et al. 1996; Koepke et al. 1985; Lee et al. 1986; Sokol and Hydick 1990). In one study, the test is regarded as very specific, but only moderately sensitive (Hein et al. 1996).

Challenge tests with sodium metabisulfite were also carried out via inhalation. The patients inhaled nebulized sodium metabisulfite solutions of 0.01 mg/ml, which in patients with sulfite-induced asthma led to a significant decrease in the FEV1of more than 20% (Jamieson and Guill 1984; Jamieson et al. 1985). In a study in which the test persons inhaled a nebulized sodium metabisulfite solution of 0.6mg/ml, significant decreases in the FEV1were observed in 3 patients with sulfite-induced asthma and in 4 of 10 asthmatics without anamnestic evidence of sulfite intolerance, but not in 10 healthy control persons (Koepke et al. 1985).

 

Effects in humans – delayed reactions:

There are several case reports and surveys about contact allergy to sodium metabisulfite. The following case reports are well documented. A 31-year-old woman with psoriasis developed allergic contact dermatitis after using a cream containing sodium metabisulfite. She reacted in the patch test to sodium metabisulfite concentrations of 0.2% and above in petrolatum and to sodium bisulfite concentrations of 0.2% and above in water, but not to sodium sulfite concentrations of up to 1% in water (Vestergaard and Andersen 1995). A baker with dermatitis on the hands had occupational contact with sodium metabisulfite. In a patch test he reacted to 2% sodium metabisulfite in water (Apetato and Marques 1986). A female employee of a photographic laboratory with existing mild dermatitis on the hands complained of a deterioration of the skin condition after contact with photographic chemicals containing sodium metabisulfite. In a patch test she reacted to 5% sodium metabisulfite in petrolatum (Jacobs and Rycroft 1995). A female patient who developed severe oedema on her face and neck after the injection of a local anaesthetic during dental treatment reacted in a patch test to the local anaesthetic used and the sodium metabisulfite contained in it (5% in petrolatum) (Dooms-Goossens et al. 1989).#

Less well documented are the reports about two patients with “burning mouth syndrome”; patch tests with 5% sodium metabisulfite in petrolatum likewise produced a reaction. The authors believed there to be a connection between the oral symptoms and the ingestion of foods containing sulfite (Levanti et al. 1996). A patient who produced wine as a hobby developed dermatitis on the hands after con- tact with potassium metabisulfite during this activity. In a patch test he reacted to sodium metabisulfite; the test concentration is, however, only insufficiently documented (Laramé et al. 1989).

At the beginning of the 1990s, 1% sodium metabisulfite in petrolatum was routinely tested for two years in the standard series in 2894 patients at the dermatological department of the university hospital of Bari, Italy. Reactions were observed in 50 patients (1.7%). All 50 patients reacted also to 1% potassium metabisulfite in petrolatum and 1% and 5% sodium bisulfite in petrolatum, but only 2 reacted to 1% sodium sulfite in petrolatum. In 12 of the 50 patients the test reaction was relevant (7 patients with occupational relevance, 5 patients with non-occupational relevance) (Vena et al. 1994).

A case study describes a female patient with contact allergies to bisulfite who became sensitized through the use of a bleaching cream which contained ammonium bisulfite. In a patch test she reacted to sodium bisulfite and ammonium bisulfite concentrations of 0.1% and above in water (Pambor 1996). There are further studies of contact allergies to sodium sulfite, which was contained in a widely-used fungicide. A total of 4 cases of contact allergy are documented. Sodium sulfite concentrations of 0.2%, 2% and 5% in petrolatum and of 0.2% and 2% in water were tested (Lodi et al. 1993; Vissers-Croughs et al. 1988). In 1989, a survey was carried out in Denmark. A concentration of 1% sodium sulfite in petrolatum was tested in the standard series in 1762 patients; 25 patients (1.4%) reacted. The relevance of these reactions could not, however, be determined (Sand Petersen and Menné 1992).

  

Animal Studies

Studies with animals of the sensitizing effects of sulfites were not available to the MAK Commission at the time of compilation of their assessment. However, it should be noted here that a guideline-conform Local Lymph Node Assay (Haferkorn, 2010) was conducted in mice under GLP with the substance disodium disulfite (i.e. sodium metabisufite). In this study, the positive control group caused the expected increases in lymph node cell count (S. I.: 2.427) and lymph node weight (S. I.: 2.160) (statistically significant at p ≤ 0.01). Therefore, the study was regarded as valid. The stimulation index for ear weight was 1.033. Treatment with sodium metabisulfite at concentrations of 10%, 25% or 50% did not reveal statistically significantly increased values for lymph node cell count, all stimulation indices for the lymph node cell count were beneath the threshold value of 1.4. In addition, the lymph node weight was not increased. Hence, the test item may be considered as lacking skin sensitising effects.

 

Conclusions of the MAK Commission with respect to classification for sensitising properties:

Ingested or inhaled sodium metabisulfite can lead in predisposed persons to urticaria and bronchospasms. An immunological pathogenesis has not been proven for these reactions and is assumed - if at all - only for a small minority of affected persons. Sodium metabisulfite is, therefore, not regarded as a respiratory sensitiser.

There are no studies available on the effects of potassium metabisulfite or sodium bisulfite on the airways. In view of the chemical properties mentioned in the introduction, it can, however, be assumed that potassium metabisulfite and sodium bisulfite have effects similar to those of sodium metabisulfite.

In the literature, many cases of contact allergies and patch test reactions to sodium metabisulfite and occasionally also to potassium metabisulfite are described. Evidently patients who react to sodium metabisulfite react also to sodium bisulfite. In view of the widespread use of sodium metabisulfite, and therefore the numerous possibilities for contact in everyday life and the occupational field, the number of persons epidermally sensitised is, however, very small. Sodium metabisulfite, potassium metabisulfite and sodium bisulfite are therefore not designated as dermal sensitisers.

 

European Scientific Committee for Food (1997):

Intolerance to sulfites: sulfiting agents are used to control microbial growth in fermented beverages and have been widely used in the food and beverage industry for over 2,000 years. They have been shown to frequently trigger attacks of asthma and more rarely other symptoms. Sulfites are also used in drugs and those administered by inhaled route have caused several asthmatic reactions. Moreover, it seems that dental local anaesthetics preserved with sulfites have induced asthma in some patients. Despite the increasing amount of data that has accumulated on sulfites as the intensity of medical interest has expanded in recent years, there are lingering, difficult, and, in most cases, as yet unresolved questions about sulfite sensitivity. These include questions about mechanism of action, prevalence and the most effective methods for protecting sulfite-sensitive patients from exposure to sulfites.

Mechanisms of sulfite intolerance: the exact mechanism of sulfite sensitive asthma is unknown but most likely involves hyperreactivity to inhaled SO2, in the great majority of cases, however, there are reports of IgE mediated reactions and other sulfite sensitive asthmatics have been found with low Ievels of sulfite oxidase, necessary to oxidise endogenous sulfite to sulfate. It is possible that patients with sulfite-induced urticaria present more commonly an IgE-mediated reaction.

Sulfite-induced asthma: most patients with sulfite-asthma have severe chronic asthma, often corticosteroid-dependent. Usually, but not always, the patient is non-atopic but may suffer from chronic vasomotor rhinosinusitis. These individuals are differentiated from aspirin-sensitive asthma because they lack nasal polyps and eosinophilia. However, some aspirin sensitive asthmatics present also a dyspnoea after ingestion of wine or foods containing sulfites. Fatal asthma has been reported afteringestion of sulfite-containing foods or drinks. There is no link between sulfite sensitivity and the degree of airways hyperresponsiveness. Due to the lack of reliable skin or in vitro tests, the diagnosis of an intolerance to sulfites is still based on placebo-controlled oral provocation tests. Sulfite-challenge may cause a severe asthmatic reaction so it should be carried out by trained investigators and when the patient is in a stable state.A screening challengeshould be conducted in a single-blind, open fashion, and when positive, confirmed by a double-blind challenge. Doses of ingested sulfites should be increased cautiously. However, sulfite-sensitive subjects with asthma will not necessarily react after ingestion of sulfited foods. The likelihood of a reaction is dependent on the nature of the food, the level of residual sulfite,the sensitivity of the patient,and perhaps on the form of residual sulfite and the mechanism of the sulfite-induced reaction.

Other symptoms: delayed eczematous and immediate urticarial reactions due to sulfites in foods have been demonstrated. However, sulfites represent a rare cause of urticaria. Anaphylaxis-likereactions are exceptionally caused by sulfites. Clinical cases of contact allergy due to sulfites have been observed.

Prevalence of adverse reactions to food: The prevalence of sulfite sensitivity in the general population is unknown and in asthma reports on a small number of patients have lead to a variable prevalence. Both adults and children can suffer from sulfite-induced asthma. It appears that patients with severe asthma, such as steroid-dependent bronchial asthma present a greater prevalence of sulfite sensitivity. Some reports suggest a prevalence of sulfite sensitivity in at least 5% of the asthmatic population. However, a very careful study proposed that on the basis of challenges, the best estimate of the prevalence of sulfite sensitivity in the asthmatic patients studied is 3.9%. This population, however, contained a larger number of steroid-dependent asthmatic patients than would be found in the general asthmatic population. It was concluded, therefore, that the prevalence of sulfite sensitivity in the asthmatic population as a whole would be less than 3.9% and that steroid-dependent asthmatic patients are most at risk. These studies combine to indicate that using double-blind food challenge, the prevalence of food allergy is far below 1% of the population in adults and may be slightly greater in children. However, this may be a low estimate since food challenge may not identify the entire population of food allergic individuals and there are some genetic and environmental factors that can increase the prevalence of food allergy.

 

Scientific Committee on Cosmetic Products and Non-Food Products Intended for Consumers (SCCNFP) 2003:

According to evaluation of test results in a corresponding CIR report (Cosmetic Ingredient Review, 2000: Safety Assessment of Sodium Sulfite, Potassium Sulfite, Ammonium Sulfite, Sodium Bisulfite, Ammonium Bisulfite, Sodium Metabisulfite, and Potassium Metabisulfite), which were officially submitted to the US governing bodies, neither sodium sulfite nor sodium metabisulfite were found to be a potent primary sensitiser. Related to this data and by in-use experience with millions of cosmetic products per year containing these sodium salts, it seems to be unlikely that a modification of the cation would increase the sensitisation potential seriously. In conclusion, inorganic salts of sulfite, bisulfite and metabisulfite are rare contact allergens.

 

Cosmetic Ingredient Review Expert Panel, (CIR, 2003):

Sulfite Sensitivity:

Many asthmatics with bisulfite sensitivity have negative allergy skin tests, suggesting a nonatopic nature. Twarog & Leung reported that immunoglobulinE (lgE), total eosinophil counts, and histamine concentrations were normal during acute reactions, suggesting the lack of an lgE mechanism.

Approximately 2-5% of asthmatics are estimated to be sulfite sensitive, and most sulfite-sensitive individuals are asthmatics. Sulfite-sensitive asthmatics react to ingestion or parenteral administration of sulfites. Asthmatics in general are more sensitive to inhaled sulfur dioxide (tested as Sodium Metabisulfite) than are non-asthmatic normal subjects (Koepke, Staudenmayer & Selner, 1985; Wright et al. 1990), but inhalation sensitivity alone is not considered indicative of sulfite sensitivity (Gunnison & Jacobsen, 1987). In the majority of instances, manifestations include dermatologic signs and symptoms such as urticaria, angioedema, hives and pruritus, flushing, tingling, and swelling. Respiratory signs and symptoms include dyspnoea, wheezing and bronchoconstriction, while gastrointestinal symptoms include nausea and gastric cramps. Bronchoconstriction is a common reaction in steroid-dependent asthmatics. Less common are hypotension, cyanosis, diaphoresis, shock, and loss of consciousness. Clinical management involves avoidance of sulfited food, beverages and pharmaceuticals by people at high risk (Jamieson et al. 1985; Simon 1986; Lester1 995).

Yang, Purchase, and Rivington ( 1986) reported that results of skin tests, provocative oral challenge test, and passive transfer tests suggested that some metabisulfite-sensitive reactions can be IgE-mediated.

Corder & Buckley (1995) studied a tertiary-referral clinic population to estimate safe exposure doses for use in epidemiological studies of acute versus allergic reactions. A positive response was defined as a 15% decrease in the amount of air expired in 1s following ingestion of the substance. The median effective molar dose for Sodium Metabisulfite was 34.4 mg (0.19 mM). The most sensitive persons (5% of population) might respond to 4.6 mg Sodium Metabisulfite and practically all (95%) susceptible persons might respond to 255.8 mg.

 

Contact Dermatitis - Sodium Sulfite

Petersen & Menm (1992) patch tested 1762 dermatologic patients with Sodium Sulfite 1% petrolatum (pet.). Following 2 days of occlusive exposure, positive reactions were observed in 25 patients (1.4% incidence). Seven of the 25 tested positive only to Sodium Sulfite (the European standard series was also tested). Only 3 of the 25 patients had previous contact with ketoconazole cream (contains Sodium Sulfite). The investigators did not consider it worthwhile to routinely patch test with Sodium Sulfite because the "clinical relevance of the positive reactions to sodium sulfite remains to be established."

A hair-colouring agent with 0.64% Sodium Sulfite was used in a repeat insult open patch test involving 100 participants. The panelists received 0.2ml or 0.2g of the test material directly onto a designated area of the back. The procedure was repeated until nine consecutive applications had been made for every Monday, Wednesday, and Friday for 3 consecutive weeks. Reactions were scored just before the next application. The panelists were then allowed a 10-14-day non-treatment period, after which a challenge or retest application was applied once to a previously unexposed site. Retest doses were equivalent to any of the original nine exposures and were scored 24 and 48 h after application. Comparisons were made between the sensitizing doses and the retest doses. No adverse reactions were observed and according to the investigators, the test material can not considered a primary irritant or primary sensitizer (Combe Incorporated 1 996).

Samples of 0.5% Sodium Sulfite in a topical feminine cream were patch tested using 100 panelists. The semi-occlusive patch, containing 0.2 ml or 0.2 g of the test material, was affixed directly onto the back and removed after 24 h. The procedure was repeated until nine consecutive applications had been made for every Monday, Wednesday, and Friday for 3 consecutive weeks. Reactions were scored just before the next application. The panelists were then allowed a 10- to 14-day non-treatment period, after which a challenge or retest application was applied once to a previously unexposed site. Retest doses were equivalent to any of the original nine exposures and were scored 24 and 48 h after application. No adverse reactions were observed and according to the investigators, the test material cannot be considered a primary irritant and primary sensitizer (Combe Incorporated 1998).

 

Contact Dermatitis - Sodium Metabisulfite

Vena, Foti & Angelini (1 994) reported the results of patch testing in 2894 eczematous patients over a 2-year period. Positive reactions to Sodium Metabisulfite 1 % pet. (following a 2-day occlusive exposure) were noted in 50 patients (1.7% incidence). All 50 patients also reacted to Potassium Metabisulfite 1% pet., and to Sodium Bisulfite 1% and 5% pet. Only two reacted to Sodium Sulfite 1 % pet. Prick and intradermal tests of 20 patients with a Sodium Metabisulfite solution ( 1 0 mg/ml) were negative and oral challenge of five patients with 30 and 50 mg Sodium Metabisulfite did not provoke a flare-up of dermatitis or patch test. The dermatitis was considered occupational in seven cases. Five of the remaining 43 cases were considered allergic contact dermatitis resulting from the use of topical preparations.

 

Conclusions of the CIR:

Between 2% and 5% of asthmatics are sulfite sensitive. The FDA established regulations regarding the use of sulfiting agents in order to minimise the hazards to this population. Clinical oral and ocular exposure studies reported no adverse effects. The sodium and potassium salts produced positive reactions in dermatologic patients under patch test.

The CIR Expert Panel concluded that sodium sulfite, potassium sulfite, ammonium sulfite, sodium bisulfite, ammonium bisulfite, sodium metabisulfite, and potassium metabisulfite are safe as used in cosmetic formulations.

 

European CommissionScientific Panel on Dietetic Products, Nutrition and Allergies (EFSA, 2004):

Background:

Sulfites can occur naturally in foods as a consequence of fermentation, but they are also added to foods as preservatives. The prevalence of sulfite sensitivity in the general population is unknown, but it appears to be rare among non-asthmatics. Most reactions to sulfites are characterised by severe bronchospasm, which can occur within minutes after ingestion of sulfite-containing foods. The smallest concentration of sulfites able to provoke a reaction in sensitive individuals has not been determined. Use of sulfating agents in fruits and vegetables intended to be consumed raw appears to be responsible for several cases of sulfite-induced bronchospasm. This led to the prohibition of sulfite use in raw fruits and vegetables by the United States Food and Drug Administration, which also requires labelling of foods containing sulfiting agents in concentrations of 10 ppm (10 mg/kg) or more. Though the threshold for sensitivity reactions may be even lower, the regulation is based on the fact that the assay used to detect the level of sulfites in foods is not sensitive enough to detect amounts less than 10 mg/kg.

Although adverse reactions to ingested sulfites were first reported in 1976 (Prenner and Stevens, 1976), it was not until after 1980 that reports on sulfite sensitivity became sufficiently common to attract the attention of the scientific community, consumer groups and regulatory agencies (Gunnison and Jacobsen, 1987). Despite the increased amount of data that has accumulated on sulfites as the interest of the medical community on the issue has grown in recent years, several questions regarding prevalence, pathogenesis and best protection of sulfite-sensitive patients remain to be settled.

Sulfites are a well-recognised part of the human natural and artificial environment. Considerable quantities of them are generated in the body by normal catabolic processing of sulfur-containing compounds, notably the amino acids cysteine and methionine (Cooper, 1983; Gunnison and Jacobsen, 1987). In foods, sulfites can occur naturally as a consequence of fermentation (e.g. during the fermentation of wine) (Tayloret al., 1986), but they are also added to foods and used as preservatives, a practice that has been implemented for centuries (Bushet al., 1986a). Sulfites can also be found in medications, including those used for the treatment of allergic reactions (Nicklas, 1989). Finally, individuals are exposed daily to SO2as part of the air pollution mix.

Sulfites can occur naturally in foods as a consequence of fermentation, but they are also added to foods as preservatives. Though sulfites were already used in antiquity by Egyptians to cleanse their wine vessels, their first recorded use as food preservatives occurred in 1664, when cider was stored in flasks of SO2 to retard spoilage. In the United States, SO2 has been widely used since the late 1800 and the sulfite salts since 1920; they were first used in the manufacture of wine and beer and their use expanded to several other products (Bushet al., 1986a). Sulfites provide a number of useful attributes when applied to foods: they inhibit enzymatic browning (especially in fresh fruits and vegetables, shrimps and raw potatoes), as well as non- enzymatic browning (especially in dried foods and dehydrated vegetables) (Tayloret al., 1997; Simon, 1998). They also have antimicrobial activity (as in wine and beer), dough- conditioning properties (as in frozen pies and pizza crusts) and bleaching effects (as in maraschino cherries) and they are used as processing aids in beet sugar (Bushet al., 1986a; Simon, 1998).

The levels of sulfites contained in foods range from under 10 mg/kg (e.g. frozen doughs, corn syrup, jellies) to 60 mg/kg (e.g. fresh shrimp, pickles, fresh mushrooms) to 100 mg/kg (e.g. dried potatoes, wine vinegar). The highest levels of sulfites (up to 1000 mg/kg) can be found in dried fruit, wine, fruit juices (e.g. lemon, lime, grape) and certain freshly prepared sauces available from retailers (Simon, 1998). In Table 28, Lester (1995) presents a list of certain foods by level of their total content in sulfites (naturally-occurring plus added).

The SCF (1994) identified a no observed effect level (NOEL) of 70 mg/kg body weight/day of sulfur dioxide equivalents (SDE) for gastric irritation in animals. Based on this and on a safety factor of 100, the Committee estimated an Acceptable Daily Intake (ADI) of 0-0.7 mg SO2/kg body weight that would ensure that gastric reactions will not occur in man. It explicitly stated, however, that “a numerical ADI would not prevent the occurrence of sulfite-induced asthma” and that it considered that “EC labelling regulations should ensure that the presence of added sulfites in foods and non-alcoholic beverages is always indicated in the list of ingredients.

 

Prevalence of sulfite sensitivity:

The prevalence of sensitivity to sulfating agents in the general population is unknown, whereas estimates of the percentage of asthmatics characterised as sensitive to oral sulfite challenge covers a broad range. The probable reasons for this wide disparity are the difference in the physical form of the orally administered sulfating agent, the differences in the selection criteria of the sample populations submitted to oral challenges and the non-uniformity in the upper limit of challenge dose as well as in the minimum decrement in pulmonary function accepted as evidence of sulfite-induced bronchoconstriction. In addition, adequate placebo challenges have not been carried out in some studies.

The SCF stated that the percentages of sulfite sensitivity reported in the literature (1-4% among all asthmatics and 5-10% among steroid-dependent asthmatics) are probably overestimates based on patients referred to allergy clinics (SCF, 1994). The FDA estimates that one out of a hundred people are sulfite sensitive, and that 5% of those who have asthma are also at risk of suffering an adverse reaction to the substance (FDA, 1996).

 

Symptoms of sulfite sensitivity:

Most reactions to sulfites are characterised by bronchospasm, occasionally severe, which can occur within minutes after ingestion of sulfite-containing foods. Bradycardia, flushing and prominent gastrointestinal symptoms, as well as urticaria, angioedema, hypotension and shock have also been observed.

 

Pathogenesis:

The pathogenesis of adverse reactions to sulfites has not been clearly documented. Three possible mechanisms have been invoked: (i) an IgE-mediated reaction, (ii) a sulfite-induced cholinergic response, and (iii) low levels of the enzyme sulfite oxidase (Bushet al., 1986a; Nicklas, 1989; Lester, 1995).

There have been reports of positive skin tests to sulfites,in vitromediator release or passive sensitivity transfer (Prenner and Stevens, 1976; Twarog and Leung, 1982; Wolf and Nicklas, 1985; Yanget al., 1986; Simon and Wassserman, 1986; Boxeret al., 1988), all of them suggesting that some reactions to sulfites may in part be IgE-mediated. A specific antibody, however, has never been identified (Lester, 1995) and the majority of studies have not been able to demonstrate an IgE-mediated mechanism (Gunnison and Jacobsen, 1987; Nicklas, 1989). It appears that IgE-mediated mechanisms, if they do exist, are very rare in sulfite sensitive individuals or are limited to a subset of patients (Bushet al., 1986a; Sainte-Laudyet al., 1994).

The broncho-constrictive effect of inhaled SO2, mediated by parasympathetic nerve endings in the bronchi, has been studied with respect to environmental pollutants. Whether gastroesophageal reflux of SO2 causes bronchospasm in sulfite sensitive patients is not clear. It has been hypothesised that it is also possible to mechanically distend the stomach, produce a cholinergic response and stimulate release of gastrin and other active mediators in sulfite sensitive patients (Nicklas, 1989).

Low levels of the mitochondrial enzyme sulfite oxidase have been demonstrated in some sulfite sensitive patients (Jacobsenet al., 1984; Stevenson and Simon, 1984). Absorbed sulfites are added to those produced endogenously and increase the demand placed on the enzyme sulfite oxidase. It is possible that when this demand is not met, sulfite sensitive patients exhibit symptoms.

Finally,it has also been hypothesised that a number of food additives, including sulfites, induce intolerance because of their aspirin-like properties (Williamset al., 1989) and an association between respiratory reactions to aspirin and those to sulfites has been reported (Sabbahet al., 1987; Hassounet al., 1994).

 

Conclusions of the Scientific Panel on Dietetic Products, Nutrition and Allergies:

The prevalence of sulfite sensitivity in the general population is unknown, but it appears to be rare among non-asthmatics. Estimates of the percentage of asthmatics characterised as sensitive to oral sulfite challenge range from less than 4% up to 66%. The SCF stated that the reported prevalences of 1-4% among all asthmatics and 5-10% among steroid-dependent asthmatics are probably overestimates based on patients referred to allergy clinics (SCF, 1994). Most reactions to sulfites are characterised by severe bronchospasm, which can occur within minutes after ingestion of sulfite-containing foods. In restaurants, the sudden choking sensation may incorrectly be attributed to aspiration of food. Assessment of consumer exposure to sulfites in foods is inadequate, as the currently available measurement methods have several shortcoming and the amounts of sulfites initially used to treat the foods do not reflect residue levels after processing. The average daily sulfite consumption has been estimated to be approximately 20 mg of sulfur dioxide equivalents. It should be noted, however, that sulfites are frequently used in restaurant foods as preservatives and an average restaurant meal may contain sulfites well in excess of 25 mg. Most sulfite sensitive individuals will react to ingested metabisulfite in quantities ranging from 20 to 50 mg. Use of sulfiting agents in fruits and vegetables intended to be consumed raw appears to be responsible for several cases of sulfite-induced bronchospasm. This led to the prohibition of sulfite use in raw fruits and vegetables by the USA FDA, which also requires labelling of foods containing sulfiting agents in concentrations of 10 mg/kg or more. Though the threshold for sensitivity reactions may be even lower, the regulation is based on the fact that the assay used to detect the level of sulfites in foods is not sensitive enough to detect amounts less than 10 mg/kg. However, threshold levels have not been systematically assessed and the smallest concentration of sulfites able to provoke a reaction in a sensitive person is unknown.

 

OECD SIDS (2001)

Regarding skin sensitization in animals, the OECD SIDS concluded at the time thatno guideline studies were available for an assessment, except for one guinea pig maximisation assay (Eastman Kodak Co., U.S. TOSCA Submission OTS 0557795, DocID 86940000099, 1980). Despite that this study was not well-documented, in this test 0/10 guinea pigs responded positively, so that there was no indication of sensitisation whatsoever. The OECD SIDS then further mentions that in humans urticaria and asthma with itching, oedema, rhinitis and nasal congestion are reported, but that an immunological pathogenesis of these is however still unclear. In a few cases, allergic contact dermatitis, as well as positive patch-testing was observed.

The OECD SIDS concluded that disodium disulfite is not considered a skin sensitiser and is also unlikely to induce respiratory sensitisation, but may enhance symptoms of asthma in sensitive individuals.

OECD SIDS when agreed by the participating member states attain “MAD” (mutual acceptance of data) status, with Hungary having joined the OECD in May 1996.


Migrated from Short description of key information:
Disodium disulfite did not reveal any sensitising properties in a local lymph node assay (Haferkorn, 2010). In an OECD SIDS (2001), a guinea pig maximisation assay with negative outcome is also reported. The prevalence of sulfite sensitivity in the general population is unknown, but it appears to be rare among non-asthmatics. Estimates of the percentage of asthmatics characterised as sensitive to oral sulfite challenge range from less than 4% up to 66%. The EU SCF (1994) stated that the reported prevalences of 1-4% among all asthmatics and 5-10% among steroid-dependent asthmatics are probably overestimates, based on the reports being associated with patients referred to allergy clinics. Most reactions to sulfites are characterised by severe bronchospasm, which can occur within minutes after ingestion of sulfite-containing foods An immunological pathogenesis has not been proven for these reactions and is assumed - if at all - only for a small minority of affected persons. In the literature, many cases of contact allergies and patch test reactions to sodium metabisulfite and occasionally also to potassium metabisulfite are described. Evidently patients who react to sodium metabisulfite react also to sodium bisulfite. In view of the widespread use of sodium metabisulfite, and therefore the numerous possibilities for contact in everyday life and the occupational field, the number of persons epidermally sensitised is, however, very small. Based on animal and human data, substances such as sodium metabisulfite, potassium metabisulfite and sodium bisulfite are therefore not considered to represent dermal sensitisers.
Sodium sulphite did not reveal any sensitising properties in the local lymph node assay (modified OECD 429, method according to Ehlings et al. 2005). Thus, sodium sulphite does not require classification as sensitiser.

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:
Migrated from Short description of key information:
There is no evidence on specific respiratory hypersensitivity in test animals following acute inhalation exposure with sodium sulfite. Based on the read-across concept (see `discussion`), it can be concluded that ammonium sulfite has no respiratory senstising properties.

Justification for classification or non-classification

Skin Sensitisation:

The reference Haferkorn (2010) is considered as key study on skin sensitisation and will be used for classification. The overall sensitisation results in this Local lymph node assay (LLNA) in mice are as follows: SIs of less than 1.4 (0.854 - 0.970) were observed for lymph node cell count at all test concentrations of disodium disulfite (10, 25, 50 %) and no dose response could be observed. In addition, SIs of less than 1.1 (0.934 - 1.086) were observed for ear weight at all concentrations of disodium disulfite (10, 25, 50 %). Thus, the classification criteria acc. to Directive 67/548/EEC and Regulation (EC) 1272/2008as skin sensitizer are not met and disodium disulfite has not being labelled as such.

One further reference Haferkorn (2010) is considered as key study on skin sensitisation and will be used for classification. The overall sensitisation results are as follows: Local lymph node assay (LLNA) in mice SIs of less than 1.4 (0.825 - 1.014) were observed for lymph node cell count at all test concentrations of sodium sulphite (10, 25, 50 %) and no dose response could be observed. In addition, SIs of less than 1.1 (1.020 - 1.093) were observed for ear weight at all concentrations of sodium sulphite (10, 25, 50 %).

Thus, the classification criteria acc. to Directive 67/548/EEC and Regulation (EC) 1272/2008 as skin sensitizer are not met and sodium sulphite has not being labelled as such.

In the corresponding OECD SIDS (2001), a guinea pig maximisation assay yielded a sensitisation rate of 0/10 animals. Thus, the classification criteria acc. to Directive 67/548/EEC and Regulation (EC) 1272/2008as skin sensitizer are not met in this case either.

Reported human incidences of sulfite sensitivity have been reviewed extensively by various scientific and regulatory bodies in the EU, concluding on an absence of sensitising effects.

Respiratory sensitisation

There is no evidence on specific respiratory hypersensitivity in humans following repeated inhalation exposure with disodium disulfite. Furthermore, none of the repeated dose toxicity studies via inhalation reports respiratory hypersensitivity in the test animals. Thus, the classification criteria acc. to Directive 67/548/EEC and Regulation (EC) 1272/2008 as respiratory sensitizer are not met.