Hydrogen peroxide

Endpoint summary

Description of key information

There is sufficient evidence from human exposure cases and animal studies that solutions exhibiting low hydrogen peroxide concentrations are not irritating to the skin and the eyes. Solutions at concentrations of equal to or greater than 5 % are irritating to the eyes, solutions at concentrations equal to or greater than 8 % can cause severe damage to eyes, solutions at concentrations equal to or greater than 35 % are irritating to the respiratory tract and the skin and solutions at concentrations of equal to or greater than 50 % are corrosive.

Additional information

Skin irritation - non-human information

Information on the skin irritation potential of hydrogen peroxide is provided by a series of animal studies investigating the irritation and corrosive potential of the substance. The irritating and corrosion properties of hydrogen peroxide to the skin and the eye varies with its concentration in aqueous solutions. Hydrogen peroxide solutions of 3% (Du Pont 1972), 6% (Du Pont 1973) and 8% (Du Pont 1974) caused mild reactions in the rabbit skin in spite of exposure under occlusion for 24 hours. The tested solutions were considered as not irritating to the skin.

The irritation potential of a 10% aqueous solution of hydrogen peroxide was tested on three male and three female New Zealand White rabbits according to OECD Guideline No. 404, 1981. After dermal administration of the test material the application sites were covered in a semi-occlusive manner for 4 hours. Test animals were fitted with Elizabethan collars throughout the study. Effects on the skin were assessed according to the method of Draize. Slight erythema occurred in two animals 24 hours post-dosing. The individual erythema scores did not exceed one. All individual oedema scores were zero throughout the study. All irritation resolved within 48 hours and the test was terminated after 72 hours. Based on the test results the tested 10% w/w solution of hydrogen peroxide is considered to be not irritating to the rabbit skin.

The skin irritation potential of a 35% aqueous solution of hydrogen peroxide was tested in six New Zealand White rabbits according to US EPA Guideline PB82-232984, August 1982, and under GLP conditions (FMC 1983). Additionally, the corrosion to the skin was tested. 0.5 mL of the undiluted substance was administered to intact and healthy skin, and test sites were held under occlusive conditions for 4 hours. The effects on the skin were then observed for 14 days and scored according to the method of Draize. The scores for erythema/eschar were not higher than two for individual animals at any timepoint and the mean scores were 1, 1.75, 0.58 and 0.58 at 4, 24, 48 and 72 hours. The scores for oedema were not higher than two for individual animals at any timepoint and the mean scores were 1.83, 0.83, 0 and 0 at 4, 24, 48 and 72 hours. After the observation period of 14 days two animals exhibited brown areas with desquamation at the sites of treatment. The test material was judged to be moderately irritating to the rabbit's skin. The test material was non-corrosive within 48 hours of dosing.

A test on primary skin irritation was carried out with a 49.2% aqueous solution of hydrogen peroxide according to guidelines equivalent to OECD Guideline No. 404 (FMC 1990b). One female White New Zealand rabbit received a single dose of 0.5 mL of undiluted test material on two sites, which were held under semi-occlusive conditions for four hours. The animal was anaesthetised in an effort to minimise pain due to the administration of the test material. Skin reaction was scored according to Draize after 4.5, 24 and 48 hours. After the last examination, the test was terminated and samples of treated and untreated skin were saved for inspection by a pathologist. Skin irritation scores and subsequent pathologic examination revealed a severe irritation of the skin by the test material that would have resulted in ulceration and necrosis. Thus, it can be concluded that a 49.2% solution of hydrogen peroxide is highly irritating to the rabbit's skin.

Skin irritation - human information

The following text on human information about skin irritation is copied from the EU risk assessment report for hydrogen peroxide (European Commission 2003, page 110):

“A retrospective review of all exposures reported to the Utah Poison Control Center over a 36-month period found that 325 cases (0.34%) were due to hydrogen peroxide. Ingestion was the most common route of exposure accounting for 83% of all exposures. The next most common routes of exposure were ocular and dermal accounting for 8.0% and 7.7% of cases, respectively. The three chief dermatologic findings were paresthesias (60%), whiteness (56%), and blistering (16%). The most common skin and ocular exposure outcome was a minor, transient effect. There were no permanent sequelae from these exposures (Dickson and Caravati 1994). A group of 32 volunteers (18 men and 14 women, 23-37 years of age) exposed one hand to hydrogen peroxide vapour at variable concentrations and for variable durations. The threshold for skin irritant action was determined (the method is not disclosed). One hand of the subjects was placed inside an exposure chamber through an opening in a rubber membrane, the other hand served as a control. Immediately following the exposure the skin was washed, and the washings were analyzed (in both the exposed and the control area) to determine the deposition of H2O2. The threshold concentrations for skin irritation (apparent LOAELs) depended on the exposure time and were as follows: 20 mg/m3 for 4-hour exposure, 80 mg/m3 for 1 hour, 110 mg/m3 for 30 min, 140 mg/m3 for 15 min, and 180 mg/m3 for 5-min exposure. The measured deposition of hydrogen peroxide at the threshold level exposures ranged 1.1-1.7 mg/dm2, deposition ranging 0.5-0.8 mg/dm2 was found to be ineffective (Kondrashov 1977).

Eye irritation - non human information

The EU Risk Assessment Report for hydrogen peroxide (European Commission 2003) mentions that solutions containing 1-3% hydrogen peroxide were used topically as an on-eye antibacterial agent historically without causing significant injury. In a series of reliable animal studies, no ocular irritation was noted after application of a 3% solution of hydrogen peroxide. Testing of eye irritation with a 5% solution indicated a slight irritation when the Draize method was used to assess effects (FMC 1987a). An 8% solution was moderately irritating (FMC 1987b) and a 10% solution was highly irritating to the rabbit eye (FMC 1985). A 35% hydrogen peroxide solution was corrosive to the rabbit eye (FMC 1983).

Eye irritation - human information

The following text on human information about eye irritation is copied from the EU risk assessment report for hydrogen peroxide (European Commission 2003, page 112-113):

“In the retrospective review of all exposures reported to the Utah Poison Control Center over a 36-month period, 8% of the reports involving hydrogen peroxide concerned eye as the route of exposure. The most frequently encountered ocular symptoms included burning (65%), redness (50%), and blurry vision (19%). The most common ocular exposure outcome was a minor, transient effect and there were no permanent sequelae (Dickson and Caravati 1994).

Historically 1 to 3% hydrogen peroxide (10,000-30,000 ppm) has been used topically as an on-eye antibacterial agent 3 to 5 times per day without causing significant injury (Grant 1993). The effect of hydrogen peroxide is dose related with 0.5% (5,000 ppm) being used previously for treatment whereas 5 and 10% (50,000 and 100,000 ppm) are definitely known to cause cloudiness in the cornea, severe pain, and intraocular inflammation (Chalmers 1989). A soft contact lens which had been stored in 3% hydrogen peroxide was placed in the eye of a 30-year-old woman without the appropriate catalase neutralisation. She had an immediate painful reaction with hyperaemia, tearing and eyelid spasm. The lens was removed at once, and the eye was treated with a topical anaesthetic. During the next 48 hours the eye became increasingly inflamed, despite the use of dexamethasone drops and the cornea began to show punctate staining. Thereafter the cornea began to clear, visual acuity was restored, and after several days there were no residual effects other than punctate keratopathy and mild discomfort (Knopf 1984).

A controlled, randomised, double-blind study with eight human subjects was conducted to determine the threshold level of eye effects by hydrogen peroxide via a high water content hydrogel contact lens. Subjective comfort, conjunctival hyperaemia, corneal and conjunctival epithelial staining, and corneal oxygen uptake were assessed in response to 5-min wear of lenses that were presoaked in isotonic saline solutions containing 0, 25, 50, 100, 200, 400 or 800 ppm hydrogen peroxide. Higher levels of H2O2 (≥200 ppm) were associated with greater discomfort and increased conjunctival hyperaemia. The highest level (800 ppm) of H2O2 did not induce significant corneal or conjunctival epithelial staining or alter the corneal aerobic response (Paugh et al. 1988). In a single-masked, controlled study the eyes of 10 volunteers were exposed to drops of hydrogen peroxide solution or to contact lenses soaked for 2-4 hours in hydrogen peroxide solutions. The mean detection threshold for drops of dilute H2O2 was 812 ppm (range 400-1,500 ppm). The mean threshold was 267 ppm for 55% water lenses and 282 ppm for 38% water lenses. Removal of 50 ppm H2O2 from a hydrogel lens was completed within 30 seconds of human wear when the eyelids were held closed, and within 60 seconds during wide-open gaze, with blinking every 5 seconds. The corneal permeability to fluorescein was determined in 10 subjects after dosing with 50 ppm H2O2, 500 ppm H2O2, as well as negative and positive controls. There was no significant difference between the negative control and the two H2O2 concentrations. Hydrogen peroxide disinfection systems are designed to have residual H2O2 concentrations in the lens of no more than 50-60 ppm (Mc Nally 1990).”

Corrosivity - non human information

The potential of hydrogen peroxide to induce skin corrosion in rabbits was tested with 50% and 70% solutions in a GLP conform study (DuPont 1994). Sites of shaved skin of New Zealand White rabbits were exposed to 50% solution of hydrogen peroxide for 3 minutes, 1 hour or 4 hours or to 70% solution for 3 minutes and covered with a semi-occlusive dressing. The skin reaction on exposure sites were evaluated using the Draize method upon removal of the test material and subsequently after 24, 48 and 72 hours and 7 and 14 days. Exposure to 50% solution for 3 minutes caused moderate dermal irritation that was fully reversible within the observation period of 14 days. On the other hand, exposure to 50 % solution for 1 and 4 hours and to 70% solution for 3 minutes caused severe skin irritation and resulted in the formation of scar tissue after 14 days. The two test solutions containing 50% and 70% hydrogen peroxide were considered as corrosive under these conditions of exposure.

Solutions containing 35% hydrogen peroxide were corrosive to the eye of rabbits in a guideline-conform eye irritation test (FMC 1983).

Effect level: empty Endpoint conclusion: Adverse effect observed

Effect level: empty Endpoint conclusion: Adverse effect observed

Effect level: empty Endpoint conclusion: Adverse effect observed

Justification for classification or non-classification

The irritating properties of hydrogen peroxide to the skin and the eye varies with its concentration: No dermal irritation was noted after application of 10% hydrogen peroxide. The application of 35% hydrogen peroxide causes slight to moderate, transient irritation and the current classification as "Skin irritant category 2" (H315) according to Regulation (EC) No 1272/2008 is supported by this finding. Hydrogen peroxide solutions ranging from 50 to 70% are corrosive to the skin and fall into the "Skin corrosion category 1B" (H314). Solutions containing more than 70 % are very corrosive to the skin and fall into the "Skin corrosion category 1A" (H314). No ocular irritation was noted after application of 3% hydrogen peroxide. The 5% and 6% “hydrogen peroxide” solutions induced signs of eye irritation which support the classification as "Eye irritant category 2" (H319) according to Regulation (EC) No 1272/2008. The effects of 8% and 10% hydrogen peroxide to the rabbit eye were severe and justified current classification with "Eye damage category 1" (H318). The evocation of respiratory irritancy in the experiments with 50% solutions of hydrogen peroxide justify the classification with “Specific target organ toxicity, single exposure 3” (H335).