Hydrogen peroxide

Endpoint summary

Description of key information

In an oral toxicity study (Freeman 1997) 35 % hydrogen peroxide was applied to mice via the drinking water for 90 days. Reduced food and water consumption were seen at ≥ 300 ppm. Body weight was reduced in mice receiving 3000 ppm during most of treatment period in male animals. Further signs of treatment indicated the duodenum as target organ with local mucosal hyperplasia at ≥ 1000 ppm. Mucosal hyperplasia in the duodenum was not found in any dose group after recovery. The no observed adverse effect level (NOAEL) was 100 ppm (26 mg/kg bw/day in males and 37 mg/kg bw/day in females). A 28-days inhalation study in the rat was performed (Kilgour 2002). Local effects appeared in the nose with necrosis and inflammation at ≥10 ppm followed by respiratory irritation and reduced body weight gain in higher exposure concentrations. The no observed adverse effect level (NOAEL) was 2.9 mg/m³ (2.03 ppm). ). In a 90-day inhalation study (Staal 2014) with narrow spaced dosing no adverse effects were observed and the No Observed Adverse Effect Level (NOAEL) was determined to be 10.3 mg/m3. Treatment related effects were not observed in the respiratory tract of any treated animal.

Additional information

Oral repeated dose toxicity - non-human information

A reliable, well-conducted 90-day study with a catalase deficient strain of mice found a decrease in body weight at doses of 3000 ppm in drinking water (Freeman 1997). The study indicated that the NOAEL of hydrogen peroxide in drinking water was 100 ppm implying a daily dose of 26 mg/kg bw for males and 37 mg/kg bw for females. The LOAEL was 300 ppm (76 mg/kg bw for males, 103 mg/kg bw for females) based on dose-related reductions in food and water consumption and on the observation of duodenal mucosal hyperplasia in one male. Hyperplasia was a consistent finding at the higher levels of 1,000 and 3,000 ppm both in males and females (corresponding daily doses were 239 mg/kg for males, 328 mg/kg for females and 547 mg/kg for males, 785 mg/kg for females, respectively), and it was completely reversible in the recovery period. At the top dose (3,000 ppm) plasma total protein and globulin concentrations were reduced.

Inhalation repeated dose toxicity - non-human information

A reliable 28-day repeated dose inhalation toxicity study was performed with male and female Alpk:APfSD (Wistar-derived) rats exposed to hydrogen peroxide vapours for 6 hours per day, 5 days per week at concentrations of 2.03, 10.3 or 23.3 ppm (Kilgour 2002). The study was carried out under GLP conditions and in accordance with OECD Guideline No. 412. Treatment of a group exposed initially to 58.1 ppm and subsequently to 27.3 ppm was terminated before schedule due to the toxicity of the test material. Clinical observations were consistent with the material being a respiratory tract irritant (reddened noses, stains around the nose, abnormal respiratory noise) and in general the time to onset, incidence and severity of clinical signs increased with exposure concentration and repeated exposure. Males exposed to 23.3 ppm hydrogen peroxide showed lower food consumption and body weight gain compared to controls. Minimal changes in albumin and total protein blood levels were found in males and females exposed to 23.3 ppm. Histopathological, treatment-related changes were seen in the anterior-most regions of the nasal cavity lined with squamous epithelium, where minimal to slight necrosis (with associated inflammation) and rhinitis were seen in animals exposed to 10.3 and 23.3 ppm hydrogen peroxide. Inflammation and epithelial erosion in the larynx and increased perivascular neutrophil infiltration in the lungs were considered unlikely to be related to treatment in the absence of a clear dose response relationship. The no observed effect level (NOEL) for the study was considered to be 2.03 ppm hydrogen peroxide (corresponding to 2.9 mg/m3).

A reliable 90 -day inhalation study (Staal 2014) was performed in Wistar outbred rats unsing nose-only exposure. No treatment-related clinical abnormalities or ocular changes were seen. No significant treatment-related effects were seen on body weight or food consumption. At necropsy, no treatment-related effects were seen on any of the haematology or white blood cell parameters tested. Of the clinical chemistry parameters tested, a statistically significant increase was seen in alkaline phosphatase concentrations in male animals of the high concentration group, which was not considered an adverse effect when compared to historical control data. In addition, liver and thymus weights (both absolute and relative to body weight) of male animals of the high concentration group were statistical significantly decreased. However, this weight change was not accompanied by microscopic abnormalities in these organs and values were consistent with historical control data.No effect was seen on the weight of any of the other organs or tissues of male animals or any organ or tissue of the female animals. No treatment-related effects were seen at microscopic examination of any organ or tissue, including the respiratory tract of all exposed animals. Based on the absence of adverse effects in this study, it is concluded that the No-Observed-Adverse-Effect-Level (NOAEL) upon subchronic exposure for male and female animals is at the high concentration level (10 mg/ m3).

Inhalation repeated dose toxicity - human information

The following text on human information about repeated dose toxicity of hydrogen peroxide is copied from the EU risk assessment report (European Commission 2003, page 118-119):

“In line with endorsement by the Technical Meeting, the Finnish Institute of Occupational Health coordinated a worker health surveillance study in one company which concerned a small group of workers exposed to hydrogen peroxide vapours in aseptic packaging of fruit juices (Riihimäki et al. 2002). Fruit juice production in the plant had gradually started in the Spring of 1998, hence all the workers were engaged with hydrogen peroxide for 3 years or less. The company occupational health personnel was alerted in the Summer of 1999 by complaints among 6 operators/maintenance workers of two packaging machines situated at one end of the factory hall concerning irritation in the eyes and airways, headaches, temporary loss of olfaction, symptoms and signs in the skin, and blanching of hair. Peak exposures up to 11 mg/m3 (8-hour TWA 2-3 mg/m3) of H2O2 in air were measured in the breathing zone of the individuals, and intermittent skin contact ensued from amending breakdowns inside the machine. Workers who handled cartons inside the machine reported on burning and pricking of fingers, drying of the hands and face, decrease of skin elasticity, and dry, rough and bleached hair. At four other machines in the factory hall low peroxide levels were detected, and no complaints had emerged among the operators.

Measures were initiated to improve the situation and peroxide levels were monitored, however, it took several months until the targeted low levels (0.5-0.7 mg/m3) were reached in the spring of 2000. At that time symptoms were ascertained with a questionnaire indicating that every other person working with the two machines causing high exposure had experienced eye and airway irritation, and asthma symptoms. Clinical histories of respiratory illness were assessed from records of the company occupational health care unit and sick leave documents over time preceding and succeeding the reduction of H2O2 levels up until the spring of 2001. Two machine operators and one maintenance worker exhibited a uniform course of recurring bronchitissinusitis which coincided with a 10-month period of high concentrations. Two patients, described in detail as case reports, exhibited even bronchoconstriction and made a full recovery only after administration of inhaled corticosteroids and the concurrent reduction of exposure. The authors conclude that the results support the hypothesis that repeated exposures to high levels of hydrogen peroxide vapour induce sustained irritation and inflammation of the airway mucosa, increase susceptibility to respiratory infections, and may even cause irritant induced asthma. They note that there are clear similarities to the respiratory effects of oxygen (hyperoxia) and ozone which enhance the biological plausibility of the novel findings. However, as the study did not include specific examinations of the lungs, possible chronic lung changes by peroxide cannot be evaluated. Any remarkable effects were unlikely as the patients monitored in the study regained good health after the exposures were reduced. Moreover, the exposure duration was short for chronic effects to be manifested. From this limited study a LOAEL of 2 mg/m3 (8-hour TWA) for the repeated inhalation toxicity by hydrogen peroxide (airway effects) can be derived. As the exposure concentrations fluctuated markedly it is possible that the peak levels (in one machine up to 11.3 mg/m3, in the other machine up to 4.2 mg/m3) played a significant role in the induction of effects.”

Dermal repeated dose toxicity

A reliable study on dermal repeated dose is not available. Reliable studies on the repeated dose toxicity of hydrogen peroxide are available for the oral and inhalation routes. In addition, a low bioavailability of hydrogen peroxide through the skin is expected. In consideration of these points a repeated dose toxicity study via the dermal route is not necessary and an appropriated descriptor for this endpoint can be derived from oral and inhalation studies.

Justification for classification or non-classification

In an oral toxicity study (Freeman 1997), 35% hydrogen peroxide was applied to mice via the drinking water for 90 days. Reduced food and water consumption were seen at ≥300 ppm. Body weight was reduced in mice receiving 3000 ppm during most of treatment period in male animals. Further signs of treatment indicated that the duodenum was affected by local mucosal hyperplasia at ≥1000 ppm. Mucosal hyperplasia in the duodenum was not found in any dose group after recovery. The oral no observed adverse effect level (NOAEL) for repeated dose toxicity was 100 ppm (26 mg/kg bw/day in males and 37 mg/kg bw/day in females). A 28-day inhalation study in the rat was performed (Kilgour 2002). Local effects appeared in the nose with necrosis and inflammation at ≥10 ppm followed by respiratory irritation and reduced body weight gain in higher exposure concentrations. The inhalation no observed adverse effect level (NOEL) for repeated dose toxicity was 2.9 mg/m³ (2.03 ppm).

There are no data on chronic toxicity available. The result may be predicted reliably from the intrinsic properties of the chemical or from test results obtained in the available 28-day and 90-day studies. It has been shown that hydrogen peroxide is rapidly metabolised in the body to oxygen and water (European Commission 2003). The substance does not bio-accumulate in the body. Hence, it can be predicted that any effect in chronic studies will be similar to those observed in the available subchronic 90-day drinking water study of Freeman (1997) or the 28-day inhalation study (Kilgour 2002). It is therefore concluded that further chronic toxicity studies will not provide any additional useful information for the risk assessment of hydrogen peroxide and should therefore not be conducted due to animal welfare reasons.

Hydrogen peroxide should not be classified for systemic effects as it is not systemically available and as the substance exhibits only local irritancy.