Chlorine

Endpoint summary

Description of key information

Oral:
No studies for oral toxicity are available for chlorine. Chlorine is handled in closed systems exclusively and an exposure to the gas is only accidentally. The only possible exposure is via inhalation. Repeated dose studies with sodium hypochlorite were conducted. Thus a read across is done from sodium hypochlorite. 
Study 1 (Hasegawa et al., 1986):
There were no mortalities throughout a 90 day study with rats. Body weight gain was statistically significantly reduced in males at 100 and 120 mg av Cl/kg bw/day dose level groups and in females at 228.8 mg av Cl/kg bw/day dose level group; there were no obvious macroscopic or histological changes in any group. A NOAEL of 50 mg/kg bw/day was identified.
Study 2 (Daniel et al., 1990):
In a 90 day study with rats up to concentrations of 16.7 mg Cl/kg bw/day, no significant differences between treated and control groups apart from sporadic deviations in haematological and clinical chemistry parameters, which were considered not to be treatment related.
Study 3 (Daniel et al., 1991):
In general, treated mice (90 days up to concentrations of 34 mg Cl/kg bw/day) exhibited a decreased water consumption and a decreased weight gain. Haematology, clinical chemistry and necropsy revealed some alterations in mice that received chlorinated water: animals of the highest concentration groups had lower enzyme activities consistent with lower liver weights. Overall, the effects are considered to be mild, non-specific, transient and secondary to e.g. nutritional deficiencies due to taste aversion of the drinking water.
Inhalation:
No systemic effects were observed in repeated dose exposure studies in rats, mice and monkeys with chlorine gas. Additionally, chlorine was discussed by SCOEL and an OEL of 0.5 ppm (1.5 mg/m3) was agreed based on these studies, with removal of the 8-hour TWA. The justification was that the effects appear to be related to concentration in the air and not to duration of exposure.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion

Dose descriptor:

NOAEL

50 mg/kg bw/day

Study duration:

subchronic

Species:

rat

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion

Dose descriptor:

NOAEC

1.5 mg/m³

Additional information

Chlorine gas is exclusively handled in closed systems and any exposure to the gas will be accidental and will be by inhalation. In aqueous solution as sodium or calcium hypochlorite at pH values in the range 6 -8, chlorine will form hypochlorus acid in equilibrium with the hypochlorite anion. Therefore, all studies investigating aqueous solutions of sodium or calcium hypochlorite can be used for evaluation and assessment of chlorine. Several repeated oral administration studies have been conducted with sodium hypochlorite and are considered to be relevant for chlorine and allow a scientific valid evaluation of systemic toxicity of chlorine after repeated applications. Futhermore studies with chlorine gas are available to address repeated dose toxicity via the inhalation route of exposure.

Oral administration of Sodium hypochlorite

In a 28-day short term oral (feeding) study (Anonymous, 1962, Doc. No. 581-001) groups of 10 male albino rats were treated with doses of 0, 30, 2500 and 7500 ppm hypochlorite (corresponding to 0, 2.66, 221, 683 mg/kg bw/d) dissolved in corn oil. No mortalities or clinical signs and no significant pathological findings apart from increased adrenal weights were observed. As these were not accompanied by other findings they are not considered to be treatment-related effects and therefore the NOAEL is considered to be ≥ 683 mg/kg bw/d.

A 90-day drinking water study (Hasegawa et al, 1986, Doc. No. 592-096) was performed in Fischer 344 rats (10 animals/sex/dose group) with 0, 0.025, 0.05, 0.1, 0.2 and 0.4 % of hypochlorite in water (corresponding to 0, 12.5, 25, 50, 100 and 200 mg/kg bw/d in males and 0, 14.3, 28.6, 57.2, 114.4 and 228.8 mg/kg bw/d in females). There were no mortalities throughout the study. Body weight gain was statistically significantly reduced in males of the 0.2 and 0.4 % groups and in females of the 0.4 % group. At autopsy there were no obvious macroscopic changes in any group, although several animals, particularly in the high dose, appeared emaciated. Absolute weights of the lung, liver and spleen of males and the salivary gland, lung, heart and brain of females were significantly lower in the highest-dose groups than in controls. No histological changes attributable to sodium hypochlorite administration were found in any of the experimental groups. Biochemical parameters of the sera showed signs of slight damage to the liver in the two highest dose groups in both sexes at this interim kill but these effects were not observed at termination of the study and therefore not regarded as relevant. The NOAEL (based on body weight reductions at the two highest dose levels) was 50 mg av Cl/kg bw/d (calculated with 0.1% solution, a water consumption of 25 ml and a body weight of 0.5 kg) in males and 57.2 mg av (available) Cl/kg bw/d in females (body weight 0.35 kg).

Another 90-day drinking water study (Daniel et. Al., 1990, Doc. No. 592-139) was performed in Sprague-Dawley rats (10 animals/sex/dose group) with 0, 25, 100, 175, 250 mg av Cl/L (test item intake: 0, 2.1, 7.5, 12.8, 16.7 mg/kg bw/d for males; 0, 3.5, 12.6, 19.5, 24.9 mg/kg bw/d for females). There were no significant differences between treated and control groups apart from sporadic deviations in haematological and clinical chemistry parameters, which were considered not to be treatment related. The NOAEL was considered to be greater than 250 mg av Cl/L (corresponding to 16.7 mg av Cl/kg bw/d for males; 24.9 mg av Cl/kg bw/d for females).

In the 90-day drinking water study (Daniel et. al, 1991, Doc. 592-144) in B6C3F1 mice (10 animals/sex/dose group) a mixture of hypochlorite and monochloramine was administered in doses of 0, 12.5, 25, 50, 100, 200 mg av Cl/L (calculated test item intake: 0, 2.7, 5.1, 10.3, 19.8, 34.4 mg/kg bw/d for males and 0, 2.8, 5.8, 11.7, 21.2, 39.2 mg/kg bw/d for females). Analysis of the test substance concentrations and measurement of the water consumption showed that the achieved intake in the highest dose was 35-40 mg/kg bw/d. In general, treated mice exhibited a decreased water consumption and a decreased weight gain. Haematology, clinical chemistry and necropsy revealed some alterations in mice that received chlorinated water. Animals of the highest concentration groups had lower enzyme activities consistently lower liver weights. Overall, the effects are considered to be mild, non-specific, transient and secondary to e.g. nutritional deficiencies due to taste aversion of the drinking water. Therefore, the NOAEL is considered to be greater than 34.4. mg av Cl/kg bw/d for males and 39.2 mg av Cl/kg bw/d for females.

For endpoint summary records of chronic toxicity/cancerogenicity tests see section 7.7:

In the chronic toxicity and carcinogenicity 104-week study (Hasegawa 1986, Doc.No.592-096) with additional 8 weeks recovery in Fischer 344 rats (50 animals/sex/dose group) hypochlorite was administered (via drinking water) at doses of 0, 0.05, 0.1 % for males (corresponding to 0, 25 and 50 mg ac Cl /kg bw/d) and 0, 0.1, 0.2 % for females (corresponding to 0, 57.2 and 114.4 mg av Cl/kg bw/d). Mean body weights of the treated rats showed a dose-related reduction compared with the control values. At the end of the treatment period body weights in the low- and high dose group were 96 and 93 % of control body weight for males and 89 and 80 % of control body weight for females. After termination of test substance administration the rats showed body-weight gains of 1-9 % in week 8 of the post-application period, the recovery being most marked in the high-dosed animals of both sexes. For males of the highest dose drinking water intake in the last 20 weeks was higher than in controls and for females in the high dose the drinking water intake was somewhat lower during the first year. 62 rats died during the period of administration of sodium hypochlorite, another 49 animals died during the 8 weeks recovery period. Survival was similar in all groups including control. Absolute liver weights were lower in all treated groups; however, relative liver weights were similar in all groups of both sexes. The other changes in absolute or relative organ weights did not show a clear dose response and are therefore considered to be of no biological significance. No significant changes due to the treatment were observed in the blood analyses. There were no treatment related increases in non-neoplastic lesions or tumour incidence. Hypochlorite is considered not to be carcinogenic in the rat. The NOAEL is derived at 0.1% hypochlorite (corresponding to 50mg av Cl/kg bw/d for males) based on reduced body weight in females in the highest dose group (0.2 %, corresponding to 114.4 mg av Cl/kg bw/d). Males were not tested at this concentration.

In another chronic toxicity/carcinogenicity study (NTP, 1992, Doc. No. 592-147) Fischer 344/N rats (70 animals/sex/dose group) were dosed at 0, 70, 140, 275 ppm hypochlorite in drinking water (corresponding to 0, 3.5, 7, and 13.75 av Cl mg/kg bw/d in males and 4, 8, and 15.7 mg av Cl/kg bw/d in females) for 103-104 weeks. Palatability was the principal factor limiting the concentrations of available chlorine in the study. There was a dose-related decrease in water consumption by animals receiving chlorinated water. Decreased water consumption was evident during the first week and continued throughout the study. Toward the end of the studies, the effect on water consumption was generally less than during the first weeks. The animals showed no physiological alterations due to decreased water consumption, and there was no clinical or haematological evidence of dehydration. Because body weight and water consumption changed as the rats aged, the amount of available chlorine ingested during the study varied. The mean daily dose (mg/kg body weight) was higher during the first 13 weeks than during the second year of the studies. High-dose rats received a mean daily dose of approximately 20 mg/kg for the first 13 weeks, which decreased to 13-14 mg/kg during the second year. Survival of rats was similar among treated groups and their respective controls. Survival of all groups of male rats was less than 50 % at the end of the studies, following a trend previously reported for all NTP studies (Rao, et al. 1990: Growth, body weight, survival, and tumour trends in F344/N rats during an eleven year period. Toxicol. Pathol. 18, 61-70.) There were no treatment-related lesions in rats at the 14-week or at the 66-week interim evaluations. There were no neoplasms or non-neoplastic lesions that were clearly attributable to the consumption of chlorinated water. However, the incidence of mononuclear cell leukaemia was marginally greater than that of controls in the mid- and high-dose groups of female rats. However, the increase in leukaemia in dosed female rats was only slight and not clearly dose related, there was no decrease in tumour latency and the incidence in concurrent controls was less than in historical controls. It is therefore concluded that the slightly increased incidences of leukaemia in treated female rats is not treatment related. Under the conditions of this 2-year drinking water study, there was no evidence of carcinogenic activity of chlorinated water in F344/N rats receiving 70, 140, or 275 ppm. The applied chlorine concentrations were well tolerated, there were no treatment related clinical signs, mortalities, haematological or histopathological findings. The NOAEL of this study was greater than 275 ppm available chlorine (corresponding to 13.75 mg av Cl /kg bw/d for males and 15.7 mg av Cl/kg bw/d for females).

Sodium hypochlorite was administered at doses of 0, 70, 140, 275 ppm (corresponding to 0, 11.7, 23.3, and 45.8 mg av Cl/kg bw/d in males and 0, 14, 28, and 55 mg av Cl/kg bw/d in females) via drinking water to groups of 70 B6C3F1 mice/sex/dose for 103-104 weeks (NTP, 1992, Doc. No. 592-147). Palatability was the principal factor limiting the concentrations of available chlorine in the study. There was a dose-related decrease in water consumption by animals receiving chlorinated water. Decreased water consumption was evident during the first week and continued throughout the study. Towards the end of the studies, the effect on water consumption was generally less than during the first weeks. The animals showed no physiological alterations due to decreased water consumption, and there was no clinical or haematological evidence of dehydration. Because body weight and water consumption changed as the mice aged, the amount of available chlorine ingested during the study varied. The mean daily dose (mg/kg body weight) was higher during the first 13 weeks than during the second year of the studies. High-dose mice received a mean daily dose of approximately 35-44 mg/kg for the first 13 weeks, which decreased to 20-23 mg/kg during the second year. Survival was similar among treated groups and their respective controls. There were no treatment-related lesions in mice at the 15-week or at the 66-week interim evaluations. There were no neoplasms or non-neoplastic lesions that were clearly attributable to the consumption of chlorinated water. Sporadically renal neoplasms occurred in the low and high dose males. This is an unusual finding in mice. Therefore, additional step sections of the kidney were prepared which revealed further incidences of renal hyperplasia in all groups including control and a carcinoma in the low dose group. Nearly all the additional neoplasms seen in the step sections were small (microscopic) adenomas believed to be the probable precursor of renal tubule carcinoma. Since no additional renal neoplasms were found in the mid and high-dose groups and since focal hyperplasia, a potential preneoplastic lesion, was found at similar incidences in the control and dosed groups, the small number of renal tubule cell neoplasms in male mice were not considered related to the consumption of chlorinated water. Under the conditions of this 2-year drinking water study, there was no evidence of carcinogenic activity of chlorinated water in male or female B6C3F1 mice receiving 70, 140, or 275 ppm. The applied available chlorine concentrations were well tolerated, there were no treatment related clinical signs, mortalities, haematological or histopathological findings. Based on the results of this chronic toxicity/carcinogenicity study with sodium hypochlorite, the NOAEL was derived greater than 275 ppm available chlorine (corresponding to 45.8 mg av Cl/kg bw/d for males and 55 mg av Cl/kg bw/d for females).

In a carcinogenicity study (Soffritti, 1997, Doc. No. 592-003) four groups of 50 male and 50 female Sprague-Dawley rats received drinking water containing sodium hypochlorite, resulting in concentrations of available chlorine of 0, 100, 500 and 750 ppm (corresponding to 5, 25 and 37.5 mg av Cl/kg bw/d for males and 5.7, 28.6 and 42.9 mg av Cl/kg bw/d for females) for 104 weeks. Mean daily water consumption was decreased in a dose-related manner. A slight decrease in mean body weight was observed in high-dose animals, especially males. However, this effect was not considered to be adverse. Males of the lowest dose had a slightly higher survival rate than all other animals. The observed increased incidence in malignant tumours was not dose-related. Higher incidences might be due to longer survival rates in low-dose male animals. All treated females showed an increase in lymphomas and leukaemias (8 to 12 %) which are within historical control data of 14.67 % (refer to “Compilation of Spontaneous Neoplastic Lesions and Survival in Crl:CD(R) (SD) Rats from Control Groups” 2004, Charles River Laboratories). Male rats in the high- and low dose but not mid-dose group, showed an increase in tumours of the stomach. 3 low dosed females developed adenomas in the lung. The higher incidence in leukaemias might be attributable to an unusually low number of cases in the control group. Depending on the concentration sodium hypochlorite is an irritating/corrosive substance. It is well known that gastric and pulmonary tumours are likely to develop when irritating substances are administered and taken up via the digestive tract and possibly aspirated at times into the lungs. Moreover, in a combined long-term toxicity and carcinogenicity study by the NTP it was shown that there is no evidence for a carcinogenic effect of chlorinated drinking water. Since no treatment related adverse effects are reported the NOAEL is greater than 37.5 mg/kg bw/d.

In a second carcinogenicity study (Kurokawa et al. 1986, Doc.No. 592-093) three dose groups of 50 male and 50 female F344 rats received drinking water containing sodium hypochlorite for 104 weeks, the doses were 0 ppm (male/female), 500 and 1000 ppm (male), 1000 and 1000 ppm (female), corresponding to 25 and 50 mg av Cl/kg bw/d for males and 57.1 and 114.3 mg av Cl/kg bw/d for females. Additionally three dose groups of 50 male and 50 female B6C3F1 mice (73 males and 72 females in case of control) received drinking water containing sodium hypochlorite for 103 weeks, the doses were 0 ppm (male/female), 500 and 1000 ppm (male/female), corresponding to 83.3 and 166.7 mg av Cl/kg bw/d for males and 100 and 200 mg av Cl/kg bw/d for females. Dose dependent inhibition of body weight increase was observed in both male and female rats/mice. This effect was not considered to be adverse. Haematological changes were not observed in rats and mice of both sexes given sodium hypochlorite. No evidence of carcinogenicity was apparent for sodium hypochlorite in both rats and mice of either sex. Since no treatment related adverse effects are reported the NOAEL is greater than 50 mg av Cl/kg bw/d.

The chronic toxicity of sodium hypochlorite has been investigated in male and female Fischer 344 rats and B6C3F1 mice following administration via the drinking water. With the exception of body weight effects, no systemic effects or morphological changes on microscopic examination could be observed.

Since the constituents of sodium hypochlorite, i.e. sodium, chloride and hydroxide, are physiologically essential elements and needed for the functionality of the living cell of all organisms irrespective of species, no further studies are required. Furthermore, sodium hypochlorite is a substance which will dissociate into (sodium), chloride and hydroxide ions when getting in contact with water and which will inevitably be the case when taken up by the body. As a consequence thereof, for an assessment of the systemic effects of sodium hypochlorite, the toxicological effects of the ions such as sodium and chloride need to be investigated rather than that of the irritating salt itself. Owing to the irritating nature of sodium hypochlorite, the primary effects are characterised by local irritation at the “port of entry” and all other effects which might be observable will be of secondary nature.

Taking into account that it is (sodium) chloride which will be available systemically since sodium hypochlorite decomposes rapidly after contact with “port of entry” only the toxicological effects of (sodium) chloride would have to be investigated. Sodium chloride is a broadly distributed nutrient from which no toxicological hazard arises considering the amounts which will be formed after uptake of sodium hypochlorite exposure during disinfection.

Dermal administration of Sodium hypochlorite

The effect of sodium hypochlorite on epidermal hyperplasia was studied in female SENCAR mice exposed cutaneously to sodium hypochlorite at concentrations of 1000 mg/l for 10 minutes/day for four days. Whole body exposure, except the head, was performed using specially designed Plexiglass chambers to prevent inhalation of any vapours or aerosols. Concentrations of 1000 mg/l sodium hypochlorite caused increases in epidermal thickness when exposed for four days. In a similar preliminary study using 1000 mg/l HClO with 10 minutes exposure for 8 consecutive days, the maximal response was observed after four daily 10 minute exposures. No effects were observed following treatment with concentrations of 1, 10 or 100 mg/l hypochlorous acid for four days. Significantly increased numbers of both total and basal epithelial cells in the skin were observed following four daily 10-minute exposures to 300 mg/l hypochlorous acid and above (Robinson, 1986). The results of this study suggest that the threshold concentration for the local skin irritant effects of sodium hypochlorite (as hypochlorous acid) is 300 mg/l, and that the effects were seen following repeated exposure to 0.1% sodium hypochlorite solution. Such local effects are dependant on the concentration of the applied irritant and not on the total dose. The study only reported the effects of a single concentration of sodium hypochlorite solution, there were inconsistencies on the reported effects of different concentrations of hypochlorous acid, suggesting that the findings of the study may be unreliable. It is noted that the non-treated control mice showed an unexplained reduction in skin thickness that will also have impacted on the outcome of the study.

Sodium hypochlorite was tested for promoting and complete carcinogenic activities in a skin carcinogenesis model using groups of 20 female SENCAR mice. A sodium hypochlorite solution (1% conc.) was applied twice a week for 51 weeks with or without initiation with dimethylbenzanthracene. Analysis of the number of skin tumours, squamous cell carcinoma and epidermal hyperplasia was performed. No animal died and no epidermal hyperplasia was observed in the group treated with sodium hypochlorite alone (Kurokawa, 1984, see carcinogenicity studies).

The effects of the repeated administration of sodium hypochlorite solutions on the skin has been studied in the guinea-pig. 0.1% and 0.5% solutions of sodium hypochlorite (950 and 4765 mg/l as available chlorine - prepared daily by diluting a proprietary household bleach (Clorox), which was a 5.25% solution of sodium hypochlorite) were applied to female guinea pig skin for up to 14 days, using gauze bandages soaked at 8 hour intervals with the solutions. The pH of the freshly prepared solutions were 7.4 and 9.65 for the 0.1% and 0.5% solutions respectively. Toxicity was assessed in terms of basal epidermal cell viability (trypan blue exclusion), the growth of single-cell cultures of epidermal cells to confluence in-vitro, and by histopathological examination. Tissue taken from areas of treated skin was compared to control adjacent tissue taken from the same animals. Basal epidermal cell viability was reduced from 85% to 65% in skin treated with the 0.5% solution for 14 days, but not for 1, 4 or 7 days and was not affected by treatment with the 0.1% solution. Similarly, the growth of epidermal cells in-vitro was affected only after treatment for 14 days with the 0.5% solution. Marked epidermal hyperplasia with an influx of inflammatory cells into the papillary dermis was observed in guinea-pigs treated with the 0.1% solution for 14 days, but not 1, 4 or 7 days and with the 0.5% solution for either 7 or 14 days, but not 1 or 4 days (Cotter et al., 1985). These results suggest that the LOAEC (expressed as a concentration) for the local irritant effects following repeated administration of this proprietary bleach solution for 14 days is 0.1%. It is noted that the test solutions were dilutions of proprietary bleach, which may have contained components in addition to sodium hypochlorite and that no adequate control was used (eg. a solution of equal osmolarity and/or pH).

A 0.1 ml solution of 0.125% (1190 mg/l av. chlorine) sodium hypochlorite was applied daily to guinea pig skin of the dorsal site of the ear for 1, 2, 4 and 8 weeks using 10 animals per group. No effect on epidermal proliferation, development and differentiation was observed. In the same study, sodium hypochlorite solution was blended with milk to assess the effects due to the formation of chloramines: no adverse effects were reported (Wohlab and Wozniak, 1982)

The immunotoxic potential of sodium hypochlorite was assessed in male Sprague-Dawley rats from weaning to 12 weeks of age given drinking water containing 0, 5, 15 or 30 mg/l sodium hypochlorite for 9 weeks. No significant effects were observed on body weights, thymus weight, antibody response, natural killer cell cytotoxicity, interleukin production and phagocytic activity. Reductions in spleen weight and delayed-type hypersensitivity reactions were observed in the rats given 30 mg/l sodium hypochlorite (about 1.5 mg/kg bw/day), while in those given 15 and 30 mg/l, a reduction in macrophage oxidative metabolism and a statistically significant increase in prostaglandin E2 production were observed. No information on the reversibility of the effects was provided (Exon, 1987).

Effects on delayed-type hypersensitivity, hemagglutination titres and reticuloendothelial clearance were studied in two groups of 30 CR1: CD-1 mice each given sodium hypochlorite in their drinking water at concentrations of 15 and 30 mg/l as available chlorine for 120 days. Other groups of mice were exposed to distilled water, tap water or other water treatment chemicals. No significant differences in antibody titres, reticulo-endothelial clearance or spleen weight were observed in the treated groups compared with the control group, suggesting that sodium hypochlorite does not significantly affect the immune function of mice (Hermann et al., 1982).

In a pharmacodynamics and toxicity study, groups of 4 Sprague-Dawley rats each were exposed to HClO as a single dose of 10, 20 and 40 mg/l (30, 60 and 120 μg, respectively) to control the rat blood glutathione after 15, 30, 60 and 120 minutes after the treatment. Other groups of 4 Sprague-Dawley rats, including a control group, were each treated daily for a period of 1 year by drinking water with 1, 10 and 100 mg/l of hypochlorous acid. Data collected were: the heparinized blood, collected at 3 and 6 months after HClO administration to control hematological parameters, chloroform content (at month 4, 6, 9 and 12), blood glutathione,and 3H-thymidine incorporation (both at month 2, 3, 4, 6, 8, 10 and 12). For the acute exposure a maximum decrease in blood GSH was observed 60 min after dosing. In the long term study a decrease of blood glutathione and increase of osmotic fragility after 6 months were observed. In animals exposed for 3 months, the 3H-thymidine incorporation into nuclei of rat kidney and testis in the 100mg/l group was increased. There was no dose-related effect and no information on the reversibility after treatment, no significant changes were observed relating the hematological parameters. Blood chloroform levels were without change during 1 year of treatment with hypoclorous acid (Abdel-Rahman and Suh, 1984).

A group of five C57BL/6N mice were given sodium hypochlorite in their drinking water at a concentration of 25-30 mg/l available chlorine for 4 weeks. A second group of mice served as controls. A decrease in the number of peritoneal exudate cells (suggestive of an effect on macrophage function) along with an overall depression of macrophage function was observed (Fidler, 1977).

Administration of chlorine via inhalation

Four groups of ten male and ten female Fischer 344 rats each were exposed to chlorine (Barrow 1979 Doc. No. 592-052) for a period of 6 hours per day for 6 weeks (exposure on weekdays only). Rats were exposed to nominal concentrations of 0, 1, 3 and 9 ppm chlorine corresponding to analytical concentrations of 1.00±0.04, 3.00±0.1 and 8.99±0.45 ppm chlorine. At the end of the treatment period, all animals were fasted overnight, and killed 1 (males) or 2 (females) days after the last day of exposure. No MMAD and GSD were determined in this study and particle size measurements were not possible to perform since chlorine is a gas. The results of this study indicated that unequivocal upper and lower respiratory tract changes were produced in Fischer 344 rats exposed to 9 ppm of chlorine. This conclusion was supported by the clinical signs of generalised respiratory tract irritation, increased lung weights and lung to body weight ratios, and the histopathological changes which extended throughout the upper and lower respiratory tract of rats exposed to 9 ppm. The increases in the total number of segmented neutrophils at 9 ppm correlated very well with the inflammation and mucopurulent exudate of the upper and lower respiratory tract. The respiratory tract effects found in animals exposed to 3 or 1 ppm chlorine were very similar and much less severe than those seen at 9 ppm. Concentration-dependent elevations were seen in urine specific gravity at all exposure concentrations in females and 9 or 3 ppm in males. Blood urea nitrogen was raised in both sexes exposed to 9 ppm . The kidneys of both sexes were darkened in appearance and the histopathological examination of males exposed to 9 ppm showed swelling of the epithelial cells of the proximal convoluted tubules with increased eosinophilic cytoplasmic homogeneity and decreased granularity. The presence of renal effects was also supported by clinical observation of urinary staining and matting of the fur around the genitalia in both sexes exposed to 9 or 3 ppm chlorine. Evidence of hepatic effects were seen from elevations of the activity of several enzymes. A concentration-related increase in alkaline phosphates was seen in rats of both sexes exposed to 9 or 3 ppm. Additionally, γ-glutamyl transpeptidase was elevated in both sexes exposed to 9 ppm and serum glutamic pyruvic transaminase was elevated in females at 9 ppm. Animals exposed to 9 ppm showed an increased hepatocellular cytoplasmic eosinophilic homogeneity. Furthermore, an increase in the degree of hepatocellular cytoplasmic vaculation was seen at 9 or 3 ppm. The spleen and thymus of rats exposed to 9 ppm showed a decreased content of lymphoid elements. This finding correlated with the statistically significant decreases in absolute and relative weights of these two organs at 9 ppm which may have been a function of the poor physical condition and decreased nutritional fate of these rats. Ulceration and oedema of the stomach wall was found in both sexes exposed to 9 ppm. A plausible explanation for this observation includes general stress or ingestion of hydrochloric and/or hypochlorous acids as a result of natural grooming or deposition in the oral cavity, either directly or via the mucus escalator. The described observations indicated that repeated exposure of Fischer 344 rats to chlorine resulted in pulmonary effects at all levels used, and hepatic and renal effects at 3 and 9 ppm chlorine. Renal and hepatic effects have not been reported in any other inhalation study with chlorine. A possible explanation was given by the author: “Ammonia, evolved from animal urine and faeces, has been shown to be present in inhalation chambers at levels ranging from approx. 0.2 to 0.4 ppm (Barrow and Dodd 1979). Furthermore, chloramines were formed in a dynamically operated exposure chamber containing chlorine and ammonia (ca. 1ppm each). These data suggest that the concentrations of ammonia present in inhalation chambers are sufficient to react with low levels of chlorine, to form chloramines. The result of the current study may have been affected by the presence of chloramines. A LOAEL in this study was set at 1.0 ppm corresponding to 3.0 mg/m3 (0.806 mg/kg bw/d assuming a body weight of 320 g and a respiratory volume of 0.086 m3/6 h).

Three groups of four male and four female Rhesus monkeys (Klonne 1987, Doc. No. 592-093) each were exposed to chlorine for a period of 6 hours per day for 52 weeks (exposure on week days only). Monkeys were exposed to nominal concentrations of 0, 0.1, 0.5 and 2.5 ppm Cl2 corresponding to analytical concentrations of 0.1±0.03, 0.5±0.1 and 2.3±0.4 ppm Cl2. Clinical signs and mortality were recorded twice daily and the animals underwent a detailed physical examination monthly during the treatment period of the study. Body weights and electrocardiograms were obtained monthly. Ophthalmological examinations were performed prior to initiation of exposures and at the termination of the study. Pulmonary function evaluations, which included pulmonary diffusing capacity of carbon monoxide and distribution of ventilation, were conducted three times prior to the start and monthly thereafter. Haematology, serum chemistry, and urinalysis tests were conducted on all monkeys prior to exposure initiation and at monthly intervals thereafter. At the end of the treatment period, all animals were fasted overnight, anaesthetised and sacrificed by exsanguination. The parasite Anatrichosoma spp. was present in many of the animals and induced moderate to severe, granulomatous rhinitis. However, the lesions associated with this nematode were generally confined to the nasal vestibule in the region lined by squamous epithelium. Furthermore, inflammatory responses to the nematode appeared to have no major effect on the shape and diameter of the vestibular lumen, and thus probably had little effect on nasal airflow patterns or consequently on regional deposition of chlorine in the nose. In this study, treatment-related responses were confined to ocular and respiratory tract irritation. Histopathological examinations revealed that treatment-induced lesions were limited to the respiratory epithelium of the nose and trachea. These lesions were characterised by mild, focal, epithelial hyperplasia in the absence of epithelial thickening with an associated loss of cilia and goblet cells in the affected areas. Nasal and tracheal lesions were induced by exposure to 2.3 ppm chlorine, while less distinct but similar changes were also present in the nasal passages of some animals in the 0.5 and 0.1 ppm groups in the absence of tracheal lesions. These findings indicate a concentration-related response relationship for chlorine-induced airway toxicity. No histological lesions were observed in this study at sites other than the nasal cavity and trachea. There was also no indication that exposure to chlorine concentrations up to 2.3 ppm for 1 year produced neither electrocardiogram abnormalities nor haematological changes. Blood gases, distribution of ventilation, diffusing capacity, and histological evaluations were conducted. There were no exposure-related effects found in blood gas or pulmonary function parameters. These findings were further verified by the lack of histological lesions. The NOAEL was considered to be 0.5 ppm corresponding to 1.5 mg/m3 (corresponding to 4.5 mg/kg bw/d assuming a body weight of 2.5 kg and a respiratory volume of 0.021 m3/min. similar to humans).

The chronic toxicity and carcinogenic potential of chlorine was examined in groups of 70 Fischer 344 rats per sex in a 104 weeks inhalation study with an additional interim sacrifice after 12 months of exposure (Wolf 1995, Doc.No. 592-158). The animals were exposed to chlorine concentrations of 0, 0.4, 1.0 and 2.5 ppm. During the experimental period all animals were observed and mortalities were recorded. Body weight was measured. All surviving animals were killed after 104 weeks of treatment. A wide range of tissues, including the entire respiratory tract was collected from all animals. Histological examination was performed on all tissues from high-concentration and control animals and selected target organs from medium- and low-concentration groups. Following observations were made:

General: Chlorine exposure resulted in decreased body weight gain. Survival to the scheduled termination of the study was not different between chlorine exposed and control animals. There were no biological or statistically significant increases in neoplasms related chlorine exposure. Chlorine-induced lesions were found only in nasal passages. No lesions were found in the remainder of the respiratory tract. The nasal lesions were generally site-specific within the nose, but their severity and incidence were not always concentration-dependent. Most of the nasal responses exhibited a clear anterior-to-posterior severity gradient and only rarely extended to the nasopharyngeal meatus.

Cancer endpoints: In this study it could be demonstrated that chlorine is not a carcinogen. This is supported by a non-classification of Chlorine by the european experts (recent EU Risk Assessments).

Non-cancer endpoints: Chlorine was clearly a nasal toxicant that affected all airway epithelial types in the nose, but no response was observed in the larynx or lower respiratory tract. Examination of the nasal lesion data reveals that the majority of lesions associated with chlorine exposure were also present in control animals, but with a lesser incidence or severity or both. This observation suggests that chlorine exacerbated the development of these background lesions.

Mechanism of toxicity: Chlorine-induced cellular injury is believed to be the result of oxidation of functional groups in cellular components after reaction of chlorine with water, forming hydrochloric and hypochlorous acid. The resulting hypochlorous acid is an oxidising agent, which can cause cytotoxicity and cytolethality. The hydrochloric acid may act as a secondary irritant. These mechanisms are consistent with the nasal tissue damage and inflammation observed in the rats. With chlorine the lateral meatus was the most severely affected region of the nose following acute and chronic exposures.

The development of eosinophilic proteinaceous accumulation was associated with varying degrees of mucous cell metaplasia and hyperplasia. This observation is consistent with the proposal that eosinophilic proteinaceous accumulations are composed of a protein associated with aberrant or excessive mucus synthesis. Female rats were markedly more sensitive to chlorine exposure than male rats. In all cases the specific localisation of the rodent lesions was in the upper respiratory tract. This observation may reflect regional airway dosimetry, regional tissue susceptibility, or a combination of these factors. No NOAEL was derived. The LOAEL was set at 0.4 ppm (corresponding to 0.35 mg/kg bw/d, assuming a mean body weight of 320 g and a respiratory volume of 240 mL/min) based on significantly increased incidence and severity in goblet cell hyperplasia and olfactory epithelium eosinophilic proteinaceous accumulation.

In another study (Wolf 1995, Doc.No. 592-158), the chronic toxicity and carcinogenic potential of chlorine was examined in groups of 70 B6C3F1 mice (except for the control group of 67 male mice) per sex in a 104 weeks inhalation study. The animals were exposed to chlorine concentrations of 0, 0.4, 1.0 and 2.5 ppm. The following results were obtained: General: Chlorine exposure resulted in decreased body weight gain. Survival to the scheduled termination of the study was not different between chlorine exposed and control animals. There were no biological or statistically significant increases in neoplasms related chlorine exposure. Chlorine-induced lesions were found only in nasal passages. No lesions were found in the remainder of the respiratory tract. The nasal lesions were generally site-specific within the nose, but their severity and incidence were not always concentration-dependent. Most of the nasal responses exhibited a clear anterior-to-posterior severity gradient and only rarely extended to the nasopharyngeal meatus. Cancer endpoints: In this study it could be demonstrated that chlorine is not a carcinogen. This is supported by a non-classification of Chlorine by the ECB. Non-cancer endpoints: Chlorine was clearly a nasal toxicant that affected all airway epithelial types in the nose, but no response was observed in the larynx or lower respiratory tract. Examination of the nasal lesion data reveals that the majority of lesions associated with chlorine exposure were also present in control animals, but with a lesser incidence or severity or both. This observation suggests that chlorine exacerbated the development of these background lesions. Some nasal lesions were only seen chlorine-treated animals, including squamous metaplasia in respiratory epithelium in female mice and septal fenestration and eosinophilic proteinaceous accumulation in squamous epithelium in male mice. The development of eosinophilic proteinaceous accumulation was associated with varying degrees of mucous cell metaplasia and hyperplasia. This observation is consistent with the proposal that eosinophilic proteinaceous accumulations are composed of a protein associated with aberrant or excessive mucus synthesis. Mice exhibited sex differences in response to comparable concentrations of chlorine. Male mice had an equivalent incidence of septal fenestration at all exposure concentrations, whereas female mice only had increased incidences at the mid and high concentrations, suggesting that male mice are more sensitive then female mice to chlorine exposure. In all cases the specific localisation of the rodent lesions was in the upper respiratory tract. This observation may reflect regional airway dosimetry, regional tissue susceptibility, or a combination of these factors. No NOAEL was derived. The LOAEL was set at 0.4 ppm based on significantly increased incidence in squamous epithelium eosinophilic proteinaceous accumulation (males) respiratory epithelium hyperplasia (males and females), and olfactory epithelium atrophy (females).

For endpoint summary records of chronic toxicity/carcinogenicity tests see IUCLID5 section 7.7.

Summary:

A 3-day human volunteer study by Schins et al. (2000) (see acute toxicity) showed no effect at 0.5 ppm (1.5 mg/m³), confirming the NOAEL of the monkey study.

No systemic effects were observed in repeated dose exposure studies in rats, mice and monkeys. Human repeated exposure to chlorine is not expected to lead to effects other than irritation observed in the study by Schins et al. (2000). For the risk characterisation the NOAEL of 0.5 ppm (1.5 mg/m³) from the human volunteers study, supported by repeated dose study in monkeys provides a reliable and robust basis. Additionally, chlorine was discussed by SCOEL and an OEL of 0.5 ppm (1.5 mg/m³) was agreed based on the presented data, with removal of the 8-hour TWA. The justification was that the effects appear to be related to concentration in the air and not to duration of exposure (SCOEL, 1998).

Justification for classification or non-classification

Based on the results obtained in the repeated dose toxicity studies and taking into account the provisions laid down in Council Directive 67/548/EEC and CLP, chlorine does not have to be classified with respect to repeated dose oral, dermal or inhalation toxicity, respectively. Also no specific target organ toxicity was detected by the available studies.