5.0 RAW DRINKING WATER

5.1 Effects

The following discussion is based on a recent review commissioned by the US Public Health Service (USPHS, 1990) on the toxicology of PAHs. Most of the toxicological information was derived from experimental animals exposed to PAHs under controlled conditions.

Mackenzie and Angevine (1981) administered B[a]P by gavage (i.e., introduction of the contaminant into the stomach by tube) to pregnant CD-1 mice during gestation at the rate of 10, 40, and 160 mg/kg body weight (bw)/d, and found that the viability of the litters at birth was significantly reduced in the highest dosed group. In all treatment groups, the mean pup weight was significantly reduced by 42 days of age. To study postnatal development and reproductive functions, these investigators bred F1 progeny (which were exposed prenatally to B[a]P) with untreated animals. It was found that the F1 progeny from the 10 mg/kg bw/d group experienced decreased fertility with associated alterations in gonadal morphology and germ-cell development. The F1 progeny from higher dose groups exhibited total sterility.

The development of forestomach tumors (papillomas and carcinomas) was studied by Neal and Rigdon (1967) in mice exposed to dietary B[a]P. In mice fed 33 mg B[a]P/kg bw/d for periods of 1 to 7 days the incidence of forestomach tumor increased following 2 or more days of exposure (total dose= 2 mg/animal), while mice fed 13.3 mg B[a]P/kg bw/d for 110 days (total dose= 4.48 mg/animal) did not develop tumors. It was suggested that there are no cumulative carcinogenic effects of B[a]P or its metabolites in mice.

Chronic oral administration of a total dose of 4.5 g anthracene/rat in the diet of BD1 or B111 rats for 78 weeks did not produce tumors (Druckrey and Schmahl, 1955). The carcinogenic potential of PAHs is shown in Table 1.

Studies related to effects on humans from exposure to PAHs, singly or collectively, are rare. Sax (1979) reported a lethal concentration of 100 µg/g naphthalene for human children (an accidental ingestion). Epidemiological studies have shown increased mortality due to lung cancer in humans exposed to coke-oven emissions, roofing-tar emissions, and cigarette smoke. Each of these mixtures contains B[a]P, chrysene, benz[a]anthracene, benzo[b]fluoranthene, and dibenzo[a,h]anthracene, as well as other potentially carcinogenic PAHs and other carcinogenic and potentially carcinogenic chemicals, tumor promoters, initiators, and cocarcinogens such as nitrosamines, coal tar pitch, and creosote (USPHS, 1990). Because of the complex nature of the mixtures, it is difficult to evaluate the contribution of any single PAH to the total carcinogenicity of these mixtures.

Humans can be exposed to PAHs via air, water, and food. In the U.S., Santodonato (1981) estimated general population exposure to total PAH, benzo[a]pyrene, and carcinogenic PAHs (i.e., total of benzo[a]pyrene+ benzo[j]fluoranthene+ indeno[1,2,3-cd]pyrene). The results (Table 14) suggested that drinking water was a minor contributor of PAH body burden in humans.

Lioy et al. (1988) conducted a multimedia study of human exposure to B[a]P in a rural town in New Jersey. The major industry in the town was a grey-iron pipe manufacturing plant which contributed to high PAH levels in the atmosphere. The mean outdoor air concentration of B[a]P was 0.0009 µg/m³ (B[a]P concentration in homes varied from 0.0001 to 0.0081 µg/m³), whereas the maximum concentration of B[a]P in samples of food was 0.001 µg/g wet weight. Ingestion of B[a]P was estimated to range between 0.01 and 4.0 µg/person/wk. B[a]P concentration in drinking water was less than the detection limit of 0.0001 µg/L. In comparing the inhalation and ingestion pathways in each home, these investigators found that potential intake could be similar in each medium.

5.2 Criteria from other jurisdictions

Drinking water criteria for PAHs from various jurisdictions are listed in Table 15.

Based on the premise that drinking water should be comparable in quality with unpolluted ground water, the World Health Organization (WHO) in 1970 and 1971 recommended a limit of 0.2 µg/L for the sum of six PAHs in drinking water (i.e., fluoranthene, benzo[a]pyrene, benzo[g,h,i]perylene, benzo[b]fluoranthene, benzo[k]fluoranthene, and indeno[1,2,3-cd]pyrene) (WHO, 1984). Concentrations of these indicator PAHs ranged from 0.01-0.05 µg/L in ground water and 0.05-0.25 µg/L in relatively unpolluted rivers, with higher levels in polluted rivers and effluents of the world. Subsequent studies, however, revealed that the concentrations of these PAHs in drinking water were considerably lower than the recommended WHO standards (WHO, 1984).

The World Health Organization (1984) also recommended a guideline of 0.01 µg/L for B[a]P alone, based on (a) available toxicity data for B[a]P (Neal and Rigdon, 1967) and its association with other PAHs of known carcinogenicity, (b) a linearized multistage model for lifetime cancer exposure risk, considering 1 in 100 000 as an acceptable risk.

The USEPA (1980) criteria, cited as total PAH concentration in untreated ambient waters, are designed to protect human health from consumption of contaminated water and contaminated organisms inhabiting the water. The USEPA criterion at the 10-5 cancer risk level is less stringent than the Health and Welfare Canada and WHO (1984) criteria, considering that the USEPA criterion, although expressed in terms of total PAH, is actually based on B[a]P.

The New York State guidelines for several lower molecular weight PAHs(e.g., naphthalene, acenaphthene, fluorene, anthracene, phenanthrene, fluoranthene, and pyrene) and high molecular weight PAHs (e.g., Benz[a]anthracene, Chrysene, Benzo[b]fluoranthene, Benzo[k]fluoranthene, and Benzo[a]pyrene) are for class AA waters. The class AA waters are designated for use as drinking water supplies which will meet the drinking water standards with treatment and/or disinfection.

Health and Welfare Canada (1989) recommended a maximum acceptable concentration of 0.01 µg/L B[a]P in drinking water, which was adopted by the British Columbia Ministry of Health.

5.3 Recommended criteria

It is recommended that the B[a]P concentration in drinking water should not exceed 0.01 µg/L. This is the current Canadian/British Columbia drinking water quality guideline.

CCREM (1987) did not recommend drinking water criteria for PAHs.

5.4 Rationale

The maximum PAH concentrations in drinking water recommended in this document are same as those recommended by the World Health Organization (total PAH and B[a]P) and Health and Welfare Canada (B[a]P)

The WHO criterion for B[a]P was based on Neal and Rigdon (1967) data, which showed a significant dose-response relationship, and on a multistage model for risk assessment; 1 in 100 000 was considered to be an acceptable lifetime cancer risk level. At the recommended level of 0.01 µg B[a]P/L, drinking water will contribute less than 30% to the acceptable body burden (at the 1:100 000 cancer risk level) computed in section 6.10.2. According to Health and Welfare Canada, the estimated lifetime risk associated with the ingestion of drinking water containing 0.01 µg/L B[a]P is 5x10-7. In their rationale, however, Health and Welfare Canada neither specified the model used in the risk calculation, nor the exposure to B[a]P from other sources.