Acrylamide in drinking-water
Background document for development of WHO Guidelines for Drinking-water Quality
Guideline value
In mutagenicity assays, acrylamide does not cause mutations in bacterial test systems but does cause chromosome damage to mammalian cells in vitro and in vivo. In a long-term carcinogenicity study in rats exposed via drinking-water, it induced tumours at various sites (Johnson et al., 1986). IARC (1994) has placed acrylamide in Group 2A (probably carcinogenic to humans).
New data on acrylamide exposure through food has led to a re-examination of the mechanism of carcinogenicity and the cancer risk associated with exposure from all sources. Controlling acrylamide exposure through drinking-water is relatively easy. A non-threshold approach was used to define safe concentrations that can be readily achieved. The upper-bound risks determined using the linearized multistage model do not equate to actual population risks through drinking-water or other routes of exposure. On the basis of combined mammary, thyroid and uterine tumours observed in female rats in a drinking-water study (Bull et al., 1984) and using the linearized multistage model, guideline values associated with upper-bound excess lifetime cancer risks of 10-4, 10-5 and 10-6 are estimated to be 5, 0.5 and 0.05 µg/litre, respectively. The most important source of drinking-water contamination by acrylamide is the use of polyacrylamide flocculants that contain residual acrylamide monomer. The practical quantification level for acrylamide is generally of the order of 1 µg/litre. However, acrylamide concentrations in drinking-water are controlled by limiting either the acrylamide content of polyacrylamide flocculants or the dose used, or both. In the event of acrylamide being present in raw water, acrylamide concentrations can be reduced by ozonation (Mallevialle et al., 1984) or treatment with potassium permanganate (Ma et al., 1994). Conventional treatment processes do not remove acrylamide.