N-Nitrosodimethylamine (NDMA)
(endorsed 2011)
Guideline
Based on health considerations, the concentration of NDMA in drinking water should not exceed 0.0001 mg/L (100 ng/L).
Action to reduce NDMA is encouraged, but must not compromise disinfection, as non-disinfected water poses significantly greater risk than NDMA.
General description
N-Nitrosodimethylamine ()(CAS No. 62-75-9) is a member of the dialkylnitrosamine family. Other names for the compound include N-methyl-N-nitrosomethanamine, dimethylnitrosamine, and nitrous dimethylamine. The compound is also referred to by the acronyms NDMA, DMN and DMNA. The most recognisable acronym in the context of water treatment and water recycling is NDMA.
NDMA is a polar compound with a molecular weight of 74.08 g/mol, a water solubility of >10 g/100 ml (at 19ºC) and a Log Octanol/Water partition coefficient of –0.57. Pure NDMA exists as yellow liquid with a density of 1.006 , a boiling point of 151-154ºC and a vapour pressure of 1080 Pa at 25ºC.
NDMA is used as an industrial solvent, an anti-oxidant, a rubber accelerator, and in the preparation of polymers, where it may be used as an initiator or a plasticiser. The compound has been used in the production of rocket fuel, as a biocide for nematodes, and an intermediate for 1,1-dimethylhydrazine to inhibit nitrification of soils.
NDMA is formed under mildly acidic conditions by the reaction of natural and synthetic secondary, tertiary or quaternary amines with nitrate and nitrite. Precursor amines include alkylamines, dimethylamine (DMA), tetramethylthiuram disulfide (thiram) and polyelectrolytes used in water and wastewater treatment. NDMA is also produced as a by-product of chloramination of drinking water (due to the presence of dimethylamine in source waters subject to wastewater discharges or the oxidation of natural organic matter by chlorine in the presence of ammonia) and to a lesser extent by chlorination. NDMA formation can be facilitated in soils by biochemical pathways in micro-organisms, and this compound is resistant to microbial degradation under both aerobic and anaerobic conditions. Ozonation of drinking water contaminated with the fungicide tolyfluamide can also lead to the formation of NDMA.
NDMA can exist in the liquid and vapour phase and may be associated with airborne particulates. The compound has been detected in indoor air contaminated with tobacco smoke at concentrations of up to 240 ng/. Detectable levels in outdoor air have been reported in the immediate vicinity of point sources (e.g. chemical production facilities). NDMA has been detected in preserved foods, such as smoked and salted fish and meat and sausages cured by nitrates. Studies conducted in the 1970s and 1980s found NDMA in foodstuffs at levels up to 17,200 ng/kg for cured meat products such as bacon, 68,000 ng/kg for smoked cheese and 9,200 ng/L for beer; although these levels should be viewed with caution as the concentrations were determined using analytical methods available at the time. Moreover, since that time, efforts have been made to reduce the amount of NDMA in foods by limiting the amount of allowable nitrate in preservation, prohibiting the use of nitrate for certain food groups, and the inclusion of nitrosation inhibitors.
In addition to pre-formed NDMA occurring in some foods, NDMA is generated in the stomach through nitrosation of secondary amines in ingested food, especially fish and meat. This process also involves reaction with nitrate and nitrite from foodstuffs and nitrate formed in the stomach, and is influenced by other food components that may enhance or inhibit nitrosation reactions. For these reasons, it is difficult to estimate the amount of NDMA formed endogenously in the human body.
NDMA is absorbed via the gastrointestinal and respiratory tracts, and may also be absorbed through the skin, but at much lower rates. Distribution in the body is uniform and rapid, and it is metabolised rapidly, with an estimated half life of 4 hours, based on observations in rodents. Excretion is primarily via carbon dioxide in expired air, with only a small percentage persisting as NDMA in the urine.
A worst case estimate for NDMA exposure from contaminated outdoor air and consumption of food and water indicated 5.0-16.0 ng/kg of body weight per day for a 29-50 year old adult (WHO 2006). Drinking water was estimated to account for 0.3-1.0 ng/kg of body weight per day based on a mean NDMA concentration of 12 ng/L and a maximum concentration of 40 ng/L in water. Food was estimated to account for 4.3-11 ng/kg of body weight per day. Cigarette smoking was a more significant source of NDMA exposure, with smokers estimated to have an intake of 1.0-80 ng/kg of bodyweight per day from mainstream smoke, and people with heavy exposure to smoke-contaminated indoor air, an intake of 40-130 ng/kg bodyweight per day from smoke. These estimates did not take into account the endogenous formation of NDMA in the digestive tract, and they indicate that drinking water forms only a minor component of exposure to exogenous NDMA (less than 10%).
Another assessment, incorporating estimates of the possible range of endogenous NDMA formation using data from in vivo and in vitro studies, indicated that drinking water contributed around 2.7% of daily NDMA intake when only exogenous sources were assessed, but only about 0.02% when endogenous NDMA formation was also taken into account (Fristachi and Rice 2007).
Typical values in Australian drinking water
There are no data in the public domain or peer reviewed literature on NDMA in Australian drinking water distribution systems and water treatment plants. Anecdotal evidence suggests a bi-modal distribution, with several water authorities indicating that NDMA is present at levels at or near the limit of determination of 1 to 2 ng/L, whereas preliminary sampling and analysis by other authorities indicates levels in the range of 60-90 ng/L. A recent report from South Australia has indicated that NDMA may originate from rubber components of newly commissioned drinking water pipelines, regardless of the disinfectant used. This may account at least partly for the divergent results reported by different water suppliers.
Measurement
Analytical methods for NDMA detection have been developed with a sensitivity at the nanogram per litre (ng/L) level. The methods developed by the United States Environmental Protection Agency (USEPA 2004) and the Ontario Ministry of the Environment (OME 2004) include a concentration and separation step prior to quantification by gas chromatography and mass spectrometry. Internal standards for each method are based on the use of the deuterated analogue of NDMA, NDMA-d6, as the surrogate. The OME method was developed specifically for use in drinking water. In this method, NDMA is extracted onto Ambersorb 572 and eluted using dichloromethane. NDMA is separated from the solvent using capillary column gas chromatography, and quantified by high resolution mass spectrometry at a detection level of 0.4 ng/L, with a reporting detection level of 1.0 ng/L. The USEPA method was developed for large scale surveys and can be used to detect NDMA and seven other nitrosamines. Following solid phase extraction and elution, the nitrosamines may be separated by gas chromatography and quantified via chemical ionisation tandem mass spectrometry, with a detection level of 0.28 ng/L and a limit of determination of 1.6 ng/L.
Health considerations
NDMA is absorbed through the gastrointestinal tract and subsequently distributes uniformly and rapidly. It can cross the placenta and may be present in breast milk. The metabolic half life in rodents is around 4 hours. It is excreted largely via exhaled carbon dioxide, with limited amounts excreted unchanged in urine.
NDMA is carcinogenic in experimental animals through several exposure routes, including ingestion in drinking water. In 1987 the International Agency for Research on Cancer (IARC) classified NDMA as a Group 2A chemical, probably carcinogenic to humans. The mechanism by which NDMA causes cancer is believed to involve biotransformation in the liver by microsomal enzymes, generating the methyldiazonium ion which subsequently forms DNA adducts.
A number of epidemiological studies have shown an association between NDMA intake from food and increased risks of gastric or colorectal cancer, although the data are not sufficient to derive a quantitative dose-response relationship for cancer risk in humans. These studies did not consider exposure to NDMA from drinking water, or endogenous generation of NDMA in the body.
Various other N-nitrosamine compounds with structures related to NDMA are known to occur in water supplies. The toxicological properties of these compounds have not been well characterised, however it is believed likely that some are also carcinogenic.
Derivation of guideline value
The World Health Organization has derived a guideline value for NDMA in drinking water based on a study of hepatic biliary cystadenomas in rats that used a wide range of NDMA exposure doses (from 0.033 mg/L to 16.896 mg/L). This dataset was used to derive a tumorigenic dose () for NDMA corresponding to a dose level that causes a 5% increase in tumour incidence over the background level. The values were used to calculate a unit risk, which represents the increase in risk per unit increase in dose. Using the most sensitive endpoint observed in the animal study and conservative assumptions, a guideline value of 100 ng/L was derived, corresponding to an excess lifetime cancer risk of 1 in 100,000. This methodology treats the risk from exposure to NDMA in drinking water in isolation from other sources of NDMA exposure.
While adopting the same numerical value, a different approach has been taken to derive the guideline value for NDMA in Australian drinking water supplies. In assessing the potential public health benefits associated with regulation of this compound, the following factors were considered:
NDMA has been demonstrated to be carcinogenic in animals, and is probably carcinogenic in humans.
NDMA levels in drinking water may be an indicator of the presence of structurally related compounds, some of which may also have carcinogenic properties.
The current level of exposure to NDMA from food is uncertain, due to lack of recent analytical data; however, even with changes in food preservation techniques since the 1970s, it is probable that exposure through food is at least 5 to 10 times greater than exposure from drinking water.
There is evidence that exposure from endogenous formation of NDMA in the stomach may greatly exceed dietary exposures from both food and water.
In these circumstances, the adoption of a guideline value corresponding to a 1 in 1,000,000 lifetime cancer risk (the usual target level for health risks from carcinogens in these Guidelines) was deemed inappropriate, as this would impose a disproportionate regulatory burden on water suppliers while having little impact on total population exposures. Nevertheless, it was judged to be prudent to limit levels of NDMA in drinking water, given that this will probably reduce exposure to a range of related but as yet largely uncharacterised N-nitrosamine compounds that may pose potential health risks. For these reasons, a guideline value of 100 ng/L has been adopted.
Limiting formation in drinking water
Reducing the occurrence of NDMA in drinking water systems may conflict with the goals of maintaining a persistent chlorine residual in distributions and controlling levels of other disinfection by-products such as trihalomethanes and haloacetic acids. The potential for NDMA formation may be reduced by avoiding chloramination through the removal of ammonia prior to disinfection, or by operating the system for breakpoint chlorination. If NDMA is a problem, treatment using UV irradiation in the prescence of hydrogen peroxide is an option to reduce NDMA levels while maintaining a chloramine residual. NDMA cannot be removed by air stripping or adsorption, due to its vapour pressure, solubility in water and limited partitioning at interfaces. It is only partially removed (<50%) by liquid phase pressure-driven separation processes such as reverse osmosis.
References
Fristachi A, Rice G (2007). Estimation of the total daily oral intake of NDMA attributable to drinking water. Journal of Water and Health 5(3):341–355.
OME (Ontario Ministry of the Environment) (2004). The determination of N-nitrosodimethylamine (NDMA) in water by gas chromatography-high resolution mass spectrometry (GC-HRMS). Toronto, Ontario, Ontario Ministry of the Environment, Laboratory Services Branch (Report No. NDMA-E3291).
USEPA (United States Environmental Protection Agency) (2004). Method 521: Determination of nitrosamines in drinking water by solid phase extraction and capillary column gas chromatography with large volume injection and chemical ionization tandem mass spectrometry (MS/MS). Washington, DC, United States Environmental Protection Agency (EPA Document No. EPA/600/R-05/054.).
WHO (World Health Organization) (2006). N-Nitrosodimethylamine in drinking water. Background document for development of WHO Guidelines on Drinking-water Quality. World Health Organization.
Last updated