The contamination of groundwater by nitrates is ubiquitous, but the magnitude of the risk posed to human health is still unclear. That the potential toxicity of nitrate in drinking water is largely based on studies using high concentrations of nitrite is a complication. Another complication is that human studies have provided little evidence that adverse health effects result from chronic exposure to low levels of nitrate in drinking water. Although the current MCL for nitrates is also in debate, the public drinking water supply rarely exceeds this limit. The lack of data on unregulated systems, however, is an important issue. The available data on the occurrences of nitrates in drinking water indicate that users of private water systems are most at risk for exposure to nitrate levels above the MCL. However, a lack of studies focusing on users of private water systems means that the extent of the problem is unknown.
Experimental animal studies on nitrate or nitrite (in the form of sodium or potassium nitrate/nitrite) and adverse reproductive and developmental outcomes provide moderate evidence for an association between exposure to nitrate and fetal loss, neonatal mortality, maternal toxicity, and decrease in number of litters and live births (Table 2). Epidemiologic evidence for increased risk for adverse reproductive and developmental outcomes in humans from exposure to nitrate in drinking water is sparse and suggestive at best. Nevertheless, the findings of excess birth defects in some studies (Table 1) suggest the need for further studies.
In most studies presented in this review, exposure to nitrates in drinking water was assessed primarily through water quality data for water systems serving women's addresses during pregnancy or at time of delivery. Although this form of exposure assessment provides information in a timely and cost-efficient way, community-based water quality data provide only a rough estimate of individual exposure. This does not account for other issues in exposure assessment such as other sources of nitrate exposure, individuals not drinking tap water, or use of private water systems. The recent study by Brender et al. (2004) is an example of the extent of exposure assessment that should be considered when evaluating nitrate exposure and reproductive effects.
Drinking water contaminants other than nitrates have been reported to be associated with increased risk of adverse pregnancy outcomes (Bove et al. 1992, 2002). Identifying which contaminant in a community water system is associated with a particular adverse reproductive outcome when multiple contaminants are present is difficult. Future studies should conduct individual exposure assessments such as maternal interviews concerning water consumption habits in and outside the home and other risk factors such as occupational exposures or smoking.
Many studies on birth defects are also limited by the time between the end of pregnancy and the maternal interview. This is difficult for individual studies to overcome because birth defects are rare outcomes and prospective cohort studies may not be feasible to conduct. Prospective cohort studies on end points such as spontaneous abortions are more feasible and would provide knowledge about the potential effects of nitrates in drinking water on this outcome. Spontaneous abortion may be a more sensitive indicator of adverse reproductive effects from relatively low levels of drinking water contamination.
States with large numbers of private wells where groundwater is vulnerable to contamination should be encouraged to increase monitoring or surveillance of such systems. Future research could include long-term monitoring or surveillance of water systems vulnerable to contamination. This could provide valuable exposure assessment information to conduct studies on drinking water contaminants such as nitrates. A discussion on the appropriateness of the current MCL is beyond the scope of this review. However, future studies with improved exposure assessment (including other dietary sources and medications), adequate sample size, and evaluating endogenous nitrate exposure could help to determine whether nitrates in drinking water at the current MCL increase the risk for reproductive and developmental effects.