IN 1985, perchlorate contamination was discovered in wells at California Superfund sites; however, perchlorate contamination of water sources nationwide was not recognized until 1997. Today, more than 11 million people have perchlorate in their public drinking-water supplies at concentrations of at least 4 ppb (4 µg/L).1 No national drinking-water standard for perchlorate exists, and the concentration at which a standard should be set is being debated. The U.S. Environmental Protection Agency (EPA), which has the responsibility for establishing national drinking-water standards, has issued draft risk assessments of perchlorate. However, the assessments have come under criticism on the grounds that the conclusions presented in them are based on flawed scientific studies and that not all available data have been incorporated appropriately into them.
In view of the controversy surrounding the concentration at which perchlorate should be regulated, EPA, the Department of Defense (DOD), the Department of Energy (DOE), and the National Aeronautics and Space Administration (NASA) asked the National Research Council (NRC) to assess independently the adverse health effects of perchlorate ingestion from clinical, toxicologic, and public-health perspectives. They also asked the NRC to evaluate the relevant scientific literature and key findings underlying EPA’s 2002 draft risk assessment, Perchlorate Environmental Contamination: Toxicological Review and Risk Characterization.
THE CHARGE TO THE COMMITTEE
In response to the agencies’ request, the NRC convened the Committee to Assess the Health Implications of Perchlorate Ingestion, which prepared this report. The members of the committee were selected for their expertise in pediatrics; endocrinology; pediatric endocrinology; thyroid endocrinology, physiology, and carcinogenesis; immunology; veterinary pathology; animal toxicology; neurotoxicology; developmental toxicology; physiologically based pharmacokinetic modeling; epidemiology; biostatistics; and risk assessment.
The committee was asked to assess the current state of the science regarding potential adverse effects of disruption of thyroid function by perchlorate in humans and laboratory animals at various stages of life. It was asked to evaluate the animal studies with particular attention to key end points, including changes in brain morphometry, behavior, thyroid hormone levels, and thyroid histopathology. On the basis of its review, the committee was asked to determine whether EPA’s findings in its 2002 draft risk assessment, Perchlorate Environmental Contamination: Toxicological Review and Risk Characterization, are consistent with the current scientific evidence. The committee was also asked to suggest specific scientific research that could reduce the uncertainty in the understanding of human health effects associated with ingestion of low concentrations of perchlorate. The committee’s complete statement of task is provided in Chapter 1 of this report.
THE COMMITTEE’S APPROACH TO ITS CHARGE
The committee held five meetings from October 2003 to July 2004. During public sessions at the first, second, and fourth meetings, the committee heard presentations from representatives of the U.S. Office of Science and Technology Policy, EPA, DOD, DOE, NASA, the Food and Drug Administration, the Agency for Toxic Substances and Disease Registry, the California Environmental Protection Agency, Congress, and other interested parties, including industry and environmental groups. At the second meeting, several noted scientists were invited to make presentations to the committee to answer questions raised at the first public meeting. The committee reviewed (1) materials submitted by EPA, DOD, DOE, NASA, industry, and private individuals, (2) studies evaluated in EPA’s 2002 draft perchlorate risk assessment, (3) findings in EPA’s 2002 draft perchlorate
risk assessment, and (4) information from publicly available scientific literature. Accordingly, the committee evaluated both published and unpublished data; however, it typically gave more weight in its deliberations to published reports. Unpublished data were considered only when the committee had sufficient information to evaluate the methods used to produce them. Overall, emphasis was given to studies with the soundest scientific methods to draw conclusions regarding the effects of perchlorate exposure.
COMMITTEE’S EVALUATION AND FINDINGS
The thyroid gland produces two hormones, thyroxine (T4) and triiodothyronine (T3), which circulate in the bloodstream primarily bound to protein. Thyroid hormone synthesis and secretion are normally maintained within narrow limits by an efficient regulatory mechanism. Specifically, decreases in serum thyroid hormone concentrations lead to increases in the secretion of thyrotropin (thyroid-stimulating hormone, TSH) by the pituitary gland, and increases in serum thyroid hormone concentrations lead to decreases in TSH secretion. TSH stimulates virtually every step of thyroid hormone production and secretion, and such tight control of TSH secretion maintains thyroid hormone production and secretion within normal or nearly normal limits and thereby protects against both hypothyroidism (deficiency of thyroid hormone production) and hyperthyroidism (excess of thyroid hormone production).
T4 is largely a precursor hormone with little or no intrinsic biologic activity, and it is converted to T3, the biologically active thyroid hormone, in most tissues of the body. T3 is required for normal development of the central nervous system in fetuses and infants. Its actions include stimulation of the development and growth of neurons (nerve cells) and glial (supporting) cells, the formation of synapses (connections) between neurons, the formation of the myelin sheaths that surround neuronal processes, and the development of neurotransmitters, which transmit signals from one neuron to another. T3 is also required for normal skeletal development and growth. In both infants and adults, T3 and T4 are critical determinants of metabolic activity and affect the function of virtually every organ system.
Iodine is a component of T4 and T3, and transfer of iodide from the circulation into the thyroid gland is an essential step in the synthesis of the
two hormones.2 Iodide transport into the thyroid is mediated by a protein molecule known as the sodium (Na+)/iodide (I−) symporter (NIS). NIS molecules bind iodide with very high affinity, but they also bind other ions that have a similar shape and electric charge. The binding of those other ions to the NIS inhibits iodide transport into the thyroid, which can result in intrathyroid iodide deficiency and consequently decreased synthesis of T4 and T3. For adverse health effects to occur in healthy adults, thyroid hormone production must fall substantially and, more importantly, must remain low for at least several weeks. The minimal prolonged decrease in thyroid hormone production that would be associated with adverse health effects is not known; any decrease is potentially more likely to have adverse effects in sensitive populations (people with thyroid disorders, pregnant women, fetuses, and infants), but data are not available to determine the magnitude of the decrease needed to cause adverse effects in those populations.
Iodide can be obtained only by ingestion of food or water that contains it. Therefore, iodide deficiency and reduction in thyroid hormone production can occur if iodide intake is very low. Because the body maintains the serum concentrations of thyroid hormones within narrow limits through feedback control mechanisms, there is remarkable compensation for iodide deficiency. Generally, thyroid hormone production is normal even when iodide intake is quite low. Hypothyroidism occurs only if daily iodide intake is below about 10 to 20 µg (about one-fifth to one-tenth of the average intake in the United States). However, in pregnant women, iodide deficiency of that severity can result in major neurodevelopmental deficits and goiter in their offspring. Lesser degrees of iodide deficiency may also cause important neurodevelopmental deficits in infants and children.
Perchlorate and the Thyroid
Perchlorate can affect thyroid function because it is an ion that competitively inhibits the transport of iodide into the thyroid by the NIS. In the 1950s and 1960s, potassium perchlorate was given for long periods to patients who had hyperthyroidism to reduce T4 and T3 production by inhibiting thyroid iodide uptake. The medical literature of that era contains reports of successful treatment of more than 1,000 hyperthyroid patients
with potassium perchlorate at doses of 400-2,000 milligrams (mg) per day for many weeks or months. Among them were 12 women who had hyperthyroidism and were treated with 600-1,000 mg of potassium perchlorate per day during pregnancy. One infant had slight thyroid enlargement that decreased soon after birth. No other abnormalities were reported; however, no thyroid-function tests or neurodevelopmental evaluations were conducted, and the infants were not followed thereafter.
Treatment of hyperthyroid patients with potassium perchlorate typically caused few side effects, although some patients had nausea, vomiting, rashes, fever, lymph node enlargement, or kidney dysfunction. The frequency of side effects was dose-dependent. Thirteen patients who had taken 400-1,000 mg of potassium perchlorate per day for 2-20 weeks developed aplastic anemia (cessation of production of red blood cells) or agranulocytosis (cessation of production of white blood cells), and seven of them died. Because of those events and the development of better antithyroid drugs, the use of perchlorate to treat hyperthyroid patients largely ceased by the late 1960s.
A study of long-term administration of potassium perchlorate reported in 1984, however, provides useful data. Eighteen people who had hyperthyroidism caused by Graves disease were treated initially with 900 mg per day. The dose of potassium perchlorate was reduced over a 12-month period to an average of 93 mg per day as thyroid function returned to normal. The patients then received 40-120 mg per day for 12 months. During that period, all the patients had normal serum T4 and T3 concentrations, and most patients no longer had high serum concentrations of TSH-receptor stimulating antibodies, which are the cause of hyperthyroidism in patients who have Graves disease. Absence of the antibodies indicated that the patients no longer had Graves disease. Thus, one could consider treatment in the latter 12 months to be equivalent to administration of perchlorate to healthy people. Therefore, the results provide evidence that moderately high doses of perchlorate given chronically to people with a history of hyperthyroidism do not cause hypothyroidism.
Overall, there have been no reports of the appearance of new thyroid disorders, thyroid nodules, or thyroid carcinomas in patients treated with potassium perchlorate for hyperthyroidism.
Perchlorate has been given to healthy men and women for up to 6 months to determine its effects on normal thyroid function. It is usually administered as potassium perchlorate, which is rapidly absorbed after ingestion and rapidly eliminated from the body unchanged primarily in urine. In those studies, perchlorate doses ranged from 0.007 to 9.2 mg per
kilogram (kg) per day, assuming a 70-kg body weight. There were no changes in serum T4, T3, or TSH concentrations to suggest that thyroid function was adversely affected. The highest dose (9.2 mg/kg per day) lowered thyroid iodide content by 25% and resulted in a very small decrease in serum free T4 concentrations after 4 weeks of perchlorate administration. However, serum TSH concentrations were also lower, not higher, as would occur if serum free T4 concentrations decreased to an important extent. Some doses of perchlorate did inhibit thyroid uptake of radioiodide. For example, a 2-week study found that inhibition of radioiodide uptake ranged from 1.8% at 0.007 mg/kg per day to 67.1% at 0.5 mg/kg per day. Uptake at the lowest dose of 0.007 mg/kg per day was not significantly different from baseline in that study. In a 6-month study, there was also no inhibition in healthy subjects given 0.007 mg/kg per day.
Because of the body’s compensation mechanisms, it is not likely that the decreases in thyroid iodide uptake reported in short-term studies would be sustained; rather, iodide uptake would be expected to return to normal. To cause declines in thyroid hormone production that would have adverse health effects, iodide uptake would most likely have to be reduced by at least 75% for months or longer. On the basis of the studies of long-term treatment of hyperthyroidism in which patients continued to be given perchlorate after their hyperthyroidism resolved and the clinical studies of healthy adults, the perchlorate dose required to cause hypothyroidism in adults would probably be more than 0.40 mg/kg per day, assuming a 70-kg body weight. However, in pregnant women, infants, children, and people with low iodide intake or pre-existing thyroid dysfunction, the dose required to cause hypothyroidism may be lower.
Epidemiologic studies have examined the association of perchlorate exposure with thyroid function and thyroid disease in newborns, children, and adults. Only one study has examined the relationship between perchlorate exposure and adverse neurodevelopmental outcomes in children. Almost all the epidemiologic studies are “ecologic.” In ecologic studies, exposure data, outcome data, or both are available only for large geographic areas, not for individuals. Exposures measured within areas are then applied to all persons living in the areas. The assumption that the geographically defined exposures reasonably represent those of all people in an area is more reasonable when one is studying agents that are parts of a common-
source exposure, such as contaminants in a municipal water supply; however, individual water consumption is still likely to vary because of the use of well water or bottled water or a nonuniform distribution of contaminants within a geographic area. Ecologic studies do not include information about exposure and outcome within individuals, so they are considered to be the weakest type of observational studies. They are subject to what is referred to as the ecologic fallacy in that relationships observed (or not observed) between exposure and outcomes at the ecologic level may not apply at the individual level. Thus, ecologic studies alone cannot provide direct evidence of causation, although their results can provide supporting data concerning a possible causal relationship.
Acknowledging that ecologic data alone are not sufficient to demonstrate whether or not an association is causal, the committee found that they can provide evidence bearing on possible associations and reached the following conclusions regarding the proposed association of perchlorate exposure with various health end points:
Congenital hypothyroidism. The available epidemiologic evidence is not consistent with a causal association between perchlorate exposure and congenital hypothyroidism as defined by the authors of the studies reviewed by the committee. All studies of that association were negative.
Changes in thyroid function in newborns. The available epidemiologic evidence is not consistent with a causal association between exposure during gestation to perchlorate in the drinking water at up to 120 ppb and changes in thyroid hormone and TSH production in normal-birthweight, full-term newborns. Most of the studies show neither significantly lower T4 production nor significantly higher TSH secretion in infants born in geographic areas in which the water supply had measurable perchlorate concentrations. However, no data are available on the association of perchlorate exposure with thyroid dysfunction in the groups of greatest concern, low-birthweight or preterm newborns, offspring of mothers who had iodide deficiency during gestation, or offspring of hypothyroid mothers.
Neurodevelopmental outcomes. The epidemiologic evidence is inadequate to determine whether or not there is a causal association between perchlorate exposure and adverse neurodevelopmental outcomes in children. Only one pertinent study has been conducted: an ecologic study that examined the association of perchlorate exposure with autism and attention-deficit-hyperactivity disorder (ADHD). Although the committee considers the inclusion of ADHD plausible, it questions the appropriateness of autism as an end point given that autism has not been observed in the spectrum of
clinical outcomes in children who had congenital hypothyroidism and were evaluated prospectively.
Hypothyroidism and other thyroid disorders in adults. The evidence from chronic, occupational-exposure studies and ecologic investigations in adults is not consistent with a causal association between perchlorate exposure at the doses investigated and hypothyroidism or other thyroid disorders in adults. In occupational studies, perchlorate doses as high as 0.5 mg/kg per day have not been associated with adverse effects on thyroid function in workers. However, the small sample sizes in some studies may have reduced the ability to identify important differences, and the studies were limited to those workers who remained in the workforce.
Thyroid cancer in adults. The epidemiologic evidence is insufficient to determine whether or not there is a causal association between exposure to perchlorate and thyroid cancer. Only two pertinent ecologic studies have been conducted. In one, the number of cancer cases was too small to have a reasonable chance of detecting an association if one existed. In the second, people were exposed to both perchlorate and trichloroethylene. In neither study was it possible to adjust for potential confounding variables. However, the committee questions the biologic plausibility of thyroid cancer as a likely outcome of perchlorate exposure.
The committee emphasizes that no studies have investigated the relationship between perchlorate exposure and adverse outcomes among especially vulnerable groups, such as low-birthweight or preterm infants. The available studies did not assess the possibility of adverse outcomes in the offspring of mothers who were exposed to perchlorate and had a low dietary iodide intake. Finally, there have been no adequate studies of maternal perchlorate exposure and neurodevelopmental outcomes in infants.
Animal Toxicology Studies
The pituitary-thyroid system of rats is similar to that of humans. For example, decreases in thyroid hormone production result in increased secretion of TSH, which then increases thyroid production and release of T4 and T3. However, differences in binding proteins, binding affinities of the proteins for the hormones, turnover rates of the hormones, and thyroid stimulation by placental hormones lead to important quantitative differences between the two species. Therefore, although studies in rats provide useful qualitative information on potential adverse effects of perchlorate exposure, they are limited in their utility for quantitatively assessing human health risk associated with perchlorate exposure.
One of the most controversial issues regarding the animal toxicology studies is the interpretation of results of rat studies that evaluated the effects of maternal perchlorate exposure on offspring brain development. In those studies, female rats were given ammonium perchlorate throughout pregnancy and into the postnatal period. Linear measurements of several brain regions of the male and female pups at several postnatal ages were compared with control values. The most consistent change observed was a statistically significant increase in the width of the corpus callosum; however, the dose at which that change was observed was not consistent between studies. Serious questions have been raised regarding the design and methods used in those studies. The committee agrees with previous reviewers that the methodologic problems, such as possible systematic differences in the plane of section across treatment groups, and the lack of a consistent dose-response relationship make it impossible to conclude whether or not perchlorate exposure causes changes in brain structure. Furthermore, the committee notes two issues concerning study design. First, although it may be appropriate to collect data in a nonblinded fashion in a comparative-morphology study, it is not appropriate to collect measurements of linear thickness of tissues in a nonblinded fashion, as was done for a set of sections measured in those studies. Second, measurements of the thickness of brain areas are not the most sensitive method of detecting alterations in neural structure. Measurements of area or volume would be more sensitive and more accurate, particularly for structures, such as the corpus callosum, that change shape across serial coronal sections. Furthermore, all measurements of size, including measurements of area and volume, are only a surrogate for changes in the cellular structure of a brain region, and it is ultimately the underlying cellular changes and their neurodevelopmental effects that are important to understand.
Other studies that have received critical attention are rat studies that investigated the effect of maternal exposure on offspring neurobehavior. In the primary study, female rats were treated with ammonium perchlorate throughout pregnancy and into the postnatal period, and the offspring evaluated with a battery of behavioral tests. Overall, the committee found that the functions evaluated (motor activity, auditory startle, learning, and memory) were appropriate in light of the suspected mode of action of perchlorate. However, the tests used were screening measures and were unlikely to detect subtle alterations in motor or cognitive functions associated with moderate reductions in circulating thyroid hormones. In addition, some important functional end points—such as auditory function, balance, and coordination—were not assessed. Therefore, it is not surprising that no significant effects of perchlorate were observed on any of the behavioral
measures except an increase in motor activity in male pups on one day of testing. Because the tests lacked the sensitivity to detect subtle effects, the committee concludes that the data are inadequate to determine whether or not gestational or lactational exposure to perchlorate affects behavioral function in rats.
Concerns have also been raised over the significance of the results of a two-generation rat study in which benign thyroid tumors were observed in two male offspring. Both the parent generation and the offspring were given ammonium perchlorate before mating, during mating, gestation, and lactation, and until sacrifice. The offspring had additional exposure during their gestation and lactation periods. On review of the original pathology data, the committee found that one control male rat in the parent generation also had a benign thyroid tumor at about the same age of those observed in the two male offspring. The committee agreed that the two male offspring did have benign thyroid tumors and noted that the observations are expected in rats given high concentrations of goitrogenic chemicals known to interfere with thyroid hormone homeostasis. The committee concludes that the thyroid tumors in the offspring were most likely treatment-related but that thyroid cancer in humans resulting from perchlorate exposure is unlikely because of the hormonally mediated mode of action and species differences in thyroid function.
On the basis of observations that high doses of perchlorate in humans with hyperthyroidism have caused side effects that could be considered immunologic responses—such as skin rashes, aplastic anemia, or agranulocytosis—some have suggested that perchlorate exposure might adversely affect the immune system. However, extensive immunotoxicity studies in mice revealed no changes in immunologic function in response to perchlorate exposure. Therefore, the committee finds that there is no evidence for a causative relationship between perchlorate ingestion and any biologically meaningful stimulatory or inhibitory effect on the immune system in rodents, and concludes that the side effects in humans were probably toxic effects of the very high doses of perchlorate given to those patients.
EPA’s 2002 Perchlorate Risk Assessment
The committee’s review of EPA’s findings in its 2002 perchlorate risk assessment focused on four points: the mode-of-action model of perchlorate toxicity, the definition of adverse effect, the point of departure, and the use of uncertainty factors to derive a reference dose (RfD) for daily oral expo-
sures to perchlorate. EPA’s proposed mode-of-action model of perchlorate toxicity represents a continuum of possible health effects of perchlorate exposure. Ultimately, EPA’s model shows birth defects in children and tumors in adults as possible effects of inhibition of thyroid iodide uptake. The committee finds that EPA’s mode-of-action model adequately represents the possible early sequence of events after perchlorate exposure, but it does not think that the model is an accurate representation of possible outcomes after changes in thyroid hormone and TSH production. Figure S-1 shows the committee’s suggested mode-of-action model of perchlorate toxicity in humans.
The committee concludes that the most reasonable pathway of events after sustained changes in thyroid hormone and TSH secretion would be thyroid hypertrophy or hyperplasia, possibly followed by hypothyroidism in people unable to compensate with an increase in thyroid iodide uptake. At that point, the pathway would diverge to two potential outcomes—metabolic sequelae (such as decreased metabolic rate and slowing of the function of many organ systems) at any age and abnormal growth and development of fetuses and children. The committee concludes that the development of thyroid tumors, as an ultimate result of perchlorate-caused inhibition of thyroid iodide uptake, is unlikely in humans.
An important point is that inhibition of thyroid iodide uptake is the only effect that has been consistently documented in humans exposed to perchlorate. Furthermore, the outcomes at the end of the continuum are not inevitable consequences of perchlorate exposure. Mechanisms exist that allow the body to compensate for decreases in T4 and T3 production. The compensatory increase in TSH secretion and thyroid iodide uptake can return T4 and T3 production to normal without causing adverse effects on human health.
Given the mode-of-action model shown in Figure S-1, the committee concludes that the first adverse effect in the continuum would be hypothyroidism. Any effects that follow and result from hypothyroidism clearly would be adverse. EPA defined changes in serum thyroid hormone and TSH concentrations as adverse effects. The committee does not agree that transient changes in serum thyroid hormone or TSH concentrations are adverse health effects; they are simply biochemical changes that might precede adverse effects.
A primary purpose of EPA’s perchlorate risk assessment was to calculate an RfD. The first step in deriving an RfD is a comprehensive review of all relevant human and animal data. Traditionally, a critical effect and a critical study are then identified that serve as the point of departure for the
risk assessment. Typically, a no-observed-adverse-effect level (NOAEL) or a lowest-observed-adverse-effect level (LOAEL) is identified from the critical study on which the RfD can be based. More recently, mathematical modeling of the dose-response data in the study has been used to provide a benchmark dose (BMD) on which the RfD can be based. The final step in the RfD process is the application of uncertainty factors to the NOAEL, LOAEL, or BMD to extrapolate from the study population to the general human population, which includes sensitive groups.
For the perchlorate risk assessment, EPA based its point of departure on reported changes in brain morphometry, thyroid histopathology, and serum thyroid hormone concentrations after oral administration of perchlorate to rats. The committee does not think that the animal data or the outcomes selected by EPA should be used as the basis of the perchlorate risk assessment. Rather, the committee recommends that inhibition of iodide uptake by the thyroid in humans, which is the key biochemical event and not an adverse effect, should be used as the basis of the risk assessment. Inhibition of iodide uptake is a more reliable and valid measure, it has been unequivocally demonstrated in humans exposed to perchlorate, and it is the key event that precedes all thyroid-mediated effects of perchlorate exposure.
The committee emphasizes that its recommendation differs from the traditional approach to deriving the RfD. The committee is recommending using a nonadverse effect rather than an adverse effect as the point of departure for the perchlorate risk assessment. Using a nonadverse effect that is upstream of the adverse effects is a conservative, health-protective approach to the perchlorate risk assessment.
The committee reviewed the human and animal data and found that the human data provided a more reliable point of departure for the risk assessment than the animal data. The committee recommends using clinical data collected in a controlled setting with the relevant route of exposure to derive the RfD. Although the data from epidemiologic studies of the general population provide some information on possible effects of perchlorate exposure, those studies are ecologic and inherently limited with respect to establishing causality and serving as a basis of quantitative risk assessment. Furthermore, those studies typically focused on changes in serum thyroid hormone and TSH concentrations or clinical manifestations of the changes, not on inhibition of iodide uptake by the thyroid. Therefore, the committee is not recommending using the available epidemiologic studies to derive the point of departure for the risk assessment. Instead, the committee recommends using the Greer et al. (2002) study in which groups of healthy men and women were administered perchlorate at 0.007-0.5 mg/kg per day for 14 days.3 The study identified a no-observed-effect level (NOEL) for inhibition of iodide uptake by the thyroid at 0.007 mg/kg per day. The committee concludes that using the NOEL for iodide uptake inhibition from Greer et al. (2002) as the point of departure provides a reasonable and transparent approach to the perchlorate risk assessment. The NOEL value from Greer et al. (2002) is consistent with other clinical studies that have investigated iodide uptake inhibition by perchlorate. That the NOEL value from Greer et al. (2002) is a health-protective and conservative point of departure is supported by the results of a 6-month study of 0.007 mg/kg per day in a small group of healthy subjects, a 4-week study of higher doses in healthy subjects, the studies of perchlorate treatment of patients with hyperthyroidism, and extensive human and animal data that demonstrate that there will be no progression to adverse effects if no inhibition of iodide uptake occurs (see Figure S-1).
If the committee’s recommendation is used as the point of departure, it
recommends using a total uncertainty factor of 10. A full factor of 10 should be used for the intraspecies factor to protect the most sensitive population—the fetuses of pregnant women who might have hypothyroidism or iodide deficiency. No additional factors are needed for duration or database uncertainties.4 First, if inhibition of iodide uptake by the thyroid is used, chronic exposure will have no greater effect than that resulting from short-term exposure. In fact, it may well have less effect because of the capacity of the pituitary-thyroid system to compensate for iodide deficiency by increasing iodide uptake. Second, the database is sufficient, given the point of departure selected—one based on inhibition of iodide uptake by the thyroid.
One committee member thought that the factor for database uncertainty should be greater than 1 and provided the following rationale:
The RfD is derived from a study in which a group of only seven healthy adults was given 0.007 mg/kg of perchlorate daily for 14 days (Greer et al. 2002). Although two other studies had similar results, the total number of subjects is still small. In addition to the small number of subjects, no chronic exposure studies have been published. An uncertainty factor of 3 could account for the uncertainty surrounding the small number of subjects and the absence of a long-term study.
The other committee members provided the following response:
Although the committee acknowledges that the low-dose group (0.007 mg/kg per day) in Greer et al. (2002) had only seven subjects, the study examined the effects of four doses in a total of 37 subjects. In addition to the Greer et al. (2002) study, there are four other studies in which healthy adults were given perchlorate. The results of all the studies are remarkably similar (see Chapter 2, pp. 65-67). In addition to those studies, the studies of long-term treatment of hyperthyroidism and the studies of occupational and environmental exposure add confidence to the overall database. The issue concerning the absence of a long-term study is discussed in the section Subchronic-to-Chronic Extrapolation Factor in Chapter 5. Briefly, the key is recognizing that the committee is recommending that the RfD be based on inhibition of iodide uptake by the thyroid, a non-adverse biochemical event that precedes any adverse effects in the mode-of-action model. If that effect is used to derive the RfD, chronic exposure will have no greater effect than that resulting from short-term exposure, and in fact, it may well have less effect because of the capacity of the pituitary-thyroid system to compensate for iodide deficiency by increasing iodide uptake (see Chapter 5, p. 175).
The committee recognizes that its recommendations would lead to an RfD of 0.0007 mg/kg per day.5 That value is supported by other clinical studies, occupational and environmental epidemiologic studies, and studies of long-term perchlorate administration to patients with hyperthyroidism. The committee concludes that an RfD of 0.0007 mg/kg per day should protect the health of even the most sensitive populations. The committee acknowledges that the RfD may need to be adjusted upward or downward on the basis of future research, such as that suggested in this report.
The committee was asked to suggest scientific research that could reduce the uncertainty in the understanding of human health effects associated with perchlorate ingestion at low concentrations, especially research that could clarify “safe” exposure for sensitive populations. Although the committee found that available data are sufficient to derive an RfD for perchlorate, new research could provide a more complete understanding of the array of effects of perchlorate, especially regarding the effects of chronic exposure and the effects on sensitive populations. Therefore, the committee recommends a series of interrelated clinical, mechanistic, and epidemiologic studies that have the potential to define more precisely “safe” perchlorate exposures.
The committee recommends a clinical study designed to provide information on the potential chronic effects of low-dose perchlorate exposure on thyroid function, with a special focus on the ability and mechanisms of thyroid compensation. If long-term studies of humans are not possible, chronic studies in nonhuman primates could provide useful information. Studies of pregnant monkeys could also provide useful information on the effects of perchlorate on fetal and neonatal development. Further toxicology studies of perchlorate in rats would be less useful for clarifying the health effects of perchlorate in humans.
Especially critical issues in perchlorate risk assessment have been the effect of perchlorate on placental and breast iodide transport and the influence of iodide status on the effects of perchlorate. The committee recommends a series of in vitro studies using human tissues and animal studies to determine the role of NIS in placental iodide transport, the susceptibility of
breast NIS to perchlorate inhibition, the role of iodide status in these effects, and the effects of perchlorate on development independently of effects on iodide transport. The committee notes that other tissues contain NIS, such as the salivary glands, gastric mucosa, and perhaps the choroid plexus. Studies of NIS in those tissues, and possible effects of perchlorate on them, might be done but at a much lower priority than studies of the placenta and mammary gland.
The primary sources of uncertainty in estimating an RfD for perchlorate in drinking water arise from the absence of data on possible effects of exposure among populations at greatest risk of adverse effects of iodide deficiency (pregnant women and their fetuses and newborns). Therefore, new epidemiologic research should assess the possible health effects of perchlorate exposure in those populations. Future epidemiologic research should focus on additional analyses of existing data, new studies of health effects in selected populations, and monitoring of frequencies of specific conditions in communities affected by efforts to reduce perchlorate in drinking water.
Finally, in its deliberations on the health effects of perchlorate in drinking water, the committee considered pregnant women and their fetuses to be particularly sensitive populations. Although iodide deficiency is believed to be rare in the United States, some pregnant women may have a low iodide intake. The committee believes that further research is needed to measure more precisely the extent of, and risk factors for, iodide deficiency, particularly in pregnant women and their offspring. However, while studies are being conducted, the committee emphasizes the importance of ensuring that all pregnant women have adequate iodide intake and, as a first step, recommends that consideration be given to adding iodide to all prenatal vitamins.