Environmental Engineering Reference
In-Depth Information
consumption patterns, and ethnic exposure patterns are
desirable for risk evaluations (Harris and Harper, 1997; Toth
and Brown, 1997; Burger and Gochfeld, 2001, 2006a, 2007;
Holloman and Newman, 2006).
Additional factors to consider in assessing mercury
exposure and toxicity are foods eaten at the same time
and co-occurring chemicals that can modify the effects
of mercury. For example, green and black tea extracts can
reduce the bio-accessibility of mercury (Shim et al. 2009).
There has been considerable interest in the ability of sele-
nium to reduce the effects of mercury through a variety
of mechanisms. Selenium is thought to be protective for
mercury exposure (Satoh et al., 1985); lower levels of non-
fatal heart attacks have been associated with higher levels
of selenium (Mozaffarian, 2009) and studies with animal
models have suggested that some of the adverse impacts
of high methylmercury exposure are a result of pathologic
effects from impaired selenium-dependent enzyme activi-
ties (Watanabe et al., 1999; Ralston, 2008, 2009; Ralston
et al., 2008).
Finally, it is customary to examine human exposure
to mercury from the viewpoints of both the general pub-
lic and workers, both in factories and in home operations
such as gold mining (Maramba et al., 2006). Gold-mining
operations provide a good example of the possible routes
of exposure. Miners are exposed to gaseous elemental mer-
cury and gold dust, infants and children ingest mercury
through hand-to-mouth activity, and fi nally, mercury can
seep into rivers, and with methylation, move up the food
chain to species that are consumed by people.
for multigenerational effects. If all of these uncertainty
factors apply, they would be multiplied to serve as the
denominator.
However, for mercury, the NOAELs are based on human
data; this greatly reduces uncertainty. The RfD for mercury
was fi rst based on retrospective studies of an incident of poi-
soning in Iraq in which grain, contaminated with a methyl-
merury fungicide, was used in the baking of bread. Although
the exposure lasted only a few months, it was quite high,
and resulted in neurologic effects in both adults and infants
(Shipp et al., 2000). When data from several child develop-
ment studies became available, the USEPA revised its chronic
oral RfD for methylmercury to 0.1 µg/kg/day (USEPA,
2001c). In this and in the following discussions, refer to
table 12.4 for concentrations and risk levels set by various
agencies.
EPA's Reference Dose for Methylmercury
The RfD is based on the assumption that thresholds
exist for certain toxic effects such as cellular necrosis. It
is expressed in milligrams per kilogram per day and is
defi ned as follows: “In general, the RfD is an estimate (with
uncertainty spanning perhaps an order of magnitude) of a
daily exposure to the human population (including sensi-
tive subgroups) that is likely to be without an appreciable
risk of deleterious effects during a lifetime” (USEPA, 2006).
In, 1985 EPA set the RfD for methylmercury at 0.0003
mg/kg/day (0.3 µg/kg/day) (Rice et al., 2000; Schierow,
2004), based on neurodevelopmental effects observed in
the 1970s Iraq poisoning epidemic (this was subsequently
lowered to 0.0001 mg/kg/day, see next paragraph). The
Iraq epidemic resulted from direct human consumption
of fungicide-treated grain intended for planting. Subse-
quent to the EPA's comprehensive mercury report (USEPA,
1997), new epidemiologic data based mainly on exposure
through fi sh consumption, became available. The EPA's
reassessment relied mainly on results from studies in New
Zealand and the Faroe Islands that detected neurobehav-
ioral defi cits related to prenatal mercury levels. At the same
time the USFDA and the ATSDR arrived at comparable
estimates of about 0.4 µg/kg/day and 0.3 µg/kg/day, respec-
tively. The FDA has set an action level of 1 ppm methyl-
mercury in commercial fi sh, above which level fi sh could
be seized. From that level FDA arrived at an ADI of 0.4 µg/
kg/day using average consumption levels (Schierow, 2004).
The ATSDR arrived at an MRL of 0.3 µg/kg/day based on
the “negative” Seychelles Island study. In 2004, the Joint
FAO/WHO Expert Committee on Food Additives (JECFA)
established a methylmercury guideline (PTWI) of 3.3 µg/kg
body weight per week for methylmercury based on adult
data from the Iraq epidemic (Schierow, 2004). It contin-
ued this level at its 1999 meeting (JECFA, 1999), based on
reassuring preliminary results from the Seychelles study
(Davidson et al., 1998). At its 2003 meeting, it revised the
PTWI to 1.6 µg/kg/day, based on an evaluation of both
Noncancer End Points and Reference Doses
Risk for noncancer effects of mercury on people focus on
neurobehavioral effects in adults or neurodevelopmental
effects on the fetus (ATSDR, 1999; NRC, 2000a; Institute
of Medicine [IOM], 2006). These approaches use the RfD
(EPA), MRL (ATSDR), or ADI (FDA) to arrive at levels of
mercury that are not expected to cause harm (Table 12.4).
Although for most risk assessments, one usually makes
assumptions about exposure that involve a 70-kg adult
over a 50- to 70-year lifespan, for mercury, risk assess-
ments for women and children are usually made because
of the sensitivity of the developing fetus and young chil-
dren (ATSDR, 1999; USEPA, 2006). The RfD, the EPA's
approach to an ADI, is computed by dividing a NOAEL
(if available), by a denominator combining all applicable
uncertainty factors. The choice of NOAEL is critical. The
uncertainty factors include: (1) extrapolating from animal
data to humans (using a default UF of 10), (2) intraspe-
cifi c variability to protect the most sensitive individuals
(such as pregnant women or children [default UF of 10]),
(3) calculating a lifetime risk from a study that used only
an acute or subacute (short-term) exposure (default UF of
10), and (4) when the RfD is based on a LOAEL rather than
a NOAEL (UF of 10). Other safety factors might account
