Environmental Engineering Reference
In-Depth Information
and public health organizations to devote a great deal
of time and attention to determining an acceptable
level of exposure (NRC, 2000). Guidelines have var-
ied across agencies, countries, and time as a function
of the key studies and uncertainty factors used in risk
assessments (Table 13.1). The US Agency for Toxic Sub-
stances and Disease Registry (ATSDR) has derived a
minimal risk level (MRL) of 0.3 µg/kg of body weight
per day (µg/kg-bw/day), based on neurodevelopmen-
tal risks assessed in the Seychelles study (ATSDR, 1999).
In 2003, the WHO Joint Expert Committee on Food
Additives (JECFA, 2003) reduced its provisional toler-
able weekly intake (PTWI) from 3.3 to 1.6 µg/kg-bw/wk
(0.23 µg/kg-bw/day), based on results from both the Faroe
Island and Seychelles studies, after considering neuro-
development to be the most sensitive health outcome.
Health Canada (2007) uses the same PTWI for women
of childbearing age and young children, but allows a
higher ingestion rate (0.47 µg/kg-bw/day) for members
of the general population. The European Union used the
Faroe Islands and Seychelles studies to arrive at a “no
observed adverse effect level” (NOAEL) of 0.1 µg/kg-bw/
day (Mahaffey et al., 2009). It is not always clear whether
exposure guidelines are meant to apply specifi cally to
sensitive subgroups (e.g., the developing child) or the
general population.
The EPA recommends using a reference dose (RfD) of
0.1 µg/kg-bw/day, based on data from the Faroe Islands,
the Seychelles and a study of prenatal exposure in chil-
dren from New Zealand (Rice et al., 2003; Rice, 2004).
The RfD is defi ned as “an estimate (with uncertainty
spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive
subgroups) that is likely to be without an appreciable
risk of deleterious effects during a lifetime” (Rice et al.,
2003). It allows a 60-kg woman a weekly consumption
of about 4 oz (120 g) of albacore tuna (assumed to con-
tain an average of 0.35 µg of methylmercury per gram)
(US FDA, 2008a). Blood levels below 5.8 µg/L are con-
sidered to be within the RfD (NRC, 2000). Some have
argued that because the RfD is based on cord blood,
which concentrates methylmercury from maternal blood
by a factor of about 1.7, an even lower screening value for
mercury in blood (e.g., 3.5 µg/L) is warranted (Stern and
Smith, 2003; Rice, 2004; Mahaffey et al., 2009).
Total mercury concentration in whole blood is usually
indicative of exposure to methylmercury unless there is
concurrent exposure to a signifi cant source of inorganic
mercury or mercury vapor (NRC, 2000; Mahaffey et al.,
2004). Blood methylmercury is present almost entirely
in the red cells (Kershaw et al., 1980). Blood concentra-
tions of 5.8 µg/L and above (other screening values are
sometimes used) are reportable in some US states as a
means of monitoring exposure in the general popula-
tion. Germany uses a similar screening value of 5.0 µg/L
(Schulz et al., 2007b).
Scalp hair is a useful matrix for measuring methylmer-
cury exposure, particularly when the purpose is to deter-
mine cumulative or long-term exposure or to determine
exposure that occurred at sequential points of time in
the past (e.g., during specifi c trimesters of pregnancy)
(Cernichiari et al., 1995; NRC, 2000). Methylmercury
makes up more than 80% of the mercury in hair, and
the remainder is likely to have been methylmercury
that has been converted to the inorganic form in the
follicle (Cernichiari et al., 1995; Phelps et al. 1980).
Concentrations of methylmercury in hair correlate well
with concentrations in blood (Grandjean et al., 1992).
The ratio of maternal hair to maternal blood concentra-
tion is approximately 250:1, but it varies (IPCS, 1990;
Gill et al., 2002; McDowell et al., 2004). Maternal hair
mercury has been shown to be a good predictor of lev-
els in the fetal brain (Cernichiari et al., 1995). Total mer-
cury concentration in hair is not an appropriate proxy
for exposure to methylmercury in a person who works
or lives in an area with high ambient mercury concen-
tration—for example, in or around mercury mines, gold-
mining operations, or chlor-alkali plants, because it may
not be possible to remove external contamination under
these circumstances (Li et al., 2008). Hair mercury con-
centrations of 1 ppm and above have been equated with
exposure exceeding the EPA's RfD for methylmercury
(NRC, 2000).
Urine testing is preferred when measuring exposure to
inorganic mercury compounds or mercury vapor because
almost all urine mercury is of the inorganic form (IPCS,
1990; ATSDR, 1999). If spot urine samples are collected,
mercury concentrations are sometimes expressed per gram
of creatinine to account for differences in urine dilution.
In the United States, some states require physicians and
laboratories to report urine mercury levels that exceed 20
µg/L. Germany has proposed a human biomonitoring value
of 7 µg/L or 5 µg/g creatinine (Schulz et al., 2007a). Urine
concentrations decline with a mean half-time of about
90 days (Roels et al., 1991).
The best approach to treating elevated blood or urine
mercury is to remove the exposure source, since levels
drop quickly in response to removal. In cases of acute and
extremely high exposure, chelation therapies have been
used, although their effectiveness depends on selecting the
appropriate agent for the form of exposure (Flomenbaum
Clinical Assessment of Exposure to Mercury
The form and timing of suspected mercury exposure
determines the appropriate laboratory test. Conversely,
results from various laboratory tests—and different bio-
logic matrices—can help to identify an exposure source.
Mercury is typically measured in blood, urine, or hair.
It has also been measured in organ tissue (i.e., autopsy
studies), cord blood, breast milk, and other matrices for
research purposes.
