Human Health Risk Assessment - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 215

• urine, for example, Kouniali et al. ( 2003 ), for measuring benzene metabolites in

urine of adults and their children in France; Standaert et al. ( 2008 ), who measured
cadmium in urine of adults in Belgium;

• nails, for example, Rodushkin and Axelsson ( 2003 ), who measured 55 metal and

metalloids in nails of adults in Sweden;

• hair, for example, Pereira et al. ( 2004 ), who measured arsenic, cadmium, cop-

per, manganese and zinc in scalp hair of the human population living near an
abandoned pyrite mine in Alentejo, Portugal.

In extreme cases, skin tissue may be sampled. Indirect sampling of human mate-
rial entails the measurement of volatile compounds in exhaled breath (e.g., Miekisch
et al. ( 2004 ), who measured volatile organic contaminants in exhaled breath in
Germany). Walker et al. ( 2003 ) based their Risk Assessment on measurements
of PCBs and several organochlorine pesticides (such as DDT, DDEs, -HCH and

hexachlorobenzene) in maternal and umbilical cord blood plasma of several eth-
nic groups in the Canadian Arctic. Signorini et al. ( 2001 ) performed, additionally
to soil and blood sampling, a health monitoring program of the population, includ-
ing measurements of the liver function, immune function, neurological impairment,
dermatological effects, reproductive pathology, and mortality due to exposure to
2,3,7,8-tetrachlorodibenzo- p -dioxin originating from the industrial Seveso accident
in Italy, in 1976.
The big advantage of sampling body fluids or body tissue is that this gener-
ally results in more reliable data concerning the actual body burden at the time
of sampling than would any calculation. Although models exist (see Section 5.6.3
on Physiologically-Based PharmacoKinetic modelling), the calculation of internal
exposure and the subsequent excretion, metabolism and bioaccumulation is complex
and relatively unreliable.
Measuring contaminants in body fluids or body tissue often pretends to be very
reliable. This may, in turn, lead to overestimation of the value that is given to these
measurements. Due to technical difficulties, or spatial and/or temporal variation,
measurements do not per definition result in more reliable values. For risk-based
soil quality assessment purposes, Biomonitoring also has the disadvantage that
the measured contaminants in body fluids or body tissue may not only originate
from the soil, but also from other sources, for example, the diet. In addition, one
needs to account for the time-scale that is represented by the measurement. As
an example, cadmium in urine represents a measure of cumulative lifelong expo-
sure, whereas cadmium in blood represents recent exposure. Cadmium in urine thus
would reflect a person's exposure history. If exposure had changed over time due
to changes in residence conditions or overall environmental quality, an erroneous
interpretation of human health risks from the contaminated site could result. In
relation to the relevant time-scale, different choices are made for the statistical inter-
pretation of the measured concentrations in body fluids and tissues (Iversen et al.
2003 ).

β