Cadmium - Toxicology of Metals - page 464

several toxic metals (i.e., combined exposure to Cd,
Hg, and Pb), and their exposure levels had decreased
constantly during the past 10 years. In a Swedish
study of an area (Skane) where exposure most prob-
ably had not decreased, Olsson et al . (2002) reported
a statistically signifi cantly increased age-adjusted
urinary beta 2 -microglobulin clearance in relation to
urinary Cd at UCd levels below 1

of dose-response relationships between urinary Cd
concentrations and the urinary excretion of low-
molecular-weight proteinuria or NAG (Oo et al ., 2000;
Suwazono et al ., 2000; Yamanaka et al ., 1998); a group of
scientists from the Kyoto Industrial Health Association
and Jichi Medical University have disputed these fi nd-
ings (Ezaki et al ., 2003; Horiguchi et al ., 2004b; Ikeda
et al ., 2000), although Ikeda et al . (2000) considered
their evidence concerning a Cd-induced effect on renal
dysfunction to be of “borderline signifi cance” (i.e.,
not a negative fi nding). In addition, Ezaki et al . (2003)
concluded that there was “no clear-cut evidence” for
Cd-induced renal dysfunction in their study of more
than 10,000 women in the Japanese general popula-
tion, but they found a statistically signifi cant ( P < 0.01)
higher level of

g/g CR. There were
no statistically signifi cant increases in other indica-
tors of renal tubular dysfunction, such as protein HC
(

µ

1-microglobulin), after adjustment for age. In another
recent study, Akesson et al . (2005) reported statisti-
cally signifi cant increases in the urinary levels of NAG
and protein HC, as well as decreases in the creatinine
clearance (from 75 to approximately 70 mL/min) and
glomerular fi ltration rate (from 100 to approximately
95 mL/min) among women in Skane, Sweden, exhib-
iting urinary Cd of 0.8

α

β 2 -microglobulin
in groups of women having the same mean age and
high (3.4

α 1 -microglobulin and

g/g CR. These changes were
statistically signifi cant when compared with women
having on average 0.48

µ g/g CR) geometric mean UCd compared

with that of those having a low (i.e., below detec-
tion limit) level of UCd. The reason why Ezaki et al .
(2003) concluded that the evidence was not clear-cut
was the absence of statistically signifi cant increases in
the number of women with

µ

µ g/g CR. Interactions with dia-

betes were indicated.
In Japan, a study of 3178 inhabitants of the Cd-pol-
luted Kakehashi River basin indicated an association
between urinary Cd concentrations and urinary

β 2 -microglobulin levels

β 2 -mg
concentrations. “Permissible values” were reported to
be 1.6-3.0

in urine as high as 1000 g/g CR in the “high”-UCd

group. In the study by Horiguchi et al . (2004b), the
geometric mean UCd value (

µ

g/g CR for
women (Hayano et al ., 1996). Because the mean urinary
Cd concentrations of adults living in the general envi-
ronment in Japan are in the range 1-4

µ g/g CR for men and 2.3-4.6 µ g/g CR) was 4.08 in the

group exhibiting the highest values. This result would
correspond to an expected increase in prevalence
of renal dysfunction of 3-4% above the background
prevalence (according to the estimates by Jarup et al .
[1998c]), and it is not surprising that the authors were
unable to detect such a small expected increase, when
they had no subgroup with UCd levels <2.6

µ

g/g CR, there
is concern about the possibility of adverse renal effects
of Cd exposure, even in so-called non-Cd-polluted
regions of Japan.
Jarup et al . (1998c) reviewed the available studies
up to 1997 on dose-response relationships between
internal doses of Cd, accumulation in renal cortex,
and renal tubular dysfunction (Table 5; Section 6.1.2).
These authors emphasized that the lowest level of Cd
in the kidneys at which renal effects can be detected
in a small percentage of the general population is
approximately 50 mg/kg, whereas the value of PCC-
10 would be approximately 125 mg/kg (6.25

µ

g/g CR
(corresponding to an expected prevalence of approxi-
mately 1% according to Jarup et al . [1998c]). This study,
therefore, is inconclusive in relation to the assessment
of risk at the present PTWI level. Uno et al . (2005), from
studies of 828 subjects living in areas in Japan with-
out any known Cd pollution, calculated the values of
LBMD-10 for increased urinary excretion of NAG and
B2M of 0.6-1.2

µ

µ g/g CR g/g CR in

women. Because they did not have a control group, it
is not completely clear how they were able to account
for such confounding factors as age and smoking; con-
fi rmatory studies are required.
In summarizing the studies of renal tubular dys-
function in Japanese areas not considered to be Cd
contaminated, it can be said that the risk of dysfunc-
tion giving rise to >1000

µ g/g CR in men and 1.2-3.6 µ

in urine). A long-term daily oral intake of 70 g of Cd

(in persons of 70-kg body weight) was suggested to
give rise to a renal dysfunction rate of 7% in a human
population, with higher percentages in particularly
sensitive groups. Furthermore, the authors asserted
that to prevent the development of renal tubular dys-
function, Cd concentrations in the renal cortex and
urine must be maintained at levels <50 mg/kg and
2.5

µ

β 2 -microglobulin
is low, but that there is evidence of a dose-response
relationship between urinary Cd levels and less pro-
nounced low-molecular-weight proteinuria. The evi-
dence is not contradictory to the estimates presented
by Jarup et al . (1998c).

µ g/g CR of

g/g CR, respectively (Table 5), and they also
recommended that the PTWI be lowered.
In investigations on residents in areas of Japan not
considered polluted by Cd, researchers from Chiba
University, Japan, have demonstrated the existence

µ