Environmental Engineering Reference
In-Depth Information
used in the studies were approximately 100 times lower than the doses of Ni and
V. The maximum dosage ratios for Ni and V exceeded one million, one to two
orders of magnitude greater than the maximum ratios for Fe, Mn, and Zn. Minimum
dosage ratios ranged from 12 (Fe) to 1,600 (V). Therefore some studies used dos-
ages up to 10,000 times as great as others.
The majority of the toxicology studies we reviewed used relatively higher doses
of Ni and V than other metals such as As, Cu, Fe, and Mn. It is possible that the
larger number of positive associations for Ni and V may partially result from the
relatively higher dosages applied. It is also the case that there is considerable varia-
tion across studies in the magnitude of the applied dose. This factor alone may
account for some of the differences in identifying signifi cant associations.
The target metals of interest to this review account for a minor portion (<1%) of
the mass of the ambient PM in the air that most members of the U.S. population
breathe. Although transition metals comprise a much larger portion of the mass of
ROFA, people in the general population do not breathe ROFA except to the extent
that oil combustion sources contribute to local and regional air quality. Relative to
the range of metal doses that people experience in ambient air, the dose levels of
metals used in laboratory animal experiments range from hundreds to more than a
million times higher than ambient levels. It should not be assumed that biological
systems, whether animal or human, handle extremely large doses of metals (or other
chemicals) in the same way as small doses are handled. Responses observed at
extremely high levels of exposure may not be observed at lower, more relevant lev-
els of exposure when normal body defense mechanisms are functional. If applica-
tion of such high exposure levels are required to produce signifi cant increases in
infl ammatory indicator parameters (as compared to unexposed controls), then the
ability of metals to produce such results at ambient levels is in doubt.
Although there are exceptions, the majority of laboratory studies suggest that the
factors present in water-soluble extracts of PM are more biologically active than
those in the insoluble fraction; moreover, in general, greater toxicity is associated
with fi ne PM
2.5
rather than coarse PM
10
particles, when applied at high doses to the
same anatomical location in experimental animals. Exceptions to this trend are
noted in the individual study discussions. Generally, for studies in which exposures
were to ambient PM, CAPs or PM extracts, multivariate analysis methods were used
to identify individual metals that were statistically associated with the responses
being measured. However, statistical signifi cance identifi ed in multivariate analysis
does not always portend biological signifi cance. Furthermore, the individual metals
most frequently associated with infl ammatory responses in human and animal stud-
ies are also those that are present in the highest concentrations in the PM sources
used in the studies (Fe, Ni, V and Zn).
Study results with individual metals (particularly Fe, Ni, V and Zn) provide con-
vincing evidence that metals, either singly or in simple combinations, induce infl am-
matory responses when administered at high doses. The activity of major metal
constituents usually differs from the activity of the complete mixture. In biological
systems, transition metals interact in ways that are not necessarily simple or additive.
Among target metals of interest that were frequently identifi ed as being signifi cantly
Search WWH ::
Custom Search