r
vs/EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/8-88/105F
January 1992
ummary of Selected
New Information on
Effects of Ozone on
Health and Vegetation:
Supplement to 1986
Air Quality Criteria for
Ozone and Other
Photochemical Oxidants
-------
-------
EPA/600/8-88/105F
JANUARY 1992
SUMMARY OF SELECTED NEW INFORMATION
ON EFFECTS OF OZONE
ON HEALTH AND VEGETATION
Supplement
to
Air Quality Criteria for Ozone
and Other Photochemical Oxidants
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency
Office of Research and Development
Research Triangle Park, NC 27711
Printed on Recycled Paper
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
11
-------
CONTENTS
LIST OF TABLES
ABSTRACT
AUTHORS
CONTRIBUTORS AND REVIEWERS
1. SUMMARY OF SELECTED NEW INFORMATION ON EFFECTS OF
OZONE ON HEALTH AND VEGETATION
1.1 INTRODUCTION
REFERENCES , . .
2. EFFECTS OF OZONE ON VEGETATION
2.1 STUDIES RELEVANT TO SELECTION OF THE
AVERAGING TIME FOR THE SECONDARY NAAQS
FOR OZONE
2.2 SUMMARY AND CONCLUSIONS: VEGETATION
EFFECTS
2.2.1 Exposure Duration
2.2.2 Peak Concentrations
2.2.3 Comparison of Exposure Indices
2.2.4 Evaluation of the 7-Hour (or 12-Hour)
Seasonal Mean
REFERENCES
3. EFFECTS OF OZONE ON HEALTH
3.1 HEALTH STUDIES RELEVANT TO SELECTION OF
THE PRIMARY NAAQS FOR OZONE
3.1.1 Human Clinical Studies
3.1.2 Field and Bpidemiological Studies
3.1.3 Laboratory Animal Studies
3.1.3.1 Effects of Multihour Exposures
3.1.3.2 Effects of Multiday Exposures
3.1.3.3 Effects of Chronic Exposure to Ozone
3.1.3.4 Animal-to-Human Extrapolation . .
3.2 SUMMARY AND CONCLUSIONS: HEALTH EFFECTS
DATA
3.2.1 Exposure Dynamics for Short-Term Ozone Exposure
Effects
3.2.2 Evaluation of Differential Susceptability of
Potential Special Risk Groups
3.2.3 Ozone Impacts on Lung Structure/Chronic Disease
Processes
3.2.4 Ozone Dosimetry Aspects .
REFERENCES
IV
v
vii
vii
1-1
1-1
1-2
2-1
2-1
2-9
2-9
2-10
2-11
2-11
2-13
3-1
3-1
3-1
3-27
3-43
3-43
3-45
3-46
3-52
3-56
3-57
3-61
3-62
3-64
3-65
111
-------
Number
LIST OF TABLES
3-1
3-2
3-3
3-4
3-5
3-6
Controlled Human Exposure Laboratory Studies Relevant to
Review of the 1-Hour NAAQS for Ozone
Key Human Studies Demonstrating Lung Function
Decrements Near the Current 1-Hour NAAQS for Ozone .
Field and Epidemiologic Studies on Effects of Ozone . . .
Experimental Animal Studies on the Relative Influence
of Ozone Concentration and Duration of Exposure . . .
Chronic Ozone Effects in Experimental Animals
Studies Relevant to Potential Animal-to-Human
Extrapolations ,
3-3
3-10
3-28
3-44
3-47
3-53
IV
-------
ABSTRACT
Since completion of the 1986 air quality criteria document for ozone, additional
information has become available that warrants consideration by the U.S. Environmental
Protection Agency (U.S. EPA) in its review of the National Ambient Air Quality Standards
(NAAQS) for ozone. This summary reviews and evaluates selected literature published from
1986 through early 1989 on the vegetation and health effects resulting from exposure to
ozone. Emphasis has been placed on evaluation of key human health effects literature and
other data most pertinent to determination by U.S. EPA of the appropriate level and form of
the primary NAAQS and the appropriate form of the secondary NAAQS.
-------
-------
AUTHORS
Dr. Daniel L. Costa, Health Effects Research Laboratory, OHR, ORD, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711.
Dr. Lawrence J. Folinsbee, C. E. Environmental, Inc., Suite 200, 800 Eastowne Drive,
Chapel Hill, North Carolina 27514.
Mr. James A. Raub, Environmental Criteria and Assessment Office, MD-52, OHEA, ORD,
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711.
Ms. Beverly Tilton, Environmental Criteria and Assessment Office, MD-52, OHEA, ORD,
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711.
Dr. David T. Tingey, Environmental Research Laboratory, OEPER, ORD,
U.S. Environmental Protection Agency, 200 S.W. 35th Street, Corvallis,
Oregon 97333.
CONTRIBUTORS AND REVIEWERS
Dr. Robert S. Chapman, Dr. Timothy R. Gerrity, Dr. Carl G. Hayes, Dr. Donald H.
Horstman, Dr. Hillel S. Eoren, Dr. William F. McDonnell, Dr. Frederick J. Miller, and
Dr. John H. Overton, Health Effects Research Laboratory, OHR, ORD, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711.
Dr. Lester D. Grant and Dr. Judith A. Graham, Environmental Criteria and Assessment
Office, OHEA, ORD, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina 27711.
Dr. Milan J. Hazucha, Center for Environmental Medicine and Lung Biology, The University
of North Carolina, Chapel Hill, North Carolina 27514.
Dr. William Hogsett, Environmental Research Laboratory, OEPER, ORD,
U.S. Environmental Protection Agency, 200 S.W. 35th Street, Corvallis, Oregon 97333.
Dr. Allan Marcus, Battelle-Applied Statistics, 200 Park Drive, P.O. Box 12056, Research
Triangle Park, North Carolina 27709.
Dr. James H. Ware, Harvard University, School of Public Health, Department of
Biostatistics, 677 Huntington Avenue, Boston, Massachusetts 02115.
VII
-------
-------
1. SUMMARY OF SELECTED NEW INFORMATION
ON EFFECTS OF OZONE ON HEALTH
AND VEGETATION
1.1 INTRODUCTION
The U.S. Environmental Protection Agency (U.S. EPA) document, Air Quality Criteria
for Ozone and Other Photochemical Oxidants, completed in August 1986 provided
comprehensive evaluation of the relevant scientific literature on ozone published through mid-
1986 (U.S. Environmental Protection Agency., 1986). The criteria document was prepared
by the Office of Research and Development for use as the scientific basis for decision
making by the Agency regarding retention or revision of primary and secondary National
Ambient Air Quality Standards (NAAQS) for ozone.
Since completion of the 1986 document, additional information has become available
that warrants consideration by the Agency in its review of the NAAQS for ozone. As a
supplement to the 1986 document, this summary reviews and evaluates published literature
concerning exposure- and concentration-response relationships observed for vegetation effects
and for health effects in humans and experimental animals. Emphasis has been placed on
evaluation of new human health effects literature and other data most directly pertinent or
useful for determining the appropriate level and form of the primary standard and the ,
appropriate form of the secondary standard. Selected important data on dosimetry and on
experimental animal studies that elucidate concentration x time (duration) exposure-response
relationships have also been included. As required by the Clean Air Act, an earlier external
review draft of this Supplement (U.S. Environmental Protection Agency, 1988) was publicly
peer-reviewed by the Clean Air Scientific Advisory Committee (CASAC) of U.S. EPA's
Science Advisory Board (McClellan, 1989).
The publications reviewed and evaluated in this supplement were selected from
approximately 500 new articles and abstracts on the health effects of ozone and from about
300 new articles and abstracts on the vegetation effects of ozone that appeared as peer-
reviewed journal publications or as proceedings papers from 1986 through early 1989. Since
1-1
-------
tile time of review by CASAC, several of the abstracts and articles previously cited in the
draft supplement as "in press" or "submitted for publication" have been published. In such
cases, the citation has been updated in this final version of the supplement, including several
revised citations of 1989 (and a few later) publications.
REFERENCES
McClellan, R. O. (1989) [Letter of CASAC closure on the 1988 ozone criteria document supplement to
William K. Reilly, Administrator, U.S. Environmental Protection Agency]. Washington, DC:
U.S. Environmental Protection Agency, Clean Air Scientific Advisory Committee; May 1.
U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report nos. EPA-600/8-84-020aF-ef. Available from: NTIS, Springfield,
VA; PB87-142949.
U.S. Environmental Protection Agency. (1988) Summary of selected new information on effects of ozone in
health and vegetation: draft supplement to air quality criteria for ozone and other photochemical
oxidants. Research Triangle Park, NC: Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office; EPA report no. EPA-600/8-88/105A. Available from:
NTIS, Springfield, VA; PB89-135123.
1-2
-------
2. EFFECTS OF OZONE ON VEGETATION
2.1 STUDIES RELEVANT TO SELECTION OF THE AVERAGING
TIME FOR THE SECONDARY NAAQS FOR OZONE
A review by Hogsett et al. (1988) discusses the biological, environmental, and
exposure-dynamic factors (e.g., concentration, duration, frequency, threshold, respite time)
that influence the magnitude of the biological responses of plants. These factors contribute to
observed variability in responses, and thus become considerations in measures of exposure
that best describe plant response to pollutant exposure. The various types of exposure indices
that have been used historically to describe pollutant exposure were also evaluated. The
ultimate goal of investigations of factors influencing plant response is to develop exposure
indices that account for all of the variation in the exposure-response relationship. However, a
second and more practical goal is that of developing or specifying indices useful for standard
setting. An index for a standard should be simple, not site-specific, and as generic as
possible.
Musselman et al. (1986) examined the influence of two different patterns of ozone (O3)
exposure on beans. The studies were conducted in a greenhouse and the plants were exposed
to either a simulated ambient or a»uniform O3 concentration. The simulated ambient
exposures followed the ambient exposure patterns of Riverside, CA (O3 concentration range:
0.058 to 0.40 ppm; peak exposure duration: 0.5 to 1.5 h; and total exposure duration: 6 h).
The uniform distribution was selected to match the total dose and peak concentration of the
simulated ambient exposure (O3 concentration: 0.30 or 0.40 ppm; exposure duration:
2.3 h). Exposures occurred weekly, and the plants received one, two, or three exposures
before being harvested 6 days after their last exposure. Both O3 exposures induced foliar
injury and reduced plant growth; and the effects of the two distributions were not statistically
different. Consequently, if the maximum concentrations and "total doses" are equal, peak
shape appears not to be an important variable.
Kohut et al. (1988) examined the effect of peak concentration and exposure frequency
on the responses of kidney beans to O3. The plants were exposed to one of four O3 exposure
regimes: (1) constant exposure to 0.05 ppm daily; (2) fluctuating exposure to 0.08 ppm on
2-1
-------
alternate days (Monday, Wednesday, and Friday); (3) cluster exposure to 0.08 ppm on three
consecutive days (Wednesday, Thursday, and Friday); and (4) exposure to 0.12 ppm on two
consecutive days (Thursday and Friday) or charcoal-filtered air. The exposure duration was
4 h and yielded weekly mean concentrations between 0.05 and 0.06 ppm. The test plants
were grown in pots and exposed to O3 in open-top chambers under field conditions. Plants
were harvested weekly throughout the study. Although there were two replicates of each
O3 treatment, the experiment was not replicated in time. In the early harvests,, the plants
receiving the peak exposures were significantly impaired. By the final harvest (12 weeks),
however, there were no significant effects of O3 on any of the plant growth or yield
parameters. The authors concluded that the plants integrated the impacts, and, consequently,
that "...the response of the plants was related to the mean rather than the peak concentration
of exposure." This conclusion is difficult to substantiate with the data, as none of the
O3 exposures produced a significant effect at the final harvest. It is therefore not possible to
determine whether the various treatments differentially affected plant response.
Heagle et al. (1986) studied the responses of soybeans to chronic doses of ozone applied
in two different ways. Plants were grown in the field using standard National Crop Loss
Assessment Network (NCLAN) methodology except for the way in which the O3 was
dispensed. In one set, various constant amounts of O3 were added to the ambient air for
7 h/day; for the second set, the O3 was increased above the ambient air by proportional
amounts. Although there was a constant addition of O3 to the ambient air, in the constant-
addition treatments, the resultant exposure regime was not "square wave" because the
O3 concentration in the ambient air varied throughout the exposure. The principle effect of
the constant addition or proportional addition treatments was to create various levels of
exposures. The O3 concentrations were expressed as the 7-h seasonal means. The authors
concluded that the two different types of exposure regimes produced identical responses.
Several trends in the data, however, cast doubt on the validity of this conclusion. The
authors clearly state that the proportional additions caused the more frequent occurrence of
elevated concentrations, but their exposure index (the 7-h seasonal mean) failed to
characterize or reflect this elevated exposure. In fact, the 7-h means for the proportional
additions were lower than those for the comparable constant-addition treatments. The authors
also reported that the slope of the "dose-response curve" for the proportional additions was
2-2
-------
greater than for the constant additions. The authors speculated that an extension of the dose
range might have shown significant results. The failure of the 7-h seasonal mean to
adequately characterize the higher concentrations is not surprising because a previous paper
(Cure et al., 1986) from the same group states that the 7-h seasonal mean was selected
specifically because it was less sensitive to variations in O3 patterns. Consequently, the
conclusions of the authors must be viewed with skepticism.
Heagle et al. (1987) also evaluated the influence on tobacco yield of daily O3 exposure
duration and fluctuations in concentrations. Plants were grown in the field using standard
NCLAN methodology except for the Cydispensing protocol. In one set of studies, various
constant amounts of O3 were added to the ambient air for 7 h/day; to the second set, the
O3 was increased above the ambient air by proportional amounts. In addition, the study
compared the effects of 7- and 12-h exposures on tobacco yield. The O3 concentrations were
expressed as the 7- and 12-h seasonal means. Yield was reduced to a greater extent by 12-h
than by 7-h exposures. The authors concluded that the two different types of 7-h exposure
regimes (7-h constant and 7-h proportional) produced identical responses.
Additional analyses of the soybean (Heagle et al., 1986) and tobacco (Heagle et al.,
1987) data sets were carried out by Rawlings et al. (1988) to evaluate various exposure
indices and the influence of exposure duration on plant response. The results from the
soybean data and the 12-h studies with tobacco suggested that the peaks should be given
greater weight. In contrast, the 7-h studies with tobacco suggested that the arithmetic mean
was sufficient and that the peaks did not require additional weighting. Rawlings et al.
acknowledged that these results must be viewed with caution, because the differences in
exposure profiles between the constant and proportional O3 additions were relatively small,
thus limiting the power of the experiment for determining the "best" exposure index. This
same caveat also applies to the conclusions reached by the authors of the soybean (Heagle
et al., 1986) and tobacco (Heagle et al., 1987) studies. The analysis of exposure duration
found that 12-h exposures caused greater effects than 7-h exposures (Rawlings et al., 1988).
The negative impact of the exposures did not increase linearly with exposure duration (i.e.,
the decrease in yield loss was not directly proportional to the increased length of exposure).
In a study by Adomait et al. (1987), white beans (Phaseolus vulgaris) were grown in
field plots throughout southern Ontario, Canada. Plants at each location were treated with a
2-3
-------
chemical protectant, ethylenediurea (EDU), to reduce or eliminate the impact of O3 on yield,
which was determined as the difference between the yields of EDU and non-EDU treated
plots at each location. Ozone exposure was expressed as the cumulative O3 concentrations
above a threshold of 0.08 ppm for the month of August. Yield decreased as the cumulative
O3 concentration increased. The addition of temperature and rainfall to the regression
equation, in an attempt to approximate O3 flux into the plant, significantly improved the fit
of the regression equation to the data. To express the experimental results, the authors
assumed that the elevated O3 concentrations (peaks) were important and that the impact was
the cumulative result of multiple exposures.
Data used in a study by Cure et al. (1986) were generated using standard open-top
chamber NCLAN protocols. The study did a three-way comparison of relationships among
the 7-h seasonal mean, the 1-h seasonal mean, and the 1-h maximum for the season. For two
of the three comparisons, the 7-h/l-h ratio was essentially constant, suggesting that these two
variables differed by a constant. For the other comparison, the ratio was less stable. The
authors concluded, however, that the 7-h and 1-h seasonal means were surrogates for each
other. The ratio of the 1-h maximum to the 7-h seasonal mean was highly unstable, which
suggests that the maximum was poorly related to the long-term mean. The authors selected a
seasonal mean for two reasons: (1) They assumed that crop yield reductions resulted from an
accumulation of daily O3 effects over the growing season; and (2) the seasonal means were
much less sensitive to peak variations in yearly O3 patterns, especially at concentrations near
the current ambient levels.
In a study by McCool et al. (1986), plants grown in a standard soil were exposed to a
range of O3 levels in closed-top exposure chambers and the yields were determined. The
authors developed yield-loss functions that related decreased crop yield to a cumulative
exposure index for 12 crops. Ozone exposure was characterized as the cumulative
concentrations greater than 0.10 ppm. The concentration threshold (0.10 ppm) was chosen
because it was the California state O3 standard. A threshold concentration for O3 was used
to avoid giving equal mathematical weight to the numerous low concentrations and to ignore
the low nighttime background in calculating the exposure.
In a field study using closed-top exposure chambers, McCool et al. (1987) assessed the
impact of O3 on four vegetables (turnip, beet, onion, lettuce). The exposure-response
2-4
-------
functions were best described as a linear function with increasing exposure. Both the sum of
i
the concentrations >0.10 ppm and the 12-h seasonal mean concentration were used in
developing the exposure-response functions. Neither exposure index was uniformly best.
A 3-year field study was conducted by Smith et al. (1987) in which the effects of O3 on
foliar injury and yield were assessed using the chemical protectant EDU. Ozone exposure
was characterized as the 7-h seasonal mean and as the cumulative exposure (using various
concentration thresholds). The EDU treatment did not significantly enhance crop yield.
Yield and foliar injury, respectively, were similar among cultivars and over years. Although
the ambient O3 exposures between 1983 and 1984 were substantially different, as indicated by
the various cumulative statistics, this difference was not reflected in the 7-h seasonal mean.
These data are another example of the lack of sensitivity of the mean to temporal variations
in O3 exposures.
Open-top chambers were used by Wang et al. (1986a) in a field study to examine the
effects of ambient O3 on the growth and foliar injury of three tree species. Ozone was
characterized as the number of daily occurrences above 0.08 and 0.12 ppm. The authors
concluded that O3 significantly impaired the growth of hybrid poplar in the absence of visible
foliar injury. There were 20 days when O3 exceeded 0.08 ppm and 1 day when the
concentration exceeded 0.12 ppm. In a 3-year study with quaking aspen, Wang et al.
(1986b) found that plant growth was reduced 12 to 24%. In only one of the years was the
current National Ambient Air Quality Standard for O3 exceeded. The observations that
growth reductions can occur in the absence of the ambient O3 concentration exceeding the
level of the current standard are consistent with the recent analysis of Lee et al. (1989),
which forecast significant effects on- crop yield when the standard was not exceeded.
Only a limited number of studies have been conducted with the specific objective of
developing or evaluating various exposure indices; several studies have reanalyzed existing
exposure-response data to evaluate a range of exposure indices. The results of these
retrospective analyses have provided useful concepts and their conclusions are in general
agreement. Because the experiments analyzed were not specifically designed to evaluate
various indices, the differences among the actual exposure treatments (frequency of
O3 occurrences) may be relatively small. Consequently, the power of these studies is less
than desirable.
2-5
-------
Reanalysis of several NCLAN data sets (soybean and wheat from Argonne, IL; cotton
from Shafter, CA; and alfalfa from Corvallis, OR) was performed by Lee et al. (1987) using
various mean and cumulative peak-weighted exposure indices (e.g., concentration threshold
and functional peak weighting). Exposure indices that included all the data (24 h) performed
better than those that used only 7 h of data. The 7-h seasonal mean was never "best" and
was near optimal in only 5 of 14 cases. From a modeling standpoint, the exposure indices
that emphasized peaks performed better than those that gave equal weighting to all
concentrations; indices that accumulated the exposures performed better than those that
averaged the exposures.
In a more extensive retrospective analysis of NCLAN data, Lee et al. (1988) fit
24 common and 589 general phenologically weighted, cumulative-impact (GPWCI) exposure
indices to the response data from seven crop studies (2 years of data for each). The "best"
exposure indices were those that displayed the smallest residual sums of square error when
the yield response data were regressed on the various O3 exposure indices using the Box-
Tidwell model. The "best" exposure index was a GPWCI with sigmoid weighting on
concentration and a gamma weighting function as a surrogate for changes in plant sensitivity
over time. Cumulative indices (with concentration thresholds) performed as well as the
GPWCIs, whereas mean indices did not perform as well. The general conclusions of the
authors were, "While no single index was deemed "best" in relating ozone exposure to plant
response, the top-performing indices were those indices that (1) cumulated the hourly ozone
concentrations over time, (2) used a sigmoid weighting scheme which emphasizes
concentrations of 0.06 ppm and higher, and (3) phenologically weighted the exposure such
that the greatest weight occurs during the plant growth stage. These findings illustrate the
importance of the duration of exposure, the importance of repeated peaks, and the time of
increased sensitivity in assessing the impact of ozone on plant growth." Although peak
concentrations should be given greater weight, the authors suggested that lower concentrations
should also be included but given lesser weight in the calculation of an exposure index.
The paper by Tingey et al. (1989) is essentially a condensation of the paper by Lee
et al. (1988) and therefore the conclusions are basically the same. However, the paper does
show the importance of exposure duration in influencing the magnitude of plant response and
the limitation of the seasonal mean to specifically incorporate varying exposure durations.
2-6
-------
For example, the mean cannot distinguish among exposures to the same average
concentrations over different durations (e.g., for 10, 50, or 100 days).
Wheat and soybean data sets (Kohut et al., 1987, 1986) collected using standard
NCLAN protocols at Ithaca, NY, were reanalyzed by Lefohn et al. (1988a) to compare
exposure indices. The authors used the 7-h mean and cumulative statistics with thresholds,
or peak weighting. The data were fit with both linear and Weibull response models. No one
exposure statistic was best for all data sets or response models. The linear model showed no
strong tendency to fit any exposure index; however, a peak-weighted statistic and the number
of occurrences >0.08 ppm or the sum of the concentrations >0.08 ppm had a higher
R2 than the 7-h seasonal statistic. When the Weibull model was used, the cumulative
statistics performed better than the seasonal means. The authors also concluded that a
sigmoid peak-weighted scheme was better than a threshold approach because it included the
effects from the concentrations below the selected threshold concentration but gave them less
weight.
The paper by Lefohn et al. (1988a) has engendered discussion in the literature about the
interpretation of the data (Runeckles, 1988; Parry and Day, 1988). Both groups of
respondents thought that the paper contained insufficient data and evidence to support the
conclusion that peak-weight exposure indices should be used in developing exposure-response
functions. However, Runeckles tempered his criticism with the observation that peak-
weighted indices performed at least as well as mean indices. Also, the respondents criticized
the compilation of the 2 years of wheat data into a single model when the exposure durations
were markedly different. In response, Lefohn et al. (1988b) stated that the wheat data
support the need to include a cumulative component in an exposure index. They concluded
that, "The cumulative index is more relevant to use in the standard-setting process than
seasonal means, which ignore the length of the exposure period."
Musselman et al. (1988) conducted a retrospective analysis of crop loss data originally
collected by Oshima et al. (1976) (see U.S. Environmental Protection Agency, 1986). The
analysis was based on data for five crops, but those data were not replicated in time. The
plants were grown in pots in standardized soil and were provided with adequate water and
nutrients. The plants were placed at 9 to 12 sites along an ambient O3 concentration gradient
in the Los Angeles Basin. The crop loss data were originally summarized by Oshima et al.
2-7
-------
(1976) on the basis of the cumulated concentration above 0.10 ppm for the growing season.
In this study, the authors tested (1) various peak indices, (2) daily mean indices, and
(3) indices based on subsets of a 24-h day. No single exposure index was "best" for the five
crops. Ozone indices utilizing a concentration threshold level performed well for most
crops, but the optimum threshold level varied with the particular index calculated. Some of
the "better" indices were (1) seasonal mean of concentrations above 0.09, 0.12, or 0.14 ppm;
(2) mean of all daily peak concentrations; (3) sum of the daily peaks squared above a
concentration of 0.15 ppm; and (4) total number of seasonal peaks above 0.12 ppm. The
7- or 12-h seasonal means were not among the better-performing indices.
Larsen et al. (1988) evaluated 14 O3 exposure indices for their ability to predict crop
yield loss. The second highest daily maximum concentration and 13 other indices, including
the effective mean O3 concentration and the summer daytime average (M7), were calculated
for 80 "agricultural" National Aerometric Data Bank sites and for multiple years, for a total
of 320 site-years. In contrast to other retrospective analyses, separate exposure-response
functions were not derived from biological data for each exposure index. Larsen et al.
(1988) used ambient air monitoring data to derive correlations between the effective mean
and other air quality indices. These correlations (based only on ambient air monitoring data)
were used to express the plant response data in terms of the different indices (i.e., the
lognormal model that expressed crop reduction as a function of the effective mean
concentration [Larsen and Heck, 1984] was used to generate crop loss estimates for the
320 site-years of ambient data). Because there was no-biological variation in the data,
correlations between the exposure indices and estimated crop reductions were, in fact,
measures of association between the (transformed) effective mean and the other indices.
Consequently, no evidence that the mean indices were better correlated with plant response
than other indices can be inferred from the analysis of Larsen et al. (1988).
Larsen et al. (1983) developed an exposure-response model that relates O3 impact on
plants to a cumulative index that they denoted as the total impact. A 75-day exposure for
7 h/day was originally assumed for calculating the estimated crop reduction for soybean (this
may not be representative of the phenological life span of soybean). In the original analysis,
the effective mean was not calculated from the hourly O3 concentrations for the NCLAN
studies but was estimated by multiplying M7 by 1.15 for charcoal-filtered and nonfiltered
2-8
-------
exposures (and supplementing the nonfiltered exposures with the constant additions for other
exposures). Further, treatment means rather than chamber means were used in estimating the
lognormal model. Consequently, this lognormal model is inaccurate and needs to be
estimated more precisely for use in the selection of exposure indices for use as a O3 standard.
The adequacy of the lognormal model using other exposure indices must also be determined.
Reich and Amundson (1985) have reviewed a series of field and controlled-environment
studies to assess the impact of O3 on photosynthesis. The authors stated, "...it may be
inappropriate as well as difficult to compare directly the response of the species on the basis
of a mean O3 exposure concentration. However, when the responses are compared on the
basis of a unit dose of O3, the results are more easily interpreted." A unit dose of O3 as
defined by the authors means cumulative exposure (i.e., total parts per million). The
O3-induced decrease in growth was directly related to reduced photosynthesis, which was
decreased by the cumulative O3 exposure.
2.2 SUMMARY AND CONCLUSIONS: VEGETATION EFFECTS
Recent literature concerning the appropriate averaging time for an exposure index for
O3 effects on vegetation was evaluated in relation to (1) the role of exposure duration, (2) the
role of peak O3 concentrations, (3) comparison of exposure indices, and (4) evaluation of the
7-h seasonal mean.
2.2.1 Exposure Duration
Increasing the duration in the exposure index from a 7-h seasonal mean to a 12-h
seasonal mean caused a greater decrease in yield. A comparison of 7- and 24-h exposure
indices showed that 24-h indices provided an even better statistical fit to the exposure-
response data. Although plant effects increased with exposure duration, the study of
Rawlings et al. (1988) showed that the increase in plant response was not proportional to the
increase in exposure duration.
All the recent studies that used impaired plant growth or yield as an adverse effect
specifically selected exposure indices that might reflect the cumulative impact of effects
throughout the growing season. For example, Cure et al. (1986) stated that crop yield
2-9
-------
reductions result from an accumulation of daily O3 effects over the growing season. Reich
and Amundson (1985) stated that the O3-induced decrease in growth was directly related to
reduced photosynthesis, which was impaired by the cumulative O3 dose. These data can be
interpreted to mean that growth and yield are reduced by repeated O3 episodes, because that
is how O3 occurred in the studies and how it occurs in nature.
These studies support the conclusion that a cumulative O3 exposure index is needed that
reflects the total exposure that the plant experiences. This conclusion is consistent with the
1986 U.S. EPA criteria document (U.S. Environmental Protection Agency, 1986), which
states, "When plant yield is considered, the ultimate impact of an air pollutant on yield
depends on the integrated impact of the pollutant exposures during the growth of the plant."
By inference or deduction, then, a mean of unspecified time (days or months) is
inappropriate because the lack of specification of "time" results in a variable duration of
O3 accumulation.
2.2.2 Peak Concentrations
Most of the recent studies, except for papers reporting results of the NCLAN program,
selected exposure indices that cumulated the exposure and preferentially weighted the peaks.
Three main peak-weighting approaches have been used: (1) a concentration threshold
approach, hi which the concentrations above the selected threshold are summed or in which
the number of days or hours above the concentration threshold are summed; (2) an allometric
or exponential weighting, in which all concentrations are raised to a specific exponential
power; and (3) a sigmoid weighting, in which all concentrations are weighted with a
multiplicative weighting factor (which depends on concentration).
The threshold weighting approach assumes that only the concentrations above the
selected concentration threshold are biologically active. Recent studies have shown that the
concentrations below the selected concentration threshold or cutoff may also have biological
importance. It is also likely that the appropriate concentration threshold differs between
species, with environmental conditions, and with endpoint measured.
Functional weighting approaches using either an allometric or sigmoid weighting are
preferred to the concentration threshold approach. These approaches do not censor
concentrations, but rather give weight, although not equal, to all concentrations in eliciting a
2-10
-------
biological response. Specific comparisons of the functional weighting to the threshold
approaches showed that they yielded better statistical fits to the data. Also, the sigmoid
weighting functions appeared to perform better than the allometric weighting approach.
The conclusions found in recent literature regarding the importance of cumulative peak
concentrations in causing vegetation responses are consistent with the data and conclusions
presented in the 1986 criteria document (U.S. Environmental Protection Agency, 1986).
2.2.3 Comparison of Exposure Indices
There have been several studies (Lee et al., 1987, 1988, 1989; Lefohn et al., 1988a;
Musselman et al., 1988) conducted that were retrospective analyses of existing plant-response
data sets. These authors evaluated the relative efficacy of existing and proposed exposure
indices, using a large number of crop data sets. The exposure indices included various means
and cumulative statistics using both threshold and functional concentration weighting. The
authors concluded that there was no single exposure index that was best for all crop species
or for all data sets. These studies are all in agreement, however, that (1) mean indices are
not among the best indices and (2) the preferred (yielded best statistical fit to the data)
exposure indices cumulated the exposure impact over the growing season and preferentially
weighted the peak concentrations.
2.2.4 Evaluation of the 7-Hour (or 12-Hour) Seasonal Mean
The 7-h seasonal mean is the most commonly used exposure index in the literature
reviewed in the 1986 U.S. EPA criteria document (U.S. Environmental Protection Agency,
1986), and it continues to be used by investigators. Mathematically, the mean contains all
hourly concentrations making up the exposure period and treats all concentrations equally,
thus implying that (1) all concentrations of O3 (across the range of concentrations to which
plants are exposed in a growing season) are equally effective in causing a response, and
(2) the contributions of the peak concentrations to the response are minimal. The mean treats
low-level, long-term exposures the same as high-concentration, short-term exposures, a
scenario that the literature does not support (e.g., the 1986 U.S. EPA criteria document). An
infinite number of hourly distributions, from those containing many peaks to those containing
none, can yield the same 7-h seasonal mean. Cure et al. (1986) reported that mean
2-11
-------
characterizations of O3 exposure were much less sensitive than the daily 1-h maximum to
variations in yearly O3 patterns. Also, Reich and Amundson (1985) stated, "...it may be
inappropriate as well as difficult to compare directly the response of species on the basis of a
mean O3 exposure concentration. However, when the responses are compared on the basis of
a unit dose of O3, the results are more easily interpreted."
The use of a mean exposure index for characterizing exposures implies certain
assumptions.
1. A seasonal mean assumes that crop yield reductions result from the
accumulation of daily O3 effects over the growing season (Cure et al
1986).
2. A mean assumes that the distribution of hourly O3 concentrations (over the
averaging time) are not highly skewed and that the distribution is unimodal.
In the ambient, the O3 concentration distributions are frequently skewed
toward the higher concentrations.
3. The mean weights all concentrations within the selected averaging time
equally.
4. The mean does not specifically include an exposure duration component; it
cannot distinguish between two exposures to the same concentration but of
different durations (e.g., 50 or 100 days).
5. The mean assumes that the selected time interval, over which the
concentrations are averaged, is the period of highest hourly occurrences of
O3 or any other pollutant being examined.
6. The mean index assumes that peak events do not need to be given special
consideration. This is not consistent with results showing that short-term
peak concentrations are important in determining vegetation response (see,
e.g., the 1986 U.S EPA criteria document).
The correlation between the 7-h seasonal mean (M7) and the second-highest daily
maximum 1-h concentrations (i.e., HDM2, the current O3 standard) was low (r = 0.54) due
to the insensitivity of peak concentrations in the M7 calculation (Lee et al., 1989). A wide
range of temporal distributions with HDM2 between 0.06 and 0.24 ppm was found at sites
with M7 values between 0.036 and 0.048 ppm. Temporal distributions of ambient O3 data at
83 nonurban sites showed large spatial differences across these sites, with the HDM2 ranging
2-12
-------
from 0.06 to 0.24 ppm. In contrast, the 7-h seasonal mean (M7 calculated from May to
September) across the 83 sites showed small differences (i.e., 90% of the sites had M7 values
between 0.03 and 0.06 ppm).
REFERENCES
Adomait, E. J.; Ensing, J.; Hofstra, G. (1987) A dose-response function for the impact of O3 on Ontario-grown
white bean and an estimate of economic loss. Can. J. Plant Sci. 67: 131-136.
Cure, W. W.; Sanders, J. S.; Heagle, A. S. (1986) Crop yield response predicted with different
characterizations of the same ozone treatments. J. Environ. Qual. 15: 251-254.
Heagle, A. S.; Lesser, V. M.; Rawlings, J. O.; Heck, W. W.; Philbeck, R. B. (1986) Response of soybeans to
chronic doses of ozone applied as constant or proportional additions to ambient air. Phytopathology
76: 51-56. '
Heagle, A. S.; Heck, W. W.; Lesser, V. M.; Rawlings, J. O. (1987) Effects'of daily ozone exposure duration
and concentration fluctuation on yield of tobacco. Phytopathology 77: 856-862.
Hogsett, W. E.; Tingey, D. T.; Lee, E. H. (1988) Ozone exposure indices: concepts for development and
evaluation of their use. In: Heck, W. W.; Taylor, O. C.; Tingey, D. T., eds. Assessment of crop loss
from air pollutants: proceedings of an international conference; October 1987; Raleigh, NC. New
York, NY: Elsevier Applied Science; pp. 107-138.
Kohut, R. J.; Amundson, R. G.; Laurence, J. A. (1986) Evaluation of growth and yield of soybean exposed to
ozone in the field. Environ. Pollut. Ser. A 41: 219-234. ;
Kohut, R. J.; Amundson, R. G.; Laurence, J. A.; Colavito, L.; Van Leuken, P.; King, P. (1987) Effects of
ozone and sulfur dioxide on yield of winter wheat. Phytopathology 77: 71-74.
Kohut, R. J.; Laurence, J. A.; Colavito, L. J. (1988) The influence of ozone exposure dynamics on the growth
and yield of kidney bean. Environ. Pollut. 53: 79-88.
Larsen, R. I.; Heck, W. W. (1984) An air quality data analysis system for interrelating effects, standards, and
needed source reductions: part 8. an effective mean O3 crop reduction mathematical model. J. Air
Pollut. Control. Assoc. 34: 1023-1034.
Larsen, R. L; Heagle, A. S.; Heck, W. W. (1983) An air quality data analysis system for interrelating effects,
standards, and needed source reductions: part 7. an O3-SO2 leaf injury mathematical model. J. Air
Pollut. Control Assoc. 33: 198-207.
Larsen, R. L; McCurdy, T. R.; Johnson, P. M.; Heck, W. W. (1988) An air quality data analysis system for
interrelating effects, standards, and needed source reductions: part 10. potential ambient O3 standards
to limit soybean crop reduction. JAPCA 38: 1497-1503.
Lee, E. H.; Tingey, D. T.; Hogsett, W. E. (1987) Selection of the best exposure-response model using various
7-hour ozone exposure statistics. Research Triangle Park, NC: U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards.
2-13
-------
Lee, E. H.; Tingey, D. T.; Hogsett, W. E. (1988) Evaluation of ozone exposure indices in exposure-response
modeling. Environ. Pollut. 53: 43-62.
Lee, E. H.; Tingey, D. T.; Hogsett, W. E. (1989) Interrelation of experimental exposure and ambient air
quality data for comparison of ozone exposure indices and estimating agricultural losses. Corvallis,
OR: U.S. Environmental Protection Agency, Environmental Research Laboratory; EPA report no.
EPA-600/3-89-047. Available from: NTIS, Springfield, VA; PB89-195Q36.
Lefohn, A. S.; Laurence, J. A.; Kohut, R. J. (1988a) A comparison of indices that describe the relationship
between exposure to ozone and reduction in the yield of agricultural crops. Atmos. Environ.
22: 1229-1240.
Lefohn, A. S.; Laurence, J. A.; Kohut, R. J. (1988b) Authors' reply [A response to comments by Runeckles,
1988b]. Atmos. Environ. 22: 1242-1243.
McCool, P. M.; Musselman, R. C.; Teso, R. R.; Oshima, R. J. (1986) Determining crop yield losses from air
pollutants. Calif. Agric. 40(July-August): 9-10.
McCool, P. M.; Musselman, R. C.; Teso, R. R. (1987) Air pollutant yield-loss assessment for four vegetable
crops. Agric. Ecosyst. Environ. 20: 11-21.
Musselman, R. C.; Huerta, A. J.; McCool, P. M.; Oshima, R. J. (1986) Response of beans to simulated
ambient and uniform ozone distributions with equal peak concentration. J. Am. Soc. Hortic. Sci.
Ill: 470-473.
Musselman, R. C.; McCool, P. M.; Younglove, T. (1988) Selecting ozone exposure statistics for determining
crop yield loss from air pollutants. Environ. Pollut. 53: 63-78.
Oshima, R. J.; Poe, M. P.; Braegelmann, P. K.; Baldwin, D. W.; Van Way, V. (1976) Ozone dosage-crop
loss function for alfalfa: a standardized method for assessing crop losses from air pollutants. J. Air
Pollut. Control Assoc. 26: 861-865.
Parry, M. A. J.; Day, W. (1988) A comparison of indices that describe the relationship between exposure to
ozone and the reduction in the yield of agricultural crops [Comments on article by Lefohn et al.,
1988]. Atmos. Environ. 22: 2057-2058.
Rawlings, J. O.; Lesser, V. M.; Heagle, A. S.; Heck, W. W. (1988) Alternative ozone dose metrics to
characterize ozone impact on crop yield loss. J. Environ. Qual. 17: 285-291.
Reich, P. B.; Amundson, R. G. (1985) Ambient levels of ozone reduce net photosynthesis in tree and crop
species. Science (Washington, DC) 230: 566-570.
Runeckles, V. C. (1988) A comparison of indices that describe the relationship between exposure to ozone and
reduction in the yield of agricultural crops [Comments on article by Lefohn et al., 1988J. Atmos
Environ. 22: 1241-1242.
Smith, G.; Greenhalgh, B.; Brennan, E.; Justin, J. (1987) Soybean yield in New Jersey relative to ozone
pollution and antioxidant application. Plant Dis. 71: 121-125.
2-14
-------
Tingey, D. T.; Hogsett, W. E.; Lee, E. H. (1989) Analysis of crop loss for alternative ozone exposure indices.
In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and
its policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988;
Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 219-227.
(Studies in environmental science 35).
U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report nos. EPA-600/8-84-020aF-ef. Available from: NTIS, Springfield,
VA; PB87-142949.
Wang, D.; Bormann, F. H.; Karnosky, D. F. (1986a) Regional tree growth reductions due to ambient ozone:
evidence from field experiments. Environ. Sci. Technol. 20: 1122-1125.
Wang, D.; Karnosky, D. F.; Bormann, F. H. (1986b) Effects of ambient ozone on the productivity of Populus
tremuloides Michx. grown under field conditions. Can. J. For. Res. 16: 47-55.
2-15
-------
-------
3. EFFECTS OF OZONE ON HEALTH
3.1 HEALTH STUDIES RELEVANT TO SELECTION OF THE
PRIMARY NAAQS FOR OZONE
3.1.1 Human Clinical Studies
The strongest and most quantifiable concentration-response data on the acute health
effects of ozone (O3) are provided by the controlled human exposure studies, in which
significant decrements in pulmonary function have been reported (U.S. Environmental
Protection Agency, 1986). In most of these studies, the greatest attention has been focused
on decrements in forced expiratory volume in 1 s (FEV^ because this measure of lung
function represents a summation of changes in both lung volume and resistance. At the
lower O3 concentrations of interest for standard-setting (<0.12 ppm), however, the observed
decrements in FEVj primarily reflect decrements in forced vital capacity (FVC), with little
or no contribution from changes in airway resistance. These changes in FEV} are caused by
a reduced inspiratory capacity that most likely results from sensitization or stimulation of
airway irritant receptors.
Scientific evidence presented in the 1986 U.S. EPA criteria document (U.S.
Environmental Protection Agency, 1986) established that pulmonary function decrements are
generally observed in healthy adults after 1 to 3 h of exposure as a function of the level of
exercise performed and the O3 concentration inhaled during the exposure. Decrements in
lung function have been reported to occur in some groups of healthy adults at the current
level of the standard (0.12 ppm) or somewhat higher. Also, pulmonary function decrements
have been observed in children and adolescents at concentrations of 0.12 ppm and 0.14 ppm,
respectively, with heavy exercise. At the lower O3 concentrations in the range 0.12 to
0.16 ppm, the average group mean changes in lung function are generally small (<6%), and
the medical significance of these changes is a matter of controversy. Some individuals,
however, are intrinsically more responsive to O3 than others and exhibit noticeably larger-
than-average pulmonary function decrements than the rest of the group. Such larger
. (> 10%) decrements in lung function may be of some medical significance to the affected
individuals.
3-1
-------
More recent controlled hitman exposure studies appearing after the completion of the
1986 criteria document have further confirmed and extended the above types of findings, as
well as demonstrating some additional new types of effects. The most pertinent of these
newer studies are summarized in Table 3-1. The newer studies noted in Table 3-1 add most
extensively to our knowledge concerning pulmonary function decrements associated with
acute O3 exposures near the current 1-h NAAQS for O3. The newer pulmonary function
decrement findings are summarized in Table 3-2 in relation to the earlier such findings
previously discussed in the 1986 criteria document".
Two of the more recent studies, by Linn et al. (1986) and Avol et al. (1987), add to the
information reviewed in the 1986 U.S. EPA criteria document (U.S. Environmental
Protection Agency, 1986) on lung function changes occurring in healthy children and young
adults exposed to low concentrations of O3 while exercising at moderate to heavy loads.
Data presented by Linn et al. (1986) in a controlled human study of healthy young adults
exercising intermittently at heavy work loads have added more detailed concentration-
response information at low O3 concentrations ranging from 0.08 to 0.16 ppm. The
O3 responsiveness of subjects in this study falls somewhere between that of subjects studied
by McDonnell et al. (1983) and of those studied by Kulle et al. (1985) under similar
exposure conditions (see Table 3-2). These subjects were also less responsive than the group
previously studied by Avol et al. (1984), who were exposed to similar concentrations of
O3 but with continuous exercise for 1 h. Although the authors of this report could not offer
a definitive explanation for differences among these studies, they pointed out that individual
biological factors such as the presence of asthma or clinical respiratory allergies and bronchial
reactivity in individual subjects, as well as external factors such as ambient exposure history
or differences in controlled exposure conditions during the study, might contribute to
differences in cohort responsiveness to O3. It is obvious that more research is needed to
better define the possible reasons for the large variations in responsiveness to O3 in
individuals and the variations in group mean responsiveness across studies.
Avol et al. (1987) presented data from a laboratory field study of healthy children (8 to
11 years old) exercising continuously for 1 h in ambient air containing a mean
O3 concentration of 0.11 ppm. The same authors (Avol et al., 1985a,b) previously studied
adolescent subjects (12 to 15 years old) under a similar protocol, although the
3-2
-------
2
H
1
|
3
3
2
3Q
1
if
£ 0
JMAN EXPOSURE LABORATOI
? THE 1-HOUR NAAQS FOR OZ
w o
Q K^;
l&d p^
G S
p g
Is
Ir"1
li^i
P
|f ^
I ^"^
1
H
i
N
H
R
r
i
i
i
i
^
ol en en
rn ^ ol en •*
t>
cn
o 1—1
0 I-H
d d
O i-H
ro
ol
S
§
0 §
« °
1 s
Pi "§
•'-' '*£>
§ a
2 M
> .9
|? VH
E,^
12
individual data indicated a significant
(p < 0.05) trend toward decreased forced
d1
2
« «
J5> ^
1 'I
•§ 5
CO
eo *H
2 rt
TO *n
'd ^
3 >r^
CO ^0
•3 .g
expiratory function with increasing O3
concentration.
«
® So"
^S
pq ^,
«
•§
>
.
& a s
8 CO 'B
"•§" a 2
g g g
IH *rf R
(D 2 : T3
2 § «
0 & I
^ 2 *
^
Group mean changes in FVC were similar
for the three exposures (-4.6, -5.8, -4.2%;
but individual responses were variable.
J 2
6 M
d} js
00 >-,
8 c?
13 o
- r-t
oo in
6"
01
! g
E9 N -§
- " h4
"^ nj8 *O
Ol '^ Ol
o in
0 •*
o" d
o ol
oo
00
PH
PH Ch
i oo
53 o\
•^ C
,22
s 13
Q t>
9
J, g
-S 5P ^
111
a? rS '»
(U -*-*
1 ?
13
I
>v< Q rl
"I a ^
tu
1
_
' ' '
\g
oo >-.
1 2
oo in
ol
g a,
^
o\
2
'"o
§ 2
O 00
ol '>
01
ol
oK
en i!
o m
o ^l~
O
-------
3
?
*J
a
o
u
I
CO
S
P
3
.g i 8 Jd^a
J>
5^
a 8
rff
<^2
r °
O
J5
»3
III
5. M3 S3
x E "
«|<
1 §.
a *
v
g
o
0«
ok
eo
2
*o
s
>>
S >p
*— ; en
1 2
O oo
»— 1 I— 1
/-% rt
C-II
VO > — Tj- 5i
82
C> 0
O m
CO
CN
*
CO
"U
o
1
.
§3 ^
i—-j CO
c3 O
O CO
CS ^H
•s
+
o
>n oo o
•* --i •*
8
ta ^
•b§
8
§
O IO
C3 CO
O O
\O
OO
VO
§§
o o
o •*
oo
t-~
"3
t>
ts?
h oo
53 01
O C
S 1
a.S's
••SdJ
lg •« "S o
CO
2
OO
i — 1
*3
§
S
•a
§
o
^ — i
0
^o
o\
(SI
* — 1
©
I
•£
^
^3
O
CN
O
C-l
O\
a
#
S
•a
S
o
•*
CO
-------
O oo O c-1
00^—*^-^
O O
-------
.00,
uy
&
a\
-3
S S
§1
II
SI
>
pq
OO M
-
-s
t£3
& 0
«S 2
s->
o o
O Ol OO
O '—I r-4
O O O
ol en
73
»P,
M
o\
s'S'S
.|.a 2
5-g
O.
r—< co "JJ
T3 . O w*
.
>
60
. 8 18
5 D^ S3 -Si
>%
ON Ctf 1-H
a
ol
l.fi
a
PL,
g4'
PH
^
X
a
I
t°
•a
^? a
S2 x x
H
00
o
3-6
-------
ncl
Observed E
a.s
-u
g«g
CO t^H
„ O
III
x S SJ
orf
60
T3 oo
Q O\
60 -3
, en
o
6n
o) ol o
ri 5
J2 en
c2 O
•^ oo
•I
a
a •§
'x
o
en
x-> 6
-< <—I Ol Ol
w
01
O OO O Ol xj- vo
O O i—t r-l *—I T—<
d d d d d d
o c- vo in •* M-
u-t a\ en c~ *— i
1-1 r-t CN e>l en
3-7
-------
g
p^
HO
< si
« o
T^£ Fr*
CQ ***^
< P
3*
||
co S
en
06
.S1
S5
CO n-4
» o
Ni
O
Ex
U
0
in
a
82
o o
O CO
H
1
Soo
i-H
o" o
o cr>
m
CO
8
I.
• pi
CD c~] co
O C3 O
O c3 oo
ON 00
CO IO
5
CN
3-8
-------
Concl
ed Effec
Ob
a.8
^o1
S C/3
'W *|_(
II
1 ts q
3 <
Q
§
CJ
of-
O cN oo •*
O 1-1 <-i Oj
O C) O C)
o vn en o
en >n f-
c*^ en Ti~
o en
o CD
oo
vo
O rf vq
O O o'
oo f-
C- <-<
•3
i <2 -a
V
.a
spi
pe
id,
^ iS S -3
CM d « c5
O £ 4) O,
..A PH M S^
d^ « S °
I -I I "3
Q « 6 O N
PL, GO ml ft p
3-9
-------
6
2 S
W O
00
o\
•a
o\
vo o\
xq c~-
C> r-i
8
•a
•s
ffi
8
o o
o c-
O *-
l-H O T—( O
i,
w
u
o o
-a
oo _ ,
01
+
IT)
a
O OO O Ol
o q ^ I-H
oo'o'o o'o'oo" oo"
8
01
01
a
01
Q—
VW f—S
I'rf
P oo
_~ **j i
2 «-< 23 £! °°
^~~l Nil—«* Vi CO ^-^
o\
g
8
8
o o o o
Ol
01
01
CO >-,
Ol I-<
01,
co Ol Cx
O O
01
o
O 01
O IT)
CO
Ol
01
in
+
01
§
s
O i-(
01
3-10
-------
!
1
«
•^a
c a g
•i ^ "f
H rt ^
*
1
.a
J" ^ " o^ « ^ o7
rH "^ p T— 1
SB ^££ ^SE
c^ d
CN OO
<-( en
31 ^^ ^ '-N
^^ s§> 1^ ^
°1 v°. ^ o °. ^ m.
9 ^ -r d - -r + 9
S~** (-•) Q>
g_ ^.| S ."
pq ^"^ r^
o ^ ? ?
rt "Tj CD O
i & 'p-. 'p..
^ CO Jf fl ^
^ -i §1
p
s
w
OP P
ScS o cs o cs oo
l-H f~> *-H f^~l r— 4 T— (
do do odd
O >o o 10 o to en
en en en >o
cN CO c-1 en
'I f
t3 vo" d- I-H
« Sa •a T! 73-
SP ^^ S1 3 ** 'SS "CD tiT
'gco r^o&JO - _, , i1^
M C^ w < CD !_l ™s
C^ "*•* ^3
do d
^ ^ V + ^-M" V
^™*^ ^f ^""^ T~H ^^ fli ^^ CO VO ''^^^ "
• ^^ C*^ • * ^" ^» C3 *™^ ^^ *""* • •
^9 +V +^2 + + + 9 t
en SS SS S $2-
a 8 8 a. 6 -
^CD^ /SS
.§ .§
Pt p t
"3 ?
_Ci ^ _CJ ® H rt _ri
*— 1 sti' > — 1 \S' t — ! CS ^ — '
P P P P
« « ** ^
^> ' ^- ^- £>• ^>- -
P ; p p p p
OCO OCO OO1 OCV1'* O^
Oi—t . O i— ( O^^ O *— 4 i— 4 O»-4'
o'd do do odd do.
Oto Oto OtO . O T> •* O1^)-
en en en en t> r^
CO CO CO CO CO CO
f
ON
^"
73
O
1
<
P
O to
0 ^c
d d
O "
-------
s
CO
s
J3
'•I
ta
§
wo
0,
Pi
"Sb
1
-H
CO
-H
o\
TH 10
+ 9
8
s
o o
CD r-l
o o
cs
a\
V
&
8
O T-t
o o
\o
ft
«
s
VO,
8
o o
S3 S3
o\ o\
Ti ^3
£f ^ °°
299
w
u
8
o o
cy>
73 73
-H
-H
JS +1
en ^
V
m
8
O en
V
+
X
c-l
o o
O en
8
o o
O en
in
CO
°i
c>>
-------
Reference0
t< HI
n ^ is
,4 — ' M
+ C?S
0?
a •
.
CM
P
P
O CS
d d
O CM
o\
en
OO £T>
a •• So _ ^
O o\ . ^
•a ^ , |5 « d^
_l s §* ..• i ^ i S3 '§ |
oc c§ -<&o<;d-t5 ^> S S
il? 11+1 1+1 i£"i£-
d OO . Hi ^ — . ^f a CM Hi CM
OO i— < Ci 1-1 CM C> t-1 G1 OO £> VO G'
vq
S ^vo
A O ' '-v
2 o o «o
o dS =?
• O o
VO \/ 1-1
x-v + + V ^ ^
? ^ • ft ? ?
cn^ai^voiJc-^^ ^
• ^^ &/) * • toO • Q OO * \JT) O
+ eni +c?g +vo . .c.
i2 tfi '"^ ^ *° °° — ^
>*~' <4— < OT C^ * * ^^
•4— t ^J C^« v-H VQ
m" oo 'a' W W W • W
H S g O O O O
-s- -«•
o$ o5
*o 'o
^s .§ ^
<1 <-| »1 ' O O
0) -H ^, ^1 ^ - l> ^x
> > > > '>
t— , 1—, )— . 1—, f— ,
J— ' (— ' L-J ^J (_J
P P P P P
OOl OCM OCMOC-1OCM
do do do do do
^^ O*l ^3 C^l C5 C*^ ^^ C1*! C3 C^
Os O\ ON O\ ON
CO CO CO CO CO
-a S
•a « . a.
« •-•><.
8 1 ^ -•
1 ^ ' 1 & *• '
.a oo p oo, o
*3 O\ 0 o\ 5.
,y i— i ^1-1 J5
fe v_^ ^1 ^s' •<
/— \
ra '"^
|1£-|. *.! |1
p CA oo -_ O3 ON ' , t+H
vo-iC CMCin
^^ C5 CO ' ^3
o o .1 • o
V V ^ V
%*TJ *ii^ ^?
o\ oo ir>: -> ^ I-1
1 — v f- \ 1 — v
H-' W h- '
8r- H C^ ^* (^^ ^J"
CM O CM O CS '
d o do do
C? C^l ^D ^3 C3 C5
T— i C^- f-
-H
o\
3-13
-------
w
i
if
I?
Is
II
is
u
5* S
g
CO
•J3 —.
% I
I
i-5
!i
i
V
a
•* in
c4 o
+ T
vo
en
<»
60
a
8^
o o o o
o o o o
11
-------
O3 concentration and exercise level were lower in the more recent study. No significant
changes in respiratory function or symptoms were found in the group, probably because of
the lower doses of O3. Regression analyses of individual data, however, suggested that
individuals receiving high doses of O3 had effects that were comparable to those found in
adolescents and young adults, although no definitive comparisons could be made because of
differing ambient exposure levels and large intersubject variability in responsiveness to O3.
This finding is also consistent with the controlled exposure study by McDonnell et al.
(1985a,b) indicating that the effects of O3 on lung spirometry in children were very similar to
those found in adults exposed under similar conditions, except that no significant increases in
symptoms were found in children. Therefore, based on the available pulmonary function
data, young children and adolescents do not appear to respond differently to O3 than do
adults.
A series of papers describing the effects of O3 on subjects greater than 50 years of age
appeared between 1986 and 1989 (Bedi et al., 1988; Bedi and Horvath, 1987; Drechsler-
Parks et al., 1987, 1989; Reisenauer et al., 1988) (see Table 3-1).
Bedi and Horvath (1987) described the decrease in pulmonary function response in a
single subject studied at age 32 and again at age 40. The major importance of this study is
that it demonstrated a decline in response of considerable magnitude (AFEVj of -25%
decreased to -5% over 8 years) that was observed longitudinally. This lends credence to the
results of the cross-sectional studies indicating a decreased response in older subjects.
Drechsler-Parks et al. (1987) compared a group of older (age 51 to 76) subjects exposed
to 0.45 ppm O3 with a group of young adults studied under the same protocol (2-h
intermittent exercise at 25 L/min). The older subjects had substantially smaller changes in
function than the younger subjects, both male and female. Changes in FVC in the older
subjects averaged -5.3% and in the young adults, -14.1%. Similar differences were
observed for other functional measurements. Similar data for O3 exposure are reported in a
second paper by Drechsler-Parks et al. (1989).
Bedi et al. (1988) reported the results of a study in which older subjects were exposed
to this same O3 concentration (0.45 ppm) on three separate occasions. The responses were
not reproducible from one exposure to the next. The group average did not change
3-15
-------
appreciably between exposure series, indicating that even though the older subjects have more
variable responses, they are less responsive to O3, as a group, than younger subjects.
Reisenauer et al. (1988) also studied a group of older subjects, age 55 to 74 years.
These O3 exposures were conducted at 0.2 and 0.3 ppm O3 using a light intermittent exercise
regime. There were no significant changes in FEVl 0. For the 0.3 ppm exposures,
however, the female subjects (n = 10) had a slight rise (13%) in total respiratory resistance
that was statistically significant.
The implication of these differences in responsiveness to O3 in older subjects is unclear.
Only standard spirometry tests have been used to evaluate responses. It is not known if
changes in airway resistance or airway responsiveness to methacholine or histamine are
similarly attenuated in older subjects. The possibility of inflammatory responses has not been
studied in these older subjects.
Four additional publications (Eschenbacher et al., 1989; Kreit et al., 1989; Koenig
et al., 1988; Koenig et al., 1987) report the results of controlled human exposure studies on
the effects of O3 on asthmatics (see Table 3-1). Also of interest is a new study of subjects
with allergic rhinitis (McDonnell et al., 1987).
Kreit et al. (1989) studied nine asthmatics exposed to 0.4 ppm O3 for 2 h while
performing intermittent exercise with a ventilation of about 53 L/min. All subjects had a
history of physician-diagnosed asthma and were sensitive to methacholine. Medications were
withheld for at least 12 h prior to exposure. Nine nonasthmatic subjects were also studied
under the same protocol. Both groups of subjects had significant decreases in FVC, FEVj,
FEVj/FVC, forced expiratory flow at 25 to 75% of FVC (FEF25_75), and inspiratory
capacity after O3 exposure. The changes in FEV1; FEVyFVC, and FEF25_75 were more
negative in the asthmatics than in the normals (e.g., A% FEVj was —13.4% in normals and
—23.1% in asthmatics). Specific airway resistance (SRaw) was not significantly increased in
normals but was in asthmatics after O3 exposure. A significant increase in SRaw also
occurred after air exposure in the asthmatics. The change in SRaw after O3 was more than
twice that after exercise in air (ASRaw-air = +3.82, ASRaw-ozone = +8.02 cm H2O/L/s).
Both groups experienced a similar relative increase in methacholine responsiveness after
O3 exposure, expressed as a decrease in the provocative methacholine concentration that
3-16
-------
causes a 100% increase in baseline SRaw. It is important tq note that these subjects
underwent methacholine challenge both 90 min before and 90 min after exposure.
It is not clear to what extent the preexposure challenge may have confounded the
results, particularly because the nonasthmatics received a substantially larger dose of
methacholine than the asthmatics. Normal subjects appeared to have a depressed FEF25_75
prior to exposure (-12% decrease after methacholine challenge). There were no differences
in O3-induced symptom responses between normals and asthmatics.
A second report of this study (Eschenbacher et al., 1989) additionally included a
description of the effects of indomethacin pretreatment in O3-exposed normal subjects. The
data for adult asthmatics were those reported by Kreit et al. (1989). Indomethacin
pretreatment in normals caused a marked decrease in O3-induced spirometry changes
(AFEVj - O3 = -21.5%; AFEVj - O3 + indomethacin = -10.6%). However, there
was also a surprising, but substantial, placebo effect, suggesting a possible behavioral
component in O3 response. Indomethacin, an inhibitor of cyclooxygenase pathways of
arachidonic acid metabolism, had no effect on the increase in airway responsiveness caused
by O3. Indomethacin appears to primarily block the "restrictive" (i.e., decreased FVC) effect
of O3 and does not alter the bronchoconstrictive or airway reactivity responses. Of additional
interest was the observation that "normal" subjects in the indomethacin study had an FEVj
decrease, after an identical protocol, which was not unlike the response of the asthmatics,
thus raising the question of the normality of the subjects or the possible confounding effect of
a preexposure methacholine challenge. Further research is, therefore, needed to fully
understand any potential differences in O3 responsiveness between asthmatic and healthy adult
subjects.
The responses of adolescent asthmatics to 0.12 ppm and 0.18 ppm O3 were tested by
Koenig et al. (1987). The mouthpiece exposure sequence consisted of 30 min rest followed
by 10 min exercise (minute ventilation [ VE] = 33 L/min). In addition to the 10 asthmatics,
10 healthy adolescents were also studied. There was a significant increase in total respiratory
resistance (forced oscillation method) in both normals and asthmatics exposed to 0.18 ppm
O3. There were no significant changes in FEVj in either subject group. At 0.12 ppm O3,
there were no significant differences that could be attributed to O3 in either asthmatics or
normals.
3-17
-------
Koenig et al. (1988) have also studied adolescent asthmatics (n = 10) and healthy
adolescents (n = 10) exposed to either air, 0.12 ppm O3, 0.3 ppm nitrogen dioxide
or the combination of O3 plus NO2. The mouthpiece exposures lasted 60 min and included
two 15-min exercise periods during which ventilation averaged about 35 L/min. Medications
were discontinued at least 4 h prior to exposure. In the asthmatics, an 11% decrease in
FEF505g was observed after 0.12 ppm O3 exposure. One of the subjects had an exceptionally
large decrease in FEF50% of —60%, which occurred approximately 20 min after the end of
exposure. This same subject did not have a large change in FEF50% when exposed to
O3 plus NO2, suggesting that the response of this individual to O3 may have been anomalous.
There were no other responses attributed to O3 in this study, either in normal or asthmatic
subjects. The authors tentatively suggested that adolescent asthmatics may be slightly more
responsive to these low levels of O3. However, replication of these observations will be
required before this suggestion can be substantiated.
McDonnell et al. (1987) studied 26 subjects with allergic rhinitis to determine if the
presence of allergies was a predisposing factor for O3 sensitivity. These allergic subjects had
airway responses to histamine that were similar to a comparable group of nonallergic
subjects. Exposure to 0.18 ppm O3 for 2 h with heavy intermittent exercise caused increased
responsiveness to histamine and a decrease in several spirometric variables. The only
apparent difference between the allergic subjects and previously exposed nonallergic subjects
was a significant increase in airway resistance in the allergic subjects. These data on allergic
and asthmatic subjects suggest that both of these groups have a greater rise in airway
resistance after O3 exposure than do normal subjects. The relative order of airway
responsiveness to O3 is normal < allergic < asthmatic.
Between 1987 and 1989, a series of reports were presented or published concerning a
study of apparent seasonal variation in O3 responsiveness in residents of Los Angeles (Avol
et al., 1988; Hackney and Linn, 1989; Hackney et al., 1989; Linn et al., 1988) (see
Table 3-1). The definitive report of this study is the journal publication by Linn et al.
(1988). From a large number of subjects tested for O3 responsiveness, 12 responsive and
13 nonresponsive subjects were selected to participate in further testing. Characteristics of
the subjects are presented below:
3-18
-------
Gender
Nonresponders
Responders
8M/5F
5M/7F
Age
5 > 30
2 > 30
Health Status
All Normal
4 Normal
6 Atopic
2 Asthmatic
Mean-AFEVj
+ 1%
-12.4%
In all tests, subjects were exposed to 0.18 ppm O3 during 2 h of intermittent heavy
exercise (ventilation = 35 L/min/m2 body surface area [BSA]) at 35 °C and 35% relative
humidity (RH). These 25 subjects participated in two more pairs of exposure to O3 and clean
air. The initial tests were conducted in late spring (1986) and the followup tests occurred in
late summer/early fall (1986) and again in winter (early 1987). A subsequent follow-up test
with a smaller number of subjects (17 of the 25) occurred in spring (1987). The differences
between responsive and nonresponsive subjects, which were of course significant at the time
of the first test, were no longer significant at the first two follow-up studies in late summer
and winter. This suggested the possibility that ambient oxidant exposure during the summer
months produced an "adaptation" response that persisted for several months. This suggestion
was further strengthened when a reduced number of subjects were exposed to O3 again, one
year later. At this time, the responsive subjects appeared to regain their sensitivity to
O3 exposure. The mean absolute changes in FEVj for the four exposures in the responsive
subjects were -385, -17, +16, -347 mL respectively for the spring, fall, winter, and
spring tests respectively. Corresponding changes for the nonresponders were +28, +90,
+34, +81 mL. Because the experimental design was not optimal, these results need to
viewed with caution and, as the authors state, "It is not clear that these results can be
generalized." Nevertheless, these findings clearly suggest that results of experimental
O3 exposures of residents of high oxidant areas must be viewed with caution if frequent
ambient exposure was a possibility during the period of experimental exposure.
Additional information presented by Hackney et al. (1989) indicated that 8 of the
12 responders were reactive to methacholine and had a history of respiratory allergies. In
addition, 10 of the 12 responders had a history of some symptomatic complaints when
exposed to "smog". The authors suggested that allergy or atopy may be a risk factor for
excess response to O3 although other studies have indicated that increased airway reactivity is
not predictive of O3 responsiveness. They further speculated that nonresponders could be at
3-19
-------
increased risk for chronic health effects of cumulative ambient O3 exposure because they
would be less likely to avoid such exposures because of their lack of symptomatic complaints.
Controlled human exposure studies reviewed earlier in the 1986 U.S. EPA criteria
document have suggested that some impairment of exercise performance may be associated
with O3 exposure. Subjective statements made by individuals engaged in these controlled
studies indicate that the perception of pain occurring with deep breathing may be an important
factor that limits performance of continuous heavy exercise at O3 concentrations > 0.18 ppm.
Studies by Gong et al. (1986) and by Schelegle and Adams (1986) substantiate these earlier
findings, whereas a third study by Linder et al. (1988) suggests that small decrements in
maximal exercise performance may occur at O3 concentrations <0.18 ppm (see Table 3-1).
Gong et al. (1986) found that maximal performance tested after exposure of endurance
athletes continuously exercising at heavy work loads (VE = 89 L/min) for 1 h in a hot
environment was impaired in 0.20 ppm O3. This level of O3 exposure also reduced
pulmonary function and enhanced respiratory symptoms and airway responsiveness to
histamine. Maximal performance was not impaired after exposure in 0.12 ppm O3, despite
small but significant group mean decrements (5.6%) in FEV^. Similarly, Schelegle and
Adams (1986) found that exercise performance, as determined by completion of the exposure
protocol, was impaired following exposure of endurance athletes who were continuously
exercising at heavy work loads (VE = 87 L/min) for 1 h at O3 concentrations >0.18 ppm,
but not at 0.12 ppm. Significant decrements in pulmonary function and increased respiratory
symptoms also occurred at ^0.18 ppm O3.
The effect of O3 inhalation on performance of maximum exercise tests was also studied
by a group of Swiss investigators (Linder et al., 1988). Twenty-four subjects (12M, 12F)
were studied while performing maximal incremental exercise tests. The maximum exposure
duration was 28 min and the minimum was 16 min. The tests were performed in clean air,
0.07 ppm O3, and 0.13 ppm O3 in an environmental chamber (24 °C; 50% RH). Small, but
significant (t-test), increases (2%) in FEVl 0 were observed after clean air exposure. Except
for women exposed to 0.13 ppm (-1.4%), no changes in FEVj were observed with
O3 exposure. Performance on the maximum exercise test was decreased 11% in women and
7% in men at 0.13 ppm and 5% and 4%, respectively, at 0.07 ppm (p < 0.05, t-test).
During the tests conducted at 0.13 ppm, there was also a small decrease (2.5 to 5%) in
3-20
-------
anaerobic threshold, defined as the workload at which the venous lactate concentration
exceeded 4 mM. It is not clear to what extent the exercise performance test results may
reflect behavioral responses to the odor of O3.
There are a number of questions that may be raised about the paper by Linder et al.
(1988). From the graphical presentation of the data on FEVj, it appears that no significant
changes would be detected by an appropriate statistical analysis (i.e., an analysis of variance
appropriate for repeated measures, rather than multiple t-tests). The authors did not indicate
whether appropriate precautions were taken to randomize or "blind" the exposures.
Furthermore, no information is provided about the selection criteria for subjects. Because the
effects were reported for very low exposure concentrations and brief exposure durations
(maximum 28 min) and because they appear to be out of line with previous studies of
exercise performance during O3 exposure, it is important to determine if these observations
can be verified.
The data currently available indicate that reduction in exercise performance may occur
in many well-conditioned athletes after performing continuous heavy exercise for 1 h at
O3 concentrations >0.18 ppm. These athletes are capable of sustaining very high exercise
VE (i.e., > 80 L/min) for 1 h. Any performance decrements occurring at O3 concentrations
<0.18 ppm are less certain and need to be verified. It must be noted, however, that other
environmental conditions, such as increased ambient temperature and/or relative humidity,
may independently affect subjective symptoms and may independently impair exercise
performance. Therefore, it may be difficult to differentiate work performance effects caused
by O3 from physiological or behavioral effects caused by other conditions in the environment.
Studies utilizing longer exposure durations, particularly at lower levels of exercise, were
not previously reviewed in the 1986 U.S. EPA criteria document. Among the newer studies,
two (Folinsbee et al., 1988; Horstman et al., 1989, 1988) address the effects of O3 exposures
for durations >2 h (see Table 3-1). The first of these was designed to determine the effects
of prolonged exposure to the present level of the 1-h NAAQS for O3 (0.12 ppm) on
10 young adult subjects that are representative of individuals who spend most of the day
outdoors exercising at moderate intensities (e.g., adults performing heavy labor). Subjects
were exposed to either 0.0 or 0.12 ppm O3 for a total of 6.6 h. During the exposure, the
subjects exercised for six periods of 50 min each; each exercise period was followed by
3-21
-------
10 min of spirometry testing and rest. An additional 35 min for lunch was interposed
between the third and fourth exercise period. The ventilation during the exercise averaged
about 41.5 L/min and heart rate ranged from 108 to 124 beats/min.
Prolonged exposure to 0.12 ppm O3 resulted in progressively larger changes in
respiratory function and symptoms with time. By the end of 6.6 h of exposure, group mean
changes were as follows: FEVl had decreased 13.0%, FVC had decreased 8.3%, and
FEF25_75% had decreased 17.4%. On forced inspiratory tests, forced inspiratory vital
capacity and forced inspiratory volume in 0.5s were decreased 12.6 and 20.7%, respectively.
Respiratory symptoms of cough and pain on deep inspiration increased with the increasing
duration of O3 exposure. There was also-a marked increase (about twofold) in airway
responsiveness to methacholine following O3 exposure. No changes were observed with
clean air exposure. The changes in lung function reported at the end of exposure were similar
in magnitude to those previously observed in healthy subjects performing at heavy levels of
exercise (VE £ 60 L/min) in much higher O3 concentrations (>0.2 ppm) for shorter
durations (i.e., <2 h).
The need for additional concentration-response information led to a subsequent study
using the same O3 exposure protocol. Twenty subjects were exposed for 6.6 h to four
O3 concentrations (0.0, 0.08, 0.10, and 0.12 ppm) in random order. The results of this
study were reported, in part, at the 1988 U.S.-Dutch symposium (Horstman et al., 1989) and
at the 1988 Annual Air Pollution Control Association Meeting (Horstman et al., 1988). The
ventilation in this study was slightly lower than in the first study, averaging 38.9 L/min. The
FEVLO decreased by 7, 7, and 12.3% at 0.08, 0.10, and 0.12 ppm, respectively. The
airway resistance response to methacholine was increased by factors of 1.56, 1.89, and 2.21,
respectively. There was also a significant increase in the symptom of pain upon deep breath,
a typical symptom of acute O3 exposure. •
The study by Folinsbee et al. (1988) is the first clinical study to demonstrate increased
airway reactivity to inhaled bronchoconstrictors in subjects exposed to low O3 concentrations
for prolonged periods of time. Other studies reported in the recent literature have identified
these effects in humans exposed to O3 for shorter durations (see Table 3-1). The study by
McDonnell et al. (1987) described an increase in airway reactivity to histamine in 26 healthy
subjects with allergic rhinitis who were exposed to 0.18 ppm O3 for 2 h while undergoing
3-22
-------
heavy (VE = 64 L/min) intermittent exercise. Seltzer et al. (1986), in a study of 10 healthy
individuals exposed for 2 h to air and to either 0.4 or 0.6 ppm O3 while undergoing moderate
intermittent exercise, observed an increase in the number of neutrophils hi bronchoalveolar
lavage (BAL) fluid 3 h after O3 exposure. Furthermore, they observed an increase in airway
reactivity to methacholine following O3 exposure and their data were suggestive of an
association between the degree of inflammation and the increase in airway reactivity.
A series of reports by Koren et al. (1989a,b; 1988a,b) described the inflammatory and
biochemical changes in the airways consequent to O3 exposure (see Table 3-1). In these
studies, subjects were exposed to 0.40 ppm for 2 h while performing intermittent exercise
(15 min exercise, 15 min -rest) at a ventilation of 70 L/min (35 L/min/m2 BSA) (i.e., the
same protocol as used by McDonnell et al., 1983). The main purpose of these studies was to
examine cellular and biochemical responses in the airways of ozone exposed subjects. To
accomplish this, BAL was performed about 18 h after the O3 exposure. Standard lung
function tests were also performed before and after exposure. A mean decrease in FEVj of
960 mL after O3 exposure was reported. An eightfold increase in polymorphonuclear
leukocytes (PMNs) was observed after O3 exposure, confirming the observations of Seltzer
et al. (1986). A twofold increase in protein, albumin, and immunoglobulin G were indicative
of increased epithelial permeability as previously suggested by the technetium-labeled
diethylene triamine pentacetic acid (DTPA) clearance studies of Kehrl et al. (1987). In
addition to confirmation of these previous findings, Koren et al. (1989b) provided evidence of
stimulation of fibrogenic processes including increases in fibronectin (6.4 x), tissue factor
(2.1 x), Factor VII (1.8x), and urokinase plasminogen activator (3.6x). There was a
twofold increase in the level of prostaglandin £2 and a similar elevation of the complement
component C3a. Levels of the leukotrienes LTC4 and LTB4 were not affected. Koren et al.
(1989a) reported that an inflammatory response, as indicated by increased levels of PMNs,
was also observed in BAL fluid from subjects exposed to 0.1 ppm O3 for 6.6 h (same
protocol as Folinsbee et al., 1988).
Further evidence supporting the hypothesis that cyclooxygenase products of arachidonic
acid metabolism (prostaglandins, thromboxane) may play a role in O3-induced spirometry
changes comes from a study by Schelegle et al. (1987). These investigators demonstrated a
significant attenuation of decrements in FVC and FEVj 0 when subjects were treated with the
3-23
-------
cyclooxygenase inhibitor indomethacin prior to O3 exposure. Subjects were exposed to
0.35 ppm for 1 h of continuous exercise (60 L/min); FEVj 0 decreased 26.3% on the
no-drug day, but only 10.6% after indomethacin pretreatment.
The above studies indicate that the inflammatory process caused by O3 exposure is
promptly initiated (Seltzer et al., 1986) and persists for at least 18 h (Koren et al., 1989b).
The time course of this inflammatory response and the O3 exposures necessary to initiate it,
however, have not yet been fully elucidated. Furthermore, these studies demonstrate that
cells and enzymes capable of causing damage to pulmonary tissues were increased and the
proteins that play a role in the fibrotic and fibrinolytic processes were elevated as a result of
O3 exposure.
Graham et al. (1988) showed an increase in PMNs in nasal lavage fluid collected from
subjects exposed to 0.50 ppm for 4 h at rest. There was a 3.5-fold increase in nasal PMNs
immediately after exposure and this increased further (6.5-fold) by the following day (i.e.,
20 h later). This study suggests that a nasal inflammatory response may serve as a qualitative
indicator of an inflammatory response in the lung.
Kehrl et al. (1987) observed an increased rate at which inhaled technetium-labeled
DTPA diffused from the airway and alveoli into the bloodstream in eight healthy subjects
who endured heavy exercise for 2 h in 0.4 ppm O3. Kehrl et al. (1989) reported results from
an additional 16 subjects studied in the same manner. For the combined group of 24 subjects
exposed for 2 h to 0.40 ppm O3, the average rate of clearance of technetium-labeled DTPA
was 1.08%/min. This clearance rate was some 60% faster than that observed after air
exposure. The average O3-induced decrement in FVC in these subjects was -10%. This
study confirms that clearance of technetium-labeled DTPA is accelerated after O3 exposure
and, in conjunction with the Koren et al. (1988a,b; 1989a,b) observations, strongly suggests
that this accelerated clearance is due, in part, to an increased epithelial permeability within
the lung. These changes in permeability are most likely associated with acute inflammation
and could potentially allow better access of inhaled antigens and other substances to the
submucosa. Results from studies of these endpoints at lower O3 levels have not been
reported.
These observations by Koren, Kehrl, and co-workers have raised the question of
whether acute inflammation occurs following exposure to low levels of O3 for prolonged
3-24
-------
periods of time (>2 h). Results from studies determining if these newly identified O3 effects
are occurring at low O3 concentrations (i.e., <0.12 ppm) remain to be more fully reported
and evaluated. This information will improve our understanding of the nature of
inflammatory responses, including the biochemical and molecular changes in the lung, that
occur in O3-exposed subjects. ,
Another series of papers by Gerrity and co-workers (Gerrity, 1987; Gerrity et al., 1988;
Gerrity and McDonnell, 1989) examining O3 uptake in the respiratory tract have important
implications for modeling the health effects of O3 exposure in humans and for extrapolating
data from animals to humans (see Table 3-1). .
Gerrity et al. (1988) studied 18 healthy young males to determine the fractional uptake
of O3 by the upper respiratory tract (URT), excluding the larynx, and by the lower
respiratory tract (LRT), including the larynx. In order to measure O3 concentrations during
the breathing cycle, a chemiluminescent O3 analyzer was modified to increase its response
time. Gas was sampled at the level of the posterior larynx from a tube inserted through the
nose. Mean inspired and mean expired (alveolar) values of pharyngeal O3 concentration were
used to compute the fractional uptake of O3 in the URT and LRT. The investigators studied
the effects of changes in O3 concentration (0.1, 0.2, 0.4 ppm), breathing frequency (12 and,
24 breaths per minute) and mode of breathing (nasal, oral, oronasal). The differences
between the various treatment conditions were small; the average URT uptake was about 40%
and average LRT uptake was about 91% (of the O3 that reached the larynx), resulting in an
average total respiratory tract uptake of approximately 95%. (In other words, of the
O3 entering the URT, about 40% was removed. Of the remaining 60% that reached the
trachea, 91% was removed. Total uptake is therefore 40% + (0.91 x 60%) = 95%.)
Increased frequency of breathing caused a decreased fractional removal of O3 in both URT
and LRT, presumably because of decreased residence time in the airway and increased flow
rate. The lowest fractional removal of O3 in the URT occurred during nasal breathing. The
differences between nasal and oral or oronasal breathing, however, were very small. The
lack of significant differences between nasal and oral breathing on O3-induced changes in
lung function and respiratory symptoms was recently reported by Hynes et al. (1988), also
suggesting that the mode of inhalation may not affect O3 uptake as much as previously
expected.
3-25
-------
In a second paper, Gerrity and McDonnell (1989) reported the influence of the
O3-induced change in breathing pattern on the O3 uptake efficiency. Subjects were exposed
to 0.4 ppm O3 during continuous 60 min exercise at a ventilation of about 40 L/min. At the
end of the exposure, there was a 25% reduction in spontaneous tidal volume and a 45%
increase in breathing frequency. Associated lung function changes included a 13% reduction
in FVC and an 18% reduction in FEVj. The change in breathing pattern was accompanied
by a 9% reduction in the LRT O3 uptake efficiency (fractional LRT uptake decreased from
68% to 62%). Total O3 uptake (about 80%) was only reduced about 4% because there was a
slight increase in O3 uptake in the URT. The reduction in LRT O3 uptake was correlated
with the decrease in tidal volume, suggesting that an increased depth of inspiration increases
the dose delivered to the LRT. The O3 uptake "efficiencies" reported in these two papers are
not strictly comparable because the methods used to make the calculations of O3 uptake were
different in each paper. The authors suggested that the reduction in tidal volume may act as
a protective mechanism for the lower airways, but that the loss of this response with repeated
exposures may permit increased O3 delivery to the lower respiratory tract.
Gerrity (1987) described a model of nasopharyngeal uptake of O3 using data from
various animal species, including humans. The conclusion reached in this analysis was that
nasopharyngeal O3 uptake decreases with increasing flow but that there was also a
considerable species variation in uptake (see Section 3.1.3.3 and Table 3-6).
The above observations of Gerrity and co-workers have important implications for
interpretation of heavy exercise studies. Increased tidal volume increased LRT O3 delivery,
but there may be a limit beyond which increases in tidal volume will not cause increased LRT
O3 delivery. Further modeling studies will hopefully address whether such a limit exists in
the physiological range of human ventilation.
Available data on respiratory tract uptake efficiency in humans appears to fit the
predicted model, making it possible to develop dose-response information from the wealth of
controlled human exposure studies that have already been published. The current likelihood of
making animal-to-human extrapolations based on this information and on the comparison of
respiratory tract uptake of O3 across different mammalian species is discussed later in
Section 3.1.3.3.
3-26
-------
3.1.2 Field and Epidemiological Studies
Field and epidemiological studies reviewed earlier in the 1986 Criteria Document (U.S.
Environmental Protection Agency, 1986) reported a variety of results concerning associations
between exposures to ambient O3 and various measures of respiratory effects. The results of
many of these studies were found to be direetionally consistent with the findings of controlled
human exposure studies, especially those providing reasonably good evidence for associations
between ambient O3 exposures and acute pulmonary function decrements. Results of other
epidemiology studies evaluating other health endpoints, e.g. exacerbation of asthma or other
chronic lung diseases, were found to be more difficult to interpret due to methodological
limitations, but some of these latter studies tended to point toward possible increases in
symptom aggravation or changes in lung function of asthmatic subjects being associated with
increased total oxidant levels, ambient O3 concentrations, or interactions between ambient
O3 levels and temperature. However, no consistent pattern of findings for aggravation of
symptoms or lung function changes emerged for patients with other types of chronic lung
disease. Newer field and epidemiology studies (summarized in Table 3-3), as with the older
literature, continue to provide somewhat mixed results across various health endpoints
measured—with most progress having been made with regard to provision of further
information concerning exposure dynamics related to the induction of pulmonary function
decrements by short-term ambient O3 exposures.
Included among the newer studies published since 1986 are reports by Raizenne and
coworkers on several aspects of field studies of children in two summer camps in Ontario
(Raizenne et al., 1987; 1989), one at Lake Couchiching (LC) about 100 km north of
Toronto, Ontario, and one at, a Girl Guide camp on the north shore of Lake Erie. In the LC
study (Raizenne et al., 1987), the strongest association between lung function and
environmental variables was found in nonasthmatics, with FVC decrements correlated
(p < 0.01) with 24-h lag functions for average SO4=, partilces <2.5 jwm (PM2 5), and
temperature. The association of peak expiratory flow rate (PEFR) with unlagged 1-h O3 was
statistically significant and the average slope of the regression line was —2.7 (mL/s/ppb).
Temperature was significantly associated with all lung indices in nonasthmatics but not in
asthmatics. The average slope of PEFR for temperature in nonasthmatics was —21.7, a
much stronger association of PEFR with temperature than with O3. Coefficients of variation
3-27
-------
CO
I
I
111
g "° •§
asp
o
s -a •=
V
S
" o
— S
i " """ S —
^ o O ~-> 75
SIM
3^rtiC —JO-,..-;., ..H *r? i-H -O
g S s -" a o.
.2 8
•
o ,
'5 o
•? ^
Q .-g
-s
•S ^-3 s.
o a
^ 5.H >. ^
^ « >.XJ S>
O
•a 1 -s
•rl W r \
o
3 o
&
-------
A
H
(2
•&
o
I
•&
B3
. d JSJSO
O +d
fc §
significant at p < 0.05 (differences from zero).
clear if slopes for data subsets significantly differ
•a
03
from each other (e.g., fimction vs. Og < 60 ppb
fimction vs. Og < 80 ppb). No formal analysis
performed for possible concentration threshold.
~Q
oo
oo
*3
t>
t-t
Q
8
P<
TO
„
*-H
Significant (p < 0.01) decrements in FVC, FEV
PEFR, FEF25_75, and FEVj 0/FVC associated
with Og. No persistence of effects seen. No
o
M .,
2 °
S£ Sco
* 11 -5
,0 M §>
& w 3-
•* • T
III
lit
o A *— '
e3 ^"' m
•2 >< 2
>+"
*H ^
w &
W> „
•a
almost daily outdoors (Tuxedo, NY) for 15 to
55 min (avg. ca. 30 min), July to early August,
calculated V-g levels, as indicated by regressing
pulmonary function changes and postexercise
function against inhaled ozone during exercise.
_5j
^
"5
'£
§
o
om
|
"S
o
o
i-
o
*&
O
1
|
-------
1
2
§
a
s
I
o
•o
VI OO
oo a\
l
O co
s|
2 -S
'
.5 S
ll
3-30
-------
G
1
Results and Comments
CO
4,000 elementary
school children. Subsamples selected from si:
.^
*5 [TI
3 0
6 +
schools in study area with consistently higher
levels of air pollution during previous 4 years
10 W 3- S '
'ft nJSf^oa —
Subsamples (3) stratified by reported sympton
One or more of following measures taken for
144 children: diaries, symptom questionnaire
spirometry. Telephone follow-up each 2 wee!
on diaries; spirometry done at school; pollutai
(including Og) measured at 1 monitor (data fr
17 monitors for SO2 generally reflected in dal
at single monitor). Diary panel study coverec
8 mo; successive PEFR spirometry studies of
9 weeks each done in respective groups of the
three subsamples.
oo
*cs
0}
.
All particle measures (TSP, PM^ 5, PMjj)
associated with substantial increases in respiratory
illness reporting rates; only PMjj associations
statistically significant. SO2 (also correlated with
particle measures) showed much weaker association
with respiratory symptoms than association with
particle measures. Weak association of NOo with
symptoms. Negative association of Og with cough
and chest illness, but annual Og positively associate
with asthma and hay fever (in contrast to neg. assoc
of latter symptoms with all other pollutants). No sij
assoc. of pulmonary function levels with any of the
pollutants, leading authors to conclude that increase
rates of illness not associated with permanent loss o
•V K^. r-
O^ 3 •" .3 S1
'*-'£(''-< .GG5*'o 5>T£S
o.So ocjo •—• .^
I S 1 ^| 1 |-| g-g 1 |
fi|iijiis|!i
iii?^iiiijf ^i
g 8 Sf2"^r«:3S § c ><'§:3
l|l 2°l-i'l«|W|«3
o c*"i*5 ^ H-t-j w ^.2*^ co(4-i
s O *& n ^ -2 5 *" "^ c ^ ®
lisl^al 11 ll-fl
•H . . *a P w
Follow-up examination in 1980-1981 of cohoi
of school children studied in six cities of U.S.
Reexamination of between- and within-city
results from earlier report (Ware et al., 1986)
Respiratory symptom questionnaires complete
by parents; spirometry done at school to obtai
FEVj Q, FEVQ 75, FVC, MMEF. Symptom
(cough, bronchitis, child illness, wheeze,
asthma) fit to multiple logistic model.
pulmonary function in preadolescent years.
P
00
o\
ci
tn
•a
G
eg
CO
C3
fH
Significant correlations found in summer between
Og, SO2, and T versus .deviations from the mean
respiratory admissions, with and without asthma, fo
M
•
>°
OO *S*
oo w ^
I* .1
.- CO M
«Pi e
S* ' ^
=1 2 -°
•0 „ J3
CO <0 *-<
?IH
.^ X" oo
O Ct, ON
g t^3
1^ CO O\
^ vo ^^
Effects of pollutants and other environmental
variables in summer (Jul.-Aug.) and winter
(Jan. -Feb.) on three categories of hospital
•a
the same day of week in the same season and year.
Stepwise multiple regression analysis based on each
year separately indicates that SOx~ and T accounte
for about 5 % of the variance from the mean in
B
K
H*
„
g
i "
in
^
o"
"&"2
S S
2> &
cr\ c3
2 |
&
O
admissions examined: total respiratory, total
respiratory excluding asthma, and nonrespirat
admissions to 79 acute-care hospitals serving
~5.9 million people in southern Ontario,
M
respiratory or asthma admissions in summer. Testii
S
c
1
i_i
«
o
c
CQ
U
of the 8-h maximum Og statistic in place of the 1-h
Og statistic did not increase the correlation
coefficient with respiratory admissions. (In winter,
asthma admissions were correlated with T only.)
examined. Pollutant data collected from
17 stations along 280-mile corridor between
Windsor and Peterborough.
3-31
-------
g
o
B
§
3
Q
a
«?
CO
9
S
Re
5
S 1
Effects of pollutants and other environmental
variables on respiratory symptoms and PEFR
evaluated in 11-mo population study of asthmatics
living in high-Og area (Glendora) of Los Angeles
County, CA. Detailed questionnaires given at
outset on medical/occupational histories and
personal factors, including general activity
patterns; psychological tests (Asthma Symptom
Checklist, State-Trait Anxiety Inventory, etc.) also
g
i
1
•e
.£2
*«3
o
o
c:
c"
o
°3j
smoggy day. Lung function (spirometry) and
bronchodilator responses measured at outset in all
subjects. Daily diaries (checked 2x/week), mini-
Wright flow meters (calibrated 2x/week), and
Nebulizer Chronology attached to metered-dose
broncho-inhaler used to record symptoms, day and
night PEFR, and medication use, respectively.
Symptom questionnaire given 2 x /week. Multiple
regression analyses for overall group; then subsets
(two groups of "responders") analyzed separately
and compared with rest of cohort.
II o
•^ U J3
>n « ffi "o .
-.?.& H J &
^ O *•* O
"
a — o S
1 .a s + §
1 > a u •§.
.§~:
& 8 * '1 o1
3 g ft § 2
» <2 „ > Z
hy
to
ty St
O4
3-32
-------
(CV%) were stable across the daily morning and evening tests of pulmonary function.
Although asthmatics had somewhat larger CV%, no statistical differences in CV% for a.m.
versus p.m. tests were seen in either group. Activity or exercise levels were not estimated,
nor was amount of indoor (as on rainy days) versus outdoor activity estimated (i.e., actual
exposure as well as proportion of higher versus lower exercise levels).
Raizenne et al. (1989) also presented preliminary data from a study of the effects of air
pollution on girls aged 8 to 14 who attended one of three consecutive 2-week sessions of the
Girl Guide camp on Lake Erie (June 29 through August 9, 1986). The health status of each
camper participating in the study (112 of 145) was characterized by questionnaires completed
by parents, by bronchial challenge with methacholine (positive responses were classified as
Mch+ and negative responses were classified as Mch—), and by skin-prick tests for atopy.
The influence of air pollution episodes on lung function was examined by comparing lung
function responses for each girl on episode days with mean responses on "control" days (the
latter defined as days with a 1-h O3 maximum of <90 ppb; SO4= < 15 /*g/m3; and sulfuric
acid, H2SO4, <10 /xg/m3). Also, lung function on the morning following an O3 episode
versus the average function on control days was examined.
Maximum decrements of 3.5% and 7% for FEVLO and PEFR, respectively, were
reported to be associated with four distinct air pollution episodes in which O3, H+, and
SO4= were all elevated. Only FEVj 0 changes were statistically significant and only on
2 episode days (one each in Camp Sessions 1 and 2). For each camp session, the mean
values for FVC, FEVj 0, and maximum mid-expiratory flow rate (MMEF) exhibited a
U-shaped pattern over time; larger first-day decrements were followed by a subsequent, more
gradual return to baseline. This pattern was not observed for PEFR. The largest FEVj 0 and
PEFR decrements were observed in methacholine responsive (Mch+) children the morning
after (July 26) the highest O3 level measured (July 25) during the study. In Mch— children,
however, the FEVj 0 change was positive and the PEFR change was negative, both on July
25 and July 26. In Camp Session 3, improvement in both FEVLO and PEFR were noted.
The authors postulated the exposure of campers in Session 3 to a regional episode prior to
their arrival in camp, with recovery occurring while at camp. No hypothesis was put
forward to explain the positive FEV1>0 change in Mch— children on the day of the highest
peak O3 level and on the day following. The lack of an aggregate analysis and the presence
3-33
-------
of largely unexplained temporal trends in pulmonary function make interpretation of these
study results difficult. This report of the study does not provide strong evidence for the
effects of O3 or of air pollution episodes on pulmonary function.
On July 25, when the 1-h O3 level was elevated (143 ppb), 12 subjects performed
pre- and postexercise spirometry (exercise level and resulting VE not estimated). For this
subset of subjects, postexercise FVC and FEV1 0 were observed to increase on control day
tests and to decrease on the episode day (results on the episode day were compared with the
mean pulmonary function test results for all control days). The function changes did not
attain statistical significance, however (Raizenne et al., 1989).
During the study of girls attending the Lake Erie residential camp, investigators
(Raizenne and Spengler, 1989) examined the use of heart rate as a surrogate for pulmonary
ventilation during daily activities. A dosimetric model was developed using heart-rate data
from a standardized exercise test and from portable heart-rate recording devices. Individual
exposure estimates were developed, based on time-activity data, and were related to changes
in lung function observed in the children. For both O3 and H2SO4, the slopes of function
(i.e., peak expiratory flow rate, PEFR) versus pollutant did not differ from zero when the
data were adjusted for dosimetry. Adjusted data for FEVl 0 were not reported.
From a field study they conducted in 1984 at a YMCA summer camp (Fairview Lake)
in northwestern New Jersey, Spektor et al. (1988a) have reported associations between
O3 and variations in respiratory functions for 91 children attending camp for at least 2 weeks.
Average slopes for the regressions between O3 concentrations and functions were significantly
negative (p < 0.05) for FVC, FEV10, MMEF, and PEFR for all children and for boys and
girls separately. Comparable data were obtained for cohort subsets (2-week campers). When
data were truncated at a heat stress index (temperature-humidity index, THI) of 78 °F, the
average slopes for girls were reduced by half for the data sets restricted to THI <78 °F,
eliminating significant differences in FEV1 0 changes between girls and boys. Little or no
comparable effect of a heat stress component was seen in boys. Activity levels were not
estimated, so that the VE component of the responses was not estimated for individual
children or for cohort subsets.
As reported by the authors, multiple regression analyses indicated that the
O3 concentration in the hour preceding spirometry, the cumulative daily O3 exposure during
3-34
-------
the hours between 9:00 a.m. and the function measurement, ambient temperature, and
humidity were the most explanatory environmental variables for daily variations in pulmonary
function, with the 1-h O3 concentration having the strongest influence. The authors
calculated predicted average functional decrements from the average slopes of the base data
set (Table 3-1, Spektor et al., 1988a), assuming the exposure-response curve to be linear, of
FVC, 4.9%; FEV, 7.7%; PEFR, 17%; and MMEF, 11% for O3 at the current standard of
120 ppb. Of the 91 children studied, 33 (36%) had individually statistically significant
FEVj 0 responses, with an average coefficient in that subset of —2.97 mL/ppb, or about a
16% decrement—again assuming linearity—at 120 ppb O3. The values for the 2-week subsets
are generally consistent directionally with O3 concentrations in the respective 2-week periods
and the total period. Likewise, slopes for data truncated at < 60 ppb and < 80 ppb O3 show
general directional consistency with the O3 concentration data, except for FEF25_75.
Several considerations should be noted. Ozone and temperature are statistically
correlated in this study (r = 0.37), with evidence of effects of heat stress on O3-associated
decrements in function. If the respiratory effects depend nonlinearly on interactions between
temperature (or THI) and O3, this may confound interpretation of the effects of O3. Data
were truncated at 60 and 80 ppb and the conclusion was drawn that O3-associated effects
occurred at < 60 ppb. A formal test for threshold would seem to be in order. The differing
number of pulmonary function test days does not appear to have been adequately accounted
for in the pooled analysis. The results of this well-conducted study would be strengthened,
however, by additional analyses. As reported, calculated decrements at the level of the
current standard should be interpreted cautiously.
Spektor et al. (1988b) conducted a field study of the effects of O3 in ambient air on
pulmonary function in 30 healthy adult nonsmokers (20 males, all Caucasian; 10 females,
2 non-Caucasian) exercising outdoors each work day (between 11:30 a.m. and 6:30 p.m.,
June 27-August 2, 1985, except for July 4 and 5) in Sterling Forest research park in Tuxedo,
New York. A respiratory questionnaire was administered before exercise and spirometry was
performed before and after exercise. The outdoor exercise regimen was selected by the
subject. Following each exercise stint, the subject measured his own pulse rate. Ventilation
(VE) for each exercise period was estimated from the subject-reported heart-rate data,
3-35
-------
calibrated from heart-rate data recorded from indoor treadmill exercise at a pace similar to
the outdoor exercise level.
For each subject, on each exercise day, pre- and postexercise function measurements
were taken, and changes in function were determined for FVC, FEV1 0, (FEVl 0/FVC)>
PEFR, and FEF25_75. Subject-specific exposures were estimated from duration of exercise,
mean O3 concentration during the exercise period, VE, and the tidal O3 inhaled during
exercise. Pollutants and environmental variables measured were O3, SO2, nitrogen oxides,
ambient aerosols (PM15 and PM2 5), aerosol acidity and other fine-particle ions, temperature,
humidity, and wind speed and direction. Pulmonary function variables were regressed on
mean O3 concentration during exercise for each subject, as well as against the
O3 concentration during exercise on the preceding day. Interactions of other environmental
variables with O3 were tested.
All pulmonary function indices showed significant (p < 0.01) O3-associated
decrements. No clear effects from other variables on the effects of O3 were seen. Mean
decrements were reported as smaller in 10 subjects with VE > 100 L/min than those in
10 subjects with VE of 60 to 100 L/min or those in 10 subjects with VE < 60 L/min. The
decrements were reported to be about twice as large as those seen in 1- to 2-h chamber
studies in which VE levels were comparable. No association was found between preexercise
lung function and mean O3 concentration during exercise on the preceding day (no evidence
of persistence of O3 effects). No symptomatic responses were reported. Analysis of lung
function changes for ventilations of 50 to 80 L/min was reported by the authors to indicate
that the influence of VE on lung function decrements peaks at about 80 L/min.
This study substantiates the effects of O3 on pulmonary function in populations engaging
in continuous exercise outdoors for short periods of time (15 to ca. 60 min; average duration
of ca. 30 min). The study further suggests that pulmonary function decrements observed
with ambient exposures to O3 at 0.12 ppm or somewhat lower may be larger than those seen
with comparable O3 exposure concentrations and exercise conditions in controlled human
exposure chamber studies, possibly due to potentiation of O3 effects by other ambient
cofactors.
In a study by Kinney, the effects of air pollutants on lung function were measured by
spirometry in children (ages 10-12, 90 male and 64 female) in Kingston and Harriman, TN,
3-36
-------
with spirometry done at least six times, >1 week apart, from February through April 1981
(Kinney, 1986; cited in Kinney et al., 1988). Ozone and other pollutants were monitored at
a single site in central Harriman. Temperature and aeroallergens were not measured. Values
for FVC, FEVg 75, MMEF, and V75% were regressed (ordinary least squares model) on the
1-h maximum O3 concentrations and on the 24-h-average fine particles (FP) and FP-SO4=
concentrations. Ozone concentrations ranged from 3 to 63 ppb during the study.
Concentrations of other pollutants (SO2, NO2, total suspended particulate (TSP), inhalable
particles (IP), respirable suspended particles (RSP), and FP) were not reported. Slopes of all
four lung function-O3 regressions were significantly negative. A positive mean slope of
MMEF on fine particle concentrations was reported. As noted in Kinney et al. (1988),
outdoor-only monitoring and lack of time-activity data compromise the specification of true
exposures; and the low O3 concentrations present during the study detract from plausibility.
Kinney et al. (1988) later published an interpretive evaluation of five epidemiological
studies of the effects on lung function of acute exposures to O3. In that review, they
compared the coefficients of O3-associated lung function declines reported in those
epidemiology studies with modeled exposure-response relationships for such effects derived
from a synthesis by Hazucha (1987) of results from controlled human exposure studies.
Hazucha (1987) modeled the effects of exercise-related ventilation rates (VE) in potentiating
the effects of O3 on pulmonary function, using pooled data from 2-h chamber studies of
healthy young adults exercising intermittently. Kinney et al. (1988) reexpressed the data of
Hazucha (1987) in units consistent with the epidemiologic study results (assuming a linear
relationship between lung function decline and concentrations up to 100 ppb and using
baseline functions obtained in Kinney, 1986). The resulting coefficients were reported as
being larger than those from controlled studies, especially for FVC (which was about five
times the mean FVC coefficients from the controlled studies). Kinney et al. (1988)
concluded that the "effective" exposures in the epidemiologic studies were cumulative over
longer periods (from 8 to 12 h versus the 2-h exposures used to generate the controlled
exposure data analyzed by Hazucha).
The Kinney et al. (1988) analysis provides further results suggesting that possibly larger
lung function decrements occur with ambient O3 exposures than those seen in controlled
human exposure studies. However, caution is warranted given several considerations. For
3-37
-------
one, no justification was given for the use of the linear model and the transformation of data
from Hazucha (1987), who had used a quadratic model. At concentrations < 100 ppb, a linear
model would tend to overestimate lung function decrements if the quadratic model is actually
more appropriate, whereas at higher O3 concentrations the linear model would underestimate
lung function decline in comparison to a quadratic model. Unfortunately, although the upper
end of the range of O3 concentrations measured in the epidemiologic studies reviewed partially
overlaps the lower end of the range used in the controlled studies modeled by Hazucha, the
mean concentrations in the respective epidemiologic studies from which data were used were
< 100 ppb—thereby not allowing for a direct comparison of the magnitudes of the effects seen
with ambient versus controlled -exposures across a fuller range of exposures of interest (e.g.,
0.05 to 0.15 ppm O3). Also complicating the comparison are possible differences in exposure
durations—leading, in fact, to the postulation by Kinney et al. that the effective O3 exposures
(including lower ambient O3 levels) were likely to have been more prolonged (8 to 12 h) in
the epidemiology studies than the exposure duration (2-h) used in the controlled exposure
studies.
Lioy and Dyba (1989) also reported a new analysis of previous data suggestive of
prolonged O3 exposures being the effective exposures in an earlier field study conducted in
Mendham, NJ, as reported on by Bock et al. (1985) and Lioy et al. (1985). In this case, Lioy
and Zyba used a time-activity pattern analysis for a hypothetical typical camper that would
have experienced a usual daily activity schedule analogous to children participating in the
Mendham summer camp study. They found that the time-activity analysis indicated a
likelihood that periods of increased exercise for the camper would have likely occurred over
prolonged periods of time (several hours) during which ambient ozone levels were elevated
(i.e., above 0.10 to 0.12 ppm). Evaluation of the total estimated O3 dosages likely
experienced on successive days of a 4-day O3 episode further led them to hypothesize that the
accumulated, multihour doses from prolonged daily exposures to ambient O3 may have been
most important in producing the pulmonary function decrements earlier reported for the
Mendham study, rather than next day residual effects being due to peak O3 concentrations.
Vedal et al. (1987) reported data from an 8-mo panel study of symptoms and concurrent
but successive 9-week PEFR studies in asthmatic and nonasthmatic school children living in
the Chestnut Ridge area of western Pennsylvania. Neither respiratory symptoms nor PEFR
3-38
-------
was strongly associated with any of the environmental variables, which included peak 1-h O3,.
NO2, SO2, and CoH, and daily temperature. Level of PEFR on the previous day was the
strongest predictor of daily PEFR. True exposures to O3 and other pollutants may have been
misspecified, because data were obtained from only one monitor for the whole area, except
for SO2, for which an average of values from 17 monitors was used; and individual
exposures and activity levels were not estimated. Further, levels of O3 during this school-
year study were very low, with the daily maximum 1-h levels ranging from 0 to 129 ppb,
with a mean of 32.4 ppb.
Results for the 1980-1981 school year have been recently reported by Dockery et al.
(1989) from an ongoing study of the effects of ambient air pollution on respiratory health in
children living in six cities in the United States: Watertown, MA; Kingston-Harriman, TN;
Steubenville, OH; Portage, WI; a geographically defined portion of St. Louis, MO; and
Topeka, KS. Previous results showed that the reported prevalence of chronic cough,
bronchitis, and chest illness increased by about a factor of two across the range of TSP and
SO2 concentrations measured in the six cities. Lung function was determined at school by
spirometry, and a respiratory illness and symptom questionnaire was completed by each
child's parents. Pollutants measured included TSP and particles <15 /j,m (PM15) and PM9 5,
O3, NO2, and SO2. Continous measurements (hourly values) of SO2, NO2, O3, and
meteorological variables, as well as daily values for various particle measures (e.g., TSP,
PM15, PM2 5, etc.), were taken. Monthly means for each pollutant were calculated from
daily means, and an estimate of air pollution exposure during the previous year was obtained
for each child by averaging the monthly means for a given city for the 12 mo preceeding the
child's spirometry test. All pollutant concentrations were highly positively correlated with
each other in all cities, except for O3, which was highest in those cities with low levels of the
other pollutants. The pulmonary function parameters measured were FVC^FEVj 0,
FEV0 75, and MMEF. Five respiratory illness or symptom categories were also considered:
bronchitis, cough, chest illness, wheeze, and asthma.
As in previously reported results from earlier years of this study, chronic cough,
bronchitis, and chest illness were positively associated with all three measures of particulate
pollution—TSP, PM15, and PM2 5—but only associations with PM15 were statistically
significant. Sulfur dioxide, which showed correlation with the particulate measures, was
3-39
-------
much more weakly associated than particles with the respiratory symptoms. The association
of NO2 with respiratory symptoms was also weak. There were negative associations of
O3 with cough and chest illness, but annual O3 showed a strong positive association with
asthma and hay fever (in contrast to negative associations of these symptoms with-all other
measured air pollutants). No association was found between air pollutant levels and the
pulmonary function measures, including FEV0 75 and MMEF, which are more sensitive
measures of small airway impairment than FEVLO and FVC. This led the authors to
conclude that the increased rates of illness observed for some pollutants were not likely
associated with permanent loss of pulmonary function—at least in preadolescent years.
Bates and Sizto (1987) have reanalyzed earlier data (Bates and Sizto, 1983; Bates, 1985)
and extended their analyses to more recent data (now covering 1974 and 1976-1983) for
examining correlations between environmental variables and hospital admissions for
79 acute-care hospitals in southern Ontario, Canada. Pollutant concentration data for
O3, NO2, sulfur dioxide (SO2), coefficient of haze (CoH), and sulfate (SO4=) were collected
at 17 sampling stations located across a 280-mile corridor between the Windsor and
Peterborough areas of Southern Ontario. Correlations were examined for relationships among
environmental variables and between environmental variables and three categories of hospital
admissions for winter (January-February) and summer (July-August), i.e. for total respiratory
admissions (TRA), TRA minus asthma (TRA—A), and nonrespiratory admissions, separately.
The authors concluded that an association exists in southern Ontario between O3 and
TRA and TRA—A in summer, but they note that these results are not in agreement with those
of Richards et al. (1981), who found no associations between O3 and admissions to children's
hospitals or emergency room visits in Los Angeles, where O3 levels are higher than those in
southern Ontario. They concluded, as well, that aerosol SO4= levels explain the highest
percentage variance in TRA from pollution in summer, but are not correlated with TRA in
winter. They also concluded that O3 and SO4= may be surrogates for one or more other
species that travel with them in summer but not in winter, such as hydrogen ions (H+) in the
fine-particle range. Lastly, it should be noted that Bates and Sizto (1987) specifically tested
the maximal 8-h O3 average for correlation'with TRA. The Pearson correlation coefficient
was not affected by substitution of the 8-h value in place of the mean of the hourly
3-40
-------
O3 maxima previously used. The correlation between the 1-h and 8-h O3 maxima across all
monitoring stations was 0.986.
Using the same methodology, Bates and Sizto (1989) examined aerometric and hospital
admissions data for June, July, and August 1983 and for June in the years 1979 through 1985
because June 1983 was observed to have O3 levels higher than those in any July or August
previously examined. Analyses showed no excess respiratory admissions in June 1983.
Furthermore, in years for which excess hospital admissions were observed in June (1982 and
1985), increased admissions were in the categories of "acute bronchitis" and "asthma," but
not in other respiratory categories, a finding inconsistent with O3-associated excess
admissions reported earlier. The authors concluded that these findings cast doubt "on the
primacy of O3 as the cause" of increased admissions, and that there are reasons against
attributing excess admissions either to O3 or SO4=.
Although asthmatics are not unequivocally more sensitive to O3 than nonasthmatics,
neither have they been shown to be less sensitive (U.S. Environmental Protection Agency,
1986). Therefore, the findings of an epidemiologic study of asthmatics reported by Gong
(1987) are of particular interest. Gong (1987) studied the relationship between air quality and
the respiratory status of 83 asthmatics living in a high-oxidant area of Los Angeles County.
The study covered February to December 1983, but data analyses were limited to a 230-day
period (April 15-November 30) because of staggered entry of subjects into the study and the
high frequency of missing or incomplete data encountered in the earlier part of the study
period.
Regression and correlation analyses between O3 and average symptom scores, asthma
medication index (AMI), and day and night PEFR across subjects showed weak,
nonsignificant relationships. These daily outcome variables were compared for days with
maximum 1-h-average O3 in three ranges: <0.12 ppm, 0.12-0.19 ppm, and >0.20 ppm;
"no statistical or clinical significance was detected." Individual exposures and activity
patterns were not estimated in these two analyses. Multiple regression analyses also indicated
the lack of a significant overall relationship between O3 (and other independent variables) and
respiratory status, despite the use of lagged variables and the inclusion of other pollutants,
meteorological variables, aeroallergens, and AMI. Total suspended particulates directly ,
affected PEFR, but the relationship was not consistent in the analysis. Aeroallergens showed
3-41
-------
significantly negative relationships to respiratory variables, but only the effect of trees was
considered clinically relevant. Temperature and humidity showed no significant effect on the
respiratory variables in this study.
Although there was no significant overall effect of O3 on respiratory variables in the
83 asthmatic subjects, multiple regression analysis of subjects whose O3 coefficients on
various days were in the top quartile for dependent variables (respiratory measures) showed
significant and consistent effects of O3 on day t and the previous day (t— 1). Multiple
regression testing of subsets for associations of symptom score or day or night PEFR on the
same day's O3 and the previous day's value of the same responses showed highly significant
O3 coefficients for-all three respiratory measures.
The clinical significance of responses in symptom scores and day and night PEFR was
evaluated for all subjects by individual regression analyses. No subject had evidence of
significant worsening of symptoms attributable to O3 during the study. Adult subjects with
high scores in fatigue, hyperventilation, dyspnea, congestion, and rapid breathing in the
Asthma Symptom Checklist had more negative slope coefficients for O3 than subjects with
low-to-moderate scores on the checklist. "Responders" (statistically identified by multiple
regression analysis) scored consistently higher in the the factors representing fatigue,
hyperventilation, and rapid breathing. The higher scores of these responders, however,
"were apparently not associated with differences in ambient ozone concentrations since the
test scores were similar during relatively low (first test) and high (second test) ozone days.
The significance of the psychological results is unclear at this time and will be the subject of
further analyses" (Gong, 1987).
Lastly, it should be noted that a reanalysis by Schwartz et al. (1988) of the Los Angeles
epidemiology study of student nurses earlier reported by Hammer et al. (1974) confirmed the
Hammer et al. results showing oxidants to be significantly associated with eye discomfort and
cough. However, earlier reported associations between oxidants and headache or chest
discomfort were not confirmed. The relationship of oxidants to eye irritation has been
previously reported by others and appears to be related to peroxyacetal nitrate (PAN) rather
than O3. Cough has been shown, however, to be a respiratory symptom related to O3, and
the present reanalysis (using logistic regression models and time-series analyses controlling
for autocorrelation effects) confirmed the presence of an apparent threshold for "cough"
3-42
-------
earlier found by the "hockey-stick" function analysis done by Hammer et al. (1974). The
upward flexure point in the dose-response curve occurs near the value reported by Hammer
et al. (ca. 20 pphm total oxidants)—a value likely including O3 levels well above the current
1-h O3 NAAQS of 0.12 ppm.
3.1.3 Laboratory Animal Studies
The more recently published reports on the animal toxicology of O3 were evaluated
according to their overall relevance to the issues of O3 toxicology described below. A report
not clearly applicable or unique in its contribution was not considered. Hence, studies that
added little or no data or insight to the issues being addressed, or that corroborated or tended
to duplicate the content of other studies contained in the Ozone Criteria Document
(U.S. Environmental Protection Agency, 1986) were eliminated in order to summarize the
newer pertinent data as briefly as possible. Additional literature has been selected for review
here that contained information on (1) the effects of multihour and multiday exposures to O3,
(2) the potential health effects of chronic O3 exposure, and (3) the conceptual and empirical
linkages between animal and human O3 toxicology (i.e., extrapolation). Information on a
less-specific, but nevertheless important, aspect of O3 toxicity (e.g., "adaptation") is given
here as well.
3.1.3.1 Effects of Multihour Exposures
Three studies on the effects in animals of multihour exposures to O3 (Table 3-4) have
been reported (Van Bree et al., 1989; Rombout et al., 1989; Costa et al., 1989). Results of
these studies point to the fact that concentration (C) dominates duration of exposure (T) in
eliciting a toxic response to the lung as determined by lavagable plasma protein on the lung
surface. All three studies suggest that exposure C and T can be modeled mathematically and
clearly demonstrate the dominance of C in eliciting effects. Although the effect of T on
response is clearly C dependent, the influence of T is apparent at all levels, with some
indication that C and T interact in a synergistic manner in the low C-long T exposures.
Although further work on this last point is needed, it appears that the C x T approach only
holds for a given C and cannot be applied in a general fashion.
3-43
-------
0\
SI
0\
o\
8
00
§ r3
*
S,
<2 I
a g
1 I B -I *
§ 11 s I
•a g fc &< .|
o PH 8 S a
o
u
P,
to
1
CO VO OO C-
8g
^g 15
11
o o
o
o o o
o 10 o
\O CM -* 00
O\ O\ oo \o
O oo
o
o-
3-44
-------
Costa et al. (1989) have attempted to address whether the apparent cumulative loss of
lung function seen with O3 exposure in human subjects also occurs in experimental laboratory
animals. As reported in humans by Folinsbee et al. (1988), FVC fell in-a linear fashion with
estimated cumulative exposure, which incorporated ventilation, but only at lower
concentrations (<0.5 ppm for up to 7 h). At 0.8 ppm, the effect of T on the C response
dramatically increased, as was seen in, their matrix studies of C x T relationships and in
similar studies by Van Bree et al. (1989). Hence, the impact of T is C dependent. It should
be noted, however, that the apparent cumulative toxicity of O3 may be endpoint dependent as
well and that the simple loss of lung volume, FVC or FEVj, may demonstrate such a
relationship (linearity) more clearly than more interdependent measures such as DLCO and
N2 washout.
3.1.3.2 Effects of Multiday Exposures
An animal model has been developed (Costa et al., 1988) that exhibits the same pattern
of attenuated response to intermittent short-term O3 (a phenomenon known as "adaptation")
as has been described in humans. More specifically, repeated 2-h exposures of rats for up to
5 days resulted in adaptation or attenuation of the O3-induced functional deficits, with
sustained but not worsening protein accumulation occurring in the lavage. However, the
histopathology of these animals appeared to worsen and evolve from an acute to a more
chronic inflammatory pattern. Recovery or exposure points beyond 5 days were not
conducted. Antioxidant levels of the lung tissues showed a slight upward trend during this
period, but their role in the pattern of response is unclear. This model demonstrates that
morphological and biochemical changes continue even while lung dysfunction attenuates with
repeated O3 exposure, suggesting that the use of lung function tests alone to assess injury can
result in misinterpretation of risk to health with repeated exposures to O3. Whatever the
precise mechanisms and attenuation events, however, animal studies have demonstrated that
chronic exposures cause effects, some of which are irreverisible.
The protein and PMN response to repeated 12 h nocturnal exposures for up to 3 days as
an analogue of an O3 "episode" appeared to be governed by the initial exposure only
(Van Bree et al., 1989). In other words, the degree of response and recovery time were
unaltered by additional exposures during the 2- or 3-day period.
3-45
-------
3.1.3.3 Effects of Chronic Exposure to Ozone
Several recent reports on O3 effects in laboratory animals have focused on the structural
alterations of the distal lung associated with prolonged, repeated exposures (see Table 3-5).
In both the adult and neonate rat (Barry et al., 1988; Grose et al., 1989; Huang et al., 1988;
Gross and White, 1987) and the monkey (Tyler et al., 1988; Hyde et al., 1989), high
(£0.25 ppm) ambient levels of O3 appear to similarly affect the junctional airways of the
distal bronchioles and the proximal alveoli. Shifts in cell population occur that result in more
cuboidal cells interfacing the airway lumen, effectively presenting less cell surface to the air,
and presumably reducing individual cell dose (Crapo et al., 1985; Barry et al., 1985, 1988;
Sherwin and Richters, 1985). Interstitial inflammation predominates over time, resulting in
thickened septal areas that do not completely recover during several weeks of postexposure
clean air (Huang et al., 1988; Barr et al., 1988; Moffatt et al., 1987). These findings are
largely consistent with the reports of enhanced collagen deposition and reduced turnover with
very high ambient levels of O3 (0.57 to 0.8 ppm) in monkeys (Reiser et al., 1987) and rats
(Hacker et al., 1986; Pickrell et al., 1987), but appear discrepant with collagen analyses in
chronically exposed rats at very low O3 concentrations (Filipowicz and McCauley, 1986;
Wright et al., 1988), unless exposure is intermittent (Tyler et al., 1988).
A preliminary report from the U.S. EPA's chronic O3 study (Grose et al., 1989)
showed that repeated daily exposure of rats to a daily episodic profile of O3 (22 h, 0.06-ppm
background with a 0.25-ppm peak; equivalent to a square wave that averaged 0.19 ppm over
9 h) for 12 mo resulted in small, but significant, decrements in lung function that were
consistent with early signs of focal fibrogenesis in the proximal bronchoalveolar junction (see
Chang et al., 1991). Augmentation of lavagable protein levels and tissue fractions of
ascorbate- and glutathione-related enzymes after 12 mo of O3 exposure were indicative of the
continued oxidant challenge. Further results of these studies through to 18 mo of exposure
and with recovery periods will be published. The functional implications of these alterations
in distal airway architecture have been explored in one higher level O3 study (0.5 ppm) in
which airflow mechanics were reversibly altered (Gross and White, 1987). Changes in
lavagable enzymes in rats (Grose et al., 1989) and lipids in monkeys (Rao et al., 1985a,b)
after prolonged exposures are consistent with shifting cell populations and/or inflammation,
3-46
-------
c
o
•-E3
3
•-1
VH
g
O
g-
§
•a
2
"g
o
C
O
U
1
S
0
t
2
-.
II
S Oi
0-1 "*
2 ol
O 0
m o
en t-
^4"
«
I 1
concentration dependi
no age-related differe
cuboidal cells. Epithelial
thickness increased,
interstitial macrophages
doubled.
oo
ON
"•••'
"S3
i?
m
1
Structure of the termi
bronchiolar cells is
ti_j
Decreased ciliated surface oi
Clara cells and number of
•^j- -u-
2! §
2 s
c
w
O4
.-i vo
a
o'
o
ON
fl W
significantly altered;
age-related difference
brush cells per square
millimeter of terminal
bronchiolar basement
membrane.
C/3
"3
•a
>
<^*s
*rt 2i:
S ^^
c3 !Hi
N j^
.y d>
? 3
O cS
J"T ( ^J>
n « ^.
o^2
Prolonged exposure t
increases the turnovei
of lung collagen and
60
rH (IN
No consistent increase in lui
collagen or total protein
content. Significant increasi
la
M
^ "
1 a
^ 3
04 en
04 U-l O
— < O4 >O
CD O O
>n o o
•^t" ON OO
O4 **• ON
protein content.
g||l
'•S & °<
2 -3 Xo ^
0 ^S ^ 04
fr2 o -;
8 -a 2 °
^ 03 «,
O C w 5?
'S S O °
llg1!
P!l3
.s .a .s «s
i
ON
Al
_
3-47
-------
3
(^
B
M
o5
g
O
u
i
1
Conclusions
Observed E
So
f^
° 8
fi£
Ex
§
c^
o
S
0
o °°
x O\
o d<
-
l
b^'
fi'l
-
iPllllllll
-
-q
s
«-a:
o S '
•3
-l 8 8
-S §8=5
.§ 1, S 1 1
S
S. S
s@
2
D
;o
.. 8
c> o
oo o
—t O\
ooo
O
t~;
O
0
§
>
ll
•s-i"
i-s-s
H « -2
U 2 ^
-
i
»«-3
S
0)
.
^ o £
oiog
>o o< c
o
00
•3
1
8|»
I.
llll
nit
- -
Q CS 0
00
o
oo
VO
3-48
-------
P
i
1
i
k
H
i
^
•1
U3j JfcFJbMJTS UN Jb
X)
1
SI
D
'J
fT
D
*^,
•pj
J
— *.
3
H(
3
w.
o
3
J
^
H
c
.0
1
s
Q
e
8
§
•§
CO
a
_o
1
Observed Effects'1
s
1
•3
Q
g
PH
S -2
o S
p, rt
w
.0
1
o
C
O
o
ID
I
O
g
a!
m
2
"eb
oo
*-H
13
4-t
nt
w
tl
ill
jtf ^. .S •
1 S 1 3
!lll
a) '" to Cu
! e? •§ £
d1! 11
*o a
« -a -2 -a
•g 1* g J & 5 §* 1
rt S S '! ! 0"1 "s - -g
i!fliii^!l
ggS^i^-S^SS*
^•^3oSgvo>
13C'£OSoC'aSca>->>%
3 O-— &,« 3 Ci,—, 6h"v3-^Ti
F^j 55 Pi X .^ Q X O ,;ij-' CO
g.
1
S
«5
e
c
0
PQ
O
t-<
c2
ft
•^ o
IT)
d
o\
Ol
oo
c,
T3
2
1
ffi
1 <|
w G
o o S
w 61) *^
o jj tf
"11
m 03 rt
? n °
^ '" ^
111
and cell changes in
ional and respiratory
ium of nose. Shortened
III
o to o!
8.-at3
.2 g vo fr
Q rt "ii y5^
§ '« e ^
||ll
ifll
oo
oo
60
a
1
>
2
q!3
d d
oo o
3-49
-------
p
•
1
w
1
i
in o
—« en
d o
rl- OO
rj- vq
O O
O\
r- o
10 •->
CJ *-*
IT) O
i-c en
ci d
ON 00
3-50
-------
e
o
"5
}-«
3
8
5
a)
o
ji3
(S
Conclusions
Observed Effects'"
w
03
1
l
o
o
o
a
e
ft S
X
W
£
O
"•S
g
£J
1
c
o
tj
§
R
0
a
ft
mci
&0
=L
j^.
QO
^
73
•4-t
v,
w
•53
P5
s collagen synthesized
ing exposure was
fil
.g
'5
g
S> •
45 60
73 .S
o 43
eo .a
§ 11
'g 0
tg T3
S g
0 £
£ 73
W
gj
^J*J
*o ^
S < ' rj 0} *O O
^ a £ '^ &J3" 2 If .
O d> *^5 O O 3 ._*. ^*
•*-* o £* *w M ^
t» C3 -7! "o >3 3 ft'*3 a,
ca * 4J 55 o fl x *a .K<
i ft'! | .i — | •- •§
j? ?•> o n ft § o *2 "3
PHfto pq8e.S§
S 60
o ^ fl "^ ^2
S 3 ^^ '^ § t?-S .
woo' 2-S'ftS • 1 ,3 "& §
CC f~*i W CO £•• ^M O ^4-J
2^S «»"£- MJ-WC
g ° &-S «M 8 g^g^
•S II g 73 o -3 • § 2 §
OwvS mJEJli) UcS.2
o3
cn ^
% "i is*
.s js g "i
w i o" 1
60 , £
• *~i OO Q ^
§"*"* ' ' "r^ S ^
j, o S oo g
8 73 H H ^3 g
'^"^•i ^.b .S -s §
iS-a^ S-^ogg
t-~iovo ooOSeoo
to
c*"i oi
d d
00 O
OO CTi
10 •*
« _
Intermittent group: Increase i
total lung collagen, chest wall
<+«
o
>
1
60
60
1
1
1
O
O
T3
c
ca
I
2
1
60
.g
JP
"o5
•o
inspiratory capacity.
^•^
00
oo
c,
73
tj
M
60
jC
^
C*l
S
O rrt 60
o 5 4S
*t -S 7?
2 73 « 8 eb
*T3 *^H fl ^t"*
5 ca '** co o
W) *j <£ 'tn -u>
1 i |JI
1 D 1
60 -g
l|!
.S |° S
g
^aj
6
f~i
\L
\D* O
». a
o o
odd
>o o o
en o\ oo
C-l Tf C^
S3 C
CO CJ
•a
incorporation into collagen an
total protein in O3 exposed
rats.
a
H
y
0 J
li ig
•S '3
VH co
§
>
^
>» to
tional residual capaci
srol acyltransferase, n
U ^W
S w
73
II ^
O .3
!i
§ n
o 3
Ǥ J.
^t
60^
fp( cti
^ 1
II S
O ^-,
y »-^,
rj
- II
_T "
§ <;
Kn ^^
VJJ PT
2g
•gfe
II &
c. «
Z >
s"3
2: ?i 8
2; S >
1^ t| *Ti
*-* ", f*j O
^ "" *S
8-s i
E .2 S
II '"^H
-b ||
1 11 J
P-i m M
3-51
-------
but remain nonspecific effects that still need to be linked with progressive injury or defensive
adjustments to the O3 challenge.
Ozone also has a significant impact on nasal epithelium and mucosal lining (Harkema
et al., 1987a,b). The health significance of this finding is uncertain, but is consistent with
the deposition data on O3 from both animal and human studies. Hence, though O3 is
relatively insoluble in water, the nose appears to provide some degree of scrubbing, and thus,
provides protection to the deeper lung. Species differences in this capability are an important
extrapolation question (see below).
3.1.3.4 Animal-to-Humaii Extrapolation
The more recently published studies cover two aspects of extrapolation: (1) models and
their validation and (2) species comparisons.
The Miller model (Miller et al., 1987a,b; Overton et al., 1987; Miller and Overton,
1989) of lower respiratory tract deposition of O3 has been enhanced with the incorporation of
both ventilatory parameters and empirically derived anatomical data (see Table 3-6). Use of
the model with input parameters from several rodents and humans indicates preferential
deposition, and presumably associated injury, in the bronchoalveolar junction, which is
consistent with empirical findings in laboratory animals. The model agrees well with the
total and partitioned uptake values determined in human studies (Gerrity et al., 1988), though
it fits less well with the rodent uptake data (Wiester et al., 1988). Although the reasons for
this are not as yet clear, the overall consistency of the predicted deposition distribution within
the lung and the approximate equality of dose rate/surface area increase confidence in the
model (Gerrity and Wiester, 1987).
Both human and animal uptake studies of O3 have been conducted (see Tables 3-1 and
3-6). Although humans (Gerrity et al., 1988) appear to retain a somewhat greater fraction of
the inhaled O3 than do rodents (Wiester et al., 1988), the biological significance of this
difference is uncertain at this time, especially considering the slight differences in technique.
Santrock et al. (1989) have shown that with continued exposure, products of O3, as indicated
by an oxygen-18 (18O) label, accumulate in-the lungs of mice with continued exposure. The
difference in total uptake between humans and laboratory rodents may result in part from
differences in nasopharyngeal removal of O3 (40% in humans, 17% in rats; as reported by .
3-52
-------
8
§
£
aj
Conclusions
Is
c^
w
•8
53
1
Is
1
oo
c
o
•B _
20
u
^ o
£ 1
w *O
0 g
P;
g
a ^
_o
1
t:
1"
^ "sfa
d1*
B
1
CD
•d
9^
w d-
w
.2
"3
-a
§1
ll
* 2
_co 'J3
VH « CO
_CD " CD
III
O
•^2 fe u.
1 7 „- M o
S e c3 "^ -§
'io B oo j« g
§'§S ?-* CO
J 8 °. &
fS ^ o o •§ •*
lie
W O r-
M p S P
5 JS * >3
CD
o
1s
o c
^^
tn 15
" |
E -* •
CO ^"
O CD
00 -*
00 00
IO t—
juitjiuu piiase reactivity anu nanna ei ai. (lyoy)
surface liquid thickness
contribute most to the O3
tissue absorption rate and,
therefore, to the
determination of local
tissue dose and its regional
distribution.
Tracing 18O in tissues and Hatch and Aissa
tissue subfractions (1987)
following 18O3 exposure is
feasible and practical.
rs CD
*SJ ' 0*8 * -3
sift
IJUtftlOUll
ii!lll!!il!l!ii
^§«^^i?^^£ f§ £ g ;3 • 3
CO
'.5 ,., CD
25 *2..g
3
3
1 • .' . * :
O
a .-;
3 •
• s
Ch
i — i
There are species Hatch et al. (1986)
differences in susceptibility
to O3. Mechanisms not
clear but antioxidants are
not inversely correlated.
"o
BAL protein increases after
03 were most marked in
guinea pigs (>0.2 ppm).
Mice, rats, and hamsters at
>1.0 ppm. Rabbits at
2.0 ppm. Not body
weight/size dependent and n<
comparable in order of
sensitivity to COC12.
r, "§ to
O *^ r*
S Q -2 a
*rj ^) *^
ro n !> rn Ci
F_> i^ to *-r
co GO . .'ti ^
GO 1 c& 2 ^ £ o
^
i
01 o
O
-------
o\
I
F-3
5
on
a .9 :§ |
« s cs s
Pi O P*5 ffi
60
I
u
-a S
|1
s
p" ill I"
o* *rf
vo
CM O
o cs
U
s
3-54
-------
_o
Is
Q
fO
1
X
a
.2
2
•5
CO
o
§
^•*
rf
w
1
§
*4-H
CO
K
Conclusions
%
w
Observed
actual dose was
concentration dependen
"c3
T3
•-B :
4-i
O
CD
60
2 §5
•« ;H
C cti
§1
00
OO
o\
^
"(3
t)
c3
4— »
J
^
.S -a
om§
<4_, ^
o cs
!l
o S
1 1
£ S
;
CO
!.s
|l
>,13
o *-<
||
^1
<0
ll
^ >^
^J" »-i
/^ ^
1 &
pL Q
^
en \o
w
fe
3-55
-------
Hatch et al., 1989), resulting in shifts in regional doses in the two species (surface area
differences and other factors are incorporated). Significant biological variations in lung tissue
concentrations of several antioxidants have also been reported (Bryan and Jenkinson, 1987;
Slade et al., 1985). How these antioxidants act individually or collectively as a defense
against exogenous oxidants is not clear, especially considering that five animal species tested
for O3 toxicity in concentration-response studies using BAL protein did not show
corresponding variations in their susceptibilities (Hatch et al., 1986). Thus, target tissue
dosimetry data, such as that being pursued with 18O are needed, along with additional species
sensitivity data to refine this issue. Nevertheless, the ability of the mathematical model to
discern relative species sensitivities is encouraging, despite its evolutionary state (Miller and
Overton, 1989). Further work is still needed,- however, to clarify various input components
of the model, such as the roles of reactive surface fluid components and regional ventilation,
for example, thereby ensuring its continued refinement and applicability to the extrapolation
issue (Hanna et al., 1989).
3.2 SUMMARY AND CONCLUSIONS: HEALTH EFFECTS DATA
Concisely summarized below are the key findings and conclusions that emerge from the
above review of the most pertinent, key O3 health effects studies published in the 1986 to
early 1989 period after completion of the 1986 U.S. EPA Ozone Criteria Document.
The newer studies reviewed from the 1986-1989 period provide further information
related to evaluation of several key issues pertinent to decision making regarding potential
revision of the primary O3 NAAQS. Such issues include: (1) clarification of exposure
dynamics (i.e., characterization of effective exposure patterns-concentrations, durations, etc.)
determining the induction of acute effects (pulmonary function decrements, respiratory
symptoms, lung inflammation, etc.) associated with short-term O3 exposures; (2) evaluation
of potential increased (or decreased) susceptibility of various population groups earlier
hypothesized as possibly being at differential risk for O3-induced health effects; (3) evaluation
of the potential for induction of chronic lung damage/disease by repeated short-term and/or
more chronic O3 exposures; and (4) additional clarification of O3 dosimetry aspects of
3-56
-------
importance in better understanding observed human exposure effects and to enhance
capabilities to carry out both qualitative and quantitative animal-to-human extrapolations.
3.2.1 Exposure Dynamics for Short-Term Ozone Exposure Effects
Newer data from 1- and 2-h controlled human exposure studies (Avol et al., 1987; Linn
et al., 1986) add to earlier existing concentration-response data indicating that lung function
decrements occur in children and young adults exposed for 1 to 2 h to low O3 concentrations
ranging from 0.12 to 0.16 ppm while performing moderate to heavy exercise. Explanations
for variations across studies in reported lowest-observed-effects-levels among individuals and
among cohorts include: subject characteristics, exposure histories of subjects, exercise levels,
and possible but presently unidentified differences in actual controlled exposure conditions.
Data from two other newer studies (Gong et al., 1986; Schelegle and Adams, 1986)
also substantiate earlier findings that statistically significant reductions in maximal exercise
performance may occur in well-conditioned athletes after performing continuous heavy
exercise (VE > 80 L/min) for 1 h at O3 concentrations >0.18 ppm, but not at 0.12 ppm.
Data from a third study (Linder et al., 1988) suggest that small decrements in maximal
exercise performance may occur at O3 concentrations <0.18 ppm, but limitations and
questions concerning this study require further verification of the results. Environmental
conditions such as high ambient temperature and/or relative humidity may affect subjective
symptoms and may independently impair exercise performance such that differentiation
between O3-induced effects and effects of other environmental conditions may be difficult.
In addition, newly emerging controlled human exposure studies of prolonged exposure
(for up to 6.6 h) to low O3 concentrations.ranging from 0.08 to 0.12 ppm report
progressively larger pulmonary function decrements and increased respiratory symptoms with
increasing duration of exposure at moderate exercise levels (VE =?= 40 L/min) (Folinsbee
et al., 1988; Horstman et al., 1988, 1989). The effects are similar in magnitude to those
previously reported for healthy subjects performing heavy exercise (VE > 60 L/min) in high
ozone concentrations (>0.2 ppm) for shorter durations, (~2 h).
In addition to the above, new data also demonstrate increased lung inflammatory and
biochemical changes from exposures to moderately high levels (0.40 ppm) of O3 for 2 h with
intermittent exercise (VE = 70 L/min), as determined from BAL 18-h post-O3-exposure
3-57
-------
(Koren et al., 1989a,b; 1988a,b). Cells and enzymes capable of causing damage to
pulmonary tissues, along with proteins involved in fibrotic and fibrinolytic processes, were
increased at 18-h postexposure. Also, evidence of increased epithelial permeability from the
air to blood compartments (as determined by clearance of technetium-labeled DPT A) was
observed (Kehrl et al., 1987). Koren et al. (1989a) further reported elevated PMNs, also as
determined by BAL, in subjects exposed for 6.6 h to 0.1 ppm O3.
Three newer studies on the effects in laboratory animals of multihour O3 exposures
provide information on relationships between concentration (C) and duration (T, time) of
exposure. Rombout et al. (1989), Van Bree et al. (1989), and Costa et al. (1988) report that
C dominates T in eliciting O3-induced changes in lavagable protein and antioxidant enzyme
levels. Preliminary modeling efforts describing these data suggest that C x T interaction
(synergism) occurs at decreased C and increased T; however, C x T relationships can only
be applied at a given C and cannot be applied in general. The time of day of exposure is
also an important determinant of oxidant toxicity because nocturnal exposures cause greater
responses than do diurnal exposures, possibly due to dosimetric differences because rats
exercise more at night. These results suggest that the primary determinants of acute O3 lung
toxicity may, therefore, be exposure concentration and time of day of exposure, followed by
the duration of exposure.
Also, certain other newer information suggests that interpretation of the results of
controlled human O3 exposures should take into account whether frequent ambient exposure
was a possibility during the period of study (Avol et al., 1988; Hackney and Linn, 1989;
Hackney et al., 1989; Linn et al., 1988). This information also suggests that further work is
still needed to resolve the implications of attenuation of pulmonary function responses to
O3. Subjects grouped according to their responses in the early spring to 0.18 ppm O3 for
2 h with intermittent exercise were tested the following fall, winter, and again the next
spring. Although "nonresponders" showed little seasonal variation in their response to
O3, "responders" showed attenuated responses in the fall, persistence of attenuation into the
winter, and a return to their initial lung function responses to O3 by the following spring.
Many of the responders were reactive to methacholine and had histories of respiratory
allergies and/or symptomatic complaints when previously exposed to smog.
3-58
-------
Newer field and epidemiological studies have employed numerous refinements over
some of the older studies, in the form of (a) better estimates of exposure, not just to O3 but
also to other pollutants and other environmental variables that can confound or otherwise
influence the outcome (e.g., Bates and Sizto, 1987; Spektor et al., 1988a,b; Raizenne et al.,
1987); (b) use of serial measurements of pulmonary function for determining correlations
with pollutants and other environmental variables (e.g., Raizenne et al., 1987, 1989; Spektor
et al., 1988a,b); and (c) better biomedical characterization of cohorts (e.g., Raizenne et al.,
1987, 1989; Gong, 1987). Despite their refinements, however, newer field and
epidemiologic studies have produced mixed results regarding the possible role of O3 versus
the roles of other agents or factors in eliciting the functional decrements and/or rates of
respiratory symptoms or respiratory disease observed. Although functional decrements and
respiratory symptoms have been shown in a number of studies to be statistically associated
with O3, other studies have shown them to be wholly attributable to particles (e.g., Dockery
et al., 1989), partially attributable to particles (e.g., Kinney, 1986), or partially attributable
to other environmental factors such as ambient temperature or humidity (e.g., Spektor et al.,
1988a) or even aeroallergens (e.g., Dockery et al., 1989; Gong, 1987).
Data reported from some of the newer field and epidemiologic studies (e.g., Raizenne
et al., 1987, 1989; Spektor et al., 1988a,b; Kinney et al., 1988) show pulmonary function
decrements that are as large or larger than those observed in human controlled (chamber)
studies. Investigators have variously interpreted these larger decrements as likely being due
to: (a) cumulative effects of O3 occurring as the result of multihour exposures;
(b) interactive effects with other pollutants (additive or synergistic effects); (c) interactive or
possibly independent effects of other environmental factors (e.g., temperature); (d) possible
misspecification of true exposures, either because of inadequate knowledge of dosimetry or
other types of inadequacies in exposure characterization; and/or (e) possible persistence of
effects from one day to the next.
Although permitting easier comparison of epidemiologic findings with chamber-study
data, the method of reporting raises several questions that need to be further evaluated. For
example, data on functional decrements have been reported as — mL/ppb O3 for measures
such as FEV1 0 and FVC and as -mL/sec/ppb for measures such as PEFR and MMEF.
Expression of data in this form assumes that (a) O3-induced changes in respiratory function
3-59
-------
are linear across all concentrations encountered in these studies (from zero up through
episodic levels); and (b) the relationships among C, T, and VE do not change with variations
in these respective components of exposure. These assumptions are open to question,
especially considering results from newer animal C x T studies. For example, the
relationships between respiratory function changes and the respective components of exposure
—C, T, and VE—have not been tested at concentrations <0.08 ppm in chamber studies; and
data obtained in chamber studies at the lowest concentration used (0.08 pprn) have not been
modeled to determine whether changes in the influence of respective components are
monotonic across ranges of C, T, or VE. Furthermore, questions of nonlinearities in the
respective effects of C, T, and VE- on O3-induced pulmonary function changes are far from
resolved. Also, Kinney et al. (1988), transformed data from controlled (chamber) studies
modeled by Hazucha (Hazucha, 1987; U.S. Environmental Protection Agency, 1986) and
compared them with data from five epidemiologic studies. The transformation assumed the
applicability of a linear model even though Hazucha had fit data from controlled (chamber)
studies to a quadratic model in describing changes in pulmonary function as a function of
VE. Mean concentrations in the five epidemiologic studies were lower than the lowest
concentration used in the controlled studies modeled.
As for respiratory symptoms, the newer field and epidemiologic studies have reported
lack of association of various respiratory symptoms with O3 more often than they have
reported demonstration of such an association with the typically low O3 levels studied.
Studies reporting no significant increases in symptoms following short (1-h to multihour)
daily exposures (over multiple days to multiple months) to ambient air containing O3 include:
(a) studies of children attending day or residential camps (Raizenne et al., 1987, 1989;
Spektor et al., 1988a), (b) at least two panel studies (Dockery et al., 1989; Vedal et al.,
1987), and (c) a study of adults exercising outdoors nearly every day (Spektor et al., 1988b).
Another study (Schwartz et al., 1988), reanalyzing data from the Hammer et al. (1974) panel
study of nurses in Los Angeles and using currently more widely accepted statistical
approaches, did show that cough was associated with total oxidants, but only at relatively
high levels (where O3 concentrations were likely well above 0.12 ppm).
3-60
-------
3.2.2 Evaluation of Differential Susceptability of Potential Special Risk
Groups
Newer data published in the 1986-1989 period provides further data useful in evaluating
possible differential sensitivity in two of several subpopulation groups of interest: (1) the
elderly and (2) asthmatics. With regard to the first group, in controlled human exposure
studies, older subjects (>50 years old) appear to have smaller changes in lung function than
younger subjects when exposed to similar O3 concentrations (Bedi et al., 1988; Bedi and
Horvath, 1987; Drechsler-Parks et al., 1987, 1989; Reisenauer et al., 1988). There were no
significant differences between the responses of men and women to O3 exposure for FEVj
and FVC, although women had a significant increase in total respiratory resistance
(Reisenauer et al., 1988). Because women had slightly lower mean exercise VE during the
studies, the data suggest that women may be somewhat more responsive to O3 than men
(Drechsler-Parks et al., 1987; Reisenauer et al., 1988). The responses to O3 may be less
reproducible, however, in older than in younger adults (Bedi et al., 1988). These results
suggest a possible dropoff in responsiveness to O3-induced pulmonary function changes
sometime in late middle-age.
/
As for asthmatics, in other new studies of adults with and without asthma (Kreit et al.,
1989; Eschenbacher et al., 1989), both groups experienced similar responses to 0.4-ppm
O3 exposure, as indicated by decrements in standard spirometric pulmonary function tests and
airway responsiveness to methacholine. Specific airway resistance was not increased in
nonasthmatics, but in asthmatics nearly twice the increase in SRaw was seen after exercise in
Q3 versus air exposures. No symptom differences were observed between adult asthmatics
and nonasthmatics. Preexposure challenge with methacholine may have confounded the
results, however. Responses were also similar for adolescent asthmatics and nonasthmatics
exposed to 0.12 and 0.18 ppm O3 (Koenig et al., 1987, 1988), although a small but
significant increase in FEF50% was observed in asthmatics after 0.12 ppm O3 exposure. In
the adult nonasthmatics studied by Eschenbacher et al. (1989), indomethacin pretreatment
blocked the restrictive but not the airway reactivity component of the effects of O3; a placebo
effect was also observed in these nonasthmatics. A study by McDonnell et al. (1987)
indicates that adults with allergic rhinitis show similar airway responsiveness to histamine
after exposure to 0.18 ppm O3 as a comparable group of nonallergic subjects. The only
3-61
-------
difference was a significant increase in airway resistance in the allergic subjects. The newer
data on allergic and asthmatic subjects suggest that both of these groups have a greater
increase in airway resistance after O3 exposure than do healthy subjects. The apparent order
of airway responsiveness to O3 from these studies is normal < allergic < asthmatic subjects.
Further research will be required, however, before this hypothesis can be substantiated.
Results from a few of the newer epidemiology studies are also suggestive of possible
increases in asthma or allergic responses among some individuals. For example, Dockery
et al. reported a positive association of annual O3 exposures to increases in asthma and hay
fever symptoms in schoolchildren, but not with other respiratory symptoms or pulmonary
function decrements. Also, in a panel study more specifically on asthmatics, Gong (1987)
reported that, although respiratory symptoms occurred during the study, they did not correlate
significantly with O3; and no worsening of symptoms attributable to O3 occurred in overall
group statistical analyses. However, other multiple regression analyses of responses of those
asthmatics in the top quartile for respiratory measures did show relationships between the
respiratory measures and O3, but these associations showed no clear concentration-response
pattern (Gong, 1987).
3.2.3 Ozone Impacts on Lung Structure/Chronic Disease Processes
Several studies, by Koren et al. (1988a,b; 1989a,b) and Kehrl et al. (1987), were noted
earlier as showing increased lung inflammatory responses and other biochemical changes
capable of causing damage to pulmonary tissue, as well as increased lung epithelial
permeability, with 2-h exposures of humans to 0.4 ppm O3 and elevated PMNs in BAL fluids
following 6.6-h exposure to low O3 levels (0.1 ppm). These results increase concern
regarding possible induction of chronic lung damage by O3 exposures and lead to questions
about effective exposure dynamics possibly related to such effects.
Costa et al. (1988) have demonstrated that laboratory animals exhibit a similar pattern
of attenuated pulmonary function changes in response to repeated short-term O3 exposure as
previously described in humans. Both morphological and biochemical changes, however,
occur while lung dysfunction attenuates with repeated O3 exposure. These results add to
concern that the attenuation phenomenon for short-term pulmonary function changes may not
be beneficial, but rather may increase the potential for more serious chronic lung damage.
3-62
-------
They also suggest that the use of lung function tests alone to assess O3-induced lung injury
may result in misinterpretation of data concerning health risks associated with multiday
O3 exposures. More research is needed, therefore, to improve our knowledge of
relationships between acute and chronic lung injury.
Other new studies in monkeys and rodents provide further support for earlier findings
that prolonged, repeated exposure to high concentrations of O3 (>0.4 ppm) lead to the
development of peribronchiolar inflammation (Barr et al., 1988; Moffatt et al., 1987),
increased lung collagen content (Reiser et al., 1987; Pickrell et al., 1987; Hacker et al.,
1986), and lung function changes (Gross and White, 1986, 1987). Even at lower
O3 concentrations (0.12 to 0.30 ppm), a lesion is still evident at the junction of the
conducting airways and the gas exchange regions of the lung, characterized by cell population
shifts along with interstitial inflammation and thickening (Huang et al., 1988; Crapo et al.,
1985; Barry et al., 1985, 1988; Sherwin and Richters, 1985), but without increased lung
collagen content in rats (Wright et al., 1988; Filipowicz and McCauley, 1986).
In a chronic exposure study (0.25 ppm, 8 h/day, 18 mo), one group of monkeys was
exposed to O3 for each day of the 18 mo, whereas another group was exposed to O3 every
other month, with the intervening month being an air exposure (Tyler et al., 1988). Thus,
on a C XT basis, the intermittent group received half the amount of O3 exposure. Both
groups of monkeys had morphometric changes, such as respiratory bronchiolitis, but only the
intermittent groups had an increase in total lung collagen and pulmonary function changes.
This suggests that under these experimental conditions, intermittent exposure could enhance
the potential for the development of fibrogenic processes and indicates that to understand the
health effects of O3, it is critical to better understand the effects ambient exposure patterns.
Preliminary information (Grose et al., 1989) from an exposure that mimics an urban
O3 pattern (0.19 ppm average concentration of O3 over 9 h) of rats for 12 mo indicates that
significant decrements in lung function also occur at these lower O3 concentrations that are
consistent with early signs of focal fibrogenesis in the centriacinar region of the lung.
Increased lavagable lipids in monkeys (Rao et al., 1985a,b) found after prolonged exposure to
ambient levels of O3 (0.15 to 0.30 ppm) are also consistent with the shifting cell populations
and/or inflammation reported at these concentrations. Multiple exposures to ambient levels of
O3 (0.15 and 0.30 ppm, 8 h/day for 6 or 90 days) also cause injury and cellular changes in
3-63
-------
transitional and respiratory epithelium of the nose of nonhuman primates (Harkema et al.,
1987a,b; Hyde et al., 1989).
3.2.4 Ozone Dosimetry Aspects
Newer studies related to the dosirnetry of O3 show that differences in mode of breathing
do not produce appreciable differences in fractional uptake of O3 in the respective regions of
the human respiratory tract. Increased frequency of breathing results in a decreased fractional
removal of O3 in both the URT and the LRT, possibly as the result of decreased residence
time in the airways and increased flow rate. The lowest fractional removal of O3 in the URT
occurred during nasal breathing, so that shifts from nasal to oronasal breathing resulting from
exercise would somewhat offset increases in delivered dose caused by increased breathing
frequency (Gerrity et al., 1988). Ozone-induced changes in tidal volume during 60-min,
continuous-exercise (VE = 40 L/min) exposures to 0.4 ppm resulted in a slight reduction in
total O3 uptake (4%) and a larger reduction in LRT O3 uptake (9%). Thus, the typical
O3-induced reduction in tidal volume may partially protect the lower airways, with possible
loss of that protection with recovery of normal tidal volume that occurs during multiday
exposure (Gerrity and McDonnell, 1989). Increased flow rates appear to reduce
nasopharyngeal uptake (Gerrity, 1987). Additional modeling is need, however, to determine
the effects of heavy exercise on regional dosirnetry, especially on O3 uptake in the LRT,
Additional data are still needed in other areas important to animal-to-human extrapolation,
namely, relative species sensitivities.
Mathematical dosirnetry models indicate preferential deposition of O3 in the
bronchoalveolar junction of several species of laboratory animals and in humans that is
consistent with laboratory animal data on the site of the O3 morphological lesion (Miller and
Overton, 1989; Miller et al., 1987a,b; Overton et al., 1987). Humans appear to retain a
greater fraction (95%) of inhaled O3 than do rats (50%), but tissue dose rates/surface area in
each species may not be that different if nasopharyngeal partitioning is considered (Wiester
etal., 1987, 1988; Gerrity and Wiester, 1987; Gerrity, 1987). Target dosirnetry data, such
as that being conducted with [18]O3 (Hatch et al., 1989; Santrock et al., 1989; Aissa and
Hatch, 1988; Hatch and Aissa, 1987) are needed, along with species sensitivity data (Bryan
and Jenkinson, 1987; Hatch et al., 1986; Slade et al., 1985) to better refine this issue.
3-64
-------
REFERENCES
Adams, W. C.; Schelegle, E. S. (1983) Ozone and high ventilation effects on pulmonary function and
endurance performance. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 55: 805-812.
Adams, W. C.; Savin, W. M.; Christo, A. E. (1981) Detection of ozone toxicity during continuous exercise via
the,,effective dose concept. J. Appl. Physiol: Respir. Environ. Exercise Physiol. 51: 415-422.
Aissa, M.; Hatch, G. E. (1988) Method for tracing oxygen-18 in vivo: application to ozone dosimetry in
animals. Research Triangle Park, NC: U.S. Environmental Protection Agency, Health Effects Research
Laboratory; EPA report no. EPA-600/D-88-001. Available from: NTIS, Springfield, VA;
PB88-153606.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Venet, T. G.; Hackney, J. D. (1983) Acute respiratory effects of
Los Angeles smog in continuously exercising adults. J. Air Pollut. Control Assoc. 33: 1055-1060.
Avol, E. L.; Linn, W. S.; Venet, T. G.; Shamoo, D. A.; Hackney, J. D. (1984) Comparative respiratory
effects of ozone and ambient oxidant pollution exposure during heavy exercise. J. Air Pollut. Control
Assoc. 34: 804-809.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Valencia, L. M.; Anzar, U. T.; Venet, T. G.; Hackney, J. D.
(1985a) Respiratory effects of photochemical oxidant air pollution in exercising adolescents. Am. Rev.
Respir. Dis. 132: 619-622.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Valencia, L. M.; Anzar, U. T.; Hackney, J. D. (1985b) Short-term
health effects of ambient air pollution in adolescents. In: Lee, S. D., ed. Evaluation of the scientific
basis for ozone/oxidants standards: transactions of an APCA international specialty conference;
November 1984; Houston, TX. Pittsburgh, PA: Air Pollution Control Association; pp. 329-336.
(APCA international specialty conference transactions: TR-4).
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Spier, C. E.; Valencia, L. M.; Venet, T. G.; Trim, S. C.;
Hackney, J. D. (1987) Short-term respiratory effects of photochemical oxidant exposure in exercising •
children. JAPCA 37: 158-162.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Hackney, J. D. (1988) Seasonal ozone reactivity in Los Angeles
residents. Presented at: 81st annual meeting of the Air Pollution Control Association; June; Dallas,
TX. Pittsburgh, PA: Air Pollution Control Association; paper no. 88-122.6.
Barr, B. C.; Hyde, D. M.; Plopper, C. G.; Dungworth, D. L. (1988) Distal airway remodeling in rats
chronically exposed to ozone. Am. Rev. Respir. Dis. 137: 924-938.
Barry, B. E.; Miller, F. J.; Crapo, J. D. (1985) Effects of inhalation of 0.12 and 0.25 parts per million ozone
on the proximal alveolar region of juvenile and adult rats. Lab. Invest. 53: 692-704.
Barry, B. E.; Mercer, R. R.; Miller, F. J.; Crapo, J. D. (1988) Effects of inhalation of 0.25 ppm ozone on the
terminal bronchioles of juvenile and adult rats. Exp. Lung Res. 14: 225-245.
Bates, D. V. (1985) The strength of the evidence relating air pollutants to adverse health effects. Chapel Hill,
NC: University of North Carolina at Chapel Hill, Institute for Environmental Studies. (Carolina
environmental essay series VI).
Bates, D. V.; Sizto, R. (1983) Relationship between air pollutant levels and hospital admissions in Southern
Ontario. Can. J. Public Health 74: 117-122.
3-65
-------
Bates, D. V.; Sizto, R. (1987) Air pollution and hospital admissions in southern Ontario: the acid summer haze
effect. Environ. Res. 43: 317-331.
Bates, D. V.; Sizto, R. (1989) The Ontario air pollution study: indentification of the causative agent. In:
International symposium on the health effects of acid aerosols: addressing obstacles in an emerging data
base; October 1987; Research Triangle Park, NC. Environ. Health Perspect. 79: 69-72.
Bedi, J. F.; Horvath, S. M. (1987) Longitudinal case study of pulmonary function response to ozone. Am.
J. Med. 82: 860-861.
Bedi, J. F.; Horvath, S. M.; Drechsler-Parks, D. M. (1988) Reproducibility of the pulmonary function
response of older men and women to a 2-hour ozone exposure. JAPCA 38: 1016-1019.
Bock, N.; Lippmann, M.; Lioy, P.; Munoz, A.; Speizer, F. E. (1985) The effects of ozone on the pulmonary
function of children. In: Lee, S. D., ed. Evaluation of the scientific basis for ozone/oxidants standards:
proceedings of an APCA international specialty conference; November 1984; Houston, TX. Pittsburgh,
PA: Air Pollution Control Association; pp. 297-308. (APCA international specialty conference
transactions: TR-4).
Bryan, C. L.; Jenkinson, S. G. (1987) Species variation in lung antioxidant enzyme activities. J. Appl. Physiol.
63: 597-602.
Chang, L.; Miller, F. J.; Ultman, J.; Huang, Y.; Stockstill, B. L.; Grose, E.; Graham, J. A.; Ospital, J. J.;
Crapo, J. D. (1991) Alveolar epithelial cell injuries by subchronic exposure to low concentrations of
ozone correlate with cumulative exposure. Toxicol. Appl. Pharmacol. 109: 219-234.
Costa, D. L.; Stevens, M. S.; Tepper, J. S. (1988) Repeated exposure to ozone (O3) and chronic lung disease:
recent animal data. Presented at: 81st annual meeting of the Air Pollution Control Association; June;
Dallas, TX. Pittsburgh, PA: Air Pollution Control Association; paper no. 88-122.3.
Costa, D. L.; Hatch, G. E.; Highfill, J.; Stevens, M. A.; Tepper, J. S. (1989) Pulmonary function studies in
the rat addressing concentration versus time relationships of ozone. In: Schneider, T.; Lee, S. D.;
Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications:
proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands.
Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 733-743. (Studies in environmental
science 35).
Crapo, J. D.; Barry, B. E.; Mercer, R. R. (1985) Morphometric studies of the effects of ozone on rodent
lungs. Research Triangle Park, NC: U.S. Environmental Protection Agency, Health Effects Research
Laboratory; EPA report no. EPA-600/1-85-008. Available from: NTIS, Springfield, VA;
PB85-207470.
Dockery, D. W.; Speizer, F. E.; Stram, D. O.; Ware, J. H.; Spengler, J. D.; Ferris, B. G., Jr. (1989) Effects
of inhalable particles on respiratory health of children. Am. Rev. Respir. Dis. 139: 587-594.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1987) Pulmonary fimction responses of older men and
women to ozone exposure. Exp. Gerontol. 22: 91-101.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1989) Pulmonary function responses of young and older
adults to mixtures of O3, NO2 and PAN. Toxicol. Ind. Health 5: 505-517.
3-66
-------
Eschenbacher, W. L.; Ying, R. L.; Kreit, J. W.; Gross, K, B. (1989) Ozone-induced lung function changes in
normal and asthmatic subjects and the effect of indomethacin. In: Schneider, T.; Lee, S. D.; Welters,
G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications: proceedings of
the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The
Netherlands: Elsevier Science Publishers; pp. 493-499. (Studies in environmental science 35).
Filipbwicz, C.; McCauley, R. (1986) The effects of chronic ozone exposure on pulmonary collagen content and
collagen synthesis in rats. J. Appl. Toxicol. 6: 87-90.
Folinsbee, L. J.; Hazucha, M. J. (1989) Persistence of ozone-induced changes in lung function and airway
responsiveness. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric
ozone research and its policy implications: proceedings of the 3rd US-Dutch international symposium;
May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers;
pp. 483-492. (Studies in environmental science 35). ;
Folinsbee, L. J.; Drinkwater, B. L.; Bedi, J. F.; Horvath, S. M. (1978) The influence of exercise on the
pulmonary function changes due to exposure to low concentrations of ozone.' In: Folinsbee, L. J.;
Wagner, J. A.; Borgia, J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F., eds. Environmental stress:
individual human adaptations. New York, NY: Academic Press; pp. 125-145.
Folinsbee, L. J.; Bedi, J. F.; Horvath, S. M. (1984) Pulmonary function changes after 1 h continuous heavy
exercise in 0.21 ppm ozone. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 57: 984-988.
Folinsbee, L. J.; McDonnell, W. F.; Horstman, D. H. (1988) Pulmonary function and symptom responses after
6.6-hour exposure to 0.12 ppm ozone with moderate exercise. JAPCA 38: 28-351
Gerrity, T. R. (1987) Nasopharyngeal uptake of ozone in humans and animals. Research Triangle Park, NC:
U.S. Environmental Protection Agency, Health Effects Research Laboratory; report no.
EPA/600/D-87/357. Available from: NTIS, Springfield, VA; PB88-140561.
Gerrity, T. R.; McDonnell, W. F. (1989) Do functional changes in humans correlate with the airway removal
efficiency of ozone? In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric
ozone research and its policy implications: proceedings of the 3rd US-Dutch international symposium;
May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers;
pp. 293-300. (Studies in environmental science 35).
Gerrity, T. R.; Wiester, M. J. (1987) Experimental measurements of the uptake of ozone in rats and human
subjects. Presented at: 80th annual meeting of the Air Pollution Control Association; June; New York,
NY. Pittsburgh, PA: Air Pollution Control Association; paper no. 87-99.3. '
Gerrity, T. R.; Weaver, R. A.; Berntsen, J.; House, D. E.; O'Neil, J. J. (1988) Extrathoracic and intrathoracic
removal of O3 in tidal-breathing humans. J. Appl. Physiol. 65: 393-400.
Gibbons, S. L; Adams, W. C. (1984) Combined effects of ozone exposure and ambient heat on exercising
females. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 57: 450-456.
Gliner, J. A.; Horvath, S. M.; Folinsbee, L. J. (1983) Preexposure to low ozone concentrations does not
diminish the pulmonary function response on exposure to higher ozone concentrations. Am. Rev.
Respir. Dis. 127: 51-55.
Gong, H., Jr. (1987) Relationship between air quality and the respiratory status of asthmatics in an area of high
oxidant pollution in Los Angeles County, [final report]. Sacramento, CA: California Air Resources
Board; contract nos. Al-151-33 and A4-135-33.
3-67
-------
Gong, H., Jr.; Bradley, P. W.; Simmons, M. S.; Tashkin, D. P. (1986) Impaired exercise performance and
pulmonary function in elite cyclists during low-level ozone exposure in a hot environment. Am. Rev.
Respir. Dis. 134: 726-733.
Graham, D.; Henderson, F.; House, D. (1988) Neutrophil influx measured in nasal lavages of humans exposed
to ozone. Arch. Environ. Health 43: 228-233.
Grose, E. C.; Stevens, M. A.; Hatch, G. E.; Jaskot, R. H.; Selgrade, M. J. K.; Stead, A. G.; Costa, D. L.;
Graham, J. A. (1989) The impact of a 12-month exposure to a diurnal pattern of ozone on pulmonary
function, antioxidant biochemistry and immunology. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.;
Grant, L. D., eds. Atmospheric ozone research and its policy implications: proceedings of the 3rd
US-Dutch international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The
Netherlands: Elsevier Science Publishers; pp. 535-544. (Studies in environmental science 35).
Gross, K. B.; White, H. J. (1986) Pulmonary functional and morphological changes induced by a 4-week _.
exposure to 0.7 ppm ozone followed by a 9-week recovery period. J. Toxicol. Environ. Health
17: 143-157.
Gross, K. B.; White, H. J. (1987) Functional and pathologic consequences of a 52-week exposure to 0.5 PPM
ozone followed by a clean air recovery period. Lung 165: 283-295.
Hacker, A. D.; Mustafa, M. G.; Ospital, J. J.; Elsayed, N. M.; Lee, S. D. (1986) Relationship of age to rat
lung collagen synthesis in response to ozone exposure. Age 9: 1-5.
Hackney, L D.; Linn, W. S. (1989) Evaluation relationships among personal risk factors, ambient oxidant
exposure, and chronic respiratory illness. In: Utell, M. J.; Frank, E., eds. Susceptibility to inhaled
pollutants. Philadelphia, PA: American Society for Testing Materials; pp. 174-181; ASTM publication
no. STP 1024.
Hackney, J. D.; Linn, W. S.; Shamoo, D. A.; Avol, E. L. (1989) Responses of selected reactive and
nonreactive volunteers to ozone exposure in high and low pollution seasons. In: Schneider, T.; Lee,
S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications:
proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands.
Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 311-318. (Studies in environmental
science 35).
Hammer, D. I.; Hasselblad, V.; Portnoy, B.; Wehrle, P. F. (1974) Los Angeles student nurse study: daily
symptom reporting and photochemical oxidants. Arch. Environ. Health 28: 255-260.
Hanna, L. M.; Frank, R.; Scherer, P. W. (1989) Absorption of soluble gases and vapors in the respiratory
system. In: Chang, H. K.; Paiva, M., eds. Respiratory physiology: an analytical approach. New York,
NY: Marcel Dekker, Inc.; pp. 277-316. (Lenfant, C., ed. Lung biology in health and disease; v. 40).
Harkema, J. R.; Plopper, C. G.; Hyde, D. M.; St. George, J. A.; Dungworth, D. L. (1987a) Effects of an
ambient level of ozone on primate nasal epithelial mucosubstances: quantitative histochemistry. Am. J.
Pathol. 127: 90-96.
Harkema, J. R.; Plopper, C. G.; Hyde, D. M.; St. George, J. A.; Wilson, D. W.; Dungworth, D. L. (1987b)
Response of the macaque nasal epithelium to ambient levels of ozone: a morphologic and morphometric
study of the transitional and respiratory epithelium. Am. J. Pathol. 128: 29-44.
3-68
-------
Hatch, G. E.; Aissa, M. (1987) Determination of absorbed dose of ozone in animals and humans using stable
isotope (oxygen-18) tracing. Presented at: 80th annual meeting of the Air Pollution Control
Association; June; New York, NY. Pittsburgh, PA: Air Pollution Control Association; paper no.
87-99.2.
Hatch, G. E.; Slade, R.; Stead, A. G.; Graham, J. A. (1986) Species comparison of acute inhalation toxicity of
ozone and phosgene. J. Toxicol. Environ. Health 19: 43-53.
Hatch, G. E.; Wiester, M. J.; Overton, J. H., Jr.; Aissa, M. (1989) Respiratory tract dosimetry of
[18]O-labeled ozone in rats: implications for a rat-human extrapolation of ozone dose. In:
Schneider, T.; Lee, S. D.; Welters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its
policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen,
The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 553-560. (Studies in
environmental science 35).
Hazucha, M. J. (1987) Relationship between ozone exposure and pulmonary function changes. J. Appl. Physiol.
62: 1671-1680.
Horstman, D. H.; McDonnell, W. F.; Abdul-Salaam, S.; Folinsbee, L.. 3,; Ives, P. J. (1988) Current USEPA
research concerning more prolonged human exposures to low ozone concentrations. Presented at: 81st
annual meeting of the Air Pollution Control Association; June; Dallas, TX. Pittsburgh, PA: Air
Pollution Control Association; paper no. 88-122.5.
Horstman, D.; McDonnell, W.; Folinsbee, L.; Abdul-Salaam, S.; Ives, P. (1989) Changes in pulmonary
function and airway reactivity due to prolonged exposure to typical ambient ozone (O3) levels. In:
Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its
policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen,
The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 755-762. (Studies in
environmental science 35).
Huang, Y.; Chang, L.-Y.; Miller, F. J.; Crapo, J. D. (1988) Lung injury caused by ambient levels of ozone.
J. Aerosol Med. 1: 180-183.
Hyde, D. M.; Plopper, C. G.; Harkema, J. R.; St. George, J. A.; Tyler, W. S.; Dungworth, D. L. (1989)
Ozone-induced structural changes in monkey respiratory system. In: Schneider, T.; Lee, S. D.;
Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications:
proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands.
Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 523-532. (Studies in environmental
science 35).
Hynes, B.; Silverman, F.; Cole, P.; Corey, P. (1988) Effects of ozone exposure: a comparison between oral
. and nasal breathing. Arch. Environ. Health 43: 357-360.
Kehrl, H. R.; Vincent, L. M.; Kowalsky, R. J.; Horstman, D. H.; O'Neil, J. J.; McCartney, W. H.;
Bromberg, P. A. (1987) Ozone exposure increases respiratory epithelial permeability in humans. Am.
Rev. Respir. Dis. 135: 1124-1128.
Kehrl, H. R.; Vincent, L. M.; Kowalsky, R. J.; Horstman, D. H.; O'Neil, J. J.; McCartney, W. H.;
Bromberg, P. A. (1989) Ozone-induced changes in the pulmonary clearance of "Tc-DTPA in man.
In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and
'its policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988;
Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers: pp. 343-351.
(Studies in environmental science 35).
3-69
-------
Kinney, P. L. (1986) Short-term lung function associations with air pollution in Kingston and Harriman,
Tennessee [dissertation]. Boston, MA: Harvard University, School of Public Health.
Kinney, P. L.; Ware, J. H.; Spengler, J. D. (1988) A critical evaluation of acute ozone epidemiology results.
Arch. Environ. Health 43: 168-173.
Koenig, J. Q.; Covert, D. S.; Morgan, M. S.; Horike, M.; Horike, N.; Marshall, S. G.; Pierson, W. E.
(1985) Acute effects of 0.12 ppm ozone or 0.12 ppm nitrogen dioxide on pulmonary function in
healthy and asthmatic adolescents. Am. Rev. Respir. Dis. 132: 648-651.
Koenig, J. Q.; Covert, D. S.; Marshall, S. G.; Van Belle, G.; Pierson, W. E. (1987) The effects of ozone and
nitrogen dioxide on pulmonary function in healthy and in asthmatic adolescents. Am. Rev. Respir. Dis.
136: 1152-1157.
Koenig, J. Q.; Covert, D. S.; Smith, M. S.; Van Belle, G.; Pierson, W. E. (1988) The pulmonary effects of
ozone and nitrogen dioxide alone and combined in healthy and asthmatic adolescent subjects. Toxicol.
Ind. Health 4: 521-532.
Keren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; Horstman, D. H.; Kozumbo, W. J.; Becker, S.;
McDonnell, W. F. (1988a) Cellular and biochemical changes in the lower airways of subjects exposed
to ozone. Research Triangle Park, NC: U.S. Environmental Protection Agency, Health Effects
Research Laboratory; report no. EPA/600/D-88/031. Available from: NTIS, Springfield, VA;
PB88-170048.
Koren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; Horstman, D. H.; Kozumbo, W. J.; Becker, S.;
McDonnell, W. F. (1988b) Cellular and biochemical changes in the lower airways of subjects exposed
to ozone, la: Sorg, C., ed. The alveolar macrophage. Werne, Federal Republic of Germany: Stiftung
Immunologische Tage; pp. 36-49. (Local immunity: v. 4).
Koren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; McDonnell, W. F. (1989a) The inflammatory
response in human lung exposed to ambient levels of ozone. In: Schneider, T.; Lee, S. D.; Welters,
G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications: proceedings of
the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The
Netherlands: Elsevier Science Publishers; pp. 745-753. (Studies in environmental science 35).
Koren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; McGee, M. P.; Horstman, D. H.; Kozumbo, W. J.;
Becker, S.; House, D. E.; McDonnell, W. F.; Bromberg, P. A. (1989b) Ozone-induced inflammation
in the lower airways of human subjects. Am. Rev. Respir. Dis. 139: 407-415.
Kreit, J. W.; Gross, K. B.; Moore, T. B.; Lorenzen, T. J.; D'Arcy, J.; Eschenbacher, W. L. (1989)
Ozone-induced changes in pulmonary function and bronchial responsiveness in asthmatics. J. Appl.
Physiol. 66: 217-222.
Kulle, T. J.; Sauder, L. R.; Hebel, J. R.; Chatham, M. D. (1985) Ozone response relationships in healthy
nonsmokers. Am. Rev. Respir. Dis. 132: 36-41.
Lauritzen, S. K.; Adams, W. C. (1985) Ozone inhalation effects consequent to continuous exercise in females:
comparison to males. J. Appl. Physiol. 59: 1601-1606.
Linder, J.; Herren, D.; Monn, C.; Wanner, H.-U. (1988) Die Wirkung von Ozon auf die koerperliche
Leistungsfaehigkeit [The effect of ozone on physical activity]. Schweiz Z. Sportmed. 36: 5-10.
3-70
-------
Linn, W. S.; Jones, M. P.; Bachmayer, E. A.; Spier, C. E.; Mazur, S. F.; Avol, E. L.; Hackney, J. D.
(1980) Short-term respiratory effects of polluted ambient air: a laboratory study of volunteers in a
high-oxidant community. Am. Rev. Respir. Dis. 121: 243-252.
Linn, W. S.; Avol, E. L.; Hackney, J. D. (1983) Effects of ambient oxidant pollutants on humans: a movable
environmental chamber study. In: Lee, S. D.; Mustafa, M. G.; Mehlman, M. A., eds. International
symposium on the biomedical effects of ozone and related photochemical oxidants; March 1982;
Pinehurst, NC. Princeton, NJ: Princeton Scientific Publishers, Inc.; pp. 125-137. (Advances in modern
environmental toxicology: v. 5).
Linn, W. S.; Avol, E. L.; Shamoo, D. A.; Spier, C. E.; Valencia, L. M.; Venet, T. G.; Fischer, D. A.;
Hackney, J. D. (1986) A dose-response study of healthy, heavily exercising men exposed to ozone at
concentrations near the ambient air quality standard. Toxicol. Ind. Health 2: 99-112.
Linn, W. S.; Avol, E. L.; Shamoo, D. A.; Peng, R.-C.; Valencia, L. M.; Little, D. E.; Hackney, J. D.
(1988) Repeated laboratory ozone exposures of volunteer Los Angeles residents: an apparent seasonal
variation in. response. Toxicol. Ind. Health 4: 505-520.
Lioy, P. J.; Dyba, R. V. (1989) The dynamics of human exposure to tropospheric ozone. In: Schneider, T.;
Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy
implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The
Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 711-721. (Studies in
environmental science 35).
Lioy, P. J.; Vollmuth, T. A.; Lippmann, M. (1985) Persistence of peak flow decrement in children following
ozone exposures exceeding the national ambient air quality standard. J. Air Pollut. Control Assoc.
35: 1068-1071.
Lippniaim, M.; Lioy, P. J. (1985) Critical issues in air pollution epidemiology. Environ. Health Perspect
62: 243-258.
Lippmann, M.; Lioy, P. J.; Leikauf, G.; Green, K. B.; Baxter, D.; Morandi, M.; Pasternack, B. S.; Fife, D.;
Speizer, F. E. (1983) Effects of ozone on the pulmonary function of children. In: Lee, S. D.; Mustafa,
M. G.; Mehlman, M. A., eds. International symposium on the biomedical effects of ozone and related
photochemical oxidants; March 1982; Pinehurst, NC. Princeton, NJ: Princeton Scientific Publishers,
Inc.; pp. 423-446. (Advances in modem environmental toxicology: v. 5).
Mage, D. T.; Raizenne, M.; Spengler, J. (1985) The assessment of individual human exposures to ozone in a
health study. In: Lee, S. D., ed. Evaluation of the scientific basis for ozone/oxidants standards:
transactions of an APCA international specialty conference; November 1984; Houston, TX. Pittsburgh,
PA: Air Pollution Control Association; pp. 238-249. (APCA international specialty conference
transactions: TR-4).
McDonnell, W. F.; Horstman, D. H.; Hazucha, M. J.; Seal, E., Jr.; Haak, E. D.; Salaam, S. A.; House,
D. E. (1983) Pulmonary effects of ozone exposure during exercise: dose-response characteristics.
J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 54: 1345-1352.
McDonnell, W. F.; Chapman, R. S.; Horstman, D. H.; Leigh, M. W.; Abdul-Salaam, S. (1985a)
A comparison of the responses of children and adults to acute ozone exposure. In: Lee, S. D., ed.
Evaluation of the scientific basis for ozone/oxidants standards: transactions of an APCA international
specialty conference; November 1984; Houston, TX. Pittsburgh, PA: Air Pollution Control
Association; pp. 317-328. (APCA international specialty conference transactions: TR-4).
3-71
-------
McDonnell, W. F., IE; Chapman, R. S.; Leigh, M. W.; Strope, G. L.; Collier, A. M. (1985b) Respiratory
responses of vigorously exercising children to 0.12 ppm ozone exposure. Am. Rev. Respir. Dis.
132: 875-879.
McDonnell, W. F.; Horstman, D. H.; Abdul-Salaam, S.; Raggio, L. J.; Green, J. A. (1987) The respiratory
responses of subjects with allergic rhinitis to ozone exposure and their relationship to nonspecific
airway reactivity. Toxicol. Lnd. Health 3: 507-517.
Miller, F. J.; Overton, J. H. (1989) Critical issues in intra- and interspecies dosimetry of ozone. In:
Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its
policy implications: proceedings of the 3rd U.S.-Dutch international symposium; May 1988; Nijmegen,
The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers: pp. 281-291. (Studies in
environmental science 35).
Miller, F. J.; Overton, J. H.; Smolko, E. D.; Graham, R. C.; Menzel, D. B. (1987a) Hazard assessment using
an integrated physiologically-based dosimetry modeling approach: ozone. Research Triangle Park, NC:
U.S. Environmental Protection Agency, Health Effects Research Laboratory; EPA report no.
EPA-600/D-87/040. Available from: NTIS, Springfield, VA; PB87-147096.
Miller, F. J.; Overton, J. H.; Gerrity, T. R.; Graham, R. C. (1987b) Interspecies dosimetry of reactive gases.
Research Triangle Park, NC: U.S. Environmental Protection Agency, Health Effects Research
Laboratory; EPA report no. EPA-600/D-87/105. Available from: NTIS, Springfield, VA;
PB87-175824.
Moffatt, R. K.; Hyde, D. M.; Plopper, C. G.; Tyler, W. S.; Putney, L. F. (1987) Ozone-induced adaptive and
reactive cellular changes in respiratory bronchioles of Bonnet monkeys. Exp. Lung Res. 12: 57-74.
Overton, J. H.; Graham, R. C.; Miller, F. J. (1987) A model of the regional uptake of gaseous pollutants in
the lung: II. the sensitivity of ozone uptake in laboratory animal lungs to anatomical and ventilatory
parameters. Toxicol. Appl. Pharmacol. 88: 418-432.
Ozkaynak, H.; Burbank, B.; Garsd, A.; Spengler, J. D. (1985) Statistical analyses of the Lake Couchiching
health and aerometric data: phase II report. Ottawa, ON, Canada: Department of National Health and
Welfare, Health Protection Branch; contract no. 1209.
Pickrell, J. A.; Hahn, F. F.; Rebar, A. H.; Horoda, R. A.; Henderson, R. F. (1987) Changes in collagen
metabolism and proteinolysis after repeated inhalation exposure to ozone. Exp. Mol. Pathol.
46: 159-167.
Raizenne, M. E.; Spengler, J. D. (1989) Dosimetric model of acute health effects of ozone and acid aerosols in
children. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone
research and its policy implications: proceedings of the 3rd U.S.-Dutch international symposium; May
1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers:
pp. 319-329. (Studies in environmental science 35).
Raizenne, M.; Stern, B.; Spengler, J. (1987) Acute respiratory function and transported air pollutants:
observational studies. Presented at: 80th annual meeting of the Air Pollution Control Association; June;
New York, NY. Pittsburgh, PA: Air Pollution Control Association; paper no. 87-32.6.
Raizenne, M. E.; Burnett, R. T.; Stern, B.; Franklin, C. A.; Spengler, J. D. (1989) Acute lung function
responses to ambient acid aerosol exposures in children. Environ. Health Perspect. 79: 179-185.
3-72
-------
Rao, G. A.; Larkin, E. C.; Harkema, J. R.; Dungworth, D. L. (1985a) Changes ia the levels of
polyunsaturated fatty acids in the lung and lecithin cholesterol acyl transferase activity in plasma of
monkeys exposed to ambient levels of ozone. Toxicol. Lett. 24: 125-129.
Rao, G. A.; Larkin, E. C.; Harkema, J. R.; Dungworth, D. L. (1985b) Changes inlipids of lung lavage in
monkeys after chronic exposure to ambient levels of ozone. Toxicol. Lett. 29: 207-214.
Reisenauer, C. S.; Koenig, J. Q.; McManus, M. S.; Smith, M. S.; Kusic, G.; Pierson, W. E. (1988)
Pulmonary response to ozone exposures in healthy individuals aged 55 years or greater. JAPCA
38: 51-55.
Reiser, K. M.; Tyler, W. S.; Hennessy, S. M.; Dominguez, J. J.; Last, J. A. (1987) Long-term consequences
of exposure to ozone: II. structural alterations in lung collagen of monkeys. Toxicol. Appl. Pharmacol.
89: 314-322.
Richards, W.; Azen, S. P.; Weiss, J.; Stocking, S.; Church, J. (1981) Los Angeles air pollution and asthma in
children. Ann. Allergy 47: 348-354.
Rombout, P. J. A.; Van Bree, L.; Heisterkamp, S. H.; Marra, M. (1989) The need for an eight hour ozone
standard. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone
research and its policy implications: proceedings of the 3rd US-Dutch international symposium; May
1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers;
pp. 701-710. (Studies in environmental science 35).
Santrock, J.; Hatch, G. E.; Slade, R.; Hayes, J. M. (1989) Incorporation and disappearance of oxygen-18 in
lung tissue from mice allowed to breathe 1 ppm 18O31. Toxicol. Appl. Pharmacol. 98: 75-80.
Schelegle, E. S.; Adams, W. C. (1986) Reduced exercise time in competitive simulations consequent to low
level ozone exposure. Med. Sci. Sports Exercise 18: 408-414.
Schelegle, E. S.; Adams, W. C.; Siefkin, A. D. (1987) Indomethacin pretreatment reduces ozone-induced
pulmonary function decrements in human subjects. Am. Rev. Respir. Dis. 136: 1350-1354.
Schwartz, J.; Hasselblad, V.; Pitcher, H. (1988) Air pollution and morbidity: a further analysis of the
Los Angeles student nurses data. JAPCA 38: 158-162.
Seltzer, J.; Bigby, B. G.; Stulbarg, M.; Holtzman, M. J.; Nadel, J. A.; Ueki, I. F.; Leikauf, G. D.; Goetzl,
E. J.; Boushey, H. A. (1986) O3-induced change in bronchial reactivity to methacholine and airway
inflammation in humans. J. Appl. Physiol. 60: 1321-1326.
Sherwin, R. P.; Richters, V. (1985) Effect of 0.3 ppm ozone exposure on type II cells and alveolar walls of
newborn mice: an image-analysis quantitation. J. Toxicol. Environ. Health 16: 535-546.
Slade, R.; Stead, A. G.; Graham, J. A.; Hatch, G. E. (1985) Comparison of lung antioxidant levels in humans
and laboratory animals. Am. Rev. Respir. Dis. 131: 742-746.
Spektor, D. M.; Lippmann, M.; Lioy, P. J.; Thurston, G. D.; Citak, K.; James, D. J.; Bock, N.; Speizer,
F. E.; Hayes, C. (1988a) Effects of ambient ozone on respiratory function in active, normal children.
Am. Rev. Respir. Dis. 137: 313-320.
Spektor, D. M.; Lippmann, M.; Thurston, G. D.; Lioy, P. J.; Stecko, J.; O'Connor, G.; Garshick, E.;
Speizer, F. E.; Hayes, C. (1988b) Effects of ambient ozone on respiratory function in healthy adults
exercising outdoors. Am. Rev. Respir. Dis. 138: 821-828.
3-73
-------
Spengler, J. D.; Garsd, A.; Ozkaynabe, H. (1985) Statistical analysis of the Lake Couchiching health and
aerometeric data. Phase I report. Ottawa, ON, Canada: Department of National Health and Welfare.
Tyler, W. S.; Tyler, N. K.; Last, J. A.; Gillespie, M. J.; Barstow, T. J. (1988) Comparison of daily and
seasonal exposures of young monkeys to ozone. Toxicology 50: 131-144.
U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report nos. EPA-600/8-84-020aF-ef. Available from: NTIS, Springfield,
VA; PB87-142949.
VanBree, L.; Rombout, P. J. A.; Rietjens, I. M. C. M.; Dormans, J. A. M. A.; Marra, M. (1989)
Pathobiochemical effects in rat lung related to episodic ozone exposure. In: Schneider, T.; Lee, S. D.;
Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications:
proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands.
Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 723-732. (Studies in environmental
science 35).
Vedal, S.; Schenker, M. B.; Munoz, A.; Sarnet, J. M.; Batterman, S.; Speizer, F. E. (1987) Daily air
pollution effects on children's respiratory symptoms and peak expiratory flow. Am. J. Public Health
77: 694-698.
Ware, J. H.; Ferris, B. G., Jr.; Dockery, D. W.; Spengler, J. D.; Stram, D. O.; Speizer, F. E. (1986) Effects
of ambient sulfur oxides and suspended particles on respiratory health of preadolescent children. Am.
Rev. Respir. Dis. 133: 834-842.
Wiester, M. J.; Williams, T. B.; King, M. E.; Menache, M. G.; Miller, F. J. (1987) Ozone uptake in awake
Sprague-Dawley rats. Toxicol. Appl. Pharmacol. 89: 429-437.
Wiester, M. J,; Tepper, J. S.; King, M. E.; Menache, M. G.; Costa, D. L. (1988) Comparative study of ozone
(O3) uptake in three strains of rats and in the guinea pig. Toxicol. Appl. Pharmacol. 96: 140-146.
Wright, E. S.; Kehrer, J. P.; White, D. M.; Smiler, K. L. (1988) Effects of chronic exposure to ozone on
collagen in rat lung. Toxicol. Appl. Pharmacol. 92: 445-452.
3-74
*U.S. GOVERNMENT PRINTING OFflCE: 19 ea -6*8 -003/60013
-------
-------
m
T3
5:
o>
o
?
O3
§
O
en
%
~u O
— CD ^*
O rj ~
o o) 2.
•Jf —
s- ro
o c
~" en
•05"
~ (D
o o m c
a' § < i--
Ill's.
CO
CD
33 >
(D CQ
W CD
CD 3
0) O
I"
o
"D
O
T3 W
m H
30 > DO
2 OC
— _mr-
p •
9
CO
01
------- |