W
United States
Environmental Protection
Agency
Health Effects
Research Laboratory
Research Triangle Park NC 27711
, / - 44
Research and Development
EPA/600/S1-85/007 Aug. 1985
Project Summary
Teratogenic Effects of
Microwave Radiation
Mary Ellen O'Connor and Robert Strattan
Pregnant CF-1 mice were exposed to
2450-MHz CW microwave irradiation at
power densities of 0,10, or 30 mW/cm2
for 6 hours daily from gestational day 1
through day 18. All exposures occurred
in an anechoic chamber maintained at
50% relative humidity with air tempera-
ture of 22°C. A group of pregnant, cage-
control colony mice was maintained for
each of the three exposure groups. Ter-
atogenic examinations were performed
on day 18 following Caesarean section
of the maternal subject. Every third
fetus was stained with Alazarin red and
examined for skeletal abnormalities.
No significant differences between the
groups were observed for fetal body
mass, fetal brain mass, resorption, live
fetuses, gross abnormalities, or skeletal
abnormalities.
Subsequent experiments were per-
formed on pregnant CF-1 mice at 0, or
30 mW/cm2 at ambient temperatures of
20°C or 25°C. Embryopathic effects at
these levels of irradiation were highly
dependent upon the ambient tempera-
ture in the chamber during the 6 hours
of daily exposure.
Pregnant albino guinea pigs were ex-
posed on days 18 through 25 of gesta-
tion for 60 minutes to 2450-MHz CW mi-
crowave radiation at a power density of
47 mW/cm2. Caesarean sections fol-
lowed by teratogenic examinations
were performed on day 60 of gestation.
Statistical analysis indicated that the
microwave exposed litters had lower
average brain mass and female concep-
tuses had a lower average body mass
than sham-exposed controls.
The full report was submitted in ful-
fillment of contract 68-02-3453 under
the sponsorship of the U.S. Environ-
mental Protection Agency.
This Project Summary was devel-
oped by EPA's Health Effects Research
Laboratory, Research Triangle Park, NC,
to announce key findings of the re-
search project that is fully documented
in a separate report of the same title
(see Project Report ordering informa-
tion at back).
Introduction
At the initiation of this investigation,
the data on the teratogenic effects of
microwave radiation had been obtained
nearly exclusively from a single acute
exposure to 2450-MHz CW radiation at
power densities well above 10 mW/cm2.
Gross teratogenic effects following
such microwave exposure did not ap-
pear. However, one observation that
was reported consistently was reduced
average fetal body mass in microwave
exposed fetuses. Such a functional de-
viation as reduced fetal body mass, al-
though mild in comparison to fetal
death or the appearance of gross struc-
tural abnormalities, is considered as an
indicator of less healthy offspring.
The second general observation re-
garding the teratogenic effects of expo-
sure to microwave radiation was that
even these relatively mild functional de-
viations were observed only at high
levels of exposure that clearly elevated
core temperature of the maternal or-
ganism. Both the rate of temperature in-
crease and the temporal duration over
which the maternal organism must en-
dure this high temperature were re-
ported to be important variables in the
production of teratogenic effects. The
maternal temperature required to in-
duce effects was reported to be 40°C-
42°C which is just below the lethal
threshold for the maternal subjects.
-------
Exposure conditions employed in the
laboratory do not resemble exposure
conditions found in the general environ-
ment. However, the conditions in the
general environment are of the highest
concern with regard to risk of exposure
of the general population. In the general
population, exposures are for long peri-
ods of time (from days to a lifetime) to
what would be considered extremely
low levels in comparison to the intensi-
ties used in the laboratory.
The purpose of the studies was to ob-
tain information with respect to rela-
tively long-term, low-level exposure to
2450-MHz CW. The CF-1 laboratory
mouse was chosen as the subject for
the study to facilitate generalization and
comparison with earlier reports. This
same rationale guided the selection of
2450-MHz CW plane wave radiation.
The full report presents results from ex-
posure of mice at 0, 10.0, and 30.0 mW/
cm2 as well as results from several addi-
tional studies performed at 0 and 30
mW/cm2 at ambient temperatures of
both 20°C and 25°C.
An additional study investigated the
teratogenic potential of microwave radi-
ation in a larger mammal, the guinea
pig. The guinea pig was chosen because
of size and because teratogenic effects
resulting from hyperthermia induced by
non-microwave sources has been re-
ported for the guinea pig. These hyper-
thermia studies provided a basis for the
design of the microwave investigations
on the guinea pig. The post-exposure
temperature of the maternal guinea pig
subject associated with the observation
of deleterious effects from hyperther-
mia was 41°C. The initial microwave
study attempted with the guinea pig re-
sulted in average maternal core temper-
atures just below this 41°C level and did
not produce any observable defects in
the conceptuses. Since the average ma-
ternal temperature did not actually
reach the same level as was reported to
induce effects from non-microwave in-
duced hyperthermia, the study was re-
peated at a higher power density.
To summarize, an investigation was
conducted to further delineate whether
low-level, long-term exposures of CF-1
mice and guinea pigs to 2450-MHz CW
microwave radiation would be embryo-
pathic or teratogenic.
Experimental Procedures
Four separate studies were con-
ducted. All experiments were per-
formed at a frequency of 2450-MHz CW
microwave radiation in an anechoic
chamber with controlled temperature
and humidity.
In the first study the exposure condi-
tions consisted of power densities of 0,
10, or 30 mW/cm2 with concurrent cage
control subjects maintained in the
colony during the exposure period for
each group. The subjects were sperm
positive CF-1 mice bred in the labora-
tory at the University of Tulsa. Each
morning the females were checked for
sperm plugs, and the presence of a
sperm plug was counted as day 1 of the
gestation period. At this time the female
was assigned randomly to either the
current exposure or cage control condi-
tion.
After the identification of a sperm
plug and on each of the subsequent 17
days each female was removed from
the group cage, assessed for body
mass, placed in a plastic container
transported to the anechoic chamber lo-
cated in an adjoining room, and placed
in one of the 18 available positions on
the styrofoam platform in the chamber.
The position of each mouse in the ane-
choic chamber was rotated daily so that
by the end of the 18-day exposure pe-
riod, each mouse had occupied each
available position for one day of expo-
sure. The subjects were exposed for 18
consecutive days for 6 hours beginning
at 0900h and ending at 1500h. Male,
nonbreeder mice of the same strain
were used for fillers if 18 sperm positive
females were not available on a given
day. Each of the exposure conditions
was continued until 25 gravid mice had
been confirmed by Caesarean section.
The ambient temperature in the cham-
ber was maintained at 22°C and the rela-
tive humidity was 62% in this experi-
ment.
Following exposure on day 18, the
maternal subjects were euthanized by
cervical dislocation. Caesarean sections
were performed by making longitudinal
abdominal incisions that exposed the
uterine horns. The number and place-
ment of resorptions and visible fetuses
was noted and the conceptuses were re-
moved. The fetuses were sexed, viabil-
ity was noted, and they were examined
for the presence of any gross structural
abnormality, including cleft palate.
Body mass was determined on a Mettler
PC440 top-loading balance. The fetal
brain was removed, placed in a weigh
boat and weighed. Every third fetus was
placed in KOH for later staining with
Alizarin red to observe skeletal struc-
ture. After staining, the fetuses were
stored in glycerin and examined under a
light microscope for the number of ster-
nal ossifications and number of ribs. At
this time any other observable skeletal
abnormality also was noted.
Experiment 2 was conducted using
CF-1 mice and the same breeding and
exposure procedures as Experiment 1,
with the exception that the anechoic
chamber was maintained at either 20°C
or 25°C and only two power density
levels were compared: 0 and 30
mW/cm2. The conditions were run se-
quentially and colony cage control ani-
mals were maintained for each of the
exposure groups. Caesarean sections,
fetal observations, and staining proce-
dures were performed as for Experi-
ment 1.
For the third experiment, the subjects,
breeding, and exposure procedures
were identical to those for Experiments
1 and 2, with the exception that the ani-
mals were exposed for only 100 min-
utes to either 0 or 30 mW/cm2 with the
temperature of the anechoic chamber
again maintained at 22°C. Colony cage
control animals were maintained for
both exposure periods. Caesarean sec-
tions, fetal observations and staining
procedures were the same as outlined
for the earlier experiments.
The subjects for the guinea pig study '
were Hartley (HA)BR guinea pigs ob-
tained from Charles Rivers Laborato-
ries. Estrus for the females was deter-
mined by examining for vaginal
opening each day until the membrane
covering had receded. The female then
was placed with the male and checked
daily for the presence of a sperm plug.
The day a plug was observed was
scored as day 1 of gestation. At this time
the animal was assigned randomly to
either the colony cage control group or
the current (0 or 47 mW/cm2) exposure
condition.
Once a sperm plug had been detected
the females were left in their home
cages and handled only for routine cage
maintenance through day 17 of gesta-
tion. The exposures occurred on days
18 through 25 of gestation. The animals
were assessed for body mass, placed in
the plastic cylindrical exposure contain-
ers, transported to the anechoic cham-
ber located in the adjoining room, and
exposed for 60 minutes. Four animals
were exposed at one time and non-
breeding males were used as fillers
when necessary. The temperature in the
anechoic chamber was maintained at
22°C and the relative humidity was set .
at 50%. The animals were placed 461
inches below the horn antenna and the
-------
power density at this location later was
determined to be 47 mW/cm2. This loca-
tion was chosen because it resulted in
maternal post-exposure core tempera-
tures of 41 °C after 60 minutes of expo-
sure. The 0 or the 47 mW/cm2 exposures
occurred during the same two-hour pe-
riod each day (1500h to 1700h) and the
two conditions were rotated so that on
one day the 0 mW/cm2 group was ex-
posed during the first hour and on the
subsequent day the 47 mW/cm2 group
was exposed during the first hour. Both
pre- and post-exposure core tempera-
tures were determined with a Bailey
BAT8C thermometer using a RET-3 rec-
tal probe. From day 26 until day 60 of
gestation, the animals remained in their
home cages and were handled only for
routine cage maintenance.
On day 60 the females from all three
groups were sacrificed by inhalation of
chloroform fumes. Caesarean sections
were performed by making a ventral
longitudinal incision that exposed the
left and right horns of the uterus. The
position of resorptions and viable con-
ceptuses was noted prior to their re-
moval from the uterus. All fetuses were
sexed, assessed for body mass, and
crown-to-rump length was measured.
The fetuses were also examined for any
observable structural abnormality, in-
cluding cleft palate. The fetal brain was
then removed, placed in a weigh boat,
and assessed for mass.
Results and Discussion
All data from this investigation were
analyzed using analysis of variance
available in statistical packages from
SPSS and BMD. All analyses, including
those pertaining to microwave expo-
sure parameters, were performed only
on data obtained for females that
proved to be gravid following the Cae-
sarean section. For each experiment,
the initial analysis was a one-way analy-
sis of variance for each endpoint. Dun-
can's multiple range tests were used to
determine differences between individ-
ual groups.
As shown in Table 1, a total of 114
gravid dams were used in the analyses
for the CF-1 mouse exposed at 0, 10, or
30 mW/cm2. After determining that
none of the cage control groups differed
significantly for any of the variables of
interest, a random sample of 27 cage
control litters was used in the one-way
analysis of variance.
No differences in average litter fetal
body mass, average litter fetal brain
mass, number of resorptions, or num-
ber of abnormal fetuses were observed.
When the ambient temperature was
raised to 25°C the fetal body mass of the
microwave exposed litters was reduced
and the number of abnormal fetuses
was increased.
Table 2 presents the data from the
guinea pig exposures and indicates that
only the femal offspring exhibited sig-
nificant differences in body mass. When
individual offspring were analyzed as
separate data points both fetal body
mass and fetal brain mass differed sig-
nificantly, with the microwave exposed
group being the smallest in both cases.
Conclusions and Recommenda-
tions
The studies did not result in observa-
tions of embryopathic or teratogenic ef-
fects in the CF-1 mouse exposed to 0,
10, or 30 mW/cm2 microwave radiation
(2450 MHz) 6 hours daily throughout 18
days of the gestation period. During the
studies the temperature in the anechoic
chamber was maintained at 22°C which
is a common temperature at which mi-
crowave investigations have been
reported. This relatively low ambient
temperature allows the subject to effec-
tively dissipate any increase in core
Table 1. Means and Standard Errors for CF-1 Mice Exposed to 2450-MHz CW Microwave
Radiation for 6 Hours Daily Throughout the Gestational Period
Ambient Fetal Body Fetal Brain
Temperature Treatment Litters Mass Mass
Number Number
Ftesorbed Abnormal
22°C
25°C
20°C
Colony
Control
0 mW/cm2
10 mW/cm2
30 mW/cm2
Colony
Control
0 mW/cm2
30 mW/cm2
Colony
Control
0 mW/cm2
30 mW/cm2
27
26
25
36
28
23
19
16
13
25
1.07
1.03
1.04
1.05
1.12
1.07
0.93
1.05
1.07
1.08
(.01)
(.03)
(.02)
(.02)
(.03)a
(.03)b
(.02)a-b
(.08)
(.04)
(.03)
0.064
0.062
0.063
0.065
.063
.064
.059
.067
.134
.066
(.002)
(.002)
(.001)
(.001)
(.002)
(.001)a
(.002)a
(.005)3
(.046)a'b
(.002)b
1.07
1.35
0.80
0.75
1.00
1.57
1.53
1.94
0.77
1.52
(.23)
(.27)
(.22)
(.21)
(.22)
(.39)
(.42)
(.61)
(.32)
(.27)
0.30
0.62
0.36
0.50
0.14
0.17
0.53
0.25
0.54
0.44
(0. 12)
(.25)
(.13)
(.16)
(.07)a
(.10)
(.21)a
(.14)
(.29)
(.12)
For each experiment, cells within a column sharing a subscript are significantly different
from one another.
Table 2. Means and Standard Errors for Guinea Pigs Exposed to 2450-MHz CW Microwave
Radiation
Number of Litters
Number of Live Fetuses
Fetal Body Mass
Fetal Brain Mass
Female Body Mass
Cage
Control
7
2.86
(0.46)
64.74
(4. 15)a
2.40
(0.07)
66.15
(2.51)a
Sham
Exposed
8
3.13
(0.29)
73.19
(1.25)a
2.52
(0.02)
75.54
(1.94)a'b
Microwave
Exposed
10
2.50
(0.31)
67.36
(1.911
2.21
(0.09)
66.82
(2. 14)b
Totals
25
2.80
(0.20)
68.49
(1.54)
2.36
(0.05)
69.39
(1.55)
Within a row, cells with the same subscripts are significantly different from one another.
3
-------
temperature that may result from mi-
crowave induced heating during expo-
sure. In the present investigation, the
sham animals were in the anechoic
chamber for 6 hours daily for 18 days
rather than a few minutes or hours on a
single day and such a condition could
conceivably induce mild cold stress in
pregnant mice.
In the second series of studies, power
densities of 0 and 30 mW/cm2 were em-
ployed but at ambient temperatures in
the anechoic chamber of 20°C and 25°C.
The ambient temperature and its inter-
action with the microwave power level
was the key to the decreases in fetal
body mass and fetal brain mass. The
results support an essentially thermal
model for accounting for embryopathic
and teratogenic effects following mi-
crowave exposure.
In the study on guinea pigs, the
microwave-exposed female fetuses had
a lower average fetal body mass than
controls. Also, overall fetal body mass
as well as fetal brain mass differed sig-
nificantly across groups if individual
fetal data was used as the endpoint for
analysis rather than the litter mean. No
statistically significant differences in the
incidence of abnormalities were ob-
served, but all of the abnormal fetuses
observed in the entire study were in the
microwave exposure group.
The results of the present investiga-
tion could be presented as differences
in fetal body mass between the mi-
crowave and sham exposure groups if
the statistical analysis excluded the
colony or cage control subjects. Such an
analysis would show a significant re-
duction in fetal body mass for our sham
subjects, which without the colony con-
trol data might be interpreted as an in-
crease in fetal body mass as a result of
microwave exposure. This type of anal-
ysis would disregard the fact that the
microwave exposed subjects had aver-
age fetal body masses that were nearly
identical to what was observed in the
cage control colony animals. Data that
does not contain some reference to
colony cage control baseline informa-
tion is of limited value and must be used
with caution in the assessment of condi-
tions that might present a health risk to
the general population.
Mary Ellen O'Connor and Robert Strattan are with University of Tulsa, Tulsa, OK
74104.
Ezra Herman is the EPA Project Officer (see below).
The complete report, entitled "Teratogenic Effects of Microwave Radiation,"
(Order No. PB85-207'462 /AS; Cost: $8.50, subject to change) will be available
only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S1-85/007
0000329 PS
U S ENVIR PROTECTION JJGENC
REGION 5 LIBRARY
230 S DEftRBCRN STREET
CHICAGO IL 60404
GOVERNMENT PRINTING OFFICE;1985—559-01b/2711
------- |