United States
 Environmental Protection
 Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
 Research and Development
EPA/600/S1-85/005  Mar. 1985
 Project Summary
 Experiments  in  Microwave
 Exposure  in the  Rat:
 Body Temperature, Serum
 Chemistry,  and  the Use  of
 Chemical  Restraint
 Ezra Berman and Hershell Carter
  A series of experiments was conduct-
 ed in rats to observe their responses to
 2450-MHz (CW) microwaves. Colonic
 temperatures were measured after ex-
 posure to 0, 20, or 30 mW/cm2 for up
 to 4 h. Exposures of 30 mW/cm2 caused
 a regular and  significant increase in
 colonic  temperature. A  plateau of
 38.9°C  was reached in 90 min and
 maintained for the rest of the 4-h
 exposure. Rats exposed to 20 mW/cm2
 for 4 h  exhibited patterns of colonic
. temperature similar to sham-exposed
 rats. Acclimation for 90 min had no
 effect on these patterns. Measurements
 (or calculations) of 26 serum chemistry
 values produced similar results in rats
 exposed to 30 mW/cm2 for 120 min
 and rats exposed  to 0 or 20 mW/cm2
 for 120 min. Only serum corticosterone
 was significantly increased in the dose-
 related response of naive rats to this
 acute exposure.
  Various injectable anesthetics were
 evaluated as restraints  in microwave
 exposure experiments. Colonic temper-
 ature was a measure of the response to
 exposure. I mmobilization was accompa-
 nied by decreased colonic temperature.
 After exposure to 30 mW/cm2 for 90
 min,  mean  colonic temperatures in
 chemically restrained rats ranged from
 39-40.4°C, and increased up to 5.5°C
 from anesthetically depressed pre-ex-
 posure levels. Limitations on the use of
 chemical restraint for this kind of
 experiment and on the use of  multi-
value serum chemistry screening tests
were determined during this research.
  This Project Summary was developed
by EPA's Health Effects Research Labo-
ratory, Research Triangle Park, NC, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).


Introduction
  Extrapolating from the results of micro-
wave-exposure studies to human health
effects depends mainly upon the develop-
ment of a base of knowledge from labo-
ratory animal experimentation. Because
such experiments are conducted under a
great variety of conditions, attempts at
replication in other laboratories often fail.
Variation in microwave-exposure condi-
tions may account for some of  these
failures, but experimental techniques
may also vary among laboratories in ways
less obvious to the researcher. Apparent-
ly, even minor variations in the tech-
niques of handling and restraint can
cause untoward reactions and results in
laboratory animals.
  The physiologic consequences of ex-
posure to microwave irradiation have
been measured by many methods. Per-
haps the most common standard of
assessment of an effective level of ex-
posure is the measurement of body or
colonic temperature. Temperature is usu-

-------
ally taken just before and just after an
exposure. Measurements taken during
an exposure have had limited interpret-
ability because of errors  introduced by
metallic temperature sensors. However,
reliance on post-exposure body temper-
ature values  is not justified without
measurements of temperature variations
encountered during the exposure itself.
  The rat was chosen as the experimental
subject for these studies because it is one
of the most commonly exployed whole
animal models  in the microwave bioef-
fects field as well as in.this laboratory. For
instance, in a review of the results from
17 reports of microwave teratogenesis in
animals,  about  half  of the studies  were
conducted  in  rats.  Another important
rationale for  the  use  of  rats in these
studies  is that the rat  may  randomly
orient itself in  a  2450-MHz frequency
microwave field without the whole-body
specific absorption rate (SAR) being af-
fected, whereas, in mice, the SAR varies
according to orientation
  The full report presents data from rats
exposed  to microwave radiation  under
"typical" experimental conditions. These
results are intended for use as reference
or baseline data.

Discussion

Body Temperature
  Experiments 1 and 2 measured  body
temperatures in the rat after exposures to
varying power densities and durations up
to  4 h. Experiment  1  compared power
densities of 0, 20, and 30 mW/cm2, the
latter being a common high level of power
density used in rats in this laboratory. The
curve of  mean body temperatures  after
increasing exposure durations showed a
rapid (15 min) response to this high power
density.  Within 60 mm, the mean  tem-
perature  had  risen to essentially 39°C,
and remained there during the longer
exposures for up to 4 h. Extended expo-
sure to a power density of 30 mW/cm2
produced a response within  the thermal
physiologic capacity of the  rat, as evi-
denced by a platea u of body temperature.
How long the plateau could be maintained
in the face of continued absorption of this
energy (SAR = 6.6 mW/g) is not known
and is beyond the scope of the full report.
  On the other hand, a power density of
20  mW/cm2  applied under identical
conditions had virtually no effect on body
temperature. The curve of body temper-
ature at  various exposure durations at
this power density (SAR  = 4.4  mW/g)
approximates that of sham exposures, ex-
cept for the initial period of  up to 1 h, in
which  lower  body temperatures were
registered for shams.  Apparently, 20
mW/cm2 did not constitute a significant
absorption of  energy by these rats,  and
was conveniently handled by  these
animals.
  The rats in all these experiments were
naive to the procedures used There are
limitations on the  interpretation  of re-
sponses in rodents in naive situations. To
help avoid these obstacles, some re-
searchers use a period of acclimation to
the experimental techniques on the basis
that  a stress  response to a new exper-
imental situation can generate a consid-
erable amount of physiologic "noise." In
turn, it  is possible to misinterpret the
"noise" as being the rats' response to
microwave exposure and, thus, a more
subtle response may be entirely masked.
To some degree, acclimation was provid-
ed in Experiment  2  Naive rats were
acclimated to the  exposure conditions
(handled, caged in  the constramer,  and
exposed in the environmental chambers
at 0 mW/cm2) for 90 mm Body temper-
atures  were  not  measured, but were
expected to follow a decreasing curve like
that of rats sham-irradiated in Experiment
1. What seems like a dramatic decrease
in the sham-irradiated rats' temperatures
results  from  the  contrast with their
initially high mean temperatures (approxi-
mately  38.2°C; pre-exposure value).
Those high temperatures were caused by
the excitement from being awakened in
the maintenance cage, and handled and
transported to the exposure facility. With-
in 60  min, the mean  temperature of
sham-irradiated rats fell below37°C, and
the  animals  were quiet and usually
dozing.
  Rats that were acclimated for 90 mm,
then microwave-irradiated for 90 min at
30 mW/cm2 (Experiment 2) had essential-
ly the same body temperatures as those
irradiated with 30 mW/cm2 with no
acclimation. Acclimation  of 90 min did
not appear to influence the final body
temperature reached by  rats exposed to
30  mW/cm2.  Even when  acclimation
reduced the temperature to <37°C and at
least 1.0°C below the initial pre-exposed
value, the final temperature still reached
levels seen in non-acclimated rats. There-
fore, the temperature at the beginning of
the exposure  did  not appear  to  be  a
significant influence on the final temper-
ature after exposure to microwave radia-
tion.
Anesthetics
  The  measurements  made in Experi-
ments 1 and 2 were taken m animals that
had normal  thermoregulation.  The rats
used in Experiment 3 were administered
six anesthetics, all of which altered phys-
iologic thermal control  Used commonly
in veterinary  and  laboratory  animals
procedures,  these chemical restraints
were  explored for their  suitability for
microwave experiments (immobilization
for 1-2 h, without effect on physiologic
thermal mechanisms).  The compounds
used were anileridme,  a synthetic nar-
cotic,  chloral hydrate and pentobarbital,
both CNS anesthetics; promazine, atran-
quilizer; and ketamine and xylazme, both
dissociative drugs.
  We administered  the  chemicals at
doses suggested in the  literature on
laboratory animals or veterinary medi-
cine. These doses provided starting points
from which  the  lowest dose producing
practical immobilization was determined.
For example, the recommended anesthet-
ic dose of pentobarbital in the rat is 25-40
mg/Kg i.p. or  50 mg/Kg  i.p. We found
that  35  mg/kg  i.p.  was  an  effective
anesthetic dose. The effective  tranquil-
izmg dose for promazine in the rat is given
as 1 -20 mg/Kg i.m. The deepest immobil-
ization was obtained from 40 mg/Kg i.m
Recommended doses of  anileridine in
animals were  not available, as  the drug
was not recommended for veterinary use.
(This  drug is  no longer  manufactured
commercially (The doses of drugs used to
obtain the greatest immobilization and
least response to touch were1 anileridine,
10  mg/Kg  s  c.,  chloral  hydrate,  300
mg/Kg i.p;  ketamine, 160 mg/Kg  i.m.;
pentobarbital, 50 mg/Kg i.p.; promazine,
40 mg/Kg i m ; and xylazme, 0.75 mg/Kg
i.p.
  Anileridine caused little loss  of wake-
fulness and produced little immobiliza-
tion  The  response to  touch was still
present during the measurement of tem-
perature, but quiet handling was  possible.
Colonic  temperatures  during  the  test
period  remained  as  high  as  in active
animals, and we  suspect that some mild
hyperthermia had resulted from the dose.
  Chloral  hydrate produced  profound
sleep  and immobilization at 300 mg/Kg
and depressed the response to touch for
more than 90 mm after injection Colonic
temperature means  were steadily de-
pressed by this drug, falling below 36°C
near the end  of the  effective  period,  a
decrease of  1.5°C. All anesthetized rats
were  observed  while  in  open-topped
plastic cages shaped like shoe boxes with
pine-shaving bedding, without this insul-
ation, temperatures might have decreased
even further than observed.

-------
  Ketamine depressed effectively the
response to touch, but caused almost no
immobilization. The  body temperature
depression was,  like  chloral hydrate,
decreased approximately 1.5°C. The de-
pression of temperature with no effective
immobilization  suggests that  ketamine
alone is not a  useful drug for restraint
during microwave exposure.
  Pentobarbital is very commonly used as
an  animal anesthetic  and  has been
thoroughly researched.  Rats responded
with deep  and constant immobilization
and with a lack of response to touch for at
least 90 min after a single i.p. injection of
35  mg/Kg. During  this  period, body
temperatures were down to  35.5°C,  a
loss of 2-2.5°C. All  three  measures
(response,  sleep, temperature) were af-
fected for almost 3 h. The use of male rats
and the bedding may have offset each
other in this experiment;  male rats are
reported to sleep longer than females,
and pine-shaving bedding is reported to
reduce the drug-induced sleep time
  Promazine (chlorpromazine;  Sparine)
was relatively  ineffective in  reducing
touch-response. There was a long delay
(>  60  min) before any depth of sleep
occurred, and the sleep was erratic. The
drug was effective (within its small effec-
tive dose range) for the entire 3-h period.
Mean body temperatures by  that time
were almost 2°C below initial tempera-
ture Promazine's  delayed and  erratic
effect suggests that this drug is not useful
as a restraint for microwave exposures.
  Xylazine  was ineffective.  No sleep
(immobilization) occurred, and response
to touch was normal. Even so, tempera-
tures at 90 min after  injection were 2°C
below initial temperatures
  When rats were given one of the four
chosen  chemicals, and  then  irradiated
with 2450-MHz microwaves at a  power
density of 30 mW/cm2 for 90 min, rectal
temperatures showed a  sharp increase.
Mean temperatures at the end of expo-
sure were 39 to < 40°C. These values
were  as much as 5.5°C above those
encountered with the drug but without
microwave exposure.  Because body tem-
peratures also exceeded those of  rats in
similar  microwave-exposure situations
(Experiment 1),  we suspect that all the
tested  drugs  interfered with normal
thermal regulation. Therefore, we do not
recommend the use  of  these drugs as
restraining agents in microwave exposure
when body temperatures are used to
determine physiologic response, unless
the compromised thermal  regulatory
functions are taken into account.
Serum Chemistries
  Chemical analyses of serum constitu-
ents  also  are common experimental
techniques for the assessment of physio-
logic response to microwave exposure in
laboratory  animals. The  use of  serum
chemistries in this  respect  may  well
increase because  of the economy and
convenience associated with modern
automated analytic techniques. For ex-
ample, the SMAC-24 tests used in Experi-
ment 4 reported 24 replicated values and
ratios  of 21 constituents on one serum
sample of less than  1-ml volume. Only
one additional milliliter was required to
do the serum Ta and T4 and corticosterone
tests  using radio-immune  assay tech-
niques. The cost of this battery of tests is
insignificant when compared to the cost
incurred by manual analyses used just a
few years ago.
  Because  these  tests are convenient
and relatively economical, one can lose
sight of their limited applicability. Data
values are  often subjected  to  no more
evaluation than statistical tests of con-
fidence (t-test; analysis of variance). A
warning of the hidden problems associ-
ated with indiscriminate use of  serum
chemistry values has suggested consid-
eration of age, sex, strain, environment,
and physiologic status (diurnal  cycle,
postural stance, age) of the animal when
interpreting values derived from blood. It
has also been stressed that in human
medicine, in  which  these  batteries of
tests find their most effective use and for
which they are designed, the test results
should be confined to decisions of diag-
nosis, treatment, and prognosis, and that
the critical aspect is interpretation. Also,
any interpretation should  be strongly
limited when there are no baseline values
derived from the same population that
can also be extrapolated to other popula-
tions.
  More important,  perhaps, is the sug-
gestion that a Bayesian approach should
be used in interpreting these diagnostic
tests.   Paraphrased for toxicologic  re-
search, e.g., microwave-exposure studies,
the requirements might be. (a) clear-cut
definition of the disease (effect) being
sought; (b) delimitation of the stage of the
effect being evaluated by the test;  and (c)
knowledge  of true-positive and  true-
negative results of the test. More  simply
stated, one should have a firm prior
hypothesis.
  Statistically, the  determination of 21,
24, or more values at once in one subject
is  expected to very  much  reduce the
independence of these values. Appropri-
ate adjustments should be made when
applying ordinary statistical tests. In our
research, the method we used extended
acceptable differences to p <.01. Another
related aspect of statistical methodology
is that serum chemistry values may not
assume normal distributions, and statis-
tical tests that require normality (t-test)
may not  be  applicable  here. Bilirubin
values, for instance, were expressed only
as 0,0.5,1.0, or 1.5; chi-square statistics
are more appropriate than t-test statistics
in  the cases of creatinine and  bilirubin
values.
  Only three serum constituents (alkaline
phosphatase, corticosterone,  and creati-
nine), exceeded the limitations of the
analytic  criteria (p < .01). However,
further evaluative tests should be put to
these values. It is  reasonable to expect
the values to follow some  consistent
direction or trend. The application of
microwaves should cause an increasing
or  decreasing trend in the  results.
Biphasic  (up-then-down, or down-then-
up) trends are  not  to be  reasonably
expected from increasing exposure. How-
ever,  alkaline phosphatase increased
from 250 mg/dl serum in the 0 mW/cm2
group to 315 in the 20 mW/cm2 group,
and then decreased in the 30 mW/cm2
group. Such an up-then-down pattern is
unexpected  and more than  difficult to
rationalize as resulting only from expo-
sure  to  microwaves. Serum  alkaline
phosphatase  increases in response to
extrahepatic blockage or to bone disease.
As  these values  did not continue to
increase with the increased power den-
sity, the rationale of microwave exposure
as the cause of the observed  changes in
serum alkaline phosphatase does not
bear scrutiny.
  Serum  creatinine values, on the other
hand, did show a trend in one direction.
They showed a sharp increase in animals
exposed at 20 mW/cm2 and 30 mW/cm2
as compared to sham animals(p = .0001).
Creatinine is elevated when renal blood
flow is reduced; however, an increase,
not a decrease, of renal blood flow would
be expected along  with raised tempera-
tures  due to microwave exposure (30
mW/cm2).  Therefore,  no  physiologic
rationale is available to  explain the in-
crease in serum creatinine. We  suspect
that we  observed a sampling error,
especially in the rather low values of 0.3-
0.4 mg/dl in the sham group. The values
are well  within normal limits for  rats
when the  values  of all animals are
averaged (0.47 mg/dl);  the  average  is
very close to the  mean of normal  rats
(0.45  mg/dl).

-------
    A corticosterone  increase due  to
   microwave irradiation is the only one of
   the statistically different serum constitu-
   ents which may lend itself to a reasonable
   physiologic rationale. Adrenal corticoid
   secretion  is stimulated as  a  result  of
   handling of animals in  naive situations.
   As the exposure to 30 mW/cm2 was
   sufficient  to increase colonic  tempera-
   tures to almost 39°C, it may also have
   caused an increase in adrenal corticoster-
   oid hormone secretion. The additional
   response to 30 mW/cm2 exposure, as a
   stress reaction, may account for the more
   than doubling of corticosterone concen-
   tration in  the high dose group as com-
   pared  to  shams  (260  vs. 116 Aig/ml,
   respectively).

   Conclusions
    When naive young adult male rats are
   exposed to 2450-MHz microwaves at a
   power density of 30 mW/cm2 with  a
   duration  of up  to 4   h,  their colonic
   temperatures rise to 39°C. Twenty
   mW/cm2 causes no obvious temperature
   increase in rats. However, acclimation to
   the same conditions, without microwaves,
   does not alter the pattern of  increased
   temperature. Except for a significant rise
   in corticosterone concentration, serum
   chemistry values in animals exposed to
   30 mW/cm2 are  not acutely altered.
    Anesthetics are not recommended for
   restraint in animals during microwave
   exposure  unless it is  appreciated how
   they  limit normal thermal physiologic
   mechanisms.
          The EPA authors Ezra Rerman and Hershell Carter are with the Health effects
           Research Laboratory. Research Triangle Park, NC 27711.
          The complete report,  entitled "Experiments in Microwave Exposure in the Rat:
           Body Temperature, Serum Chemistry, and the Use of Chemical Restraint,"
           (Order No. PB 85-156 834/AS; Cost: $8.50, subject to change) will be available
           only from:
                 National Technical Information Service
                 5285 Port Royal Road
                 Springfield, VA 22161
                  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
                                            US GOVERNMENT PRINTING OFFICE 1985-559-016/27003
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
                                     OC00329    PS
                                     U  S  ENVIR  PROTECTION  HGINCY
                                     REGION  5  LIBRARY
                                     230  S  DEARBORN  STREfT
                                     CHICAGO                IL

-------