TAB D

                   DRAFT   FOR DISCUSSION PURPOSES ONLY   

MEMORANDUM

TO:       Members of the Knoxville Small Panel

FROM:     Advisory Committee Staff

DATE:     May 19, 1995

RE:       Tennessee-Vanderbilt Nutrition Study and Follow-up: 
Research on Pregnant Women



1.  The Original Nutrition Survey (1946-1949)

In 1946, a cooperative study between the Vanderbilt University
and the Tennessee State Health department was initiated to study
nutrition in pregnancy.  The study received annual grants from
the Public Health Service of approximately $9,000 per year,
beginning in November 1946.  (See  Attachments #1 and #2.)  The
study investigators, intending to study only white women,
enrolled all white patients at the Vanderbilt Obstetrics Clinic
who were prior to three weeks before (anticipated) delivery.  In
addition to their routine prenatal care, subjects were asked to
complete dietary surveys, and have additional blood tests
performed to determine vitamin and protein levels.  They were
given test doses of B vitamins, and had to collect a 2 hour urine
sample in order to determine their B vitamin status.  

A portion of the study was devoted to the examination of iron
absorption in pregnancy.  There had been studies of iron
requirements during pregnancy prior to the availability of
radioactive isotopes.  The Food and Nutrition Board of the
National Research Council had set a  Recommended Daily Allowance
of 12 mg for non pregnant women, and 15 mg for women in the
second half of pregnancy and for lactating women.  In the
Tennessee-Vanderbilt study, the women were administered a dose of
radioactive iron, Fe 59, during their second prenatal visit,
prior to receiving a routine dose of therapeutic iron.  On their
third prenatal visit, blood was drawn and tests were performed to
determine the percentage of administered iron that was absorbed
by the mother.  The blood of the infants was examined at birth to
determine the percentage of radioactivity that was absorbed by
the infant.  Approximately 819 women were administered iron, and
466 women completed this portion of the study.  The maternal dose
of radiation delivered varied between 200,000 and 1,000,000
countable counts per minute, and was mixed with between 1.8-120
mg of elemental iron.  This was administered orally.  The fetal
dose was estimated to be 2-3% of the maternal dose of iron at the
time of iron administration.  In a 1963 personal communication,
one of the investigators would estimate the fetal dose to be
between 5-15 rads.

2.   The Follow-Up Study (1963-1964)

In 1963-64, a group of researchers at Vanderbilt University who
were interested in the effects of prenatal exposure to low levels
of ionizing radiation received grants from the Public Health
Service and the Atomic Energy Commission to do a followup study
of the children who were exposed prenatally during the Vanderbilt
iron study.  None of the researchers from the original study were
involved in this investigation.  But Paul Hahn, one of the
original investigators, and the investigator who was responsible
for the iron absorption portion of the study, was responsible for
administering funds for this project from the Public Health
Service.  The followup study investigators were able to locate
the records of 751 of the women who were fed radioactive iron,
and were able to find a group of 771 women who had been involved
in the nutrition study, but did not intake radioactive iron.  The
researchers found errors with the assumptions that were used to
calculate fetal dose in the original study, and incompleteness of
the records that were available.  

The exposed women and the control women were located for
followup, and written questionnaires were administered via mail. 
The recruitment letters and questionnaires mentioned that this
was a followup of the 1946-49 nutrition study, but did not
mention the investigators hypothesis or purpose for the study. 
679 of the exposed mothers and 705 of the non-exposed mothers
participated in the study.  Four cancer deaths were found in the
children of the non-exposed group, and none in the children of
the non-exposed group.  One of the cancer deaths was of a rare
type that was not known to be related to radiation exposure, and
that had a familial pattern.  This case was not included in the
statistical analysis of the data.  According to the
investigators, the expected incidence of all malignancies in the
exposed population , based on the number of years of observation,
and the number of years of followup, would be 0.65 cases.  The
probability of observing three or more cases when one was
expected was calculated to be 3 in 1000.  The investigators
concluded that these data were suggestive of a cause and effect
relationship.

In 1969, one of the researchers  in the followup study, in
conjunction with another investigator, conducted their own
studies using the administration of radioactive iron and
radioactive iodine  to women presenting for therapeutic abortion,
to gain a better estimation of fetal dose at given ages of
gestation.  This study was funded by the Atomic Energy
Commission.  Nine fetuses were studied, and the results were used
to calculate doses received by the three cancer cases from
original Vanderbuilt Study that were among the exposed.  They
found that theradiation doses for these cancer cases
wererelatively  low, and concluded that the relationship between
the radiation exposure and the cancer outcome may not have been
causal.