Wearing a special thermometric brassiere, selected women self-measured
their breast surface temperature. These measurements were made during
one hour each evening at home for one menstrual cycle under standard
conditions of overclothing and room temperature. To stage their cycle
they also collected daily samples of saliva in their freezer for immun
e-assay of progesterone concentration in the laboratory A total of 82
women participated, most having young families. This total included fo
ur groups, a control group (N = 25) and three 'disease' groups, namely
: family history of breast cancer (14); benign breast disease (12); an
d a 'cancer-associated' group (31) who had had previous cancer surgery
. A significant breast temperature rhythm with a period at or about 28
days was found not only in the controls but also in the three groups
of breasts designated 'disease'. Nevertheless, consistent rhythm abnor
malities were found in all the disease groups. Most evident was a hype
rthermia throughout the cycle, a reduction in the rhythm amplitude, an
d a tendency for the breast temperature rhythm to be manifest 1-2 days
earlier in the menstrual cycle.