Abstract
The aim of this study was to investigate arcadian and ultradian variations in menopausal hot flash. The number of hot flashes per 2-hr period was collected from 25 diurnally-active, perimenopausal women for 1 week in January or February of each year for 3 consecutive years. Fourteen women were experiencing natural menopause (NM) (mean age 51.9 years) and 11 were experiencing surgically-induced menopause (SIM) (mean age 52.0 years). The difference in the number of hot flashes between the two types of menopause at each clock time was not statistically significant; neither was the mean number of hot flashes per 24 hr different between the two groups (Student's f-test). Data when normalized for each woman and placed end-to-end revealed by cosinor analysis circadian rhythmicity in the SIM group (P =0.02) but not in the NM group. A 12-hr periodicity was detected in both groups (P< 0.001 for both). An 8-hr rhythm was detected only for the NM group (P = 0.04). Both groups combined exhibited statistically significant rhythmicities with periods of 24 hr (p= 0.003), 12 hr (P<0.001) and 8 hr (P= 0.005). Regardless of the type of menopause, the women could be separated into two groups based on the temporal pattern of hot flashes during the day. One group was defined by the occurrence of peak frequency of flashes during the morning (0400–0959), while the second group was defined by the occurrence of the peak in the evening (1600–2159). The difference between the two groups in the number of hot flashes during the morning wasstatistically significant (Student's r-test, P = 0.009) as was the number of hot flashes in the evening (Student's r-test, P= 0.03).