of work: As of July 1, 1997, 106 women have been enrolled in the BLSA Perimenopausal Study and have completed their first inpatient assessments and at least one quarterly outpatient GCRC visit. A total of 631 outpatient visits have been completed, ranging from 1 to 16 per woman. A total of 27 inpatient GCRC visits have been completed by 13 women for overnight frequent blood sampling. Of these women, two are classified as perimenopausal (based on measurement of the follicle stimulating hormone, FSH) but they are within two years of their last menses. Analyses of nocturnal GH secretion has been completed for data obtained from ten of the premenopausal women during both early and late phases of one menstrual cycle. There was no difference in GH secretion in between phases of the menstrual cycle. A preliminary analysis of the longitudinal data will be performed when plasma FSH levels are high in at least 6 women. Two more premenopausal women have agreed to be studied within the next three months to complete the study cohort of 15 women. We have also analyzed longitudinal bone mineral density (BMD) measurements at the lumbar spine and hip. The 27 women who remain premenopausal (PRE), were followed > 1 yr (mean 2.5 yrs), and had a mean age of 48.7+/-0.3 yrs. They were compared to 21 women classified as perimenopausal (PERI),with at least one visit > 0.3 yrs before (mean 0.84 yrs) and at least one visit >1 yr after (mean 1.90 yrs) their initial high FSH values. They were followed an average of 2.7 yrs, and had a mean age of 50.3+-0.6 yrs. The PRE group had a significant increase in BMD in the spine while the PERI group had a significant decrease. The BMD values of the two groups were also significantly different . Both the PRE and the PERI groups were losing bone at the femoral neck and trochanteric region. There was no significant difference between PRE and PERI at any hip site.These data confirm the expected accelerated vertebral bone loss during the perimenopausal transition, whereas the similar losses of hip BMD in pre- and perimenopausal women appear not to be distinguishable, suggesting that hip fracture risk may not start to increase until the post-menopausal period. The assays to measure bone biomarkers have been completed for visits up to 2/1/97. Data analyses are underway to determine if serial changes correlate with changes in bone density. Body composition is measured by DEXA and anthropometrics. To date, complete body composition data are available for 21 women who remain premenopausal and 24 who are classified as perimenopausal, as defined above. None of the parameters of body composition measured (weight, body mass index, waist circumference, waist-hip ratio, total fat mass by DEXA scan, or % fat) differed by ANOVA between PRE and PERI groups at baseline. The PERI women exhibit significant longitudinal increases in the estimates of fatness (total fat and % fat by DEXA) compared with the PRE group as well as differences in indices of fat distribution (waist circumference and WHR, p<0.05), and in serum leptin levels (p<0.03), but not in weight or body mass index. These data suggest that changes in body composition may occur early in the menopausal transition.
