Adequate levels of reproductive and pituitary hormones are needed for the i
nitiation and maintenance of regular menstrual cycles as well as for the ac
hievement of peak bone mineral density (BMD). Therefore, in the absence of
direct hormone measures, menstrual history may serve as a surrogate for the
adequacy of hormonal functioning and be a marker for bone status in young
women. In our cross-sectional study of white college women aged 19-26 years
, we examined the association of six characteristics of menstrual history w
ith bone density at the lumbar spine and the femoral neck. To characterize
associations, we used multiple linear regression models that also accounted
for the contribution of body mass index, dietary calcium intake, height, l
evel of physical activity, smoking, and alcohol use. The associations betwe
en each of the six menstrual characteristics and BMD were stronger at the l
umbar spine than at the femoral neck. Age at menarche explained the most va
riance at both the lumbar spine (partial r(2) 100 = 5.9%) and the femoral n
eck (partial r(2) 100 = 2.1%). For each year that menarche was delayed, bon
e density was lower by -0.023 g/cm(2) (p = 0.0024) at the lumbar spine and
-0.0129 g/cm(2) (p = 0.0565) at the femoral neck. At the lumbar spine, a hi
gher number of lifetime menstrual cycles was also significantly associated
with increased bone density (adjusted beta = 0.0010, p = 0.0052, partial r(
2).100 = 4.4%). This association was not significant after adjusting for ag
e at menarche. Neither reproductive years (age - age at menarche) nor a his
tory of irregular cycles (either at menarche, in the past year, or ever) wa
s associated with bone density at either site. Menstrual function appears t
o affect the bone density of these young women. Studies that include measur
es of reproductive and pituitary hormones are needed to further explore the
role of hormones in the potential link between menstrual history and bone
density.