Mj. Grainge et al., Reproductive, menstrual and menopausal factors: Which are associated with bone mineral density in early postmenopausal women?, OSTEOPOR IN, 12(9), 2001, pp. 777-787
The associations between a number of reproductive and menopausal factors an
d bone mineral density (BMD) were studied in a sample of early postmenopaus
al women. The study included 580 women aged 45-61 years who completed a ris
k factor questionnaire containing sections on obstetric and menstrual histo
ry. BMD measurements were taken at the anteroposterior (AP) spine, greater
trochanter, femoral neck, total radius and whole body, along with whole bod
y bone mineral content (BMC). In analyses adjusting for key confounders, nu
mber of pregnancies was more strongly associated with increased BMD than nu
mber of live births at all sites (p <0.05 at femoral neck and total radius)
, and menstrual years was more strongly associated with increased BMD than
years since menopause (p <0.05 at all sites). Hysterectomized women had a s
ignificantly higher adjusted mean BMD than non-hysterectomized women at all
sites (AP spine: 0.999 g/cm(2) vs 0.941 g/cm(2), p <0.001), although there
were no significant differences in BMD between hysterectomized women who h
ad a bilateral oophorectomy and those whose ovaries were preserved. Negativ
e associations between the duration of hot flushes and BMD were statistical
ly significant (p <0.05) at the three non-hip sites. In multiple regression
analyses containing all reproductive terms, duration of hormone replacemen
t therapy (HRT) use, menstrual years and hysterectomy status were significa
ntly associated with BMD at all five sites, whilst oral contraceptive use b
efore the age of 23 years was significantly associated with increased BMD a
t all sites except the total radius. Breastfeeding duration, the duration o
f oral contraceptive use and premenopausal amenorrhea were found to have no
association with BMD. Results for whole body BMC were consistent with thos
e for the five BMD sites, across all the variables considered here. These f
indings confirm the importance of HRT use and duration of menses as predict
ors of BMD, whilst the results for hysterectomy status and early oral contr
aceptive use require further consideration.