Influence of pregnancy and breast-feeding on quantitative ultrasonometry of bone in postmenopausal women

Abstract

Objective: Reproductive factors such as parity and breast-feeding may be associated with low bone mass and osteoporotic fractures in later years. In this study, os calcis quantitative ultrasonometry was used to elucidate the relationship between parity, lactation and bone mass in postmenopausal women.

Design: This was a comparison study using subsequent matched pairs analysis as well as multiple linear regression analysis. The study was carried out at five centers in Germany. The study included 2,080 postmenopausal women (age (mean +/- SD) 58.8 +/- 8.2 years), who were attending for routine check-up and in whom diseases and drug treatments known to affect bone metabolism had been excluded.

Methods and outcome measures: Women underwent quantitative ultrasonometry (QUS) measurement at the heel. Values of the ultrasonometry variables -speed of sound, broadband ultrasound attenuation and stiffness index -were calculated and compared for nulliparous and parous women and for women who had and had not breast-fed. Because of some significant intergroup differences, and to determine any effect of the number of live births and the duration of breast-feeding on ultrasonometry results, second analyses were undertaken using equally sized samples, matched for possible confounding variables such as age and body mass index (matched pairs). In these analyses, nulliparous women were compared with parous women, grouped according to number of live births, and women who had never breast-fed were compared with women who had breast-fed, grouped according to duration of breast-feeding. Furthermore, a multiple linear regression analysis was performed to examine the combined effects of reproductive factors on QUS variables.

Results: No statistically significant associations were found between ultrasonometry variables and parity or breast-feeding, even after controlling for confounding variables in matched-pairs analysis or in a multiple linear regression analysis.