| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Medical Research Council Epidemiology Resource Centre (M.K.J., S.R.C., N.C.H., H.M.I., K.M.G., C.C.), University of Southampton, Southampton General Hospital, Southampton SO16 6YD, United Kingdom; and Department of Medical Physics and Bioengineering (P.T.), Southampton University Hospitals National Health Service Trust, Southampton General Hospital, Southampton SO16 6YD, United Kingdom
Address all correspondence and requests for reprints to: Cyrus Cooper, M.A., D.M., F.R.C.P, FMedSci, Professor of Rheumatology, Medical Research Council Epidemiology Resource Centre, University of Southampton, Southampton General Hospital, Southampton SO16 6YD, United Kingdom. E-mail: cc{at}mrc.soton.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subjects and Methods: We conducted a population-based longitudinal study of 307 term pregnancies using a cohort of 307 pregnant women living in Southampton, United Kingdom. During early and late pregnancy, skeletal status was measured at the left calcaneus using quantitative ultrasound (QUS).
Results: There was a significant (P < 0.001) decline in both speed of sound and broadband ultrasound attenuation during pregnancy. Those women who were pregnant for the first time (P = 0.001), had low milk intake prepregnancy (P = 0.01), and reduced measures of fat mass (P = 0.01) showed the greatest decline in calcaneal bone measurements. Furthermore, those women who were pregnant over winter months had greater losses in calcaneal QUS (P = 0.02).
Conclusion: Maternal lifestyle, fat stores, and seasonality of early pregnancy influence maternal calcaneal QUS loss during pregnancy; the findings support a role for vitamin D supplementation of women pregnant during winter, especially those with low calcium intakes who are pregnant for the first time.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, there is marked between-individual variation in the changes in bone mass observed during pregnancy, with some mothers even reported to gain bone mineral as measured by QUS. We therefore assessed the extent and determinants of maternal QUS during pregnancy in a population-based cohort of healthy women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A research nurse administered the questionnaire at the initial interview at the woman’s home. The questionnaire included sociodemographic characteristics, lifestyle, milk intake, previous obstetric history, and recalled birthweight. Those women becoming pregnant were then assessed in early (11 wk) and late (34 wk) pregnancy. At each visit, they completed another interviewer-administered questionnaire where information on lifestyle characteristics, smoking habit, alcohol consumption, and the level of physical activity were obtained. Measurements were made of the following: height using a stadiometer, weight using calibrated electronic scales, triceps skinfold thickness, and mid-upper arm circumference (MUAC).
The mother’s skeletal status during pregnancy was measured twice by QUS of the left foot using a calcaneal ultrasound device (Sahara, Hologic Inc., Bedford, MA). The QUS instrument measures speed of sound (SOS), broadband ultrasound attenuation (BUA), and calcaneal width. The instrument was calibrated daily using its own phantom. In a repeatability study of 40 healthy nonpregnant women, the coefficients of variation were found to be 0.8% (SOS) and 3.0% (BUA). Calcaneal scans with a 
2 value of greater than 50 were excluded as per manufacturer’s guidelines.
The local research ethics committee approved the study, and all women gave written informed consent.
Statistical analysis
A sample size of 300 conferred 95% power to detect a 0.2 SD difference in SOS during pregnancy at the 5% significance level. The data were analyzed using STATA version 7.0. The dynamic measurement ranges of SOS and BUA differ markedly; hence, change in SOS and BUA during pregnancy was expressed as a standard deviate (Z) score using the SD of the measurements at 11 wk. Change in SOS and BUA measurements during pregnancy were found to be associated with changes in heel width; hence, both SOS and BUA were adjusted for mean heel width during early and late pregnancy where appropriate.
The effect of season during the first trimester of pregnancy was investigated using the following classification: spring, March–May; summer, June–August; autumn, September–November; and winter, December–February. Further information on the hours of sunshine per month of pregnancy was obtained from The Meteorological Office weather station (Leckford, Hampshire, UK). The data provided were adjusted for seasonal energy variation in UV B radiation using the SoDa-IS web service for professionals in solar energy and radiation (http://www.soda-is.com/index.html). The estimated cumulative UV B exposure was calculated for two time points, during pregnancy, before the 11-wk scan and between the two QUS measurements (11–34 wk).
Univariate analysis of the determinants of baseline QUS was performed, and the significant univariate predictors were used to generate a multiple linear regression model of determinants of both baseline QUS and the change observed during pregnancy.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The anthropometric and lifestyle characteristics of the 307 mothers are shown in Table 1
. The mean age of the mothers at early pregnancy was 29.4 yr, and the median interval between prepregnancy interview and early pregnancy assessment was 1.1 yr [interquartile range (IQR), 0.6–1.8 yr]. The mean interval between early (11 wk) and late (34 wk) pregnancy calcaneal measurements was 22.8 wk (SD 0.7 wk). This sample did not differ from the group as a whole in body build or lifestyle.
|
During pregnancy, there was a significant (P < 0.001) decline in calcaneal SOS and BUA (Table 2). Calcaneal width increased during this time, and although there was no significant relationship between either baseline SOS or BUA with calcaneal width, the change in calcaneal width was significantly (P < 0.01) positively correlated with change in SOS and negatively with BUA measurements (Table 3). The observed reductions in both calcaneal SOS and BUA during pregnancy persisted after adjustment for change in calcaneal width. Furthermore, mothers who had less than a 0.25 SD change in calcaneal width during pregnancy had reductions in both SOS (–0.32 SD) and BUA (–0.32 SD) of a similar magnitude to the remainder of the cohort.
|
|
A positive correlation was observed between maternal age (range, 20.4–37.1 yr) and calcaneal QUS measurements at both early (SOS, r = 0.19, P = 0.001; BUA, r = 0.15, P = 0.01) and late (SOS, r = 0.22, P < 0.001; BUA, r = 0.18, P = 0.002) pregnancy. This relationship was independent of calcaneal width and maternal parity. Maternal age, however, did not predict calcaneal QUS change during pregnancy.
Higher maternal parity was associated with lower early pregnancy maternal SOS (r = –0.14, P = 0.01) and an attenuated reduction in calcaneal SOS during pregnancy (P = 0.01) (Fig. 1). Maternal educational level was also positively correlated with calcaneal SOS in early (r = 0.16, P = 0.005) and late (r = 0.15, P = 0.01) pregnancy but not SOS change during pregnancy. The effects of maternal educational level on calcaneal SOS were independent of maternal age and parity. Age at menarche was not correlated with any measure of calcaneal QUS.
|
). This was independent of changes in calcaneal width. The increment in maternal adiposity during pregnancy did not predict change in calcaneal BUA. Reported maternal birthweight was not associated with QUS measurements during pregnancy. Those mothers reporting physical activity sufficient to cause subjective breathlessness and a rapid heart beat during early pregnancy had higher early pregnancy SOS (+0.3 SD, P = 0.01) and higher BUA (+0.2 SD, P = 0.06) measurements. Reported vigorous activity before pregnancy, but not during pregnancy, was also associated with a greater reduction in SOS (–0.16 SD, P = 0.01) and BUA (–0.13 SD, P = 0.06) compared with those not reporting strenuous activity. The change in reported vigorous activity during pregnancy did not significantly (P > 0.3) influence QUS changes at the heel. Maternal occupational status (full-time, part-time, or none) was also not associated with calcaneal QUS during pregnancy.
Maternal smoking status before or during pregnancy was not associated with either baseline or change in calcaneal QUS measurements. Women who smoked during early pregnancy did have a greater increment in heel width (0.43 SD, P = 0.03) during pregnancy, independent of weight gain during pregnancy. Although there was no overall significant effect of smoking on calcaneal QUS measurements, in those who smoked, the number of reported cigarettes smoked per day during early pregnancy was associated with lower calcaneal SOS during early (r = –0.41, P = 0.01) and late pregnancy (r = –0.34, P = 0.04) and attenuated the reduction in SOS during pregnancy (r = 0.32, P = 0.05).
Maternal milk intake during pregnancy was not associated with change in calcaneal QUS. However, those mothers drinking more than 1 pint milk/d before pregnancy tended to preserve calcaneal SOS during pregnancy (+0.32 SD, P = 0.01) (Fig. 1). Although maternal use of any nutritional supplement during early pregnancy was significantly correlated with maternal education (P < 0.001), there was no association between supplement use and calcaneal SOS. Those mothers who continued to use supplements into late pregnancy did have higher late pregnancy BUA measurements (P = 0.01) but did not differ in the change in BUA during pregnancy.
Seasonal effects on calcaneal QUS
Calcaneal width was 5 mm (1.2 SD, P < 0.001) greater for measurements performed during summer compared with winter. Although the season at the time of calcaneal measurement did not influence SOS or BUA measurements in early or late pregnancy, season during early pregnancy did predict the change in both SOS (P = 0.06) and BUA (P = 0.03). Pregnancies during spring and summer were associated with blunted reductions in SOS and BUA, whereas those in autumn and winter had greater reductions in SOS and BUA (Fig. 2). These effects persisted after adjustment for changes in calcaneal width. There was a weaker, nonsignificant relationship with season at the time of the second QUS measurement in late pregnancy.
|
Multiple linear regression models
Using multiple linear regression models (Table 4), the mutually independent predictors of early pregnancy SOS were age (P < 0.001), vigorous activity (P = 0.03), and lower parity (P = 0.01). Calcaneal BUA during early pregnancy was determined by age (P = 0.01) with a weaker effect of maternal fat stores (P = 0.06). The changes in both SOS and BUA were influenced by season at the time of the early pregnancy visit. Change in calcaneal SOS during pregnancy was also independently predicted by parity and milk intake (>1 pint/d) before pregnancy. Although maternal fat stores varied by season, both were independent predictors of calcaneal BUA change during pregnancy.
|
In 106 pregnancies, the neonate underwent whole-body DXA using a Lunar DPX-L instrument (coefficient of variation < 1%). Figure 3 shows the relationship between change in calcaneal SOS and neonatal whole-body bone mineral density. There were significant negative associations between calcaneal QUS and each of the two neonatal outcomes (whole-body bone mineral content and whole-body bone area, P 
0.05). Thus, babies born to mothers in the highest quarter of the distribution of calcaneal bone loss had bone mineral values 9.7% higher than those born to mothers who showed least change in calcaneal ultrasound. These effects remained significant after adjustment was made for heel width. The associations were not observed for calcaneal BUA.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To meet the increase in calcium demand during pregnancy, there are a number of maternal physiological adaptations, including mobilization of calcium from the maternal skeleton to that of the fetus during pregnancy (1). Bone histomorphometric studies of women during early pregnancy and at term have demonstrated changes in bone structure evident as early as 8 wk gestation (2). In early pregnancy, bone volume decreases with an increase in resorption cavities, whereas in late pregnancy, bone volume recovers with an increase in osteoid and seam width and postulated mineralization rate.
The biphasic response during pregnancy is mirrored by corresponding changes in bone resorption and formation markers, with a progressive increase in bone resorption markers throughout pregnancy (6) and the markers of bone formation only rising in late pregnancy (7, 8). Although the change in maternal bone markers may be due to changes in the developing fetal skeleton, using isomers specific to fetal tissue, the fetal contribution is less than 10% (7).
The reduction in maternal bone mass during pregnancy has also been demonstrated using DXA. In a study of women wishing to become pregnant for the first time, there was a 2.1% reduction in lumbar spine and a 3.8% reduction in distal radial BMD between prepregnancy and postdelivery (3). Similar reductions in BMD at trabecular sites have been reported in another longitudinal study using whole-body DXA measurements (7).
Previous work has identified a progressive decline in SOS and BUA measurements with the greatest loss in the last trimester, the time of greatest fetal demand for mineral (4, 6, 9). The magnitude of the decline in QUS measurements in our study is in accord with that reported by Sowers et al. (10), who demonstrated a 3.6% decline in BUA between 16 wk pregnancy and 6 wk postpartum. However, they were unable to demonstrate a significant decline in SOS, and differences in and adjustment for ankle edema are not mentioned.
Changes in heel width accounted for a proportion of the observed change in QUS during pregnancy. Total heel width comprises the bone volume, marrow volume, and extraosseous soft tissue. Although in a study of patients with dependent pitting edema of on average 6.3 mm, the edema was associated with a reduction in both SOS and BUA measurements (11), in this longitudinal study, whereas BUA measurements were reduced with increased ankle edema, SOS measurements were slightly but significantly increased. The cause for this is not as yet apparent. The relationship between increased heel width and change in calcaneal QUS during pregnancy may be due to factors other than soft tissue edema. From prospective studies in older women (12), the differences in QUS we have observed during pregnancy would approximate to a 28% increase in relative risk of fracture.
We have shown that increased maternal adiposity rather than weight gain during pregnancy is associated with higher calcaneal BUA measurements and an attenuated loss during pregnancy. This is in accord with a cross-sectional study of children and young adults (13), suggesting that increased loading of the calcaneus increases BUA measurements. We were unable to detect a significant independent effect of maternal smoking status on maternal QUS. A previous cross-sectional study has also demonstrated increased bone loss in nulliparous compared with parous young women and adolescents (10), and this is replicated in our observations. Although the average decline in BUA during pregnancy was similar (3.6 vs. 4.0% in this study), the effect of parity was restricted to change in BUA and did not affect SOS, as demonstrated in this study.
Season at the time of the early pregnancy assessment and estimated UV-B exposure influenced the subsequent change in maternal calcaneal SOS and BUA during pregnancy. This is likely to be due to seasonal variation in vitamin D status influencing maternal bone loss rates during pregnancy. Although UV-B exposure is the principal determinant of vitamin D status, there is large interindividual variation in sunlight exposure, and, in the absence of serum maternal vitamin D concentrations at the different stages of pregnancy, observed seasonal influences on pregnancy-related change in bone mass are likely to be underestimated.
The endocrine mechanism underlying dissociated bone resorption in early pregnancy is not fully characterized. In addition to the increase in weight, pregnancy is a high-estrogen state that should, through inhibition of osteoclast recruitment and activity, maintain bone mass. Higher maternal calcitonin levels during pregnancy also protect the maternal skeleton from increased bone resorption (14). During early pregnancy, maternal serum PTH levels are suppressed (8), and although there is an increase in 1,25 (OH)2 vitamin D from placental 1
OHase activity, this is matched by an increase in vitamin D binding protein during pregnancy. Although the free 1,25 (OH)2 vitamin D concentration only rises in late pregnancy, there is some evidence to support activity of bound vitamin D (15). The seasonal effect on changes in maternal bone during pregnancy suggests that the maternal skeleton at this time is still sensitive to changes in vitamin D status. High levels of 1,25 hydroxyvitamin D would inhibit classic PTH secretion from the parathyroid glands and increase absorption of calcium from the maternal gut.
The placenta also produces PTHrP, which has a PTH-like N-terminal end able to stimulate bone resorption. PTHrP stimulates renal 1 hydroxylation of vitamin D and may be responsible for the increase in maternal 1,25 (OH)2 vitamin D concentration during pregnancy. However, changes in PTHrP concentration in the maternal serum have not been consistently demonstrated (8). Other candidate hormones include ßhCG, which has been associated with osteolytic tumors (16), and IGF-1, whose concentration in the maternal serum rises in pregnancy, preceding the rise in bone formation markers (8), and is negatively associated with changes in maternal BMD during pregnancy (7). Serum prolactin, secreted by both maternal pituitary and uterine decidua, rises during pregnancy, and the inhibition of prolactin secretion during pregnancy is associated with reduced bone turnover (17). Leptin, a marker of adiposity, inhibits prolactin production, and this may explain the protective effect of maternal adiposity on calcaneal QUS changes that we have demonstrated (18). Recovery of bone mass is associated with resumption of menses, with further bone loss during postpartum amenorrhea (19).
There are several limitations in our study. We were unable to measure calcaneal SOS and BUA before pregnancy and have used the measurements recorded at 11 wk as baseline. There is histological evidence of increased bone resorption even before this point of pregnancy (2); therefore, the observed changes in QUS during pregnancy may be an underestimate. Although there were significant reductions in both SOS and BUA during pregnancy, these were less than the least significant change for each measure of calcaneal QUS. The greater reproducibility error in QUS measurements may account for the small proportion in the variance of QUS accounted for by the final independent models. Also, although we have demonstrated that those mothers consuming less than 1 pint milk/d had higher rates of QUS loss at the heel, in the absence of data on other dietary sources of calcium or vitamin D, we are unable to estimate an adequate calcium intake needed to maintain maternal bone mass during pregnancy.
In summary, maternal calcaneal BUA and SOS measurements fall during pregnancy, indicating a loss of bone mass. These changes are accentuated in women pregnant for the first time, reporting vigorous activity in early pregnancy, consuming less than 1 pint milk/d, and those who were pregnant during autumn or winter for the first stage of the pregnancy. In addition, the seasonal variation in vitamin D status is likely to influence the maternal bone response. Hence, calcium and vitamin D supplementation of mothers pregnant during winter, especially those with low milk intakes and pregnant for the first time, may help preserve maternal skeletal status during pregnancy.
|
Acknowledgments |
|---|
We thank the mothers who gave us their time, I. Cameron and T. Wheeler for allowing us to include their patients, and a team of dedicated research nurses and ancillary staff for their assistance. We also thank G. Hood and the Meteorological Office (Exeter, UK) and L. Wald (SoDa, Armines, France) for their invaluable assistance in calculating estimated UV B exposure. Participants were drawn from a cohort study funded by the Medical Research Council. We thank Mrs. G. Strange for helping prepare the manuscript.
|
Footnotes |
|---|
First Published Online June 28, 2005
1 See Acknowledgments for names of members of the Southampton Women’s Survey Study Group. 
Abbreviations: BUA, Broadband ultrasound attenuation; DXA, dual-energy x-ray absorptiometry; IQR, interquartile range; MUAC, mid-upper arm circumference; QUS, quantitative ultrasound; SOS, speed of sound.
Received January 28, 2005.
Accepted June 20, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Olausson, M A. Laskey, G. R Goldberg, and A. Prentice Changes in bone mineral status and bone size during pregnancy and the influences of body weight and calcium intake Am. J. Clinical Nutrition, October 1, 2008; 88(4): 1032 - 1039. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Harvey, M. K. Javaid, J. R. Poole, P. Taylor, S. M. Robinson, H. M. Inskip, K. M. Godfrey, C. Cooper, E. M. Dennison, and Southampton Women's Survey Study Group Paternal Skeletal Size Predicts Intrauterine Bone Mineral Accrual J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1676 - 1681. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M Inskip, K. M Godfrey, S. M Robinson, C. M Law, D. J. Barker, C. Cooper, and and the SWS Study Group Cohort profile: The Southampton Women's Survey Int. J. Epidemiol., February 1, 2006; 35(1): 42 - 48. [Full Text] [PDF]
|
|
|
|
 |
|
 G. F. Cunningham and L. G. Raisz Screening for osteoporosis. N. Engl. J. Med., November 3, 2005; 353(18): 1975 - 1975. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
