| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR |
MRC Human Nutrition Research
Elsie Widdowson Laboratory
Fulbourn Road
Cambridge CB1 9NL
United Kingdom
E-mail: ann.prentice{at}mrc-hnr.cam.ac.uk
MRC Keneba
The Gambia
Dear Sir:
We thank Koo for his comments on our recent article (1) and the opportunity to clarify and expand on certain points. First, we need to make clear that the reported study involved a subset of a calcium intervention study in The Gambia, as yet unpublished, and was not part of the recent World Health Organization (WHO) randomized trial of 8325 pregnant women with low calcium intakes (2).
We agree with Koo that the study was limited by the instrumentation available and by the relatively small number of measurements made with dual-energy X-ray absorptiometry (DXA). We drew attention to this in the methods and discussion sections of our article. However, the observed lack of a significant effect of calcium supplementation during pregnancy on infant length and radial bone mineral content (RadBMC) in all subjects supports the finding of no significant difference in whole-body BMC (WBBMC) between the intervention groups in the subset. The number of infants with DXA measurements at each age (
50) was similar to that in the American study, in which a significant effect of maternal calcium supplementation on infant WBBMC was noted in the subgroup of mothers with a low calcium intake (3). This finding suggests that the sample size was adequate in the Gambian study to detect an equivalent effect of the supplement, if it were present. In addition, in the Gambian study, the tendency toward a lower WBBMC at each age in the infants of supplemented mothers was in the opposite direction of that originally hypothesized, and it is unlikely that, had DXA measurements been made in all infants, it would have materially altered the overall conclusion of the study, ie, that there had been no positive benefit of the supplement on infant bone growth and mineral accretion.
Koo discusses the point that RadBMC increases at a different rate than that of WBBMC during early life; thus, the ratio between them is influenced by gestational and postnatal age. This should not have affected the evaluation of the effect of maternal calcium supplementation on bone mineral accretion, given that the infants in the 2 intervention groups did not differ significantly in gestational age, that the infants were measured at the same postnatal ages, and that the effects on RadBMC and WBBMC were evaluated separately. Similarly, although the DXA software was designed for use in infants weighing >5 kg and tissue depth may have affected bone edge detection and hence absolute DXA values in lighter infants, no significant differences in the weights of the infants in the 2 intervention groups were observed at any age. Thus, potential artifacts caused by the small size of the Gambian infants were unlikely to have affected the between-group comparisons. Furthermore, at 52 wk, the infants had weights within the recommended range, and the DXA data were internally consistent across infancy; very strong intraindividual correlations were observed (1). This suggests that major problems with the measurements at weights <5 kg that would have obscured a supplement effect at 2 and 13 wk were unlikely.
We stand by our comments about DXA scanning in infants. Although we recognize the valuable studies that Koo and others have conducted to validate the accuracy and precision of DXA instruments and software for use in infants, there continues to be a lack of comparability between systems. In addition, the assessment of accuracy involves comparisons of WBBMC measured by DXA with carcass ash weight in animal models and provides no information about the accuracy of bone size measurements. Comparisons between studies using different instruments are therefore problematic, and it is particularly difficult to draw conclusions about whole-body bone area, and hence bone mineral density (bone mineral content ÷ bone area), as we discussed in our article. In our Table 5 we draw attention to this problem by referencing published data derived by using a variety of different instruments. We acknowledge that Table 5 is not a comprehensive listing of all DXA studies in infants, but, by providing a summary of published DXA data for infants aged <3 wk, we sought to provide comparative data for the Gambian infants at this age while illustrating the difficulties of comparing bone area and bone mineral density obtained with different systems.
Finally, Koo makes an important point about the possibility that an inadequate supply of energy and other nutrients may have limited the beneficial effects of the extra calcium on fetal and infant bone growth and mineralization. The nutritional status of this rural Gambian population has been characterized in many previous studies, and it is well known that energy restriction is common among pregnant and lactating women (4) and children (5), although adult protein intakes tend to be in line with WHO recommendations (6). The pregnancy weight gains recorded for the mothers in the supplement and placebo groups (1), their protein intakes (supplement = 50 g/d, placebo = 60 g/d; P = 0.1; 7), and the growth rates of their infants (1) suggest that this was also the case in our study but that the 2 intervention groups were well-matched for maternal and infant nutritional status. Poor intakes of vitamins and other dietary factors are also well recorded in this population (8), but we have no reason to believe that these differed between the intervention groups. It is therefore unlikely that differences in dietary intake obscured an effect of the calcium supplement, but it is possible that a limited supply of energy or nutrients other than calcium in the group as a whole prevented a significant response to additional calcium. This possibility needs to be tested, but our data suggest that, in the prevailing dietary environment of many poor countries, an increase in calcium intake alone during pregnancy does not have significant benefits on the birth weight, growth, or bone mineral status of infants in the first year of life.
ACKNOWLEDGMENTS
None of the authors had a conflict of interest.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
