inherent risk of kidney stones associated with pregnancy. Thus, hypercalciuria may be of some utility in DRI development, in that it indicates that dietary intake of calcium is more than adequate.

In sum, although no studies have directly explored levels of calcium intake sufficient for pregnant women, indirect measures suggest that the maternal calcium requirement is not increased over the non-pregnant state because of the physiological changes in calcium absorption and possibly, to some extent, bone turnover during pregnancy. The majority of epidemiological and long-term prospective studies that have examined the effect of parity on BMD, risk of osteoporosis, and incidence of fracture have found that parity is associated with a neutral or even a protective effect relative to these outcomes (Sowers, 1996; Kovacs and Kronenberg, 1997). In short, pregnancy does not impair long-term BMD or skeletal health of the mother.

POTENTIAL INDICATORS FOR PREGNANCY: VITAMIN D Key physiologic changes that occur in pregnancy to assure delivery of adequate calcium to meet fetal needs are relevant for DRI development. Potential indicators for vitamin D requirements during pregnancy are described below.

• Calcium absorption Although the efficiency of calcium absorption doubles in pregnancy, evidence from studies in the Vdr-null mouse shows that this up-regulation occurs independently of vitamin D or calcitriol (Van Cromphaut et al., 2001; Fudge and Kovacs, 2010). Mechanistic evidence is not available from humans; indeed, if such data were available they would still be difficult to interpret because of the known concomitant physiological adaptations. Thus, this measure is not useful for an integrated bone health indicator for vitamin D in pregnancy.

• Maternal, fetal, and childhood BMC/BMD Regarding biological plausibility, fetal calcium homeostasis, skeletal development, and bone mineralization appear independent of vitamin D, the VDR, and calcitriol, based on animal models, and human genetic mutations, as discussed above. Regarding AHRQ-Ottawa, this analysis identified three cohort studies and found insufficient evidence on the association of serum 25OHD levels with maternal BMD during pregnancy. No additional studies were identified addressing vitamin D and maternal BMD. One RCT (Delvin et al., 1986) found no effect of vitamin D supplementation on fetal calcium homeostasis. One observational study (Akcakus et al., 2006) reported no relationship between maternal 25OHD level and fetal BMC or BMD. A number of observational studies found normal fetal skeletal devel-