Calcium supplementation and body mass index modify associations between prenatal phthalate exposure and perinatal bone ultrasound measures among pregnant women.

Phthalates have endocrine activity that may interfere with bone health, particularly during pregnancy and the early postpartum period, when bone resorption increases. We evaluated associations between prenatal phthalate exposure and perinatal bone health among 289 mothers in the ELEMENT birth cohort in Mexico City who were randomized upon recruitment to receive 1,200 mg daily calcium supplementation or placebo throughout pregnancy. Spot urine samples at up to three timepoints during pregnancy were assayed for 9 phthalate metabolites. Bone integrity was assessed by quantitative ultrasound speed of sound (SOS) measurements of the phalange and distal radius at 3, 6, and 8 months of pregnancy and 1, 3, 7, and 12 months postpartum. Geometric means of specific gravity-corrected phthalate concentrations were used as overall measures of prenatal exposure. Linear mixed effect models estimated associations between phthalate exposure and repeated perinatal bone SOS measures, adjusting for age, pre-pregnancy body mass index (BMI), education, parity, calcium supplementation, and month of pregnancy/postpartum. Effect modification by calcium supplementation and BMI were assessed in sensitivity analyses. An interquartile range increase in MEP and MiBP increased pregnancy phalange z-scores (ß: 0.11; 95%CI: 0.003, 0.31 and ß: 0.15; 95%CI: 0.00,0.42, respectively). Higher concentrations of several phthalate metabolites resulted in lower SOS measures among women who received calcium supplements (compared to placebo group) but higher SOS measures among women with a BMI≥25 (compared to BMI<25). These results suggest that phthalate exposure may interfere with bone remodeling during pregnancy, and that consideration of effect modifiers is paramount to fully understand the effects of environmental exposures on bone health.

Vitamin D receptor polymorphisms and bone mineral density in Mexican women without osteoporosis.

Polymorphisms corresponding to Apa I, Bsm I, and Taq I restriction endonucleases at the vitamin D receptor (VDR) gene and bone mineral density (BMD) at lumbar spine (L2-L4) and proximal femur (neck, Ward's triangle and trochanteric region) sites were examined in a sample of 98 Mexican women (age 55 +/- 10 years). None of the subjects were pregnant or nursing and none had a previous diagnosis of osteoporosis. Polymorphisms were assessed by the restriction fragment length polymorphism - polymerase chain reaction (RFLP-PCR) technique. Alleles were denoted with capital letters for the absence of the RFLP site (A, B, or T) and with small letters for its presence (a, b, or t). BMD was assessed by dual energy X-ray absorptiometry (DXA). A structured, self-administrated questionnaire was used to obtain data on age, menopause, number of pregnancies, breast-feeding, fractures, exercise, smoking, alcohol, estrogens, calcium supplement, height, weight, and BMI. There were no differences between BMD at the skeletal sites and the genotypes disclosed by Apa I (Allele A = 0.43), Bsm I (Allele B = 0.26) and Taq I (Allele T = 0.76). The present study provides data for comparison with other studies to determine the possible value of genotyping VDR to predict predisposition for osteoporosis in Mexican or Mexican-American women. Am. J. Hum. Biol. 11:793-797, 1999. Copyright 1999 Wiley-Liss, Inc.