Vitamin D status in early pregnancy and risk of preeclampsia

Abstract

Objective: We sought to examine the association between maternal serum 25-hydroxyvitamin D (25[OH]D) concentration in early pregnancy and the subsequent diagnosis of preeclampsia (PE).

Study design: This was a nested case-control study from 2 prospective Canadian cohorts conducted in Quebec City, Quebec, and Halifax, Nova Scotia, from 2002 through 2010. Participants were pregnant women (n = 169 cases with PE and 1975 controls). Maternal serum was drawn <20 weeks of gestation, and 25(OH)D measurement was performed. Cases were ascertained from medical records. Logistic regression analysis was used to estimate adjusted odds ratios with 95% confidence intervals.

Results: Women who developed PE had a significantly lower 25(OH)D concentration at a mean gestational age of 14 weeks compared with women in the control group (mean ± SD 25[OH]D 47.2 ± 17.7 vs 52.3 ± 17.2 nmol/L, P < .0001). Women with 25(OH)D <30 nmol/L compared to those with at least 50 nmol/L had a greater risk of developing PE (adjusted odds ratio, 2.23; 95% confidence interval, 1.29-3.83) after adjustment for prepregnancy body mass index, maternal age, smoking, parity, season and year of blood collection, gestational week at blood collection, and cohort site. Exploratory analysis with cubic splines demonstrated a dose-response relationship between maternal 25(OH)D and risk of PE, up to levels around 50 nmol/L, where the association appeared to plateau.

Conclusion: Maternal vitamin D deficiency early in pregnancy defined as 25(OH)D <30 nmol/L may be an independent risk factor for PE. The relevance of vitamin D supplementation for women of childbearing age should be explored as a strategy for reducing PE and for promoting a healthier pregnancy.

Keywords: 25-hydroxyvitamin D; preeclampsia; vitamin D.

FIGURE. Dose-response association between maternal serum 25(OH)D concentrations at <20 weeks’ gestation and risk of preeclampsia

Solid line represents the odds ratio (OR) relative to 50 nmol/L, and dotted lines represent 95% confidence intervals (CIs). Dose-response relationship was modeled with restricted cubic splines using logistic regression. Adjustment is made for prepregnancy body mass index, parity, and variables forced in model maternal age, smoking during pregnancy, season of blood collection, year of blood collection, gestational age at blood collection, and study site (Halifax, Nova Scotia, or Quebec City, Quebec). 25(OH)D, 25-hydroxyvitamin D.