Prediction of Iron Deficiency Anemia in Third Trimester of Pregnancy Based on Data in the First Trimester: A Prospective Cohort Study in a High-Income Country

Abstract

Background: Systematic iron supplementation may be harmful in pregnant women with non-depleted iron. Our objectives were to estimate the prevalence of anemia at the third trimester of pregnancy (T3) and to identify the parameters at the first trimester (T1), which best predict anemia at T3. Methods: This prospective cohort study in France included pregnant women at T1 without non-iron deficiency anemia. Clinical and social characteristics, health-related quality of life, blood count, and a frozen blood sample were collected at T1 and/or T3. Secondly, a matched nested case–control study was built for women with anemia at T3 but not at T1. Multivariate analyses and ROC curves were used to identify the best predictive parameter(s) of anemia at T3. Results: The prevalence of anemia at T3 in the cohort (629 women) was 21.9% (95% CI 18.7–25.2%). In the matched nested case–control study (256 women), hemoglobin (Hb), serum ferritin (SF) and the SF/soluble transferrin receptor ratio at T1 were predictive of anemia at T3 (p < 0.001); however, clinical and social characteristics, as serum hepcidin were not. In multivariate analyses, Hb at T1 was the best predictive biomarker of anemia at T3 with a cut-off value of 120 g/L (specificity 87.5%). Conclusions: The prevalence of anemia at the end of pregnancy remained high in a High-Income Country. Clinical, social, and biochemical parameters did not seem useful to predict anemia at T3 and could not guide iron supplementation. We suggest systematically performing a simple blood count in the first trimester of pregnancy and offering oral iron supplementation for women with Hb < 120 g/L.