Anemia of Inflammation during Human Pregnancy Does Not Affect Newborn Iron Endowment

Abstract

Background: To our knowledge, no studies have addressed whether maternal anemia of inflammation (AI) affects newborn iron status, and few have addressed risk factors for specific etiologies of maternal anemia.

Objectives: The study aims were to evaluate 1) the contribution of AI and iron deficiency anemia (IDA) to newborn iron endowment, 2) hepcidin as a biomarker to distinguish AI from IDA among pregnant women, and 3) risk factors for specific etiologies of maternal anemia.

Methods: We measured hematologic biomarkers in maternal blood at 12 and 32 wk of gestation and in cord blood from a randomized trial of praziquantel in 358 pregnant women with Schistosoma japonicum in The Philippines. IDA was defined as anemia with serum ferritin <30 ng/mL and non-IDA (NIDA), largely due to AI, as anemia with ferritin ≥30 ng/mL. We identified cutoffs for biomarkers to distinguish IDA from NIDA by using area under the curve (AUC) analyses and examined the impact of different causes of anemia on newborn iron status (primary outcome) by using multivariate regression modeling.

Results: Of the 358 mothers, 38% (n = 136) had IDA and 9% (n = 32) had NIDA at 32 wk of gestation. At 32 wk of gestation, serum hepcidin performed better than soluble transferrin receptor (sTfR) in identifying women with NIDA compared with the rest of the cohort (AUCs: 0.75 and 0.70, respectively) and in identifying women with NIDA among women with anemia (0.73 and 0.72, respectively). The cutoff that optimally distinguished women with NIDA from women with IDA in our cohort was 6.1 µg/L. Maternal IDA, but not NIDA, was associated with significantly lower newborn ferritin (114.4 ng/mL compared with 148.4 µg/L; P = 0.042).

Conclusions: Hepcidin performed better than sTfR in identifying pregnant women with NIDA, but its cost may limit its use. Maternal IDA, but not NIDA, is associated with decreased newborn iron stores, emphasizing the need to identify this cause and provide iron therapy. This trial was registered at www.clinicaltrials.gov as NCT00486863.

FIGURE 1

Study flow diagram.

FIGURE 2

Iron status of pregnant women with schistosomiasis by anemia type at 12 and 32 wk of gestation. Values are geometric means (95% CIs). The sample size, n, represents the number of participants with serum biomarker measures at 12 and 32 wk, by maternal anemia type at 12 wk. P values were adjusted for praziquantel treatment (yes or no), age (<30 y or ≥30 y), parity (<3, 3–4, or >4), maternal height (centimeters), infection with hookworm at 12 wk of gestation, and fetal sex (male or female). IDA, iron-deficiency anemia; NIDA, non–iron deficiency anemia; sTfR, soluble transferrin receptor.

FIGURE 3

ROC curves for serum hepcidin, IL-6, sTfR, and CRP concentrations in the identification of maternal NIDA, IDA, and anemia at 32 wk of gestation in pregnant women with schistosomiasis. (A) NIDA compared with the others (IDA and no anemia), (B) IDA compared with others, (C) NIDA compared with IDA, and (D) anemia compared with no anemia. Values in parentheses after the variable names indicate AUCs. CRP, C-reactive protein; IDA, iron-deficiency anemia; NIDA, non–iron deficiency anemia; ROC, receiver operating characteristic; sTfR, soluble transferrin receptor; ZPP, zinc protoporphyrin.

FIGURE 4

Cord blood iron status by maternal anemia type (A, D), maternal serum hepcidin tertiles (B, E), and cord blood plasma hepcidin tertiles (C, F) of pregnant women with schistosomiasis. Values are geometric means (95% CIs). Models were adjusted for praziquantel treatment (yes or no), age (<30 y or ≥30 y), parity (<3, 3–4, or >4), maternal height (centimeters), infection with hookworm at 12 wk of gestation, fetal sex (male or female), and gestational age at birth (weeks). *P < 0.05, **P < 0.01, ***P < 0.001. IDA, iron-deficiency anemia; NIDA, non–iron deficiency anemia; sTfR, soluble transferrin receptor; T, tertile.