Iron-Deficiency Prevalence and Supplementation Practices Among Pregnant Women: A Secondary Data Analysis From a Clinical Trial in Vancouver, Canada

ABSTRACT

Background

North American public health guidelines recommend supplementation with an iron-containing prenatal multivitamin throughout pregnancy to meet the RDA of 27 mg of elemental iron daily. However, whether supplementation with standard prenatal multivitamins is sufficient to prevent maternal iron deficiency is unclear, as needs increase substantially with advancing gestation.

Objectives

This study aimed to assess iron status in early and late pregnancy among 60 pregnant women receiving 27 mg/day of elemental iron as part of a randomized trial in Vancouver, Canada.

Methods

Study visits were conducted at 8–21 (baseline) and 24–38 (endline) weeks of gestation. Venous blood specimens were collected for a complete blood count and measurement of iron and inflammatory biomarkers. Supplementation with any additional iron (beyond 27 mg/day) was reported by participants (treatment with additional iron is recommended if ferritin is <30 μg/L). Quantile regression was used to explore predictors of endline ferritin concentrations, including ethnicity, education, income, and baseline ferritin measurement.

Results

Overall, 60 and 54 women participated in baseline and endline visits, respectively. Rates of probable iron deficiency (ferritin <30 μg/L) at baseline and endline were 17 (28%) and 44 (81%), respectively. Less than half (n = 18; 41%) of participants with probable iron deficiency at endline reported supplementation with additional iron. Ethnicity was the only significant modifier of endline ferritin, with higher concentrations in those of South, East, and Southeast Asian ethnicity compared to those of European ethnicity (β: 10.4 μg/L; 95% CI: 0.3–20.5).

Conclusions

Pregnant individuals may require additional supplemental iron beyond 27 mg to meet requirements in later pregnancy, given the high rates of iron deficiency observed in this clinical trial, despite consumption meeting 100% of the RDA. This trial was registered at clinicaltrials.gov as NCT04022135.

See corresponding editorial on page 2184.

Introduction

Iron requirements increase with advancing gestation and are ∼10-fold higher by late pregnancy to accommodate the expansion of red blood cell mass and iron transfer to the placenta and fetus (1). Maternal iron deficiency, anemia, and iron-deficiency anemia are most prevalent in the third trimester, and may result in symptoms of fatigue, restless leg syndrome, pica, hair loss, and irritability (often dismissed as normal symptoms of pregnancy), and have been associated with impaired neonatal growth and neurocognitive development (2–8) and with perinatal morbidity and mortality (9). To meet increased iron needs, the RDA increases to 27 mg of elemental iron daily throughout pregnancy (10). As this is difficult to obtain via diet alone, public health agencies in the United States and Canada recommend daily supplementation with an iron-containing prenatal multivitamin (11,12). Although the majority of pregnant individuals in the United States (∼70%) (13) and Canada (∼90%) (14) report consumption of prenatal multivitamins and standard formulations typically contain 20–27 mg of elemental iron, there are concerns that this may not be sufficient to prevent iron deficiency with advancing gestation (15,16).

Recent reports note high rates of iron deficiency in otherwise healthy, nonanemic women in North America (16–18). In a prospective study conducting assessments of iron status among 102 pregnant women attending initial prenatal appointments (8–10 weeks gestation) in Baltimore, Maryland, 42% were iron deficient (serum ferritin concentrations <30 μg/L) (17). In 2 retrospective cohort studies among pregnant individuals in Toronto, Canada [n = 1307 (16) and n = 44,552 (18)], ∼80% (16) and ∼50% (18) of those screened for iron deficiency had serum ferritin concentrations <30 μg/L. Although rates of iron supplementation were not reported in any of these studies (16–18) and the number of weeks of gestation at the time of ferritin assessment in Canadian investigations was unknown (16,18), this is very concerning given the high rates of prenatal multivitamin supplementation reported across North America (13,14). Additionally, high rates of iron deficiency in the first trimester (17) are alarming as, again, iron requirements increase significantly with advancing gestation, particularly into the third trimester (1). Overall, whether the RDA (27 mg/day) for iron during pregnancy or the quantity of iron consumed and absorbed from standard prenatal multivitamins is enough to prevent iron deficiency in pregnancy is unclear and warrants further investigation.

Standards for the assessment of iron status (e.g., ferritin monitoring) during pregnancy are variable across North America. The US Preventative Services Task Force concluded that there is insufficient evidence to recommend iron-deficiency anemia screening in asymptomatic, pregnant women (19–21); however, they note that if testing does take place, the measurement of serum hemoglobin or hematocrit is often the first step (19,20). The American College of Obstetricians and Gynecologists recommend routine assessment of hemoglobin in the first trimester and at ∼24 weeks of gestation; if hemoglobin is low, ferritin is measured (22) [as iron deficiency is the most common cause of anemia (4)]. Similar guidelines are in place across Canada (18,23,24); in British Columbia, hemoglobin monitoring occurs at initial and 28-week prenatal visits, followed by assessment of ferritin concentrations in those with low hemoglobin (25,26) and treatment with iron (in addition to the RDA of 27 mg/day) if ferritin is <30 μg/L (27). These protocols are concerning, because iron deficiency is first characterized by the depletion of iron stores (ferritin), followed by progression to iron-deficiency anemia (characterized by iron deficiency and reduced hemoglobin concentrations) (28). Thus, iron deficiency may progress unnoticed until the onset of anemia, given that erythropoiesis is maintained until iron stores are exhausted (2, 4,18). Given these clinical practice guidelines, it is likely that iron deficiency (in nonanemic women) is being overlooked in a high proportion of pregnancies.

While previous cohort studies report high rates of gestational iron deficiency among low-risk pregnant women, there remains substantial controversy regarding optimal maternal iron intakes, with some who argue that supplementation may not be required throughout pregnancy (29). The current literature is lacking data on iron supplementation practices (16–18), leading to a lack of ability to evaluate sufficiency of the RDA. The aim of this secondary data analysis was to assess iron statuses and prevalences of iron deficiency and anemia among healthy, pregnant women participating in a clinical trial in which they were provided a daily prenatal multivitamin containing 27 mg of elemental iron.

Methods

The full study protocol of the original trial is published elsewhere (30). Briefly, pregnant individuals (n = 60) in Vancouver, British Columbia, Canada, were randomized to supplementation with folic acid or (6S)-5-methyltetrahydrofolic acid for 16 weeks of pregnancy. All participants also received a standard prenatal multivitamin (NPN: 80025456) with folate removed, which provided 27 mg of elemental iron (ferrous fumarate) total (daily dose, 2 capsules per day; 13.5 mg elemental iron per capsule). Participants were instructed to take 1 folate capsule and 1 prenatal capsule in the morning and 1 prenatal capsule in the evening, to improve tolerability and absorption (31). Participants were generally healthy individuals with low-risk, singleton pregnancies, and a reported pre-pregnancy BMI of <30 kg/m². Two study visits took place during pregnancy: a baseline and an endline visit between 8–21 and 24–38 weeks of gestation, respectively. Participants self-reported demographic characteristics, including age, ethnicity, education, annual household income, and parity. Supplementation with other micronutrients (other than the study supplements, e.g., additional supplemental iron) was recorded. Ethics approval was obtained from the University of British Columbia Children and Women's Research Ethics Board (H18-02635).

Blood collection and biomarker quantification

Blood specimens were collected at baseline and endline for a complete blood count and for the assessment of iron and inflammatory biomarkers in a 2-mL EDTA tube and 4-mL serum tube (Beckton Dickinson), respectively. The complete blood count, including measurements of hemoglobin (g/L), the mean corpuscular volume (fL), and reticulocyte hemoglobin (pg), was performed using an automated hematology analyzer (Sysmex XNL-550, Sysmex Corp.). Serum tubes were left at room temperature for ∼30 minutes after collection and then centrifuged at 1308 × g for 15 minutes at 4°C. Serum aliquots were collected and stored at −80°C. Serum ferritin (μg/L), serum soluble transferrin receptor (sTfR; mg/L), C-reactive protein (mg/L), and α-1-acid glycoprotein (g/L) values were quantified via a sandwich ELISA technique (32). Ferritin is an acute-phase reactant that is elevated in the presence of inflammation, making it difficult to interpret in pregnancy (33–35). To account for this, serum ferritin concentrations were adjusted for inflammation using the internal regression correction approach outlined by the Biomarkers Reflecting the Inflammation and Nutritional Determinants of Anemia project (36). All biochemical analyses were conducted for research purposes after the study was complete. Prenatal care providers did not have access to the results before delivery, and clinical care decisions were based on routine perinatal bloodwork.

Statistical analyses

Descriptive statistics were used to summarize participant demographic characteristics and biochemical outcomes. Continuous variables were reported with a mean ± SD or median (IQR) if not normally distributed, and categorical variables were reported as n (%). Serum ferritin concentrations were adjusted for inflammation in all subsequent analyses (as applicable). The following cutoffs were used for iron deficiency: ferritin <15 μg/L (iron deficiency) and 15–30 μg/L (probable iron deficiency) (33). Anemia was characterized using trimester-specific hemoglobin cutoffs of <110 g/L in the first and third trimesters and <105 g/L in the second trimester (37). Iron-deficiency anemia was characterized by the presence of anemia and iron deficiency. Differences in biochemical outcomes and rates of iron deficiency and iron-deficiency anemia from baseline to endline were assessed with a paired t-test (or Wilcoxon signed-rank test if not normally distributed) and McNemar's test. As treatment with oral iron is recommended in those with probable iron deficiency (ferritin <30 μg/L) (27), rates of those who reported supplementation with additional iron were described. As demand for iron increases with advancing gestation and ferritin concentrations are generally the most depleted in the third trimester, we explored predictors of median endline serum ferritin values (μg/L), including ethnicity, education, and income, adjusting for each other and baseline serum ferritin values (μg/L) using multivariable quantile regression. Covariates were determined a priori, as pertinent characteristics associated with iron-deficiency screening (ethnicity, income, education) (18) and baseline values may be predictive of values in later pregnancy. A P value < 0.05 was considered to be statistically significant. Statistical analyses were conducted using Stata 16.1 (Stata Corp.).

Results

Overall, 60 and 54 women participated in baseline and endline visits, respectively. Reasons for the lack of 6 endline visits were concerns related to in-person visits during the coronavirus disease 2019 pandemic (n = 1), preterm delivery (n = 1), general sickness (n = 2), and pregnancy loss (n = 2). All participants reported supplementation with a standard prenatal multivitamin or with iron/folic acid prior to study enrollment, which was discontinued and replaced with the study prenatal multivitamin at baseline (note: for 1 participant, supplementation only took place before conception and was discontinued prior to baseline). The baseline characteristics of participants are presented in Table 1. The average age of participants was 33 years and the mean gestational weeks at baseline and endline were 16 and 32 weeks, respectively. Participants were predominantly of European ethnicity (57%), had postsecondary education (97%), had an annual household income >CAD$100,000 (52%), and were nulliparous (73%).

TABLE 1.

Baseline characteristics of participants

Participant characteristics	N = 60
Age, years (mean ± SD)	33.0 ± 3.4
Ethnicity, n (%)
European	34 (57)
South, East, and Southeast Asian	13 (22)
Hispanic/Latino	7 (12)
Middle Eastern	2 (3)
Mixed ethnicity	4 (6)
Education, n (%)
High school	2 (3)
College	9 (15)
Undergraduate	25 (42)
Graduate	24 (40)
Household income per year, CAD$, n (%)
<20,000	1 (2)
20,000–50,000	6 (10)
50,000–100,000	21 (36)
>100,000	30 (52)
Parity, n (%)
Nulliparous	44 (73)
Multiparous	16 (27)
Weeks of gestation at baseline (mean ± SD)	16 ± 4

Levels of iron and inflammatory biomarkers and rates of iron deficiency, anemia, and iron-deficiency anemia are summarized in Table 2. Iron stores (as per inflammation-adjusted serum ferritin) became more depleted from baseline (median, 49 μg/L; IQR, 28–49 μg/L) to endline (median, 17 μg/L; IQR, 12–24 μg/L), while concentrations of sTfR increased significantly [baseline median, 3.5 mg/L (IQR, 3.1–4.0 mg/L); endline median, 4.1 mg/L (IQR, 3.6–5.1 mg/L)], indicating depletion of transport iron. Inflammation (C-reactive protein >5 mg/L, indicative of acute inflammation; α-1-acid glycoprotein >1 g/L, indicative of chronic inflammation) remained relatively low throughout the study, decreasing slightly from baseline to endline visits. Rates of iron deficiency (inflammation-adjusted ferritin <15 μg/L) and probable iron deficiency (inflammation-adjusted ferritin <30 μg/L) increased from 8% to 41% and from 28% to 81% from baseline to endline, respectively. Rates of anemia and iron-deficiency anemia were very low, with one participant with both anemia and iron-deficiency anemia at baseline. In the exploratory quantile regression analysis, being of South, East, or Southeast Asian ethnicity was a significant determinant of inflammation-adjusted serum ferritin at endline, compared to being of European ethnicity (β: 10.4 μg/L; 95% CI: 0.3–20.5). Other ethnicities, education, income, and inflammation-adjusted serum ferritin values at baseline were not significantly associated with inflammation-adjusted serum ferritin values at endline (see full results in Supplemental Table 1).

TABLE 2.

Summary of iron and inflammatory biomarkers and rates of iron deficiency, anemia, and iron-deficiency anemia¹

Biomarkers	Baseline (n = 60)	Endline (n = 54)	P value
Serum ferritin,2,3 μg/L, median (IQR)	49 (28–49)	17 (12–24)	<0.001
Serum sTfR,3 mg/L, median (IQR)	3.5 (3.1–4.0)	4.1 (3.6–5.1)	<0.001
Hemoglobin,4 g/L, mean ± SD	123 ± 9	128 ± 14	0.13
Mean corpuscular volume,4 fL, mean ± SD	89 ± 3	91 ± 3	<0.001
Reticulocyte hemoglobin,4 pg, mean ± SD	31 ± 1.3	31 ± 1.4	0.90
CRP,3 mg/L, median (IQR)	2.2 (1.0–5.1)	2.0 (0.9–4.5)	0.78
AGP,4 g/L, mean ± SD	0.53 ± 0.11	0.48 ± 0.09	0.003
Iron deficient,2,5n (%)
Ferritin <15 μg/L	5 (8)	22 (41)	<0.001
Ferritin <30 μg/L	17 (28)	44 (81)	<0.001
Anemia,6,7n (%)	1 (2)	0 (0)
Iron-deficiency anemia,6,7n (%)
Anemia and iron deficiency	1 (2)	0 (0)

¹Abbreviations: AGP, α-1-acid glycoprotein; CRP, C-reactive protein; sTfR, serum soluble transferrin receptor.

²Ferritin is adjusted for inflammation using Biomarkers Reflecting the Inflammation and Nutritional Determinants of Anemia methods (36).

³Ferritin, sTfR, and CRP were assessed with a Wilcoxon signed-rank test.

⁴Hemoglobin, mean corpuscular volume, reticulocyte hemoglobin, and AGP values were assessed with a paired t-test.

⁵Rates of iron deficiency was assessed with a McNemar's test.

⁶Unable to test difference in rates of anemia and IDA due to low cell counts in the 2 × 2 table.

⁷The same participant (n = 1) had anemia and IDA at baseline; the participant discontinued participation in the study prior to endline.

In total, 22 participants reported supplementation with additional oral iron. Seven (12%) reported supplementation with additional iron before enrolment (e.g., in addition to their prenatal multivitamin), with an intention to be continued throughout the study. At endline, 21 (39%) reported additional iron supplementation. This included 15 participants who initiated additional iron supplementation between baseline and endline visits; of the 7 participants who reported additional iron supplementation at baseline, 6 continued supplementation until endline and 1 ended their participation in the study prior to endline analyses. The demographic characteristics of those who reported supplementation with additional iron and of those with probable iron deficiency at endline are presented in Supplemental Table 2.

Iron deficiency (inflammation-adjusted serum ferritin <15 μg/L) and probable iron deficiency (inflammation-adjusted serum ferritin <30 μg/L) at baseline persisted to endline in most participants. In those with iron deficiency at baseline (5 women), 1 remained iron deficient at endline, 2 had ferritin levels indicative of probable iron deficiency, and 2 ended their participation in the trial prior to endline. Of note, the 1 participant who remained iron deficient from baseline to endline did not report any additional iron supplementation, while the 2 whose ferritin concentrations improved slightly (progressed from iron deficiency to probable iron deficiency) did report additional iron supplementation. In those with probable iron deficiency at baseline (17 women), serum ferritin concentrations remained below 30 μg/L in 12 participants at endline (in 5 of the 12, ferritin concentrations dropped below 15 μg/L), 1 participant no longer had any indication of iron deficiency (ferritin concentrations increased to >30 μg/L),  and 4 women discontinued their participation in the trial. While the 1 participant who no longer had ferritin concentrations indicative of probable iron deficiency at endline reported supplementation with additional iron, there were also 6 women who did report additional iron supplementation but whose ferritin concentrations remained below 30 μg/L (5 women) or fell below 15 μg/L (1 woman). Changes in ferritin concentrations from baseline to endline in those who did and did not report supplementation with additional iron are presented in Supplemental Table 3. Most importantly, we report that of participants with probable iron deficiency at endline, fewer than half (n = 18; 41%) reported additional iron supplementation (as is recommended in clinical practice guidelines).

Discussion

High rates of iron deficiency were observed in this secondary analysis of generally healthy, low-risk, pregnant individuals in both early and later pregnancy. At baseline (mean gestational age, 16 weeks), 28% (n = 17) had ferritin concentrations indicative of probable iron deficiency. This increased significantly by endline (mean gestational age, 32 weeks), at which time the vast majority (n = 44; 81%) were likely iron deficient. This occurred despite almost ubiquitous supplementation before study initiation and the provision of 27 mg of elemental iron daily throughout the study (meeting 100% of the RDA). While iron deficiency did not progress to iron-deficiency anemia in this cohort, increases in sTfR concentrations are indicative of progression to iron-deficient erythropoiesis (the phase of iron depletion prior to iron-deficiency anemia) (28). Of those participants with probable iron deficiency, only 18 (41%) reported intake of additional supplemental iron, despite public health guidelines in British Columbia for treatment with oral iron when serum ferritin concentrations are <30 μg/L (27).

These findings contribute to the body of evidence indicating that iron deficiency remains a significant concern for pregnant women in high-resource settings. Comparisons to previous Canadian cohorts are challenging given the lack of data on gestational weeks at the time of ferritin assessments (16,18). Teichman et al. (18) and Tang et al. (16) reported iron deficiency (ferritin concentrations <15 μg/L) in ∼24% and ∼50% of women, respectively, and probable iron deficiency (ferritin concentrations <30 μg/L) in ∼50% and ∼80% of women, respectively, on at least 1 occasion in pregnancy. The proportion of women with ferritin concentrations < 30 μg/L reported by Auerbach et al. (17) at 8–10 weeks of gestation was 42%. Teichman et al. (18) reported that the majority (71%) of ferritin tests were performed in the first trimester, a time when rates would be expected to be at their lowest. Rates found in this study were lower in early pregnancy, with 8% of women having iron deficiency and 28% having probable iron deficiency. This may be attributed to differences in study designs. Randomized clinical trials are the gold standard for evaluating an intervention and may have superior methodological rigor, as all participants were provided the same prenatal multivitamin and the numbers of gestational weeks at the time of ferritin assessments were recorded. However, clinical trials often lack external validity, as participants generally have higher socioeconomic statuses and better adherence to an intervention or improved health behaviors while participating in the study (38). This means that the true prevalence of iron deficiency in an unselected population may be even higher than the 80% in later pregnancy found in this clinical trial population, underpinning the urgency for action to improve iron-deficiency screenings and treatment.

There are several factors that likely contributed to the high rates of iron deficiency and probable iron deficiency in this investigation. As screening for iron deficiency is not a part of routine, perinatal bloodwork across Canada (18,23), it is possible that health-care providers were not aware of the low ferritin concentrations, given that the hemoglobin concentrations remained within normal limits. If no ferritin assessment took place [which according to current clinical practice guidelines, would not have been necessary (26,27)], there would have been no indication to recommend additional supplementation with iron. Although we did not ascertain whether ferritin was assessed by a health-care provider between study visits, it has been reported that ferritin screening varies greatly by clinicians, and a recent study of 44,552 pregnant individuals from Toronto, Canada, found that those with lower socioeconomic statuses are ∼20% less likely to be screened (18). Overall, iron-screening practices across North America are poorly documented (20).

This investigation raises concerns that standard prenatal multivitamins are insufficient to maintain ferritin concentrations throughout pregnancy. While commercial prenatal multivitamins typically contain 20–27 mg of elemental iron, doses of iron vary between brands and bioavailability varies by supplement form (13). Gastrointestinal discomfort is commonly experienced following intake of ferrous iron salts, which are the most cost-effective and commonly prescribed supplemental iron form (39). Thus, the bioavailability of iron in standard prenatal multivitamin formulations may be poor or compliance to daily intake may be sporadic. Additionally, there are an increasing number of “gummy” multivitamins, which may be labeled as “prenatal” but do not contain any iron. While this variability in commercially available supplements and potentially poor bioavailability of certain iron salts may account for the findings at baseline, all participants were provided the same prenatal multivitamin containing 27 mg of elemental iron (ferrous fumarate) throughout the study. We note that the study-provided prenatal multivitamins included 250 mg of calcium carbonate (125 mg per capsule). While supplementation with calcium may reduce iron bioavailability (40,41), it remains unclear whether the impact on the iron status is clinically meaningful (42), particularly following long-term supplementation (43) and in calcium doses <800 mg (44). We note that most commercially available prenatal multivitamin formulations contain calcium. Overall, given that no major issues regarding tolerance or adherence to daily intake were reported, it is possible that the RDA of 27 mg/day into later pregnancy requires reevaluation.

Although we observed that iron deficiency and probable iron deficiency at baseline generally persisted until endline, baseline ferritin values were not significantly associated with endline ferritin values as per the quantile regression analysis (β: 0.06 μg/L; 95% CI: −0.04 to 0.17). Depletion of iron stores and declining ferritin concentrations with advancing gestation are expected (29,45); thus, we'd expect the ferritin concentrations at endline to be lower than the concentrations at baseline. While this was the case overall in this investigation, as median concentrations among the whole group decreased significantly from baseline to endline, there was a high degree of interindividual variation: the median change from baseline to endline was 33 μg/L (IQR, 12–61 μg/L; minimum, −84 μg/L; maximum, 138 μg/L). This variation explains why baseline ferritin concentrations were not significant in our model. In total, 7 women had increased ferritin concentrations from baseline to endline, and the majority of these also reported additional iron supplementation (n = 5; 71%). In the 2 women who did not report any additional iron supplementation, ferritin concentrations only increased modestly (by 2.4 μg/L and 0.7 μg/L). In these cases, perhaps the 27 mg/day of elemental iron was sufficient to sustain their iron stores, or it is possible that they forgot to report additional iron supplementation.

Optimal iron status during pregnancy has recently been the subject of much debate (29). While the ferritin cutoffs used in this study are well accepted and widely used in routine clinical practice, it has been argued that declining maternal serum ferritin concentrations represent a “normal” physiological response in pregnancy (29), possibly attributed to increases in plasma volume (46). However, the fetus does rely on maternal iron stores throughout gestation, with iron from maternal red blood cell catabolism serving as the major source (47). While there are mechanisms to upregulate iron delivery to the fetus in those with lower iron stores (47) and maternal iron absorption increases by approximately 2-fold from the second to third trimester of pregnancy (45,47), the threshold at which the maternal supply can no longer meet the placental and fetal demand is unknown. We reiterate that iron deficiency may have a significant effect on maternal well-being (contributing to fatigue, restless leg syndrome, pica, hair loss, and irritability) and, if left untreated, may progress to iron-deficiency anemia in the postpartum period or for any subsequent pregnancies (which may be a concern for many women in this study, as 73% were nulliparous). This may lead to more serious, long-term health concerns, including cardiac failure, a poor thyroid function, decreased wound healing, a requirement for blood transfusion, and possibly death (3,4,9,48). Inadequate maternal iron stores throughout gestation also have implications for infant development after delivery, as exclusively breastfed infants rely on stores acquired during gestation to meet requirements until ∼6 months of age (49,50). Iron deficiency in early life may impact infant brain development and is a significant risk factor for poorer cognitive, motor, and social-emotional development in later life (4). Overall, further, prospective trials are needed to better understand optimal maternal iron statuses throughout gestation. Until that time, we note that healthy, pregnant women are at high risk for deficiency as per the current cutoffs, despite meeting recommendations for iron supplementation.

Recent clinician-focused publications have advocated for the revision of iron-deficiency screening and treatment guidelines. A 2019 review by anemia researchers and clinicians based at the University of Washington (Seattle, WA), Thomas Jefferson University (Philadelphia, PA), and Georgetown University (Washington, DC) suggests iron-deficiency screening for all pregnant women, regardless of their hemoglobin concentrations (4). Similar recommendations are expressed in a 2021 Canadian Medical Association Journal article urging clinicians to implement hemoglobin and ferritin monitoring at all initial and 28-week prenatal visits (51). A novel toolkit (IRON MOM) was established in Toronto, Canada, to assist clinicians with iron-deficiency diagnoses and management (24). Its implementation at an obstetrics clinic and inpatient delivery ward (St. Michael's Hospital) resulted in a 10-fold increase in ferritin testing, and the incidence of antenatal hemoglobin concentrations <100 g/L was reduced from 13.5% to 10.6% (P < 0.001) (24). The authors note that IRON MOM could be easily scaled to other settings (24).

Limitations to consider in the interpretation of this study are noted below. We note a small sample size in this secondary analysis [which was powered for the original trial purpose (30)]. Additionally, as the primary aim for this analysis was to evaluate folate supplementation, capsule counts were completed for folate capsules only. Although we suspect high adherence to daily supplementation with the prenatal capsules, as participants were given a tracking diary to serve as a daily reminder and the mean adherence to daily supplementation with folate was >95%,  we acknowledge this lack of data as a limitation. However, this also improves the generalizability of findings to real-world supplementation practice. We are also limited by the lack of data around additional oral iron intake, including the iron form, dose, and adherence. While we found that being of South, East, or Southeast Asian ethnicity was associated with having higher ferritin concentrations in later pregnancy, and an increase in ferritin of 10 μg/L appears clinically meaningful, further validation is required, as the 95% CI is wide and approaching the null (95% CI: 0.3–20.5 μg/L). Finally, while adjustment of ferritin concentrations for inflammation helps to improve accuracy, the calculation requires further validation in pregnancy.

In conclusion, iron deficiency remains a significant and largely neglected concern for pregnant individuals. Not only do our findings further support the previously reported, high rates of iron deficiency among pregnant women in North America (16–18), but provision of 27 mg/day of elemental iron to all participants and assessments of iron status at 2 points in pregnancy help to further understand changes from early to late pregnancy and evaluate the sufficiency of the current RDA in maintaining iron stores. Pregnant women may believe that they are meeting their iron needs throughout gestation by consuming a standard prenatal multivitamin. However, this study provides further evidence that women may require additional iron, beyond 27 mg, to support needs into later pregnancy. Further large, longitudinal investigations are needed to understand optimal iron intakes in pregnancy and determine the clinical significance of declining maternal iron stores for mom and baby. Most importantly, our findings support incorporating the assessment of iron status into routine perinatal clinical practice guidelines to ensure that iron deficiency is addressed and that an optimal iron status is maintained throughout all of pregnancy.

Notes

This work was supported by a Healthy Starts Catalyst Grant, BC Children's Hospital Research Institute. KMC is supported by a Frederick Banting and Charles Best Canada Graduate Scholarship Doctoral Award from the Canadian Institute of Health Research. CDK is supported by a Michael Smith Foundation for Health Research Scholar Award, and a Canada Research Chair in Micronutrients and Human Health. JAH holds a Canada Research Chair in Perinatal Population Health.

Author disclosures: The authors report no conflicts of interest. CDK is an editorial board member for the Journal of Nutrition.

Supplemental Tables 1–3 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/jn/.

Data Availability

Requests for data sharing may be considered and should be directed to CDK.