Original Research

The impact of response to iron therapy on maternal

and neonatal outcomes among pregnant women with
anemia
Sarah E. Detlefs, MD; Michael D. Jochum, PhD; Bahram Salmanian, MD; Jennifer R. McKinney, MD, MPH;
Kjersti M. Aagaard, MD, PhD

BACKGROUND: Anemia during pregnancy is associated with iron therapy, 2695 (36.3%) had a successful response to therapy, and
increased risks of preterm birth, preeclampsia, cesarean delivery, and 3402 (45.9%) were untreated. Successfully treated patients with anemia
maternal morbidity. The most prevalent modifiable cause of pregnancy- had a significant reduction in the odds of preterm birth (5.1% vs 8.3%;
associated anemia is iron deficiency. However, it is still unclear whether adjusted odds ratio, 0.59; 95% confidence interval, 0.47−0.72) and pre-
iron therapy can reduce the risks of adverse outcomes in women with eclampsia (5.9% vs 8.3%; adjusted odds ratio, 0.75; 95% confidence
anemia. interval, 0.61−0.91). Refractory and untreated patients had significantly
OBJECTIVE: This study aimed to determine whether response to iron increased odds of preterm birth (adjusted odds ratio, 1.44 [95% confi-
therapy among women with anemia is associated with a change in odds dence interval, 1.16−1.76] and 1.45 [95% confidence interval, 1.26
of adverse maternal and neonatal outcomes. −1.67], respectively) and preeclampsia (adjusted odds ratio, 1.54 [95%
STUDY DESIGN: This was a population-based cohort study (2011 confidence interval, 1.24−1.89] and 1.44 [95% confidence interval, 1.25
−2019) using an institutional database composed of obstetrical patients −1.67], respectively). All groups of women with anemia had increased
from 2 delivery hospitals. Patients with adequate prenatal care were clas- odds of postpartum hemorrhage and decreased odds of delivering a small
sified as being anemic or nonanemic (reference). Patients with anemia for gestational age neonate. There was no difference in composite neona-
were further stratified by success or failure of treatment with oral iron ther- tal morbidity.
apy using the American College of Obstetricians and Gynecologists criteria CONCLUSION: Successful treatment of anemia with oral iron therapy
for anemia at the time of admission for delivery: successfully treated was associated with a reduction in the odds of preterm birth and pre-
(Hgb≥11 g/dL) or unsuccessfully treated (“refractory;” Hgb<11 g/dL). All eclampsia. Women with refractory anemia had similar outcomes to those
categories of women with anemia categories were compared with the ref- who were untreated, emphasizing the importance of monitoring response
erence group of women without anemia using chi-square and logistic to iron therapy during pregnancy.
regression analyses. The primary outcomes were preterm birth and
preeclampsia. Key words: iron deficiency anemia, maternal morbidity, neonatal mor-
RESULTS: Among the 20,690 women observed, 7416 (35.8%) were bidity, preeclampsia prevention, preterm birth prevention
anemic. Among women with anemia, 1319 (17.8%) were refractory to

Introduction <11.0 g/dL in the first and third trimes- unit (ICU) admissions, and postpartum

A nemia during pregnancy is one of

the most common diagnoses
among gravidae, given that it is esti-
ters of pregnancy and a hemoglobin
level of <10.5 g/dL in the second tri-
mester of pregnancy.5 The ACOG rec-
depression.3 A 2011 Cochrane review
evaluated different treatments of iron
deficiency anemia in pregnancy for the
mated to affect 40% of women world- ommends screening all pregnant prevention of perinatal morbidity and
wide.1 The prevalence of anemia among women for anemia and treating women concluded that there is insufficient evi-
pregnant women in the United States with suspected iron deficiency with iron dence to suggest an efficacious reduc-
has been estimated to be approximately supplementation. Of note, 1 study esti- tion in the incidence of perinatal
10% to 12%.2−4 The American College mated that only 50% of pregnant complications.9 They concluded that
of Obstetricians and Gynecologists women with anemia respond to iron further studies are needed to evaluate
(ACOG) has defined pathologic anemia therapy.1 whether treatment of iron deficiency
in pregnancy as a hemoglobin level of Studies have shown that beyond anemia can improve maternal and neo-
physiological dilutional anemia of preg- natal outcomes.
Cite this article as: Detlefs SE, Jochum MD, Salmanian
nancy, iron deficiency is the most com- Given these knowledge gaps, we
B, et al. The impact of response to iron therapy on mater- mon cause of anemia in pregnancy and undertook a large population-based
nal and neonatal outcomes among pregnant women with is associated with an increased risk of cohort study to evaluate whether treat-
anemia. Am J Obstet Gynecol MFM 2022;4:100569. adverse pregnancy outcomes, such as ment of anemia during pregnancy with
2589-9333/$36.00 preterm delivery, preeclampsia (PE), oral iron therapy was associated with a
© 2022 The Author(s). Published by Elsevier Inc. This is cesarean delivery, and perinatal change in the occurrence of associated
an open access article under the CC BY-NC-ND license death.2,6−8 In addition, women with perinatal outcomes, such as preterm
(http://creativecommons.org/licenses/by-nc-nd/4.0/) anemia have an increased risk of perina- birth (PTB) and PE. We further evalu-
http://dx.doi.org/10.1016/j.ajogmf.2022.100569
tal blood transfusions, intensive care ated whether this effect was modified by

March 2022 AJOG MFM 1

Original Research

AJOG MFM at a Glance prenatal visits based on gestational age at

delivery. Patients enrolled in PeriBank
Why was this study conducted? receive prenatal care at several clinics
Iron deficiency anemia during pregnancy is associated with maternal and neona- before delivering at our 2 enrollment
tal morbidities and affects 40% of women worldwide. We do not yet know and collection sites, where they were
whether iron therapy improves outcomes and if all women will be equally recruited at the time of delivery. Their
responsive to iron therapy. prenatal care followed the ACOG rec-
Key findings ommendations and institutional guide-
Successful response to iron therapy was associated with a reduced odds of devel- lines for prenatal care. At our institution,
oping preeclampsia and/or preterm delivery. Anemia left untreated or refractory patients were initially treated with sup-
to iron therapy was associated with a considerably increased odds of occurrence plemental iron therapy if their hemoglo-
of these outcomes. bin is below the ACOG cutoffs for
anemia. If iron studies were performed
What does this add to what is known? at a later time and indicated iron defi-
Anemia was associated with increased maternal morbidity, and the impact of ciency, they were continued on iron sup-
iron therapy on these outcomes was dependent on treatment response. Patients plementation for the remainder of the
who received iron therapy and had a resolution of their anemia had better out- pregnancy. Moreover, most clinics per-
comes than those who remained untreated or anemic despite therapy. formed hemoglobin electrophoresis to
rule out hemoglobinopathies in all
patients.
treatment and response, as measured by data accuracy, and any adjudication of Patients with another known cause of
correction of the maternal hemoglobin cases was performed by maternal-fetal anemia, such as vitamin B12 deficiency,
at delivery. medicine physicians. The database spe- thalassemia, or sickle cell disease, and
cifically captured medication prescrip- those who were actively bleeding on
tions by maternal interview and review admission to labor and delivery were
Materials and Methods of prescriptions or medication dis- excluded from the study. Furthermore,
Data sources pensed within the electronic medical patients with an antenatal diagnosis of
We used an institutional review board record.10−12 Women were specifically known or suspected placenta accreta
(IRB)-approved, subject-consented asked about prenatal vitamin and iron spectrum disorder (including previa,
perinatal database, which is collected therapy as examples of medications accreta, increta, and percreta) were
and maintained by trained full-time used in pregnancy during interviews excluded as they often received iron
research personnel at Baylor College of with study personnel during enrollment prophylactically at our institution.
Medicine in Houston, Texas (PeriBank; to labor and delivery. Maternal demographics, diagnoses,
H-26364). All gravid patients who medications, and outcomes were uni-
delivered at our 2 institutional hospitals formly abstracted for patients.
(Ben Taub Hospital and Texas Child- Selection criteria for study patients Those patients who received iron
ren’s Pavilion for Women) were The use of the PeriBank database for therapy other than that included in a
approached and consented to partici- this study was approved by the Baylor prenatal vitamin were considered to
pate in our perinatal database. In addi- College of Medicine IRB (H-45877). All have a diagnosis of iron deficiency. Sup-
tion to detailed data abstraction from women with singleton pregnancies and plements included the brand names and
the electronic medical record into the sufficient prenatal care from August generic forms of ferrous sulfate, ferrous
database, additional information was 2011 to November 2019 were included. gluconate, ferrous fumarate, and ferrous
obtained by direct maternal interview in For women with ≥2 pregnancies within glycinate. The administration of intra-
their native language. In the current this time frame, the first delivery in the venous (IV) iron was not reported con-
study, more than 100 variables relevant PeriBank database was chosen for inclu- sistently and so was not included in this
to the hypothesis and study aims were sion for the overall analysis. Sufficient analysis. The sample size for this analy-
used.10−12 prenatal care was defined by the Kotel- sis was exhaustive for the number of
The PeriBank database and biorepo- chuck Adequacy of Prenatal Care participants meeting the criteria in the
sitory involved active consent for partic- Index.11,13,14 This model has been previ- database.
ipation from all participants and did not ously validated and took into account
rely on volunteers to self-identify or the number of prenatal visits and gesta- Criteria for patient study group
enter data.10 The rate of enrollment did tional age at initiation of prenatal care allocation
not change considerably throughout the and delivery.11,15 Sufficient prenatal care Patients were categorized into 1 of 2
study period; consent rates ranged from was defined as beginning prenatal care groups: anemic and nonanemic.
86% in 2012 to 90% in 2019. Regular before 20 weeks of gestation and attend- Patients without anemia (eg, women
quarterly audits were done to ensure ing 50% to 100% of recommended who were pregnant without anemia)

2 AJOG MFM March 2022

 Original Research

were used as the reference group. defined as below the 10th percentile psychiatric disease. For example, if a
Patients with anemia were considered using national birthweight reference patient had 3 of these conditions, then
as those who were treated with an iron data, which is stratified by race and eth- they would receive a value of 3 for this
supplement outside of prenatal vitamin nicity and gestational age at delivery.16 composite. A P value of <.05 or CI
or presented to labor and delivery with Large for gestational age (LGA) neo- excluding 1.0 was considered signifi-
anemia as defined by the ACOG crite- nates were defined as above the 95th cant. All statistical analyses were per-
ria.5 This included a hemoglobin level percentile using the same reference. formed, and graphs were created using
of <11 g/dL in the third trimester of the SAS software (version 9.4; SAS
pregnancy or 10.5 g/dL if delivered in Covariates Institute Inc, Cary, NC).
the second trimester of pregnancy. Baseline maternal characteristics,
Patients with anemia were further sub- comorbidities, demographics, and Results
categorized by whether (1) they received adverse perinatal outcomes were ana- Cohort characteristics
iron therapy and (2) whether they had lyzed for all eligible participants. These At the time of this analysis, a total of
normal hemoglobin at the time of included maternal age, parity, marital 43,580 pregnancies had been enrolled in
admission for delivery. Categories were status, body mass index at time of the perinatal database since its inception
labeled as “untreated and anemic” for delivery, educational achievement, race (Figure 1). After excluding individuals
women who were anemic on admission and ethnicity, income, and insurance with missing data from key variables, a
to labor and delivery and did not receive type. Furthermore, maternal comorbid- final study cohort of 20,690 patients was
iron supplementation, “successfully ities were evaluated, which included retained for further analysis.
treated” for women who arrived with chronic hypertension (CHTN), gesta- Information regarding prenatal care is
normal hemoglobin and reported taking tional diabetes mellitus (GDM), type 1 listed in Table 1. The median hemoglobin
iron supplementation, and “refractory diabetes mellitus, type 2 diabetes melli- was 10.2 (9.5−10.6), 12.0 (11.5−13.2),
anemic” for those who were anemic on tus, deep venous thrombosis and pul- 10.3 (9.7−10.7), and 12.3 (11.7−13.0) for
admission to labor and delivery despite monary embolus, hypothyroidism, refractory, successful, untreated, and ref-
taking an iron supplement. Recorded hyperthyroidism, cardiac disease, sei- erence groups, respectively. Maternal
hemoglobin for this analysis was deter- zure disorder, asthma, endometriosis, demographics and comorbidities are rep-
mined at the time of admission to labor cancer, psychiatric disease, and sub- resented in Tables 2 and 3, respectively.
and delivery unit on all patients in both stance use. Successfully treated patients were more
institutions. likely to be White non-Hispanic, older,
Statistical analysis nulliparous, and married with a high
Maternal and neonatal outcomes Descriptive statistics were used to report school education and have private insur-
The primary maternal outcomes desig- all variables of interest. Continuous var- ance (Table 2). Untreated and refractory
nated a priori as of interest were PTB iables were presented as median (inter- patients had similar demographics, and
before 37 weeks of gestation and PE. quartile range) using the Kruskal-Wallis specifically, they were less likely to be nul-
Other maternal outcomes included test. Categorical variables were pre- liparous and married or have a high
cesarean delivery, placenta abruption, sented as number (percentage) and school education and more likely to be a
intrapartum hemorrhage, postpartum were evaluated using the chi-square test minority, make <$35,000 per year, and
hemorrhage (PPH), and composite of association. All groups were initially have government insurance.
maternal morbidity (CMM). CMM compared as a whole to evaluate for sig-
included any of the following: hyper- nificance. Primary outcomes
tensive disorders of pregnancy, cho- Odds ratios (ORs) were calculated by Total number and percent and cORs
rioamnionitis, endometritis, placental comparing all patients with anemia to and aORs for each outcome are listed in
abruption, blood transfusion, maternal the reference group with anemia. Crude Tables 4 to 7. The odds of PTB were sig-
ICU admission, hysterectomy, pulmo- ORs (cORs) and adjusted ORs (aORs) nificantly increased among refractory
nary edema, or maternal death. were reported. aORs and 95% confi- (aOR, 1.44; 95% CI, 1.16−1.76) and
The primary neonatal outcome was dence intervals (CIs) were calculated untreated (aOR, 1.45; 95% CI, 1.26
small for gestational age (SGA). Other using logistic regression. Variables −1.67) patients but was significantly
neonatal outcomes evaluated included included in the regression analyses were decreased for successfully treated
transient tachypnea of the newborn age, nulliparity, education, race and eth- patients even after controlling for con-
(TTN), retinopathy of prematurity, and nicity, composite medical comorbidity, founders (aOR, 0.59; 95% CI, 0.47
composite neonatal morbidity (CNM). and tobacco use. A numeric composite −0.72) (Figure 2; Tables 4 and 5).
CNM included any of the following: 5- of medical comorbidities was created, Moreover, 85.1% of preterm deliveries
minute Apgar score of ≤3, respiratory which included CHTN, pregestational among all groups occurred at >32
distress syndrome, suspected or proven diabetes mellitus and GDM, asthma, weeks of gestation. There was a signifi-
newborn sepsis, seizure, stillbirth, or thyroid disease, seizure disorder, throm- cantly decreased odds of very PTB (<32
neonatal death. SGA neonates were boembolism, cardiac disease, and weeks of gestation) among successfully

March 2022 AJOG MFM 3

Original Research

FIGURE 1 and untreated (aOR, 1.60; 95% CI, 1.23

Study population −2.08) patients.
These findings were similar when
evaluating the odds of developing PE at
all gestational ages. The odds of PE sig-
nificantly decreased among successfully
treated patients (aOR, 0.75; 95% CI,
0.61−0.91). There was a significant
increase in the odds of PE among
refractory (aOR, 1.54; 95% CI, 1.24
−1.89) and untreated (aOR, 1.44; 95%
CI, 1.25−1.67) patients.

Secondary outcomes
CMM was significantly increased
among refractory (aOR, 1.70; 95% CI,
1.44−2.01) and untreated (aOR, 1.71;
95% CI, 1.52−1.92) patients (Tables 4
and 5; Figure 2). The odds of CMM was
not significantly different between suc-
cessfully treated patients and patients
with anemia after controlling for con-
founders (aOR, 0.89; 95% CI, 0.77
−1.03). Refractory and untreated
a
Inadequate prenatal care as defined by the Kotelchuck Adequacy of Prenatal Care Index; bMissing patients had considerably increased
data included hemoglobin and gestational age at delivery. odds of hysterectomy and blood trans-
PO, orally. fusion. Moreover, untreated patients
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022. had considerably increased odds of cho-
rioamnionitis, endometritis, and need
for ICU care. There was no maternal
treated patients (aOR, 0.32; 95% CI, PTB because of preterm labor (PTL) death in the cohort. All anemic catego-
0.14−0.62). There was no difference in (aOR, 0.49; 95% CI, 0.33−0.70), pre- ries had an increased odds of PPH after
very PTB for refractory (aOR, 1.11; 95% term premature rupture of membranes controlling for confounders. There was
CI, 0.62−1.88) or untreated (aOR, 1.33; (PPROM) (aOR, 0.60; 95% CI, 0.38 a significant increase in the odds of
95% CI, 0.92−1.91) patients compared −0.88), and PE (aOR, 0.47; 95% CI, cesarean delivery for refractory (aOR,
with the population without anemia. 0.28−0.74) among successfully treated 1.53; 95% CI, 1.33−1.75) and untreated
The causes of PTB varied by subcate- patients. The odds of PTB because of (aOR, 1.38; 95% CI, 1.25−1.51)
gory (Figure 3). There was a significant PE significantly increased among refrac- patients, but there was no difference for
reduction in the odds of occurrence of tory (aOR, 1.75; 95% CI, 1.17−2.53) successfully treated patients (aOR, 1.09;

TABLE 1
Characteristics of prenatal care among the study groups
Patients with anemiaa
Refractory to treatment Successful treatment Untreated Reference group without anemia
Characteristic (n=1319) (n=2695) (n=3402) (n=13,274)
Component
Hemoglobin level 10.2 (9.5−10.6) 12.0 (11.5−13.2) 10.3 (9.7−10.7) 12.3 (11.7−13.0)
Planned pregnancy 459 (36.2) 1379 (52.5) 987 (31.3) 5581 (46.1)
Prenatal vitamin 1242 (94.2) 2598 (96.4) 2887 (91.6) 11,649 (93.3)
Data are presented median (interquartile range) or number (percentage). Chi-square analysis was used to compare categorical variables. Continuous variables were assessed using the Kruskal-Wallis
test.
a
Anemia was defined as receiving iron therapy or having a hemoglobin level below the American College of Obstetricians and Gynecologists cutoff by gestational age on admission to the labor and
delivery department.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

4 AJOG MFM March 2022

 Original Research

TABLE 2
Maternal demographics by study group
Patients with anemiaa
Refractory to treatment Successful treatment Untreated Reference group without anemia
(n=1319) (n=2695) (n=3402) (n=13,274)
Demographic n (%) n (%) n (%) n (%)
Maternal demographic
Age at delivery (y)
<18 17 (1.3) 17 (0.6) 76 (2.2) 122 (0.9)
18−35 1078 (81.8) 2095 (77.9) 2594 (76.3) 10,112 (76.2)
≥35 222 (16.9) 579 (21.5) 730 (21.5) 3029 (22.8)
Nulliparous 325 (24.6) 956 (35.5) 867 (25.5) 3923 (29.6)
Marital status, married 872 (66.1) 2147 (79.7) 2205 (64.8) 10,064 (75.8)
2
BMI (kg/m )
<18.5 1 (0.1) 1 (0.0) 0 (0.0) 6 (0.1)
18.5−25.0 112 (9.1) 253 (9.9) 223 (7.0) 1063 (8.6)
25.0−30.0 372 (30.3) 929 (36.5) 922 (28.9) 4136 (33.3)
30.0−40.0 593 (48.3) 1135 (44.5) 1634 (51.2) 6019 (48.4)
>40.0 151 (12.3) 231 (9.1) 415 (13.0) 1215 (9.8)
High school education 962 (76.0) 2229 (85.4) 1994 (64.7) 8703 (73.1)
Race and ethnicity
African American 307 (23.5) 437 (16.4) 546 (16.3) 1170 (9.0)
Hispanic 726 (55.5) 1150 (43.3) 2122 (63.6) 7094 (54.4)
White 235 (18.0) 863 (32.5) 591 (17.7) 3899 (30.0)
Asian 37 (2.8) 201 (7.6) 75 (2.3) 838 (6.4)
b
Other 2 (0.2) 5 (0.2) 4 (0.1) 22 (0.2)
Income<$35,000 607 (56.9) 811 (35.3) 1759 (67.6) 4932 (48.4)
Insurance type
Federal 895 (70.0) 1236 (46.9) 2320 (72.0) 6947 (55.0)
Private 380 (29.7) 1391 (52.8) 880 (27.3) 5621 (44.5)
No insurance 4 (0.3) 7 (0.3) 22 (0.7) 47 (0.5)
Chi-square analysis was used to compare categorical variables.
BMI, body mass index.
a
Anemia was defined as receiving iron therapy or having a hemoglobin level below the American College of Obstetricians and Gynecologists cutoff by gestational age on admission to the labor and
delivery department; b Other included Native American and Pacific Islanders.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

95% CI, 0.98−1.21) compared with SGA neonate even after controlling for therapy (“successful treatment”) affected
patients without anemia. confounders. There was a considerable maternal and neonatal outcomes in
The evaluation of neonatal outcomes increase in the odds of delivering an LGA women with anemia with adequate pre-
revealed that refractory and untreated neonate for successfully treated and natal care in a large population-based
patients with anemia were considerably untreated patients. There was no differ- cohort. Successfully treated patients had
more likely to have a baby with TTN after ence in the odds of CNM. a considerable reduction in adjusted odds
controlling for preterm status, but this of PE and PTB. Similar to untreated
was not substantial for successfully Comment women with anemia, refractory women
treated patients (Tables 6 and 7; Figure 2). Principal findings had higher odds of most maternal out-
All anemic categories had a considerable This study was designed to evaluate comes, including PTB, CMM, PE, blood
reduction in the odds of delivering an whether successful response to iron transfusion, endometritis, and cesarean

March 2022 AJOG MFM 5

Original Research

TABLE 3
Maternal comorbidities by study group
Patients with anemiaa
Refractory to treatment Successful treatment Untreated Reference group without anemia
(n=1319) (n=2695) (n=3402) (n=13,274)
Variable n (%) or median (IQR) n (%) or median (IQR) n (%) or median (IQR) n (%) or median (IQR)
Maternal comorbidity
CHTN 86 (6.5) 106 (3.9) 248 (7.4) 632 (4.8)
Gestational DM 92 (7.0) 219 (8.2) 344 (10.2) 1399 (10.6)
Type 1 DM 5 (0.4) 10 (0.4) 35 (1.0) 92 (0.7)
Type 2 DM 21 (1.6) 27 (1.0) 124 (3.7) 334 (2.5)
DVT or pulmonary embolus 4 (0.3) 12 (0.5) 9 (0.3) 35 (0.3)
Hypothyroid 40 (3.0) 136 (5.1) 136 (4.0) 636 (4.8)
Hyperthyroid 2 (0.2) 15 (0.6) 17 (0.5) 57 (0.4)
Cardiac disease 17 (1.3) 29 (1.1) 17 (0.5) 118 (0.9)
Seizure disorder 10 (0.8) 32 (1.2) 29 (0.9) 110 (0.8)
Asthma 115 (8.7) 236 (8.8) 233 (6.9) 914 (6.9)
Endometriosis 18 (1.4) 53 (2.0) 28 (0.8) 167 (1.3)
Cancer 9 (0.7) 31 (1.2) 25 (0.7) 101 (0.8)
Psychiatric disease 176 (13.4) 368 (13.7) 366 (10.8) 1462 (11.1)
Cigarette
Ever use 184 (14.0) 435 (16.1) 347 (10.2) 1583 (11.9)
Current 14 (1.1) 12 (0.5) 16 (0.5) 76 (0.6)
Alcohol
Ever use 706 (53.6) 1717 (63.7) 1097 (32.6) 5723 (43.1)
Current 5 (0.4) 29 (1.1) 35 (1.0) 166 (1.3)
Marijuana
Ever use 89 (6.8) 201 (7.5) 145 (4.3) 687 (5.2)
Current 4 (0.3) 4 (0.2) 11 (0.3) 23 (0.2)
b
Illicit drugs
Ever use 10 (0.8) 21 (0.8) 21 (0.6) 86 (0.7)
Current 1 (0.1) 4 (0.2) 2 (0.1) 7 (0.1)
c
Composite comorbidity 1.0 (0.0−1.0) 1.0 (0.0−1.0) 1.0 (0.0−1.0) 1.0 (0.0−1.0)
Chi-square analysis was used to compare categorical variables.
CHTN, chronic hypertension; DM, diabetes mellitus; DVT, deep venous thrombosis; IQR, interquartile range; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; Tx, treatment.
a
Anemia was defined as receiving iron therapy or having a hemoglobin level below the American College of Obstetricians and Gynecologists cutoff by gestational age on admission to the labor and
delivery department; b Illicit drugs included heroin, methamphetamine, and cocaine; c Numeric composite of medical comorbidities was created, which included chronic hypertension, pregestational
and gestational DM, asthma, thyroid disease, seizure disorder, thromboembolism, cardiac disease, and psychiatric disease.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

delivery. All groups with anemia had a after adjusting for potential confound- therapy can reduce the odds of maternal
higher odds of PPH. Interestingly, there ers. Moreover, it uniquely demonstrated and neonatal complications.
was a considerable reduction in the odds that risk is dependent on (1) receipt of
of all women with anemia delivering an iron supplementation and (2) “success- Comparison with other studies and
SGA neonate. ful treatment” under a considerable and clinical implications
This study confirmed that pregnant clinically relevant increase in hemoglo- This study describes a comprehensive
persons with iron deficiency anemia bin. Specifically, we showed that suc- evaluation of outcomes associated with
have higher odds of perinatal morbidity cessful treatment of anemia with iron anemia and response to treatment and

6 AJOG MFM March 2022

 Original Research

TABLE 4
Composite and individual maternal outcomes by subgroup
Patients with anemiaa
Outcome Nonanemic
Outcome Refractory to treatment Successful treatment Untreated P valuea
(n=1319) (n=2695) (n=3402) (n=13,274)
n (%) n (%) n (%) n (%)
Cesarean delivery 473 (35.9) 874 (32.4) 1182 (34.8) 3858 (29.1) <.0001b
Noncephalic (1) 27 (2.1) 71 (2.6) 89 (2.6) 405 (3.1) —
RCD (13) 193 (14.6) 318 (11.8) 473 (13.9) 1368 (10.3)
Because of NRFHT (9) 69 (5.2) 203 (7.5) 167 (4.9) 647 (4.9)
Because of preeclampsia (11) 15 (1.1) 10 (0.4) 37 (1.1) 83 (0.6)
Failed IOL (7) 45 (3.4) 68 (2.3) 111 (3.3) 306 (2.3)
Abruption 4 (0.3) 15 (0.6) 28 (0.8) 77 (0.6) .1722
Preterm birth 145 (11.0) 136 (5.1) 410 (12.1) 1106 (8.3) <.0001b
<32.0 wk 19 (1.4) 12 (0.4) 65 (1.9) 157 (1.2)
32.0−36.9 wk 125 (9.5) 124 (4.6) 343 (10.1) 939 (7.1)
Composite maternal morbidity 377 (29.6) 517 (19.8) 935 (30.0) 2709 (21.7) <.0001b
Preeclampsia 136 (11.5) 147 (5.9) 385 (12.4) 1014 (8.3) <.0001b
GHTN 137 (10.4) 214 (7.9) 292 (8.6) 1098 (8.3) .0560
Sepsis 0 (0) 0 (0) 2 (0.1) 3 (0.0) .4610
Chorioamnionitis 67 (5.1) 159 (5.9) 222 (6.6) 666 (5.1) .0031b
Endometritis 16 (1.2) 19 (0.7) 40 (1.2) 85 (0.6) .0034b
Hysterectomy 6 (0.5) 5 (0.2) 10 (0.3) 14 (0.1) .0047b
ICU care 9 (0.7) 4 (0.2) 15 (0.6) 23 (0.2) .0002b
Blood transfusion 72 (5.5) 41 (1.5) 118 (3.5) 130 (1.0) <.0001b
Pulmonary edema 4 (0.3) 4 (0.2) 11 (0.3) 35 (0.3) .5925
Maternal death 0 (0) 0 (0) 0 (0) 0 (0) —
Postpartum readmission 14 (1.1) 22 (0.8) 32 (0.9) 101 (0.8) .5458
Intrapartum hemorrhage 51 (3.9) 74 (2.8) 89 (2.6) 262 (2.0) <.0001b
PPH 47 (3.6) 69 (2.6) 101 (3.0) 242 (1.8) <.0001b
ICU, intensive care unit; IOL, induction of labor; GHTN, gestational hypertension; PPH, postpartum hemorrhage; NRFHT, nonreassuring fetal heart tones; RCD, repeat cesarean delivery.
a
Chi-square analysis was used to compare all groups for significance; b Indicates significant difference.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

supplementation. A search for the terms not surprising given that most of our noted that our patients with anemia had
“iron therapy,” “iron supplementation,” patients had other risk factors for ane- higher odds of PTB as a group, but this
“iron deficiency anemia,” “maternal mia, such as concurrent medical effect differed in each subgroup. Our
morbidity,” “neonatal morbidity,” and comorbidities.17,18 Anemia has been patients with anemia who were success-
“maternal outcomes” in PubMed rang- associated with PTB in several previous fully treated had a considerably reduced
ing from 1980 revealed no similar study and smaller case-control studies.2,6−8 odds of PTB, which seemed to be
inclusive of the approach and data with The underlying true causes of this asso- related to a reduction in PTL, PPROM,
findings herein. ciation are poorly understood, but we and PE. Our refractory and untreated
Approximately one-third of our pop- speculate that it could be attributable to women had an increased odds of PTB,
ulation was composed of women who associated placental hypoxia and/or which may potentially be related to per-
were pregnant and with anemia; increased oxidative stress that is sistent increased oxidative stress and
although this was higher than the typi- hypothesized to lead to hypertensive placental hypoxia. However, this study
cal prevalence in North America, it was disorders and PTL and PPROM.19 We did not test that hypothesis.

March 2022 AJOG MFM 7

Original Research

FIGURE 2
aORs and cORs of adverse maternal and neonatal outcomes of the study group compared with the reference
population

There was a significant reduction in the odds of preterm birth (aOR, 0.59; 95% CI, 0.47−0.72) and preeclampsia (aOR, 0.75; 95% CI, 0.61−0.91)
among successfully treated patients.
aOR, adjusted odds ratio; CI, confidence interval; cOR, crude odds ratio.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

A single retrospective study in Hun- studies stratified by treatment response was refractory to therapy, and their PTB
gary reported that iron therapy mitigated nor success of treatment (which may be rate was increased. Therefore, iron alone
the increased risk of PTB in their popu- a reflection of the duration of therapy), did not seem to provide this benefit
lation with anemia.19 Conversely, a large which may explain our differing inter- unless occurring in the context of cor-
meta-analysis of 48 randomized con- pretations. It is possible that iron therapy recting anemia. Although one might
trolled trials found that iron therapy was would reduce placental and neonatal have expected to see an improvement in
not considerably associated with a reduc- stress by supporting hematopoiesis and a CNM with a reduction in the risk of
tion in PTB among women who were oxygen-carrying capacity. However, this PTB, this was not observed here as there
pregnant with anemia.20 However, benefit was not realized among women was no difference in CNM for our suc-
unlike our study, neither of these other who were pregnant and had anemia that cessfully treated group.

8 AJOG MFM March 2022

 TABLE 5
Crude and adjusted odds ratios for maternal outcomes by subgroup of patients with anemia
Refractory to treatment Successful treatment Untreated
Variable
Outcome cOR (95% CI) aOR (95% CI) cOR (95% CI) aOR (95% CI) cOR (95% CI) aOR (95% CI)
a a,b a a,b a
Preterm birth 1.36 (1.13−1.63) 1.44 (1.16−1.76) 0.58 (0.49−0.70) 0.59 (0.47−0.72) 1.51 (1.34−1.70) 1.45 (1.26−1.67)a,b
Because of PTL 0.96 (0.58−1.33) 0.85 (0.55−1.25)b 0.46 (0.33−0.64)a 0.49 (0.33−0.70)a,b 1.27 (1.04−1.55) 1.27 (0.88−1.43)b
b a a,b
Because of PPROM 1.16 (0.78−1.73) 1.11 (0.69−1.71) 0.66 (0.46−0.96) 0.60 (0.38−0.88) 1.21 (0.93−1.57) 1.27 (0.93−1.72)b
Because of preeclampsia 1.45 (1.04−2.05)a 1.75 (1.17−2.53)a,b 0.44 (0.29−0.66)a 0.47 (0.28−0.74)a,b 1.69 (1.36−2.11)a 1.60 (1.23−2.08)a,b
Because of NRFHT 1.85 (0.91−3.78) 2.38 (1.01−4.99)a,b 0.60 (0.26−1.41) 0.52 (0.16−1.33)b 0.56 (0.25−1.23) 0.67 (0.25−1.51)b
Composite maternal 1.51 (1.33−1.72)a 1.70 (1.44−2.01)a,b 0.89 (0.80−0.99)a 0.89 (0.77−1.03)b 1.54 (1.41−1.68)a 1.71 (1.52−1.92)a,b
a a,b a a,b a
Preeclampsia 1.43 (1.18−1.73) 1.54 (1.24−1.89) 0.69 (0.58−0.83) 0.75 (0.61−0.91) 1.56 (1.38−1.77) 1.44 (1.25−1.67)a,b
GHTN 1.29 (1.07−1.55)a 1.54 (0.63−3.28)b 0.96 (0.82−1.11) 0.75 (0.31−1.57)b 1.04 (0.91−1.19) 1.28 (0.68−2.28)b
d d d d
Sepsis 2.60 (0.43−15.58)
c c
Chorioamnionitis 1.01 (0.79−1.31) 0.93 (0.67−1.25) 1.18 (0.99−1.41) 1.11 (0.90−1.37) 1.32 (1.13−1.55)a 1.43 (1.18−1.72)a,c
a c c a
Endometritis 1.90 (1.11−3.25) 1.70(0.86−3.11) 1.10 (0.67−1.82) 0.93 (0.49−1.66) 1.85 (1.27−2.69) 1.78 (1.12−2.80)a,c
Hysterectomy 4.24 (1.63−11.06)a 6.63 (2.17−18.97)a,c 1.70 (0.61−4.71) 2.05 (0.55−6.42)c 2.92 (1.30−6.59)a 2.90 (1.02−7.92)a,c
ICU care 3.93 (1.81−8.51)a 2.31 (0.74−6.06)c 0.86 (0.30−2.48) 0.83 (0.19−2.54)c 2.55 (1.33−4.90)a 3.01 (1.39−6.40)a,c
Blood transfusion 5.84 (4.35−7.83)a 6.05 (4.29−8.48)a,c 1.56 (1.10−2.22)a 1.49 (0.97−2.23)c 3.63 (2.82−4.67)a 3.70 (2.76−4.98)a,c
c c
Pulmonary edema 1.14 (0.40−3.21) 1.78 (0.51−4.88) 0.56 (0.20−1.58) 0.55 (0.09−1.94) 1.23 (0.62−2.42) 1.25 (0.48−2.92)c
d d d d d d
Maternal death
Postpartum readmission 1.40 (0.80−2.45) 1.21 (0.56−2.33)c 1.07 (0.68−1.71) 1.20 (0.70−1.97)c 1.24 (0.83−1.85) 1.23 (0.74−1.98)c
Intrapartum hemorrhage 1.99 (1.47−2.70)a 2.03 (1.41−2.86)a,c 1.40 (1.08−1.82)a 1.43 (1.06−1.90)a,b 1.33 (1.05−1.70)a 1.31 (0.98−1.73)c
a a,c a a,c a
PPH 1.98 (1.44−2.73) 2.04 (1.40−2.89) 1.42 (1.08−1.86) 1.44 (1.05−1.94) 1.65 (1.30−2.08) 1.56 (1.18−2.05)a,c
Cesarean delivery 1.36 (1.21−1.54)a 1.53 (1.33−1.75)a,c 1.17 (1.07−1.28)a 1.09 (0.98−1.21)c 1.30 (1.20−1.41)a 1.38 (1.25−1.51)a,c
Abruption 0.52 (0.19−1.42) 0.54 (0.13−1.49)c 0.96 (0.55−1.67) 0.77 (0.35−1.50)c 1.42 (0.92−2.20) 1.97 (1.19−3.18)a,c
aOR, adjusted odds ratio; CI, confidence interval; cOR, crude odds ratio; GHTN, gestational hypertension; ICU, intensive care unit; NRFHT, nonreassuring fetal heart tones; PPH, postpartum hemorrhage; PPROM, preterm premature rupture of membranes; PTL, preterm

Original Research
labor.
a
Indicates significant difference; b Multivariate logistic regression was adjusted for confounders of age, nulliparity, education status, race and ethnicity, composite medical condition, and tobacco use; c Multivariate logistic regression was adjusted for confounders of age,
March 2022 AJOG MFM

nulliparity, education status, race and ethnicity, composite medical condition, preeclampsia, and tobacco use; d Not reported as minimum number of events not met.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.
9
Original Research

FIGURE 3
Incidence of PTB stratified by cause in anemia

There was a significant reduction in the odds of PTB for successfully treated patients. This can be attributed to a significant reduction in the odds of PTL,
PPROM, and PreE. aOther causes included placenta previa, placental abruption, oligohydramnios, growth restriction, fetal demise, and unknown.
NRFHT, nonreassuring fetal heart tones; PPROM, preterm premature rupture of membranes; PreE, preeclampsia; PTB, preterm birth; PTL, preterm labor.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

When assessing all patients with ane- deficiency anemia. It is possible that informed by our general approach to
mia, there was a considerable increase this association is related to the resolu- maternal care. Moreover, patients who
in the adjusted odds of PE, which has tion of anemia with subsequent were either untreated or refractory had
been demonstrated in previous improved oxygen-carrying capacity of an increased odds of cesarean delivery.
studies.2,21 Moreover, 1 Cochrane maternal blood and, therefore, reduced Although there was an increased odds of
review noted no significant difference in oxidative stress at the placental-mater- cesarean delivery among our successfully
the rate of PE among women receiving nal interface.23,24 treated patients, this was not substantial
daily iron therapy vs controls.22 How- In addition, there were several interest- after controlling for confounders. We
ever, this was applied to all pregnant ing findings from our secondary out- speculate that the association between
women regardless of anemia status, and comes. The rate of PPH was considerably anemia and cesarean delivery could also
many of the studies they included were increased in our cohort with anemia. arise from other factors or comorbidities,
from low-income countries, which may Interestingly, even successful treatment including decreased placental reserve
not apply to our population. These was associated with a 40% increase in the from recent or current anemia and
studies did not evaluate the effect of odds of intrapartum and PPH, even after increased maternal fatigue during labor.19
iron therapy on outcomes. We noted adjusting for confounders. The cause for Interestingly, the rate of blood transfu-
that there was a 25% reduction in the this association was likely multifactorial sion and hysterectomy were, like cesar-
adjusted odds of PE with successful iron and more likely to represent those at risk ean delivery, only considerably increased
therapy compared with the reference of PPH being repeatedly advised and among refractory and untreated patients
group. This benefit was not seen among counseled regarding benefits and efficacy with anemia after controlling for con-
refractory patients. Our findings sug- of iron therapy (given their at-risk sta- founders. This may have been, in part,
gested that the association between ane- tus). However, we could not decipher related to lower average starting hemo-
mia and PE may be reversed among from our database whether this had globin in addition to the increased risk of
women with adequate treatment of iron occurred and remains speculative but PPH and subsequent need for surgical

10 AJOG MFM March 2022

 Original Research

TABLE 6
Composite and individual neonatal outcomes for patients with anemia based on subgroup
Patients with anemia
P valuea
Refractory to treatment Successful treatment Untreated Nonanemic
(n=1319) (n=2695) (n=3402) (n=13,274)
Variable N z(%) or median (IQR) n (%) or median (IQR) n (%) or median (IQR) n (%) or median (IQR)
Gestational age at delivery (wk) 39.0 (39.0−40.4) 39.3 (38.6−40.0) 39.0 (37.7−39.9) 39.1 (38.3−40.0) <.0001b
Birthweight (g) 3269 (2945−3569) 3340 (3035−3655) 3315 (2975−3646) 3302 (2990−3625) <.0001b
TTN 37 (2.8) 55 (2.1) 87 (2.6) 200 (1.5) <.0001b
ROP 4 (0.3) 1 (0.0) 19 (0.6) 48 (0.4) .0074b
SGA 231 (17.5) 502 (18.6) 655 (19.3) 2914 (22.0) <.0001b
LGA 154 (11.7) 309 (11.5) 502 (14.8) 1391 (10.5) <.0001b
Composite neonatal morbidity 158 (12.3) 261 (10.0) 387 (11.7) 1,378 (10.7) .0537
5-min Apgar score of ≤3 2 (0.2) 0 (0) 10 (0.3) 19 (0.2) .0316b
RDS 53 (4.0) 65 (2.4) 154 (4.5) 383 (2.9) <.0001b
Vent or CPAP 2 (0.20) 3 (0.11) 10 (0.30) 32 (0.20) .4328
Seizures 4 (0.3) 2 (0.1) 0 (0) 18 (0.1) .0327b
Suspected or proven sepsis 101 (7.7) 196 (7.3) 223 (6.6) 962 (7.3) .4479
Stillbirth 2 (0.2) 1 (0.0) 5 (0.2) 11 (0.1) .4510
Neonatal death 3 (0.2) 5 (0.2) 10 (0.3) 27 (0.2) .7601
CPAP, continuous positive airway pressure; IQR, interquartile range; LGA, large for gestational age; RDS, respiratory distress syndrome; ROP, retinopathy of prematurity; SGA, small for gestational age;
TTN, transient tachypnea of newborn; Vent, mechanical ventilator.
a
Chi-square analysis was used to compare all groups for significance. Continuous variables were assessed using the Kruskal-Wallis test; b Indicates significant difference.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.

treatment of hemorrhage. Women with g/dL).2 The adjusted odds of delivering this study emphasized the need to ade-
untreated anemia were not only 3 times an SGA neonate were considerably quately treat and monitor response to
more likely to undergo hysterectomy but reduced among our untreated and iron therapy among women with ane-
also 3 times more likely to receive ICU refractory populations, both of which mia in pregnancy. This included appro-
care after adjusting for confounders. This had a median hemoglobin in this range. priate counseling on when to take the
suggested that anemia in pregnancy is Interestingly, the odds of SGA also supplement to optimize absorption and
associated with major maternal morbid- decreased among our gravidae with suc- reduce side effect profile. Given the
ity and increased healthcare costs, cessfully treated anemia, which may findings associated with refractory and
although the causal relationship cannot indicate these patients, like our other untreated anemia compared with suc-
be determined in the current study. groups with anemia, had a mild anemia cessfully treated women with anemia,
Multiple studies have noted an asso- in the third trimester of pregnancy. Our anemia unresponsive to oral iron sup-
ciation between both occurrence and results and previous studies suggested plementation warrants consideration
severity of maternal anemia in associa- that fetal growth and placental nutrient for evaluation of coexistent comorbid-
tion with risk of delivery of an SGA exchange may be contributory factors ities and potential responsiveness to IV
neonate.21,25,26 This may be because of in mild maternal iron depletion and iron therapy. Moreover, we cannot
the effect of considerably lower hemo- anemia.27 explain why women with sufficient pre-
globin on placental perfusion with oxy- We are unsure why our refractory natal care as included in our study did
genated blood, leading to decreased population did not respond to iron not receive treatment of iron deficiency
fetal growth.25 A recent study found therapy. This could be related to unmet anemia in the untreated group. This
that although the odds of SGA consid- challenges to compliance, an undiag- may be related to unmet challenges
erably increased among gravidae with nosed secondary micronutrient defi- with compliance with recommended
anemia compared with the reference ciency, or coexisting morbidities that care, but that is speculative. Our lack of
population, the odds considerably are associated with reduced absorption detailed knowledge emphasized the
decreased for women who were preg- (ie, celiac disease, Helicobacter pylori need to directly query women in preg-
nant and only manifested mild anemia infection, gastritis, or inflammatory nancy for important social determi-
(hemoglobin level between 9 and 10.9 bowel disease).28,29 The findings from nants of health that not only impact

March 2022 AJOG MFM 11

12 AJOG MFM March 2022

Original Research
TABLE 7
Neonatal crude and adjusted odds ratios for neonatal outcomes among patients with anemia
Refractory to treatment Successful treatment Untreated
Outcome
a a a
cOR (95% CI) aOR (95% CI) cOR (95% CI) aOR (95% CI) cOR (95% CI) aOR (95% CI)a
TTN 1.89 (1.32−2.69)b 1.82 (1.21−2.66)b,c 1.36 (1.01−1.84)b 1.32 (0.93−1.83) 1.72 (1.33−2.22)b 1.54(1.14−2.06)b,c
ROP 0.84 (0.30−2.32) 0.39 (0.09−1.13) 0.10 (0.01−0.74) 0.15 (0.01−0.73) 1.55 (0.91−2.64) 0.99 (0.51−1.83)
SGA 0.76 (0.65−0.88)b 0.71 (0.59−0.84)b 0.82 (0.74−0.91)b 0.82 (0.72−0.93)b 0.85 (0.78−0.94)b 0.71 (0.63−0.80)b
LGA 1.13 (0.95−1.35) 1.16 (0.94−1.43) 1.11 (0.97−1.26) 1.17 (1.01−1.37)b 1.48 (1.33−1.65)b 1.42 (1.24−1.63)b
Composite neonatal morbidity 1.17 (0.98−1.40) 1.08 (0.88−1.32) 0.93 (0.81−1.07) 1.02 (0.87−1.19) 1.11 (0.98−1.25) 0.98 (0.85−1.13)
5-min Apgar of ≤3 1.07 (0.25−4.59) 1.05 (0.06−5.95) — — 2.06 (0.96−4.44) 2.67 (0.84−7.93)
RDS 1.41 (1.05−1.89)b 1.04 (0.72−1.48) 0.83 (0.64−1.09) 1.03 (0.75−1.40) 1.60 (1.42−1.93)b 1.29 (1.01−1.63)
Vent or CPAP 0.63 (0.15−2.61) 0.33 (0.02−1.57) 0.46 (0.14−1.50) 0.56 (0.09−1.95) 1.22 (0.60−2.49) 0.53 (0.18−1.29)
Seizures 2.24 (0.76−6.62) 1.03 (0.16−3.83) 0.55 (0.13−2.36) 0.66 (0.10−2.44) — —
Sepsis 1.06 (0.86−1.32) 1.08 (0.84−1.37) 1.00 (0.86−1.18) 1.04 (0.87−1.24) 0.90 (0.77−1.04) 0.85 (0.71−1.01)
Stillbirth 1.83 (0.41−8.27) 3.71 (0.52−17.84) 0.45 (0.06−3.47) 1.07 (0.05−7.09) 1.77 (0.62−5.11) 2.51 (0.61−9.67)
Neonatal death 1.12 (0.34−3.69) 0.95 (0.22−2.87) 0.91 (0.35−2.37) 0.99 (0.28−2.70) 1.45 (0.60−2.99) 1.60 (0.68−3.52)
aOR was not reported as minimum number of events was not met.
aOR, adjusted odds ratio; CI, confidence interval; CPAP, continuous positive airway pressure; cOR, crude odds ratio; LGA, large for gestational age; RDS, respiratory distress syndrome; ROP, retinopathy of prematurity; SGA, small for gestational age; TTN, transient
tachypnea of newborn; Vent, mechanical ventilator.
a
Multivariate logistic regression was adjusted for confounders, which included race, education, nulliparity, preterm birth, preeclampsia, education, composite of maternal medical conditions, delivery route, and substance use; b XXX; c Adjusted only for preterm delivery
and delivery route as incidence was low and minimal number of events was not met for adjustment of all confounders.
Detlefs. The impact of response to iron therapy in pregnancy. Am J Obstet Gynecol MFM 2022.
 Original Research

their care but are also associated with therapy. However, these limitations spontaneous preterm birth. Am J Obstet Gyne-
perinatal morbidity.30 were reflective of the “real-world” clini- col 1991;164:59–63.
9. Reveiz L, Gyte GM, Cuervo LG, Casasbue-
cal setting, and cohort stratification and
nas A. Treatments for iron-deficiency anaemia
Research implications cohort size minimized these limitations. in pregnancy. Cochrane Database Syst Rev
Although this study demonstrated an 2011:CD003094.
association between several maternal Conclusions 10. Antony KM, Hemarajata P, Chen J, et al.
and neonatal outcomes with responsive- Using a large, diverse, population-based Generation and validation of a universal perina-
tal database and biospecimen repository: Peri-
ness to iron therapy for the treatment of cohort, we observed a considerable asso- Bank. J Perinatol 2016;36:921–9.
anemia in pregnancies, cause and effect ciation between unsuccessful treatment 11. Chu DM, Aagaard J, Levitt R, et al. Cohort
could not be determined, and several of iron deficiency anemia during preg- analysis of immigrant rhetoric on timely and reg-
questions remain. Further research is nancy and adverse perinatal outcomes, ular access of prenatal care. Obstet Gynecol
necessary to identify the mechanisms inclusive of PTB. Accurate diagnosis of 2019;133:117–28.
12. Mendez-Figueroa H, Chauhan SP, Tolcher
by which iron deficiency anemia con- the underlying cause of anemia during MC, et al. Peripartum outcomes before and
tributes to these pathologies. In addi- pregnancy, and enabling correct choice after hurricane Harvey. Obstet Gynecol
tion, future studies need to evaluate the and duration of treatment, may play a 2019;134:1005–16.
impact of iron deficiency without ane- key role in reducing maternal and neo- 13. Kotelchuck M. An evaluation of the Kess-
mia on similar outcomes, which was natal morbidities and mortalities. & ner Adequacy of Prenatal Care Index and a pro-
posed Adequacy of Prenatal Care Utilization
not assessed in this study. Moreover, Index. Am J Public Health 1994;84:1414–20.
this study reiterated some urgency in Supplementary materials 14. Alexander GR, Kotelchuck M. Quantifying
determining the optimal iron regimen Supplementary material associated with the adequacy of prenatal care: a comparison of
for iron deficiency anemia in pregnancy this article can be found in the online ver- indices. Public Health Rep 1996;111:408–19.
and evaluating why some compliant 15. Debiec KE, Paul KJ, Mitchell CM, Hitti JE.
sion at doi:10.1016/j.ajogmf.2022.100569.
Inadequate prenatal care and risk of preterm
women fail to respond to adequate ther- delivery among adolescents: a retrospective
apy. Lastly, our results provided further References study over 10 years. Am J Obstet Gynecol
rationalization for future clinical 1. Stevens GA, Finucane MM, De-Regil LM, 2010;203:122.e1−6.
research into how protocols for the et al. Global, regional, and national trends in 16. Duryea EL, Hawkins JS, McIntire DD,
management of iron deficiency anemia haemoglobin concentration and prevalence of Casey BM, Leveno KJ. A revised birth weight
total and severe anaemia in children and preg- reference for the United States. Obstet Gynecol
in pregnancy could be of benefit. 2014;124:16–22.
nant and non-pregnant women for 1995-2011:
a systematic analysis of population-representa- 17. Goonewardene M, Shehata M, Hamad A.
Strengths and limitations tive data. Lancet Glob Health 2013;1:e16–25. Anaemia in pregnancy. Best Pract Res Clin
Our study benefited from a large sample 2. Smith C, Teng F, Branch E, Chu S, Joseph Obstet Gynaecol 2012;26:3–24.
KS. Maternal and perinatal morbidity and mor- 18. Breymann C. Iron deficiency anemia in
size and a population-based approach,
tality associated with anemia in pregnancy. pregnancy. Semin Hematol 2015;52:339–47.
which created an ethnically and socio- Obstet Gynecol 2019;134:1234–44. 19. Ba
N, Puho

nhidy F, Acs
EH, Czeizel AE.
economically diverse cohort. The data- 3. Beckert RH, Baer RJ, Anderson JG, Jelliffe- Iron deficiency anemia: pregnancy outcomes
base had extensive demographic and Pawlowski LL, Rogers EE. Maternal anemia with or without iron supplementation. Nutrition
comorbidity information, which could and pregnancy outcomes: a population-based 2011;27:65–72.
study. J Perinatol 2019;39:911–9. 20. Haider BA, Olofin I, Wang M, et al. Anae-
be considered and accounted for in our
4. World Health Organization. Prevalence of mia, prenatal iron use, and risk of adverse preg-
analysis. However, 1 limitation was that anaemia in pregnant women (aged 15−49) (%). nancy outcomes: systematic review and meta-
our uniform working definition of ane- 2021. Available at: https://www.who.int/data/ analysis. BMJ 2013;346:f3443.
mia was based on nadir hemoglobin on gho/data/indicators/indicator-details/GHO/prev- 21. Young MF, Oaks BM, Tandon S, Martorell
admission to labor and delivery and/or alence-of-anaemia-in-pregnant-women-(-). Ac- R, Dewey KG, Wendt AS. Maternal hemoglobin
cessed December 6, 2021. concentrations across pregnancy and maternal
the use of ferrous sulfate. It was
5. American College of Obstetricians and and child health: a systematic review and meta-
assumed that women treated with iron Gynecologists. ACOG Practice Bulletin No. 95: analysis. Ann N Y Acad Sci 2019;1450:47–68.
had iron deficiency as iron studies were anemia in pregnancy. Obstet Gynecol 22. Pen ~a-Rosas JP, De-Regil LM, Garcia-
not abstracted on every patient within 2008;112:201–7. Casal MN, Dowswell T. Daily oral iron supple-
this database. In our experience, nearly 6. Levy A, Fraser D, Katz M, Mazor M, Sheiner mentation during pregnancy. Cochrane Data-
E. Maternal anemia during pregnancy is an base Syst Rev 2015:CD004736.
all women with anemia in our cohort
independent risk factor for low birthweight and 23. Yoo JH, Maeng HY, Sun YK, et al. Oxida-
have abnormal iron studies suggestive preterm delivery. Eur J Obstet Gynecol Reprod tive status in iron-deficiency anemia. J Clin Lab
of iron deficiency. However, this may Biol 2005;122:182–6. Anal 2009;23:319–23.
have incorrectly classified some women 7. Lin L, Wei Y, Zhu W, et al. Prevalence, risk 24. Bozkaya VO, € Oskovi-Kaplan ZA, Erel O,
within the cohort. In addition, there factors and associated adverse pregnancy out- Keskin LH. Anemia in pregnancy: it’s effect on
comes of anaemia in Chinese pregnant women: oxidative stress and cardiac parameters. J
was no compliance or dosing data, and
a multicentre retrospective study. BMC Preg- Matern Fetal Neonatal Med 2021;34:105–11.
some providers likely failed to address nancy Childbirth 2018;18:111. 25. Bakacak M, Avci F, Ercan O, et al. The
challenges, which limited our patient’s 8. Klebanoff MA, Shiono PH, Selby JV, Trach- effect of maternal hemoglobin concentration
ability to comply with recommended tenberg AI, Graubard BI. Anemia and on fetal birth weight according to trimesters.

March 2022 AJOG MFM 13

Original Research

J Matern Fetal Neonatal Med 2015;28:2106– 29. Hershko C, Camaschella C. How I treat Received Nov. 3, 2021; revised Dec. 22, 2021;
10. unexplained refractory iron deficiency anemia. accepted Jan. 9, 2022.
€ u
26. Yildiz Y, Ozg € E, Unlu SB, Salman B, Eyi Blood 2014;123:326–33. The authors report no conflict of interest.
EG. The relationship between third trimester 30. Wang E, Glazer KB, Howell EA, Janevic TM. The authors gratefully acknowledge the support of the
maternal hemoglobin and birth weight/ Social determinants of pregnancy-related mortal- National Institutes of Health’s (NIH) National Institute of
length; results from the tertiary center in Tur- ity and morbidity in the United States: a system- Nursing Research (grant number R01NR014792;
key. J Matern Fetal Neonatal Med 2014; atic review. Obstet Gynecol 2020;135:896–915. K.M.A.), the NIH’s Eunice Kennedy Shriver National Insti-
27:729–32. tute of Child Health and Human Development (grant num-
27. Drukker L, Hants Y, Farkash R, Ruchlemer ber R01HD091731), the Burroughs Wellcome Fund
R, Samueloff A, Grisaru-Granovsky S. Iron defi- Author and article information Preterm Birth Initiative (K.M.A.), and the March of Dimes
ciency anemia at admission for labor and delivery From the Division of Maternal-Fetal Medicine, Depart- Preterm Birth Research Initiative (K.M.A.). The sponsors
is associated with an increased risk for cesarean ment of Obstetrics and Gynecology, Baylor College of did not have any role in the study design, data collection,
section and adverse maternal and neonatal out- Medicine and Texas Children’s Hospital, Houston, TX analysis, interpretation, or writing of the manuscript or
comes. Transfusion 2015;55:2799–806. (Drs Detlefs, Jochum, Salmanian, McKinney and decision to submit for publication.
28. Weyermann M, Rothenbacher D, Gayer L, Aagaard); Departments of Molecular and Human Genet- This study was presented as a poster at the 40th
et al. Role of Helicobacter pylori infection in iron ics, Baylor College of Medicine, Houston, TX annual meeting of the Society for Maternal-Fetal Medi-
deficiency during pregnancy. Am J Obstet (Dr Aagaard); Molecular and Cell Biology, Baylor College cine, Dallas, TX, February 3−8, 2020.
Gynecol 2005;192:548–53. of Medicine, Houston, TX (Dr Aagaard). Corresponding author: Kjersti M. Aagaard, MD, PhD
[email protected]

14 AJOG MFM March 2022