Routine Iron/Folate Supplementation during Pregnancy: Effect on Maternal Anaemia and Birth Outcomes

Authors


Zulfiqar A. Bhutta, Husein Lalji Dewraj Professor & Head, Division of Women & Child Health, The Aga Khan University, Karachi 74800, Pakistan. E-mail: zulfiqar.bhutta@aku.edu

Abstract

Iron deficiency is the most common nutritional deficiency globally. Children and women of reproductive age are at a particular risk of iron deficiency. Anaemia during pregnancy is a specific risk factor for adverse maternal and perinatal outcomes. The objective of this review was to assess the impact of routine iron supplementation on maternal anaemia and perinatal outcomes. A literature search was conducted for published randomised and quasi-randomised trials on PubMed and the Cochrane Library. Only those studies were included in the review that assessed the preventive effect of iron supplementation during pregnancy. Data from selected studies were double abstracted in a standardised excel sheet. The studies were graded according to study design, limitations, intervention specifics and outcome effects. Meta-analyses were conducted where data were available from more than one study for an outcome. After screening 5209 titles, 30 studies were selected for inclusion in this review. Daily iron supplementation resulted in 69% reduction in incidence of anaemia at term in the intervention group compared with control [relative risk (RR) 0.31 [95% confidence interval (CI) 0.22, 0.44]] and 66% reduction in iron deficiency anaemia at term (RR 0.44 [95% CI 0.28, 0.68]; random model) compared with no intervention/placebo. The quality grade for these outcomes was that of ‘moderate’ level. Routine daily iron supplementation during pregnancy resulted in a significant reduction of 20% in incidence of low birthweight in the intervention group compared with control (RR 0.80 [95% CI 0.71, 0.90]). Preventive iron supplementation during pregnancy has a significant benefit in reducing incidence of anaemia in mothers and low birthweight in neonates.

According to World Health Organization, around 2 billion people, amounting to over 30% of the world's population are anaemic. Iron deficiency is the most common cause of anaemia and is the most widespread nutritional disorder in the world. It is not only prevalent in developing countries but also in developed countries (Figure 1). The most commonly affected populations are children and women.5 It is estimated that 56 million pregnant women (41.8% of the total) are affected with anaemia globally, largely because of iron deficiency.6 In developing countries, this proportion can be as high as 85% in South-East Asia, making pregnant mothers especially susceptible to increased risk of mortality and reduced work capacity. It is the poorest, most vulnerable and least educated who are disproportionately affected by iron deficiency, and it is this group that stands to gain the most by its reduction.6

Figure 1.

Worldwide prevalence of anaemia: pregnant women. Source: World Health Organization, World prevalence of anaemia 1993–2005.

Iron deficiency anaemia is a risk factor for perinatal complications like pre-eclampsia, low birthweight, prematurity and perinatal mortality.7,8 Earlier studies have provided strong evidence to show that iron supplementation with or without folic acid results in a significant reduction in the incidence of anaemia during pregnancy.9,10 Recently it has been shown that iron supplementation during pregnancy can also help reduce perinatal mortality.11 The purpose of this review was to synthesise the up-to-date evidence on effects of iron supplementation during pregnancy.

Methods

Searching

We searched PubMed, the Cochrane Library and World Health Organization regional databases, to identify studies of routine iron supplementation with or without folic acid during pregnancy and included publications in any language. The last date of search was 2nd June 2011. Following search strategy was used for PubMed: (‘Iron’[Mesh] OR ‘Folic Acid’[Mesh] OR iron OR folic acid OR folate) AND (‘Anemia’[Mesh] OR ‘Anemia, Iron-Deficiency’[Mesh] OR anemia OR neonate OR Low birth weight OR Preterm Birth OR mortality) AND (pregnancy OR maternal). Additional studies were obtained through hand search of references from identified studies and previous reviews.

Inclusion criteria

Following were the inclusion criteria for this review

  • • Type of studies. Randomised and quasi-randomised trials.
  • • Type of intervention. Prenatal iron or iron + folic acid for prevention of anaemia in pregnant women at any gestational age.
  • • Type of comparison. ‘Placebo’ or ‘no intervention’. In case where studies assessed the effects of multiple combinations of vitamins and minerals, those groups were included where iron/iron–folate was the only difference among the study groups (arms).
  • • Type of outcomes. Maternal: anaemia at term (Hb < 110 g/L), iron deficiency anaemia at term (Hb < 110 g/L and at least one additional laboratory indicator), maternal mortality, postpartum haemorrhage, need for blood transfusion and pre-eclampsia. Infant: low birthweight, weight at birth, preterm birth (in grams), perinatal death and congenital malformation.
  • • Location. Studies from both developed and developing countries were included.

Exclusion criteria

  • • Studies of peri-conceptional or postpartum iron/iron–folate supplementation were excluded.
  • • Studies of fortification of iron/iron–folate in food or studies in which iron was given in forms other than oral supplements like powder or parenteral form were excluded.
  • • Studies where iron was given in combination with multiple micronutrients supplementation were excluded.
  • • Studies where iron/iron–folate was given as a therapeutic intervention for anaemic pregnant women were also excluded.
  • • Studies where clinical outcomes were not reported were excluded.

Abstraction, analyses and summary measures

All studies that met final inclusion and exclusion criteria were double-data abstracted into a standardised form for each outcome of interest. We abstracted key variables with regard to the study identifiers and context, study design and limitations, intervention specifics, and outcome effects. Each study was assessed and graded according to the Child Health Epidemiology Review Group adaptation of the Grading of Recommendations, Assessment, Development and Evaluation (GRADE).12,13 Studies received an initial score of high if they were randomised controlled trials or cluster randomised controlled trials. The grade was decreased by one point for each study design limitation like inadequate methods of sequence generation, allocation concealment and attrition >20% etc. In addition, studies reporting an intent-to-treat analysis or with statistically significant strong levels of association (>80% reduction) received 0.5–1.0 grade increase. Any study with a final grade of very low was excluded on the basis of inadequate study quality.

Meta-analyses were generated where data were available from more than one study for an outcome. For cluster randomised trials, we used the stated cluster adjusted relative risk (RR) and 95% confidence interval (CI), irrespective of the method used. We adjusted the results for cluster design if not stated in the study. The assessment of statistical heterogeneity among trials was done by visual inspection, that is the overlap of the CI among the studies, and by the chi-square (P-value) of heterogeneity in the meta-analyses. A low P-value (<0.10) or a large chi-squared statistic relative to its degree of freedom was considered as providing evidence of heterogeneity. The I2 values were also looked into, and an I2 greater than 50% was taken to represent substantial heterogeneity. In situations of substantial heterogeneity being present, causes were explored by sensitivity analysis and random effects model was used. Results of pooled estimates are described as RR with 95% CI. All meta-analyses were conducted using the software RevMan version 5.14

The ‘overall’ evidence was summarised according to the GRADE criteria. The qualitative evaluation of the overall (pooled) evidence was based on the volume and consistency of the evidence across studies, the size of pooled RR and the strength of the statistical evidence for an association between the intervention and the health outcome as reflected in the P-value.

Results

Literature search

The search generated 5209 hits on PubMed and 290 in Cochrane Library that were screened and after removing duplicates, 77 abstracts were preliminarily selected. After a detailed review of full text, 30 studies were selected for inclusion in the review.1–5,12–34Figure 2 shows the flow diagram of literature search whereas the Web Table S1 lists the excluded studies with pertinent information.

Figure 2.

Flow chart of the literature search.

Characteristics of included studies

Studies were conducted both in developed and in developing countries. Iron supplementation started in all studies no later than 28 weeks of gestation. Dose ranged from 20 to 300 mg per day. There were 18 studies that reported anaemia at term.11,38–45 Fourteen of these evaluated effects of daily iron supplementation alone,4,7,11,12,14–16,22,24,26–28,31–33,35,41 while in four studies the combined effect of daily iron and folic acid was evaluated. There were 12 studies that reported data on low birthweight.

Maternal and infant outcomes

A combined analysis of 18 studies that included studies with any iron vs. no iron showed a significant reduction of 69% in the intervention group compared with control (RR 0.31 [95% CI 0.22, 0.44]). A subgroup analysis for iron supplementation alone and in combination with folic acid showed similar results (Figure 3). There was no statistically significant difference between intermittent iron/iron folic acid supplementation and daily supplementation based on data from three studies (RR 1.61 [95% CI 0.82, 3.14]; random model, data not shown). Table 1 gives details of other maternal outcomes.

Figure 3.

Effect of iron/iron–folate supplementation during pregnancy on anaemia at term. M-H, Mantel Haenszel.

Table 1.  Effect of any iron vs. no iron supplementation during pregnancy on maternal outcomes
OutcomeNo. of studiesSummary estimate [95% CI]Analysis modelStatistical heterogeneity
  • a

    Only one study contributed data, zero events in rest of studies.

  • RR, relative risk, CI, confidence interval.

Iron deficiency anaemia at term7RR 0.44 [0.28, 0.68]RandomI2 = 85%, P < 0.00001
Severe anaemia at terma11RR 4.83 [0.23, 99.88]FixedI2 = 0%, P = 0.0
Severe anaemia at any time during second or third trimester13RR 0.25[0.03, 2.48]RandomI2 = 66%, P = 0.05
Postpartum haemorrhage6RR 0.88 [0.57, 1.36]FixedI2 = 18%, P = 0.30
Transfusion provided3RR 0.61 [ 0.38, 0.96]FixedI2 = 0%, P = 0.0
Pre-eclampsia2RR 2.58 [0.81, 8.22]FixedI2 = 0%, P = 0.43
Maternal death1No events in any study group

Combined results from twelve studies showed that iron supplementation during pregnancy has a significant effect on incidence of low birthweight (RR 0.80 [95% CI 0.71, 0.90]) (Figure 4). Table 2 shows results of some of other infant outcomes. Iron supplementation had a significant effect on mean birthweight (mean difference 42.18 [9.27, 75.09]). There was no effect on incidence of preterm birth or that of small for gestation age birth.

Figure 4.

Effect of any iron vs. no iron supplementation during pregnancy on incidence of low birthweight. M-H, Mantel Haenszel.

Table 2.  Effect of any iron vs. no iron supplementation during pregnancy on infant outcomes
OutcomeSubgroupNo. of studiesSummary estimate [95% CI]Analysis modelStatistical heterogeneity
  1. RR, relative risk; CI, confidence interval; MD, mean difference.

Birthweight (g)All studies13MD 42.18 [9.27, 75.09]RandomI2 = 54%, P = 0.01
Iron alone10MD 45.42 [−3.78, 94.62]RandomI2 = 61%, P = 0.006
Iron + folic acid3MD 34.94 [−1.69, 71.58]RandomI2 = 31%, P = 0.24
Preterm birthAll studies12RR 0.94 [0.84, 1.06]FixedI2 = 40%, P = 0.09
Iron alone8RR 0.90 [0.77, 1.06]FixedI2 = 30%, P = 0.20
Iron + folic acid4RR 0.99 [0.84, 1.18]FixedI2 = 65%, P = 0.09
Perinatal mortalityAll studies4RR 0.82 [00.65, 1.05]FixedI2 = 0%, P = 0.0
Iron alone3RR 0.83 [0.59, 1.18]FixedI2 = 0%, P = 0.0
Iron + folic acid1RR 0.82 [0.58, 1.15]FixedI2 = 0%, P = 0.0
Small-for-gestational ageAll studies6RR 0.94 [0.81, 1.09]RandomI2 = 69%, P = 0.007
Iron alone4RR 0.87 [0.58, 1.30]RandomI2 = 75%, P = 0.007
Iron + folic acid2RR 0.95 [0.80, 1.13]RandomI2 = 74%, P = 0.05

Table 3 gives summary estimates and grading according to GRADE criteria.

Table 3.  Quality assessment of trials of iron folate supplementation during pregnancy
No. of studies (ref.)Quality assessmentDirectnessSummary of findings
No. of eventsRelative risk [95% CI]
DesignLimitationsConsistencyGeneralisability to population of interestGeneralisability to intervention of interestInterventionControl
  1. RR, relative risk; CI, confidence interval; RCT, randomised controlled trial.

Daily iron vs. no intervention/placebo: Anaemia at term: Moderate outcome-specific quality
18RCT/quasi-RCTStudies with unclear or inadequate sequence generation and high loss to follow-upAll studies show direction of benefit, but high heterogeneityStudies from both developed and developing countriesYes218495RR (random) 0.31 [0.22, 0.44]
Daily iron vs. no intervention/placebo: Low birthweight: High outcome-specific quality
12RCT/quasi-RCTStudies with unclear or inadequate sequence generation and high loss to follow-up8 studies show direction of benefit, with borderline heterogeneityStudies from both developed and developing countriesYes590749RR (fixed) 0.80 [0.71, 0.90]
Daily iron folic acid vs. no intervention/placebo: Preterm birth: Moderate outcome-specific quality
12RCT/quasi-RCTsHigh loss to follow-up and unclear sequence generation in few of the included studies5 studies show direction of benefit; I2 = 40%Studies from both developed and developing countriesYes468508RR (fixed) 0.94 [0.84, 1.06 ]
Daily iron and folic acid vs. no intervention/placebo: Small-for-gestational age: Moderate outcome-specific quality
6RCT/quasi-RCTAllocation concealment unclear and blinding inadequate in few of the included studies4 studies show direction of benefit; I2 = 69%Studies from both developed and developing countriesYes10141122RR (random) 0.94 [0.81, 1.09]
Daily iron and folic acid vs. no intervention/placebo: Perinatal mortality: Low outcome-specific quality
4RCT/quasi-RCTStudies with unclear or inadequate sequence generation and high loss to follow-up2 studies show direction of benefit; I2 = 0%Studies from both developed and developing countriesYes108146RR (fixed) 0.82 [0.65, 1.05]

Discussion

It has been shown that iron supplementation alone or in combination with folic acid is associated with the well-being of the mother and fetus.19,30,44 It leads to a significant reduction in anaemia incidence during pregnancy and, thus, plays a vital role in reducing maternal morbidity and mortality. The results of this review for effect of iron supplementation on maternal anaemia are consistent with those of an earlier review by our group46 and the Cochrane review by Pena Rosas and Viteri9 that also showed a significant reduction in incidence of anaemia and iron deficiency anaemia at term. The results for low birthweight are however different because of the addition of a study from China.11 The updated results showed that iron supplementation during pregnancy have a significant effect on incidence of low birthweight (Figure 4). A similar corresponding significant increase in birthweight (g) was also observed (mean difference 42.18 [95% CI 9.27, 75.09]). There was no significant effect on incidence of preterm birth or that of small-for-gestational age birth. The latter effect may be due to the fact there were not enough study to show a statistically significant effect.

There was significant statistical heterogeneity in the pooled analyses of effect of daily iron supplementation vs. control for most of the outcomes. This can be expected as the included studies have been conducted in different setting with differences in baseline prevalence of anaemia. An important observation to make is that the direction of effect in all the studies was in the same direction. This was demonstrated by the subgroup analysis in our previous paper11 where an analysis based on baseline anaemia status showed less heterogeneous and less prominent results for non-anaemic women compared with that of a mixed population that had more heterogeneous and more prominent results. This supports a strong biological effect in favour of the intervention and also indicates that the effects of iron supplementation depend on the degree of baseline anaemia in the study population.

Subgroup analyses showed that there was not much difference in effects between supplementation with iron alone and iron–folate combinations. The effect sizes were similar in both instances but the CI were wider than for iron/folate supplementation, mainly because of a lesser number of studies and events. We did not look at side-effects of oral iron supplements however it has been shown previously that iron supplementation is associated with any side-effects compared with control. The most commonly reported side-effects are gastrointestinal symptoms.9

Notwithstanding the significant heterogeneity of findings, the current and previous reviews confirm the efficacy of routine iron supplementation during pregnancy for prevention of anaemia and low birthweight. Further research should focus on delivery mechanisms of iron supplementation during pregnancy. It has been proposed that weekly iron and folic acid supplementation is in synchrony with the turnover of mucosal cells and may be a promising substitute for daily iron supplementation.46 The World Health Organization advocates that weekly iron and folic acid supplementation should be considered a strategy for the prevention of iron deficiency in population groups where the prevalence of anaemia is above 20% among women of reproductive age and mass fortification programmes of staple foods with iron and folic acid are unlikely to be implemented within 1–2 years. Weekly dosage may have benefits of reduced side-effects and increased compliance, but the overall evidence is mixed.47–54 Further field randomised controlled trials are needed to establish the efficacy of weekly supplementation compared with a daily regimen.

Conflicts of interest

The authors declare no conflicts of interest.

Ancillary