Effects and safety of preventive oral iron or iron+folic acid supplementation for women during pregnancy

 

Background

Iron-deficiency anaemia, the late manifestation of chronic iron deficiency, is thought to be the most common nutrient deficiency among pregnant women (WHO 1992). Iron deficiency involves an insufficient supply of iron to the cells following depletion of the body's reserves (Viteri 1998) and its main causes are a diet poor in absorbable iron, an increased requirement for iron (e.g. during pregnancy), a loss of iron due to parasitic infections, particularly hookworm, and other blood losses (Crompton 2002; INACG 2002a).

Although haemoglobin and haematocrit tests are commonly used to screen for iron deficiency, low haemoglobin or haematocrit values are not specific to iron deficiency. While iron deficiency is the most common cause of anaemia, other causes such as acute and chronic infections that cause inflammation; deficiencies of folate and of vitamins B₂, B₁₂, A, and C; and genetically inherited traits such as thalassaemia and drepanocytosis may be independent or superimposed causal factors (WHO 2001). According to the most recent estimate, the global prevalence of anaemia among pregnant women is 41.8% (McLean 2007). According to criteria of U.S. Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO), anaemia during pregnancy should be diagnosed if a woman's haemoglobin (Hb) concentration is lower than 110 g/L during the first or third trimester, or lower than 105 g/L during the second trimester. Iron-deficiency anaemia is defined as anaemia accompanied by depleted iron stores and signs of a compromised supply of iron to the tissues (WHO 2001).

Iron deficiency in non-pregnant populations can be measured quite precisely using laboratory tests such as serum ferritin, serum iron, transferrin, transferrin saturation and transferrin receptors. However, these tests are often not readily available and their results may be of limited value in some settings and under some conditions, particularly among pregnant women and where different infections (e.g. malaria, HIV/AIDS, vaginosis) are prevalent. Furthermore, the results of those tests do not correlate closely with one another because each reflects a different aspect of iron metabolism. Serum ferritin concentration is an indicator of iron reserves. During pregnancy, however, serum ferritin levels as well as levels of bone marrow iron fall even in women who ingest daily supplements with high amounts of iron, which casts doubts about their true significance in pregnancy and suggests the need to review cut-off values (Puolakka 1980; Romslo 1983; Svanberg 1975). Currently, a serum ferritin concentration of less than 12 µg/L in adults is accepted as an indication of depleted iron stores, even among pregnant women. Interestingly, the nadir of maternal serum ferritin occurs by week 28, before higher iron demands are believed to occur, a decrease only partially explained by the normal plasma volume expansion that occurs during pregnancy (Taylor 1982). Other indicators of iron status are also distorted during pregnancy, even among women who are administered supplements containing 200 mg of iron daily (Puolakka 1980). Recently it has been suggested that the ratio of serum transferrin receptors to serum ferritin, a seemingly good estimator of iron nutrition among non-pregnant adults, could also be used to estimate the iron nutritional status of pregnant women. However, this ratio does not seem to differentiate clearly between an iron-deficient and an iron-sufficient population of pregnant women (Cook 2003). An important consistent finding in all the studies referenced above is that pregnant women who received iron supplementation had differences in indicators of iron nutritional status during pregnancy in comparison with those who did not (i.e. their Hb, iron, transferrin saturation and ferritin levels declined less, and low serum transferrin and erythrocyte protoporphyrins among them increased less). There is a need to better define the distribution of serum transferrin receptors during pregnancy in populations that are not iron deficient (Nair 2004), as has been done in some populations from better environments (Milman 2007).

Recently, a WHO and CDC Technical Consultation on the Assessment of Iron Status at the Population Level concluded that Hb and ferritin were the most efficient combination of indicators for monitoring change in the iron status of a population as a consequence of iron supplementation (WHO/CDC 2005). Unfortunately, only two of the very differing studies on pregnant women were included, and only one of them demonstrated changes with iron supplementation. The use of multiple indicators (Hb, ferritin and serum transferrin receptors) is useful for population-based assessments of iron-deficiency anaemia, when this is feasible. However, further studies will be needed to determine how levels of iron nutrition and its indicators among women change during pregnancy under different circumstances, and the extent to which the iron nutritional status of pregnant women influences maternal health and pregnancy outcomes in different populations.

The consequences of iron-deficiency anaemia are serious and can include diminished intellectual and productive capacity (Hunt 2002) and possibly increased susceptibility to infections (Oppenheimer 2001). During pregnancy, low Hb levels, indicative of moderate (between 70 and 90 g/L) or severe (less than 70 g/L) anaemia, are associated with increased risk of maternal and child mortality and infectious diseases (INACG 2002b). The lowest rates of low birthweight and premature birth appear to occur when maternal Hb levels are between 95 and 105 g/L during the second trimester of gestation (Steer 2000; Murphy 1986) and between 95 and 125 g of Hb/L at term (Hytten 1964; Hytten 1971). However, the results of several studies suggest that near-term Hb levels below 95 g/L or even below 110 g/L may be associated with low birthweight, heavier placentas and increased frequency of premature births (Garn 1981; Godfrey 1991; Kim 1992; Klebanoff 1989; Klebanoff 1991; Murphy 1986). There is little doubt that maternal Hb levels below 95 g/L before or during the second trimester of gestation are associated with increased risk of giving birth to a low birthweight infant and with premature delivery. Favourable pregnancy outcomes occur 30% to 45% less often in anaemic mothers, and probably their infants have less than one-half of normal iron reserves (Bothwell 1981). Unfortunately, the time between birth and umbilical cord clamping has not been considered in the estimates of impact of maternal iron status and anaemia on the infant's iron reserves, even though delayed cord clamping has been shown to provide significant iron reserves to infants (Chaparro 2006; Chaparro 2007; Mercer 2001; van Rheenen 2004). Iron deficiency adversely affects the cognitive performance and development and physical growth of infants (WHO 2001) even in the long term (Lozoff 2006), and moderate or severe iron deficiency during infancy has been shown to have irreversible cognitive effects (Gleason 2007). Haemoglobin levels greater than 130 g/L at sea level have also been associated with negative pregnancy outcomes (Hytten 1964; Hytten 1971; Murphy 1986; Scholl 1997; Steer 2000).

Large epidemiologic retrospective studies (Murphy 1986; Steer 2000; Xiong 2000) and one prospective study in China (Zhou 1998) have shown that both low and high prenatal haemoglobin concentrations are associated with increased risks for premature delivery and low birthweight. In fact, the incidence of these negative consequences increases dramatically when women's haemoglobin levels, at sea level, are below 95 to 105 g/L at any time in pregnancy or above 130 to 135 g/L after mid-pregnancy. A prospective study in Mexico has shown associations between prenatal daily iron supplement intake at recommended doses to be associated with high haemoglobin concentrations and the risk for both low birthweight and premature delivery (Casanueva 2003a). Ziaei 2007 also showed that women whose haemoglobin concentration at gestational weeks 32 to 36 was >132 g/L had more low birth weight babies and also higher blood pressure than women with lower haemoglobin concentrations. Unfortunately any woman considered anaemic were excluded from the study. Observational studies have shown that among iron supplemented pregnant women, and particularly among those who are anaemic early in pregnancy, a failure of haemoglobin and/or ferritin levels to decline during the second and third trimesters and overall high ferritin levels during pregnancy, not due to infection, are associated with adverse pregnancy outcomes. However, when some confounding factors are controlled for, the association between high serum ferritin concentrations and the risk for premature delivery was not significant (Scholl 1998; Scholl 2000; Scholl 2005).

The association between iron deficiency without anaemia and adverse perinatal outcomes is less clear, although some studies have shown an association between iron deficiency to be associated with inadequate pregnancy weight gain, decreased defence against infections, preterm delivery and low birthweight (Garn 1981; Kandoi 1991; Prema 1982; Scholl 1992).

Interventions to control iron deficiency and iron-deficiency anaemia include iron supplementation and iron fortification, health and nutrition education, control of parasitic infections, and improvement of sanitation (INACG 1977). Delayed clamping of the umbilical cord has also been shown to be effective in preventing iron deficiency among infants and young children (Chaparro 2007; Mercer 2001; van Rheenen 2004).

The results of some studies suggest that the amount of iron that can be absorbed from diet alone is insufficient to cover women's increased iron requirements during pregnancy except when women can draw enough iron from pre-pregnancy iron reserves. The Institute of Medicine recommends that women consume 27 mg/day of iron during pregnancy (IOM 2001). Because most women would need additional iron as well as sufficient iron stores to prevent iron deficiency (Bothwell 2000), direct iron supplementation for pregnant women has been used extensively in most low- and middle-income countries as an intervention to prevent and correct iron deficiency and anaemia during pregnancy. It has been recommended that iron supplements also contain folic acid, an essential B-vitamin, because of the increased requirements of pregnancy, due to the rapidly dividing cells in the fetus and elevated urinary losses. Other micronutrients for which deficiencies are documented may justify their addition to the supplementation formula.

Several studies have shown that iron supplementation, with or without folic acid during pregnancy, results in a substantial reduction in women's risk of having haemoglobin levels less than 100 g/L in late pregnancy, at delivery and six weeks postpartum (Mahomed 2000a; Mahomed 1997; Villar 2003). However, the overall impact of iron supplementation interventions under field conditions has been limited and the effectiveness of these interventions has been questioned (Beaton 1999). The limited success has been attributed to inadequate infrastructure and poor compliance (Mora 2002), although few studies have evaluated this issue adequately. The effectiveness of iron supplementation for pregnant women has been evaluated mostly in terms of improvement in haemoglobin concentration, rather than improvements in maternal or infant health (Beaton 2000). This narrow scope may have been an important omission in most studies addressing the efficacy, effectiveness and safety of antenatal iron and iron with folic acid supplementation during pregnancy.

International organisations have been advocating routine iron and folic acid supplementation for every pregnant woman in areas of high anaemia prevalence (Beard 2000; Villar 1997). While iron supplementation with or without folic acid has been used in a variety of doses and regimens, some current recommendations for pregnant women include the provision of a standard daily dose of 60 mg of iron and 400 ug of folic acid for six months or, if six months of treatment cannot be achieved during pregnancy, either continued supplementation during the postpartum period or increased dosage to 120 mg iron daily during pregnancy (WHO 2006), or if iron deficiency prevalence in the country is high or the pregnant women are anaemic (INACG 1998). Gastrointestinal side effects have been selected as the critical adverse effect on which to base the tolerable upper intake level for iron, as gastrointestinal distress is observed commonly in women consuming high levels of supplemental iron on an empty stomach. High-dose iron supplements are commonly associated with constipation and other gastrointestinal effects including nausea, vomiting and diarrhoea, with frequency and severity varying according to the amount of elemental iron released in the stomach. The Institute of Medicine has established the tolerable upper limit for iron during pregnancy based on gastrointestinal side effects as 45 mg/day of iron, a daily dose much lower than international recommendations (IOM 2001). In most industrialised countries, the decision to prescribe or recommend antenatal iron with folic acid supplementation to women during pregnancy is left to the health care personnel, and is based on the individual maternal condition. In the United States, iron supplementation as a primary prevention intervention involves smaller daily iron doses (i.e. 30 mg/day) but higher doses, up to 120 mg daily are recommended in the presence of anaemia (CDC 1998).

Less frequent regimens of iron supplementation, such as once weekly or twice weekly with iron alone or in conjunction with folic acid, have been evaluated in the last decade as a promising innovative regimen. The weekly iron administration is based on two lines of evidence: (1) daily iron supplementation, by maintaining an iron-rich environment in the gut lumen and in the intestinal mucosal cells, produces oxidative stress, reduces the long-term iron-absorption efficacy and is prone to increasing the severity and frequency of undesirable side effects (Srigirihar 1998; Srigiridhar 2001; Viteri 1997; Viteri 1999a); (2) the concept that exposing intestinal cells to supplemental iron less frequently, every week based on the rate of mucosal turnover that occurs every five to six days in the human, may improve the efficiency of iron utilisation. Recently, the mucosal block phenomenon of iron absorption has been demonstrated when enterocytes have high iron levels, as occurs with daily iron intake (Anderson 2005; Frazer 2003a; Frazer 2003b). Additionally, compliance could increase due to fewer side effects and the costs of supplementation may be favourable if provided outside of the medical context (Viteri 1995; Viteri 1999b). However, some authors have questioned this belief, indicating that the main reason for the poor compliance with programs is the unavailability of iron supplements for the targeted women (Galloway 1994). Recently, lower birthweight and premature delivery have been associated with excess iron intake (which may cause cell damage from the production of reactive oxygen species) and with higher levels of haemoglobin concentrations late during the second trimester and early into the third trimester but not at term (Casanueva 2003b).

This review combines and updates the three currently published Cochrane Reviews on iron and iron+folic acid supplementation (Mahomed 2000a; Mahomed 1998a; Pena-Rosas 2006) that have clearly shown improvements on biochemical and haematological parameters, and evaluates the issues related to periodicity as well as the potential benefits and hazards of these interventions.

 

Objectives

To assess the effectiveness and safety of preventive daily and intermittent use of iron supplements by pregnant women, either alone or in conjunction with folic acid.

The effectiveness of different treatments for iron-deficiency anaemia among pregnant women (Reveiz 2007) and the effects of supplementation with iron and vitamin A during pregnancy (Van den Broek 2002) are covered in other Cochrane Reviews. The effectiveness of vitamin C supplementation in pregnancy is covered in another Cochrane Review (Rumbold 2005). The effects of supplementation with folic acid alone (Mahomed 1998a), the effectiveness of periconceptional use of folic acid on the prevalence of neural tube defects (Lumley 2003) and the effects of multiple vitamin and mineral supplements during pregnancy have also been reviewed elsewhere (Haider 2006).

 

Methods

 

Criteria for considering studies for this review

 

Types of studies

We reviewed randomised and quasi-randomised trials comparing the effects of daily prenatal oral iron or iron+folic acid supplements among pregnant women with the effects of no treatment/placebo or intermittent supplementation regimens. We excluded studies that assessed the effects of multiple combinations of vitamins and minerals, except studies that examined the "additional effect" of iron or iron+folic acid supplements when all arms of the study group were provided with the same other micronutrient supplements (with the exception of iron or iron+folic acid). We have not reviewed the effects of supplementation with multiple micronutrients containing iron or iron+folic acid in comparison to supplementation with iron or iron+folic acid alone here. We also excluded studies dealing with iron supplementation for anaemic women as a medical treatment.

 

Types of participants

Pregnant women of any gestational age, parity, and number of fetuses.

 

Types of interventions

1. Daily universal oral supplementation with iron or iron+folic acid compared with no iron or iron+folic acid supplementation/placebo.
   
2. Daily universal oral supplementation with iron or iron+folic acid compared with universal intermittent (weekly or twice weekly) regimens.
   
3. Intermittent oral iron or iron+folic acid supplementation compared with no iron+folic acid supplementation/placebo.
   

 

Types of outcome measures

Maternal, perinatal and postpartum clinical and laboratory outcomes and infant clinical and laboratory outcomes as described below.

 

Primary

 

Infant

1. Low birthweight (less than 2500 g)
   
2. Birthweight (g)
   

 

Maternal

1. Premature delivery (less than 37 weeks' gestation)
   
2. Hb concentration at term in g/L
   
3. Anaemia at term (Hb less than 110 g/L)
   
4. Haemoconcentration at term (defined as Hb greater than 130 g/L)
   
5. Haemoconcentration at any time during second or third trimesters (defined as Hb greater than 130 g/L)
   
6. Iron deficiency at term (based on two or more laboratory indicators)
   
7. Iron-deficiency anaemia at term (Hb less than 110 g/L and at least one additional laboratory indicator)
   
8. Side effects (any)
   

 

Secondary

 

Infant

1. Very low birthweight (less than 1500 g)
   
2. Perinatal mortality
   
3. Hb concentration at 1 month in g/L
   
4. Ferritin concentration at 1 month in ug/L
   
5. Hb concentration at 3 months in g/L
   
6. Ferritin concentration at 3 months in ug/L
   
7. Hb concentration at 6 months in g/L
   
8. Ferritin concentration at 6 months in ug/L
   
9. Long-term infant developmental (as defined by trial authors)
   
10. Admission to special care unit
    

 

Maternal

1. Very premature delivery (less than 34 weeks' gestation)
   
2. Severe anaemia at term (Hb less than 70 g/L)
   
3. Moderate anaemia at term (Hb greater than 70 g/L and less than 110 g/L)
   
4. Severe anaemia at any time during second or third trimesters (Hb less than 70 g/L)
   
5. Moderate anaemia at any time during second or third trimesters (Hb greater than 70 g/L and less than 110 g/L)
   
6. Infection during pregnancy (including urinary tract infections and others as specified by trial authors)
   
7. Puerperal infection (as defined by trial authors)
   
8. Antepartum haemorrhage (as defined by trial authors)
   
9. Postpartum haemorrhage (intrapartum and postnatal, as defined by trial authors)
   
10. Transfusion given (as defined by trial authors)
    
11. Hb concentration within 1 month postpartum in g/L
    
12. Severe anaemia postpartum (Hb less than 80 g/L)
    
13. Moderate anaemia at postpartum (Hb greater than 80 g/L and less than 100 g/L)
    
14. Diarrhoea
    
15. Constipation
    
16. Nausea
    
17. Heartburn
    
18. Vomiting
    
19. Maternal death (any known)
    
20. Maternal well being/satisfaction (as defined by trial authors)
    
21. Placental abruption (as defined by trial authors)
    
22. Premature rupture of membranes (as defined by trial authors)
    
23. Pre-eclampsia (as defined by trial authors)
    

We recorded other relevant outcomes reported by trial authors and labelled them as 'not prespecified'.

 

Search methods for identification of studies

 

Electronic searches

We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co-ordinator (March 2009).

The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from:

1. quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
   
2. weekly searches of MEDLINE;
   
3. hand searches of 30 journals and the proceedings of major conferences;
   
4. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
   

Details of the search strategies for CENTRAL and MEDLINE, the list of hand searched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.

Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords. 

We did not apply any language restrictions.

 

Searching other resources

For assistance in identifying ongoing or unpublished studies, we also contacted the Departments of Reproductive Health and Research and Nutrition for Health and Development from the World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC).

 

Data collection and analysis

We assessed trials for methodological quality using the criteria in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008) for adequate, unclear and inadequate allocation concealment. We also collected information on blinding of outcome assessment and loss to follow up and incorporated them in the additional table of methodological quality. We classified blinding as "adequate" if both the trial participants and care providers/assessors were blind to the treatment, as "unclear" if either the participants or care providers/assessors were blind to the treatment and as "inadequate" if the blinding status of a trial was unclear or the trial was open. We considered follow up to be adequate if more than 80% of participants initially randomised in a trial were included in the analysis, unclear if the percentage of initially randomised participants included in the analysis was unclear, and inadequate if less than 80% of those initially randomised were included in the analysis.

Two authors independently assessed the eligibility of identified studies. The contact authors extracted data from the reports. We then assessed the number of losses to follow up and post-randomisation exclusions systematically for each trial. We included quasi-randomised studies and conducted a sensitivity analysis. We also included cluster-randomised studies and adjusted their samples sizes (Higgins 2008) if sufficient information was available to allow for this. If possible, we estimated the intra-cluster correlation coefficients for each outcome from original data provided by the authors and contacted authors of the original reports for additional data as required.

We designed a form to facilitate the process of data extraction and to request additional (unpublished) information from the authors of the original reports. We entered data onto Review Manager software (RevMan 2008). We resolved any disagreements among the authors of this review concerning what studies to include or exclude studies or concerning data extraction by discussion, and, if necessary, sought clarification from the authors of the original reports. We analysed dichotomous data in terms of risk ratio (RR) and we analysed continuous data in terms of mean difference (MD), unless the trials used in a particular analysis reported outcomes on different scales that could not be converted to a common scale, in which case, we used the standard mean difference.

We described the proportion of variability in the data due to between-study heterogeneity using the I² statistic test available in RevMan 2008. Because of the high heterogeneity among trials, we pooled trial results using a random-effects model and were cautious in our interpretation of the pooled results.

 

Sensitivity analysis

We considered substantial heterogeneity when the I² statistic was greater than 50% among trials used in a particular meta-analysis. This statistic quantifies inconsistency and describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance). We conducted a sensitivity analysis based on the quality of the studies. We considered a study to be of high quality if it was graded as adequate in both the randomisation and allocation concealment and in either blinding or loss to follow up. We conducted an available case analysis and reinstated previously excluded cases when possible.

We investigated publication bias on outcomes with more than 10 trials by examining the funnel plots for signs of asymmetry, although we gave consideration to reasons other than publication bias that could explain the asymmetry, when present.

We defined iron and iron+folic acid supplementation regimens as either daily (a dose taken every day either as a single or repeated dose) or intermittent (any dose taken less frequently than daily such as alternate days, twice a week or weekly).

Our original goal in this review was to compare the effects of eight pairs of supplementation regimens: (1) any iron alone compared to no intervention/placebo; (2) daily iron alone compared with no intervention/placebo; (3) intermittent iron alone compared with no intervention/placebo; (4) intermittent iron alone compared with daily iron alone; (5) any iron+folic acid compared with no intervention/placebo; (6) daily iron+folic acid compared with no intervention/placebo; (7) intermittent iron+folic acid compared with no intervention/placebo; and (8) intermittent iron+folic acid compared to daily iron+folic acid. However, to avoid presenting repetitive data and because we found no studies for some of the comparisons, we limited our review to the effects of (1) daily iron alone compared with no intervention/placebo; (2) intermittent iron alone compared with daily iron alone; (3) daily iron+folic acid compared with no intervention/placebo;and (4) intermittent iron+folic acid compared with daily iron+folic acid.

We conducted both overall analysis of the effects of various supplementation regimens on primary outcomes and subgroup analysis on the primary outcomes based on the following criteria:
(1) early gestational age (supplementation started before 20 weeks' gestation or prior to pregnancy);
(2) late gestational age (supplementation started at 20 weeks of gestation or later);
(3) unspecified gestational age or mixed gestational ages at the start of supplementation;
(4) anaemic (Hb below 110 g/L during first and third trimesters or below 105 g/L in second trimester) at start of supplementation;
(5) non-anaemic (Hb 110 g/L or above during first and third trimesters or Hb 105 g/L or above if in second trimester) at start of supplementation;
(6) unspecified/mixed anaemia status at start of supplementation;
(7) low daily dose of iron(60 mg elemental iron or less);
(8) higher daily dose of iron (more than 60 mg elemental iron).

 

Results

 

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies.

We identified 209 references corresponding to 145 trials. Of these, we included 49 trials, excluded 92 trials, classified two trials as 'awaiting assessment' and confirmed that two trials are still ongoing. We treated a trial in Guatemala that included two sub-studies as two separate trials: one with supervised intake (Chew 1996a) and one with unsupervised intake (Chew 1996b). We also treated another study carried out collaboratively in two different sites as two different trials, one conducted in Rotterdam (Wallenburg 1983) and one conducted in Antwerp (Buytaert 1983). One trial in China (Liu 1996) involved three comparison groups: one receiving weekly doses of iron, one receiving daily doses of iron and a control group. However, since the allocation of the control group was not randomised, we included this study only in our comparisons of the effects of intermittent versus daily iron supplementation.

Thirty-six trials compared the effects of daily iron supplementation with the effects of no iron or placebo (Batu 1976; Butler 1968; Buytaert 1983; Cantlie 1971; Chanarin 1971; Charoenlarp 1988; Chisholm 1966; Christian 2003; Cogswell 2003; De Benaze 1989; Eskeland 1997; Hankin 1963; Harvey 2007;Hemminki 1991Holly 1955; Hood 1960; Kerr 1958; Makrides 2003; Meier 2003; Menendez 1994; Milman 1991; Paintin 1966; Pita Martin 1999; Preziosi 1997; Pritchard 1958; Puolakka 1980; Romslo 1983; Siega-Riz 2001; Svanberg 1975; Tura 1989; Van Eijk 1978; Wallenburg 1983; Willoughby 1967; Wills 1947; Ziaei 2007; Ziaei 2008). Of these, 13 trials were of high quality according to our pre-established criteria (Buytaert 1983; Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007; Hemminki 1991; Makrides 2003; Preziosi 1997; Siega-Riz 2001;Tura 1989; Wallenburg 1983; Ziaei 2007; Ziaei 2008).

Two studies compared the effects of intermittent supplementation with iron alone with the effects of daily supplementation with iron alone (Pita Martin 1999; Yu 1998). However, neither of these met our criteria for high quality. No study compared the effects of intermittent supplementation with iron alone or iron+folic acid with the effects of no treatment or placebo.

Nine trials compared the effects of daily iron+folic acid supplementation with the effects of no treatment (Barton 1994; Batu 1976; Butler 1968; Charoenlarp 1988; Chisholm 1966; Lee 2005; Liu 1996; Taylor 1982; Willoughby 1967). Only one of them (Barton 1994) met the criteria for high quality.

Nine trials compared the effects of intermittent iron+folic acid supplementation with the effects of daily iron+folic acid supplementation (Chew 1996a; Chew 1996b; Ekstrom 2002; Liu 1996; Mukhopadhyay 2004; Ridwan 1996; Robinson 1998; Winichagoon 2003; Young 2000).

See the table of Characteristics of included studies for a detailed description of the studies. All included studies met the pre-stated inclusion criteria.

 

Risk of bias in included studies

Twenty-five trials adequately randomised the participants to the treatment groups (Barton 1994; Butler 1968; Buytaert 1983; Charoenlarp 1988; Chew 1996a; Chew 1996b; Christian 2003; Cogswell 2003; Ekstrom 2002; Eskeland 1997; Harvey 2007; Hemminki 1991; Kerr 1958; Lee 2005; Makrides 2003; Meier 2003; Mukhopadhyay 2004; Preziosi 1997; Ridwan 1996; Siega-Riz 2001; Tura 1989; Wallenburg 1983; Young 2000; Ziaei 2007; Ziaei 2008). Eighteen trials did not report or did not state clearly the randomisation method used (Batu 1976; Cantlie 1971; Chisholm 1966; De Benaze 1989; Holly 1955; Hood 1960; Liu 1996; Menendez 1994; Milman 1991; Paintin 1966; Pritchard 1958; Puolakka 1980; Romslo 1983; Svanberg 1975; Taylor 1982;Van Eijk 1978; Willoughby 1967; Winichagoon 2003). Six trials were quasi-randomised using alternate or sequence allocation (Chanarin 1971; Hankin 1963; Pita Martin 1999; Robinson 1998; Wills 1947; Yu 1998). Six trials used cluster randomisation (Christian 2003; Ekstrom 2002; Makrides 2003; Mukhopadhyay 2004; Ridwan 1996; Winichagoon 2003).

Twenty-one trials reported using sealed envelopes or coded or opaque bottles when allocating women to treatment groups (Barton 1994; Butler 1968; Buytaert 1983; Chew 1996a; Chew 1996b; Chisholm 1966; Christian 2003; Cogswell 2003; De Benaze 1989; Eskeland 1997; Harvey 2007; Hemminki 1991; Liu 1996; Makrides 2003; Paintin 1966; Preziosi 1997; Siega-Riz 2001; Tura 1989; Wallenburg 1983; Ziaei 2007; Ziaei 2008). The method of concealing allocation used in 17 trials was unclear (Batu 1976; Cantlie 1971; Charoenlarp 1988; Holly 1955; Hood 1960; Kerr 1958; Lee 2005; Meier 2003; Milman 1991; Pritchard 1958; Puolakka 1980; Robinson 1998; Romslo 1983; Svanberg 1975; Taylor 1982; Willoughby 1967; Young 2000). Some trials used an inadequate method or did not use any allocation concealment at all (Chanarin 1971; Ekstrom 2002; Hankin 1963; Mukhopadhyay 2004; Menendez 1994;Pita Martin 1999; Ridwan 1996; Van Eijk 1978; Wills 1947; Winichagoon 2003; Yu 1998). Clearly, studies comparing the effects of intermittent supplementation regimens with the effects of daily supplementation regimens would have had difficulty keeping participants blinded as to what treatment they were receiving without obscuring participants' adherence to any intermittent supplementation regimen, given that doing so would require participants on an intermittent regimen to receive placebo for some days.

See the table of "Methodological quality assessment of included trials" ( Table 1) for a summary of the trials' quality.

 

Effects of interventions

We have included 49 trials, involving 23,200 women, in this review. We have organised the summary results by supplementation regimens compared and by primary and secondary outcomes. Most of the included studies focused on haematological indices and few reported on any of the other outcomes prespecified in the review protocol. Because all the results showed significant heterogeneity that could not be explained by standard sensitivity analyses including quality assessment, we used a random-effects model to analyse the results.

See the Data and analyses section for detailed results on primary and secondary outcomes.

 

(1) Daily iron alone compared with no intervention/placebo

 

Infant outcomes

 

Low birthweight (less than 2500 g)

Overall, we found no significant difference in the prevalence of low birthweight (less than 2500 g) between newborns of mothers in these two groups ( Analysis 1.1). Among 6275 women in nine trials (Christian 2003; Cogswell 2003; Eskeland 1997; Makrides 2003; Meier 2003; Menendez 1994; Siega-Riz 2001; Ziaei 2007; Hemminki 1991), 9.6% of those who took daily iron supplementation during pregnancy had a baby with birthweight below 2500 grams versus 11.9% of those who received no iron or placebo (risk ratio (RR) 0.79; 95% CI 0.61 to 1.03) ( Analysis 1.1). When we limited our analysis to high-quality studies (Christian 2003; Cogswell 2003; Eskeland 1997; Makrides 2003; Menendez 1994; Siega-Riz 2001; Ziaei 2007), the difference in the percentage of mothers with low birthweight babies remained non-significant. Results from the three trials where 1823 women had unspecified/mixed anaemic status at start of supplementation suggest that women who received iron supplements had a lower risk of having a low birthweight baby in comparison to those who did not receive the iron supplements (risk ratio (RR) 0.82; 95% CI 0.71 to 0.94).

 

Birthweight (g)

We found no significant difference in birthweight ( Analysis 1.3) in children from mothers of the two groups. Among infants born to 5956 participants in ten trials (Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007; Hemminki 1991; Makrides 2003; Preziosi 1997; Puolakka 1980; Siega-Riz 2001; Ziaei 2007) the mean difference (MD) in birthweight between those whose mothers had taken iron supplements and those whose mothers had not was 36.05 g and was not statistically significant (95% CI -4.84 to 76.95) ( Analysis 1.3). Results from the three trials where 1577 women had unspecified/mixed anaemic status at start of supplementation suggest that women who received iron supplements had a lower risk of having a heavier baby in comparison to those who did not receive the iron supplements (MD 52.33 g; 95% CI 10.16 to 94.51).

We did find evidence of significant differences between treatment groups in the following infant secondary outcomes:

 

Infant ferritin concentration at 3 months in ug/L

The MD was 19.0; 95% confidence interval (CI) 2.75 to 35.25 (one trial involving 197 women) (Preziosi 1997) ( Analysis 1.25).

 

Infant ferritin concentration at 6 months in ug/L

The MD was 11.0 ug/L; 95% CI 4.37 to 17.63 ug/L (one trial involving 197 women) (Preziosi 1997)( Analysis 1.27). However, the results of another study that evaluated infant serum ferritin concentration at six months (Makrides 2003) showed that the geometric mean of concentrations among infants from mothers who had received 20 mg elemental iron during pregnancy did not differ significantly from that of infants whose mothers had received placebo (32.5 +/- 2.0 and 30.8 +/- 2.0 ug/L respectively). The Makrides study was conducted among women from a well-nourished industrialised population (Makrides 2003), whereas the Preziosi study was conducted among women in a developing country (Preziosi 1997).

 

Birth length in cm (not pre-specified)

We found significant difference in birth length ( Analysis 1.94) in children from mothers of the two groups. Among infants born to 2140 participants in five trials (Christian 2003; Cogswell 2003; Eskeland 1997; Makrides 2003; Preziosi 1997), the MD in birth length between those whose mothers had taken iron supplements and those whose mothers had not was 0.38 cm (95% CI 0.10 to 0.65) ( Analysis 1.94).

We found no evidence of significant difference by treatment group in the following secondary outcomes:
very low birthweight (less than 1500 g) ( Analysis 1.20); perinatal death ( Analysis 1.21); infant Hb concentration at 3 months in g/L ( Analysis 1.24); infant haemoglobin concentration at six months ( Analysis 1.26); admission to special care unit ( Analysis 1.29); percentage of infants who were small for gestational age ( Analysis 1.91).

No trials reported on the remaining infant outcomes.

 

Maternal outcomes

 

Premature delivery (less than 37 weeks' gestation)

We found no evidence of significant difference in rates of premature delivery between women who received daily iron supplementation and those who received no iron supplementation ( Analysis 1.5).

 

Maternal haemoglobin concentration at term in g/L

Among 2463 women who participated in 17 trials (Batu 1976; Butler 1968; Buytaert 1983; Cantlie 1971; Chanarin 1971; Cogswell 2003; De Benaze 1989; Eskeland 1997; Makrides 2003; Milman 1991;Puolakka 1980; Romslo 1983; Tura 1989; Van Eijk 1978; Wallenburg 1983; Ziaei 2007; Ziaei 2008), those who took iron supplements had a mean haemoglobin concentration 8.83 g/L higher at term in comparison to those who took no iron supplements at all (MD 8.83; 95% CI 6.55 to 11.11 g/dL) ( Analysis 1.7). However, because the heterogeneity among the treatment effects found in individual studies was substantial (I² greater than 50%), our results have to be interpreted with caution. The difference in haemoglobin concentration was slightly higher among those starting supplementation after 20 weeks of gestation ( Analysis 1.8) and the difference was maintained among those who received different doses of elemental iron. The MD did not change significantly when we included only high-quality trials (Buytaert 1983; Cogswell 2003; Eskeland 1997; Makrides 2003; Tura 1989; Wallenburg 1983; Ziaei 2007; Ziaei 2008) (MD 6.00; 95% CI 2.75 to 9.25) and heterogeneity remained high. A forest plot showed no asymmetry, suggesting that there may not be publication bias (Figure 1).

Figure 1. Funnel plot of comparison: 1 Daily iron alone versus no intervention/placebo, outcome: 1.7 Maternal Hb concentration at term (g/L) (ALL).

 

Anaemia at term (Hb less than 110 g/L)

Among 4390 women in 14 trials (Batu 1976; Chanarin 1971;Chisholm 1966; Cogswell 2003; De Benaze 1989; Eskeland 1997; Hemminki 1991; Holly 1955; Makrides 2003; Milman 1991; Preziosi 1997; Pritchard 1958; Puolakka 1980; Romslo 1983), 5.08% of those who received daily iron supplements during pregnancy and 14.6% who did not had anaemia at term (RR 0.27; 95% CI 0.17 to 0.42 ( Analysis 1.9)). However, because the heterogeneity in study results was substantial (I² greater than 50%), our results have to be interpreted with caution. The results of a sensitivity analysis of data from the four high-quality studies involving a total of 787 women (Cogswell 2003; Eskeland 1997; Hemminki 1991; Makrides 2003; Preziosi 1997) showed a slightly weaker association between use of iron supplements and anaemia status at term (4.7% versus 9.98%; RR 0.46; 95% CI 0.29 to 0.72), although the heterogeneity I² value increased. A forest plot showed certain asymmetry, suggesting that there may be some publication bias (Figure 2).

Figure 2. Funnel plot of comparison: 1 Daily iron alone versus no intervention/placebo, outcome: 1.9 Anaemia at term (Hb less than 110 g/L).

 

Haemoconcentration at term (defined as Hb greater than 130 g/L)

Data from 10 trials involving 4643 women (Butler 1968; Chisholm 1966; Cogswell 2003; Eskeland 1997; Hemminki 1991; Holly 1955; Makrides 2003; Milman 1991; Pritchard 1958; Ziaei 2007) indicated that 53.8% of women who took daily iron supplementation during pregnancy and 38.0% of those who did not had haemoconcentration at term (RR 2.62; 95% CI 1.21 to 5.67) ( Analysis 1.10). The heterogeneity between the treatment effects was substantial (I² greater than 90%) and the results have to be interpreted with caution ( Analysis 1.10). When we limited the analysis to the five high-quality trials involving a total of 1317 women (Cogswell 2003; Eskeland 1997; Hemminki 1991; Makrides 2003; Ziaei 2007) the difference in haemoconcentration prevalence was no longer significant (62.9% versus 51.9%; RR 1.73; 95% CI 0.64 to 4.66) and the heterogeneity in study increased (I² = 97%) (not shown). The risk of haemoconcentration at term was higher among women who received daily higher doses of more than 60 mg elemental iron ( Analysis 1.11)

 

Haemoconcentration (Hb greater than 130 g/L) at any time during second or third trimester

Ten trials involving 4841 women evaluated the effects of oral routine supplementation with iron alone and haemoconcentration at any time during the second or third trimesters (Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007; Hemminki 1991; Holly 1955; Makrides 2003; Milman 1991; Pritchard 1958; Ziaei 2007). Among women who received daily iron supplements, 25.1% were found to have haemoconcentration at some time during their second or third trimesters, compared with 9.3% of those who received no iron supplements (RR 2.27; 95% CI 1.40 to 3.70) ( Analysis 1.12). However, because the heterogeneity in study results was substantial (I² = 89%), the results have to be interpreted with caution ( Analysis 1.12). When we limited our analysis to the seven trials of high quality (Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007;Hemminki 1991; Makrides 2003; Ziaei 2007), the association between use of iron supplements and risk for haemoconcentration was still significant (RR 2.22; 95% CI 1.28 to 3.85), and the heterogeneity remained high (91%). The effect was maintained in women who started the supplementation at an early gestational age (less than 20 weeks) and among women who were non-anaemic at the start of the intervention. The effects were also similar in the women who received higher or lower doses of elemental iron as defined in this review ( Analysis 1.13).

 

Iron deficiency at term (based on two or more laboratory indicators)

Data from six trials involving 1108 women (Cogswell 2003; Eskeland 1997; Makrides 2003; Milman 1991; Preziosi 1997; Tura 1989) showed that 30.7% of women who received daily iron supplements had iron-deficiency anaemia at term, compared with 54.8% of those who received no iron supplements (RR 0.44; 95% CI 0.27 to 0.70) ( Analysis 1.14). The heterogeneity between the treatment effects is substantial (I² greater than 50%) and the results have to be interpreted with caution ( Analysis 1.14). Five of the trials were of high quality.

 

Iron-deficiency anaemia at term (Hb below 110 g/L and at least one additional laboratory indicator)

Data from six trials involving 1667 women (Cogswell 2003; Eskeland 1997; Makrides 2003; Milman 1991; Tura 1989; Ziaei 2007) showed that 4.9% of women who received daily iron supplements and 15.5% of those who did not had iron-deficiency anaemia at term (RR 0.33; 95% CI 0.16 to 0.69) ( Analysis 1.16). The heterogeneity between the treatment effects was small (I² less than 50%) ( Analysis 1.16). These results were similar for the different subgroups, including those who start supplementation early in the gestation and those who are non-anaemic at start of the study, and in any iron dose. The effect was similar (3.1% versus 7.9%); when only five trials of high quality involving 1547 women (Cogswell 2003; Eskeland 1997; Makrides 2003; Tura 1989; Ziaei 2007) were compared: RR 0.39; 95% CI 0.20 to 0.74 and a test of heterogeneity (I² = 40.4%) (not shown).

 

Side effects (any)

Data from eight trials involving 3667 women (Charoenlarp 1988; Cogswell 2003; De Benaze 1989; Eskeland 1997; Harvey 2007; Hemminki 1991; Hood 1960; Kerr 1958) suggest that women who receive daily oral iron supplementation are more likely to report side effects of any kind than women taking placebo or not taking any iron supplements at all (24.7% versus 4.3%; RR 3.92; 95% CI 1.21 to 12.64) ( Analysis 1.18). However, the heterogeneity between the treatment effects is substantial (I² greater than 50%) and the results have to be interpreted with caution ( Analysis 1.18). When only the four high-quality trials involving 2897 women were included (Cogswell 2003; Eskeland 1997; Harvey 2007; Hemminki 1991), the effect is no longer significant (24.5% versus 3.4%; RR 3.27; 95% CI 0.39 to 27.31 (data not shown)) with high heterogeneity (I² = 98%). The effect was significantly consistent in women who received daily higher doses of elemental iron (more than 60 mg).

There was evidence of significant differences found in the following secondary outcomes:

 

Moderate anaemia (defined as Hb > 70/L and < 90 g/L)at any time during the second or third trimester

Data from 10 trials involving 2266 women (Butler 1968; Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007; Holly 1955; Makrides 2003; Milman 1991; Paintin 1966; Ziaei 2007) suggest that women who routinely receive daily iron supplementation are less likely to have moderate anaemia at any time during the second or third trimesters than women receiving no iron or placebo (2.1% versus 3.3%; RR 0.42; 95% CI 0.19 to 0.92). The heterogeneity between treatment effects was small (I² less than 50%) ( Analysis 1.34). When only the six high-quality trials involving 1668 women were included (Christian 2003; Cogswell 2003; Eskeland 1997; Harvey 2007; Makrides 2003; Ziaei 2007) the effect remained significant (2.1% versus 3.3%; RR 0.23; 95% CI 0.08 to 0.66 (data not shown)) with no heterogeneity (I² = 0%) as only two trials had cases.

 

Transfusion provided

The data from three trials involving 3453 women (Hemminki 1991; Puolakka 1980; Ziaei 2007) suggest that women that routinely receive daily iron supplementation have a lower risk of receiving transfusion in comparison with women who did not receive iron supplementation (1.6% versus 2.7%); RR 0.61; 95% CI 0.38 to 0.96).

 

Maternal haemoglobin concentration within one month postpartum in g/L

The data from six trials involving 904 women (Cantlie 1971; Hankin 1963; Lee 2005; Menendez 1994; Milman 1991; Wills 1947) suggest that women that routinely receive daily iron supplementation have a higher concentration of haemoglobin after one month postpartum than those taking placebo or not taking any iron supplements at all (WMD 7.08 g/L; 95% CI 4.70 to 9.47 g/L). The I² statistic shows that heterogeneity of the results is less than 50% ( Analysis 1.40). None of the trials met the criteria for high quality.

 

Diarrhoea

Data from three trials involving 1088 women (Paintin 1966; Siega-Riz 2001; Ziaei 2007) showed that 3.9% of women who received daily iron supplements and 5.1% of those who did not presented with diarrhoea (RR 0.55; 95% CI 0.32 to 0.93) with no heterogeneity in individual study results (I² = 0%). In our analysis of data from only the two high-quality trials involving 915 women (Siega-Riz 2001; Ziaei 2007) the association between supplement use and risk for diarrhoea remained significant (4.4% versus 5.5%; RR 0.53; 95% CI 0.31 to 0.91; data not shown), and there was no heterogeneity in study results (I² = 0%) ( Analysis 1.43).

There was no evidence of significant difference between women receiving daily iron supplementation and women receiving placebo or not taking any iron supplements at all, in the following secondary outcomes:
very premature delivery (less than 34 weeks' gestation), placental abruption, pre-eclampsia, severe anaemia at term, at any time during second or third trimesters or postpartum, moderate anaemia at term; and in the postpartum, puerperal infection, antepartum haemorrhage and postpartum haemorrhage, transfusion given, constipation, nausea, heartburn, vomiting, maternal death, maternal Hb concentration at four to eight weeks postpartum in g/L (not pre-specified).

 

Maternal wellbeing/satisfaction

A maternal index of wellbeing was measured in one trial (Makrides 2003) through the use of a self-administered questionnaire at 36 weeks gestation and at six weeks and six months postpartum. There were not significant differences in any of the eight health concepts measured by this methodology between the women in the iron supplemented group or those in the placebo group at 36 weeks' gestation, six weeks and six months postpartum. Another trial (Eskeland 1997) assessed maternal wellbeing at 28 and 36 weeks' gestation, and found no differences between the iron supplemented mothers or those receiving placebo ( Analysis 1.49).

No trials reported on the remaining secondary outcomes.

 

(2) Intermittent iron alone compared to daily iron alone

 

Infant outcomes

 

Low birthweight (less than 2500 g)

No trials reported on this outcome.

 

Birthweight (g)

No evidence of significant differences was found between these groups of infants in birthweight. Only one study (Pita Martin 1999) with 41 women provided data for this outcome.

No trials reported on the remaining secondary outcomes.

 

Maternal outcomes

 

Premature delivery (less than 37 weeks' gestation)

No evidence of significant differences was found between these groups of women.

 

Hb concentration at term in g/L

No trials reported on this outcome.

 

Anaemia at term (Hb less than 110 g/L) (not prespecified)

No trials reported on this outcome.

 

Haemoconcentration at term (defined as Hb greater than 130 g/L)

No trials reported on this outcome.

 

Haemoconcentration at any time during second or third trimesters (defined as Hb greater than 130 g/L)

No evidence of significant differences was found between these groups of women.

 

Iron deficiency at term (based on two or more laboratory indicators)

No trials reported on this outcome.

 

Iron-deficiency anaemia at term (Hb less than 110 g/L and at least one additional laboratory indicator)

No trials reported on this outcome.

 

Side effects (any)

No trials reported on this outcome.

No evidence of significant differences was found between these groups of women for this secondary outcome: moderate anaemia at any time during the second or third trimesters. The effect of the intervention on severe anaemia at any time during second or third trimesters could not be estimated ( Analysis 2.33).

No trials reported on the remaining secondary outcomes.

 

(3) Daily iron-folic acid compared to no intervention/placebo

 

Infant outcomes

 

Low birthweight (less than 2500 g)

No evidence of significant differences was found between infants from these groups of women receiving daily iron+folic acid supplementation and those taking placebo or not taking any supplements at all.

 

Birthweight (g)

Data from two trials involving 1365 women (Christian 2003; Taylor 1982) suggest that infants whose mothers received daily iron+folic acid are 57.7 g heavier than infants born from mothers who received no iron+folic acid in pregnancy MD 57.73 (95% CI 7.66 to 107.79) ( Analysis 3.3).

 

Small for gestational age (less than 10th percentile weight at birth for gestational age)

One trial including 1318 women (Christian 2003) suggested that women routinely receiving iron and folic acid supplementation are less likely to give birth to a small for gestational age baby, in comparison to those receiving no iron and folic acid supplementation (RR 0.88; 95% CI 0.80 to 0.97) ( Analysis 3.91).

No evidence of significant differences was found between infants from these groups of women receiving daily iron+folic acid supplementation and those taking placebo or not taking any supplements at all in the following secondary outcomes:
very low birthweight (less than 1500 g), perinatal mortality, admission to special care unit, small for gestational age (less than 10th percentile weight for gestational age) or birth length.

No trials reported on the remaining secondary outcomes.

 

Maternal outcomes

 

Premature delivery (less than 37 weeks' gestation)

No evidence of significant differences was found between women who received daily iron and folic acid supplements and those receiving no treatment or placebo.

 

Haemoglobin concentration at term in g/L

The data from four trials including 179 women (Barton 1994; Batu 1976; Butler 1968; Taylor 1982) suggest that women who routinely receive daily iron and folic acid supplementation reach term with higher Hb concentration than women taking placebo or not taking any iron and folic acid supplement at all (MD 12.00 g/L; 95% CI 2.93 to 21.07). However, the heterogeneity between the treatment effects is substantial (I² greater than 50%) and the results have to be interpreted with caution ( Analysis 3.7). The effect of iron-folic acid supplementation did not change significantly after including only the one high-quality trial (MD 17.10; 95% CI 8.44 to 25.76 ) (data not shown). The subgroup analysis provided similar trends ( Analysis 3.8).

 

Anaemia at term (Hb less than 110 g/L) (not prespecified)

The data from three trials including 346 women (Barton 1994; Batu 1976; Chisholm 1966) suggest that women who routinely receive daily iron and folic acid supplementation during pregnancy are less likely to have anaemia at term than those not taking any iron and folic acid supplements at all (defined as Hb less than 110 g/L) (8.2% versus 35.5%; RR 0.27; 95% CI 0.12 to 0.56) ( Analysis 3.9). However, the heterogeneity between the treatment effects is substantial (I² greater than 50%) and the results have to be interpreted with caution. No studies met the prespecified criteria for high quality.

 

Haemoconcentration at term (defined as Hb greater than 130 g/L)

No evidence of significant differences was found between women who received daily iron and folic acid supplements and those receiving no treatment or placebo.

 

Haemoconcentration at any time during second or third trimesters (defined as Hb greater than 130 g/L)

No evidence of significant differences was found between women who received daily iron and folic acid supplements and those receiving no treatment or placebo.

 

Iron deficiency at term (based on two or more laboratory indicators)

Data from one trial involving 131 women (Lee 2005) suggest that women who routinely receive daily oral supplementation with iron are less likely to have iron deficiency at term than women taking placebo or not taking any iron and folic acid supplements at all (3.6% versus 15%; RR 0.24; 95% CI 0.06 to 0.99) ( Analysis 3.14). The effect was not significant in the trial. and arms were presented separately in the subgroup analysis for early or late start of supplementation, although the trends followed the same direction ( Analysis 3.15).

 

Iron-deficiency anaemia at term (Hb less than 110 g/L and at least one additional laboratory indicator)

No evidence of significant differences was found between women who received daily iron and folic acid supplements and those receiving no treatment or placebo.

 

Side effects (any)

One trial including 456 women (Charoenlarp 1988) suggests that women routinely receiving iron and folic acid supplementation are more likely to report any side effects in comparison to none from those receiving no supplementation (RR 44.32; 95% CI 2.77 to 709.09) ( Analysis 3.18).

 

Severe anaemia at any time during second and third trimester (Hb less than 70 g/L)

The data from four trials involving 523 women (Barton 1994; Butler 1968; Christian 2003; Lee 2005) suggest that women who receive iron+folic acid supplements during pregnancy have lower risk of severe anaemia at any time during second or third trimester, in comparison to those receiving no iron and folic acid (RR 0.11; 95% CI 0.01 to 0.83) ( Analysis 3.33).

 

Haemoglobin concentration within one month postpartum in g/L

One study (Taylor 1982) involving 45 women reported this outcome. The data from this trial suggest that women receiving daily iron+folic acid supplementation achieve a higher concentration of haemoglobin at one month postpartum than women not taking any supplements at all (MD 10.40; 95% CI 4.03 to 16.77) ( Analysis 3.40) but no firm conclusions can be made given the scarcity of the data.

 

Severe anaemia at postpartum (Hb less than 80 g/L)

The data from three trials including 525 women suggest that women who received iron+folic acid supplementation during pregnancy were less likely to present moderate anaemia at postpartum RR 0.05; 95% CI 0.00 to 0.76 ( Analysis 3.41). The scarcity of data makes it difficult to draw any conclusion.

 

Moderate anaemia at postpartum (Hb more than 80 g/L and less than 100 g/L)

The data from three trials including 525 women suggest that women who received iron+folic acid supplementation during pregnancy were less likely to present moderate anaemia at postpartum (3.5% versus 12.9%; RR 0.34; 95% CI 0.17 to 0.69) ( Analysis 3.42). The scarcity of data makes it difficult to draw any conclusion.

No evidence of significant differences was found in the following secondary outcomes:
very premature delivery, severe anaemia at term, moderate anaemia at term, severe anaemia at any time during second or third trimesters, moderate anaemia at any time during second or third trimesters, infection during pregnancy, puerperal infection, antepartum haemorrhage, postpartum haemorrhage, maternal death, placental abruption, pre-eclampsia, haemoglobin concentration at four to eight weeks postpartum.

No trials reported on the remaining secondary outcomes.

 

(4) Intermittent iron-folic acid compared to daily iron-folic acid

 

Infant outcomes

 

Low birthweight (less than 2500 g)

The data from four trials (Chew 1996a; Chew 1996b; Mukhopadhyay 2004; Winichagoon 2003) involving 730 women suggest that women who take intermittent iron+folic acid supplementation during pregnancy are as likely to have a baby with birthweight below 2500 grams (5.6% versus 5.9%; RR 1.05; 95% CI 0.58 to 1.91) ( Analysis 4.1).

 

Birthweight (g)

The data from four trials (Chew 1996a; Chew 1996b; Mukhopadhyay 2004; Winichagoon 2003) involving 730 women suggest that there is no significant effect in birthweight of newborns born from women who had taken daily supplementation with iron+folic acid during pregnancy or from those being supplemented intermittently (MD -7.10; 95% CI -67.20 to 53.01 g) ( Analysis 4.3).

 

Infant ferritin concentration at six months in ug/L

One study (Winichagoon 2003) including 88 women reported this outcome ( Analysis 4.27). The data from this trial suggest that the infants from women receiving intermittent iron+folic acid supplementation achieve a higher concentration of serum ferritin at six months (MD 0.09; 95% CI 0.05 to 0.13 ug/L) ( Analysis 4.27) but no firm conclusions can be made given the scarcity of the data.

No evidence of significant differences was found in the following secondary outcomes:
very low birthweight (less than 1500 g) was not estimable.

No trials reported on the remaining secondary outcomes.

 

Maternal outcomes

 

Premature delivery (less than 37 weeks' gestation)

No evidence of significant differences was found between these groups of women.

 

Hb concentration at term in g/L

No evidence of significant differences was found between these groups of women.

 

Anaemia at term (Hb less than 110 g/L) (not prespecified)

No evidence of significant differences was found between these groups of women.

 

Haemoconcentration at term (defined as Hb greater than 130 g/L)

No evidence of significant differences was found between these groups of women.

 

Haemoconcentration at any time during second or third trimesters (defined as Hb greater than 130 g/L)

Six trials involving 1111 women (Ekstrom 2002; Liu 1996; Mukhopadhyay 2004; Ridwan 1996; Robinson 1998; Winichagoon 2003) suggest that women who routinely receive intermittent iron+folic acid supplementation during pregnancy are less likely to have haemoconcentration at any time during the second or third trimesters than those receiving the daily regimen (7.7% versus 18.8 %; RR 0.43; 95% CI 0.24 to 0.77) ( Analysis 4.12). None of the trials met the criteria for high quality. The difference trend was maintained in women who started supplementation late in pregnancy. However, the heterogeneity between the treatment effects is substantial (I² greater than 50%) and the results have to be interpreted with caution.

 

Iron deficiency at term (based on two or more laboratory indicators)

No trials reported on this outcome.

 

Iron-deficiency anaemia at term (Hb less than 110 g/L and at least one additional laboratory indicator)

No evidence of significant differences was found between these groups of women.

 

Side effects (any)

No evidence of significant differences was found between these groups of women.

There was no evidence of significant difference in the following secondary outcomes:
severe anaemia at term, moderate anaemia at term, severe anaemia at any time during 2nd or 3rd trimesters, moderate anaemia at any time during 2nd or 3rd trimesters, antepartum haemorrhage, anaemia at postpartum, moderate anaemia at postpartum, diarrhoea, constipation, nausea, heartburn, vomiting, placental abruption, premature rupture of membranes.

No trials reported on the remaining secondary outcomes.

 

Discussion

In this review, we addressed the effects of the use of iron or iron+folic acid by pregnant women, either provided alone or in combination with other micronutrients. The effects can be determined if the differences between the comparison groups relies only in the presence of iron or iron+folic acid, that is we are estimating the effects of the addition of iron or iron+folic acid to the pregnant women independently of any other interventions given to both groups being compared. The effects of the use of multiple micronutrient supplementation by pregnant women were addressed by another Cochrane Review (Haider 2006).

Unfortunately, most of the studies we reviewed provided very limited information about the clinical outcomes for women or their babies. Instead, most focused primarily on maternal changes in haemoglobin and on some haematological indices after a certain period of supplementation. In addition, few studies provided much outcome data at term or postpartum except for maternal haematology and on the longer term effects of the supplementation.

The interpretation of data from studies with a high level of heterogeneity remains a challenge. Simply pooling the results from multiple studies may not be a good way to understand the effects of a particular intervention if the conditions of the studies involved are too dissimilar. For example, in a U.S study we reviewed (Cogswell 2003), women received either 30 mg of iron daily or a placebo from week 20 to week 28 of gestation, after which both groups were given iron supplements in accordance with Institute of Medicine guidelines. In an Australia study we reviewed (Makrides 2003), however, women received either 20 mg/d of iron or a placebo from week 20 until delivery. Therefore, although both studies were designed to assess the effects of low doses of iron supplements among pregnant women, both the precise iron doses given to women in the treatment groups and the length of time that they were given differed, making it difficult to analyse pooled data from the two studies.

Women who received iron alone or iron+folic acid had higher haemoglobin concentration at term, and had lower risk of having anaemia or iron deficiency at term than women who did not. There were no effects of iron or iron+folic acid in risk of low birth weight of the newborns. However, neonates born from mothers who had received iron+folic acid during pregnancy were heavier than those who had not received these micronutrients. In most studies we reviewed, the iron dosages given to women in treatment groups were high. None of them assessed the effects of low-dose iron supplements in combination with folic acid.

There were no differences in the effects of iron alone or iron+folic acid supplements for the risk of low birthweight or prematurity. Women who received the doses daily or intermittently had similar risk of developing anaemia at term and their Hb concentrations did not differ at term. None of the studies we reviewed compared the effects of intermittent iron supplementation with the effects of no iron supplementation because all the studies involving intermittent supplementation were carried out in developing countries whose legislatures require all pregnant women to be given iron supplements. In addition, few studies compared the effects of intermittent iron alone with the effects of daily iron alone because most developing countries routinely include folic acid in iron supplements for pregnant women.

 

Adverse effects

Side effects are a clear drawback to most current iron compounds used as supplements, either alone or with folic acid. The results of this review confirm that daily iron or iron+folic acid doses are associated with a higher risk for side effects, as has been recognised for many years. As a result, investigators are searching for highly bioavailable iron compounds that produce fewer side effects and that can be administered at low doses or intermittently (please see below). Most daily iron supplementation regimens for pregnant women involve doses of more than 45 mg/day, the upper tolerable limit suggested by the Institute of Medicine (IOM 2001). There is no stated upper limit for intermittent iron supplementation.

There is a debate on the benefits of routine daily iron supplementation during pregnancy at the currently high levels recommended by various agencies. It appears that small daily doses as recommended by the US Food and Nutrition Board, the U.S. Centers for Disease Control and Prevention and the Institute of Medicine (Anderson 1991; CDC 1998; IOM 1993), as well as weekly dosing, are essentially as efficacious as daily iron at current doses in preventing significant anaemia and iron deficiency anaemia at term, defined as that having health and functional consequences.

Although we found that the overall risk for haemoconcentration in the second and/or third trimester was lower among women who received intermittent iron+folic acid supplementation than those who received daily supplementation, these findings were confounded by the fact that in some studies low iron doses were administered from the start to nonanaemic women and high iron doses were administered to women with undefined anaemia.

This review suggests that haemoconcentration at term as well as in/or during both the second and third trimester of pregnancy is associated with daily iron supplementation, particularly when doses are high and started early in pregnancy. Haemoconcentration secondary to excessive erythropoiesis during pregnancy in association with iron supplementation has been previously suggested by researchers in Newcastle and others (Hytten 1971; Hytten 1985; Lund 1961; Letsky 1991; Mahomed 1989). Low haemoglobin levels but also high haemoglobin levels have been associated with low birthweight (Garn 1981; Huisman 1986; Koller 1979; Murphy 1986; Scanlon 2000; Steer 1995; Zhou 1998). Further associations were reported between preterm birth and low haemoglobin during the first and second trimesters, and low birthweight due to intrauterine growth retardation and high haemoglobin concentrations also during the first two trimesters (Scanlon 2000). Only haemoglobin levels during the third trimester had erratic consequences regarding birthweight. Importantly, the odds ratios for small-for-gestational-age babies were lower when haemoglobin concentrations were low-normal or low (Z scores less than -1 and greater than -2, and less than -2 and greater than -3 for haemoglobin, respectively) during the second and third trimesters than among women whose haemoglobin concentrations were above 130 g/L. These data emphasise the fact that the iron nutritional and haematological status in the first (Scholl 1992) and second trimesters are the ones that have an influence in the outcome of pregnancy, rather than the status at term or at the third trimester.

It would appear that the normal haemodilution reaching a nadir during the second and early third trimester of pregnancy favours the uneventful course of pregnancy and fetal growth and well being, resulting in normal newborns. In many instances antenatal iron supplementation, at doses currently recommended for developing nations (60 mg to 300 mg of iron/day) and commonly prescribed by obstetricians in industrial societies, may annul the normal haemodilution and even produce abnormally elevated haemoglobin levels in pregnancy (Ziaei 2007). In addition to exploring the possible association between high-dose iron supplementation and the risk for haemoconcentration, researchers also need to examine other possible adverse consequences of high iron supplementation doses, including poor placental perfusion and oxidative stress, as suggested by different studies (Casanueva 2003b). The relationship between iron supplementation and abnormally high haemoglobin levels also merits research because numerous studies have shown that haemoconcentration among pregnant women is associated with an increased risk of their child having a low birth weight.

High haemoglobin levels during pregnancy have also been associated with plasma volume depletion, pre-eclampsia, eclampsia, pregnancy complications, and low birthweight (Gallery 1979; Goodlin 1981; Koller 1979; Silver 1998) and low plasma volume appears to precede late pregnancy hypertension and low birthweight (Gallery 1979; Huisman 1986). A recent trial that studied both volumes of plasma and red blood cells simultaneously showed that both plasma and red cell volumes were reduced, plasma volume reduction averaging 16% was present only in pre-eclampsia (hypertension with albuminuria) but not in non-albuminuric gestational hypertension and was associated with a greater risk of small-for-gestational-age babies (Silver 1998). Other studies involving low birthweight babies where maternal plasma volume was measured failed to demonstrate a level of haemoconcentration that resulted in haemoglobin levels greater than or equal to 135 g/L (Gallery 1979; Hytten 1971; Hytten 1985; Koller 1979; Letsky 1991; Poulsen 1990). These results may suggest that, in otherwise normal pregnant women, haemoconcentration defined as haemoglobin greater than 135 g/L cannot be wholly explained by reduction in maternal plasma volume.

Can haemoconcentration of the levels reported in the studies included in this review result in hyperviscosity, poor placental perfusion and placento/fetal hypoxia? This seems possible based on the data presented by some authors (Erslev 2001; LeVeen 1980). On the one hand, blood viscosity increases essentially in a linear form by about 45% (from 3.2 to 4.3 units relative to H₂O) between a hematocrit of 30% and 47% (corresponding to haemoglobin concentrations of 89 and 140 g/L), but oxygen transport declines only by about 4% between the optimum at hematocrit of 30% to that of 45% (corresponding to haemoglobin concentration of 134 g/L) (LeVeen 1980).

 

Authors' conclusions

 

Implications for practice

Daily or intermittent supplementation with iron or with iron+folic acid by pregnant women clearly results in a substantial reduction in the prevalence of maternal anaemia at term and increased maternal Hb concentrations at term. However, the current sparse data fail to demonstrate that supplementation with iron alone or in combination with folic acid among women without anaemia or with mild or moderate anaemia by current cut-off criteria is significantly associated with any other substantial beneficial or adverse effects on maternal health, fetal health, or pregnancy outcomes. Available data also indicate that weekly supplementation is as effective as daily supplementation in preventing low haemoglobin levels associated with negative outcome consequences and that the use of either daily or weekly iron supplements (with or without folic acid) may be beneficial where iron deficiency and anaemia are prevalent pre-gestationally or in early pregnancy. Starting supplementation early in pregnancy must be stressed, together with interventions to improve pregnancy iron and folate status prior to conception. The evidence suggests that iron supplementation schemes providing more iron than women need may not be desirable, and doses and formulations that can reduce side effects should be encouraged. However, current preventive antenatal iron supplementation doses recommended for populations in developing countries, based on therapeutic doses, appear to be excessive even where moderate/mild anaemia is found. Consequently, current recommended iron and folic acid supplementation schemes during pregnancy should be reviewed to adjust iron doses to levels that are effective and safe for the mothers and newborns and that can assure compliance. Current preventive antenatal iron supplementation doses recommended for populations in developing countries appear to be excessive. Starting supplementation early in pregnancy must be stressed, together with interventions to improve pregnancy iron and folate status prior to conception. The evidence suggests that iron supplementation schemes providing more iron than women need may not be desirable, and doses and formulations that can reduce side effects should be encouraged. Intermittent supplementation with iron could be considered in anaemia prevention strategies not only prior to but also during pregnancy.

 

Implications for research

On the basis of the results of this review, we recommend that researchers investigating the use of iron or iron+folate supplements by pregnant women attempt to undertake the following.

1. Establish a solid basis for defining ranges of iron and folate nutrition and haematological conditions among pregnant woman that are associated with safe and desirable pregnancy outcomes of clinical relevance.
   
2. Identify the mechanisms involved in haemoconcentration during various gestational ages and its functional consequences.
   
3. Establish effective and safe doses of supplemental iron with folic acid and possibly other nutrients for pregnant women using either daily or intermittent preventive supplementation; established doses may need to vary with the early nutritional and haematological status of the mothers and with the particular setting in which supplements are used (e.g. recommended doses may need to be initially higher where iron deficiency and anaemia early in pregnancy are highly prevalent.
   
4. Find effective, safe, and affordable iron compounds with fewer side effects than current iron supplements for use in public health pre-pregnancy and prenatal preventive supplementation programs..
   
5. Conduct large multicenter studies to define effective and safe prenatal supplementation strategies and modalities according to the health and nutritional conditions of women of fertile age, with emphasis on maternal health and pregnancy outcomes. This research should explore supplementation strategies involving different iron doses, different supplementation starting times relative to conception, and both daily and less frequent administration of supplements. These studies could also be used to assess the influence of altitude on pregnant women and newborns' health under different iron or iron+folate supplementation regimens.
   
6. Ideally, given the benefits of women entering pregnancy with iron reserves, a large prospective multicenter study exploring the same independent variables and pregnancy outcomes as in point (5) should be undertaken, including women supplemented in preventive schemes with iron, folic acid and if necessary, with other micronutrients prior to and during pregnancy.
   

 

Acknowledgements

We would like to thank the trial authors who have contributed additional data for this review; and Richard Riley who provided statistical advice. In addition, we would like to thank the staff at the editorial office of the Cochrane Pregnancy and Childbirth Group in Liverpool for their support in the preparation of this review and, in particular, Professor Zarko Alfirevic.

As part of the pre-publication editorial process, this update has been commented on by three peers (an editor, and two referees who are external to the editorial team) and the Group's Statistical Adviser.

 

Data and analyses

Download statistical data

 

Comparison 1. Daily iron alone versus no intervention/placebo

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Low birthweight (less than 2500 g) (ALL)

9

6275

Risk Ratio (M-H, Random, 95% CI)

0.79 [0.61, 1.03]

---

 2 Low birthweight (less than 2500 g) (BY SUBGROUPS)

9

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

2.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

7

5771

Risk Ratio (M-H, Random, 95% CI)

0.81 [0.61, 1.08]

---

2.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

2

504

Risk Ratio (M-H, Random, 95% CI)

0.55 [0.22, 1.38]

---

2.5 Non-anaemic at start of supplementation

6

4452

Risk Ratio (M-H, Random, 95% CI)

0.76 [0.46, 1.25]

---

2.6 Unspecified/mixed anaemic status at start of supplementation

3

1823

Risk Ratio (M-H, Random, 95% CI)

0.82 [0.71, 0.94]

---

2.7 Daily lower dose (60 mg elemental iron or less)

7

3247

Risk Ratio (M-H, Random, 95% CI)

0.79 [0.53, 1.18]

---

2.8 Daily higher dose (more than 60 mg elemental iron)

2

3028

Risk Ratio (M-H, Random, 95% CI)

0.75 [0.45, 1.26]

---

 3 Birthweight (g) (ALL)

10

5956

Mean Difference (IV, Random, 95% CI)

36.05 [-4.84, 76.95]

---

 4 Birthweight (g) (BY SUBGROUPS)

11

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

4.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

9

5822

Mean Difference (IV, Random, 95% CI)

30.89 [-13.87, 75.65]

---

4.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

2

251

Mean Difference (IV, Random, 95% CI)

39.72 [-67.69, 147.12]

---

4.5 Non-anaemic at start of supplementation

8

4496

Mean Difference (IV, Random, 95% CI)

29.32 [-27.08, 85.71]

---

4.6 Unspecified/mixed anaemic status at start of supplementation

3

1577

Mean Difference (IV, Random, 95% CI)

52.33 [10.16, 94.51]

---

4.7 Daily low dose (60 mg elemental iron or less)

6

2804

Mean Difference (IV, Random, 95% CI)

37.22 [-17.82, 92.27]

---

4.8 Daily higher dose (more than 60 mg elemental iron)

6

3382

Mean Difference (IV, Random, 95% CI)

17.99 [-41.28, 77.26]

---

 5 Premature delivery (less than 37 weeks of gestation) (ALL)

8

5730

Risk Ratio (M-H, Random, 95% CI)

0.85 [0.67, 1.09]

---

 6 Premature delivery (less 37 weeks of gestation) (BY SUBGROUPS)

8

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

6.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

6

2989

Risk Ratio (M-H, Random, 95% CI)

0.89 [0.66, 1.20]

---

6.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

47

Risk Ratio (M-H, Random, 95% CI)

0.32 [0.01, 7.48]

---

6.5 Non-anaemic at start of supplementation

6

1775

Risk Ratio (M-H, Random, 95% CI)

0.78 [0.54, 1.12]

---

6.6 Unspecified/mixed anaemic status at start of supplementation

1

1261

Risk Ratio (M-H, Random, 95% CI)

1.04 [0.85, 1.28]

---

6.7 Daily lower dose (60 mg elemental iron or less)

6

3023

Risk Ratio (M-H, Random, 95% CI)

0.89 [0.68, 1.18]

---

6.8 Daily higher dose (more than 60 mg elemental iron)

2

2707

Risk Ratio (M-H, Random, 95% CI)

0.71 [0.48, 1.06]

---

 7 Maternal Hb concentration at term (g/L) (ALL)

17

2463

Mean Difference (IV, Random, 95% CI)

8.83 [6.55, 11.11]

---

 8 Maternal Hb concentration at term (g/L) (BY SUBGROUPS)

17

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

8.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

13

2306

Mean Difference (IV, Random, 95% CI)

8.14 [5.61, 10.67]

---

8.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

3

130

Mean Difference (IV, Random, 95% CI)

10.24 [2.45, 18.04]

---

8.3 Unspecified/mixed gestational age at start of supplementation

1

27

Mean Difference (IV, Random, 95% CI)

14.0 [8.07, 19.93]

---

8.5 Non-anaemic at start of supplementation

11

2002

Mean Difference (IV, Random, 95% CI)

7.77 [4.86, 10.69]

---

8.6 Unspecified/mixed anaemic status at start of supplementation

6

461

Mean Difference (IV, Random, 95% CI)

11.03 [7.51, 14.56]

---

8.7 Daily low dose (60 mg elemental iron or less)

8

1956

Mean Difference (IV, Random, 95% CI)

7.16 [4.14, 10.17]

---

8.8 Daily higher dose (more than 60 mg elemental iron)

9

507

Mean Difference (IV, Random, 95% CI)

10.94 [7.06, 14.82]

---

 9 Anaemia at term (Hb less than 110 g/L) (ALL)

14

4390

Risk Ratio (M-H, Random, 95% CI)

0.27 [0.17, 0.42]

---

 10 Haemoconcentration at term (Hb more than 130 g/L) (ALL)

10

4643

Risk Ratio (M-H, Random, 95% CI)

2.62 [1.21, 5.67]

---

 11 Haemoconcentration at term (Hb more than 130 g/L) (BY SUBGROUPS)

10

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

11.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

5

4173

Risk Ratio (M-H, Random, 95% CI)

1.99 [0.72, 5.45]

---

11.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

3

198

Risk Ratio (M-H, Random, 95% CI)

3.94 [0.31, 50.47]

---

11.3 Unspecified/mixed gestational age at start of supplementation

2

272

Risk Ratio (M-H, Random, 95% CI)

4.67 [2.53, 8.60]

---

11.5 Non-anaemic at start of supplementation

5

4011

Risk Ratio (M-H, Random, 95% CI)

1.73 [0.64, 4.66]

---

11.6 Unspecified/mixed anaemic status at start of supplementation

5

632

Risk Ratio (M-H, Random, 95% CI)

4.03 [1.39, 11.72]

---

11.7 Daily low dose (60 mg elemental iron or less)

4

1317

Risk Ratio (M-H, Random, 95% CI)

1.44 [0.54, 3.86]

---

11.8 Daily higher dose (more than 60 mg elemental iron)

6

3326

Risk Ratio (M-H, Random, 95% CI)

3.55 [1.89, 6.66]

---

 12 Haemoconcentration during second or third trimester (ALL)

10

4841

Risk Ratio (M-H, Random, 95% CI)

2.27 [1.40, 3.70]

---

 13 Haemoconcentration during second or third trimester (BY SUBGROUPS)

10

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

13.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

7

4522

Risk Ratio (M-H, Random, 95% CI)

2.62 [1.49, 4.60]

---

13.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

47

Risk Ratio (M-H, Random, 95% CI)

1.44 [0.72, 2.86]

---

13.3 Unspecified/mixed gestational age at start of supplementation

2

272

Risk Ratio (M-H, Random, 95% CI)

1.94 [0.30, 12.29]

---

13.5 Non-anaemic at start of supplementation

6

4088

Risk Ratio (M-H, Random, 95% CI)

2.10 [1.13, 3.90]

---

13.6 Unspecified/mixed anaemic status at start of supplementation

4

753

Risk Ratio (M-H, Random, 95% CI)

2.67 [0.99, 7.17]

---

13.7 Daily low dose (60 mg elemental iron or less)

5

1655

Risk Ratio (M-H, Random, 95% CI)

2.35 [1.21, 4.57]

---

13.8 Daily higher dose (more than 60 mg elemental iron)

5

3186

Risk Ratio (M-H, Random, 95% CI)

2.14 [1.05, 4.37]

---

 14 Iron deficiency at term (as defined by two or more indicators) (ALL)

6

1108

Risk Ratio (M-H, Random, 95% CI)

0.44 [0.27, 0.70]

---

 15 Iron deficiency at term (as defined by two or more indicators) (BY SUBGROUPS)

6

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

15.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

4

867

Risk Ratio (M-H, Random, 95% CI)

0.56 [0.35, 0.90]

---

15.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

2

241

Risk Ratio (M-H, Random, 95% CI)

0.28 [0.17, 0.44]

---

15.5 Non-anaemic at start of supplementation

4

944

Risk Ratio (M-H, Random, 95% CI)

0.60 [0.41, 0.90]

---

15.6 Unspecified/mixed anaemic status at start of supplementation

2

164

Risk Ratio (M-H, Random, 95% CI)

0.14 [0.07, 0.29]

---

15.7 Daily low dose (60 mg elemental iron or less)

4

944

Risk Ratio (M-H, Random, 95% CI)

0.60 [0.41, 0.90]

---

15.8 Daily higher dose (more than 60 mg elemental iron)

2

164

Risk Ratio (M-H, Random, 95% CI)

0.14 [0.07, 0.29]

---

 16 Iron deficiency anaemia at term (ALL)

6

1667

Risk Ratio (M-H, Random, 95% CI)

0.33 [0.16, 0.69]

---

 17 Iron deficiency anaemia at term (BY SUBGROUPS)

6

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

17.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

5

1622

Risk Ratio (M-H, Random, 95% CI)

0.37 [0.18, 0.76]

---

17.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

45

Risk Ratio (M-H, Random, 95% CI)

0.07 [0.00, 1.13]

---

17.5 Non-anaemic at start of supplementation

5

1547

Risk Ratio (M-H, Random, 95% CI)

0.39 [0.20, 0.74]

---

17.6 Unspecified/mixed anaemic status at start of supplementation

1

120

Risk Ratio (M-H, Random, 95% CI)

0.04 [0.00, 0.72]

---

17.7 Daily low dose (60 mg elemental iron or less)

5

1547

Risk Ratio (M-H, Random, 95% CI)

0.39 [0.20, 0.74]

---

17.8 Daily higher dose (more than 60 mg elemental iron)

1

120

Risk Ratio (M-H, Random, 95% CI)

0.04 [0.00, 0.72]

---

 18 Side-effects (Any) (ALL)

8

3667

Risk Ratio (M-H, Random, 95% CI)

3.92 [1.21, 12.64]

---

 19 Side-effects (Any) (BY SUBGROUPS)

8

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

19.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

4

3034

Risk Ratio (M-H, Random, 95% CI)

3.69 [0.42, 32.71]

---

19.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

3

428

Risk Ratio (M-H, Random, 95% CI)

1.75 [0.99, 3.08]

---

19.3 Unspecified/mixed gestational age at start of supplementation

1

205

Risk Ratio (M-H, Random, 95% CI)

62.79 [3.89, 1013.31]

---

19.5 Non-anaemic at start of supplementation

4

2897

Risk Ratio (M-H, Random, 95% CI)

3.27 [0.39, 27.31]

---

19.6 Unspecified/mixed anaemic status at start of supplementation

4

770

Risk Ratio (M-H, Random, 95% CI)

3.94 [1.09, 14.28]

---

19.7 Daily low dose (60 mg elemental iron or less)

4

566

Risk Ratio (M-H, Random, 95% CI)

1.36 [0.99, 1.87]

---

19.8 Daily higher dose (more than 60 mg elemental iron)

5

3148

Risk Ratio (M-H, Random, 95% CI)

12.12 [1.76, 83.43]

---

 20 Very low birthweight (less than 1500 g) (ALL)

5

2687

Risk Ratio (M-H, Random, 95% CI)

0.73 [0.31, 1.74]

---

 21 Perinatal death (ALL)

3

5036

Risk Ratio (M-H, Random, 95% CI)

0.93 [0.67, 1.29]

---

 24 Infant Hb concentration at 3 months (g/L) (ALL)

1

197

Mean Difference (IV, Random, 95% CI)

Not estimable

---

 25 Infant serum ferritin concentration at 3 months (ug/L) (ALL)

1

197

Mean Difference (IV, Random, 95% CI)

19.0 [2.75, 35.25]

---

 26 Infant Hb concentration at 6 months (g/L) (ALL)

2

533

Mean Difference (IV, Random, 95% CI)

-1.25 [-8.10, 5.59]

---

 27 Infant serum ferritin concentration at 6 months (ug/L) (ALL)

1

197

Mean Difference (IV, Random, 95% CI)

11.0 [4.37, 17.63]

---

 29 Admission to special care unit (ALL)

2

2805

Risk Ratio (M-H, Random, 95% CI)

0.95 [0.73, 1.23]

---

 30 Very premature delivery (less than 34 weeks' gestation) (ALL)

4

1417

Risk Ratio (M-H, Random, 95% CI)

0.44 [0.16, 1.24]

---

 31 Severe anaemia at term (Hb less than 70 g/L) (ALL)

8

1751

Risk Ratio (M-H, Random, 95% CI)

4.83 [0.23, 99.88]

---

 32 Moderate anaemia at term (Hb more than 70 g/L and less than 90 g/L) (ALL)

9

1868

Risk Ratio (M-H, Random, 95% CI)

0.94 [0.55, 1.62]

---

 33 Severe anaemia at any time during second and third trimester (ALL)

9

2089

Risk Ratio (M-H, Random, 95% CI)

0.48 [0.01, 34.52]

---

 34 Moderate anaemia at any time during second or third trimester (ALL)

10

2266

Risk Ratio (M-H, Random, 95% CI)

0.42 [0.19, 0.92]

---

 35 Infection during pregnancy (including urinary tract infections) (ALL)

2

3421

Risk Ratio (M-H, Random, 95% CI)

1.16 [0.83, 1.63]

---

 36 Puerperal infection (ALL)

2

2169

Risk Ratio (M-H, Random, 95% CI)

0.72 [0.25, 2.10]

---

 37 Antepartum haemorraghe (ALL)

2

1157

Risk Ratio (M-H, Random, 95% CI)

1.48 [0.51, 4.31]

---

 38 Postpartum haemorraghe (ALL)

5

1554

Risk Ratio (M-H, Random, 95% CI)

0.95 [0.51, 1.78]

---

 39 Transfusion provided (ALL)

3

3453

Risk Ratio (M-H, Random, 95% CI)

0.61 [0.38, 0.96]

---

 40 Haemoglobin concentration within one month postpartum (ALL)

6

904

Mean Difference (IV, Random, 95% CI)

7.08 [4.70, 9.47]

---

 41 Severe anaemia at postpartum (Hb less than 80 g/L) (ALL)

7

1094

Risk Ratio (M-H, Random, 95% CI)

0.06 [0.00, 1.05]

---

 42 Moderate anaemia at postpartum (Hb more than 80 g/L and less than 100 g/L) (ALL)

4

831

Risk Ratio (M-H, Random, 95% CI)

0.55 [0.12, 2.51]

---

 43 Diarrhoea (ALL)

3

1088

Risk Ratio (M-H, Random, 95% CI)

0.55 [0.32, 0.93]

---

 44 Constipation (ALL)

4

1495

Risk Ratio (M-H, Random, 95% CI)

0.95 [0.62, 1.43]

---

 45 Nausea (ALL)

4

1377

Risk Ratio (M-H, Random, 95% CI)

1.21 [0.72, 2.03]

---

 46 Heartburn (ALL)

3

1323

Risk Ratio (M-H, Random, 95% CI)

1.19 [0.86, 1.66]

---

 47 Vomiting (ALL)

4

1392

Risk Ratio (M-H, Random, 95% CI)

0.88 [0.59, 1.30]

---

 48 Maternal death (death while pregnant or within 42 days of termination of pregnancy) (ALL)

1

47

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 49 Maternal wellbeing/satisfaction (ALL)

2

2604

Risk Ratio (M-H, Random, 95% CI)

1.00 [0.91, 1.09]

---

 50 Placental abruption (ALL)

2

2169

Risk Ratio (M-H, Random, 95% CI)

1.14 [0.32, 4.06]

---

 51 Premature rupture of membranes (ALL)

1

727

Risk Ratio (M-H, Random, 95% CI)

0.99 [0.74, 1.34]

---

 52 Pre-eclampsia (ALL)

2

774

Risk Ratio (M-H, Random, 95% CI)

2.58 [0.81, 8.22]

---

 91 Small for gestational age (less than 10th percentile weight at birth for gestational age) (not pre-specified)

4

2511

Risk Ratio (M-H, Random, 95% CI)

0.87 [0.58, 1.30]

---

 93 Cesarean delivery (not pre-specified)

7

4283

Risk Ratio (M-H, Random, 95% CI)

0.94 [0.78, 1.13]

---

 94 Birth length in cm (not pre-specified)

5

2140

Mean Difference (IV, Random, 95% CI)

0.38 [0.10, 0.65]

---

 95 Forceps or vacuum delivery (not pre-specified)

2

477

Risk Ratio (M-H, Random, 95% CI)

1.50 [0.94, 2.40]

---

 96 Breastfeeding at least 4 months (not pre-specified)

1

48

Risk Ratio (M-H, Random, 95% CI)

1.00 [0.89, 1.13]

---

 97 Haemoglobin concentration at 4-8 weeks' postpartum (g/L) (not pre-specified)

10

1188

Mean Difference (IV, Random, 95% CI)

5.13 [0.24, 10.02]

---

 98 Apgar score < 7 at 5 minutes (not pre-specified)

2

475

Risk Ratio (M-H, Random, 95% CI)

0.74 [0.17, 3.28]

---

 99 Apgar Score at 5 min (not pre-specified)

2

228

Mean Difference (IV, Random, 95% CI)

0.27 [-0.07, 0.62]

---

 

Comparison 2. Intermittent iron alone versus daily iron alone

---

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3 Birthweight (ALL)	1	41	Mean Difference (IV, Random, 95% CI)	-68.0 [-398.33, 262.33]
5 Premature delivery (less than 37 weeks of gestation) (ALL)	1	41	Risk Ratio (M-H, Random, 95% CI)	0.46 [0.02, 8.96]
12 Haemoconcentration during second or third trimester (Hb more than 130 g/L) (ALL)	2	64	Risk Ratio (M-H, Random, 95% CI)	0.54 [0.18, 1.58]
33 Severe anaemia at any time during second and third trimester (Hb less than 70 g/L) (ALL)	2	64	Risk Ratio (M-H, Random, 95% CI)	Not estimable
34 Moderate anaemia at any time during second or third trimester (ALL)	2	64	Risk Ratio (M-H, Random, 95% CI)	2.42 [0.16, 35.56]

---

 

Comparison 3. Daily iron-folic acid versus no intervention/placebo

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Low birthweight (less than 2500 g) (ALL)

2

1368

Risk Ratio (M-H, Random, 95% CI)

1.06 [0.28, 4.02]

---

 2 Low birthweight (less than 2500 g) (BY SUBGROUPS)

2

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

2.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

1368

Risk Ratio (M-H, Random, 95% CI)

1.06 [0.28, 4.02]

---

2.6 Unspecified/mixed anaemic status at start of supplementation

2

1368

Risk Ratio (M-H, Random, 95% CI)

1.06 [0.28, 4.02]

---

2.7 Daily low dose (60 mg elemental iron or less)

1

1320

Risk Ratio (M-H, Random, 95% CI)

0.79 [0.69, 0.91]

---

2.8 Daily higher dose (more than 60 mg elemental iron)

1

48

Risk Ratio (M-H, Random, 95% CI)

5.0 [0.25, 98.96]

---

 3 Birthweight (ALL)

2

1365

Mean Difference (IV, Random, 95% CI)

57.73 [7.66, 107.79]

---

 4 Birthweight (BY SUBGROUPS)

2

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

4.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

1365

Mean Difference (IV, Random, 95% CI)

57.73 [7.66, 107.79]

---

4.6 Unspecified/mixed anaemic status at start of supplementation

2

1365

Mean Difference (IV, Random, 95% CI)

57.73 [7.66, 107.79]

---

4.7 Daily low dose (60 mg elemental iron or less)

1

1320

Mean Difference (IV, Random, 95% CI)

65.0 [17.46, 112.54]

---

4.8 Daily higher dose (more than 60 mg elemental iron)

1

45

Mean Difference (IV, Random, 95% CI)

-32.0 [-213.62, 149.62]

---

 5 Premature delivery (less than 37 weeks of gestation) (ALL)

3

1497

Risk Ratio (M-H, Random, 95% CI)

1.55 [0.40, 6.00]

---

 6 Premature delivery (less than 37 weeks of gestation) (BY SUB GROUPS)

2

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

6.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

1362

Risk Ratio (M-H, Random, 95% CI)

1.13 [0.92, 1.39]

---

6.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

6.6 Unspecified/mixed anaemic status at start of supplementation

2

1449

Risk Ratio (M-H, Random, 95% CI)

1.13 [0.92, 1.39]

---

6.7 Daily low dose (60 mg elemental iron or less)

2

1449

Risk Ratio (M-H, Random, 95% CI)

1.13 [0.92, 1.39]

---

 7 Haemoglobin concentration at term (ALL)

4

179

Mean Difference (IV, Random, 95% CI)

12.00 [2.93, 21.07]

---

 8 Haemoglobin concentration at term (BY SUBGROUPS)

4

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

8.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

93

Mean Difference (IV, Random, 95% CI)

15.65 [11.84, 19.46]

---

8.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

2

86

Mean Difference (IV, Random, 95% CI)

7.92 [-11.68, 27.52]

---

8.5 Non-anaemic at start of supplementation

1

48

Mean Difference (IV, Random, 95% CI)

17.10 [8.44, 25.76]

---

8.6 Unspecified/mixed anaemic status at start of supplementation

3

131

Mean Difference (IV, Random, 95% CI)

10.47 [-1.07, 22.00]

---

8.7 Daily low dose (60 mg elemental iron or less)

0

0

Mean Difference (IV, Random, 95% CI)

Not estimable

---

8.8 Daily higher dose (more than 60 mg elemental iron)

4

179

Mean Difference (IV, Random, 95% CI)

12.00 [2.93, 21.07]

---

 9 Anaemia at term (Hb less than 110 g/L) (ALL)

3

346

Risk Ratio (M-H, Random, 95% CI)

0.27 [0.12, 0.56]

---

 10 Haemoconcentration at term (Hb more than 130 g/L) (ALL)

3

353

Risk Ratio (M-H, Random, 95% CI)

1.74 [0.34, 8.94]

---

 11 Haemoconcentration at term (Hb more than 130 g/L) (BY SUBGROUPS)

3

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

11.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

1

75

Risk Ratio (M-H, Random, 95% CI)

3.37 [0.19, 60.03]

---

11.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

3

298

Risk Ratio (M-H, Random, 95% CI)

1.97 [0.32, 12.08]

---

11.6 Unspecified/mixed anaemic status at start of supplementation

1

131

Risk Ratio (M-H, Random, 95% CI)

4.31 [0.26, 70.41]

---

11.7 Daily low dose (60 mg elemental iron or less)

1

131

Risk Ratio (M-H, Random, 95% CI)

4.31 [0.26, 70.41]

---

11.8 Daily higher dose (more than 60 mg elemental iron)

2

222

Risk Ratio (M-H, Random, 95% CI)

1.28 [0.24, 6.78]

---

 12 Haemoconcentration during second or third trimester (ALL)

2

446

Risk Ratio (M-H, Random, 95% CI)

1.78 [0.63, 5.04]

---

 13 Haemoconcentration during second or third trimester (BY SUBGROUPS)

2

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

13.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

390

Risk Ratio (M-H, Random, 95% CI)

1.45 [0.33, 6.32]

---

13.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

76

Risk Ratio (M-H, Random, 95% CI)

1.34 [0.50, 3.56]

---

13.3 Unspecified/mixed gestational age at start of supplementation

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

13.5 Non-anaemic at start of supplementation

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

13.6 Unspecified/mixed anaemic status at start of supplementation

2

446

Risk Ratio (M-H, Random, 95% CI)

1.78 [0.63, 5.04]

---

13.7 Daily low dose (60 mg elemental iron or less)

2

446

Risk Ratio (M-H, Random, 95% CI)

1.78 [0.63, 5.04]

---

13.8 Daily higher dose (more than 60 mg elemental iron)

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 14 Iron deficiency at term (as defined by two or more indicators) (ALL)

1

131

Risk Ratio (M-H, Random, 95% CI)

0.24 [0.06, 0.99]

---

 15 Iron deficiency at term (as defined by two or more indicators) (BY SUBGROUPS)

1

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

15.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

1

75

Risk Ratio (M-H, Random, 95% CI)

0.12 [0.01, 1.10]

---

15.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

76

Risk Ratio (M-H, Random, 95% CI)

0.36 [0.08, 1.63]

---

15.5 Non-anaemic at start of supplementation

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

15.6 Unspecified/mixed anaemic status at start of supplementation

1

131

Risk Ratio (M-H, Random, 95% CI)

0.24 [0.06, 0.99]

---

15.7 Daily low dose (60 mg elemental iron or less)

1

131

Risk Ratio (M-H, Random, 95% CI)

0.24 [0.06, 0.99]

---

15.8 Daily higher dose (more than 60 mg elemental iron)

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 16 Iron deficiency anaemia at term (ALL)

1

131

Risk Ratio (M-H, Random, 95% CI)

0.43 [0.17, 1.09]

---

 17 Iron deficiency anaemia at term (BY SUBGROUPS)

1

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

17.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

1

75

Risk Ratio (M-H, Random, 95% CI)

0.36 [0.12, 1.12]

---

17.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

76

Risk Ratio (M-H, Random, 95% CI)

0.5 [0.18, 1.40]

---

17.5 Non-anaemic at start of supplementation

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

17.6 Unspecified/mixed anaemic status at start of supplementation

1

131

Risk Ratio (M-H, Random, 95% CI)

0.43 [0.17, 1.09]

---

17.7 Daily low dose (60 mg elemental iron or less)

1

131

Risk Ratio (M-H, Random, 95% CI)

0.43 [0.17, 1.09]

---

17.8 Daily higher dose (more than 60 mg elemental iron)

0

0

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 18 Side effects (Any) (ALL)

1

456

Risk Ratio (M-H, Random, 95% CI)

44.32 [2.77, 709.09]

---

 20 Very low birthweight (less than 1500 g) (ALL)

1

48

Risk Ratio (M-H, Random, 95% CI)

5.0 [0.25, 98.96]

---

 21 Perinatal death (ALL)

3

1862

Risk Ratio (M-H, Random, 95% CI)

0.83 [0.58, 1.17]

---

 29 Admission to special care unit (ALL)

1

48

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 30 Very premature delivery (less than 34 weeks' gestation) (ALL)

2

92

Risk Ratio (M-H, Random, 95% CI)

5.0 [0.25, 98.96]

---

 31 Severe anaemia at term (Hb less than 70 g/L) (ALL)

3

180

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 32 Moderate anaemia at term (Hb more than 70g/L and less than 90 g/L) (ALL)

3

180

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 33 Severe anaemia at any time during second and third trimester (Hb less than 70 g/L) (ALL)

4

523

Risk Ratio (M-H, Random, 95% CI)

0.11 [0.01, 0.83]

---

 34 Moderate anaemia at any time during second or third trimester (ALL)

4

523

Risk Ratio (M-H, Random, 95% CI)

0.34 [0.11, 1.04]

---

 35 Infection during pregnancy (including urinary tract infections) (ALL)

1

48

Risk Ratio (M-H, Random, 95% CI)

1.0 [0.15, 6.53]

---

 36 Puerperal infection (ALL)

1

2863

Risk Ratio (M-H, Random, 95% CI)

0.55 [0.13, 2.28]

---

 37 Antepartum haemorrhage (ALL)

2

145

Risk Ratio (M-H, Random, 95% CI)

1.25 [0.22, 7.12]

---

 38 Postpartum haemorrhage (ALL)

1

68

Risk Ratio (M-H, Random, 95% CI)

0.12 [0.00, 2.71]

---

 40 Haemoglobin concentration within one month postpartum (ALL)

1

45

Mean Difference (IV, Random, 95% CI)

10.40 [4.03, 16.77]

---

 41 Severe anaemia at postpartum (Hb less than 80 g/L) (ALL)

3

525

Risk Ratio (M-H, Random, 95% CI)

0.05 [0.00, 0.76]

---

 42 Moderate anaemia at postpartum (Hb more than 80 g/L and less than 100 g/L) (ALL)

3

525

Risk Ratio (M-H, Random, 95% CI)

0.34 [0.17, 0.69]

---

 48 Maternal death (death while pregnant or within 42 days of termination of pregnancy) (ALL)

1

131

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 50 Placental abruption (ALL)

1

2863

Risk Ratio (M-H, Random, 95% CI)

8.19 [0.49, 138.16]

---

 52 Pre-eclampsia (ALL)

1

48

Risk Ratio (M-H, Random, 95% CI)

3.00 [0.13, 70.16]

---

 91 Small for gestational age (less than 10th percentile weight at birth for gestational age) (not pre-specified)

1

1318

Risk Ratio (M-H, Random, 95% CI)

0.88 [0.80, 0.97]

---

 92 Oedema during pregnancy (not pre-specified)

1

67

Risk Ratio (M-H, Random, 95% CI)

2.82 [0.99, 8.09]

---

 93 Caesarean delivery (not pre-specified)

1

97

Risk Ratio (M-H, Random, 95% CI)

0.83 [0.22, 3.13]

---

 94 Birth length in cm (not pre-specified)

1

1320

Mean Difference (IV, Random, 95% CI)

0.20 [-0.06, 0.46]

---

 97 Haemoglobin concentration at 4-8 weeks postpartum (not prespecified)

3

526

Mean Difference (IV, Random, 95% CI)

4.88 [-0.85, 10.62]

---

 

Comparison 4. Intermittent iron-folic acid versus daily iron-folic acid

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Low birthweight (less than 2500 g) (ALL)

4

730

Risk Ratio (M-H, Random, 95% CI)

1.05 [0.58, 1.91]

---

 2 Low birthweight (less than 2500 g) (BY SUBGROUPS)

4

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

2.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

455

Risk Ratio (M-H, Random, 95% CI)

1.29 [0.55, 3.01]

---

2.3 Unspecified/mixed gestational age at start of supplementation

2

275

Risk Ratio (M-H, Random, 95% CI)

0.85 [0.36, 1.99]

---

2.5 Non-anaemic at start of supplementation

1

80

Risk Ratio (M-H, Random, 95% CI)

1.25 [0.36, 4.32]

---

2.7 Daily low dose (60 mg elemental iron or less)

3

650

Risk Ratio (M-H, Random, 95% CI)

0.99 [0.50, 1.97]

---

2.8 Daily higher dose (more than 60 mg elemental iron)

1

80

Risk Ratio (M-H, Random, 95% CI)

1.25 [0.36, 4.32]

---

 3 Birthweight (ALL)

4

730

Mean Difference (IV, Random, 95% CI)

-7.10 [-67.20, 53.01]

---

 4 Birthweight (BY SUBGROUPS)

4

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

4.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

455

Mean Difference (IV, Random, 95% CI)

1.19 [-70.50, 72.87]

---

4.3 Unspecified/mixed gestational age at start of supplementation

2

275

Mean Difference (IV, Random, 95% CI)

-26.71 [-137.00, 83.58]

---

4.5 Non-anaemic at start of supplementation

1

80

Mean Difference (IV, Random, 95% CI)

Not estimable

---

4.7 Daily low dose (60 mg elemental iron or less)

3

650

Mean Difference (IV, Random, 95% CI)

-8.36 [-73.56, 56.85]

---

4.8 Daily higher dose (more than 60 mg elemental iron)

1

80

Mean Difference (IV, Random, 95% CI)

Not estimable

---

 5 Premature delivery (less than 37 weeks of gestation) (ALL)

1

80

Risk Ratio (M-H, Random, 95% CI)

2.0 [0.39, 10.31]

---

 7 Haemoglobin concentration at term (ALL)

3

475

Mean Difference (IV, Random, 95% CI)

-0.83 [-4.74, 3.08]

---

 8 Haemoglobin concentration at term (BY SUBGROUPS)

3

Mean Difference (IV, Random, 95% CI)

Subtotals only

---

8.3 Unspecified/mixed gestational age at start of supplementation

2

301

Mean Difference (IV, Random, 95% CI)

-2.19 [-6.85, 2.47]

---

8.7 Daily low dose (60 mg elemental iron or less)

3

422

Mean Difference (IV, Random, 95% CI)

-0.10 [-5.15, 4.95]

---

8.8 Daily higher dose (more than 60 mg elemental iron)

1

109

Mean Difference (IV, Random, 95% CI)

-0.82 [-4.99, 3.35]

---

 9 Anaemia at term (Hb < 110 g/L) (ALL)

3

475

Risk Ratio (M-H, Random, 95% CI)

1.20 [0.78, 1.83]

---

 10 Haemoconcentration at term (Hb more than 130 g/L) (ALL)

3

475

Risk Ratio (M-H, Random, 95% CI)

0.93 [0.47, 1.82]

---

 11 Haemoconcentration at term (Hb more than 130 g/L) (BY SUBGROUPS)

3

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

11.7 Daily low dose (60 mg elemental iron or less)

3

422

Risk Ratio (M-H, Random, 95% CI)

1.24 [0.42, 3.66]

---

11.8 Daily higher dose (more than 60 mg elemental iron)

1

109

Risk Ratio (M-H, Random, 95% CI)

0.63 [0.19, 2.11]

---

 12 Haemoconcentration during second or third trimester (Hb more than 130 g/L) (ALL)

6

1111

Risk Ratio (M-H, Random, 95% CI)

0.43 [0.24, 0.77]

---

 13 Haemoconcentration during second or third trimester (Hb more than 130 g/L) (BY SUBGROUPS)

6

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

13.1 Early gestational age (less than 20 weeks of gestation or pre-pregnancy) at start of supplementation

2

250

Risk Ratio (M-H, Random, 95% CI)

0.52 [0.26, 1.01]

---

13.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

166

Risk Ratio (M-H, Random, 95% CI)

0.24 [0.10, 0.55]

---

13.3 Unspecified/mixed gestational age at start of supplementation

3

695

Risk Ratio (M-H, Random, 95% CI)

0.46 [0.13, 1.65]

---

13.4 Anaemic at start of supplementation (Hb <110 g/L if in first or <105 g/L if in second trimester)

1

47

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

13.5 Non-anaemic at start of supplementation

1

80

Risk Ratio (M-H, Random, 95% CI)

0.6 [0.15, 2.34]

---

13.6 Unspecified/mixed anaemic status at start of supplementation

4

966

Risk Ratio (M-H, Random, 95% CI)

0.41 [0.21, 0.80]

---

13.7 Daily low dose (60 mg elemental iron or less)

5

953

Risk Ratio (M-H, Random, 95% CI)

0.41 [0.20, 0.82]

---

13.8 Daily higher dose (more than 60 mg elemental iron)

2

247

Risk Ratio (M-H, Random, 95% CI)

0.74 [0.42, 1.31]

---

 16 Iron deficiency anaemia at term (based on two or more indicators) (ALL)

1

156

Risk Ratio (M-H, Random, 95% CI)

0.71 [0.08, 6.63]

---

 18 Side effects (any) (ALL)

7

1307

Risk Ratio (M-H, Random, 95% CI)

0.69 [0.45, 1.04]

---

 19 Side effects (any) (BY SUBGROUPS)

7

Risk Ratio (M-H, Random, 95% CI)

Subtotals only

---

19.1 Early gestational age (less than 20 weeks or pre-pregnancy) at start of supplementation

1

80

Risk Ratio (M-H, Random, 95% CI)

0.2 [0.08, 0.53]

---

19.2 Late gestational age (20 weeks or more of gestation) at start of supplementation

1

172

Risk Ratio (M-H, Random, 95% CI)

1.0 [0.79, 1.27]

---

19.3 Unspecified/mixed gestational age at start of supplementation

5

1055

Risk Ratio (M-H, Random, 95% CI)

0.72 [0.42, 1.26]

---

19.4 Non-anaemic at start of supplementation

1

80

Risk Ratio (M-H, Random, 95% CI)

0.2 [0.08, 0.53]

---

19.7 Daily low dose (60 mg elemental iron or less)

6

1171

Risk Ratio (M-H, Random, 95% CI)

0.92 [0.71, 1.19]

---

19.8 Daily higher dose (more than 60 mg elemental iron)

2

253

Risk Ratio (M-H, Random, 95% CI)

0.14 [0.07, 0.25]

---

 20 Very low birthweight (less than 1500 g) (ALL)

4

737

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 21 Perinatal death (ALL)

1

80

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 27 Infant ferritin concentration at 6 months (ug/L) (ALL)

1

88

Mean Difference (IV, Random, 95% CI)

0.09 [0.05, 0.13]

---

 30 Very premature delivery (less than 34 weeks of gestation) (ALL)

1

111

Risk Ratio (M-H, Random, 95% CI)

0.98 [0.06, 15.31]

---

 31 Severe anaemia at term (Hb less than 70 g/L) (ALL)

4

555

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 32 Moderate anaemia at term (Hb more than 70g/L and less than 90 g/L) (ALL)

3

475

Risk Ratio (M-H, Random, 95% CI)

1.03 [0.07, 16.23]

---

 33 Severe anaemia at any time during second and third trimester (Hb less than 70 g/L) (ALL)

6

1240

Risk Ratio (M-H, Random, 95% CI)

Not estimable

---

 34 Moderate anaemia at any time during second or third trimester (ALL)

6

1111

Risk Ratio (M-H, Random, 95% CI)

2.54 [0.63, 10.17]

---

 37 Antepartum haemorraghe (ALL)

1

110

Risk Ratio (M-H, Random, 95% CI)

1.0 [0.06, 15.59]

---

 41 Severe anaemia at postpartum (Hb less than 80 g/L) (ALL)

1

169

Risk Ratio (M-H, Random, 95% CI)

0.43 [0.04, 4.64]

---

 42 Moderate anaemia at postpartum (Hb more than 80 g/L and less than 100 g/L) (ALL)

1

169

Risk Ratio (M-H, Random, 95% CI)

1.14 [0.26, 4.95]

---

 43 Diarrhoea (ALL)

4

553

Risk Ratio (M-H, Random, 95% CI)

0.94 [0.38, 2.34]

---

 44 Constipation (ALL)

4

553

Risk Ratio (M-H, Random, 95% CI)

0.99 [0.48, 2.06]

---

 45 Nausea (ALL)

5

854

Risk Ratio (M-H, Random, 95% CI)

0.59 [0.30, 1.16]

---

 46 Heartburn (ALL)

3

473

Risk Ratio (M-H, Random, 95% CI)

0.78 [0.29, 2.06]

---

 47 Vomiting (ALL)

5

854

Risk Ratio (M-H, Random, 95% CI)

1.44 [0.82, 2.53]

---

 50 Placental abruption (ALL)

1

110

Risk Ratio (M-H, Random, 95% CI)

0.33 [0.01, 8.01]

---

 51 Premature rupture of membranes (ALL)

1

80

Risk Ratio (M-H, Random, 95% CI)

0.33 [0.01, 7.95]

---

 68 Ln (serum ferritin concentration) 4-8 wk postpartum (not pre-specified)

1

160

Mean Difference (IV, Random, 95% CI)

-0.13 [-0.42, 0.16]

---

 70 Low serum ferritin concentration at postpartum (4-8 wk) (not pre-specified)

1

146

Risk Ratio (M-H, Random, 95% CI)

1.19 [0.40, 3.57]

---

 71 High serum transferrin receptors at 6 weeks postpartum (not pre-specified)

1

146

Risk Ratio (M-H, Random, 95% CI)

0.69 [0.36, 1.33]

---

 93 Caesarean delivery (not pre-specified)

1

80

Risk Ratio (M-H, Random, 95% CI)

0.89 [0.38, 2.07]

---

 97 Haemoglobin concentration at 4-8 weeks postpartum (not pre-specified)

1

146

Mean Difference (IV, Random, 95% CI)

2.0 [-3.86, 7.86]

---

 

Feedback

 

Hemminki, June 2008

 

Summary

My trial, Hemminki 1989a, is excluded from this review and it is not clear why. The comment in Characteristics of excluded studies is "Only women who were anaemic received iron in the unsupplemented group thus making any comparisons among the groups biased for the purposes of this review."

What bias is being referred to? Hemminki 1989a was in the previous version of this review. It was a randomised trial, analysed by intention to treat, having outcome data for all women randomised, and a high compliance (about 80% of women in both groups received the treatment they were allocated to). The 20% of women who received iron in the non-routine supplementation group was as expected.

There are two options for dealing with women whose haemoglobin falls below a pre-specified cut-off in the non-routine supplemented group:
1.  give them iron, as in my study where 20% of women in the non-routine treatment group had iron; or
2.  call those who take iron non-compliant and do the analysis by intention to treat, as did some of the included studies.

What is the difference between these two strategies? They seem to me to be essentially the same.

The effect of routine iron therapy on substantive health outcomes remains unclear. It is a real pity that you have excluded Hemminki 1989a, based on criteria I consider inappropriate: it had a large number of women, several health outcomes including long term follow up, and was well conducted.

A minor issue is that it is misleading to call this trial Hemminki 1989a. Although the study design was published in 1989, the main results were not published until 1991. Hence a more appropriate study identifier would be ‘Hemminki 1991’.

(Summary of feedback from Elina Hemminki, June 2008)

 

Reply

We agree that your trial was well conducted, had a large number of women and looked at several health outcomes including long term follow up.  We did review all publications on the work you have conducted on assessing the effects of routine versus selective iron supplementation during pregnancy. This systematic review aims to assess the effectiveness and safety of daily and intermittent use of iron supplements by pregnant women, either alone or in conjunction with folic acid given as a preventive universal measure. Your trial provided 100 mg of elemental iron daily with various choice of iron compounds and dosage as determined individually by the midwives to all women in the routine iron supplementation group. For women in the "selective iron supplementation group", treatment with iron supplements as slow release form for two months or until the hematocrit increased to 0.32 was provided only to those whose hematocrit was lower than 0.30 on two consecutive visits. Consequently, we have included your trial in the included studies and we thank you for the additional data you have provided us for this analysis. Your study compared the effects of routine versus selective iron supplementation, an issue that certainly deserves better understanding and that reflects current practices.   

We have changed the study identifier to Hemminki 1991 as requested.

 

Contributors

Juan Pablo Pena-Rosas, MD, PhD, MPH

 

What's new

Last assessed as up-to-date: 24 June 2009.

---

Date	Event	Description
16 June 2009	New citation required but conclusions have not changed	In this update, trials assessing the effect of iron or folic acid when given in combination with other micronutrients were included as long as both groups being compared in the daily regimens received the same other micronutrient interventions. This has resulted in four trials previously excluded now being included (Cantlie 1971; Christian 2003; Hemminki 1991; Siega-Riz 2001).
16 June 2009	New search has been performed	Search updated. Ten new trials included (Cantlie 1971; Christian 2003; Hemminki 1991; Harvey 2007; Lee 2005; Meier 2003; Mukhopadhyay 2004; Siega-Riz 2001; Ziaei 2007; Ziaei 2008). One trial included is now excluded (Ortega-Soler 1998). Twenty-seven new trials excluded.
20 October 2008	Feedback has been incorporated	Feedback from Elina Hemminki added with response from author.

---

 

History

Protocol first published: Issue 2, 2004
Review first published: Issue 3, 2006

---

Date	Event	Description
15 April 2008	Amended	Converted to new review format.

---

 

Contributions of authors

Juan Pablo Pena-Rosas and Fernando Viteri co-wrote the protocol, the review and the update. Juan Pablo Pena-Rosas abstracted the trial data and carried out the analysis with the technical support and guidance of Fernando Viteri. Both took primary responsibility in producing the final manuscript.

Disclaimer: Juan Pablo Pena-Rosas is currently a staff member of the World Health Organization. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the decisions, policy or views of the World Health Organization.

 

Declarations of interest

We certify that we have no affiliations with or involvement in any organisation or entity with a direct financial interest in the subject matter of the review (e.g. employment, consultancy, stock ownership, honoraria, expert testimony).

Fernando Viteri was involved in some included studies with intermittent iron supplementation. Juan Pablo Pena-Rosas was author of an excluded study on iron and folic acid intermittent supplementation.

 

Sources of support

 

Internal sources

• Children's Hospital and Oakland Research Institute (CHORI), USA.
  
• International Micronutrient Malnutrition Prevention and Control Program (IMMPaCt) - U.S. Centers for Disease Control and Prevention (CDC), USA.
  

 

External sources

• Department of Reproductive Health and Research, World Health Organization (WHO), Switzerland.
  

 

Differences between protocol and review

This review aims to evaluate the effectiveness of supplementation with iron alone or in combination with folic acid on functional outcomes in the mother and the infant rather than just haematological indicators. It also evaluates the regimen schedules by comparing intermittent (less frequent than daily) supplement intake with the standard daily regimens and the effects of these interventions on side-effects and haemoconcentration. Unlike the previous version of this review, in this updated version, trials assessing the effect of iron or folic acid when given in combination with other micronutrients were included as long as both groups being compared in the daily regimens received the same other micronutrient interventions, which has led to some previously excluded studies now being included.

The comparisons in this review were reduced to four instead of eight as stated in the original protocol. The subgroup analyses were considered only for the primary outcomes.

 

Index terms

Medical Subject Headings (MeSH)

*Dietary Supplements [adverse effects]; Anemia, Iron-Deficiency [prevention & control]; Folic Acid [*administration & dosage]; Iron [*administration & dosage]; Pregnancy Complications, Hematologic [prevention & control]; Pregnancy Outcome; Prenatal Care [methods]; Randomized Controlled Trials as Topic

MeSH check words

Female; Humans; Pregnancy

* Indicates the major publication for the study