1. Objectives

To evaluate the effects and safety of iron supplementation, with or without folic acid, in children living in areas hyperendemic or holoendemic for malaria transmission

2. How studies were identified

The following databases were searched to February 2015:

• Cochrane Infectious Diseases Group Specialized Register
• CENTRAL (The Cochrane Library, February 2015)
• MEDLINE
• EMBASE
• LILACS
• WHO International Clinical Trials Registry Platform (ICTRP)
• metaRegister of Controlled Trials (mRCT)

Reference lists were also searched and the authors directly contacted the primary investigators of included trials, ongoing trials and those awaiting assessment for unpublished data

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials and cluster-randomized controlled trials

(Cluster-randomized trials were only considered eligible if they included two or more units per trial arm)

3.2 Study participants

Children aged less than 18 years, regardless of anaemia or malaria status at baseline, living in hyperendemic or holoendemic areas for malaria transmission

(Studies were excluded if they were conducted in hypoendemic or mesoendemic areas unless malaria outcomes were reported, or if the trial was conducted in an area or time period with no malaria activity)

3.3 Interventions

Oral iron, with or without folic acid or antimalarial treatment, compared to placebo, no treatment, or antimalarial treatment alone

(Interventions where iron was administered as a supplement or a fortified food or drink providing a dose of at least 80% of the WHO recommended dietary allowance for prevention of anaemia by age were included)

(Studies with micronutrient co-interventions were included only if both treatment arms received the co-intervention)

3.4 Primary outcomes

Malaria

• Clinical malaria: uncomplicated malaria, defined as a history of fever with parasitological confirmation
• Severe malaria: cerebral malaria or acute Plasmodium falciparum malaria with signs of severity, or evidence of vital organ dysfunction, or both

(Cases of severe malaria could be included with clinical malaria cases if they were not reported separately)

Other outcomes

• Deaths

Secondary outcomes included malaria parasitaemia (asymptomatic parasitaemia) or parasitological failure, malaria parasite density, hospitalizations for any cause, clinic visits, haemoglobin levels, the prevalence of study-defined anaemia, infections other than malaria, weight and height

4. Main results

4.1 Included studies

Thirty-five randomized controlled trials, enrolling 31,955 children, were included in this review:

• Twenty-six studies were individually randomized and nine were cluster-randomized
• Unpublished data on malaria and death outcomes were obtained for one cluster-randomized trial
• All trials assessed iron or iron plus folic acid for the prevention or treatment of anaemia among children without an acute illness; mean iron supplementation was 2 mg/kd/day and mean duration was 4.5 months (1 to 12 months)
• Co-interventions included antimalarial treatment in the iron arm or both study arms (27 trials), anthelminthics in both arms (12 trials), and micronutrients in both arms (8 trials)
• The mean baseline parasitaemia rate was 45% (0 to 70%); mean baseline haemoglobin was <10 g/dL in 17 trials and ≥10 g/dL in 22 trials
• Children were less than two years of age in 12 trials, two to five years in 11 trials and over five years in 16 trials
• Twelve trials reported predominantly (>80%) on P. falciparum

4.2 Study settings

• Benin (2 trials), Cambodia, Côte d’Ivoire, Ethiopia (2 trials), Gambia (2 trials), Ghana (2 trials), Indonesia, Kenya (6 trials), Malawi (2 trials), Mali (2 trials), Nigeria, Papua New Guinea, Peru, Togo, the United Republic of Tanzania (6 trials), Viet Nam (3 trials), and Zambia
• Trials were conducted in mesoendemic, hyperendemic, and holoendemic areas for malaria

4.3 Study settings

How the data were analysed
Analyses were stratified according to intervention type: i) iron versus placebo/no treatment; ii) iron plus folic acid versus placebo; iii) iron with or without folic acid versus placebo; and iv) iron plus antimalarial versus placebo. Where multiple episodes of an outcome were reported for individuals, the number of individual children with at least one event was extracted, except for the outcomes infectious episodes other than malaria, hospitalizations, and clinic visits, where repeated episodes were counted. Pooled risk ratios (RR) and 95% confidence intervals (CI) were calculated for dichotomous data, while the risk difference (RD) was calculated for the outcome all-cause mortality to permit the inclusion of studies with no deaths in either intervention arm in the analysis. Standardized mean differences (SMD) were calculated for the outcomes of weight and height, to enable the combined analysis of absolute values and Z-scores. Trials with several arms could be included more than once for different comparisons, but individual trial arms were not included more than once in the same meta-analysis. Adjustments for clustering were made using standard methods. To investigate heterogeneity, the following subgroup analyses were planned:

• By anaemia at baseline: mean haemoglobin <10 g/dL, mean haemoglobin ≥10 g/dL
• By age group: < two years, two to five years, and > five years
• By malaria management strategy: treated bed nets, prophylactic antimalarials, standardized diagnosis, standardized treatment

Sensitivity analyses were also conducted by restricting studies to those in which >85% of malaria species diagnosed were P. falciparum, and by method of allocation concealment

Results
Iron versus placebo/no treatment for the treatment or prevention of anaemia (31 trials)
Clinical and severe malaria
There was no evidence of a statistically significant difference between treatment and control groups in the risk of clinical malaria (RR 0.93, 95% CI [0.87 to 1.00], p=0.054; 14 trials/7168 children). In subgroup and sensitivity analyses, the effect became statistically significant in children below two years of age (RR 0.89, 95% CI [0.82 to 0.97]), when restricted to trials describing malaria caused by P. falciparum (RR 0.91, 95%CI [0.84 to 0.99], 9 trials/5503 children), and in trials reporting on severe malaria (clinical malaria with high-grade parasitaemia) (RR 0.90, 95% CI [0.81 to 0.98], 6 trials/3421 children). No meaningful difference in effect by anaemia status was found.

Deaths
Overall, there was no evidence of a difference in risk of all-cause mortality between the iron and control groups (RD 0.00, 95% CI [0.00 to 0.01], p=0.56; 21 comparisons/7576 children).

Additional outcomes
The risk of any parasitaemia at end of follow-up was 23% greater with iron treatment in five in trials of 1150 children (RR 1.23, 95% CI [1.09 to 1.40]), and was also 22% higher at end of treatment in trials without adequate allocation concealment (RR 1.22, 95% CI [1.06 to 1.40], p=0.005; 5 trials). In one trial of 111 children, the number of days with fever per month was increased more than eight-fold with iron treatment (RR 8.37, 95% CI [1.91 to 36.58]). Diarrhoeal episodes per month were increased with iron supplementation (RR 1.15, 95% CI [1.06 to 1.26], 8 trials/23912 children); this finding was rendered non-significant when iron without zinc was considered, while the effect size increased for iron plus zinc (RR 1.29, 95% CI [1.15 to 1.44], 3 trials/6346 children). Haemoglobin at the end of treatment was greater by 0.75 g/dL among those receiving iron (95% CI [0.48 to 1.01 g/dL], 16 comparisons/5261 individuals), and the change in haemoglobin was also greater with iron treatment (MD 0.67 g/dL, 95% CI [0.42 to 0.92], 12 trials/2462 individuals). The risk of anaemia was reduced by 37% among those receiving iron (RR 0.63, 95% CI [0.49 to 0.82], 15 comparisons/3784 children). Change in weight was improved in four trials including 486 children (SMD 0.31, 95% CI [0.13 to 0.49]). No evidence of a significant difference between groups was found for other secondary outcomes.

Iron plus folic acid versus placebo or no treatment (6 trials)
Clinical malaria
Opposing effects were noted between two groups in the single study that reported on clinical and severe malaria, with the main study group showing a higher risk for severe malaria with iron plus folic acid and the substudy group showing a lower risk. The original researchers noted that children in the substudy were monitored and treated for malaria whereas the children in the main study were not.

Deaths
The RD for all-cause mortality was 0.00 per 1000 children (95%CI [-0.00 to 0.01], p=0.31; 5 trials/18,034 children).

Additional outcomes
Haemoglobin at end of treatment was 0.90 g/dL greater with allocation to iron and folic acid (95% CI [0.51 to 1.29 g/dL], 1 trial/124 children), and the risk of anaemia was halved (RR 0.49, 95% CI [0.25 to 0.99], 3 trials/633 children).

Iron with or without folic acid versus placebo for treatment or prevention of anaemia (15 trials)
Clinical malaria
Overall, there was no evidence of a difference between treatment groups in the risk of clinical malaria (RR 0.97, 95% CI [0.91 to 1.03], p=0.26; 16 comparisons). When restricted to settings in which malaria prevention or management services were present, a statistically significant reduction in risk was observed (RR 0.91, 95% CI [0.84 to 0.97], p=0.0075; 11 comparisons), while for settings in which these services were absent, a statistically significant increase in risk was found (RR 1.16, 95% CI [1.02 to 1.31], p=0.019; 5 trials).

Iron plus antimalarial versus placebo (4 trials)
Clinical malaria
Three trials with antimalarial treatment in both iron and placebo arms reported on clinical malaria. Relative to the control group, the iron group had a statistically significantly reduced risk of malaria (RR 0.54, 95% CI [0.43 to 0.67], p<0.00001; 728 children).

Deaths
No difference in the risk of all-cause mortality was found between treatment and control groups in pooled analysis (RR 1.05, 95% CI [0.52 to 2.11], p=0.90; 3 trials/728 children).

Additional outcomes
Hospitalizations and clinic visits were statistically significantly reduced with iron treatment in two trials of 5904 children (hospitalizations: RR 0.59, 95% CI [0.48 to 0.73]; clinic visits: RR 0.88, 95% CI [0.82 to 0.95]). The MD in haemoglobin at the end of treatment was 0.91 g/dL in one trial of 151 children (95% CI [0.47 to 1.35 g/dL), and the risk of anaemia at the end of treatment was reduced by 56% in two trials of 295 children (RR 0.44, 95% CI [0.28 to 0.70]), and by 63% in one trial of 420 children (RR 0.36, 95% CI [0.26 to 0.54]).

5. Additional author observations*

The overall quality of evidence for the main comparison (iron versus placebo/no treatment) was considered to be high for the outcome clinical malaria and severe malaria and low for deaths, as few trials reported on mortality, and most of those that did reported on death only among children who were not lost to follow-up. Although the combination of iron and folic acid resulted in a higher rate of admissions for malaria in one large trial, the relative contributions of the addition of folic acid, the poor system for diagnosis and surveillance of malaria, and the opposing finding in a substudy of this trial render these findings difficult to interpret.

Overall, iron treatment alone did not result in an increased risk of clinical malaria, and iron treatment with or without folate was also safe where malaria prevention or management programmes have been implemented. However, iron plus folate elevated the risk of clinical malaria in settings where neither prevention nor antimalarial treatments were available. Benefits of iron supplementation in malaria-endemic areas include improved haemoglobin levels and a reduced risk of anaemia, and based on this review, iron administration for the prevention or treatment of anaemia is safe in malaria-endemic areas if malaria is adequately prevented, diagnosed, and treated.

Future meta-analyses using individual patient data may allow a better assessment of the effect of covariates such as iron and haemoglobin levels at baseline and age on the occurrence of malaria during iron supplementation. Longer-term studies would also better assess the effect of iron treatment on the outcomes of growth and development in malaria-endemic areas.