1. Objectives

To determine the effects of zinc supplementation in healthy pregnant women and pregnant women likely to be zinc deficient on maternal, fetal, neonatal and infant outcomes

2. How studies were identified

The following databases were searched in October 2014:

• Cochrane Pregnancy and Childbirth Group’s Trials Register
• CENTRAL (The Cochrane Library 2014)
• MEDLINE
• EMBASE

Reference lists were also hand searched

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials, excluding quasi-randomized trials

3.2 Study participants

Pregnant women with no systemic illnesses, including those with normal zinc status and those who were likely to be zinc deficient

3.3 Interventions

Zinc supplementation initiated before 27 weeks’ gestation compared to no zinc supplementation or placebo

3.4 Primary outcomes

Maternal and pregnancy outcomes

• Preterm labour or birth (<37 weeks’ gestation), or both

Neonatal outcomes

• Stillbirth or neonatal death
• Birth weight
• Small-for-gestational age (birth weight less than 10^th centile for gestational age)
• Low birth weight (<2500 g)

Secondary maternal and pregnancy outcomes included antepartum haemorrhage, pregnancy-induced hypertension, prelabour rupture of membranes, post-term delivery, induction of labour, any maternal infection, meconium in liquor, Caesarean section, instrumental vaginal birth, retained placenta, postpartum haemorrhage, and smell or taste dysfunction. Secondary fetal neurodevelopmental assessment outcomes included baseline fetal heart rate, baseline variability, number of accelerations, number of fetal movements, fetal activity level (minutes), and movement amplitude. Secondary neonatal outcomes included gestational age at birth, high birth weight (>4500 g), Apgar score of less than five at five minutes, head circumference, hypoxia, neonatal sepsis, neonatal jaundice, respiratory distress syndrome, neonatal intraventricular haemorrhage, necrotising enterocolitis, neonatal length of hospital stay, and congenital malformation (a post hoc outcome). Secondary infant and child outcomes included episodes of disease, weight-for-age Z-score, weight-for-height Z-score, mid-upper arm circumference, mental development index, psychomotor development index, and any other measures of infant or child development

4. Main results

4.1 Included studies

Twenty-one randomized controlled trials, enrolling more than 17,000 women, were included in this review

• Fifteen studies compared zinc to placebo; two compared zinc to iron plus folate; co-interventions were given to all participants in some trials and included iron, folate, vitamins or combinations of these; four trials had more than two arms
• Initiation of supplementation ranged from prior to pregnancy to 26 weeks’ gestation, and daily dose of zinc ranged from 5 mg to up to 90 mg
• One identified trial did not contribute data to analyses

4.2 Study settings

• Bangladesh, Chile, China, Denmark, Egypt (did not contribute data), Ghana, Indonesia (2 trials), Iran, Nepal, Pakistan, Peru (2 trials), South Africa, the United Kingdom of Great Britain and Northern Ireland (3 trials), and the United States of America (4 trials)
• Eighteen studies were conducted in low- and middle-income settings, and one of the four studies in high- or mixed-income settings only recruited women at risk of delivering small-for-gestational age infants
• Three of the 21 studies included women who were likely to have had normal zinc concentrations; in the remaining trials women were likely to have had, or did have, low zinc concentrations

4.3 Study settings

How the data were analysed
The effect of zinc supplementation was compared with placebo or no treatment on maternal, pregnancy and infant outcomes. Treatment effects were presented as summary risk ratios (RR) with 95% confidence intervals (CI) for dichotomous data, and mean differences (MD) with corresponding 95% CI were generated for continuous data. Data from the single cluster-randomized trial had already been adjusted for clustering. Data from the zinc arms in factorial trials were combined and compared to data from arms without zinc supplementation to create a single measure of the treatment effect for each trial. Fixed-effects meta-analysis was used except where substantial clinical or statistical heterogeneity was detected, in which case random-effects meta-analysis was used. Heterogeneity was assessed in subgroup analysis for primary outcomes as follows:

• By risk of zinc deficiency: no/low risk of zinc deficiency versus high risk
• By study setting: high-income versus low-income settings

Results
Maternal and pregnancy outcomes
Preterm labour or birth
The risk of preterm birth was reduced by 14% among women receiving zinc compared to those who did not receive zinc (RR 0.86, 95% CI [0.76 to 0.97], p=0.012; 16 trials/7637 women). In subgroup analysis, the effect was unchanged in populations with low zinc concentrations (RR 0.87, 95% CI [0.77 to 0.98], p=0.02; 14 trials/7099 women), but became non-significant when restricted to populations with normal zinc levels (RR 0.64, 95% CI [0.31 to 1.32], p=0.23; 2 trials/538 women).

Neonatal outcomes
Stillbirth and neonatal death
Zinc supplementation did not reduce the risk of stillbirth or neonatal death (RR 1.12, 95% CI [0.86 to 1.46], p=0.38; 5100 participants) and did not differ in subgroup analysis by underlying risk of zinc deficiency (p=0.72 for subgroup differences).

Birth weight, small-for-gestational age, low birth weight
Overall, there was no effect of zinc supplementation on birth weight (MD 0.90 g, 95% CI [-22.23 to 24.02], p=0.94; 17 trials/6757 infants), the risk of having a small-for-gestational age infant (RR 1.02, 95% CI [0.94 to 1.11], p=0.58; 8 trials/4252 infants), or the risk of low birth weight (RR 0.93, 95% CI [0.78 to 1.12], p=0.46; 14 trials/5634 infants). There was no evidence for a difference in effect by underlying risk of zinc deficiency except for the outcome small-for-gestational age, which showed that women with normal zinc concentrations might have achieved a greater benefit from zinc supplementation, although the effect remained statistically non-significant (RR 0.24, 95% CI [0.05 to 1.10], p=0.066; 1 trial/52 women).

Additional outcomes
Maternal and pregnancy outcomes
Fewer women in the zinc supplementation group required induction in one trial of 52 women deemed to be at risk of delivering small-for-gestational age infants (RR 0.27, 95% CI [0.10 to 0.73], p=0.0098). For other maternal and pregnancy outcomes (hypertension, pre-eclampsia, prelabour rupture of membranes, antepartum haemorrhage, post-term birth, retention of placenta, meconium in liquor, instrumental vaginal birth, smell or taste dysfunction, postpartum haemorrhage, maternal infection, gestational age at birth, Caesarean section), no significant differences were found between zinc and control groups.

Neonatal and infant outcomes
Weight-for-age Z-score was reduced at 13 months of age among children born to mothers in the zinc group compared to those born to mothers in the control group in one trial of 168 infants (MD -0.40 Z-score, 95% CI [-0.70 to -0.10], p=0.0095), although this difference had not been present at six months of age. In the same trial, there were significantly fewer episodes of impetigo per infant in the zinc group over the first six months (MD -0.30, 95% CI [-0.44 to -0.16], p<0.0001; 1 trial/410 infants), and significantly fewer episodes of acute diarrhoea per infant over the same time period (MD -0.40 Z-score, 95% CI [-0.79 to -0.01], p=0.043; 1 trial/410 infants). Fetal heart rate variability (MD 0.60 beats/minute, 95% CI [0.04 to 1.16], p=0.037), and the number of fetal accelerations (MD 1.90, 95% CI [0.91 to 2.89], p<0.001) were statistically significantly higher in the treatment group in one trial of 176 participants. No other significant differences were found between treatment and control groups for the following outcomes: high birth weight, head circumference, mid-upper arm circumference, congenital malformations, Apgar score less than five at five minutes, neonatal hypoxia, jaundice, fever, infant umbilical infection, neonatal sepsis, respiratory distress syndrome, neonatal intraventricular haemorrhage, necrotising enterocolitis, neonatal hospital stay, weight-for-height Z-score at six months, and persistent diarrhoea, dysentery, cough and acute lower respiratory infection over the first six months.

Child development
At 13 months follow-up in one trial of 168 infants, those in the zinc group scored lower in measures of psychomotor development, mental development, emotional tone, and co-operation (all p≤0.044), while no differences were observed for infant approach or vocalisation. In a different trial including 355 infants at five years follow-up, no differences were found between zinc and control groups for any measure of child development, nor did the intelligence quotient differ in a further trial of 181 infants at 54 months follow-up.

5. Additional author observations*

Due to design limitations intrinsic to the included studies, the quality of the evidence for preterm birth, small-for-gestational age and low birth weight was rated as moderate, while for stillbirth or neonatal death and birth weight the quality of the evidence was considered to be of low quality. No evidence of publication bias was observed in funnel plots for preterm birth and birth weight; however, there was some asymmetry in the funnel plot for low birth weight indicating potential reporting bias.

There was moderate quality evidence of a reduction in preterm birth among women receiving zinc supplementation, primarily in low-income settings and in populations with a poor underlying zinc status. However, insufficient evidence currently exists to demonstrate a beneficial effect of zinc supplementation on other pregnancy and infant outcomes, and inconsistency between trials was also apparent for some outcomes.

Future research should be aimed at improving the overall nutritional status, including protein-calorie nutrition, of pregnant women in low-income settings, as it is unlikely that zinc deficiency exists in isolation of other nutritional deficits.