1. Objectives

To evaluate the effects of giving energy or protein supplements, or dietary advice to increase energy and protein intake, during pregnancy on maternal and infant health outcomes

2. How studies were identified

The Cochrane Pregnancy and Childbirth Group’s Trials Register was searched to January 2015 which includes trials identified from the following databases:

• CENTRAL
• MEDLINE
• EMBASE
• CINAHL

Additionally, email alerts from BioMed Central, hand searches of 30 journals and proceedings of key conferences and weekly awareness alerts of 44 other journals were also included in the register. Furthermore, authors were contacted for additional information

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials, including cluster-randomized trials

(Quasi-randomized trials and crossover trials were excluded)

3.2 Study participants

Pregnant women free from systemic illness

3.3 Interventions

Four types of interventions were assessed:

i. Nutritional education to increase dietary energy and protein intake versus no intervention or a different form of information

ii. Balanced energy and protein supplements (defined as an energy supplement whereby <25% of the energy is from protein) versus no supplement or a placebo

iii. High-protein supplements (defined as an energy supplement whereby >25% of the energy is from protein) versus low or no protein supplements

iv. Isocaloric protein supplements (defined as an energy supplement whereby the protein content provides less than 25% of the total energy content) versus an equivalent quantity of non-protein energy

3.4 Primary outcomes

• Perinatal mortality (defined by trialists)
• Stillbirth (death after 20 weeks’ gestation and before birth)
• Neonatal death (death within the first 28 days of life)

Secondary outcomes were divided into three main categories:

• Maternal outcomes: pre-eclampsia, protein intake (g/day), gestational weight gain (kg), energy intake (kcal/day), duration of labour (hours), mode of birth, number of antenatal admissions, exclusive breastfeeding at six months
• Foetal/infant outcomes: birth weight (g), small-for-gestational age, low birth weight (<2500 g), birth head circumference (cm), preterm birth (<37 weeks’ gestation), macrosomia (birth weight ≥4 kg and birth injury), neurological development, respiratory distress syndrome, admission to neonatal intensive care unit, chronic lung disease, periventricular leukomalacia, intraventricular haemorrhage, necrotizing enterocolitis, retinopathy of prematurity, growth (weight, height, head circumference, body mass index (BMI))
• Child outcomes: growth (weight, height, head circumference, BMI), neurological development

4. Main results

4.1 Included studies

Seventeen randomized controlled trials, enrolling 9030 women, were included in this review:

• Five trials assessed nutritional advice to increase energy and protein intake: four trials compared counselling or classes to deliver nutritional education versus no intervention, whilst one study compared counselling to home visits without counselling
• Twelve trials evaluated the impact of balanced energy and protein supplements: nine compared supplements to control supplements, and three studies assessed supplementation against no intervention
• One study compared high-protein supplements with supplements containing vitamins and minerals
• Two studies compared isocaloric supplements to flavoured carbonated water containing iron and vitamin C

4.2 Study settings

• Ten studies were conducted in low- and middle-income settings: Bangladesh, Burkina Faso, Colombia, Gambia, Ghana, Greece, India, Indonesia, South Africa and Taiwan, China
• Seven studies were conducted in high-income settings: the United Kingdom of Great Britain and Northern Ireland (3 trials) and the United States of America (4 trials)
• Ten studies were conducted in well-nourished populations and seven were conducted in disadvantaged areas in under-nourished populations

4.3 Study settings

How the data were analysed
Four comparisons were made: i) nutritional education versus no nutritional education or normal care; ii) balanced protein/energy supplementation versus control or no intervention, iii) high-protein supplementation versus low or no protein supplements; and iv) isocaloric balanced protein supplementation versus protein replaced an equal quantity of non-protein energy. Data from cluster-randomized trials were adjusted using standard methods based on the trial’s intra-class correlation coefficients, if reported, or based on another similar trial’s intra-class correlation coefficients. Fixed effects meta-analysis was used to produce pooled estimates of risk ratios (RR) for dichotomous data, and mean differences (MD) for continuous data, with corresponding 95% confidence intervals (CI). When substantial statistical heterogeneity was detected or expected based on clinical heterogeneity, random-effects models were used. In addition, the following subgroup analyses were planned to explore sources of heterogeneity:

• by pre-pregnancy or early pregnancy weight: low weight women, well-nourished women
• by sex of infant: boy, girl

Results
Nutritional education versus no nutritional education or normal care (5 trials)
Primary outcomes
Nutritional education was found to have no statistically significant effect on stillbirth (RR 0.37, 95% CI [0.07 to 1.90], p=0.24; 1 trial/430 women) or neonatal death (RR 1.28, 95% CI [0.35 to 4.72], p=0.71; 1 trial/448 women) in comparison to no nutritional education, and perinatal mortality was not reported in any of the included trials.

Additional outcomes
While the intervention was found to increase birth weight among infants born to undernourished women by almost half a kilogram relative to controls (MD 489.76 g, 95% CI [427.93 to 551.59], p<0.00001; 2 trials/320 women), no significant effect was found among infants born to well-nourished women (MD 15.00 g, 95% CI [-76.30 to 106.30], p=0.75; 1 trial/406 women). Nutritional education reduced the risk of having a low birth weight infant in one trial of 300 women (RR 0.04, 95% CI [0.01 to 0.14]). In one trial of 384 poor urban Bangladeshi women, gestational body weight gain from seven to nine months was significantly greater in the intervention group (8.6 ± 2.3 kg versus 5.4 ± 2.3 kg), and head circumference at birth was significantly larger with nutritional education in another trial of 389 women in Greece (MD 0.99 cm, 95% CI [0.43 to 1.55]). The risk of preterm birth appeared to be reduced with nutritional education (RR 0.46, 95% CI [0.21 to 0.98], p<0.05; 2 trials/449 women), although this finding was not upheld in an analysis of mean gestational age at birth (MD -0.10 weeks, 95% CI [-0.48 to 0.28], 1 trial/399 women). Nutritional education improved maternal protein intake (MD 6.99 g/day, 95% CI [3.02 to 10.97], 3 trials/632 women), but had no significant effect on energy intake (MD 105.61 kcal/day, 95% CI [-18.94 to 230.15], 3 trials/342 women). There was no evidence of a difference between treatment groups in birth length, small-for-gestational age, or total gestational weight gain.

Balanced protein/energy supplementation versus control or no intervention (12 trials)
Primary outcomes
The risk of stillbirth was reduced by 40% among women provided with balanced protein or energy supplements in comparison to controls (RR 0.60, 95% CI [0.39 to 0.94], p=0.024; 5 trials/3408 women). There was no evidence of a significant difference between groups in the risk of neonatal death (RR 0.68, 95% CI [0.43 to 1.07], p=0.095; 5 trials/3381 women).

Additional outcomes
Balanced energy and protein supplementation resulted in a significantly greater in birth weight in comparison to controls (MD 40.96 g, 95% CI [4.66 to 77.26], p=0.015; 11 trials/5385 infants). The risk of small-for-gestational age was also reduced with balanced energy and protein supplementation (RR 0.79, 95% CI [0.69 to 0.90], p=0.00038; 7 trials/4408 women). No significant differences between groups were detected for the outcomes birth length, head circumference, preterm birth, gestational age, weekly gestational weight gain, pre-eclampsia, Bayley mental score at one year, Stanford-Binet IQ score at five years, infant growth at one year, infant mortality, duration of labour, maternal postpartum weight, cognitive ability, offspring cardiometabolic outcomes, or child growth from 11 to 17 years; however, BMI Z-score at 11 to 17 years was significantly greater (MD 0.16, 95% CI [0.01 to 0.31], 1 trial/855 participants).

High-protein supplementation versus low or no protein supplements (1 trial)
Primary outcomes
High-protein supplementation had no statistically significant effect on stillbirth (RR 0.81, 95% CI [0.31 to 2.15], p=0.67; 1 trial/529 women) or neonatal death (RR 2.78, 95% CI [0.75 to 10.36], p=0.13; 1 trial/529 women).

Additional outcomes
A 58% increase in the risk of giving birth to a small-for-gestational age infant was observed in women receiving high-protein supplementation (RR 1.58, 95% CI [1.03 to 2.41], p=0.036; 1 trial/505 women), although there were no significant differences between groups in birth weight (MD -73.00 g, 95% CI [-171.26 to 25.26], 1 trial/504 women) or preterm birth (RR 1.14, 95% CI [0.83 to 1.56], 1 trial/505 women). There was no evidence of a treatment effect on gestational weight gain, infant growth at one year, or Bayley mental score at one year.

Isocaloric balanced protein supplementation versus protein replaced an equal quantity of non-protein energy (2 trials)
Neither of the two included trials reported data on any primary outcomes. For the outcomes birth weight and gestational weight gain, no evidence of a difference between groups was detected.

5. Additional author observations*

The included trials were of varying methodological quality, with two-thirds failing to report adequate methods of allocation concealment, and almost all trials having high or unclear risk of bias for at least one category assessed. The quality of the evidence according to GRADE assessment was also variable, with evidence related to nutritional education being rated as very low to low quality, the evidence related to balanced energy and protein supplementation as very low to moderate quality, the evidence related to high-protein supplementation as low to moderate quality, and the evidence related to isocaloric protein supplementation as very low quality. The findings of this review should therefore be interpreted with caution.

Nutritional education was found to be effective in reducing the risk of preterm birth and low birth weight, improving head circumference at birth, and in improving birth weight among infants born to undernourished women. Balanced energy/protein supplementation reduced the risk of stillbirth and small-for-gestational age, and improved birth weight. High-protein supplements and isocaloric supplements are unlikely to be of any benefit, with high-protein supplements increasing the risk of small-for-gestational age, although the data included in these analyses are few.

Further trials of high methodological quality targeting undernourished women are needed, and long-term follow-up of these trials is required. Clarification of effective interventions, such as the content and frequency of nutritional education, is needed, while additional trials of high-protein and isocaloric supplementation should not be considered.