Antenatal dietary advice and supplementation to increase energy and protein intake

  • Conclusions changed
  • Review
  • Intervention

Authors


Abstract

Background

Gestational weight gain is positively associated with fetal growth, and observational studies of food supplementation in pregnancy have reported increases in gestational weight gain and fetal growth.

Objectives

To assess the effects of advice during pregnancy to increase energy and protein intake, or of actual energy and protein supplementation, on energy and protein intakes, and the effect on maternal and infant health outcomes.

Search methods

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (22 July 2011) and contacted researchers in the field. We updated the search on 12 July 2012 and added the results to the awaiting classification section of the review.

Selection criteria

Randomised controlled trials of dietary advice to increase energy and protein intake, or of actual energy and protein supplementation, during pregnancy.

Data collection and analysis

Two review authors independently assessed trials for inclusion and assessed risk of bias. Two review authors independently extracted data and checked for accuracy. Extracted data were supplemented by additional information from the trialists we contacted.

Main results

We examined 110 reports corresponding to 46 trials. Of these trials, 15 were included, 30 were excluded, and one is ongoing. Overall, 15 trials involving 7410 women were included.

Nutritional advice (four trials, 790 women)

Women given nutritional advice had a lower relative risk of having a preterm birth (two trials, 449 women) (risk ratio (RR) 0.46, 95% CI 0.21 to 0.98 ), head circumference at birth was increased in one trial (389 women) (mean difference (MD) 0.99 cm, 95% CI 0.43 to 1.55) and protein intake increased (three trials, 632 women) (protein intake: MD +6.99 g/day, 95% CI 3.02 to 10.97). No significant differences were observed on any other outcomes.

Balanced energy and protein supplementation (11 trials, 5385 women)

Risk of stillbirth was significantly reduced for women given balanced energy and protein supplementation (RR 0.62, 95% CI 0.40 to 0.98, five trials, 3408 women), mean birthweight was significantly increased (random-effects MD +40.96 g, 95% CI 4.66 to 77.26 , Tau2= 1744, I2 = 44%, 11 trials, 5385 women). There was also a significant reduction in the risk of small-for-gestational age (RR 0.79, 95% CI 0.69 to 0.90, I2 = 16%, seven trials, 4408 women). No significant effect was detected for preterm birth or neonatal death.

High-protein supplementation (one trial, 1051 women)

High-protein supplementation (one trial, 505 women), was associated with a significantly increased risk of small-for-gestational age babies (RR 1.58, 95% CI 1.03 to 2.41).

Isocaloric protein supplementation (two trials, 184 women)

Isocaloric protein supplementation (two trials,184 women) had no significant effect on birthweight and weekly gestational weight gain.

Authors' conclusions

This review provides encouraging evidence that antenatal nutritional advice with the aim of increasing energy and protein intake in the general obstetric population appears to be effective in reducing the risk of preterm birth, increasing head circumference at birth and increasing protein intake, there was no evidence of benefit or adverse effect for any other outcome reported.

Balanced energy and protein supplementation seems to improve fetal growth, and may reduce the risk of stillbirth and infants born small-for-gestational age. High-protein supplementation does not seem to be beneficial and may be harmful to the fetus. Balanced-protein supplementation alone had no significant effects on perinatal outcomes.

The results of this review should be interpreted with caution, the risk of bias was either unclear or high for at least one category examined in several of the included trials and the quality of the evidence was low for several important outcomes. Also the anthropometric characteristics of the general obstetric population is changing, therefore, those developing interventions aimed at altering energy and protein intake should ensure that only those women likely to benefit are included. Large, well designed randomised trials are needed to assess the effects of increasing energy and protein intake during pregnancy in women whose intake is below recommended levels.

Résumé scientifique

Supplémentation et conseils nutritionnels prénatals pour augmenter l'apport en énergie et en protéines

Contexte

La prise de poids pendant la grossesse est associée de manière positive à la croissance du fœtus, et les études observationnelles sur la supplémentation alimentaire pendant la grossesse ont mentionné une augmentation de la prise de poids pendant la grossesse et une augmentation de la croissance du fœtus.

Objectifs

Évaluer les effets des conseils durant la grossesse pour augmenter l'apport en énergie et en protéines ou d'une supplémentation effective en énergie et en protéines sur les apports en énergie et en protéines, ainsi que l'effet sur les critères de santé chez la mère et le nourrisson.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre des essais du groupe Cochrane sur la grossesse et la naissance (22 juillet 2011) et nous avons contacté des chercheurs spécialisés dans ce domaine. Nous avons mis à jour les recherches le 12 juillet 2012 et ajouté les résultats à la section de classification en attente de la revue.

Critères de sélection

Les essais contrôlés randomisés sur les conseils diététiques visant à accroître l'apport en énergie et en protéines ou sur une supplémentation en énergie et en protéines, pendant la grossesse.

Recueil et analyse des données

Deux auteurs de la revue ont indépendamment évalué les essais à inclure et les risques de biais. Deux auteurs ont extrait les données et vérifié leur exactitude de façon indépendante. Les données ont été complétées avec des informations supplémentaires fournies par les investigateurs que nous avons contactés.

Résultats principaux

Nous avons examiné 110 rapports correspondant à 46 essais. Parmi ces essais, 15 ont été inclus, 30 ont été exclus et un essai est en cours. Au total, 15 essais impliquant 7 410 femmes ont été inclus.

Conseils nutritionnels (quatre essais, 790 femmes)

Les femmes recevant des conseils nutritionnels avaient un risque relatif plus faible d'accouchement prématuré (deux essais, 449 femmes) (risque relatif (RR) 0,46, IC à 95 % 0,21 à 0,98), le périmètre crânien à la naissance a été augmenté dans un essai (389 femmes) (différence moyenne (DM) 0,99 cm, IC à 95 % 0,43 à 1,55) et l'apport en protéines a été augmenté (trois essais, 632 femmes) (apport en protéines : DM +6,99 g/jour, IC à 95 % 3,02 à 10,97). Il n'a été observé aucune différence significative concernant les autres critères de jugement.

Supplémentation équilibrée en énergie et en protéines (11 essais, 5 385 femmes)

Le risque de mortinaissance a été réduit de façon significative pour les femmes recevant une supplémentation équilibrée en énergie et en protéines (RR 0,62, IC à 95 % 0,40 à 0,98, cinq essais, 3 408 femmes), le poids de naissance moyen a été augmenté de façon significative (effets aléatoires DM +40,96 g, IC à 95 % 4,66 à 77,26, Tau2= 1 744, I2 = 44 %, 11 essais, 5 385 femmes). Il a également été observé une réduction significative du risque de petite taille pour l'âge gestationnel (RR 0,79, IC à 95 % 0,69 à 0,90, I2 = 16 %, sept essais, 4 408 femmes). Aucun effet significatif n'a été détecté pour les accouchements prématurés ou les décès néonatals.

Supplémentation riche en protéines (un essai, 1 051 femmes)

Une supplémentation riche en protéines (un essai, 505 femmes) a été associée à une augmentation significative du risque de bébés de petite taille pour l'âge gestationnel (RR 1,58, IC à 95 % 1,03 à 2,41).

Supplémentation protéinée isocalorique (deux essais, 184 femmes)

La supplémentation protéinée isocalorique (deux essais, 184 femmes) n'a eu aucun effet significatif sur le poids de naissance et la prise de poids hebdomadaire pendant la grossesse.

Conclusions des auteurs

Cette revue fournit des preuves encourageantes indiquant que les conseils nutritionnels prénatals visant à augmenter l'apport en énergie et en protéines dans la population obstétrique générale semblent efficaces pour réduire le risque d'accouchement prématuré, augmenter le périmètre crânien à la naissance et augmenter l'apport en protéines, il n'a été fourni aucune preuve de bénéfice ou d'effet indésirable concernant les autres critères de jugement indiqués.

La supplémentation équilibrée en énergie et en protéines semble améliorer la croissance fœtale et peut réduire le risque de mortinaissance et de nourrissons de petite taille pour l'âge gestationnel. La supplémentation riche en protéines ne semble pas bénéfique et peut être nocive pour le fœtus. La supplémentation équilibrée en protéines seule n'a eu aucun effet significatif sur les critères de jugement périnatals.

Les résultats de cette revue doivent être interprétés avec prudence, le risque de biais était soit incertain ou élevé pour au moins une catégorie examinée dans plusieurs des essais inclus et la qualité des preuves était faible pour plusieurs critères de jugement importants. De plus, les caractéristiques anthropométriques de la population obstétrique générale varient, par conséquent, les personnes développant des interventions visant à modifier l'apport en énergie et en protéines doivent s'assurer que seules les femmes susceptibles d'en bénéficier soient incluses. Des essais randomisés, bien conçus, doivent être réalisés à grande échelle pour évaluer les effets de l'augmentation de l'apport en énergie et en protéines pendant la grossesse chez les femmes dont l'apport est inférieur aux niveaux recommandés.

アブストラクト

エネルギー及び蛋白質摂取増加のための妊娠中食事のアドバイス及び補充

背景

妊娠期の体重増加は胎児の発育と正の相関を示し、妊娠中の栄養補助食品に関する観察研究では、妊娠期の体重増加と胎児発育が報告されている。

目的

妊娠中のアドバイスが、エネルギー及び蛋白質摂取を増加する効果又は実際のエネルギー、及び蛋白質補充のエネルギー及び蛋白質摂取に対する効果並びに、母子健康アウトカムへの効果を評価する。

検索戦略

Cochrane Pregnancy and Childbirth Group's Trials Register(2011年7月22日)を検索し、当該分野の研究者と連絡をとった。2012年7月12日に検索を更新し、本レビューにある分類セクションに当該結果を追加した。

選択基準

妊娠中のエネルギー及び蛋白質摂取を増加するための食事アドバイス又は実際のエネルギー及び蛋白質補充に関するランダム化比較試験。

データ収集と分析

レビューア2名が別々に選択する試験を評価し、バイアスのリスクを評価した。レビューア2名が別々にデータを抽出し、精度をチェックした。抽出したデータに、連絡をとった試験実施者からその後追加された情報を補足した。

主な結果

46件の試験に相当する110件の報告書を検討した。上記試験のうち、15件が組み入れられ、30件が除外、1件が継続中であった。全体で、女性7,410名を対象とする15件の試験を組み入れた。

著者の結論

本レビューにより、一般産科集団におけるエネルギー及び蛋白質摂取増加を目的とする出産前の栄養アドバイスが、早期産のリスクを低減し、出産時頭囲を増加、蛋白質摂取を増加する上で有効と思われる強力なエビデンスが得られたが、その他報告されたアウトカムの利益又は有害な作用のエビデンスはなかった。

Plain language summary

Energy and protein intake in pregnancy

During pregnancy, a baby developing inside the womb receives all its nutrition from its mother. Therefore, advising women on their diet and providing food supplements in pregnancy may help babies to grow and thrive. This review of randomised controlled trials examined several aspects of dietary advice and supplementation and produced the following four findings.

(1) Providing nutritional advice resulted in an increase in the mother's protein intake, fewer preterm births in two trials involving 449 women and increases in birth head circumference in one trial involving 389 women.
(2) Giving the mothers balanced energy and/protein supplements was associated with clear increases in mean birthweight (11 trials, 5385 women) with fewer stillbirths (five trials, 3408 women) and fewer small-for-gestational age births (seven trials, 4408 women), but the impact on the long-term health of the baby was uncertain, including among undernourished women.
(3) High-protein supplementation: one trial involving 1051 women showed no benefit for women and potential harm for the baby.
(4) Isocaloric protein supplementations (i.e. balanced supplements in which the protein replaces an equal quantity of other nutrients, e.g. macronutrients, fat and carbohydrate): in two trials involving 184 women this intervention showed no benefit for women or their babies.

Providing nutritional advice or balanced energy and protein supplements to women during pregnancy may be beneficial; high-protein supplements and Isocaloric protein supplements given in to women in pregnancy may be unhelpful or harmful.

Résumé simplifié

Apport en énergie et en protéines pendant la grossesse

Au cours de la grossesse, un bébé qui se développe in utero reçoit tous les nutriments de sa mère. Par conséquent, conseiller les femmes sur leur régime alimentaire et leur apporter des compléments alimentaires pendant leur grossesse peut aider leur bébé à grandir et à se développer. Cette revue d'essais contrôlés randomisés a examiné plusieurs aspects des conseils nutritionnels et de la supplémentation et a produit les quatre résultats suivants.

(1) Le fait de prodiguer des conseils nutritionnels a entraîné une augmentation de l'apport en protéines chez la mère, une baisse des accouchements prématurés dans deux essais portant sur 449 femmes et une augmentation du périmètre crânien à la naissance dans un essai portant sur 389 femmes.
(2) Le fait de donner aux mères des suppléments énergétiques et protéinés équilibrés a été associé à une augmentation nette du poids de naissance moyen (11 essais, 5 385 femmes) avec une baisse de la mortinaissance (cinq essais, 3 408 femmes) et une baisse des accouchements de bébés de petite taille pour l'âge gestationnel (sept essais, 4 408 femmes), mais l'impact sur la santé à long terme du bébé était incertain, notamment chez les femmes sous-alimentées.
(3) Supplémentation riche en protéines : un essai portant sur 1 051 femmes a démontré une absence de bénéfice pour les femmes et un risque potentiel pour le bébé.
(4) Supplémentations protéinées isocaloriques (c'est-à-dire des suppléments équilibrés dans lesquels la protéine remplace une quantité égale d'autres nutriments, par ex. des macro-nutriments, des matières grasses et des glucides) : dans deux essais portant sur 184 femmes, cette intervention n'a démontré aucun bénéfice pour les femmes ou leurs bébés.

Le fait de dispenser des conseils nutritionnels ou des suppléments énergétiques et protéinés équilibrés aux femmes pendant la grossesse peut être bénéfique ; les suppléments riches en protéines et les suppléments protéinés isocaloriques donnés aux femmes au cours de la grossesse peuvent être inutiles, voire dangereux.

Notes de traduction

Traduit par: French Cochrane Centre 30th October, 2012
Traduction financée par: Ministère du Travail, de l'Emploi et de la Santé Français

平易な要約

妊娠期におけるエネルギー及び蛋白質の摂取

妊娠中、子宮内で発育する胎児は、全栄養分を母親から得る。従って、女性に食事について助言し妊娠中の補助食品を提供することは、胎児の発育及び成長の助けとなる。ランダム化比較試験に関する本レビューでは、食事アドバイス及び栄養補充を多面的に検討し、下記の4所見を得た。 (1)栄養アドバイスをすることによって、女性449名を対象とする2件の試験で母親の蛋白質摂取が増加し、早期産が減少し、女性389名を対象とする1件の試験で出産時頭囲が増大した。(2)母親へのバランスのとれたエネルギー及び蛋白質補助食品は、平均出産時体重の明らかな増加(11件の試験、女性5,385名)、死産の減少(5件の試験、女性3,408名)及び在胎期間に比して軽小児の減少(7件の試験、女性4,408名)との関連を示したが、栄養不良の女性などの乳児の長期健康への影響は不明である。(3)高蛋白質補充は、女性1,051名を対象とする1件の試験で、母親への利点は認められず、乳児に対する有害な作用の可能性が認められた。(4)等カロリー蛋白質補充 (主要栄養素、脂質及び炭水化物などその他の栄養物等量と蛋白質を置換するバランス補助食品)という介入は、女性184名を対象とする2件の試験で母子に利益を認めなかった。 妊婦への栄養アドバイスの提供、又はバランスのとれたエネルギー及び蛋白質補助食品は有益であり、妊婦への高蛋白質の補充及び等量カロリー蛋白質の補充は無効又は有害と思われる。

訳注

監  訳: 江藤 宏美,2013.1.30

実施組織: 厚生労働省委託事業によりMindsが実施した。

ご注意 : この日本語訳は、臨床医、疫学研究者などによる翻訳のチェックを受けて公開していますが、訳語の間違いなどお気づきの点がございましたら、Minds事務局までご連絡ください。Mindsでは最新版の日本語訳を掲載するよう努めておりますが、編集作業に伴うタイム・ラグが生じている場合もあります。ご利用に際しては、最新版(英語版)の内容をご確認ください。

Summary of findings(Explanation)

Summary of findings for the main comparison. Nutritional advice compared to no counselling or advice during pregnancy for perinatal outcomes
  1. 1 Wide 95% CI.
    2 Sample size is smaller than optimal information size.
    3 Allocation concealment, blinding, incomplete outcome reporting is high risk of bias in one study.
    4 Random sequence, allocation concealment is unclear in some studies.

Nutritional advice compared to no counselling or advice during pregnancy for perinatal outcomes
Patient or population: Pregnant women
Settings:
Intervention: Nutritional advice during pregnancy
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlNutritional advice during pregnancy
StillbirthStudy populationRR 0.37
(0.07 to 1.9)
431
(1 study)
⊕⊕⊝⊝
low 1,2
 
24 per 10009 per 1000
(2 to 46)
Moderate
24 per 10009 per 1000
(2 to 46)
Neonatal deathStudy populationRR 1.28
(0.35 to 4.72)
448
(1 study)
⊕⊕⊝⊝
low 1,2
 
18 per 100023 per 1000
(6 to 83)
Moderate
18 per 100023 per 1000
(6 to 85)
Birthweight (g) The mean birthweight (g) in the intervention groups was
205.75 higher
(242.54 lower to 654.03 higher)
 426
(2 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Birth head circumference (cm) The mean birth head circumference (cm) in the intervention groups was
0.99 higher
(0.43 to 1.55 higher)
 389
(1 study)
⊕⊕⊕⊝
moderate 2
 
Small-for-gestational ageStudy populationRR 0.97
(0.45 to 2.11)
404
(1 study)
⊕⊕⊝⊝
low 1,2
 
60 per 100058 per 1000
(27 to 127)
Moderate
60 per 100058 per 1000
(27 to 127)
Preterm birthStudy populationRR 0.46
(0.21 to 0.98)
449
(2 studies)
⊕⊕⊝⊝
low 2,3
 
85 per 100039 per 1000
(18 to 84)
Moderate
92 per 100042 per 1000
(19 to 90)
Protein intake (g/day) The mean protein intake (g/day) in the intervention groups was
6.99 higher
(3.02 to 10.97 higher)
 632
(3 studies)
⊕⊕⊝⊝
low 2,4
 
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 2 Balanced protein and energy supplementation compared to control or no intervention in pregnancy for perinatal and maternal outcomes

Summary of findings 2. Balanced protein and energy supplementation compared to control or no intervention in pregnancy for perinatal and maternal outcomes
  1. 1 Allocation concealment, blinding, incomplete outcome reporting is high risk of bias in some studies.
    2 Wide 95% CI.
    3 Random sequence, allocation concealment is unclear in some studies.
    4 Sample size is smaller than optimal information size.

Balanced protein and energy supplementation compared to control or no intervention in pregnancy for perinatal and maternal outcomes
Patient or population: Pregnant women
Settings:
Intervention: Balanced protein/energy supplementation in pregnancy
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlBalanced protein/energy supplementation in pregnancy
StillbirthStudy populationRR 0.62
(0.4 to 0.98)
3408
(5 studies)
⊕⊕⊕⊝
moderate 1
 
28 per 100017 per 1000
(11 to 27)
Moderate
25 per 100015 per 1000
(10 to 25)
Neonatal deathStudy populationRR 0.68
(0.43 to 1.07)
3381
(5 studies)
⊕⊕⊝⊝
low 1,2
 
26 per 100018 per 1000
(11 to 28)
Moderate
17 per 100012 per 1000
(7 to 18)
Birthweight (g) The mean birthweight (g) in the intervention groups was
40.96 higher
(4.66 to 77.26 higher)
 5385
(11 studies)
⊕⊕⊕⊝
moderate 1
 
Small-for-gestational ageStudy populationRR 0.79
(0.69 to 0.9)
4408
(7 studies)
⊕⊕⊕⊝
moderate 1
 
173 per 1000137 per 1000
(120 to 156)
Moderate
163 per 1000129 per 1000
(112 to 147)
Preterm birthStudy populationRR 0.96
(0.8 to 1.16)
3384
(5 studies)
⊕⊕⊕⊝
moderate 3
 
112 per 1000108 per 1000
(90 to 130)
Moderate
113 per 1000108 per 1000
(90 to 131)
Pre-eclampsiaStudy populationRR 1.48
(0.82 to 2.66)
463
(2 studies)
⊕⊝⊝⊝
very low 1,2,4
 
73 per 1000108 per 1000
(60 to 195)
Moderate
38 per 100056 per 1000
(31 to 101)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 3 High protein supplementation in pregnancy and perinatal outcomes

Summary of findings 3. High protein supplementation in pregnancy and perinatal outcomes
  1. 1 Wide 95% CI.
    2 Sample size is smaller than optimal information size.

High protein supplementation in pregnancy and perinatal outcomes
Patient or population: Pregnant women
Settings:
Intervention: High protein supplementation in pregnancy
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlHigh protein supplementation in pregnancy
StillbirthStudy populationRR 0.81
(0.31 to 2.15)
529
(1 study)
⊕⊕⊝⊝
low 1,2
 
33 per 100027 per 1000
(10 to 72)
Moderate
33 per 100027 per 1000
(10 to 71)
Neonatal deathStudy populationRR 2.78
(0.75 to 10.36)
529
(1 study)
⊕⊕⊝⊝
low 1,2
 
11 per 100031 per 1000
(8 to 115)
Moderate
11 per 100031 per 1000
(8 to 114)
Small-for-gestational ageStudy populationRR 1.58
(1.03 to 2.41)
505
(1 study)
⊕⊕⊕⊝
moderate 2
 
117 per 1000185 per 1000
(121 to 282)
Moderate
117 per 1000185 per 1000
(121 to 282)
Birthweight (g) The mean birthweight (g) in the intervention groups was
73 lower
(171.26 lower to 25.26 higher)
 504
(1 study)
⊕⊕⊝⊝
low 1,2
 
Preterm birthStudy populationRR 1.14
(0.83 to 1.56)
505
(1 study)
⊕⊕⊝⊝
low 1,2
 
219 per 1000249 per 1000
(182 to 341)
Moderate
219 per 1000250 per 1000
(182 to 342)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 4 Isocaloric balanced protein supplementation in pregnancy and outcomes

Summary of findings 4. Isocaloric balanced protein supplementation in pregnancy and outcomes
  1. 1 Two trials are unclear risk of random sequence generation, allocation concealment, blinding, and selective reporting.
    2 I-square is 84%, P value = 0.01.
    3 Sample size is smaller than optimal information size.
    4 I-square is 85% with P value 0.01.

Isocaloric balanced protein supplementation in pregnancy and outcomes
Patient or population: Pregnant women
Settings:
Intervention: Isocaloric balanced protein supplementation in pregnancy
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlIsocaloric balanced protein supplementation in pregnancy
Birthweight (g) The mean birthweight (g) in the intervention groups was
108.25 higher
(220.89 lower to 437.4 higher)
 184
(2 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Weekly gestational weight gain (g/week) The mean weekly gestational weight gain (g/week) in the intervention groups was
110.45 higher
(82.87 lower to 303.76 higher)
 184
(2 studies)
⊕⊝⊝⊝
very low 1,3,4
 
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Pregnancy requires an increased intake of macronutrient and micronutrients for maternal and fetal needs, and malnourishment or inadequate dietary intake during pregnancy can lead to adverse perinatal outcomes. Observational studies (IOM 1990; Kramer 1987; Rush 2001) have indicated that both gestational weight gain and energy intake are strongly and positively associated with fetal growth, and possibly associated with a reduced risk of preterm birth. Moreover, these associations are stronger in undernourished women, i.e. those with low pre-pregnancy body mass index (BMI) (Ota 2011). Fetal development complications, such as low birthweight (LBW) and infants born small-for-gestational age, are associated with increases in perinatal mortality and morbidities (Ashworth 1998; Kramer 1987). Globally, it is estimated that more than approximately 20 million low birthweight infants are born each year, and more than 95% of these babies are born in developing countries (Unicef-WHO 2004). The effects of poor maternal nutrition on both immediate birth outcomes and longer term health has been well described in many epidemiological studies, including the effects from the Dutch winter famine of 1944 to 1945 (Stein 1975). Recognised longer-term health risks associated with poor infant growth include type 2 diabetes, hypertension, cardiovascular disease and obesity (Barker 1998; Barker 2002; Eriksson 2001).

Description of the interventions and how the interventions might work

Undernourished maternal nutritional status at conception and inadequate maternal nutritional status during pregnancy can result adverse perinatal outcomes (Viswanathan 2008). Dietary advice to pregnant women and balanced protein energy supplementation aim to achieve appropriate energy intakes lead to increase in maternal weight gain during pregnancy and fetal growth (de Onis 1998; Kulier 1998; Viller 1998). Protein generally comprises about 10% to 15% of dietary energy (Garlick 2000). Balanced protein energy supplementation (i.e. supplements in which protein provides less than 25% of the total energy content) has been shown to have significant positive impacts on maternal and perinatal birth outcomes, such as reductions in the incidences of preterm birth (Viller 1998), stillbirth (Imdad 2011) and intrauterine growth restriction (de Onis 1998). Furthermore, non-randomised trials have reported beneficial effects on fetal growth (Lechtig 1975; Prentice 1983), although the evidence from properly randomised trials suggests more modest benefits (Rush 1989; Rush 2001). On the other hand, data from severe dietary carbohydrate restriction with very high animal protein intake was counselled as part of routine antenatal care in a moderately affluent area suggest that high-protein dietary supplementation may have depressed birthweight by 400 g or more (Grieve 1979: Rush 1989). Isocaloric protein supplementation denotes a supplement, in which the protein content is 'balanced', i.e. provides less than 25% of its total energy content, but the protein replaced an equal quantity of non-protein energy in the control group. The observational findings reported for a non-randomised trial in Guatemala (Lechtig 1975) also suggest that protein supplementation is unlikely to benefit pregnant women or their infants.

Why it is important to do this review

Reliable high-quality information is required about the benefits and harms of energy/protein supplementation during pregnancy both for the woman and her infant. This review updates the review by one additional trial (Huybregts 2009), in order to aid clinical decisions and health policy-making.

Objectives

To assess the benefits and harms of dietary advice, supplementation or restriction on health outcomes for women and their infants. More specifically, the purpose of this review was to evaluate the five items listed below.

  1. Effects of advising pregnant women to increase their energy and protein intakes on gestational weight gain and outcomes of pregnancy, including fetal growth, gestational duration, and maternal and fetal/infant morbidity and mortality.

  2. Effects of balanced energy and protein supplements during pregnancy on gestational weight gain and outcomes of pregnancy.

  3. Effects of high-protein nutritional supplements during pregnancy on gestational weight gain and outcomes of pregnancy.

  4. Effects of isocaloric protein supplements (i.e. where the protein replaces an equal quantity of non-protein energy) during pregnancy on gestational weight gain and outcomes of pregnancy.

Methods

Criteria for considering studies for this review

Types of studies

We included all randomised controlled trials with randomisation at either individual or cluster level. We did not include quasi-randomised trials or cross-over trials.

For assessing dietary advice to increase energy and protein intakes: randomised controlled trials of such advice, whether administered on a one-to-one basis or to groups of women.

For assessing dietary supplementation: randomised controlled trials of energy and protein supplementation, with or without placebo.

Types of participants

All pregnant women with no systematic illness.

Types of interventions

Specific advice to increase dietary energy and protein intakes, energy and protein supplementation. The types of supplements included those that were 'balanced' energy and protein supplements (i.e. an energy supplement in which less than 25% of the energy is from protein), high-protein supplements (i.e. an energy supplement in which more than 25% of the energy is from protein), and isocaloric protein supplements (i.e. a supplement in which the protein content is 'balanced', i.e. provides less than 25% of total energy content, but the protein replaced an equal quantity of non-protein energy in the control group).

Types of outcome measures

Primary outcomes
  • Perinatal mortality (defined by trialists)

  • Stillbirth (death after 20 weeks' gestation and before birth)

  • Neonatal death (death of a live infant within the first 28 days of life)

Secondary outcomes
Maternal outcomes
  • Pre-eclampsia (defined by trialists)

  • Energy intake (kcal/day)

  • Protein intake (g/day)

  • Gestational weight gain (kg)

  • Duration of labour (hours)

  • Mode of birth

  • Number of antenatal hospital admissions

  • Exclusive breast feeding at six months (defined by trialists)

Fetal/infant outcomes
  • Birthweight (g)

  • Small-for-gestational weight (defined by trialists)

  • Low birthweight (less than 2500 g)

  • Macrosomia (birthweight > 4 kg and birth injury

  • Birth length (cm)

  • Birth head circumference (cm)

  • Neurological development

  • Preterm birth (prior to 37 weeks' gestation) 

  • Respiratory distress syndrome

  • Admission to neonatal intensive care unit

  • Chronic lung disease 

  • Periventricular leukomalacia

  • Intraventricular haemorrhage

  • Necrotising enterocolitis

  • Retinopathy of prematurity

  • Child growth (weight, height, head circumference, BMI)

Child outcomes
  • Child growth (weight, height, head circumference, BMI)

  • Neurological development

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 (22 July 2011). We updated this search on 12 July 2012 and added the results to Studies awaiting classification.

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. weekly searches of EMBASE;

  4. handsearches of 30 journals and the proceedings of major conferences;

  5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.

Details of the search strategies for CENTRAL, MEDLINE and EMBASE, the list of handsearched 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.  

Searching other resources

We contacted authors for additional data.

We did not apply any language restrictions.

Data collection and analysis

For the methods used when assessing the trials identified in the previous version of this review, see Appendix 1.

For this update, we used the following methods when assessing the trials identified by the updated search.

Selection of studies

Review authors Erika Ota (EO) and Ruoyan Tobe-Gai (RT) independently assessed all the potential studies we identified as a result of the updated search strategy for inclusion and resolved any disagreements through discussion or, if required, through consultation with Rintaro Mori (RM).

Data extraction and management

For eligible studies, EO and RT extracted the data independently and entered them into Review Manager software (RevMan 2011). We resolved any discrepancies through discussion or, if required, through consultation with RM. Data were checked for accuracy.

When information regarding any of the above was unclear, we attempted to contact the authors of the original reports to provide further details.

Assessment of risk of bias in included studies

Two review authors (EO and RT) independently assessed risk of bias for the one new study included in this update ( Huybregts 2009) using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved discrepancies through discussion.

EO, RM and RT independently re-assessed the risk of bias for additional columns newly required for all the studies already included in the previous version with respect to changes in the methods (Higgins 2011).

(1) Sequence generation (checking for possible selection bias

For each included study, we described the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the methods as indicated below:

  • low risk of bias: any truly random process (e.g. random number table; computer random number generator);

  • high risk of bias: any non-random process (e.g. odd or even date of birth; hospital or clinic record number);

  • unclear risk of bias.   

 (2) Allocation concealment (checking for possible selection bias)

For each included study, we described the method used to conceal the allocation sequence in sufficient detail and determined whether the intervention allocation could have been foreseen in advance of or during recruitment, or changed after assignment.

We assessed the methods as indicated below:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • high risk of bias (e.g. open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);

  • unclear risk of bias. 

(3) Blinding of participants, personnel and outcome assessment (checking for possible performance bias and detection bias)

For each included study, we described the methods used to blind study participants and personnel from knowledge of which intervention a participant received. Studies were judged at low risk of bias if they were blinded, or if we judged that the lack of blinding could not have affected the results. Blinding was assessed separately for different outcomes or classes of outcomes.

We assessed the methods as indicated below:

  • low, high or unclear risk of bias for participants;

  • low, high or unclear risk of bias for personnel;

  • low, high or unclear risk of bias for outcome assessment.

(4) Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations)

For each included study, and each outcome or class of outcomes, we described the completeness of the data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported or supplied by the trial authors, we re-included missing data in the analyses that we undertook.

We assessed methods as indicated below:

  • low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);

  • high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; 'as treated' analysis done with substantial departure of intervention received from that assigned at randomisation);

  • unclear risk of bias.

(5) Selective reporting bias

For each included study, we described how we investigated the possibility of selective outcome reporting bias and what we found.

We assessed the methods as indicated below:

  • low risk of bias (where it was clear that all of the study's pre-specified outcomes and all expected outcomes of interest to the review had been reported);

  • high risk of bias (where not all of the study's pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest were reported incompletely and could not be used; study failed to include results of a key outcome that would have been expected to have been reported);

  • unclear risk of bias.

(6) Other bias (checking for bias caused by problems not covered by section (1) to (5) above)

For each included study, we described any important concerns we have about other possible sources of bias.

We assessed whether each study was free of other problems that could put it at risk of bias as indicated below:

  • low risk of other bias;

  • high risk of other bias;

  • unclear risk of other bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). With reference to points (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it was likely to impact on the findings. 

Measures of treatment effect

Dichotomous data

For dichotomous data, we presented the results as the summary risk ratio with the 95% confidence intervals. 

Continuous data

For continuous data, we used the mean difference if the outcomes were measured in the same way between trials. If necessary, we planned to use the standardised mean difference to combine trials that measured the same outcome, but used different methods.  

Unit of analysis issues

Cluster-randomised trials

We included cluster-randomised trials in the analyses along with individually-randomised trials. We adjusted their sample sizes using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) using an estimate of the intra-cluster correlation co-efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we used ICCs from other sources, we reported this and conducted sensitivity analyses to investigate the effect of variation in the ICC. When we wanted to identify both cluster-randomised trials and individually-randomised trials, we planned to synthesise the relevant information. We would have considered it reasonable to combine the results from both if there was little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit was considered to be unlikely. Two of the trials (Ceesay 1997; Kafatos 1989) gave no published or unpublished data on the outcome-specific ICC. Therefore, we assumed a value of 0.01 and adjusted the corresponding sample sizes according to the design effect, i.e. by dividing the crude (individual) sample sizes by 1+ (m -1)r, where m is the average cluster size and r is the ICC (assumed to be 0.01). We conducted sensitivity analyses to investigate the effect of variation in the ICC (Figure 1; Figure 2).

Figure 1.

Sensitivity analysis of the effect of clustering : Nutritional advice during pregnancy (1.7 preterm birth) Kafatos 1989

Figure 2.

Sensitivity analysis of the effect of clustering : Balanced protein/energy supplementation in pregnancy (2.1 Stillbirth) Ceesay 1997

Cross-over trials

Cross-over trials were not considered in this review.

Dealing with missing data

For included studies, the levels of attrition were noted. The impact of including studies with high levels of missing data in the overall assessment of the treatment effect was explored using a sensitivity analysis.

All outcomes analyses were carried out, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses. The denominator for each outcome in each trial was the randomised number minus any participants whose outcomes were known to be missing.

Assessment of heterogeneity

We assessed the statistical heterogeneity in each meta-analysis using T², I² and Chi² statistics. We regarded heterogeneity as substantial if the I² was greater than 30% and either T² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity. 

Assessment of reporting biases

If there were 10 or more studies in the meta-analysis, we investigated the reporting biases (such as publication bias) using funnel plots. We assessed the funnel plot asymmetry visually, and used formal tests for funnel plot asymmetry. For continuous outcomes, we used the test proposed by Egger 1997, and for dichotomous outcomes, we used the test proposed by Harbord 2006. If asymmetry was detected in any of these tests or was suggested by a visual assessment, we performed exploratory analyses to investigate this.

Data synthesis

We carried out statistical analysis using Review Manager software (RevMan 2011). We used a fixed-effect inverse variance meta-analysis for combining data where trials were examining the same intervention, and the trials' populations and methods were judged to be sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we used a random-effects meta-analysis to produce an overall summary when an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average range of possible treatment effects and we discussed the clinical implications of treatment effects differing between trials. Where the average treatment effect was not clinically meaningful, we did not combine trials.

When we performed random-effects analyses, the results were presented as the average treatment effects with 95% confidence intervals, and the estimates of T² and I².

Subgroup analysis and investigation of heterogeneity

When we identified substantial heterogeneity, we investigated it using subgroup analyses and sensitivity analyses. We considered whether an overall summary was meaningful, and if it was, we used a random-effects analysis to produce such a summary.

Since observational studies (IOM 1990; Kramer 1987) suggest a stronger association between gestational weight gain and fetal growth in women who were under-nourished before pregnancy, we stratified the analysis of the effects on mean birthweight into those trials in which the majority of women had low pre-pregnancy (or early pregnancy) weight (Ceesay 1997; Girija 1984; Kardjati 1988; Mora 1978; Rush 1980), and those in which the participants appeared adequately nourished (Elwood 1981; Ross 1985; Viegas 1982a). For the Taiwan trial (Blackwell 1973) and (Huybregts 2009; Viegas 1982b), within-trial stratification was possible, based on data contained in the published reports.

For fixed-effect inverse variance meta-analyses, we assessed differences between subgroups by interaction tests. For random-effects and fixed-effect meta-analyses using methods other than inverse variance, we assessed differences between subgroups by inspection of the subgroups' confidence intervals, in which non-overlapping confidence intervals indicated a statistically significant difference in the treatment effects between the subgroups.

Because growth varies with differences in sex (Onis 2007), it is desirable to compare growth between groups after adjusting for variations by sex. We conducted subgroup analysis on the children, separated by sexes for follow-up results of balanced protein and energy supplementation at the age of 11 to 17 years (height, weight, systolic blood pressure, diastolic blood pressure, BMI z-score, and body fat).

Sensitivity analysis

We carried out sensitivity analyses to explore the effects of fixed- or random-effects analyses for outcomes with statistical heterogeneity.

Results

Description of studies

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

Results of the search

Initially we examined 110 reports corresponding to 46 trials. Of these trials, 15 were included and 30 were excluded. One trial is ongoing (Moore 2011) with the results expected in 2013 (see Characteristics of ongoing studies). (Three reports from an updated search in July 2012 have been added to Studies awaiting classification.)

Included studies

We included 15 trials published between 1973 to 2009 that met the inclusion criteria. Most of the trials focused on assessing the effects of dietary advice, supplementation, or restriction on gestational weight gain, pre-eclampsia and/or pregnancy outcomes, or the development of the children. Four trials (Briley 2002; Hunt 1976; Kafatos 1989; Sweeney 1985) evaluated nutritional advice to increase energy and protein intake. Eleven trials (Blackwell 1973; Ceesay 1997; Elwood 1981; Girija 1984; Huybregts 2009; Kardjati 1988; Mora 1978; Ross 1985; Rush 1980; Viegas 1982a; Viegas 1982b) assessed the impact of balanced energy/protein supplementation. Only one trial assessed the effects of high-protein nutritional supplements (Rush 1980). Two trials (Viegas 1982a; Viegas 1982b) investigated the effects of isocaloric protein supplements. Seven trials were from high-income countries such as the USA and UK, and eight trials from low- and middle-income countries such as Ganbia, Taiwan, India, Burkina Faso, Greece, Indonesia, Colombia, South Africa. Four trials were conducted in an economically disadvantaged area including under-nourished populations; the other 11 trials included well-nourished populations. Interventions for nutritional advice included counselling or classes versus no interventions (three trials; Hunt 1976; Kafatos 1989; Sweeney 1985) and counselling versus home visits without counselling (one trial; Briley 2002). Interventions for balanced energy and protein supplementation included supplementation versus control supplements (eight trials; Blackwell 1973; Huybregts 2009; Kardjati 1988; Mora 1978; Ross 1985; Rush 1980; Viegas 1982a; Viegas 1982b) and supplementation versus no intervention (three trials; Ceesay 1997; Elwood 1981; Girija 1984). Intervention for high-protein nutritional supplements included supplementation versus supplement containing vitamins/minerals (Rush 1980). Intervention for isocaloric-protein nutritional supplements included supplementation versus supplement of flavoured carbonated water containing iron and vitamin C (Viegas 1982a; Viegas 1982b).

For details of the included studies, see the Characteristics of included studies table.

Excluded studies

We excluded 30 trials. Of these trials, eight trials did not match the interventions in this review, nine involved participants who were outside the scope of the review, one did not have a control group, one trial's analysis was based on individual women despite randomising by village, and 11 trials did not involve randomisation, or the designs were outside the scope of the review.

For details of the excluded studies, see the Characteristics of excluded studies table.

Risk of bias in included studies

See Figure 3 and Figure 4.

Figure 3.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 4.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation (selection bias)

Sequence generation

Six trials had a low risk of bias because of adequate randomisation of participants to the intervention groups (Elwood 1981; Huybregts 2009; Kafatos 1989; Kardjati 1988; Rush 1980; Sweeney 1985). For nine trials, risk of bias could not be adequately judged because no detailed information was provided about allocation sequence generation (Blackwell 1973; Briley 2002; Ceesay 1997; Girija 1984; Hunt 1976; Mora 1978; Ross 1985; Viegas 1982a; Viegas 1982b).

Allocation concealment

Five trials had a low risk of bias through the use of sequentially numbered, opaque, sealed envelopes or drug containers of identical appearance (Elwood 1981; Huybregts 2009; Kardjati 1988; Rush 1980; Sweeney 1985). Ten trials provided no information about allocation concealment (Blackwell 1973; Briley 2002; Ceesay 1997; Girija 1984; Hunt 1976; Kafatos 1989; Mora 1978; Ross 1985; Viegas 1982a; Viegas 1982b).

Blinding (performance bias and detection bias)

Six trials had a low risk of bias using single or double blinding, or used no blinding but reported outcomes and outcome measurements that were not likely to be influenced by the lack of blinding. Seven trials had a high risk of bias owing to a lack of blinding. Two trials could not be judged for the risk because no information was provided (Viegas 1982a; Viegas 1982b).

Incomplete outcome data (attrition bias)

Losses to follow-up ranged from 1.5% in Viegas 1982b to 25.9% in Briley 2002. Eleven trials had a low risk of bias, two trials had a high risk of bias and two trials were judged as unclear bias owing to insufficient information.

Selective reporting (reporting bias)

Fourteen trials were judged as unclear risk because the protocol was not available for judgment of this bias. Only one trial (Huybregts 2009) had a protocol and was judged as low risk for selective reporting.

Other potential sources of bias

Eight trials had a high risk of bias because no data were presented on compliance or substitution, and for other reasons. Three trials had a low risk of bias and four trials had insufficient information and were judged as unclear risk.

The funnel plots (Figure 5; Figure 6) did not show any publication bias.

Figure 5.

Funnel plot of comparison: 2 Balanced protein/energy supplementation in pregnancy, outcome: 2.3 Birthweight (g).

Figure 6.

Funnel plot of comparison: 2 Balanced protein/energy supplementation in pregnancy, outcome: 2.9 Weekly gestational weight gain (g/week).

Effects of interventions

See: Summary of findings for the main comparison Nutritional advice compared to no counselling or advice during pregnancy for perinatal outcomes; Summary of findings 2 Balanced protein and energy supplementation compared to control or no intervention in pregnancy for perinatal and maternal outcomes; Summary of findings 3 High protein supplementation in pregnancy and perinatal outcomes; Summary of findings 4 Isocaloric balanced protein supplementation in pregnancy and outcomes

Nutritional advice to increase energy and protein intakes

Four trials of nutritional advice, involving 790 women, were included. For the primary outcomes, there was no significant effect on stillbirth (risk ratio (RR) 0.37, 95% confidence interval (CI) 0.07 to 1.90; one trial, 431 women- Analysis 1.1) or neonatal death (RR 1.28, 95% CI 0.35 to 4.72; one trial, 448 women- Analysis 1.2 ). There was no significant effects on the outcomes of birthweight (Analysis 1.3), birth length (Analysis 1.4), and small-for-gestational age (Analysis 1.6). Because the results of total gestational weight gain (Briley 2002; Kafatos 1989) were inconsistent and showed high heterogeneity we have not combined the studies in the analysis (no total shown in Analysis 1.11). Birth head circumference (cm) was significantly increased in the intervention group (mean difference (MD) 0.99 cm, 95% CI 0.43 to 1.55; one trial, 389 women - Analysis 1.5). The 'significant' reduction in preterm birth associated with advice (RR 0.46, 95% CI 0.21 to 0.98, P < 0.05; two trials, 449 women - Analysis 1.7) was not consistent with the total absence of effect on mean gestational age (MD -0.10 weeks, 95% CI -0.48 to 0.28; one trial, 399 women - Analysis 1.8). Sensitivity analyses (Figure 1) was conducted in using values of 0.01 for the ICC did not qualitatively change the relative risks for preterm in Kafatos 1989. Within the methodological limitations discussed above, advice to increase protein intake seems to be successful in achieving its goal (protein intake: MD +6.99 g/day, 95% CI 3.02 to 10.97, P < 0.05; three trials, 632 women- Analysis 1.9), but there was no significant increase in energy intake (energy intake: MD +105.61 kcal/day, 95% CI -18.94 to 230.15, P = 0.10; three trials, 342 women - Analysis 1.10).

Balanced energy/protein supplementation

Eleven trials, involving 5385 women, were included. Providing balanced energy and protein supplementation significantly reduced the risk of stillbirth (RR 0.62, 95% CI 0.40 to 0.98; five trials, 3408 women - Analysis 2.1). Neonatal death was unaffected (RR 0.68, 95% 0.43 to 1.07; five trials, 3381 women - Analysis 2.2). Sensitivity analyses (Figure 2) for ICCs of 0.02 to 0 made little difference, using values of 0.01 for the ICC did not qualitatively change the relative risks for stillbirth in Ceesay 1997.

Supplementation was also associated with significant increases in mean birthweight (random-effects MD +40.96 g, 95% CI 4.66 to 77.26 , Tau2 = 1744, I2 = 44%, P = 0.03; 11 trials, 5385 infants - Analysis 2.3 ). Although clinically small, birth length (cm) was statistically significantly increased (fixed-effect MD +0.16 cm, 95% CI 0.01 to 0.31; five trials, 3370 women - Analysis 2.4), while no significant difference was found for birth head circumference (Analysis 2.5). The incidence of small-for-gestational age birth was significantly reduced (RR 0.79, 95% CI 0.69 to 0.90, I2 = 16%; seven trials, 4408 women - Analysis 2.6). There were no significant effects observed on preterm birth (Analysis 2.7), gestational age (week) (Analysis 2.8), or weekly gestational weight gain (g/week) (Analysis 2.9). The rather meagre data on pre-eclampsia did not suggest a reduction in risk with supplementation (RR 1.48, 95% CI 0.82 to 2.66; two trials, 463 women - Analysis 2.10).

Although postnatal follow-up was limited to a small number of trials, the enhancement of fetal growth observed in those trials was not reflected in larger size or improved neurocognitive development at one year. Bayley mental score at one year had no significant effect in one trial (Rush 1980; Analysis 2.11). The Taiwan trial (Blackwell 1973) detected no effect on Stanford-Binet IQ score at five years (Analysis 2.12), and weight at one year (Analysis 2.13). The data of the standard deviation of length at one year for Blackwell 1973 were not credible compared with the Rush 1980 study, we have omitted this trial from the analysis and only showed the data from Rush 1980 (Analysis 2.14). There was no significant effect on head circumference at one year from either the Taiwan (Blackwell 1973) or Harlem trials (Rush 1980) (Analysis 2.15).

Maternal outcomes other than weight gain were reported infrequently. Only one trial each reported results on other outcomes. The Bogota trial (Mora 1978) detected no significant reduction in duration of labour with supplementation (Analysis 2.16). The East Java trial (Kardjati 1988) found neither an increase in maternal weight at four weeks postpartum (Analysis 2.17).

Follow-up at 6.5 to 9.5 years of age for approximately 25% of the children randomised in Ceesay 1997 found no difference in immune function (delayed-hypersensitivity skin tests, antibody responses to pneumococcal and rabies vaccines, and salivary IgA concentration) between the intervention and control groups (data not shown in data and analysis table). Follow-up at 11 to 17 years of age for approximately two-thirds of the children who were still alive found no significant differences in height (Analysis 2.18), weight (Analysis 2.19), or systolic or diastolic blood pressure (Analysis 2.20; Analysis 2.21), but did find a small increase in the mean BMI z-score (MD +0.16, 95% CI +0.01 to +0.31; one trial, 855 children - Analysis 2.22) in the control group. However, the difference in BMI was in contrast with the absence of the effect on per cent body fat (Analysis 2.23).

High-protein supplementation

Only one trial (Rush 1980), involving 1051 women, was included. For primary outcomes, the Harlem trial (Rush 1980) reported non-significant effects in stillbirth (RR 0.81, 95% CI 0.31 to 2.15; one trial, 529 women - Analysis 3.1) and neonatal death (RR 2.78, 95% CI 0.75 to 10.36; one trial, 529 women - Analysis 3.2) with high-protein supplementation. The only available trial (Rush 1980) provided the evidence of significant increases in infants born small-for-gestational age (RR 1.58, 95% CI 1.03 to 2.41, P = 0.04; one trial, 505 women - Analysis 3.3), although no significant effect for birthweight (Analysis 3.4) or preterm birth (Analysis 3.5).

High-protein supplementation had no effect on weekly gestational weight gain (MD +4.50 g/week, 95% CI -33.55 to +42.55; one trial, 486 women - Analysis 3.6). At one-year follow-up in the Harlem trial (Rush 1980), high-protein supplementation was not associated with detectable differences in weight (Analysis 3.7), length (Analysis 3.8), head circumference (Analysis 3.9) or Bayley mental score (Analysis 3.10).

Isocaloric protein supplementation

Two trials, involving 184 women, were included. Owing to the significant heterogeneity in the results for birthweight and gestational weight gain, the data were pooled using a random-effects model. There was no significant effect on birthweight or gestational weight gain of isocaloric protein supplementation. For mean birthweight the MD was +108.25 g (95% CI -220.89 to 437.40 , Tau2 = 47211, I2 = 84%; two trials, 184 infants - Analysis 4.1), while for gestational weight gain, the MD was +110.45 g/week (95% CI 82.87 to 303.76 , Tau2 = 16542, I2 = 85%; two trials, 184 women - Analysis 4.2).

Subgroup analysis in balanced energy/protein supplementation

Since observational studies (IOM 1990; Kramer 1987) suggested a stronger association between gestational weight gain and fetal growth in women who were under-nourished before pregnancy, we stratified the analysis of the effects on mean birthweight into those trials in which the majority of women had low pre-pregnancy (or early pregnancy) weight (Ceesay 1997; Girija 1984; Kardjati 1988; Mora 1978; Rush 1980) and those in which the participants appeared adequately nourished (Elwood 1981; Ross 1985; Viegas 1982a). For the Taiwan trial (Blackwell 1973), (Huybregts 2009) and (Viegas 1982b), within-trial stratification was possible, based on the data contained in the published reports. Only the mean birthweight in balanced energy/protein supplementation were analysed for the subgroups of undernourished and nourished women. However, there was no evidence of a subgroup differences between the malnourished and adequately nourished groups (test for subgroup differences: Chi2 = 2.35. df = 1(P = 0.12), I2 = 57.5%).

Discussion

Summary of main results

Nutritional advice was successful in reducing the risk of preterm birth, increasing head circumference at birth and increasing protein intake, however, there was no evidence of benefit or adverse effect for any other outcome reported.

Balanced energy/protein supplementation was associated with significantly reduced risk of stillbirth, increased mean birthweight, and a significant reduction in the risk of small-for-gestational-age birth. No significant effect was detected for preterm birth, or neonatal death.

High-protein supplementation was associated with a significantly increased risk of infants born small-for-gestational age, but this is based on only one trial including 1051 women. Isocaloric protein supplementation had no significant effect on birthweight or weekly gestational weight gain, based on two trials including only 184 women.

Overall completeness and applicability of evidence

Nutritional advice appears effective in increasing pregnant women's protein intake, reducing preterm birth and significant increases in birth head circumference. No data have been reported on other important maternal pregnancy outcomes, such as duration of labour, caesarean section, or postpartum weight retention.

The modest increase in birthweight associated with balanced energy/protein supplementation may well be explained by the rather small net increases in energy intake achieved in most of the trials. Noncompliance and dietary substitution are likely explanations for these small net increases, and the much higher energy supplement provided in the Gambia trial (Ceesay 1997) appeared to have a much larger effect on mean birthweight. Of the seven sizeable trials with the highest methodological quality (Blackwell 1973; Ceesay 1997; Elwood 1981; Huybregts 2009; Kardjati 1988; Mora 1978; Rush 1980), only the East Java trial (Kardjati 1988) failed to show any benefit for mean birthweight (Analysis 2.3), despite convincing evidence that the trial participants were under-nourished prior to the intervention. Owing to the large sample size, chance is an unlikely explanation for the absence of benefit in the East Java trial( Kardjati 1988), and an undetected substitution of the normal home diet by the supplement seems more likely. Due to the significant effect on mean birthweight (Analysis 2.3), the reduction in the risk of infant born in small-for-gestational age (Analysis 2.6) was substantial. Nonetheless, that reduction did not appear to be associated with long-term benefits for child growth or development, but long-term follow-up was only reported in two trials (Ceesay 1997; Rush 1980). Of greatest importance is the evidence indicating reduced risk of stillbirth (Analysis 2.1). However, this evidence is based on five trials and the evidence is classified as low quality, and the biological mechanism for such risk reduction remains unclear, given the modest effects observed on the indices of fetal growth.

Most of the supplements/dietary manipulations also involved changes to the micronutrient (vitamins and minerals) content of the diet in the both intervention and control. As micronutrient supplementation may also alter some pregnancy outcomes independent of protein and energy, it is difficult to separate the contribution to the effects, particularly in the "balanced protein and energy" studies.

The available evidence from one trial provides no justification for prescribing high-protein nutritional supplements to pregnant women. Not only do such supplements appear to lack beneficial effects, but the evidence suggests that they may even be harmful. Furthermore, the data derived from these trials suggest that isocaloric protein supplementation alone (i.e. without energy supplementation) is unlikely to be of benefit to pregnant women or their infants. The included two trials had high heterogeneity, probably because amounts of energy supplementation were different (273 kcal in Viegas 1982a; 425 kcal in Viegas 1982b). The finding of the excluded trial of Mardones 1988, which reported increases in the risk of infant born small-for-gestational age, remains uncertain, given the methodological limitations of the trial. Moreover, the normal-protein "control" supplement in Mardones 1988 contained much higher quantities of iron and other micronutrients than the high protein supplement.

The results of this review should be interpreted with caution considering that the majority of trials were published in 1970/1980s. The incidence of inadequate nutrition and overweight and obesity is likely to be different today and most trials included a mixed population of those considered to have poor nutritional status and potentially those with adequate nutrition or over-nutrition. Indeed seven trials were from high-income countries where recent reports suggest two-thirds of the general population and half of pregnant women are overweight or obese (Haslehurst 2006; Wang 2011).

Quality of the evidence

We included 15 trials involving 7410 women. The quality of the evidence in this review is assessed using the GRADE approach (Guyatt 2008) and the results are presented in Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3; and Summary of findings 4. The GRADE uses four levels of quality (very low, low, moderate and high) over several domains covering limitations in the design and implementation of the studies, indirectness of evidence, unexplained heterogeneity or inconsistency in the results, imprecision of the results and high probability of publication bias. In the nutritional advice during pregnancy studies (Summary of findings for the main comparison), the evidence was judged to be of moderate quality (birth head circumference), and low quality (Stillbirth, neonatal death, preterm birth, infants born small-for-gestational age and protein intake) which includes two of the primary outcomes, to very low quality (birthweight) suggesting that the estimates were very uncertain. In the balanced protein and energy supplementation in pregnancy studies (Summary of findings 2), with significant reductions in stillbirth and infants born small-for-gestational age, and significant increase in birthweight were considered to be of moderate quality. Preterm birth was moderate quality. Neonatal death were of low quality and pre-eclampsia was of very low quality. In the high-protein supplementation in pregnancy studies (Summary of findings 3), the significant increase in infants born small-for-gestational age was of moderate quality in only one study (Rush 1980). In the isocaloric balanced protein supplementation in pregnancy studies (Summary of findings 4), the evidence was judged to be of very low quality (birthweight, weekly gestational weight gain) meaning that the estimates were very uncertain.

Potential biases in the review process

There were several potential biases in the review process. We made efforts to limit the bias in several ways: two review authors assessed the eligibility for inclusion and assessed the risks of bias independently. Although the authors' views varied, we decided to accept the final conclusions after extensive discussion and reaching a consensus. Carrying out reviews, however, may require a number of subjective judgements, and it is possible that a different review team may have reached different decisions regarding the assessments of eligibility and risks of bias. Feedback from readers will serve to improve the next review update.

Agreements and disagreements with other studies or reviews

We have included only randomised controlled trials (RCTs) and excluded the quasi-RCTs previously included in the review (Kramer 2003). The new findings of this review are that balanced energy and protein supplementation was associated with significant increases in mean birthweight, while the other major findings are consistent with those of the previous Cochrane Review (Kramer 2003). Prenatal supplementation with multi-micronutrients was associated with a significantly reduced risk of low-birthweight infants and with improved birthweight when compared with iron-folic acid supplementation, although there was no effect on the risk of preterm birth or small-for gestational-age infants (Shah 2009). Researchers should aim to include only those women in trials to increase energy and protein intake who have the potential to benefit. Observational data suggest women who are overweight or obese or who exceed their daily energy and protein requirements during pregnancy are at increased risk of adverse pregnancy outcomes including: stillbirth and large-for-gestational age and macrosomia (birthweight > 4 kg) (Chen 2009; Heslehurst 2008; Thangaratinam 2012), therefore, the effect of increasing protein and energy intakes could have opposite effects on different populations within the same trial if those included are not adequately defined and selected.

Authors' conclusions

Implications for practice

This review provides encouraging evidence that nutritional advice to increase protein and energy intake and balanced energy and protein supplementation may reduce some perinatal adverse outcomes. The long-term effects are unclear and it seems likely that targeting undernourished women rather than the whole obstetric population would convey the most benefit. For most of the included trials in this review, the risk of bias was either unclear or high for at least one category examined, and the results of this review should therefore be interpreted with caution.

Nutritional advice appears to be effective in increasing pregnant women's protein intake, and increases fetal growth such as birth head circumference. The 54% relative reduction in preterm birth for nutritional advice in energy and protein compared with no nutritional counselling may be beneficial to pregnant women.

Balanced energy and protein supplementation appears to reduce the risks of stillbirth, although the biological mechanisms underlying these reductions remain unclear. Furthermore, balanced protein and energy intervention, as provided in most trials, results in significant increases in maternal weight gain and infant birthweight, and decreases the risk of infants born in small-for-gestational age. These effects do not seem to confer long-term benefits to the child in terms of growth, neurocognitive development, and adiposity or blood pressure. The available evidence is inadequate to evaluate the potential effects on preterm birth, neonatal death or maternal health.

Based on the available evidence, there is no justification for prescribing high-protein and isocaloric nutritional supplements to pregnant women, although the number of trials and women included are few.

Implications for research

High-quality randomised trials are needed that target those women who are nutritionally deprived or underweight with reduced energy intake; long-term follow-up is required.

Given the modest benefits in preterm delivery documented for balanced energy and protein advice during pregnancy, future randomised trials need to assess the effects on perinatal outcomes such as stillbirth, neonatal death and birthweight. Effective interventions, such as the content and frequency of nutritional advice, need to be clarified.

Future energy and protein supplementation trials should focus their attention on outcomes other than fetal growth, especially in undernourished women and particularly on confirming the evidence of intervention on reduced risks of stillbirth and infants born small-for-gestational age. Such trials will require large sample sizes. Any future trials should also assess the effects on women, including duration of labour, caesarean section, macrosomia and postpartum weight retention.

The lack of evidence of benefit, coupled with the possibility of harm, suggests that future trials of high-protein supplementation, and isocaloric protein supplementation should not be considered.

Acknowledgements

We are grateful to Michael S. Kramer and Ritsuko Kakuma who developed the original review (Kramer 2003) and subsequent updates upon which this updated review is based.

The review authors would like to acknowledge the Pregnancy and Childbirth team for assistance with the preparation of the original review and its update, including the Trials Search Co-ordinator for assistance in developing the search strategy, the editors, co-editors and other staff within the team. We also acknowledge Tommy Tang and Stuart Gilmour, who supported the 2012 update.

As part of the pre-publication editorial process, this review 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. Nutritional advice during pregnancy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Stillbirth1431Risk Ratio (M-H, Fixed, 95% CI)0.37 [0.07, 1.90]
2 Neonatal death1448Risk Ratio (M-H, Fixed, 95% CI)1.28 [0.35, 4.72]
3 Birthweight (g)2426Mean Difference (IV, Random, 95% CI)205.75 [-242.54, 654.03]
4 Birth length (cm)1399Mean Difference (IV, Fixed, 95% CI)0.17 [-0.72, 1.06]
5 Birth head circumference (cm)1389Mean Difference (IV, Fixed, 95% CI)0.99 [0.43, 1.55]
6 Small-for-gestational age1404Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.45, 2.11]
7 Preterm birth2449Risk Ratio (M-H, Fixed, 95% CI)0.46 [0.21, 0.98]
8 Gestational age (week)1399Mean Difference (IV, Fixed, 95% CI)-0.10 [-0.48, 0.28]
9 Protein intake (g/day)3632Mean Difference (IV, Fixed, 95% CI)6.99 [3.02, 10.97]
10 Energy intake (kcal/day)3342Mean Difference (IV, Fixed, 95% CI)105.61 [-18.94, 230.15]
11 Total gestational weight gain (kg)2 Mean Difference (IV, Random, 95% CI)Totals not selected
Analysis 1.1.

Comparison 1 Nutritional advice during pregnancy, Outcome 1 Stillbirth.

Analysis 1.2.

Comparison 1 Nutritional advice during pregnancy, Outcome 2 Neonatal death.

Analysis 1.3.

Comparison 1 Nutritional advice during pregnancy, Outcome 3 Birthweight (g).

Analysis 1.4.

Comparison 1 Nutritional advice during pregnancy, Outcome 4 Birth length (cm).

Analysis 1.5.

Comparison 1 Nutritional advice during pregnancy, Outcome 5 Birth head circumference (cm).

Analysis 1.6.

Comparison 1 Nutritional advice during pregnancy, Outcome 6 Small-for-gestational age.

Analysis 1.7.

Comparison 1 Nutritional advice during pregnancy, Outcome 7 Preterm birth.

Analysis 1.8.

Comparison 1 Nutritional advice during pregnancy, Outcome 8 Gestational age (week).

Analysis 1.9.

Comparison 1 Nutritional advice during pregnancy, Outcome 9 Protein intake (g/day).

Analysis 1.10.

Comparison 1 Nutritional advice during pregnancy, Outcome 10 Energy intake (kcal/day).

Analysis 1.11.

Comparison 1 Nutritional advice during pregnancy, Outcome 11 Total gestational weight gain (kg).

Comparison 2. Balanced protein/energy supplementation in pregnancy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Stillbirth53408Risk Ratio (M-H, Fixed, 95% CI)0.62 [0.40, 0.98]
2 Neonatal death53381Risk Ratio (M-H, Fixed, 95% CI)0.68 [0.43, 1.07]
3 Birthweight (g)115385Mean Difference (IV, Random, 95% CI)40.96 [4.66, 77.26]
3.1 Undernourished women82903Mean Difference (IV, Random, 95% CI)66.96 [13.13, 120.78]
3.2 Adequately nourished women62482Mean Difference (IV, Random, 95% CI)15.93 [-20.83, 52.69]
4 Birth length (cm)53370Mean Difference (IV, Fixed, 95% CI)0.16 [0.01, 0.31]
5 Birth head circumference (cm)53352Mean Difference (IV, Random, 95% CI)0.04 [-0.08, 0.17]
6 Small-for-gestational age74408Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.69, 0.90]
7 Preterm birth53384Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.80, 1.16]
8 Gestational age (week)63471Mean Difference (IV, Fixed, 95% CI)-0.10 [-0.22, 0.01]
9 Weekly gestational weight gain (g/week)92391Mean Difference (IV, Random, 95% CI)18.63 [-1.81, 39.07]
10 Pre-eclampsia2463Risk Ratio (M-H, Fixed, 95% CI)1.48 [0.82, 2.66]
11 Bayley mental score at 1 year1411Mean Difference (IV, Fixed, 95% CI)-0.74 [-1.95, 0.47]
12 IQ at 5 years1153Mean Difference (IV, Fixed, 95% CI)0.0 [-4.98, 4.98]
13 Weight at 1 year (g)2623Mean Difference (IV, Fixed, 95% CI)30.43 [-139.67, 200.53]
14 Length at 1 year (cm)1428Mean Difference (IV, Fixed, 95% CI)0.0 [-5.69, 5.69]
15 Head circumference at 1 year (cm)2627Mean Difference (IV, Fixed, 95% CI)-0.13 [-0.35, 0.10]
16 Duration of labour (hours)1345Mean Difference (IV, Fixed, 95% CI)-0.09 [-1.18, 1.00]
17 Maternal weight 4 weeks' postpartum (kg)1354Mean Difference (IV, Fixed, 95% CI)-0.90 [-1.92, 0.12]
18 Height at age 11-17 years (cm)1855Mean Difference (IV, Fixed, 95% CI)-0.39 [-1.73, 0.94]
18.1 Boys1445Mean Difference (IV, Fixed, 95% CI)0.60 [-1.40, 2.60]
18.2 Girls1410Mean Difference (IV, Fixed, 95% CI)-1.20 [-3.00, 0.60]
19 Weight at 11-17 years (kg)1855Mean Difference (IV, Fixed, 95% CI)0.46 [-0.77, 1.69]
19.1 Boys1445Mean Difference (IV, Fixed, 95% CI)0.70 [-0.89, 2.29]
19.2 Girls1410Mean Difference (IV, Fixed, 95% CI)0.10 [-1.86, 2.06]
20 Systolic blood pressure at age 11-17 years (mmHg)1855Mean Difference (IV, Fixed, 95% CI)0.60 [-0.61, 1.81]
20.1 Boys1445Mean Difference (IV, Fixed, 95% CI)1.10 [-0.61, 2.81]
20.2 Girls1410Mean Difference (IV, Fixed, 95% CI)0.10 [-1.60, 1.80]
21 Diastolic blood pressure at age 11-17 years (mmHg)1855Mean Difference (IV, Fixed, 95% CI)-0.08 [-1.10, 0.93]
21.1 Boys1445Mean Difference (IV, Fixed, 95% CI)0.5 [-0.98, 1.98]
21.2 Girls1410Mean Difference (IV, Fixed, 95% CI)-0.60 [-1.99, 0.79]
22 BMI z-score at age 11-17 years1855Mean Difference (IV, Fixed, 95% CI)0.16 [0.01, 0.31]
22.1 Boys1445Mean Difference (IV, Fixed, 95% CI)0.20 [0.00, 0.40]
22.2 Girls1410Mean Difference (IV, Fixed, 95% CI)0.10 [-0.13, 0.33]
23 % body fat at 11-17 years1847Mean Difference (IV, Fixed, 95% CI)0.06 [-0.41, 0.52]
23.1 Boys1440Mean Difference (IV, Fixed, 95% CI)0.0 [-0.54, 0.54]
23.2 Girls1407Mean Difference (IV, Fixed, 95% CI)0.20 [-0.68, 1.08]
Analysis 2.1.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 1 Stillbirth.

Analysis 2.2.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 2 Neonatal death.

Analysis 2.3.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 3 Birthweight (g).

Analysis 2.4.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 4 Birth length (cm).

Analysis 2.5.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 5 Birth head circumference (cm).

Analysis 2.6.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 6 Small-for-gestational age.

Analysis 2.7.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 7 Preterm birth.

Analysis 2.8.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 8 Gestational age (week).

Analysis 2.9.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 9 Weekly gestational weight gain (g/week).

Analysis 2.10.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 10 Pre-eclampsia.

Analysis 2.11.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 11 Bayley mental score at 1 year.

Analysis 2.12.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 12 IQ at 5 years.

Analysis 2.13.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 13 Weight at 1 year (g).

Analysis 2.14.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 14 Length at 1 year (cm).

Analysis 2.15.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 15 Head circumference at 1 year (cm).

Analysis 2.16.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 16 Duration of labour (hours).

Analysis 2.17.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 17 Maternal weight 4 weeks' postpartum (kg).

Analysis 2.18.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 18 Height at age 11-17 years (cm).

Analysis 2.19.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 19 Weight at 11-17 years (kg).

Analysis 2.20.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 20 Systolic blood pressure at age 11-17 years (mmHg).

Analysis 2.21.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 21 Diastolic blood pressure at age 11-17 years (mmHg).

Analysis 2.22.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 22 BMI z-score at age 11-17 years.

Analysis 2.23.

Comparison 2 Balanced protein/energy supplementation in pregnancy, Outcome 23 % body fat at 11-17 years.

Comparison 3. High protein supplementation in pregnancy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Stillbirth1529Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.31, 2.15]
2 Neonatal death1529Risk Ratio (M-H, Fixed, 95% CI)2.78 [0.75, 10.36]
3 Small-for-gestational age1505Risk Ratio (M-H, Fixed, 95% CI)1.58 [1.03, 2.41]
4 Birthweight (g)1504Mean Difference (IV, Fixed, 95% CI)-73.0 [-171.26, 25.26]
5 Preterm birth1505Risk Ratio (M-H, Fixed, 95% CI)1.14 [0.83, 1.56]
6 Weekly gestational weight gain (g/week)1486Mean Difference (IV, Fixed, 95% CI)4.5 [-33.55, 42.55]
7 Weight at 1 year (g)1409Mean Difference (IV, Fixed, 95% CI)61.0 [-184.60, 306.60]
8 Length at 1 year (cm)1412Mean Difference (IV, Fixed, 95% CI)0.20 [-5.59, 5.99]
9 Head circumference at 1 year1412Mean Difference (IV, Fixed, 95% CI)0.11 [-0.19, 0.41]
10 Bayley mental score at 1 year1396Mean Difference (IV, Fixed, 95% CI)0.32 [-0.91, 1.55]
Analysis 3.1.

Comparison 3 High protein supplementation in pregnancy, Outcome 1 Stillbirth.

Analysis 3.2.

Comparison 3 High protein supplementation in pregnancy, Outcome 2 Neonatal death.

Analysis 3.3.

Comparison 3 High protein supplementation in pregnancy, Outcome 3 Small-for-gestational age.

Analysis 3.4.

Comparison 3 High protein supplementation in pregnancy, Outcome 4 Birthweight (g).

Analysis 3.5.

Comparison 3 High protein supplementation in pregnancy, Outcome 5 Preterm birth.

Analysis 3.6.

Comparison 3 High protein supplementation in pregnancy, Outcome 6 Weekly gestational weight gain (g/week).

Analysis 3.7.

Comparison 3 High protein supplementation in pregnancy, Outcome 7 Weight at 1 year (g).

Analysis 3.8.

Comparison 3 High protein supplementation in pregnancy, Outcome 8 Length at 1 year (cm).

Analysis 3.9.

Comparison 3 High protein supplementation in pregnancy, Outcome 9 Head circumference at 1 year.

Analysis 3.10.

Comparison 3 High protein supplementation in pregnancy, Outcome 10 Bayley mental score at 1 year.

Comparison 4. Isocaloric balanced protein supplementation in pregnancy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Birthweight (g)2184Mean Difference (IV, Random, 95% CI)108.25 [-220.89, 437.40]
2 Weekly gestational weight gain (g/week)2184Mean Difference (IV, Random, 95% CI)110.45 [-82.87, 303.76]
Analysis 4.1.

Comparison 4 Isocaloric balanced protein supplementation in pregnancy, Outcome 1 Birthweight (g).

Analysis 4.2.

Comparison 4 Isocaloric balanced protein supplementation in pregnancy, Outcome 2 Weekly gestational weight gain (g/week).

Appendices

Appendix 1. Methods used to assess trials included in previous versions of this review

The following methods were used to assess Atton 1990a; Badrawi 1993; Blackwell 1973; Briley 2002; Campbell 1975; Campbell 1983; Campbell Brown 1983a; Ceesay 1997; Elwood 1981; Girija 1984; Hankin 1962a; Hunt 1976; Kafatos 1989; Kardjati 1988; Mardones 1988a; Mora 1978; Ross 1938a; Ross 1985; Rush 1980; Sweeney 1985; Viegas 1982a; Viegas 1982b; Wolff 2008.

Selection of studies  

We assessed all potential studies we identified as a result of the search strategy for inclusion in the review, without consideration of the results. We resolved any disagreement through discussion.

Data extraction and management  

Both review authors extracted the data using the agreed form. We used the Review Manager software (RevMan 2003) to double-enter all data.

When information regarding any of the above was unclear, we attempted to contact the authors of the original reports to obtain further details.

Measures of treatment effects  

We carried out statistical analysis using the Review Manager software (RevMan 2003). We used a fixed-effect meta-analysis for combining data in the absence of significant heterogeneity if trials were sufficiently similar. If heterogeneity was found, we used random-effects meta-analysis.

Dichotomous data

For dichotomous data, we presented the results as the summary relative risk with the 95% confidence intervals.

Continuous data

For continuous data, we the weighted mean difference when the outcomes were measured in the same way between trials.

Unit of analysis issues  

Cluster-randomised trials

We included cluster-randomised trials in the analyses along with individually-randomised trials. Their sample sizes were adjusted by using the methods described in Gates 2005 using an estimate of the intra-cluster correlation co-efficient (ICC) derived from the trial (if possible) or from another source. Two of the trials (Ceesay 1997; Kafatos 1989) gave no published or unpublished data on the outcome-specific ICC. Therefore, we assumed a value of 0.01 and adjusted the corresponding sample sizes according to the design effect, i.e. by dividing the crude (individual) sample sizes by 1+ (m -1)r, where m is the average cluster size and r is the ICC (assumed to be 0.01).

Dealing with missing data  

We analysed data for all participants with available data in the group to which they were allocated, regardless of whether or not they received the allocated intervention. If the participants were not analysed in the group to which they were randomised in the original report, and there was sufficient information in the trial report, we attempted to restore them to the correct group.

Assessment of heterogeneity  

We applied tests of heterogeneity between trials, if appropriate, using the I² statistic. If we identified high levels of heterogeneity among the trials (exceeding 50%), we explored it by performing sensitivity analysis. A random-effects meta-analysis was used as an overall summary when considered appropriate.

Since observational studies (Kramer 1987; IOM 1990) suggest a stronger association between gestational weight gain and fetal growth in women who were under-nourished before pregnancy, we stratified the analysis of the effects on mean birthweight into those trials in which the majority of women had low pre-pregnancy (or early pregnancy) weight (Atton 1990a; Campbell Brown 1983a; Ceesay 1997; Girija 1984; Kardjati 1988; Mora 1978; Rush 1980), and those in which the participants appeared adequately nourished (Elwood 1981; Ross 1985; Viegas 1982a). For the Taiwan trial (Blackwell 1973) and Viegas 1982b, within-trial stratification was possible, based on data contained in the published reports.

What's new

DateEventDescription
22 July 2011New citation required and conclusions have changedNutritional advice to increase energy and protein intakes is associated with significant increases in protein intake. Balanced energy and protein supplementation is associated with significant increases in mean birthweight, although this difference disappeared after excluding one trial of weekly gestational weight gain. The other findings have not changed.
22 July 2011New search has been performed

A new team of review authors prepared this updated review.

Search updated. Ten new trials identified: one has been included (Huybregts 2009) and eight excluded (Aaltonen 2011; Behrman 2009; Eneroth 2010; Fung 2010; Guelinckx 2010; Laitinen 2009; Luoto 2010; Rasmussen 2010). One trial is ongoing (Moore 2011).

The methods section has changed to include only RCTs and exclude quasi-RCTs or cross-over trials.

Six trials (Atton 1990; Campbell Brown 1983; Hankin 1962; Iyengar 1967; Mardones 1988; Ross 1938), previously included in the analysis, have now been excluded because of their quasi-RCT design.

The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for two different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain".

Five trials (Badrawi 1993; Campbell 1975; Campbell 1983; Guelinckx 2010; Wolff 2008), previously included in the analysis, have now been excluded because the target population was out of focus.

Three reports from an updated search in July 2012 have been added to Studies awaiting classification for consideration at the next update.

History

Protocol first published: Issue 2, 1997
Review first published: Issue 2, 1997

DateEventDescription
22 December 2009New search has been performedSearch updated. One new trial included (Wolff 2008) and two excluded (Aaltonen 2005; Kinra 2008).
2 September 2008AmendedConverted to new review format.
30 November 2006New search has been performedNew search conducted in November 2006 identified eight new reports to evaluate (Anderson 1995; an additional report of Clapp 1997; Fard 2004; Kaseb 2002; Moses 2006; additional reports of Lechtig 1975; Woods 1995), none of which were eligible for inclusion in the update. We have substantially updated the Methods of the review section.
1 August 2003New search has been performed

This updated review combines and replaces five previous Cochrane reviews entitled ''Balanced protein/energy supplementation in pregnancy', 'Energy/protein restriction for high weight-for-height or weight gain during pregnancy' (CDSR 1996a), 'High protein supplementation in pregnancy' (CDSR 1996b), 'Isocaloric balanced protein supplementation in pregnancy' (CDSR 1996c) and 'Nutritional advice in pregnancy' (CDSR 1996d).

This combination was suggested by colleagues in the field, the PCG editors, and by the Cochrane Pregnancy and Childbirth Group's Consumer Panel.

Contributions of authors

Erika Ota (EO) and Rintaro Mori (RM) independently rated all the included studies for the risk of bias tables from the previous review. EO and Rupuam Tobe-Gai (RT) jointly applied the study selection criteria and extracted data from the included studies for updated trials. EO edited the updated results. RT, RM and Diane Farrar (DF) revised the manuscript. All the authors read and approved the final version to be published.

Declarations of interest

None known.

Sources of support

Internal sources

  • The University of Tokyo, Department of Global Health Policy, Graduate School of Medicine, Japan.

External sources

  • Department of Reproductive Health and Research and Department of Technical Cooperation among Countries, World Health Organization, Geneva, Switzerland.

Differences between protocol and review

The methods section has changed to include only randomised controlled trials (RCTs) and to exclude quasi-RCTs or cross-over trials. Six trials (Atton 1990; Campbell Brown 1983; Hankin 1962; Iyengar 1967; Mardones 1988; Ross 1938), previously included in the analysis, have now been excluded because of their quasi-RCT design. The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for two different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". Five trials (Badrawi 1993; Campbell 1975; Campbell 1983; Guelinckx 2010; Wolff 2008), previously included in the analysis, have now been excluded because the target population was out of focus. Six trials (Atton 1990; Campbell Brown 1983; Hankin 1962; Iyengar 1967; Mardones 1988; Ross 1938), previously included in the review, have now been excluded because of quasi-RCT design.

We have added subgroup analysis from original protocol since observational studies (IOM 1990; Kramer 1987) suggest a stronger association between gestational weight gain and fetal growth in women who were under-nourished before pregnancy, we stratified the analysis of the effects on mean birthweight into those trials in which the majority of women had low pre-pregnancy (or early pregnancy) weight (Ceesay 1997; Girija 1984; Kardjati 1988; Mora 1978; Rush 1980), and those in which the participants appeared adequately nourished (Elwood 1981; Ross 1985; Viegas 1982a). For the Taiwan trial (Blackwell 1973) and (Huybregts 2009; Viegas 1982b), within-trial stratification was possible, based on data contained in the published reports. Because growth varies with differences in sex (Onis 2007), it is desirable to compare growth between groups after adjusting for variations by sex. We conducted subgroup analysis separated by sexes for follow-up results of balanced protein and energy supplementation at the age of 11 to 17 years (height, weight, systolic blood pressure, diastolic blood pressure, BMI z-score, and body fat).

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Blackwell 1973

MethodsInterventions 'assigned randomly and blindly', but method not specified.
ParticipantsWell-nourished rural Taiwanese women with 'marginal' diets (estimated daily energy intake is approximately 2000 kcal and protein intake <= 40 g for adult women in this area from preliminary food survey in 1965).
InterventionsExperimental: chocolate-flavoured liquid supplement given twice daily beginning after prior birth and continuing during index pregnancy; supplement contained 40 g protein and 800 kcal energy plus vitamins/minerals.
Control: supplement containing vitamins and minerals only, but given at same times and for same duration.
OutcomesGestational weight gain, preterm birth, birthweight, small-for-gestational age, length, head circumference, and IQ at age 5.
Notes1) Data presented on dietary substitution, but based on meal survey only.
2) High alleged net energy supplement not associated with significantly higher gestational weight gain.
3) Discrepancies in first-infant LBW rates in 1981 vs 1973 reports.
4) Significant correlation between birthweight and energy (and supplement) intake in controls only.
5) Supplementation continued until 15 months postpartum; data on maternal postpartum weight therefore, omitted from review.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskAlthough details are not described, it says the study participants were randomly assigned.
Allocation concealment (selection bias)Unclear riskInsufficient information, the method of concealment is not described.
Blinding (performance bias and detection bias)
All outcomes
Low riskThe 2 supplements were similar and no women were considered to be able to distinguish them.
Incomplete outcome data (attrition bias)
All outcomes
Low risk506 out of 524 (96.5%) were with complete data for the analysis and this was not ITT.
Selective reporting (reporting bias)Unclear riskNo description on this. Not mentioned on registered protocol.
Other biasUnclear riskNo data were provided for background characteristics.

Briley 2002

MethodsRandomisation method not reported.
Participants27 low-income African-American women. Mean pre-pregnancy body mass index is within the normal range for both groups (intervention, 24.7±3.4, control, 23.2 ±4.1 kg/m2)
InterventionsExperimental: minimum of 6 individualised in-home nutrition assessment and counselling visits.
Control: 2 home visits without counselling.
OutcomesEnergy intake, gestational weight gain, birthweight, and preterm birth.
Notes1) 7 of 27 randomised women dropped out and not included in analysis.
2) Neither participants nor observers apparently blind to allocation.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskWomen were randomly assigned, though no detailed methods on randomisation were described.
Allocation concealment (selection bias)Unclear riskInsufficient information, the method of concealment is not described.
Blinding (performance bias and detection bias)
All outcomes
Low riskCounselling group is evident and interventions could not be blinded.
Incomplete outcome data (attrition bias)
All outcomes
High risk7 women dropped out of 27 women (74.1%) and no ITT.
Selective reporting (reporting bias)Unclear riskUncertain, as if the protocol was registered, etc. was not described.
Other biasLow riskThere was no significant difference in demographic background between the groups.

Ceesay 1997

MethodsCluster randomisation by village "using a stratified design according to village size", but no details provided on method of random allocation or concealment.
ParticipantsRural Gambian women from 28 villages with "chronically" marginal nutrition. Undernutrition more pronounced from June to October (the 'hungry' season involving low food supply and heavy agricultural work) than from November to May (the dry harvest season with adequate food supply and less strenuous work).The mean maternal body mass index measured after delivery was 20.7± 2.3 kg/m2 in the control group and 21.3 ± 2.8 kg/m2 in the intervention group.
InterventionsExperimental villages: 2 supplement biscuits containing roasted groundnuts, rice flour, sugar, and groundnut oil (4250 kJ (1017 kcal) energy, 22 g protein, 56 g fat, 47 mg calcium, and 1.8 mg iron)] consumed daily in presence of birth attendants. Supplementation began at 20 weeks' gestation.
Control villages: no supplement.
OutcomesGestational weight gain, GA, birthweight, birth length, head circumference, stillbirth, and neonatal death.
Notes

1) Randomisation by cluster (village), but effects reported for individual births, based on multilevel (3-stage random-effects) modelling with separate error terms for village, mother, and (for mothers with more than 1 pregnancy during study) baby.
2) Results reported both overall and stratified by season (hungry vs harvest), but this review based on overall data. Note that definitions of seasons are not entirely consistent with previous (non-randomised) studies from this group and were chosen because 'post hoc analysis indicated that this selection yielded the greatest discrimination between hungry and harvest season effects'.
3) Many outcome analyses are based on individual women and therefore, do not account for the intra-class correlation among women living in the same village. Sample sizes in these outcomes have therefore, been adjusted downward to the nearest integer by dividing by 1+(m-1)r, where m is the average number of women per village and r = 0.01 is the (assumed) intra-class correlation co-efficient.

4) Data on LBW used in analysis of infant born small-for-gestational age.
5) Number of intervention and control participants reversed in column headings of Table 5.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskVillages were randomly assigned, but no details provided on method of random allocation.
Allocation concealment (selection bias)Unclear riskInsufficient information, the method of allocation concealment is not described.
Blinding (performance bias and detection bias)
All outcomes
High riskThe supplement biscuits provided in intervention group only. Intervention was evident.
Incomplete outcome data (attrition bias)
All outcomes
Low riskOver 95% agreed and remained in the trial throughout,The analysis presented here covers 2047 normal singleton live births from 1460 different women who delivered during October 1989 to October 1994.
Selective reporting (reporting bias)Unclear riskNot clear if the protocol was registered prior to the study.
Other biasLow riskSimilar between the groups and multilevel multiple regression was employed.

Elwood 1981

MethodsRandomisation based on random numbers with sealed envelopes.
Participants1251 pregnant Welsh women in 2 small towns recruited at time of first reporting of pregnancy in South Wales, UK. No information available for pregnant women's pre-pregnancy BMI.
InterventionsExperimental: free tokens worth ½ pint milk each.
Control: no intervention.
OutcomesGA, preterm birth, birthweight, LBW, length, and head circumference.
Notes1) 24% of women lost to follow-up, with evidence of higher losses in control group.
2) No adjustment for higher percentage of smokers in control group.
3) Trial also includes postnatal milk supplement (tokens) in children; all data on postnatal growth in children therefore, omitted from review.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation based on random numbers with sealed envelopes.
Allocation concealment (selection bias)Low riskRandomisation based on random numbers with sealed envelopes.
Blinding (performance bias and detection bias)
All outcomes
High riskAllocation was evident.
Incomplete outcome data (attrition bias)
All outcomes
Low risk212 were loss of follow-up. 82% were analysed.
Selective reporting (reporting bias)Unclear riskNot enough information was provided.
Other biasUnclear riskNot enough information was provided.

Girija 1984

MethodsRandomly allocated.
Participants20 poor Indian women in last trimester. Pregnant women's weight at last trimester was approximately 47 kg in both intervention group and control group.
InterventionsExperimental: supplement containing 50 g sesame cake, 40 g jaggery, and 10 g oil (417 kcal energy and 30 g protein).
Control: normal (unsupplemented) diet.
OutcomesGestational weight gain, birthweight, length, head circumference, breast milk output, and weight, length, and head circumference, through 3 months of age.
Notes1) Large losses to follow-up for breast milk output.
2) No SDs reported on postnatal anthropometric outcomes, so data not included in review.
3) No data reported on compliance or dietary substitution.
4) Energy and protein intakes appear higher before supplementation, even in supplemented group.
5) Mean GA (between 36 and 37 weeks in both groups) is incompatible with reported rates of preterm birth (0 of 10 in both groups), so data on preterm birth are omitted from review.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe participants were randomly assigned though no other details were provided.
Allocation concealment (selection bias)Unclear riskInsufficient information, the method of concealment is not described.
Blinding (performance bias and detection bias)
All outcomes
High riskThe allocation was evident.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo information was given.
Selective reporting (reporting bias)Unclear riskNo enough information was given.
Other biasUnclear riskNo description on demographic characteristics and others.

Hunt 1976

MethodsMethod of randomisation not reported.
Participants344 Spanish-speaking women with first prenatal clinic visit <= 21 weeks' gestation in Los Angeles. Pre-pregnancy self-report weight for intervention group was 127 ± 19 lb, and control group was 126 ± 23 lb.
InterventionsExperimental: nutrition classes (average of 3 per woman). Control: no classes.
OutcomesProtein and energy intakes; no data on gestational weight gain or pregnancy outcome.
Notes1) 65 women excluded or lost (not interviewed) post-randomisation.
2) Possible 'contamination' via contact between women in 2 groups.
3) No blinding.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe women were randomly assigned to a control or treatment group. Method of randomisation not reported.
Allocation concealment (selection bias)Unclear riskInsufficient information, the method of concealment is not described.
Blinding (performance bias and detection bias)
All outcomes
High riskNo blinding, possible ‘Contamination’ via contact between women in the groups.
Incomplete outcome data (attrition bias)
All outcomes
Low risk279 (81%) women were followed. Reasons for missing outcome data balanced in numbers across groups.
Selective reporting (reporting bias)Unclear riskNo protocol, insufficient information to permit judgement.
Other biasUnclear riskInsufficient information to assess whether an important risk of bias exist.

Huybregts 2009

MethodsA non-blinded,  individually randomised controlled trial. A randomisation scheme was generated by a computer program in permuted blocks of 4. Randomisation numbers were sealed in opaque envelopes by administrative staff.
Participants1296 Pregnant women in 2 villages In rural Burkina Faso. BMI at entry of the trial for intervention group was 20.8 ± 2.2 kg/m2, and control group was 21.0 ± 2.2 kg/m2.
InterventionsIntervention: prenatal multiple micronutrient (MMN) + fortified food supplement (FFS) Control: MMN.
OutcomesAnthropometric measures at birth, LBW, infant born small-for-gestational age, LGA, GA, preterm.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA randomisation scheme was generated by a computer program in permuted blocks of 4. Randomisation numbers were sealed in opaque envelopes by administrative staff.
Allocation concealment (selection bias)Low riskSequentially numbered, opaque, sealed envelopes.
Blinding (performance bias and detection bias)
All outcomes
Low riskNo blinding but care was taken to blind staff who performed the anthropometric measurements at delivery; measurement bias was therefore unlikely.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAnalysis for 87% of the 1175 live singleton deliveries enrolled.
Selective reporting (reporting bias)Low riskThe trial was registered at clinical trials.gov as NCT00909974.
Other biasLow riskThe compliance was closely verified by using a community-based network of home visitors.

Kafatos 1989

MethodsRandomisation of 20 clinics using computer-generated random numbers.
Participants568 pregnant women in rural area in Northern Greece < 27 weeks' gestation. Initial BMI was 23.10 ± 0.2 kg/m2in intervention group, and 22.7 ± 0.2 kg/m2 in control group.
InterventionsExperimental: nutrition counselling to improve 'quality' of diet ('high nutrient value'). Control: no counselling.
OutcomesEnergy and protein intake, serum vitamin and mineral levels, gestational weight gain, birthweight, birth length and head circumference, GA, LBW, infant born small-for-gestational age, preterm birth, stillbirth, and neonatal death.
Notes1) Analysis based on individual women, rather than clinic. To account for the intra-class correlation among women attending the same clinic, sample sizes have been adjusted downward to the nearest integer by dividing by 1+(m-1)r, where m is the average number of women per clinic (30.0 intervention and 26.8 control) and r = 0.01 is the (assumed) intra-class correlation.
2) Dietary intake unblinded, and energy intake higher in experimental group prior to intervention.
3) Inconsistent results: lower preterm rate, yet no difference in mean GA; higher head and chest circumferences but no difference in birthweight.
4) Discrepancies in sample sizes for different outcomes, even birthweight vs LBW rate.
5) SEM of GA in intervention (experimental) group assumed to be 0.10, not the 0.01 shown in Table 3.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA cluster randomisation of 20 clinics using computer-generated random numbers.
Allocation concealment (selection bias)Low riskRandomisation by clinic using computer-generated random numbers, clinic enrolled all women to minimise selection bias for allocation concealment.
Blinding (performance bias and detection bias)
All outcomes
Low riskNo blinding but the possible contamination effects of the educational program in that women from the same village or neighbourhood attending the same clinic would be enrolled in the same group.
Incomplete outcome data (attrition bias)
All outcomes
Low riskFor dietary records, Intervention group 216 (86.7%), control group 180(94.2%) were followed up after allocation.
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement.
Other biasHigh riskEnergy intake higher in experimental group prior to intervention.

Kardjati 1988

Methods"Blind" randomisation based on household numbers, with use of random-numbers table.
Participants747 women in 3 villages in rural East Java (an area known to be 'nutritionally vulnerable'(Kardjati 1983) at 26-28 weeks' gestation. Total mean ± SD pre-pregnant BMI was 18.7 ± 2.0 kg/m2.
InterventionsExperimental: supplement containing a dry powder (50% fat, 10% casein, and 40% glucose) providing 465 kcal energy and 7.1 g protein ('high energy').
Control: supplement containing 52 kcal energy and 6.2 g protein ('low energy').
OutcomesGestational weight gain, birthweight, and breast milk output.
Notes1) Although data on birthweight were not analysed according to ITT, they are included in this review because birthweight was similar in the 2 study groups and in non-compliers (both groups combined).
2) Data on gestational weight gain are based on the combined results in all 3 compliance strata but are missing for approximately one-third of study participants.
3) Data on breast milk output based on a selection of 50% of 'randomly'-selected study participants (only 10% of total study sample). Data excluded on 16 'uncooperative' or 'repeatedly absent' participants.
4) Data on postnatal infant growth reported in Kusin 1992 have been excluded from review, because poor compliers were excluded from the analysis (i.e., not based on ITT).
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskThe household numbers were the basis for allocation using random number tables.
Allocation concealment (selection bias)Low riskUsing random number tables.
Blinding (performance bias and detection bias)
All outcomes
Low riskWhile the study was not blind, the randomisation was, since the characteristics of the pregnant women cannot be inferred from the household numbers.
Incomplete outcome data (attrition bias)
All outcomes
High risk

Birthweight was recorded in 419 liveborn singletons (87%).

Gestational weight gain is missing for approximately one-third of study participants.

Selective reporting (reporting bias)Unclear riskProtocol not available.
Other biasHigh riskThe absence of a difference in mean birthweight between the HE and LE groups as a whole may be attributed to a masking effect of the better home diet in the experimental period.

Mora 1978

MethodsAllocation method not reported.
Participants456 poor first-or second-trimester women from Bogota slum for whom at least 50% of previous children had weight-for-height < 85% of Colombian standard. No information about maternal anthropometry (weight or BMI) provided.
InterventionsExperimental: supplement containing 60 g dried skim milk, 150 g enriched bread, and 20 g vegetable oil (856 kcal energy and 38.4 g protein) beginning in third trimester.
Control: normal (unsupplemented) diet.
OutcomesPre-eclampsia, GA, preterm birth, birthweight, LBW, stillbirth, perinatal mortality, neonatal mortality.
Notes1) Compliance assessed but data not presented.
2) Substitution assessed by single 24-hour recall 8 weeks after starting supplement.
3) Preterm birth rate not increased, but higher mortality reported among those born preterm.
4) Data on term LBW used in analysis of infant born small-for-gestational age.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomly assigned either to a supplemented or an unsupplemented group.
Allocation concealment (selection bias)Unclear riskAllocation method not reported.
Blinding (performance bias and detection bias)
All outcomes
High riskNo blinding.
Incomplete outcome data (attrition bias)
All outcomes
Low riskSupplemented group (186/226, 82.3%), unsupplemented group(173/230, 75.2%) were followed. Total sample and subsample in table2 showed no significant difference in characteristics.
Selective reporting (reporting bias)Unclear riskNo protocol available, insufficient information to judge this.
Other biasHigh riskCompliance not mentioned.

Ross 1985

MethodsAllocation method not reported.
Participants127 Black South African women < 20 weeks' gestation.Study women averaged > 70 kg at 20 weeks.
InterventionsExperimental: supplement containing 700-800 kcal energy and 36-44 g protein. 2 types of supplements were given: a high-bulk mixture of beans and maize, given as mush with added vitamins, and a low-bulk porridge containing dried skimmed milk, maize, flour, vitamins, and minerals; the high- and low-bulk groups are combined in the experimental group for this review.
Control: placebo pills (zinc-supplemented group is excluded from review).
OutcomesGestational weight gain (after 20 weeks), GA, and birthweight.
Notes1) Higher gestational weight gain in control group argues against causal association with birthweight.
2) No data presented on compliance or substitution.
3) Number of women originally randomised not reported ('90% continued ... to delivery').
4) Original sample size not given nor its justification.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk127 Zuru women were randomly assigned to 4 groups.
Allocation concealment (selection bias)Unclear riskAllocation method not reported.
Blinding (performance bias and detection bias)
All outcomes
High riskNo blinding.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNumber of women originally randomised not reported, or no missing case.
Selective reporting (reporting bias)Unclear riskProtocol is not available.
Other biasHigh riskNo data presented on compliance or substitution.

Rush 1980

MethodsStratified randomisation based on table of random numbers, with allocation in sealed envelope and blinding of all research staff.
Participants1051 low-income black women in Harlem (New York City) <= 30 weeks' gestation 'at risk' for LBW based on 1 or more of the following criteria: 1) pre-pregnancy weight < 110 lbs; 2) pre-pregnancy weight 110-139 lbs plus low gestational weight gain as of recruitment;
3) pre-pregnancy weight 110-139 lbs plus previous history of LBW; or
4) pre-pregnancy weight 110-139 lbs plus protein intake < 50 grams in the 24 hours preceding registration.
InterventionsExperimental (1): balanced energy/protein 16-oz beverage supplement containing 322 kcal energy, 6 g protein, and vitamins/minerals ('complement').
Experimental (2): high-protein 16-oz beverage supplement containing 470 kcal + 40 g protein per day + vitamins and minerals.
Control: supplement containing vitamins/minerals only.
OutcomesGestational weight gain, GA, preterm birth, infant born small-for-gestational age, birthweight, LBW, stillbirth, neonatal mortality, and weight, length, head circumference, and Bayley scores at 1 year.
NotesAlmost no data presented on the (approximately) 25% of participants who failed to comply, dropped out, or moved away.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom assignment.
Allocation concealment (selection bias)Low riskBased on table  of random numbers, with allocation in sealed envelope.
Blinding (performance bias and detection bias)
All outcomes
Low riskBlinding of all research staff.
Incomplete outcome data (attrition bias)
All outcomes
Low risk94% follow-up, 48 whose mothers chose to discontinue supplements and 1 infant with Down’s syndrome excluded.
Selective reporting (reporting bias)Unclear riskNo protocol available.
Other biasHigh riskAlmost no data presented on the (approximately) 25% of participants who failed to comply, dropped out, or moved away.

Sweeney 1985

MethodsStratified randomisation 'using biased coin methodology'.
Participants47 healthy women < 20 weeks' gestation in Salt Lake, USA. Maternal height ranged from 152 to 180 cm, and pre-pregnant weight ranged from 41 to 113 kg.
InterventionsExperimental: Higgins' method of protein/energy 'prescription' (i.e., advice only, no supplementation) Control: no advice.
OutcomesProtein and energy intake, gestational weight gain, birthweight, and GA.
Notes1) Slight discrepancy in number of women allocated.
2) Mean and SD weight gain, birthweight, and GA not reported by allocation group.
3) Probable non-blinding of intake (protein and energy) histories.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskStratified random allocation using a biased coin methodology.
Allocation concealment (selection bias)Unclear riskInsufficient information to permit judgement for allocation concealment.
Blinding (performance bias and detection bias)
All outcomes
High riskProbable non blinding.
Incomplete outcome data (attrition bias)
All outcomes
Low risk43 out of 47 (91.5%) were analysed.
Selective reporting (reporting bias)Unclear riskNo protocol available.
Other biasHigh riskSlight discrepancy in number of women allocated.

Viegas 1982a

MethodsAllocation method not reported.
Participants153 Asian women in Birmingham, UK < 20 weeks' gestation who appeared well-nourished based on their weight and height. The mean ± SD height and weight for intervention group was 154.6 ± 4.4 cm, 53.0 ± 9.1 kg and control group was 156.5 ± 5.8 cm, 56.3 ± 10.6 kg.
InterventionsExperimental: supplement of flavoured carbonated glucose drink providing 273 kcal energy (with 11% of energy as protein) plus vitamins from 18 to 38 weeks.
Control: supplement of flavoured carbonated water containing iron and vitamin C.
OutcomesGestational weight gain and birthweight, placental weight, maternal skin folds and arm circumference.
Notes1) Designed as 3-arm trial, but group receiving supplement with 11% of energy provided as protein combined with energy-only group for this review.
2) No evidence that study women were undernourished.
3) No data presented on compliance or dietary substitution.
4) Results presented only in graphic form; extracted data are therefore approximate.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskMinimisation. Allocation to a particular regimen was designed to give as closely as possible the same distribution of parity, abnormal past obstetric history and history of early bleeding in the current pregnancy.
Allocation concealment (selection bias)Unclear riskAllocation method not reported.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskBlinding not reported.
Incomplete outcome data (attrition bias)
All outcomes
Low riskProtein energy vitamin group: 47/51(92.1%), energy vitamin group: 50/57(87.7%), vitamin group 45 no missing.
Selective reporting (reporting bias)Unclear riskNo protocol available.
Other biasHigh riskNo data presented on compliance.

Viegas 1982b

  1. a

    BMI: body mass index
    GA: gestational age
    IQ: intelligence quotient
    ITT: intention to treat
    IVGTT: intravenous glucose tolerance test
    LBW: low birthweight
    LGA: large-for-gestational age
    SD: standard deviation
    SEM: standard error of the mean
    UK: United Kingdom
    vs: versus

MethodsAllocation method not reported.
Participants130 Asian women in Birmingham, UK < 20 weeks' gestation (who appeared well-nourished (based on height and weight) prior to pregnancy, 45 of whom were later considered "nutritionally at risk" based on inadequate increase in triceps skin folds between 18 and 28 weeks) stratified at 28 weeks according to increase in triceps skinfold during second trimester (<= 0.02 vs > 0.02 mm/week).
InterventionsExperimental: supplement of flavoured carbonated glucose drink + skim milk powder providing 425 kcal energy (with 10% of energy as protein), plus vitamins from 28 to 38 weeks.
Control: supplement of flavoured carbonated water containing iron and vitamin C.
OutcomesGestational weight gain, GA, birthweight, length, and head circumference, placental weight, and maternal skin folds.
Notes1) Designed as 3-arm trial, but group receiving supplement with 10% of energy provided as protein combined with energy-only group for this review.
2) No data presented on compliance or dietary substitution.
3) Results for gestational weight gain presented only in graphic form; extracted data are therefore approximate.
4) Probable misprint in Table II: mean GA in supplemented (EnVi = energy plus vitamins) group assumed to be 39.2 weeks, rather than the 29.2 weeks indicated in the table.
5) Data on outcomes stratified according to increase in triceps skin folds from 18-28 weeks. Because of harmful effect in those with normal skin folds and no statement that threshold was established a priori, outcomes from both strata have been combined in review.
6) Data from ITT analysis extracted from graph; not presented in tabular form.
7) Probable misprint in GA for control group (adequate skinfold stratum) in Table II.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskMembers of each group were then assigned at random to 1 of 3 supplement groups.
Allocation concealment (selection bias)Unclear riskAllocation method not reported.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskBlinding not reported.
Incomplete outcome data (attrition bias)
All outcomes
Low riskOnly 2 missing in EnVi group, reason for perinatal death with detailed. 128 out of 130 (98.5%)
Selective reporting (reporting bias)Unclear riskProtocol not available.
Other biasHigh riskNo data for compliance.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    RCT: randomised controlled trial
    vs: versus

Aaltonen 2005Intervention involved advice to alter fat composition of the diet, but not to change its energy or protein content.
Adams 1978Participants were high-risk women only.
Anderson 1995The nutritional advice studied does not relate to energy or protein intake, or both.
Atton 1990Not randomised, alternate allocation.
Badrawi 1993The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for 2 different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". This trial previously included in the analysis, have now been excluded because the target population was out of focus.
Campbell 1975The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for 2 different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". This trial previously included in the analysis, have now been excluded because the target population was out of focus.
Campbell 1983The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for 2 different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". This trial, previously included in the analysis, has now been excluded because the target population was out of focus.
Campbell Brown 1983Not randomised, alternate allocation.
Clapp 1997Experimental intervention involved no change in energy or protein intake, but only in the type of carbohydrate in the diet. Moreover, the only outcomes studied were glycaemic (blood glucose) responses to diet and exercise.
Dirige 1987The nutritional advice studied does not relate to energy or protein intake, or both.
Ebbs 1941Not randomised.
Eneroth 2010Follow-up analysis of Matlab (MINIMat) study, intervention is not relevant to our review.
Fard 2004RCT of maternal dietary fat modification with no net supplementation of energy or protein.
Fung 2010Participants are not only pregnant women and outcome is not relevant to our review.
Guelinckx 2010The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for 2 different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". This trial previously included in the analysis, have now been excluded because the target population was out of focus.
Hankin 1962Not randomised, allocation by day of week.
Iyengar 1967Not randomised.
Kaseb 2002Not randomised, quasi-randomised study.
Kinra 2008Intervention was not randomised and included both prenatal and postnatal (for the infant/child) supplementation.
Lechtig 1975Despite the original RCT design, the reported results were based on observational analyses of the data. In 1 report of this trial (Delgado 1982), the results were indeed presented according to randomised treatment. This report was also excluded, however, because the analysis was based on individual women despite randomisation by village, was limited to women with data on length of gestation, and showed evidence of major problems in validity of gestational age measurements. Stein 2003, Webb 2005, Stein 2006, and Hodinott 2008 were also based on the treatment allocation as randomised but were excluded from analysis of long-term outcomes because the offspring were also supplemented, making it impossible to distinguish effects of prenatal maternal supplementation from those of postnatal supplementation of the infant/child.
Luke 2001Not randomised or quasi-randomised trial.
Mardones 1988Not randomised, alternate allocation.
Metcoff 1985Participants were high-risk women only.
Moses 2006RCT of diets with high vs low glycaemic index, with no net supplementation of energy or protein.
Qureshi 1973Not randomised, alternate allocation.
Ross 1938Not randomised, alternate allocation.
Tompkins 1954The target population was out of focus.
Tontisirin 1986Not randomised.
Wolff 2008The current inclusion of information about caloric restriction for women who are overweight or obese only serves to increase confusion as it requires discussion of the clinical implications for 2 different populations, thus we excluded the outcome of "energy and protein restriction in women who were overweight or showed high weight gain". This trial previously included in the analysis, have now been excluded because the target population was out of focus.
Woods 1995Small (n = 10) cross-over trial of high- vs low-protein diet without control group, but no pregnancy or offspring outcomes are analysed. The only outcomes reported are renal haemodynamic responses to a meat meal.

Characteristics of ongoing studies [ordered by study ID]

Moore 2011

Trial name or titleInvestigating the effects of pre-natal and infancy nutritional supplementation on infant immune development in The Gambia: the early nutrition and immune development (ENID) trial.
Methods3-way randomised controlled trial 
ParticipantsWomen (aged 18 to 45 years) resident in Kiang West Region, The Gambia, with pregnancy confirmed by urine test and ultrasound examination and with gestational age approximately 10-20 weeks. 
Interventions4 pregnancy interventions, to be given daily from 12 weeks' gestation until delivery: 
1. FeFol: iron-folate, 60 mg iron 400 µg folate, representing the usual standard of care during pregnancy, as per Gambian Government guidelines (control group). 
2. Multiple micronutrients: a combination of 15 micronutrients, specifically designed for use during pregnancy, and as formulated by UNICEF. A single tablet provides the recommended dietary allowance for each micronutrient, but we will supplement women in this arm of the trial with 2 daily MMN tablets. 
3. PE + FeFol: protein-energy and iron-folate. A food-based supplement developed by Valid International, providing a comparable level of iron and folate to the FeFol only arm, but with the addition of energy, protein and lipids. 
4. PE + MMN: protein-energy and multiple micronutrients. A micronutrient fortified food-based supplement also developed by Valid International, and providing comparable levels of micronutrients to the MMN arm (including FeFol), in addition to the energy and protein and lipid content. 
OutcomesInfant immunity development
Starting date01/10/2009
Contact information

Dr  Sophie  Moore

smoore@mrc.gm

NotesISRCTN49285450, Anticipated end date 30/09/2013

Ancillary