Effects of dietary interventions on neonatal and infant outcomes: a systematic review and meta-analysis.

See corresponding editorial on page 1220.

Effects of dietary interventions on neonatal and infant outcomes: a systematic review and meta-analysis1–3 Ellie Gresham, Julie E Byles, Alessandra Bisquera, and Alexis J Hure

INTRODUCTION

Nutrition before conception and during pregnancy is important to ensure a healthy pregnancy outcome (1). Poor maternal nutrition during pregnancy has been shown to lead to adverse birth outcomes (2–5) and long-term negative consequences for the developing fetus (5–8). A review by Abu-Saad and Fraser (2) summarized the observational evidence for maternal dietary intake and birth outcomes. The review concluded that maternal nutrition plays a crucial role in influencing fetal growth and birth outcomes, particularly low birth weight, preterm birth, and intrauterine growth restriction (2). The challenge is to identify

1298

whether dietary intake during pregnancy plays a causal role in the incidence of adverse birth and infant outcomes with consideration given to the mechanisms at play. Research has shown the importance of diet as a prevention strategy for some adverse neonatal outcomes, particularly the role of folic acid for the prevention of neural tube defects (9). The establishment of healthy eating behaviors in women, including adequately feeding women before and during pregnancy, is a priority for maternal health and birth outcomes (10). “The first 1000 d” is a current global movement and international public health priority aimed at improving nutrition for the mother and child between a woman’s pregnancy and her child’s second birthday (11, 12). The provision of the right nutrition during this 1000-d window can help ensure the child lives a healthy life (11, 12) and reduces the financial and social costs of poor health over the life course. There is a clear need to identify the best dietary interventions for pregnant women aimed at preventing adverse neonatal and infant outcomes. Even with small effect sizes, a simple dietary change as a public health nutrition strategy has the potential to affect a large number of women. Therefore, the objective of this study was to synthesize the best of the available evidence by conducting a systematic review and metaanalysis to determine whether dietary interventions before or during pregnancy have any effect on neonatal or infant outcomes. Dietary interventions could consist of dietary counseling, food and fortified food products, or a combination of both. METHODS

The review protocol was developed by using the Cochrane Handbook for Systematic Reviews of Interventions (13) and Preferred Reporting Items for Systematic Reviews and Meta-Analyses 1 From the Research Centre for Gender, Health and Ageing, School of Medicine and Public Health, Faculty of Health (EG, JEB, and AJH) and the Clinical Research Design Information Technology and Statistical Support Unit (AB), University of Newcastle, Callaghan, Australia. 2 Supported by an Australian Postgraduate Award (PhD scholarship; to EG) provided by the University of Newcastle, New South Wales. 3 Address reprint requests and correspondence to E Gresham, Research Centre for Gender, Health and Ageing, Hunter Medical Research Institute Level 4, University of Newcastle, University Drive, Callaghan, New South Wales, 2308 Australia. E-mail: [email protected]. Received November 24, 2013. Accepted for publication July 30, 2014. First published online September 3, 2014; doi: 10.3945/ajcn.113.080655.

Am J Clin Nutr 2014;100:1298–321. Printed in USA. Ó 2014 American Society for Nutrition

Downloaded from ajcn.nutrition.org at UNIV OF PRINCE EDWARD ISLAND on October 30, 2014

ABSTRACT Background: Nutrition plays a fundamental role in fetal growth and birth outcomes. Objective: We synthesized effects of dietary interventions before or during pregnancy on neonatal and infant outcomes. Design: Randomized controlled trials that assessed the whole diet or dietary components and neonatal or infant outcomes were included. Two authors independently identified articles to be included and assessed the methodologic quality. A meta-analysis was conducted separately for each outcome by using a random-effects model. Results were reported by dietary intervention as follows: 1) counseling, 2) food and fortified food products, or 3) a combination (counseling plus food) intervention, and 4) collectively for all dietary interventions. Results were subanalyzed by the nutrient of interest, country income, and BMI. Results: Of 2326 abstracts screened, a total of 29 randomized controlled trials (31 publications) were included in this review. Food and fortified food products were effective in increasing birth weight [standardized mean difference (SMD): 0.27; 95% CI: 0.14, 0.40; P , 0.01] and reducing the incidence of low birth weight (SMD: 20.22; 95% CI: 20.37, 20.06; P , 0.01). All dietary interventions and those focused on macronutrient intake also increased birth weight (P , 0.01) and length (P , 0.05) and reduced the incidence of low birth weight (P , 0.01). Dietary interventions in low-income countries and underweight or nutritionally at-risk populations increased birth weight (P , 0.05) and reduced the incidence of low birth weight (P = 0.01). No effects were seen for the following other outcomes: placental weight, head circumference, macrosomia, Apgar score, small for gestational age, large for gestational age, and perinatal mortality. Conclusion: Additional high-quality randomized controlled trials that test different dietary interventions are required to identify maternal diet intakes that optimize neonatal and infant outcomes. Am J Clin Nutr 2014;100:1298–321.

1299

DIETARY INTERVENTIONS DURING PREGNANCY

RESULTS

Description of studies The trial-selection process is summarized in Figure 1. Of 2326 articles that were screened, data were included from 29 randomized controlled trials (RCTs) (31 publications), which included 10,026 participants. The earliest published study was in 1978 (31) and the latest published study was in 2011 (32–35). Twenty-three publications were performed in high-income Organisation for Economic Co-operation and Development countries as follows: 10 4 Abbreviations used: RCT, randomized controlled trial; RMD, raw mean difference; SMD, standardized mean difference.

TABLE 1 Criteria for the selection of studies Criteria Inclusion

Exclusion

Studies that reported any neonatal or infant outcomes in pregnant women of any age, weight, or BMI without date limits Three or more randomized controlled trials per birth or infant outcome Any healthy human population Randomized or pseudorandomized controlled trials on dietary interventions, including counseling, food, or both provided by any health professional Any intensity, frequency, or timing of intervention Positive or neutral methodologic quality1 Health conditions that may influence dietary intake (ie, unrepresentative sample such as gestational diabetes) Studies published in languages other than English Studies in animals Case study, editorial, or conference proceeding Trials solely on nutrient supplementation (ie, tablet form; no macronutrient change) Multiple births Studies in duplicate populations (ie, data set reporting the smaller number of participants for the same neonatal or infant outcome was excluded) Studies not meeting the Quality Criteria Checklist (ie, negative quality)1

1

American Dietetic Association Quality Criteria Checklist for Primary Research (17).

publications in the United States, 4 publications in the Netherlands, 3 publications in Denmark, 2 publications in Finland, and one publication each in Australia, Greece, Italy, and Norway (Table 2). The methodologic quality and characteristics of included studies are shown in Table 2. Sixteen publications compared dietary counseling with standard antenatal care (no active intervention) (20, 33–47), 2 publications of which came from a trial in the Netherlands that examined the effect of sodium restriction (45, 46). Fourteen publications evaluated the effect of specific foods and fortified food products (18, 21–23, 31, 32, 48–55), 2 publications of which came from the Columbian Longitudinal Study of Malnutrition and Intellectual Development (31, 54). One publication from the Finnish Mother-Infant Nutrition and Probiotic Intervention assessed the effect of combined dietary counseling and food and fortified food products (56). All but one publication (52) studied the effect of dietary intervention during pregnancy on neonatal and infant outcomes. McDonald et al (52) included a prepregnancy (preconception) dietary intervention. Preconception was defined as #3 mo before pregnancy. Twenty-three (of 31) publications included 2 groups of dietary intake (intervention) compared with usual care or dietary intake (control). Dietary interventions ranged in duration from 10 wk (23) to .40 wk (prepregnancy intervention) (52). Dietitians or nutritionists were the most frequent dietary intervention providers (11 publications), and macronutrients (25 publications) were more commonly targeted in the intervention than were micronutrients (6 publications). Fifteen publications did not report dietary data (18, 22, 32, 34–37, 42–45, 48, 49, 53, 55), and 13 of these publications did not conduct nutritional assessments during the intervention period (18, 22, 35–37, 42– 45, 48, 49, 53, 55). Eight trials (9 publications) recruited women


for reporting methods and outcomes (14, 15). Details of the review protocol, including eligibility criteria, search strategy, selection process, and data extraction have been described previously (16). Briefly, the inclusion and exclusion criteria for the selection of studies and flow diagram are presented in Table 1 (17) and Figure 1, respectively. The main measure of effect was the standardized mean difference (SMD)4. The SMD was used to express the size of the intervention effect in each publication relative to the variability observed within that study. The SMD is calculated by taking the difference in the mean outcome between the intervention and control groups in one publication divided by the pooled SD for the outcome across the whole trial. This method resulted in all measurements standardized to SD units. A secondary measure of effect, the raw mean difference (RMD), was used to present results in a scale that can be more easily interpreted (eg, grams for birth weight and centimeters for length). CIs have not been included for the RMD to emphasize the appropriateness of the SMD as the primary measure of effect. The single trial that contained more than one control group (18) had its effects averaged because there were no significant differences on neonatal outcomes (19). Trials that contained more than one intervention group (20–23) had their effects taken from the group and showed statistically significant results on neonatal (20–22) and infant outcomes (23). Dichotomized outcomes had log odds and SEs approximated to the SMD (24, 25), and means and variance were approximated for trials that reported medians (26). Trials that reported outcomes with zero counts for both intervention and control groups were excluded from the analysis. For outcomes that reported zero counts for one intervention group only, ORs and 95% CIs were approximated (27, 28). The I2 statistic was used to describe heterogeneity, and a random-effects model was used for meta-analyses for each neonatal and infant outcome, with random weights applied for each study. Additional analyses were performed for each outcome by dietary intervention type, combined dietary interventions, and subgroup analyses performed for each outcome by the nutrient of interest, BMI, and country income. Funnel plots were used to investigate the presence of a potential bias, and Egger’s test was used to test for funnel plot asymmetry (29). Statistical analyses were performed with the metan command in the statistical software package Intercooled Stata (version 12; StataCorp LP) (30). P , 0.05 was considered statistically significant.

1300

GRESHAM ET AL

who were underweight or nutritionally at risk (21, 22, 31, 40, 49, 51–54), whereas 5 trials recruited overweight or obese women as their target population (32, 34, 35, 44, 47). Neonatal outcomes included placental weight, birth weight, length, head circumference, low birth weight, macrosomia, Apgar score (at 5 min of age), Apgar score (,7 at 5 min of age), large for gestational age, small for gestational age, and perinatal mortality (including stillbirth and early fetal, fetal, neonatal, and perinatal mortality). Infant outcomes included body-size measures (including weight and height). Effects of dietary intervention Effects of dietary intervention components and all dietary intervention trials on meta-analyzed neonatal and infant outcomes are shown in Table 3. There was one trial in our review on a combination (counseling plus food) dietary intervention (56). Aaltonen et al (56) analyzed the effect of an intervention on the neonatal outcomes birth weight, length, head circumference, and Apgar score (at 5 min of age) and was meta-analyzed with all dietary intervention trials only. Neonatal outcomes Twenty-nine RCTs (31 publications) (n = 10,026 women) studied the effect of dietary intervention on neonatal outcomes (18, 20–23, 31–56). Fifteen RCTs studied the effect of dietary counseling on birth weight (n = 3287 women) (20, 33–45, 47), and 12 RCTs studied the effect of food and fortified food products (n = 5547 women) (18, 21–23, 32, 48–52, 54, 55). A meta-analysis showed a significant effect for an increase in birth weight (SMD: 0.27; 95% CI: 0.14, 0.40; P , 0.01, I2 = 74%) (Figure 2) for food and fortified food products. With the use of an RMD, this result translated to a mean increase of 125 g from a mean birth weight of 3169 g. A meta-analysis showed no

effect of dietary counseling on birth weight. However, overall there was a significant increase in birth weight for all dietary intervention trials (28 trials) (SMD: 0.19; 95% CI: 0.06, 0.32; P , 0.01, I2 = 87%), which was largely attributable to interventions that provided food and fortified food products (Figure 2). With the use of an RMD, this result translated to a mean increase of 62 g from a starting weight of 3277 g. Data showed a moderate heterogeneity for food and fortified food products and a high heterogeneity for dietary counseling and all dietary intervention trials. Pooled trials showed no evidence of bias. Six RCTs studied the effect of dietary counseling on length (n = 1383 women) (37, 39, 40, 43, 46, 47). Seven RCTs studied the effect of food and fortified food products on length (n = 3656 women) (18, 32, 48, 49, 51, 52, 55). There were no effects for either type of dietary intervention on length. However, when pooled in a meta-analysis, all dietary intervention trials (14 trials) showed a significant increase in length (SMD: 0.08; 95% CI: 0.01, 0.15; P = 0.03, I2 = 25%) (Figure 3). This amount translated to a mean increase of 0.07 cm from a starting length of 50 cm. Data showed no evidence of heterogeneity or presence of bias. Two RCTs studied the effect of dietary counseling on low birth weight (n = 908 women) (40, 42). Eight trials studied the effect of food and fortified food products (n = 4530 women) on low birth weight (18, 21, 49, 52–55, 57), showing a significant reduction in the incidence (SMD: 20.22; 95% CI: 20.37, 20.06; P , 0.01, I2 = 24%) (Figure 4). With the use of an RMD, this result translated to an OR of 0.73 or a 27% reduction in odds of having a low-birth-weight infant. A meta-analysis of all dietary intervention trials (10 trials) also showed a significant reduction in the incidence of low birth weight (SMD: 20.19; 95% CI: 20.32, 20.05; P , 0.01, I2 = 19%) with reduced odds of w24% (Figure 4). Data showed slight heterogeneity and no evidence of bias.


FIGURE 1. Flow diagram for the selection of included studies. EMBASE, http://www.elsevier.com/online-tools/embase; PreMedline, http://ovidsp.tx.ovid.com/ sp3.12.0b/ovidweb.cgi?&S=LDALFPMOPDDDLABINCMKADFBGBMNAA00&New+Database=Single%7c11; MEDLINE, http://www.ncbi.nlm.nih.gov/pubmed; Proquest, http://www.proquest.com/; Web of Science, http://thomsonreuters.com/thomson-reuters-web-of-science/; CINAHL, http://www.ebscohost.com/academic/ cinahl-plus-with-full-text; Scopus http://www.scopus.com/; Cochrane Library, http://www.thecochranelibrary.com; Mosby Index, http://www.mosbysindex.com/home; Maternity and Infant Care, http://ovidsp.tx.ovid.com/sp-3.12.0b/ovidweb.cgi?&S=LDALFPMOPDDDLABINCMKADFBGBMNAA00&New+Database=Single%7c9.

First author, year of publication, quality rating (ref)

Aaltonen, 2008, positive (56)

Briley, 2002, neutral (38)

RCT, Finland, Mother-Infant Nutrition and Probiotic Intervention, during pregnancy and #12 mo after birth; postnatal data excluded

RCT, United States, WIC, during pregnancy

Evaluate the impact of an in-home nutrition intervention during pregnancy on low birth weight

Evaluate the impact of maternal dietary intervention during pregnancy and after birth on 6-mo-old infant blood pressure

Study aims Whole diet, with an emphasis on fat and fiber. I1 had higher intakes of MUFA, PUFA, fiber, vitamin D, and vitamin E and lower intakes of SFA and percentages of energy from protein, riboflavin, and calcium than those in C. Assessed by 3-d food diaries, including one weekend day, with Micro-Nutrica3 to compute nutrient intake

Diet + placebo (I1) Individual dietitian counseling at 14, 24, and 34 wk of gestation. Aimed to comply with recommendations on the amount and type of fat, increasing unsaturated fatty acids, reducing SFAs; increase the consumption of vegetables, fruit, wholegrain products, lean meats, and low-fat dairy and the use of vegetable oil or margarine as spread and in food preparation. Food products with favorable fat and fiber compositions were provided for home use (eg, plant stanol ester spreads, low–erucic acid rapeseed oil–based spreads and salad dressing, fiber-enriched pasta and cereals). Diet + probiotic (I2, excluded) Placebo C Counseling (I1) Individual nutritionist counseling, commencing at #24 wk of gestation, occurring weekly for the first 4 wk and monthly for 2 more visits. Structured sessions on nutrition and special nutrient needs during pregnancy and breastfeeding. C Whole diet. I1 had higher increases in iron consumption than those in C. Assessed by 24-h dietary recall, with Food Processor Plus4 to compute nutrient intakes

Dietary modification, assessment, and compliance measures

Intervention (eg, provider and duration)

I1: n = 15 (25) C: n = 12 (22) (n = 7 did not complete study)

I1: n = 86 (27) C: n = 85 (26) (n = 13 withdrawn before birth)

Subjects, withdrawals

7% withdrawal from I1 + C before birth. No intentionto-treat analysis. Self-reported prepregnancy weight and food diary. Ethnically homogenous.

Dietary intervention in trimesters 2 and 3 did not change any neonate outcomes, except that, in subgroup analyses, women with and without GD in I1 delivered infants of the same birth weight and length, whereas C women with GD had heavier (426 g) and longer (1.7 cm) infants (P = 0.001)

I1 women produced heavier infants (480 g) than those of C (P , 0.05)

Neonate outcomes: birth weight, length, head circumference, and Apgar score (at 5 min)

Neonate outcome: birth weight

Age (y): 30.0 (17.6– 44.2); higherdegree education: 73%; smoking before pregnancy: 41%; prepregnancy BMI (in kg/m2): 23.6 6 3.8; white: 100%

Age ,21 y: 70%; high school education or less: 75%; annual income (,$15,000): 75%; unemployed: 60%

(Continued)

21% withdrawal from I1 + C before study completion. No intention-to-treat analysis. Small sample size. No blinding or allocation concealment reported. Low income, African American population. Self-reported prepregnancy weight

Limitations

Conclusion

Outcomes

Participant characteristics2


Study design, country, trial name, intervention period

TABLE 2 Characteristics of included studies of whole-diet interventions during pregnancy on neonate and infant outcomes1


1301


Quinlivan, 2011, positive (34)

Vinter, 2011, positive (35)

Kafatos, 1989, neutral (40)


RCT, Australia, during pregnancy

RCT, Denmark, LiP, during pregnancy

RCT, Greece, during pregnancy

Evaluate the impact of nutrition education during pregnancy on maternal weight gain and pregnancy outcomes

Evaluate the impact of maternal lifestyle intervention during pregnancy on gestational weight gain and obstetric and neonatal complications

Evaluate the impact of maternal dietary intervention during pregnancy on the incidence of GD

Study aims Whole diet. I1 increased their consumption of water, fresh fruit and vegetables, and home-cooked meals and decreased their intake of carbonated fizzy drinks, juices, and fast foods (fresh and frozen) compared with those in C. Assessed by using itemized dietary recall Whole diet. I1 reported morehealthy eating habits as a result of participating in the study. Compliance monitored by attendance at sessions. Assessed by using a studyspecific questionnaire Whole diet. I1 had higher intakes of protein, carbohydrate, and fats than did C. Assessed by 24-h dietary recall and food weighing. USDA’s Food Composition used to compute nutrient intake

Intervention (I1) Brief (5 min), individual food technologist counseling at every antenatal appointment, commencing in second trimester (13–15 wk of gestation). Provided information on reading food labels, shopping lists of affordable foods in local shops, and recipes for a healthy pregnancy. C

Intervention (I1) Individual in-home nurse counseling, for bimonthly visits, commencing at #21 wk of gestation. Instructed on the basics of nutrition during pregnancy; included food sources and methods for selecting a balanced diet and encouraged to consume locally grown foods. C

Intervention (I1) Individual dietitian counseling at 15, 20, 28, and 35 wk of gestation. Aimed to comply with national recommendations and individual energy requirements calculated. C



I1: n = 300 C: n = 268

I1: n = 180 (230) C: n = 180 (226) (n = 56 dropout after random assignment)

Withdrawals not disclosed. Blinding and allocation concealment not reported. Inclusion and exclusion criteria not disclosed. Low SES population from a rural, farming, and animal husbandry district.

Women in I1 produced infants with larger head circumference [0.99 cm (P , 0.001)] but had a higher number of early fetal deaths [n = 8 (P , 0.04)] than women in C did Neonate outcomes: stillbirth, birth weight, low birth weight, SGA, head circumference, length, early fetal death, and neonatal death

Age (y): I1, 23.2 6 (0.31); C, 22.9 6 (0.31); initial weight (kg): I1, 57.6 6 (4.8); C, 58.2 6 (5.2). Data presented by intervention group

(Continued)

16% withdrawal from I1 + C after randomization. No intention-to-treat analysis. Women recruited from 2 university hospitals. Obese population, women aged 18–40 y only. No nutritional assessment, analysis, or dietary data available. I1 women produced heavier infants (149 g) than those of women in C (P = 0.039) Neonate outcomes: stillbirth, birth weight, LGA, and macrosomia

Obese: 100%; age (y): I1, 29 (27–32); C, 29 (6–31); smoker: I1, 7.3%; C, 11.7%. Data presented by intervention group

Limitations 6% withdrawal from I1 + C after randomization. No intention-to-treat analysis. Women recruited from one public hospital in a socioeconomically disadvantaged area. Overweight and obese population. Dietary data not available for C.

Conclusion Dietary intervention in trimesters 2 and 3 did not change neonate outcome

Overweight or obese: 100%; age (y): I1, 28.3 (0.63); C, 29.5 (0.71); smoker: I1, 22%; C, 18%. Data presented by intervention group

I1: n = 67 (24) C: n = 65 (24) (n = 8 withdrawn after random assignment)

Outcomes Neonate outcome: birth weight




TABLE 2 (Continued )

1302 GRESHAM ET AL


Asbee, 2009, positive (36)

Bonomo, 2005, neutral (37)

Thornton, 2009, positive (44)


RCT, United States, during pregnancy

RCT, Italy, during pregnancy


Evaluate the effect of a nutritional intervention during pregnancy in obese pregnant women on perinatal outcomes

Evaluate the effect of a therapeutic intervention during pregnancy in women with very mild gestational glucose intolerance on pregnancy outcome

Evaluate the impact of maternal dietary and lifestyle intervention during pregnancy on weight gain

Study aims

Intervention (I1) Individual dietitian counseling, commencing at initial obstetric appointment (12–28 wk of gestation). Aimed to comply with GD guidelines (energy, carbohydrate, protein, and fat) and placed on a 18–24-kcal/kg diet. C

Whole diet with an emphasis on carbohydrate, protein, and fat intakes

Intervention (I1) Initial individual dietitian counseling, followed by individual nurse visits, commencing at initial obstetric appointment (#16 wk of gestation). Instructed on pregnancy-specific dietary and lifestyle choices, including recommendations for carbohydrate, protein, and fat. C BGGI intervention: treatment (I1) Initial individual counseling, then biweekly visits with doctors and nurses commencing at third trimester (w28 wk of gestation). Provided with dietary advice 24– 30-kcal/kg diet to be divided into 3 meals and 2–3 snacks. BGGI intervention: no treatment (I2) C (excluded) Whole diet with an emphasis on carbohydrate, protein, and fat. Compliance measured by using a daily food intake logbook

Whole diet with an emphasis on carbohydrate, protein, and fat. Assessed by discussion of dietary habits



Dietary intervention during pregnancy did not change any neonate outcomes Neonate outcomes: birth weight, Apgar score (,7 at 5 min), and macrosomia

Age (y): I1, {26.8}; C, {27.3}; married/defacto: I1, 75.9%; C, 73.3%. Data presented by intervention group


(Continued)

6.5% replacement from I1 + I2 after randomization. Participants recruited from one hospital. I1 and I2 had elevated glucose challenge test and normal oral glucose tolerance test to be classified as BGGI. Control group contained subjects from I1 + I2. No allocation concealment or blinding. Ethnically homogenous. No nutritional assessment, compliance measure, analysis, or dietary data available. 10% withdrawal from I1 + C after randomization. No blinding reported. Study conducted at principal investigator’s institutions. Obese population. No nutritional assessment, analysis, or dietary data available. Dietary intervention in the third trimester did not change any pregnancy outcomes Neonate outcome: birth weight, length, Apgar score (at 5 min), LGA, SGA, and macrosomia

Age (y): I1, 31.1 6 4.7; I2, 30.7 6 5.1; BMI: I1, 23.1 6 4.4; I2, 23.0 6 4.5; white: 100%. Data presented by intervention group

Limitations

I1: n = 165 (215) I2: n = 156 (26) (n = 21 replaced after random assignment)

Conclusion

30.5% withdrawal from I1 + C after randomization. No intention-to-treat analysis. Participant recruitment and prenatal care conducted in one Obstetric Clinic. No nutritional assessment, analysis, or dietary data available. No measure of participant compliance.

Outcomes Dietary intervention commencing in the second trimester did not change any neonate outcome

I1: n = 57 C: n = 435

5

Neonate outcome: birth weight

Participant characteristics2 Age (y): I1, 26.7 6 6.0; C, 26.4 6 5.0; BMI: I1, 25.5 6 6.0; C, 25.6 6 5.1; smoker: I1, 5.4%; C, 14%; high school graduate or less: I1, 68.4%; C, 65.1%. Data presented by intervention group





1303


Wolff, 2008, positive (47)

MardonnesSantander, 1988, neutral (21)


RCT, Denmark, during pregnancy

RCT, Chile, National Supplementary Food Program, during pregnancy

Whole diet with an emphasis on carbohydrate, protein, and fat. I1 lowered their intakes of energy, fat and carbohydrate and increased protein compared with those of C. Assessed by using 7-d weighed food records, Dankost 30006 to compute nutrient intakes Emphasis on protein; fat; carbohydrate; vitamins A, C, E, and B-3; folate; calcium; zinc; and iron. I1 had higher protein intake derived from supplements and lower iron intake than did I2. Compliance monitored by counting empty boxes and measuring the quantity in the most-recently opened box. Assessed by using 24-h dietary recall, and a Chilean-based computer program9 to compute nutrient intake

Intervention (I1) Ten 1-h individual dietitian counseling sessions, recruited at 15 6 3 wk of gestation. Aimed to comply with recommendations; was based on estimated individual requirements and energetic cost of fetal growth for carbohydrate, protein, and fat. C

Purita Powdered milk7 (I1) Midwives commenced supplementation at ,20 wk of gestation. Provided 2 kg milk supplement/mo with an information booklet detailing preparation and consumption. Nutritionist reinforced instructions and consumption once postrecruitment. Per 100 g provided 498 kcal, 27.9 g protein, and 26 g fat (26 g milk fat). Vita-Nova Motherfood Milk8–based fortified product (I2) Same protocol as I1. Per 100 g provides 470 kcal, 14.5 g protein, and 21 g fat (10.5 g milk fat) C (nonconsumers)

Evaluate the effect of dietary counseling during pregnancy in obese pregnant women on glucose metabolism

Evaluate the effect of milk-based food supplementation during pregnancy on fetal growth

Study aims



I1: n = 565 (2168) I2: n = 570 (2177) C: n = 38 (n = 345 dropout after allocation)



Data presented by intervention group; subclassified as follow-up losses; with and without complications, therefore not presented

Age (y): I1, 28 6 4; C, 30 6 5; BMI: I1, 34.9 6 4; C, 34.6 6 3. Data presented by intervention group


31% withdrawal from I1 + I2 after randomization. No concurrent control group. Control group was made up of nonconsumers of the fortified food product. No blinding or allocation concealment reported. Participants alternately assigned. Perinatal death data not reported for control group. Lowincome, underweight population. Study conducted during severe economic depression.

I2 women produced heavier infants (73.3 g), and had a lower incidence of infants born small birth weight (7.42%) than women in I1 did (P , 0.05) Neonate outcomes: birth weight, low birth weight, length, head circumference, and perinatal death

(Continued)

Limitations 24% withdrawal from I1 + C after randomization. No intention-to-treat analysis. Women excluded based on complications known to affect fetal growth. Ethnically homogenous. Small sample size. Obese population, aged 18–45 y only. Self-reported weight.

Conclusion Dietary intervention did not change any neonate outcomes

Outcomes Neonate outcomes: birth weight, length, placental weight, and head circumference



1304 GRESHAM ET AL

Evaluate the effect of a dietsupplementation program during pregnancy on birth weight and perinatal mortality

Evaluate the effect of a supplement beverage during pregnancy and lactation on offspring

Ceesay, 1997, positive (49)

McDonald, 1981, neutral (52)

RCT, Africa, during pregnancy and for 20 wk after birth for control group; postnatal data excluded

RCT, Taiwan, Bacon Chow Study, during lactation of first child until conception, gestation, and lactation of second child; first study infants excluded

Study aims


Dietary modification, assessment, and compliance measures Emphasis on energy, protein, fat, calcium, and iron. Compliance measured by birth attendants recording consumption

Emphasis on energy, protein, fat, and carbohydrate. I1 had higher intakes of energy (all trimesters) and protein (trimesters 1 and 2) compared with those of I2. Compliance monitored by measuring the volume of beverage remaining after delivery. Assessed by using weighed food samples, processed in Taipei

Intervention (eg, provider and duration) Intervention (I1) Midwives/birth attendants commenced supplementation from 20 wk of gestation. Supplied 2 biscuits/d that contained roasted groundnuts, rice flour, sugar, and groundnut oil, providing 4250 kJ energy, 22 g protein, 56 g fat, 47 mg Ca, and 1.8 mg Fe. C

Beverage (I1) Nurses commenced supplementation 3 wk postdelivery of first child. Supplied with one beverage twice daily, providing 800 kcal energy, 40 g protein, 26 g fat, and 100 g carbohydrates. Placebo C

I1: n = 10811 C: n = 10511 (n = 81 dropout after random assignment)

I1: n = 1061 C: n = 102110

10


Age (y) (at birth first study child): I1, 26.56 6 2.33; C, 26.52 6 2.79; prepregnant weight (kg): I1, 49.01 6 4.72; C, 48.4 6 4.87. Data presented by intervention group

Age (y): I1, 24.0 6 6.2; C, 23.7 6 6.4; parity: I1, 4.3 6 2.6; C, 4.2 6 2.4. Data presented by intervention group


(Continued)

38% withdrawal from I1 + C after randomization. Withdrawals not reported per group. No intentionto-treat analysis. Length of gestation and stillbirth data not reported. Farming area. Study conducted in low SES, nutritionally atrisk women, aged 19–30 y. Underestimation of food intake as assessment of dietary intake was limited to meal time consumption only.

No blinding or allocation concealment reported. Withdrawals not reported. Study conducted in rural, subsistence farming communities. Women aged 15–45 y, who were chronically undernourished. Ethnically homogenous. No nutritional assessment, analysis, or dietary data available. I1 women had reduced prevalence of low birth weight infants [5.9% (P , 0.001)] and stillbirths [n = 13 (P , 0.05)] and produced infants that were heavier [136 g (P , 0.001)] and had larger head circumference [3.1 mm (P , 0.01)] than did women in C Dietary intervention during conception and trimesters 1, 2, and 3 did not change any neonate outcomes Neonate outcomes: stillbirth, birth weight, low birth weight, length, and head circumference

Neonate outcomes: stillbirth, birth weight, low birth weight, length, and head circumference

Limitations

Conclusion

Outcomes





1305


Metcoff, 1985, positive (53)

Mora, 1978, neutral (31) Mora, 1979, neutral (54)


RCT, United States, The Special Supplemental Food Program for WIC, during pregnancy

RCT, Columbia, the longitudinal study of malnutrition and intellectual development, during pregnancy and #3 y after birth; postnatal data excluded

· ·

Intervention (I1) Project personnel commenced weekly supplementation from third trimester (w28 wk of gestation). Supplied with dry skim milk, protein-enriched bread, vegetable oil, and vitamin mineral supplement to meet large proportion of dietary allowance and instructed to consume in addition to the usual diet, providing 856 cal and 38.4 g protein/d. C

Evaluate the effect of supplementation during pregnancy and lactation on: perinatal and neonatal mortality birth weight Whole diet, with an emphasis on energy, protein, vitamin A, and iron. I1 increased their intakes of energy and protein (NS) and consumed more than did C. Compliance monitored by project personnel with random home visits and empty containers returned before qualifying for next supplement. Assessed by using 24-h dietary recall

Emphasis on energy and protein

Food vouchers (I1) Study personnel exchanged WIC vouchers monthly for WIC food supplement commencing from midpregnancy (w20 wk of gestation). Vouchers provided for milk, eggs and cheese, providing 900–1000 kcal energy and 40–50 g protein/d. C Additional nonstudy group (C1, excluded)

Evaluate the effect of food supplementation during pregnancy on birth weight

Study aims



I1: n = 226 (25) C: n = 230 (28) (n = 13 withdrawn from I1 + C) I1: n = 226 (219) C: n = 230 (230) (n = 49 withdrawn from I1 + C)

I1: n = 238 C: n = 17212 (n = 61 withdrawn from I1 + C)

13% withdrawal from I1 + C after random assignment. Withdrawals not reported per group. No blinding or allocation concealment reported. No intention-totreat analysis. Length of gestation, length, and head circumference data not reported. Recruited from one hospital. Enrollment based on prediction of birth weight. Participants are known for adverse socioeconomic factors, poor nutritional pattern and inadequate health care. Compliance was not assessed. No nutritional assessment, analysis or dietary data available 5% and 19% withdrawal respectively from I1 + C after random assignment. No allocation concealment reported. No intention-totreat analysis. Length of gestation not reported. Study conducted in a slum, urban poverty district. Participants at high risk of malnutrition.

I1 women produced heavier infants [91 g (P = 0.0395, adjusted)]13 than those of women in C

I1 women produced heavier full-term male infants [95 g (P , 0.05)] than those of women in C

Neonate outcomes: birth weight, low birth weight, length, and head circumference

Neonate outcomes: stillbirth, perinatal mortality, and neonatal mortality Neonate outcomes: birth weight, and low birth weight

Age (y): 21.9 6 4.4; education (y): 11.3 6 1.7; married: 57%; smoker: 46%; white: 74%

Age (y): I1, 25.6 6 5.4; C, 26.9 6 6.1; education (y): I1, {3.01}; C, {2.96}; family size: I1, 5.2 6 2.0; C, 5.9 6 2.5. Data presented by intervention group

(Continued)

Limitations

Conclusion

12

Outcomes





1306 GRESHAM ET AL


Ross, 1985, neutral (22)

Rush, 1980, positive (23)

Courville, 2011, neutral (32)


RCT, South Africa, during pregnancy



Evaluate the effect of DHA-containing functional food during pregnancy on intrauterine growth

Evaluate the effect of nutritional supplementation during pregnancy on birth weight and mental competence

Evaluate the effect of a low and high bulk supplement during pregnancy on fetal growth

Study aims

DHA-functional food (I1) Supplementation commenced from 24 wk of gestation. Provided with 3, 5, or 7 DHA cereal-based bars to be consumed weekly. Participants averaged 5/wk, providing 214 mg DHA/d. Each DHA bar contained 300 mg DHA [EPA (20: 5n23):DHA ratio of 1:8]. Placebo C

Emphasis on protein and energy. Compliance monitored by a dietitian who supervised the consumption of the supplement. Assessed by using 24-h dietary recall history (baseline only)

High-bulk supplement (I1) Dietitian commenced daily supplementation (5 d/ wk) from 20 wk of gestation. Supplement to be consumed in addition to usual diet, consisting of a mixture of beans and maize, providing 36 g vegetable protein, 3247 kJ energy, and 9.3 mg Fe. Low-bulk supplement (I2) Same protocol as I1. Porridge, providing 44 g protein (36 g animal), 2927 kJ, and 2.8 mg Fe. Zinc supplement (I3 excluded) Placebo C Supplement (I1) Nutritionists supplied two 8-oz cans/d to participants ,30 wk of gestation to be consumed in addition to the usual diet, providing 470 cal and 40 g animal protein. Complement (I2) Same protocol as I1. Providing 322 cal and 6 g animal protein. Placebo C

Emphasis on fat (DHA). Compliance through selfreported logs and uneaten bars returned. Assessed by using 24-h diet recall, with Nutrition Data Systems for Research15 to compute nutrient intake

Emphasis on protein. I1 had higher intake of protein than that of I2 and C. Assessed by using 24-h diet recalls and urinary excretion (riboflavin added to beverage to act as a marker)



(Continued)

Dietary intervention during trimester 2 and 3 did not change any neonate outcomes

Prepregnancy BMI: I1, 25.12 6 5.49; C, 26.08 6 5.17; parity: I1, 1.4 6 0.6; C, 2.2 6 1.2. Data presented by intervention group

Neonate outcomes: birth weight, length, head circumference, and Apgar score (at 5 min)

I1: n = 22 C: n = 25

6% withdrawal from I1, I2 + C after random assignment. Recruited from one urban hospital. Participants were poor and had high risk of low birth weight infant. Participants had to weigh ,140 pounds and have low weight gain, previous low birth weight infant, or protein intake ,50 g in 24 h to be enrolled. No intention-to-treat analysis. Infant weight and height were not reported accurately and, therefore, not metaanalyzed. Ethnically homogenous. Withdrawals not accounted for. Cohort derived from a larger, unpublishedstudy.Recruited from one hospital only. Participants aged 18–35 y. Ninety-three percent of participants participated in WIC concurrently. Overweight population. Withdrawals not reported. Small sample size. Selfreported prepregnancy weight and record of consumption. No dietary data reported.

I1 women produced lighter infants ,37 wk than did I2 women [323 g (P , 0.02)] and C [333 g (P , 0.01)] Neonate outcomes: birth weight, low birth weight, length, head circumference, placental weight, fetal death, and neonatal death Infant outcomes (1 y): weight and height

Age (y): I1, 21.5 6 4.7; I2, 21.9 6 4.8; C, 21.8 6 4.7; baseline weight (lb): I1, 120.9 6 14.7; I2, 120.0 6 14.0; C, 120.0 6 14.2; parity: I1, 1.9 6 1.1; I2, 1.8 6 1.2; C, 2.1 6 1.7; welfare recipient: I1, 26.5%; I2, 24.6%; C, 24.6%. Data presented by intervention group

I1: n = 263 (214)14 I2: n = 272 (216)14 C: n = 279 (215)14 (n = 46 withdrawn from I1, I2 + C)

Limitations Ethnically homogenous. Recruited from one hospital. Women malnourished before study commencing. Withdrawals not reported per group. Confounding factors not accounted for. No nutritional assessment, analysis, or dietary data available during or after supplement.

Conclusion I2 women produced heavier infants than those of women in C [205 g (P , 0.05)] and I1 [294 g (P , 0.005)]

Outcomes Neonate outcome: birth weight

Baseline weight (kg): I1, 71 6 10; I2, 73 6 8; C, 71 6 14; zulu: 100%. Data presented by intervention group


I1: n = 31 I2: n = 31 C: n = 33 (10% withdrew before delivery)





1307


Smuts, 2003a, positive (18)

Smuts, 2003b, positive (55)




Evaluate the effect of consuming highDHA eggs during the third trimester of pregnancy on pregnancy and birth outcomes

Evaluate the effect of consuming highDHA eggs during pregnancy on pregnancy outcome

Study aims

Dietary modification, assessment, and compliance measures Emphasis on fat (DHA) Compliance through self-reported logs and uneaten eggs returned. Assessed by using food-frequency questionnaire and the Harvard Nutrient Database to compute nutrient intakes (baseline only)

Emphasis on fat (DHA). Compliance through self-reported biweekly logs. Assessed by using food-frequency questionnaire, with the USDA Food and Nutrient Database to compute nutrient intakes (baseline only)

Intervention (eg, provider and duration) High DHA eggs (I1) Supplementation commenced from 24 to 28 wk of gestation. Provided with 2 dozen high-DHA eggs fortnightly to refrigerate immediately and cook before use. Placebo C (I2) C

High-DHA eggs (I1) Supplementation commenced from 24 to 28 wk of gestation. Provided with 12 highDHA eggs/wk (133 6 15 mg DHA/egg) to refrigerate immediately, consume as many as possible, and cook before use. Placebo C

I1 women produced infants with longer length at birth [0.66 cm (P = 0.048, unadjusted)] than those of women in C Neonate outcomes: birth weight, low birth weight, length, head circumference, and placental weight

Age (y): I1, 21.7 6 4.3; C, 21.6 6 4.2; BMI (at enrollment): I1, 29.4 6 5.9; C, 28.6 6 5.4; smoker (during pregnancy): I1, 27.0%; C, 21.5%. Data presented by intervention group I1: n = 176 (234) C: n = 174 (225) (n = 59 withdrawn from I1 + C)

(Continued)

Limitations Martek Biosciences Boulder Corp provided financial support (to EB) and provided eggs for study use. 27% withdrawal from I1, I2 + C after random assignment. No allocation concealment reported. No intention-to-treat analysis. Recruited from one prenatal clinic, aged 16– 35 y. Small sample size. Feasibility/pilot trial. Participants were mostly of African American descent. Participants were randomly assigned to I1 or C1 if they ate eggs. Selfreported record of consumption. No nutritional assessment, analysis, or dietary data available during or after supplement. 17% withdrawal from I1 + C after random assignment. No allocation concealment reported. No intention-totreat analysis. Recruited from one prenatal clinic. Aged 16–35 y. Participants were mostly black (73%) and received government assistance for medical care. High DHA eggs were commercially available in a number of regions during study. Self-reported record of consumption. No nutritional assessment, analysis, or dietary data available during or after supplement.

Conclusion Dietary intervention during trimester 3 did not change any neonate outcomes

Outcomes Neonate outcomes: birth weight, low birth weight, length, head circumference, and placental weight

Age (y): I1, 19.9 6 4.1; I2, 24.8 6 7.8; C, 21.3 6 4.8; prepregnancy weight (kg): I1, 68.6 6 19.1; I2, 61.5 6 11.5; C, 64.8 6 19.8; number of pregnancies: I1, 1.9 6 1.1; I2, 2.3 6 1.9; C, 1.8 6 0.9. Data presented by intervention group


I1: n = 27 (29) I2: n = 25 (26) C: n = 21 (25) (n = 20 withdrawn from I1, I2 + C)




1308 GRESHAM ET AL


de Groot, 2004, positive (50)

Fard, 2004, positive (39)


RCT, Netherlands, during pregnancy

RCT, Iran, during pregnancy and lactation

Evaluate the effect of modifying maternal dietary fat in the antenatal period and during infancy on serum lipids of infants at birth and 1 y of age

Evaluate the effect of consuming a-linolenic acid supplementation during pregnancy on maternal and neonatal PUFA status and pregnancy outcome

Study aims Emphasis on fat (ALA, LA precursors of DHA, and AA). I1 and C increased intakes of PUFAs and LA, whereas I1 increased ALA and total fat intakes compared with those of C. Compliance monitored by weighing leftovers to estimate consumption. Assessed by using food-frequency questionnaire and KOMEET software16 to compute nutrient intake Whole diet with an emphasis on fat. I1 and C increased their intakes of PUFAs and LA, whereas I1 increased ALA and total fat, and C decreased ALA. Assessed by using a 4-d food record, with an unspecified nutrient analysis to compute nutrient intake

ALA-enriched high-LA margarine (I1) Supplementation commenced from 14 wk of gestation. Provided with margarine weekly. Instructed to maintain usual diet, replacing butter or margarine with study margarine and use $25 g/d, primarily on bread and not in baking. If consumption was low, instructed to add to potatoes and pasta. Placebo C

Fat-modified diet (I1) Dietary modification commenced from 16 wk of gestation. Aimed to comply with recommendations on cholesterol; saturated, monounsaturated, polyunsaturated, and total fat; corn, safflower, and olive oil to replace hydrogenated fats; to increase consumption of seed oils, fish, low-fat dairy products, fruit, and vegetables, limiting meat and animal intakes at required intake and egg yolks to 3/wk. C



(Continued)

13% withdrawal from I1 + C after random assignment. Undisclosed recruitment procedure or setting. Aged 18–35 y. Participants with BMI $40 were excluded. Participants were included if they had atherogenic diets. No allocation concealment reported. Withdrawals not reported per group. No adjustments for confounding factors reported. Compliance measure not reported. Dietary intervention during trimester 2 and 3 did not change any neonate or infant outcomes Neonate outcomes: birth weight and length Infant outcomes (1 y): weight and height

Age (y): I1, 22.2 6 8.4; C, 23.7 6 9.1; weight (baseline) (kg): I1, 61.8 6 20.3; C, 64.4 6 19.8; BMI: I1, 24.7 6 6.4; C, 25.6 6 7.1; number of pregnancies: I1, 1.8 6 0.8; C, 1.7 6 0.5. Data presented by intervention group I1: n = 8618 C: n = 9418 (n = 24 withdrawn from I1 + C)

Limitations Unilever provided financial support and donated the margarines for study use. No allocation concealment reported. Small sample size. Ethnically homogenous. Participants had to consume fish ,2 times per week to be enrolled. Self-reported height, weight, and pregnancy outcomes.

Conclusion I1 women produced heavier infants [364 g (P = 0.043, adjusted)] than those of women in C17

Outcomes Neonate outcomes: birth weight and stillbirth

Age (y): I1, 30.0 6 3.3; C, 29.2 6 3.8; prepregnancy weight (kg): I1, 73.1 6 16.7; C, 70.4 6 13.0; 100% white. Data presented by intervention group


I1: n = 40 (211) C: n = 39 (210) (n = 21 withdrawn from I1 + C)





1309


Luoto, 2011, positive (33)


RCT, Finland, (NELLI), during pregnancy

Evaluate the effect of lifestyle counseling in pregnant women at high risk of GD on GD or newborns’ high birth weight

Study aims

Dietary modification, assessment, and compliance measures Whole diet with an emphasis on fat, fiber, and saccharose. I1 lowered their intakes of saturated fat and saccharose (baseline to 26–28 of wk gestation), saturated fat (36–37 wk of gestation), and saccharose (baseline to 36–37 wk of gestation) and increased their intakes of dietary fiber and polyunsaturated fat (baseline to 36–37 wk of gestation) compared with those of C. Compliance monitored by weekly log books. Assessed by using food-frequency questionnaire and Fineli software19 to compute nutrient intakes

Intervention (eg, provider and duration) Lifestyle counseling (I1) Individual nurse (public health) counseling sessions at 16–18, 22–24, 32–34, and 36–37 wk of gestation. Aimed to comply with recommendations on saturated, polyunsaturated, and total fat; saccharose; and fiber. Encouraged to consume vegetables, fruit, and berries; to select high-fiber bread and other whole-meal products, fat-free or low-fat versions of milk and milk products and of meat and meat products; to eat fish at least twice per week, to use moderate amounts of soft table spreads on bread, oil-based salad dressing in salad, and oil in cooking and baking, consume in small portions and seldom high-fat foods and snacks containing high amounts of sugar and fat. Provided with a leaflet highlighting recommendations. C I1: n = 246 (227) C: n = 196 (216) (n = 43 withdrawn from I1 + C)

Subjects, withdrawals Age (y): I1, 29.5 6 4.8; C, 30.0 6 4.7; BMI (prepregnancy): I1, 26.3 6 4.9; C, 26.4 6 4.3. Data presented by intervention group


Participants .18 y and had to have at least one GD risk factor to be included in study (previous GD, glucose intolerance or newborn macrosomia in previous pregnancy, or type 1 or 2 diabetes in first- or second-grade relative). No blinding reported. Self-reported prepregnancy weight. I1 women produced lighter [127 g (P = 0.035, adjusted)] and had a lower number of LGA infants [n = 8 (P = 0.043, adjusted)] than those of women in C

Neonate outcomes: birth weight, head circumference, LGA, SGA, and macrosomia

(Continued)

Limitations

Conclusion

Outcomes



1310 GRESHAM ET AL


Khoury, 2005, positive (41)

Kusin, 1992, neutral (51)


RCT, Norway, CARRDIP, during pregnancy

RCT, East Java, EJPS, during pregnancy

Evaluate the effect of high- and lowenergy supplements in pregnancy on the child’s growth for the first 5 y of life

Evaluate the effect of a cholesterollowering diet in pregnancy on maternal, cord, and neonatal cholesterol concentrations, and incidence of pregnancy-related complications

Study aims Whole diet, with an emphasis on fat and cholesterol. I1 consumed less energy, fat, saturated fat, carbohydrates as sugar, cholesterol, and calcium and had higher intakes of monounsaturated and polyunsaturated fat; protein; vitamins D, E, and C; carbohydrate; and magnesium compared with those of C. Compliance was assessed by using weighed dietary records and Beregn software20 to compute nutrient intake

Intervention diet (I1) Individual dietitian counseling at 24, 30, and 36 wk of gestation. Aimed to limit cholesterol and reduce saturated fat by replacing with polyunsaturated and monounsaturated fat. Encouraged intake of fatty fish, vegetable oil (olive and rapeseed), nuts, nut butters, margarine (olive or rapeseed oils), and avocado to replace meat, butter, cream, and fatty dairy products. Advised to consume fresh fruit and vegetables, skimmed and low-fat dairy products, meat, legumes, vegetable main dishes, fatty fish, or poultry with fat trimmed. Coffee limited to 2 filtered cups/d. Cooking lessons provided for special foods (ie, legumes and olive oil). C High-energy supplement (I1) Field workers commenced daily supplementation from 26 to 28 wk of gestation. Supplement was to be consumed in addition to the usual diet and provided 465 kcal and 7.1 g protein. Low-energy supplement (C) Same protocol as I1. Supplement provided 52 kcal and 6.2 g protein. Noncomplier (excluded) Emphasis on energy and protein. I1 and C had similar intakes of energy and protein. Compliance measured by field workers who weighed supplement not consumed. Assessed by using 3-d weighed food record and the Indonesian Food and Agricultural Organization food composition table to compute nutrient intakes



I1: n = 276 (2208) C: n = 266 (2182) (n = 390 excluded from I1 + C)

I1: n = 141 (214) C: n = 149 (27) (n = 21 withdrawn from I1 + C)


72% participants excluded from I1 + C because of incomplete records. Study conducted in a semiarid area. Participants were considered nutritionally at risk, with poor environmental hygiene and socioeconomic status. Underweight population, with participants short and lean. No concurrent control. Food records were recorded at cooking and eating times only.

I1 women produced heavier infants at 3 [236 g (P = 0.02)], 6 [359 g (P , 0.01)], 9 [463 g (P , 0.01)] and 12 mo [421g (P = 0.02)] and taller infants at 3 [0.8 cm (P = 0.05)], 6 [1.0 cm (P = 0.01)], 9 [1.2 cm (P = 0.01)], and 12 mo [1.3 cm (P , 0.01)] than those of women in C Neonate outcomes: birth weight and height Infant outcomes (3, 6, 9, 12 mo): weight and height

Age (y): 25.7 6 6.0; prepregnancy weight (kg): 42.7 6 4.8; primipara 20%. Data presented as total (n = 747) including noncomplier (n = 205) group

(Continued)

Limitations Funding provided by the Norwegian Council on Cardiovascular Disease. Recruited from one hospital. Participants aged 21–38 y, nonsmoking or quit within past 5 y or more, and excluded if underweight or obese. Ethnically homogenous. Computed intakes did not include vitamin or mineral supplementation. No adjustments for confounding factors reported.

Conclusion Dietary intervention during trimesters 2 and 3 did not change any neonate outcomes

Outcomes Neonate outcomes: stillbirth, birth weight, length, and head circumference

Age (y): I1, 29.6 6 3.7; C, 29.8 6 3.4; BMI (baseline): I1, 24.3 6 2.9; C, 24.3 6 2.7; education .12 y: I1, 79.4%; C, 84.6%; nulliparous: I1, 70.9%; C, 65.1%. Data presented by intervention group





1311


Chan, 2006, positive (20)

O’Connor, 2007, neutral (43)

Bech, 2007, positive (48)



RCT, United States, PHFE-WIC, during pregnancy

RCT, Denmark, during pregnancy

Evaluate the impact of reducing caffeine intake during pregnancy on birth weight and length of gestation

Evaluate the impact of brief intervention as a technique to achieve abstinence during pregnancy on newborn outcomes

Evaluate the impact of dietary calcium intervention during pregnancy on adolescent pregnant mothers and their newborns

Study aims

Decaffeinated coffee (I1) Project coordinator commenced supplementation from 18 wk of gestation. Provided with 6 boxes of decaffeinated coffee to replace usual. No restriction on other caffeinated products, ie, participants were not asked to avoid regular coffee offered by others, or to limit their intake of other caffeinated beverages such as tea, cocoa, or cola. Placebo C

Emphasis on caffeine. Compliance monitored by regular interviews at 20, 25, and 34 wk gestation and 4 wk after expected delivery date

Emphasis on calcium. I1 and I2 had higher intakes of calcium, phosphorus, and iron than those of C. I1 had higher intakes of potassium and zinc, and I2 had higher intake of vitamin D than those of C. I1 had higher intake of vitamin D compared with that of I2. Compliance measured by weekly monitoring of dietary intake and unscheduled 24-h dietary recalls. Assessed by 2-d dietary records and Nutritrac software21 to compute nutrient intakes Emphasis on alcohol. Compliance monitored by questionnaire on current drinking status (current drinker or abstinent) and goal setting

Orange juice and calcium (I1) Commenced from 20 wk of gestation. Counseled to consume $4 servings orange juice plus calcium (.1200 mg Ca)/d. Protocol changed to calcium-carbonate tablets Dairy (I2) Same protocol as I1; counseled to consume dairy products (.1200 mg Ca) daily (ie, yogurt, cheese, and milk). C

Brief intervention (I1) Brief (10–15 min) individual nutritionist counseling monthly until alcohol abstinence, commencing in first or second trimester (w13 wk of gestation). C Provided advice to cease drinking during pregnancy only



I1: n = 629 (254) C: n = 568 (228) (n = 82 dropped out before giving birth)

I1: n = 162 (245) C: n = 183 (245) (n = 90 did not complete study)

I1: n = 24 I2: n = 25 C: n = 23


26% withdrawal from I1 + C after random assignment. No blinding or allocation concealment reported. Low income minority population. Participants had to be currently drinking alcohol to be enrolled. The National Institute on Alcohol Abuse and Alcoholism collaborated on the research. No nutritional assessment, analysis or dietary data available. Participants recruited from one hospital. Participants had to be currently drinking at least 3 cups of coffee a day to be enrolled. No nutritional assessment, analysis or dietary data available.

Dietary intervention did not change any neonate outcomes, except that in subgroup analyses women with high consumption of alcohol in I1 delivered heavier (180.45 g) (P , 0.06) and shorter (1.69 cm) (P , 0.03) infants Dietary intervention during trimesters 2 and 3 did not change any neonate outcomes Neonate outcomes: stillbirth, birth weight, length, and fetal mortality

Neonate outcomes: birth weight, length, head circumference, SGA, placental weight, Apgar score (,7 at 5 min), and fetal death

Age (y): I1, 28.52 6 5.84; C, 27.9 6 6.09 married or has partner: I1, 71.9%; C, 71%; income $15000 or less: I1, 63.9%; C, 69.6%. Data presented by intervention group

Age (y): I1, 30.5 6 4.3; C, 30.7 6 4.3 prepregnancy weight (kg): I1, 67.8 6 13.3; C, 66.8 6 12.4; smoker: I1, 38%; C, 37.7%. Data presented by intervention group

(Continued)

Limitations Small sample size. Participants aged 15–17 y of age. No blinding reported. Withdrawals not reported. Confounding factors not accounted for.

Conclusion I2 women produced heavier infants than those of women in I1 (225 g) and C (240 g) (P , 0.001)

Outcomes Neonate outcomes: birth weight, length, and head circumference

Age (y): I1, 16.6 6 0.6; I2, 16.6 6 0.6; C, 16.7 6 0.6; BMI: I1, 26 6 5; I2, 25 6 4; C, 25 6 5. Data presented by intervention




1312 GRESHAM ET AL


Knuist, 1998, positive (42)

Van Buul, 1997, neutral (45) van der Maten, 1997, positive (46)


RCT, Netherlands, during pregnancy

RCT, Netherlands, during pregnancy I1: n = 133 (223) C: n = 137 (25) (n = 28 did not complete study) I1: n = 57 (216) C: n = 64 (211) (n = 27 did not complete study)

Emphasis on sodium. Compliance measured by 24-h urine samples Emphasis on sodium. I1 had lower intakes of total energy, protein, fat, carbohydrates, calcium, sodium, zinc, magnesium, iron, and cholesterol than those of C. Compliance measured by 24-h urine samples. Assessed by using 7-d weighed food records. Stichting Nevo software19 was used to compute nutrient intake

Low-sodium diet (I1) Individual dietitian counseling, commencing from 14 wk of gestation. Instructed to consume ,20 mmol sodium/d and given practical advice regarding shopping locations and commercial products available. Provided written dietary instructions on: no added salt during cooking or at the table; replacing milk, cheese, butter, and other dairy with low-sodium substitutes; bread and other bakery products baked without salt; consumption of readymade products if no salt added during preparation; recipes; forbidden and allowable food products and list of sodium content of several products. C

Evaluate the effectiveness of sodium restriction during pregnancy on: incidence of gestational hypertension blood pressure

·

·

I1: n = 184 C: n = 177

Emphasis on sodium. Compliance measured by 24-h urine samples

Low-sodium diet (I1) Individual midwife counseling, commencing from ,20 wk of gestation. Instructed to consume ,50 mmol sodium/d and provided written dietary instructions. C


Evaluate the effectiveness of sodium restriction during pregnancy on mild pregnancyinduced hypertension

Study aims



Age (y): I1, 28.1 (19.8–41.3); C, 28.3 (18.1–40.5); BMI: I1, 25.4 (20.0–43.2); C, 26.5 (19.5–45.8). Data presented by intervention group Age (y): I1, 29 6 5; C, 28 6 4; prepregnancy BMI: I1, 22.9 6 2.1; C, 23.5 6 2.9; smoker: I1, 34%; C, 21%. Data presented by intervention group

Age (y): I1, 27.6 6 4.2; C, 27.5 6 4.8 weight (kg): I1, 70.4 6 12.5; C, 69.7 6 12.1; smoker: I1, 34%; C, 31%; 86% white. Data presented by intervention group


Limitations Ethnically homogenous. Participants were nulliparous and excluded if increased risk of pregnancy-induced hypertension. Participants were randomly assigned only if increase in blood pressure, excessive weight gain, or edema. Adjustments not made for confounding factors. Low compliance (24%) with I1. No nutritional assessment, analysis, or dietary data available. 10% withdrawal from I1 + C after random assignment. Ethnically homogenous. Participants recruited from 2 hospitals (University Hospital, Saint Radboud and Bosch Medical Center). No blinding reported. No intention-totreat analysis. Small sample size. No nutritional assessment, analysis, or dietary data reported. The trial was ceased due to the detrimental side effects in the absence of beneficial effects. 22% withdrawal from I1 + C after random assignment. Ethnically homogenous. Nulliparous women only. Participants recruited from one hospital (Bosch Medical Center). No blinding reported. No intention-to-treat analysis. Small sample size.

Conclusion Dietary intervention during trimesters 2 and 3 did not change any neonate outcomes

Dietary intervention during trimesters 2 and 3 did not change any neonate outcomes

Outcomes Neonate outcomes: birth weight, low birth weight, Apgar score (,7 at 5 min), and perinatal mortality

Neonate outcomes: birth weight and SGA placental weight Neonate outcomes: length and head circumference




1313

GRESHAM ET AL

There was no effect of dietary intervention components or all dietary intervention trials on placental weight, head circumference, macrosomia, Apgar score (at 5 min of age and ,7 at 5 min of age), small for gestational age, and large for gestational age (Table 3). Perinatal mortality Ten RCTs (n = 6125 women) studied the effect of dietary intervention on perinatal mortality (23, 31, 35, 40–43, 48–50). Five RCTs studied the effect of dietary counseling on perinatal mortality (n = 1778 women) (35, 40–43); the other 5 trials provided food and fortified food products (n = 4347 women) (23, 31, 48–50). A meta-analysis showed no effect of dietary intervention components or all dietary intervention trials on perinatal mortality with no evidence of heterogeneity or bias. Infant body-size measures Three RCTs (n = 1773) studied the effect of dietary intervention on infant body-size measures (23, 39, 51). All trials examined the effect of dietary intervention on weight and height measures; however, a meta-analysis could not be performed because of the insufficient reporting of results. Subanalysis by nutrient of interest, BMI, and country income There were significant differences in effects of dietary interventions on birth weight, length, and low birth weight on the basis of the nutrient of interest, BMI, and country income (Table 4). Trials that altered the macronutrient composition of dietary interventions showed increases in birth weight (P , 0.01) and length (P = 0.03) and a reduction in the incidence of low birth weight (P = 0.01). With the use of an RMD, these results translated to an increase of 72 g in birth weight (from a starting weight of 3240 g), 0.2 cm in length (from a starting length of 50 cm), and a 0.76 OR or 24% reduction in odds for low birth weight, respectively. Dietary interventions conducted with underweight or nutritionally at-risk pregnant women, showed an increase in birth weight (P , 0.01) and a reduction in the incidence of low birth weight (P = 0.03). With the use of an RMD, these results translated to an increase of 101 g from a starting weight of 3112 g and a 0.76 OR or 24% reduction in odds for low birth weight. Trials that involved all weight categories showed a significant increase in length (P = 0.03) only, which corresponded to a 0.5cm increase from a starting length of 51 cm. Dietary interventions conducted in low-income countries showed an increase in birth weight (P = 0.04) and a reduction in the incidence of low birth weight (P , 0.01). With the use of an RMD, these results translated to an increase of 94 g (from a starting weight of 3086 g) and a reduction of 37% in odds (or a 0.63 OR), respectively. Trials in high-income countries showed increases in birth weight (P = 0.04) and lengths (P = 0.04) that corresponded to 49-g (from a starting weight of 3406 g) and 0.5cm (from a starting length of 51 cm) increases, respectively. The neonatal outcomes Apgar score (at 5 min of age and ,7 at 5 min of age) and large for gestational age were not sub-analyzed because they contained only 3 RCTs. There was no effect on any other neonatal outcome.


1 AA, arachidonic acid; ALA, a-linolenic acid; BGGI, borderline gestational glucose intolerance; C, control group; CARRDIP, Cardiovascular Disease Risk Reduction Diet in Pregnancy; C1, control group 1; EJPS, East Javan Pregnancy Study; GD, gestational diabetes; I1, intervention group 1; I2, intervention group 2; LA, linoleic acid; LGA, large for gestational age; LiP, Lifestyle in Pregnancy; NELLI, neuvonta, elintavat, liikunta (counseling, lifestyle, physical activity); PHFE-WIC, Public Health Foundation Enterprises Management Solutions Special Supplemental Nutrition Program for Women, Infants, and Children; RCT, randomized controlled trial; ref, reference; SES, socioeconomic status; SGA, small for gestational age; WIC, Women, Infants, and Children. 2 Unless otherwise indicated, values are means 6 SDs, means 6 (SEs), medians (IQRs), or {medians}. 3 Version 2.5; Research Centre of the Social Insurance Institution. 4 1995; ESHA Research. 5 Total n = 144 randomly assigned; participant numbers not reported per intervention group. 6 Version 3000; Danish Catering Centre A/S. 7 Purita Powdered milk manufactured by Ministry of Health, Santiago, Chile. 8 Vita-Nova Motherfood milk manufactured by Melkunie Holland, Wuerden, Netherlands. 9 6th Edition (1979); Editorial, Antarctica. 10 Participant numbers reported for singleton births and not the number of women (n = 1460). 11 Total n = 294 randomly assigned; participant numbers not reported per intervention group. 12 Total n = 471 randomly assigned; participant numbers not reported per intervention group. 13 NS once adjusted for prepregnant weight. 14 Numbers are for total participants active to delivery and analyzed n = 6335 recruited and n = 1051 registered in the study initially. 15 University of Minnesota. 16 Version 2.0 (1996); Wageningen University. 17 Once adjusted for gestational age plus pregnancy weight, smoking, and height of father, there was no significant difference. 18 Withdrawals not reported per group; therefore, numbers presented per group at the time of random assignment. 19 Version and manufacturer information not available for nutrient software. 20 Institute for Nutrition Research, University of Oslo. 21 Mosby.

1314

(56) (56) (56) (56) 6 28 14 10 10 4 3 3 5 3 10

(18, 23, 48, 55) (18, 21–23, 32, 48–52, 54, 55) (18, 32, 48, 49, 51, 52, 55) (18, 32, 48, 52, 55) (18, 21, 23, 49, 52–55) (32) (48) (48) (23, 31, 48–50)

4 12 7 5 8 1 1 1 5

1 1 1 1

(45, 47) (20, 33–45, 47) (37, 39, 40, 43, 46, 47) (33, 40, 46, 47) (40, 42) (33, 35, 37, 44) (37) (42, 44) (33, 37, 40, 45) (33, 35, 37) (35, 40–43)

2 15 6 4 2 4 1 2 4 3 5

1011 4717 2584 1419 2981 631 258 878 1360 515 3113

78 78 78 86

878 2983 1823 813 2496 22 578 598 2221

133 1656 683 528 475 631 150 300 652 515 892

Intervention

1005 4273 2611 1366 2467 593 260 820 1257 477 3012

78 78 77 85

846 2564 1833 797 2034 25 527 552 2126

159 1631 700 492 433 593 150 293 573 477 886

Control

603.88 6 125.53 3276.75 6 491.76 50.35 6 3.06 34.66 6 1.73 226 95 8.99 6 1.19 17 84 58 71

6 132.12 6 547.13 6 3.68 6 2.01 312 95 8.95 6 0.98 11 66 73 97 605.66 3221.58 50.15 34.44

3600 51.00 35.10 7.98

1033.63 4.44 3.08 1

6 6 6 6

758.59 3.11 2.22 1.17

6 6 6 6

3628 51.30 35.10 7.74

6 130.13 6 475.80 6 2.38 6 1.36 286 8.90 6 0.30 4 25 76

607.72 6 123.65 3168.95 6 436.44 50.35 6 2.21 34.61 6 1.09 199 9 6 0.30 6 28 48 610.18 3061.32 50.14 34.45

6 140.25 6 525.50 6 5.80 6 2.58 26 95 9.50 6 0.50 7 41 73 21

581.57 3455.41 50.10 34.33

Control2

6 135.06 6 510.68 6 4.59 6 2.34 27 95 9.70 6 0.50 11 56 58 23

578.57 3454.38 50.23 34.67

Intervention2

(20.12, 0.26) (0.14, 0.40) (20.01, 0.15) (20.17, 0.38) (20.37, 20.06) (20.25, 0.92) (20.53, 0.87) (20.29, 0.32) (20.51, 0.03) (20.28, (20.24, (20.31, (20.53,

(20.12, 0.20) (0.06, 0.32) (0.01, 0.15) (20.19, 0.19) (20.32, 20.05) (20.27, 0.19) (20.31, 0.62) (20.21, 0.62) (20.06, 0.30) (20.60, 0.17) (20.39, 0.08)

0.07 0.27 0.07 0.11 20.22 0.33 0.17 0.02 20.24 0.03 0.08 0.00 20.23 0.04 0.19 0.08 0.00 20.19 20.04 0.15 0.20 0.12 20.21 20.16

0.35) 0.39) 0.31) 0.06)

(20.68, 0.41) (20.11, 0.36) (20.08, 0.25) (20.58, 0.24) (20.33, 0.28) (20.27, 0.19) (0.17, 0.63) (20.30, 0.74) (20.05, 0.41) (20.60, 0.17) (20.28, 0.43)

20.14 0.13 0.09 20.17 20.02 20.04 0.40 0.22 0.18 20.21 0.07

SMD (95% CI)3


12.85 62.37 0.07 0.11 0.76 0.94 0.07 1.42 1.20 0.75 0.76

28.00 0.30 0.00 20.25

28.38 125.30 0.25 0.32 0.73 0.10 1.36 1.03 0.67

220.58 15.13 0.31 20.23 0.96 0.94 0.20 1.44 1.30 0.75 1.13

RMD4

0.66 ,0.01 0.03 1.0 ,0.01 0.75 0.52 0.34 0.20 0.28 0.19

0.85 0.63 1.00 0.12

0.45 ,0.01 0.08 0.44 ,0.01 0.26 0.63 0.90 0.09

0.62 0.30 0.31 0.43 0.88 0.75 ,0.01 0.41 0.13 0.28 0.69

P5

58 87 25 81 19 32 82 0 0 70 27

NA NA NA NA

65 74 13 79 24 NA NA NA 33

71 91 52 88 0 32 NA 0 0 70 0

I2

The main measure of effect was the SMD. The SMD was determined by taking the difference in the mean of an outcome between the intervention and control groups in one publication and dividing it by the pooled SD for the outcome across the whole trial. The Apgar score was measured by using a scale from 1 to 10. NA, not applicable; ref, reference(s); RMD, raw mean difference; SMD, standardized mean difference. 2 All values are numbers or means 6 SDs. 3 The trial by Metcoff et al (53) was not included in the meta-analyzed outcomes of birth weight, length, and head circumference because the authors did not present results. 4 The secondary measure of effect was RMD, which is the unadjusted mean difference in the outcome between the intervention and control group. RMD is presented in the common units for each outcome. Categorical outcomes are reported as ORs. 5 P values apply to SMDs only.

1

Dietary counseling Placental weight (g) Birth weight (g) Length (cm) Head circumference (cm) Low birth weight Macrosomia Apgar score (at 5 min of age) Apgar score (,7 at 5 min of age) Small for gestational age Large for gestational age Perinatal mortality Food and food products Placental weight (g) Birth weight (g) Length (cm) Head circumference (cm) Low birth weight Apgar score (at 5 min of age) Apgar score (,7 at 5 min of age) Small for gestational age Perinatal mortality Combination Birth weight (g) Length (cm) Head circumference (cm) Apgar score (at 5 min of age) All dietary interventions Placental weight (g) Birth weight (g) Length (cm) Head circumference (cm) Low birth weight Macrosomia Apgar score (at 5 min of age) Apgar score (,7 at 5 min of age) Small for gestational age Large for gestational age Perinatal mortality

No. of trials (ref)

No. of participants

TABLE 3 Effect of dietary interventions during pregnancy on neonatal and infant outcomes1


1315

1316

GRESHAM ET AL

DISCUSSION

Summary of main findings This review showed dietary intervention to be an effective strategy for increasing the size of the infant at birth. In particular, food or fortified food products increased birth weight (by w125 g) and reduced the incidence of low birth weight (by an w27% decrease in odds). After all dietary interventions were combined,

both the increase in birth weight and reduction in low birth weight remained significant (62 g and 24% decreased odds, respectively), and length also increased (0.07 cm). From subgroup analyses, it appeared that the largest gains in birth size as a result of dietary intervention were made in underweight and nutritionally at-risk populations in both high- and low-income countries and dietary interventions that focused on macronutrients. To our knowledge, this is the largest systematic review of

FIGURE 3. The SMD for length between dietary interventions in pregnancy. The overall effect size was estimated by the SMD. Black dots represent point estimates of studies, square sizes represent weights of studies in the meta-analysis, and horizontal lines represent 95% CIs. The vertical solid line represents the line of no effect. The vertical dashed line represents the overall measure of effect. Diamonds represent overall pooled estimates of effects of dietary interventions on length. SMD, standardized mean difference.


FIGURE 2. The SMD for birth weight between dietary interventions in pregnancy. The overall effect size was estimated by the SMD. Black dots represent point estimates of studies, square sizes represent weights of studies in the meta-analysis, and horizontal lines represent 95% CIs. The vertical solid line represents the line of no effect. The vertical dashed line represents the overall measure of effect. Diamonds represent overall pooled estimates of effects of dietary interventions on birth weight. SMD, standardized mean difference.


1317

dietary interventions during pregnancy on neonatal and infant outcomes. Interpretation Macronutrient or balanced protein-energy food supplementation is considered one of the most promising dietary interventions in the prevention of adverse perinatal outcomes (58, 59). Results from our previous work on the effect of dietary intervention on pregnancy outcomes (16) as well as our current review reinforce this view. These reviews showed that food or fortified food products and macronutrient dietary intervention trials increase the size of the infant at birth and reduce the incidence of preterm delivery, with no effect on macrosomia or a prolonged length of gestation. There were 15 trials that reported on preterm delivery, and 28 trials that reported on birth weight. The combination of both preterm delivery and birth-weight outcomes were reported in 14 trials. Of these trials, 43% (n = 6) of them showed a reduction in preterm delivery and a corresponding increase in birth weight, which would be expected with an increase in the length of gestation. Despite decreased odds of a preterm delivery as a categorical variable, our pregnancy review also showed no increase in the length of gestation as a continuous variable (16). There were 20 trials that reported on the length of gestation, but of these studies, only 10 trials reported both preterm delivery and length of gestation. There was relatively high heterogeneity in trials that reported on the length of gestation (I2 = 61%) than all trials that included preterm delivery (I2 = 15%) (16). In this review, less than one-half of individual food or fortified food product trials (5 of 12) showed significant effects on birth weight. The combination of studies in our meta-analysis increased the statistical power and improved point estimates. All trials of food and fortified food products targeted macronutrients

(primarily energy, protein, and fat) in their interventions with the exception of Bech et al (48) who evaluated the impact of reducing caffeine intake during pregnancy. Dietary interventions that were effective in increasing the size of the infant included a milk-based fortified product (21), high-energy and -protein supplement biscuits (49), and a DHA cereal-based bar (32). The provision of a food or fortified food product in addition to the usual diet during pregnancy can increase overall energy intake and subsequently increase the size of the infant at birth. The effect of dietary counseling during pregnancy was not consistent across trials. There were no significant results of dietary counseling on birth weight because one-half of included trials (8 trials) showed an increase in birth weight with dietary counseling, and the other one-half of trials (7 trials) showed a decrease in birth weight. Dietary counseling interventions shown to significantly increase birth weight and showed the strongest effects in the meta-analysis included trying to increase dairy intake in adolescent pregnancies (20), providing recommendations on nutritional needs during pregnancies of low-income women (38), and dietary advice to achieve national recommendations (35). In comparison, dietary counseling interventions shown to decrease birth weight included guidelines for specific energy requirements (36, 37, 44, 47), recommendations to modify fat intake (33, 39), and providing counseling to reduce sodium intake (45). Of trials that provided counseling that focused on achieving national dietary recommendations (6 trials), one trial showed an increase in birth weight (35). An increase in birth weight may be from a combination of a woman beginning her pregnancy obese and exceeding upper thresholds for gestational weight gain (60, 61). High birth weight (births .4000 g) has been associated with increased risk of obesity later in life (62); however, Risnes et al (63) have shown an overall reduction in adult mortality with a higher birth weight (6% reduction per kilogram increase in


FIGURE 4. The SMD for low birth weight between dietary interventions in pregnancy. The overall effect size was estimated by the SMD. Black dots represent point estimates of studies, square sizes represent weights of studies in the meta-analysis, and horizontal lines represent 95% CIs. The vertical solid line represents the line of no effect. The vertical dashed line represents the overall measure of effect. Diamonds represent overall pooled estimates of effects of dietary interventions on low birth weight. SMD, standardized mean difference.

1318

GRESHAM ET AL TABLE 4 Subgroup analyses for pregnancy outcomes in evaluation of dietary intervention during pregnancy1 RMD3

P4

Birth weight (g) Nutrient of interest Macronutrients Micronutrients BMI Underweight/nutritional risk Overweight and obese All weight categories5 Country income Low High Length (cm) Nutrient of interest Macronutrients Micronutrients BMI Underweight/nutritional risk Overweight and obese All weight categories Country income Low High Head circumference (g) Nutrient of interest Macronutrients Micronutrients BMI Underweight/nutritional risk Overweight and obese All weight categories Country income Low High Placental weight (g) Nutrient of interest Macronutrients Micronutrients BMI Underweight/nutritional risk Overweight and obese All weight categories Country income (high) Low birth weight Nutrient of interest Macronutrients Micronutrient6 BMI Underweight/nutritional risk All weight categories Country income Low High Macrosomia Nutrient of interest (macronutrients) BMI Overweight and obese All weight categories Country income (high) Small for gestational age Nutrient of interest Macronutrients Micronutrients

24 4

0.23 (0.07–0.38) 0.02 (20.07–0.10)

72.10 210.11

,0.01 0.73

8 4 16

0.26 (0.09–0.41) 0.31 (20.54–1.17) 0.05 (20.07–0.17)

100.65 40.24 42.02

,0.01 0.47 0.40

7 21

0.24 (0.02–0.47) 0.17 (0.01–0.33)

94.09 48.80

0.04 0.04

11 3

0.07 (0.01–0.14) 0.21 (20.09–0.50)

0.18 0.85

0.03 0.17

4 1 9

0.06 (20.01–0.14) 20.40 (20.94–0.14) 0.14 (0.01–0.26)

0.16 21.00 0.54

0.11 — 0.03

4 10

0.06 (20.02–0.14) 0.12 (0.00–0.24)

0.15 0.46

0.17 0.04

8 2

0.10 (20.09–0.29) 20.30 (20.53 to 20.06)

0.22 21.00

0.30 —

2 1 7

0.22 (20.08–0.51) 20.62 (21.15 to 20.09) 20.02 (20.22–0.18)

0.52 21.00 0.10

— — 0.85

1 9

0.05 (20.22–0.32) 20.01 (20.22–0.21)

0.70 0.10

— 0.96

4 2

0.08 (20.22–0.38) 20.03 (20.16–0.11)

14.03 10.16

0.60 —

1 1 4 6

0.05 20.47 0.09 0.04

4.40 270.00 31.88 12.85

— — 0.38 0.66

9 1

20.19 (20.34 to 20.04) 20.10 (20.52–0.31)

0.76 0.85

0.01 —

7 3

20.19 (20.37 to 20.02) 20.13 (20.42–0.17)

0.76 0.82

0.03 0.40

4 6

20.31 (20.54 to 20.08) 20.08 (20.25–0.08)

0.63 0.88

,0.01 0.32

4

20.04 (20.27–0.19)

0.94

0.75

2 2 4

0.15 (20.22–0.51) 20.17 (20.41–0.07) 20.04 (20.27–0.19)

1.22 0.79 0.94

— — 0.75

3 2

0.15 (20.12–0.42) 0.09 (20.16–0.34)

1.24 1.16

0.27 —

(20.15–0.25) (21.01–0.07) (20.12–0.30) (20.12–0.20)

(Continued)


SMD (95% CI)2

No. of trials

1319

DIETARY INTERVENTIONS DURING PREGNANCY TABLE 4 (Continued ) RMD3

BMI Underweight/nutritional risk All weight categories Country income (high) Perinatal mortality Nutrient of interest Macronutrients Micronutrients BMI Underweight/nutritional risk Overweight and obese All weight categories Country income Low High

P4

1 4 5

0.03 (20.36–0.42) 0.15 (20.06–0.36) 0.12 (20.06–0.30)

1.04 1.25 1.20

— 0.17 0.20

7 3

20.18 (20.44–0.09) 20.08 (20.75–0.58)

0.74 0.88

0.19 0.81

4 1 5

20.18 (20.50–0.14) 0.40 (20.93–1.73) 20.12 (20.59–0.36)

0.74 2.04 0.82

0.28 — 0.63

2 8

20.45 (20.72 to 20.18) 0.06 (20.18–0.30)

0.47 1.11

— 0.63

1 The main measure of effect was the SMD. The SMD was determined by taking the difference in the mean of an outcome between the intervention and control groups in one publication and dividing it by the pooled SD for the outcome across the whole trial. The meta-analysis focused on outcomes with $3 trials that contributed data to pooled results. There were no studies for subgroup analyses for BMI (overweight and obese) on placental weight, macrosomia, and small for gestational age. There were no studies of low-income countries on low birth weight and small for gestational age. RMD, raw mean difference; SMD, standardized mean difference. 2 The trial by Metcoff et al (53) was not included in meta-analyzed outcomes of birth weight, length, and head circumference because the authors did not present results. 3 The secondary measure of effect was the RMD, which was the unadjusted mean difference in the outcome between intervention and control groups. RMDs are presented in common units for each outcome. Categorical outcomes are reported as ORs. 4 P values apply to SMDs only. 5 Included trials that did not restrict BMI within the target population. 6 Micronutrient of interest was sodium.

birth weight). Overall, the appropriateness of implementing dietary strategies to increase birth weight must be considered in the context of the target population. Furthermore, birth weight is only a surrogate marker for body composition, which has a much stronger biological basis for longer-term disease propensity. Specifically, increases in birth weight that increase visceral fat disproportionately to lean muscle mass are likely to be more harmful than protective (64–67). In an attempt to provide a more clinically relevant measure of the effect size for birth weight (because the SMD cannot be used to interpret absolute or relative differences in clinically meaningful outcomes) we have presented raw data for one trial. Smuts et al (55) showed an increase in birth weight of 103 g from an average birth weight of 3158 g. The SMD in this trial (SMD: 0.19; 95% CI: 0.06, 0.32) was similar to the SMD for the pooled effect for this outcome in this review. However the absolute effect may vary depending on the population’s characteristics (eg, the average birth weight and SD for the population). Implications for practice and research Infant size at birth (small for gestational age, low birth weight, and intrauterine growth restriction) has been associated with increased risk and development of both short (perinatal mortality and morbidity) (68) and long-term (diabetes and cardiovascular disease) (32, 68) health problems compared with infants born appropriate for gestational age (2500–4000 g) (69–72). In our review, dietary interventions during pregnancy were associated with increased birth weight and length and a reduced incidence

of low birth weight, with no strong evidence for any other neonatal or infant outcome. This review advances our understanding of the role of nutrition for a healthy birth outcome. The provision of nutrition education as well as food or fortified food products to pregnant women, particularly those who are underweight, at nutritional risk, or come from a low-income country, is likely to increase the size of the infant at birth with important health and financial ramifications. Limitations Despite the broad scope of this review, very few dietary intervention trials contributed data to each neonatal and infant outcome with the exception of birth weight (n = 28). For this reason, some of the outcomes were not reported, and other outcomes were underpowered. There were very few trials on dietary interventions, particularly a combination (both counseling and food) intervention, and some neonatal and infant outcomes. Trials should measure and report on a range of neonatal and infant outcomes so that effects of diet on neonatal and infant outcomes can be determined even if only by a meta-analysis. The quality across included trials was mostly positive (61%; 39% neutral). However, many trials did not report on how effective their dietary interventions were on achieving a dietary change. There was a moderate to high heterogeneity for the neonatal outcomes placental weight, birth weight, head circumference, Apgar score (at 5 min of age), and large for gestational age with no evidence of bias for any outcome. Subgroup


SMD (95% CI)2

No. of trials

1320

GRESHAM ET AL

We thank Debbie Booth for her assistance with database searches, Samantha Diamond and Jun Lai for their assistance with the assessment of the methodologic quality and data extraction, and Peta Forder for statistical advice. The authors’ responsibilities were as follows—EG and AJH: conducted the research; EG and AB: analyzed the data; EG: wrote the manuscript and had primary responsibility for the final content of the manuscript; and all authors: designed the research and read and approved the final manuscript. None of the authors had a conflict of interest.

15.

16.

17.

18.

19. 20.

21.

22.

23. 24. 25. 26.

27.

REFERENCES 1. Imdad A, Bhutta ZA. Maternal nutrition and birth outcomes: effect of balanced protein-energy supplementation. Paediatr Perinat Epidemiol 2012;26:178–90. 2. Abu-Saad K, Fraser D. Maternal nutrition and birth outcomes. Epidemiol Rev 2010;32:5–25. 3. Abrams B, Altman SL, Pickett KE. Prepregnancy weight gain: still controversial. Am J Clin Nutr 2000;71(5 suppl):1233S–41S. 4. Kanade AN, Rao S, Kelkar RS, Gupte S. Maternal nutrition and birth size among urban affluent and rural women in India. J Am Coll Nutr 2008;27:137–45. 5. Victora CG, Adair L, Fall C, Hallal PC, Martorell R, Richter L. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008;371:340–57. 6. Godfrey K, Robinson S. Maternal nutrition, placental growth and fetal programming. Proc Nutr Soc 1998;57:105–11. 7. Moore VM, Davies M, Willson K, Worsley A, Robinson J. Dietary composition of pregnant women is related to size of the baby at birth. J Nutr 2004;134:1820–6. 8. Northstone K, Emmett P, Rogers I. Dietary patterns in pregnancy and associations with socio-demographic and lifestyle factors. Eur J Clin Nutr 2008;62:471–9. 9. Pitkin RM. Folate and neural tube defects. Am J Clin Nutr 2007;85: 285S–8S. 10. Kant AK, Schatzkin A, Graubard BI, Schairer C. A prospective study of diet quality and mortality in women. JAMA 2000;283:2109–15. 11. 1,000 days. Why 1,000 days. Available from: http://www.thousanddays.org/about/ (cited 10 November 2013). 12. Scaling Up Nutrition. SUN movement global gathering. Available from: http://scalingupnutrition.org/ (cited 10 November 2013). 13. Higgins JPT, Green S, eds. Cochrane handbook for systematic reviews of interventions version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from: http://handbook.cochrane.org/ (cited 28 November 2012). 14. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA

28. 29. 30. 31.

32.

33.

34.

35.

36.

37.

38.

statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6:e1000100. Moher D, Liberati A, Tetzlaff J, Altman DG, The PG. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. Gresham E, Bisquera A, Byles JE, Hure AJ. The effects of dietary interventions on pregnancy outcomes: a systematic review and metaanalysis. Matern Child Nutr (Epub ahead of print 22 July 2014). Academy of Nutrition and Dietetics. Evidence analysis manual: steps in the academy evidence analysis process. Chicago, IL: Research and Strategic Business Development, 2012. Smuts CM, Borod E, Peeples JM, Carlson SE. High-DHA eggs: feasibility as a means to enhance circulating DHA in mother and infant. Lipids 2003;38:407–14. Lipsey MW, Wilson DB. Practical meta-analyisis. Thousand Oaks, CA: Sage Publications Inc, 2001. Chan GM, McElligott K, McNaught T, Gill G. Effects of dietary calcium intervention on adolescent mothers and newborns: a randomized controlled trial. Obstet Gynecol 2006;108:565–71. Mardones-Santander F, Rosso P, Stekel A, Ahumada E, Llaguno S, Pizarro F, Salinas J, Vial I, Walter T. Effect of a milk-based food supplement on maternal nutritional status and fetal growth in underweight Chilean women. Am J Clin Nutr 1988;47:413–9. Ross SM, Nel E, Naeye RL. Differing effects of low and high bulk maternal dietary supplements during pregnancy. Early Hum Dev 1985; 10:295–302. Rush D, Stein Z, Susser M. A randomized controlled trial of prenatal nutritional supplementation in New York City. Pediatrics 1980;65:683–97. Chinn S. A simple method for converting an odds ratio to effect size for use in meta-analysis. Stat Med 2000;19:3127–31. Hasselblad V, Hedges LV. Meta-analysis of screening and diagnostic tests. Psychol Bull 1995;117:167–78. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005;5:13. Firth D. Bias reduction of maximum likelihood estimates. Biometrika 1993;80:27–38. Heinze G, Schemper M. A solution to the problem of separation in logistic regression. Stat Med 2002;21:2409–19. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. StataCorp. Stata release 12. College Station, TX: Stata Press, 2011. Mora JO, Clement J, Christiansen N, Suescun J, Wagner M, Herrera MG. Nutritonal supplementation and the outcome of pregnancy. III. Perinatal and neonatal mortality. Nutr Rep Int 1978;18:167–75. Courville AB, Harel O, Lammi-Keefe C. Consumption of a DHAcontaining functional food during pregnancy is associated with lower infant ponderal index and cord plasma insulin concentration. Br J Nutr 2011;106:208–12. Luoto R, Kinnunen T, Aittasalo M, Kolu P, Raitanen J, Ojala K, Mansikkama¨ki K, Lamberg S, Vasankari T, Komulainen T, et al. Primary prevention of gestational diabetes mellitus and large-for-gestationalage newborns by lifestyle counseling: a cluster-randomized controlled trial. PLoS Med 2011;8:e1001036. Quinlivan JA, Lam LT, Fisher J. A randomised trial of a four-step multidisciplinary approach to the antenatal care of obese pregnant women. Aust N Z J Obstet Gynaecol 2011;51:141–6. Vinter CA, Jensen DM, Ovesen P, Beck-Nielsen H, Jørgensen JS. The LiP (Lifestyle in Pregnancy) study: a randomized controlled trial of lifestyle intervention in 360 obese pregnant women. Diabetes Care 2011;34:2502–7. Asbee SM, Jenkins TR, Butler JR, White J, Elliot M, Rutledge A. Preventing excessive weight gain during pregnancy through dietary and lifestyle counseling: a randomized controlled trial. Obstet Gynecol 2009;113:305–12. Bonomo M, Corica D, Mion E, Goncalves D, Motta G, Merati R, Ragusa A, Morabito A. Evaluating the therapeutic approach in pregnancies complicated by borderline glucose intolerance: a randomized clinical trial. Diabet Med 2005;22:1536–41. Briley C, Flanagan NL, Lewis N. In-home prenatal nutrition intervention increased dietary iron intakes and reduced low birthweight in low-income African-American women. J Am Diet Assoc 2002; 102:984–7.


analyses were undertaken, which did not explain the heterogeneity, which may have been due to the duration and varying intensity of included trials. A small number of trials (n = 4) had their effects taken from the intervention group exhibiting significant results with the nonsignificant group removed from the meta-analysis. This method would have biased the results in favor of a greater effect size than if the nonsignificant intervention was included. In conclusion, there is evidence that dietary intervention during pregnancy can increase the size of the infant at birth. Interventions that provide food and fortified food products or target pregnant women who are underweight, nutritionally at risk, or from a low-income country are the most-promising strategies to increase the size of the infant at birth. However, size is just a surrogate marker for other health and economic outcomes. In addition, large, high-quality RCTs that investigate combination (counseling plus food) dietary intervention and micronutrient provision from food are needed. Future trials spanning preconception, the duration of pregnancy, and even between pregnancies are needed to advance our understanding of optimal maternal nutrition for maternal-child health.


55. Smuts CM, Huang M, Mundy D, Plasse T, Major S, Carlson SE. A randomized trial of docosahexaenoic acid supplementation during the third trimester of pregnancy. Obstet Gynecol 2003;101:469–79. 56. Aaltonen J, Ojala T, Laitinen K, Piirainen TJ, Poussa TA, Isolauri E. Evidence of infant blood pressure programming by maternal nutrition during pregnancy: a prospective randomized controlled intervention study. J Pediatr 2008;152:79–84. 57. Rush D. Maternal nutrition and perinatal survival. J Health Popul Nutr 2001;19:S217–64. 58. Kramer MS, Kakuma R. Energy and protein intake in pregnancy. Cochrane Database Syst Rev 2003;4:CD000032. 59. de Onis M, Villar J, Gulmezoglu M. Nutritional interventions to prevent intrauterine growth retardation: evidence from randomized controlled trials. Eur J Clin Nutr 1998;52(suppl 1):S83–93. 60. Ludwig DS, Currie J. The association between pregnancy weight gain and birthweight: a within-family comparison. Lancet 2010;376:984–90. 61. Hunt KJ, Alanis M, Johnson E, Mayorga M, Korte J. Maternal prepregnancy weight and gestational weight gain and their association with birthweight with a focus on racial differences. Matern Child Health J 2013;17:85–94. 62. Yu ZB, Han SP, Zhu GZ, Zhu C, Wang XJ, Cao XG, Guo XR. Birth weight and subsequent risk of obesity: a systematic review and metaanalysis. Obesity Rev 2011;12:525-42. 63. Risnes KR, Vatten LJ, Baker JL, Jameson K, Sovio U, Kajantie E, Osler M, Morley R, Jokela M, Painter RC, et al. Birthweight and mortality in adulthood: a systematic review and meta-analysis. Int J Epidemiol 2011;40:647–61. 64. Yliharsila H, Kajantie E, Osmond C, Forsen T, Barker DJ, Eriksson JG. Birth size, adult body composition and muscle strength in later life. Int J Obes (Lond) 2007;31:1392–9. 65. Kuzawa CW, Hallal PC, Adair L, Bhargava SK, Fall CH, Lee N, Norris SA, Osmond C, Ramirez-Zea M, Sachdev HS, et al. Birth weight, postnatal weight gain, and adult body composition in five low and middle income countries. Am J Hum Biol 2012;24:5–13. 66. Demerath EW, Reed D, Choh AC, Soloway L, Lee M, Czerwinski SA, Chumlea WC, Siervogel RM, Towne B. Rapid postnatal weight gain and visceral adiposity in adulthood: the Fels Longitudinal Study. Obesity (Silver Spring) 2009;17:2060–6. 67. Bann D, Wills A, Cooper R, Hardy R, Aihie Sayer A, Adams J, Kuh D. Birth weight and growth from infancy to late adolescence in relation to fat and lean mass in early old age: findings from the MRC National Survey of Health and Development. Int J Obes (Lond) 2014;38:69–75. 68. Goldenberg RL, Culhane JF. Low birth weight in the United States. Am J Clin Nutr 2007;85:584S–90S. 69. Haggarty P. Placental regulation of fatty acid delivery and its effect on fetal growth – a review. Placenta 2002;23:S28–38. 70. Oken E, Gillman M. Fetal origins of obesity. Obes Res 2003;11:496–506. 71. Barker DJP, Eriksson JG, Forsen T, Osmond C. Fetal origins of adult disease: strength of effects and biological basis. Int J Epidemiol 2002; 31:1235–9. 72. Boney CM, Verma A, Tucker R, Vohr BR. Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 2005;115:e290–6.


39. Fard NM, Mehrabian F, Sarraf-Zadegan N, Sajadi F. Fat-modified diets during pregnancy and lactation and serum lipids after birth. Indian J Pediatr 2004;71:683–7. 40. Kafatos AG, Vlachonikolis I, Codrington C. Nutrition during pregnancy: the effects of an educational intervention program in Greece. Am J Clin Nutr 1989;50:970–9. 41. Khoury J, Henriksen T, Christophersen B, Tonstad S. Effect of a cholesterol-lowering diet on maternal, cord, and neonatal lipids, and pregnancy outcome: a randomized clinical trial. Am J Obstet Gynecol 2005;193:1292–301. 42. Knuist M, Bonsel GJ, Zondervan HA, Treffers PE. Low sodium diet and pregnancy-induced hypertension: a multi-centre randomised controlled trial. Br J Obstet Gynaecol 1998;105:430–4. 43. O’Connor MJ, Whaley SE. Brief intervention for alcohol use by pregnant women. Am J Public Health 2007;97:252–8. 44. Thornton YS, Smarkola C, Kopacz SM, Ishoof SB. Perinatal outcomes in nutritionally monitored obese pregnant women: a randomized clinical trial. J Natl Med Assoc 2009;101:569–77. 45. Van Buul BJA, Steegers EAP, Van Der Maten GD, Delemarre FMC, Jongsma HW, Oosterbaan HP, De Jong PA. Dietary sodium restriction does not prevent gestational hypertension: a Dutch two-center randomized trial. Hypertens Pregnancy 1997;16:335–46. 46. van der Maten GD, van Raaij JM, Visman L, van der Heijden LJ, Oosterbaan HP, de Boer R, Eskes TK, Hautvast JG. Low-sodium diet in pregnancy: effects on blood pressure and maternal nutritional status. Br J Nutr 1997;77:703–20. 47. Wolff S, Legarth J, Vangsgaard K, Toubro S, Astrup A. A randomized trial of the effects of dietary counseling on gestational weight gain and glucose metabolism in obese pregnant women. Int J Obes (Lond) 2008; 32:495–501. 48. Bech BH, Obel C, Henriksen TB, Olsen J. Effect of reducing caffeine intake on birth weight and length of gestation: randomised controlled trial. BMJ 2007;334:409–12. 49. Ceesay SM, Prentice AM, Cole TJ, Foord F, Weaver LT, Poskitt EM, Whitehead RG. Effects on birth weight and perinatal mortality of maternal dietary supplements in rural Gambia: 5 year randomised controlled trial. BMJ 1997;315:786–90. 50. de Groot RHM, Hornstra G, van Houwelingen AC, Roumen F. Effect of alpha-linolenic acid supplementation during pregnancy on maternal and neonatal polyunsaturated fatty acid status and pregnancy outcome. Am J Clin Nutr 2004;79:251–60. 51. Kusin JA, Kardjati S, Houtkooper JM, Renqvist UH. Energy supplementation during pregnancy and postnatal growth. Lancet 1992;340: 623–6. 52. McDonald EC, Pollitt E, Mueller W. The bacon chow study: maternal nutritional supplementation and birth weight of offspring. Am J Clin Nutr 1981;34:2133–44. 53. Metcoff J, Costiloe P, Crosby WM, Dutta S, Sandstead HH, Milne D, Bodwell CE, Majors SH. Effect of food supplementation (WIC) during pregnancy on birth weight. Am J Clin Nutr 1985;41:933–47. 54. Mora JO, de Paredes B, Wagner M, de Navarro L, Suescun J, Christiansen N, Herrera MG. Nutritional supplementation and the outcome of pregnancy. I. Birth weight. Am J Clin Nutr 1979;32:455–62.

1321

Effects of dietary interventions on pregnancy outcomes: a systematic review and meta-analysis.

Interventions for postnatal depression assessing the mother-infant relationship and child developmental outcomes: a systematic review.

Impact of single- vs double-layer closure on adverse outcomes and uterine scar defect: a systematic review and metaanalysis.

Diagnostic markers for hyperemesis gravidarum: a systematic review and metaanalysis.

The effects of morbid obesity on maternal and neonatal health outcomes: a systematic review and meta-analyses.

The impact of interventions on appointment and clinical outcomes for individuals with diabetes: a systematic review.

A systematic review of interventions on body image and disordered eating outcomes among women in midlife.

Effectiveness of family and caregiver interventions on patient outcomes in adults with cancer: a systematic review.

The effects of motivating interventions on rehabilitation outcomes in children and youth with acquired brain injuries: a systematic review.

Effects of Different Dietary Interventions on Blood Pressure: Systematic Review and Meta-Analysis of Randomized Controlled Trials.

The Impact of Dietary Interventions on the Symptoms of Inflammatory Bowel Disease: A Systematic Review.

Salutogenically focused outcomes in systematic reviews of intrapartum interventions: a systematic review of systematic reviews.

Simulation-based neonatal and infant resuscitation teaching: a systematic review of randomized controlled trials.

[Potential dietary interventions for prevention and treatment of infant allergy].

Dietary interventions and quality of life: a systematic review of the literature.

Interventions on bullying and cyberbullying in schools: a systematic review.

Effects of Exercise Interventions and Physical Activity Behavior on Cancer Related Cognitive Impairments: A Systematic Review.

Effects of relaxation interventions on depression and anxiety among older adults: a systematic review.

The effectiveness of workplace dietary interventions: protocol for a systematic review and meta-analysis.

A systematic review of weight loss, physical activity and dietary interventions involving African American men.

Dietary educational interventions for management of hyperphosphatemia in hemodialysis patients: a systematic review and meta-analysis.

Systematic Review of Physical Activity Outcomes of Rural Lifestyle Interventions.

Are dietary interventions effective at increasing fruit and vegetable consumption among overweight children? A systematic review.

Effects of dietary interventions on incidence and progression of CKD.