|Home | About | Journals | Submit | Contact Us | Français|
Maternal nutritional status during pregnancy will affect the outcomes for the mother and the baby. Many analyses of the relationship between diet and outcome are often based on a single or a few food items or nutrients. However, foods are not consumed in isolation and dietary patterns can be used to assess the whole diet consumed. The use of dietary pattern analysis to understand nutritional intake and pregnancy outcome is becoming more and more popular. Many published studies have showed the association between maternal dietary patterns and pregnancy outcome. This review examined articles about the relationship between maternal dietary patterns and pregnancy outcome. As a modifiable factor, dietary patterns may be more applicable to clinical and pregnant health interventions.
Nutritional status during pregnancy will affect the outcomes for the mother and the baby, but the evidence for a beneficial effect of specific nutritional supplements for women of reproductive age is inconclusive [1,2,3]. Thus, identifying what foods should be eaten, in what quantities, and even how often these foods should be eaten, remain important questions—the answers will enable accurate nutritional assessment and facilitate appropriate counseling in pregnancy.
Many analyses of the relationship between diet and an outcome are based on a single or a few food items or nutrients. Examples such as a high consumption of saturated fatty acids [4,5], total carbohydrates and soft drinks [6,7,8] and a low intake of polyunsaturated fatty acids [9,10] have been associated with increased risk of gestational diabetes mellitus (GDM). The calcium supplementation has been found to decrease preeclampsia  and result in fewer preterm births . However, foods are not consumed in isolation and people eat foods containing a mix of nutrients and nonnutrients. Moreover, the individual nutrient approach does not take biological complexity resulting from interactions between nutrients into consideration. For this reason, methods that identify patterns of food intake instead of looking at single food have become popular and are being used to examine potential disease risk prediction .
Dietary patterns study the whole diet and provide a simple but helpful, comprehensive and complimentary approach way to deliver meaningful results to the respective population . Dietary patterns have potential to be used as a valid tool in assessing the relationship between diet and pregnancy outcomes .
The purpose of this review is to examine articles about maternal dietary patterns before or during their pregnancies and pregnancy outcome. For this purpose, a PubMed Central search was conducted up to November 2015. The terms used to search abstracts were (dietary pattern or diet or dietary) AND (pregnancy or pregnant women or infant or offspring). This review included studies of dietary patterns and pregnancy outcome meeting the following criteria: studies based on dietary patterns rather than any single food and studies relating diet during pregnancy with health outcomes in the mother and the infant. The PubMed Central search strategy yielded a total amount of 2927 articles. After applying the inclusion criteria by screening titles and abstracts, 54 articles were selected.
The information obtained from food frequency questionnaires (FFQs) can be used to assess dietary patterns. Use of the FFQs has shown the validity of identifying maternal dietary patterns and of studying the relationship between dietary patterns . FFQs have become the most common measure of diet to obtain dietary patterns.
Two different approaches for observing overall dietary exposure were evident in the published literature. One approach is based on the popular hypotheses and guidance about the role of nutrients in disease prevention, and the diet is evaluated in accordance with this guidance (Table 1). In this approach, the diet assessed for certain foods or nutrient characteristics, and then the score resulting from the particular dietary exposure is the variable.
The alternative approach is data driven, using statistical analyses that seek to minimize bias (Table 2). In epidemiologic studies, statistical analyses play a central role in nutritional epidemiology and are used to promote objectivity in the conclusions drawn . Statistical methods like factor analysis or cluster analysis are common a posteriori approaches to the identification of dietary patterns .
Factor analysis can be considered a pattern detection method that reduces the number of dietary variables by finding correlated dietary variables. Most published studies that have used factor analysis to describe dietary patterns have used principal components analysis (PCA). The PCA is used to reduce a large amount of detailed information into a smaller set of interpretable factors that have the characteristic of explaining the largest amount of variability in the specific patterns .
Cluster analysis is a subject-oriented method with the objective of joining together variables placed into different nonoverlapping groups on the basis of some shared characteristics. Members of a cluster tend to be similar but distinct members of another cluster .
Another analytical approach in reflecting complex relationships between diet and disease is latent class analysis (LCA). LCA is useful to study unobserved heterogeneity characterized by several unidentified groups that behave differently. Thus group membership is an unobserved random variable and individuals have a predicted probability for belonging to each class, which reflects the uncertainty of class membership . However, it has rarely been used due to its complexity, making it difficult to interpret and apply the results.
In addition, there is clear evidence of social confounders associated with the dietary patterns. Social factors, such as educational levels, financial difficulties and age, need to be accounted for in studies using dietary patterns .
Infertility is defined as the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse . Its prevalence rates range from 6.9% to 9.3% in developing nations and from 3.5% to 16.7% in developed nations .
From the Nurses’ Health Study II a prospective cohort study, a diet score was calculated based on variables previously found to predict ovulatory disorder infertility. This ‘fertility diet’ was found to have favorable effects on fertility and was associated with a 69% lower risk of ovulatory disorder infertility (95% CI 29%–86%) . In a Dutch cohort of couples seeking to conceive by means of vitro fertilization or intracytoplasmic sperm injection, researchers found that a preconception Mediterranean diet contributed to successful pregnancy with a 40% increased probability (OR 1.4, 95% CI 1.0–1.9) . Similar results were obtained in a Spain nested case-control study; a greater adherence to the Mediterranean-type dietary pattern enhanced fertility (OR 0.56, 95% CI 0.35–0.95) .
Gestational diabetes mellitus is characterized as glucose intolerance is first diagnosed during pregnancy, is one of the most common pregnancy complications, and has serious, long-term consequences for both mother and baby .
A prospective study included 13,110 American women who were free of cancer, cardiovascular disease, type 2 diabetes and history of GDM and reported at least one singleton pregnancy between 1992 and 1998 in the Nurses’ Health Study II. This study found that the Western dietary pattern high in red and processed meat was positively associated with the risk of GDM (OR 1.63, 95% CI 1.20–2.21), however, a prudent dietary pattern high in fruit, green leafy vegetables and fish was negatively associated . Another prospective study included 21,411 singleton pregnancies in the Nurses’ Health Study II between 1991 and 2001 and reported that a pre-pregnancy low-carbohydrate dietary pattern with high protein and fat from animal food sources was positively associated with the risk of GDM (OR 1.36, 95% CI 1.13–1.64), whereas association for a pre-pregnancy low-carbohydrate dietary pattern with high protein and fat from vegetable food sources was not found .
In a study in 10 Mediterranean countries, the GDM was interpreted both by the American Diabetes Association (ADA) 2010 and the more stringent International Association of the Diabetes and Pregnancy Study Groups (IADPSG) 2012 criteria which require only one abnormal glucose value, instead of two, for the diagnosis. Adherence to a Mediterranean dietary pattern was associated with better glucose tolerance and a lower incidence of GDM (8.0% vs. 12.3%, OR = 0.618, by ADA_2010 and 24.3% vs. 32.8%, OR = 0.655, by IADPSG_2012 criteria) . Another study concluded that a prudent dietary pattern in pregnancy was associated with decreased risk of GDM (OR 0.54, 95% CI 0.30–0.98), especially among women who were either overweight or obese pre-pregnancy (OR 0.31, 95% CI 0.13–0.75) . In China, He et al. found that the vegetable dietary pattern was significantly associated with a lower risk of GDM (RR 0.79, 95% CI 0.64–0.97), while the sweets and seafood pattern was associated with increased risk of GDM (RR 1.23, 95% CI 1.02–1.49) . In an Australian population-based prospective cohort study, a Mediterranean-style dietary pattern pre-pregnancy was inversely associated with GDM risk (RR 0.85, 95% CI 0.76–0.98), moreover, a dietary pattern of meats, snacks and sweets was associated with higher GDM risk (RR 1.38, 95% CI 1.02–1.86) .
A randomized clinical trial aimed at identifying the effects of the Dietary Approaches to Stop Hypertension (DASH) diet (rich in fruits, vegetables, whole grains and low-fat dairy products, and contained lower amounts of saturated fats, cholesterol and refined grains with a total of 2400 mg Na/day) on glucose tolerance and lipid profiles in GDM found that in pregnant women with GDM, the DASH eating pattern for four weeks resulted in beneficial effects on glucose tolerance and lipid profiles compared with a control diet . The same authors demonstrated that consumption of the DASH diet by pregnant women with GDM had beneficial effects on fasting plasma glucose, serum insulin levels, Homeostasis Model of Assessment-Insulin Resistance score, plasma total antioxidant capacity, and total glutathione levels .
The results from the Nurses’ Health Study II indicated that healthy dietary patterns including the alternate Mediterranean Diet (aMED), DASH, and alternate Healthy Eating Index (aHEI) were significantly associated with a lower risk of GDM (aMED:RR 0.76, 95% CI 0.60–0.95; DASH:RR 0.66, 95% CI 0.53–0.82; aHEI:RR 0.54, 95% CI 0.43–0.68) . Moreover, they observed increased adherence to the aMED, DASH, and aHEI dietary patterns with a lower risk of type 2 diabetes mellitus (T2DM) in this large prospective study of women with a history of GDM (aMED:RR 0.60, 95% CI 0.44–0.82; DASH:RR 0.54, 95% CI 0.39–0.73; aHEI:RR 0.43, 95% CI 0.31–0.59) . All three dietary patterns include a high intake of fruit, vegetables, nuts, legumes, soy, fish, seafood, whole grains and low intake of red and processed meats. aMED includes moderate alcohol and higher ratio of monounsaturated fatty acid to saturated fatty acid (MUFA:SFA). aHEI also includes moderate alcohol, in addition to multivitamin use, higher ratio of polyunsaturated fatty acid to saturated fatty acid (PUFA:SFA) and lower intakes of transfat. DASH assesses high intakes of low-fat dairy, lower sweetened beverages and sodium.
Hypertensive disorders of pregnancy (HDPs), including gestational hypertension and preeclampsia, are obstetric complications . The frequency of preeclampsia ranges between 2% and 7% in healthy nulliparous women. Preeclampsia is associated with adverse health outcomes (maternal mortality, perinatal deaths, preterm birth, and intrauterine growth restriction) for both the mother and child .
Research in nulliparous pregnant Norwegian women, found that a dietary pattern characterized by high intake of vegetables, plant foods and vegetable oils resulted in a lower risk of preeclampsia (OR 0.72, 95% CI 0.62–0.85), whereas a dietary pattern characterized by a high consumption of meat, sweet drinks, and snacks increased the risk (OR 1.21, 95% CI 1.03–1.42) . In a prospective cohort study, researchers showed that low adherence to a Mediterranean-style dietary pattern and high adherence to a traditional dietary pattern during pregnancy were positively associated with higher blood pressure during pregnancy, but were not associated with gestational hypertension or preeclampsia outcomes . In a population-based study of Australian women, pre-pregnancy consumption of a Mediterranean-style dietary pattern was associated with lower risk of developing HDPs (RR 0.58, 95% CI 0.42–0.81) .
In contrast to the above studies, in a Brazilian cohort, investigators did not observe associations between dietary patterns and prospective changes in blood pressure during pregnancy and the early postpartum period. The difference may be explained on the basis that this study only included healthy pregnant women and excluded those who developed hypertension or preeclampsia. The investigators hypothesized that the powerful mechanisms involved in maternal hemodynamic adaptations, such as the integrity of endothelium, were not disrupted in normotensive pregnant women, and maintained blood pressure despite the different dietary intakes .
The prevalence of depressive symptoms during pregnancy varies from 8% to 26% around the world . Depressive symptoms have been associated with adverse outcomes for both mother and fetus, such as insufficient weight gain, premature birth, low birth weight, obstetric complications, and postpartum depression [78,79,80,81].
From a ‘Rhea’ cohort study in Crete, Greece, intake of the ‘health-conscious’ pattern comprised vegetables, fruit, nuts, pulses, fish and seafood, olive oil and dairy products led to a protective effect against depressive symptoms in postpartum women (RR 0.51, 95% CI 0.25–1.05) . Another study found that traditional (OR 0.84, 95% CI 0.73–0.97) and health-conscious (OR 0.77, 95% CI 0.65–0.93) dietary patterns were associated with protective effects on symptoms of anxiety whereas there was an increased possibility of anxiety symptoms with the vegetarian pattern (OR 1.25, 95% CI 1.08–1.44) . This is due to the reduced animal sources meat and fish in the vegetarian pattern; meat is the richest source of Vitamin B12 and fish is the main source of long-chain essential n-3 PUFA, especially docosahexaenoic acid (DHA; 22:6n-3). These nutrients are known to be necessary components for optimal neurological function . The direction of causality between observed association between the vegetarian diet and increased maternal anxiety requires further exploration. In a prospective cohort of Brazilian women, researchers found that a pre-pregnancy healthy pattern (the common-Brazilian or healthy patterns) was inversely associated with depressive symptoms throughout pregnancy, even to early postpartum [49,50].
In the Osaka Maternal and Child Health Study, researchers hypothesized that a dietary pattern high in the main food sources of nutrients that are potentially protective against postpartum-depressive symptoms would be related to a lower risk of postpartum depression. However, there was no association between dietary patterns and the risk of depressive symptoms .
Preterm birth, defined as delivery before 37 weeks or 259 days of gestation occurs in about 10% of pregnancies globally, is significantly associated with neonatal mortality and morbidity, and has long-term adverse consequences for health [83,84].
In Denmark, a Mediterranean diet during pregnancy was associated with reduced risk of preterm birth (OR 0.61, 95% CI 0.35–1.05) . Results from a large prospective cohort study in Norway, showed that a “traditional” dietary pattern or a “prudent” dietary pattern during pregnancy was associated with a reduced risk of preterm birth, 9% (OR 0.91, 95% CI 0.83–0.99) and 12% (OR 0.88, 95% CI 0.80–0.97) respectively . In another large, prospective cohort of close to 60,000 Danish women followed during pregnancy, researchers found that a high intake of Western-type diet led to an increased risk of preterm birth (OR 1.30, 95% CI 1.13–1.49), while a seafood diet had a modestly protective effect (OR 0.90, 95% CI 0.72–1.11) . Moreover, a study of the association between preconception dietary patterns and preterm birth demonstrated that a high-protein/fruit pattern reduced the likelihood of induced preterm birth (OR 0.31, 95% CI 0.13–0.72) and a high-fat/sugar/takeaway was positively associated with preterm birth (OR 1.54, 95% CI 1.10–2.15) and shorter gestational ages . In contrast, in other studies there was no association between Mediterranean diet adherence during pregnancy and the risk of preterm birth [28,29], although a mother-child cohort study found that adherence to the Mediterranean diet was associated with a lower risk of preterm birth, specifically in overweight and obese women (OR 0.7, 95% CI 0.6–0.9) .
A high adherence to a maternal prudent diet (characterized by a high intakes of fruit and vegetables, whole meal bread, rice, and pasta, yogurt, and breakfast cereals and low intakes of chips and roast potatoes, sugar, white bread, processed meat, crisps, tinned vegetables, and soft drinks) was found to be associated with greater offspring bone size and bone mineral density; this study suggested that maternal dietary pattern during pregnancy was an independent determinant of bone mineral accrual in the offspring . Through a prospective study of the Danish National Birth Cohort, researchers indicated that the maternal Western diet was positively associated with offspring forearm fractures (hazard ratio 1.11, 95% CI 1.01–1.23) .
Four studies have observed the relationship between maternal dietary patterns during pregnancy and subsequent small for gestational age (SGA) infants. A Danish study has found that the Western diet led to a significantly higher risk of having a SGA infant and lower birth weight . In New Zealand, Thompson et al. found that the ‘traditional’ diet in early pregnancy was associated with a lower risk of having a SGA infant (OR 0.86, 95% CI 0.75–0.99) . A Dutch study indicated that high adherence to a Mediterranean dietary pattern was associated with decreased risk of SGA . In Japan, pregnant women in the ‘wheat products’ pattern had significantly higher probabilities of having a SGA infant (multivariate OR 5.2, 95% CI 1.1–24.4) .
Studies conducted in UK and Brazil both found a positive association between dietary pattern and birth weight, a ‘health conscious’ pattern in UK and a snack dietary pattern in Brazil [60,61]. Moreover, another study from the Netherlands suggested that high adherence to an energy-rich dietary pattern was associated with increased crown–rump length in the first trimester .
In contrast, two studies found that no dietary pattern was significantly associated with any of the birth outcomes [29,63]. An adherence to Mediterranean diet was not significantly associated with the risk of delivering an infant with fetal growth restriction . No dietary pattern was significantly associated with any of the birth outcomes, the mixed dietary patterns may provide antagonistic relationships between foods and nutrients that result in null associations with birth outcomes .
The prevalence of asthma and allergies in children has been increasing globally during the last few decades .
Four studies have examined the relationship between a maternal Mediterranean diet during pregnancy and asthma or other atopic conditions in children. Two studies found a protective effect of a high level of adherence to a Mediterranean diet during pregnancy against childhood atopic conditions at ages 6–7 years [31,32]. However, the other two studies suggested that a Mediterranean diet during pregnancy was not associated with infantile wheeze and eczema [33,34].
A Japanese study examined three maternal dietary patterns (Healthy, Western, and Japanese). The maternal Western pattern during pregnancy may have had a preventive effect against wheeze in offspring (OR 0.59, 95% CI 0.35–0.98), however, neither the Healthy diet nor Japanese diet during pregnancy was related to wheeze or eczema in offspring . Two other studies investigated the association between maternal dietary patterns during pregnancy and wheeze, but none of the maternal dietary patterns was meaningfully associated with wheeze in the offspring [35,65].
A further study investigated the relationship between maternal diet and maternal asthma control during pregnancy. This study found pre-pregnancy dietary patterns influenced maternal asthma control .
A small number of studies have investigated the association between dietary patterns and other pregnancy/offspring outcomes.
There are two global domains of child problem behavior: internalizing problems consist of emotionally reactive, anxious/depressed symptoms, somatic complaints and symptoms of being withdrawn; externalizing problems consist of attention problems and aggressive behavior. Both high adherence to the traditionally Dutch diet characterized by a high intakes of fresh and processed meat, potatoes, margarines and low intake of soy and diet products (OR 1.11, 95% CI 1.03–1.21) and low adherence to the Mediterranean diet (OR 0.90, 95% CI 0.83–0.97) during pregnancy are significantly associated with an increased risk of childhood externalizing problems; there were no associations between dietary patterns and child internalizing problems .
There are some studies of the congenital defect about babies. In contrast to the prudent diet (characterized by a high intakes of fish, garlic, nuts, and vegetables and a higher frequency of hot meals per day) users, mothers with the high adherence to the Western diet increased the risk of having an offspring with a cleft lip or cleft palate approximately two fold (OR 1.9, 95% CI 1.2–3.1) . The Mediterranean dietary pattern seemed to have a protective effect in reducing the risk of offspring being affected by spine bifida . A more ‘Non-health conscious’ maternal food pattern was associated with an increased risk of hypospadias in baby boys (OR 1.54, 95% CI 1.06–2.26) . Women who adhered to a prudent dietary pattern, even when folate supplementation was administered, may have lower risk of neural tube defects (NTDs) and some heart defects .
When cancer in offspring was studies, Musselman et al. found that higher adherence to a diet that is high in fruits and vegetables was associated with a lower likelihood of childhood germ cell tumors (OR 0.83, 95% CI 0.69–0.99) . A case-control study provided preliminary evidence that a diet higher in fruit and lower in fried foods and cured meats during pregnancy may decrease the risk of unilateral retinoblastoma in offspring (OR 0.75, 95% CI 0.61–0.92) .
The validity of identifying maternal dietary patterns and of studying the relationship between dietary patterns using FFQs has been demonstrated . The majority of studies in this review used FFQs rather than diet scores to obtain a comprehensive account of dietary information. However, there are inevitable misclassifications in dietary intake; this is likely to bias the magnitude of the observed effects towards the null hypothesis. Moreover, as most of the diet information was obtained from memory, limitations of recall bias are also inevitable.
In a mother–child ‘Rhea’ cohort in Crete, Greece, two dietary patterns during pregnancy were identified using principal component analysis: ‘health conscious’ and ‘Western’. The ‘Western’ dietary pattern comprised mainly meat and meat products, potatoes, sugar and sweets, cereals, fats except olive oil, salty snacks, eggs, beverages and sauces . In contrast, the Western pattern in a Japan study was characterized by low consumption of soft drinks and sweets . The Western pattern in Greece may thus have been more extreme than the Western pattern in Japan. This is just one example of different dietary patterns in the different studies having the same name. In contrast, some patterns that have similar content have an alternative name in different studies. There is a variety of names for the ‘Mediterranean Diet’, such as Mediterranean-type diet, Mediterranean Diet Index, alternate Mediterranean Diet and Mediterranean Diet Score. Confusingly, the main diet content are similar, but the methods for calculation are subtly different, and dietary criteria to define the Mediterranean Diet was used for some studies [27,28] while much research used the method of scoring according to the consumption of various foodstuffs [22,25,26,29,30,31,32,33,34,35]. Therefore, a comparison between various research studies is both difficult and challenging and formal development of taxonomy and classification is required.
This review has highlighted many research studies of dietary patterns associated with pregnancy outcome, however, some pregnancy outcomes have yet to be investigated, such as intrahepatic cholestasis of pregnancy.
In addition, low-glycemic-index (GI) diet has also impact on pregnancy outcomes [89,90,91]. In recent years, low-GI diet has received increased interest from the public and researchers as an important strategy in GDM management . However, low-GI diet mainly focuses on the glycemic index of food and its carbohydrate content, not the overall diet. We did not include it in our review. Also, there is a comprehensive evaluation of GI and pregnancy outcomes .
The studies detailed above highlight the importance of emphasising healthy dietary choices in preconception counseling to optimise not only reproductive outcomes but also general maternal health. Current guidelines of preconception care emphasise that nutrition and certain lifestyle factors play an important role in pregnancy. This review finds evidence that, for European countries, the Mediterranean diet is a relatively healthy diet. Importantly, the diets with higher intake of fruits, vegetables, legumes and fish have positive pregnancy outcomes in general and this conclusive evidence should be communicated to women specifically. As a modifiable factor, diet is a key area for intervention in pregnant women, but the precise content of the intervention is yet to be elucidated.
This work was supported by the National Natural Science Foundation of China (Hua Zhang, grant numbers 81370732, 81571453), (Yinyin Xia, grant number 81502777); the Natural Science Foundation Project of CQ CSTC (Yinyin Xia, grant number 2012jjA10070, 2015jcyjA10032); and the Science-Technology Research Program of the CQ Municipal Commission of Education (Yinyin Xia, grant number KJ1400231).
The following abbreviations are used in this manuscript:
|GDM||gestational diabetes mellitus|
|FFQs||food frequency questionnaires|
|PCA||principal components analysis|
|LCA||Latent Class Analysis|
|DASH||Dietary Approaches to Stop Hypertension|
|aMED||alternate Mediterranean Diet|
|aHEI||alternate Healthy Eating Index|
|MUFA:SFA||the ratio of monounsaturated fatty acid to saturated fatty acid|
|PUFA:SFA||the ratio of polyunsaturated fatty acid to saturated fatty acid|
|T2DM||Type 2 diabetes mellitus|
|HDPs||Hypertensive disorders of pregnancy|
|SGA||small for gestational age|
|NTDs||neural tube defects|
XC, DZ and XM: were involved in writing the manuscript and contributed to critical revision of the manuscript; YX, PB and HZ had primary responsibility for final content.
The authors declare no conflict of interest.