Multiple lines of evidence from epidemiological observations have implicated that the early-life environment is linked to the risk of noncommunicable diseases later in life [1
]. The Developmental Origin of Health and Disease (DOHaD) determined that the developing conditions in utero modify the long-lasting bodily functions and physiology of the offspring [3
]; thus, the maternal diet can influence the life-course health of the child.
Insufficient maternal dietary intake that does not meet the increased demands during pregnancy is a risk factor for adverse birth outcomes, such as low birth weight, preterm birth, and intrauterine growth restriction [4
]. Previous studies have focused on the relationship between the intake of specific nutrients and newborn height and weight [5
]. Appropriate maternal intake of methyl-donor nutrients, micronutrients, and omega-3 fatty acids is particularly important for fetal neurodevelopment [7
Recently, comprehensive assessments of maternal diets have been conducted using the Healthy Eating Index, Alternate Healthy Eating Index, and Dietary Approaches to Stop Hypertension (DASH) [9
], as well as by the application of dietary pattern analysis to identify the effects of diet on various pregnancy outcomes [12
]. However, these studies have mainly focused on certain foods or food groups. Moreover, studies involving comprehensive dietary quality assessments based on multiple nutrients, rather than an individual nutrient, are less common.
Both food-based and nutrient-based dietary indices are useful in assessing the overall quality and/or properties of a pregnant woman’s diet. However, the international use of food-based indices will require harmonization of the food database because foods frequently consumed are unique to each country’s dietary culture. In contrast, nutrients are universal so that nutrient-based dietary indices are globally used without further processing. Therefore, in this study, we used two dietary indices that are commonly used in the world, which cover a wide range of nutrients: the Nutrient-Rich Food Index 9.3 (NRF9.3) and the energy-adjusted dietary inflammatory index (E-DII) (Figure 1
The NRF9.3 was originally developed to score the nutritional value of foods [21
] and has recently been used to assess the nutritional value of the diet consumed by individuals [22
]. The NRF9.3 is useful for assessing the overall dietary quality; however, it has not yet been applied to assess the dietary quality of pregnant women.
The DII is a comprehensive index of diet-derived inflammatory capacity [19
] and was designed to be universally applicable across all human studies for dietary assessment [20
]. Since the relationship between energy and nutrient consumption varies across populations, the energy-adjusted DII (E-DII) was developed, which is often used as an improved version of DII [20
]. Multiple studies have reported that E-DII scores are not only associated with cardiovascular disease, obesity [24
], and inflammatory biomarkers [25
] but also maternal and child health [11
]. It has also been documented that the unadjusted DII score is linked to the risk of preterm birth and low birth weight [29
]. However, with the exception of one study [30
], all previous investigations have focused on the total E-DII (or DII) score and do not identify which parameters contribute most to the variability of DII scores in each analyzed population.
Both NRF9.3 and E-DII are nutrient-based dietary metrics. We applied them to pregnant women, because the main nutritional problems during pregnancy are nutrient intake below the requirements for pregnancy and/or excessive intake of nutrients associated with low-grade chronic inflammation [31
In this study, we aimed to assess and report the overall dietary quality and inflammatory potential of pregnant women in a single-center birth cohort in Japan using the NRF9.3 and the E-DII. We also intended to the nutrients that contribute to the variability of each score and the relationship between these scores and the intake patterns of food groups.
This study evaluated the maternal dietary quality and inflammatory potential of a cohort of pregnant Japanese women using the NRF9.3 and E-DII indices. The mean (SD) NRF9.3 and E-DII scores were 602 (106) and 1.00 (1.55), and the NRF9.3 and E-DII scores depicted a significant inverse correlation (Figure 2
). Based on the tertile stratification of each index, the nutrients that had profound effects on the scores of each index were identified. In case of the NRF9.3 score, dietary fiber, iron, potassium, magnesium, and vitamin C contributed to the variation of total score across the tertiles (Figure 3
). On the other hand, for the E-DII score, dietary fiber, vitamin A, niacin, vitamin E, β-carotene, magnesium, vitamin B1, vitamin C, zinc, vitamin B6, and folate contribute to the variation of the total score across the tertiles (Figure 4
). The food groups whose intake were positively associated with dietary quality, as assessed using either NRF9.3 or E-DII, were vegetables and fruits even after adjustment for potential confounders and after considering the influence of misreporting EI. In addition, intake of wheat flour and wheat products was positively associated with inflammatory potential, as assessed using E-DII (Table 2
, Table 3
, Table 4
and Table 5
The NRF9.3 score represents the adherence to the national RDV. To the best of our knowledge, this is the first study to apply the NRF9.3 index to assess the dietary quality of pregnant women. A closer look at the breakdown of the NRF9.3 and analysis of the component scores revealed the extent of insufficient intake of qualifying nutrients as well as the extent of excess intake of disqualifying nutrients, compared to the RDV (Figure 3
). Among the qualifying nutrients, the component scores for iron were low in all the tertile groups. The component scores of vitamins A and D were much lower than 100 in the highest tertile, suggesting insufficient intake of these nutrients. Regarding the disqualifying nutrients, the differences between the three tertile groups were small. As shown in Table S4
, the participants in the lowest NRF9.3 tertile group (T1) were also more likely to have insufficient intake of other essential nutrients, such as folate, vitamins B1, B2, and B6; however, these nutrients are not components of the NRF9.3 score.
The DII score characterizes the inflammatory potential of a diet based on the literature-based evidence, which is globally applicable. In recent years, the DII of pregnant mothers has been reported to be associated with birth size and childhood health, and has received much attention. However, previous studies that have investigated the DII score of pregnant women have focused only on the score, and have failed to examine which nutrients contribute to the score [11
]. Since the nutrients that contribute to the DII score may differ depending on the target population, it may be desirable to describe not only the total score but also the parameter-specific scores. As shown in Figure 4
, the parameter-specific scores for dietary fiber, vitamins A, B1, B2, D, and niacin were positive in all the tertile groups. The parameter-specific scores of most anti-inflammatory nutrients increased as the E-DII scores increased, suggesting that the major cause for an increase in final E-DII score was the low intake of dietary fiber, vitamins, and minerals. The parameter-specific scores of pro-inflammatory nutrients were below 0 in all tertile groups, except for cholesterol. Interestingly, the intake of saturated fatty acid, total fat, and carbohydrates were not major factors in increased E-DII scores in our cohort.
As shown in Table 2
and Table 3
, the food groups whose intake were positively associated with higher dietary quality in either NRF9.3 or E-DII were legumes, vegetables, fruits, fish, and shellfish. A previous study stated that a diet high in bread, confectioneries, and soft drinks and low in fish and vegetables during pregnancy might be associated with low birth weight [12
]. Another study reported that adherence to vegetable dietary patterns may be associated with a lower risk of preeclampsia [15
]. A recent review reported that a high consumption of vegetables and fruits is associated with lower risks of preterm birth and low birth weight [49
]. These findings signify that a high intake of fruits and vegetables is characteristic of high dietary quality during pregnancy. Our nutrient-based dietary assessment identified that a high NRF9.3 score and a low E-DII score were associated with a high intake of vegetables and fruits, a finding which is consistent with the previous studies.
The purpose of this study was to examine the quality of the diet, through the intake of foods, not supplements. Nutrients in dietary supplements are not metabolized in the same way as those in foods, and this puts supplement users at risk of excess intake. Also, from the viewpoint of sustaining a healthy dietary habit, we decided to exclude intake from supplements when assessing dietary quality in this study. However, we were aware that pregnant women often take vitamins and minerals through dietary supplements. It would be a challenge in the future to determine accurate nutrient intake from dietary supplements and evaluate both food-derived and supplement-derived nutrients.
The main strength of our study is that it is the first time that the NRF9.3 index has been applied to assess the overall dietary quality of pregnant women. In particular, using the comprehensive database for sugars [36
], we were able to deduce the added sugar intake and NRF9.3 scores, which is in contrast to the previous study that calculated the nutritional score without considering sugars [50
]. Importantly, the anthropometric and overall energy/nutrient intake characteristics of the participants were almost identical to those of the pregnant women cohort in National Health and Nutrition Survey (NHNS 2015–2017) [35
]; therefore, the results are generalizable to some extent, although our cohort was a single-center cohort from an urban area in Japan (n
= 108). Additionally, we assessed the overall dietary quality and inflammatory potential using two globally established metrics, NRF9.3 and E-DII, and found there was a strong negative correlation between them. The analysis of the component score or the parameter-specific score clarified that the major nutrients contributing to both indices were similar: dietary fiber, magnesium, vitamin C, and vitamin A. This result is especially important because the most of the previous DII studies have not revealed which nutrients are associated with the final score.
This study has several limitations. First, three-day food dietary records were used for dietary assessment; however, the minimum number of days required for estimating an individual’s average EI is usually longer than 3 days [52
]. Second, the possibility of seasonal variation, which might have introduced bias in the assessment of average dietary intake, was not considered. Third, self-reported dietary records have the potential disadvantage of under-reporting of the dietary intake [45
]. Nevertheless, all analyses of nutritional score calculations and food intake were adjusted for EI. Furthermore, sensitivity analyses were performed by excluding the under-reporters.