In the present large cohort study, we investigated the association of maternal fruit and vegetable and vitamin C intakes during mid-pregnancy with increased fetal growth and infant growth up to 6 months of age, in a Korean population. We found maternal fruit and vegetable intake had a positive association with the biparietal diameter of the fetus and infant weight at birth to 6 months. Also, maternal vitamin C intake was positively associated with the abdominal circumference of the fetus and infant birth length.
Similar results, demonstrating fruit and vegetable consumption during pregnancy favorably influences fetal growth, have been reported by previous investigators. Rao et al.  and Mikkelsen et al.  documented a significant increase in birth weight with increased fruit and vegetable consumption (as a food group). Ramón et al.  observed an increased intake of vegetables, but not fruit, during pregnancy was associated with birth length and weight. Loy et al.  noted that vegetable intake was linked to birth length and head circumference, and fruit intake was correlated with birth weight and length, and head circumference.
In our study, the multiple logistic regression analysis revealed a significant inverse relationship between fruit and vegetable consumption and the risk of low growth (<25th percentile) of biparietal diameter and weight at birth. In line with this result, there was a significant inverse relationship between consumption of vitamin C (an abundant nutrient in fruit and vegetables) and the risk of low (<25th percentile) weight at birth, weight from birth to 6 months, and length at birth. Prior research has demonstrated that the vitamin C contained in fruit and vegetables may contribute to both, placental functions and optimal immune system functioning , which are paramount for fetal development.
Along with maternal vitamin consumption, we investigated maternal oxidative stress (MDA level), considering that it might be increased under normal pregnancy environments and cause adverse birth outcomes. Although the correlation coefficient value is very low, the maternal fruit and vegetable, and vitamin C intakes were negatively correlated with the oxidant marker (MDA) levels at mid-pregnancy, in agreement with the literature . This factor could partly explain the underlying reason for our findings that women consuming low amounts of fruit and vegetables, and vitamin C had low birth outcomes’, with the offspring having low growth up to 6 months. Antioxidant defense systems are vital to protecting tissues and cells from damage caused by oxidative stress. Consequently, an imbalance between increased oxidative stress and decreased antioxidant defense can adversely affect pregnancy outcomes, including fetal growth retardation . Increased vitamin C consumption (supplement or via fruits and vegetables) during pregnancy may be advantageous to birth size due to its role in the endogenous antioxidant defense system. In a previous epidemiological study in Spain of 586 newborns, an increase in dietary vitamin C intake during pregnancy was implicated to reduce the association between estimated maternal dietary benzo[a]pyrene (a cigarette smoke carcinogen) intake and infant size at birth . Also, maternal fruit and vegetable intake may modify the association between birth weight and bulky DNA adduct levels, which are purported to be a sensitive biomarker of genotoxic agents .
Another possibility is that vitamin C, an essential cofactor for two key enzymes (lysyl and prolyl hydroxylase) in collagen biosynthesis, is beneficial for cartilage and bone development . Some experimental studies have found that vitamin C inadequacy can result in a decreased proliferation of chondrocytes in the growth plate and impaired matrix synthesis  in mice, and low bone mineral density and bone length in a scorbutic guinea pig model . However, further investigation is needed to unveil the true effects of vitamin C intake during pregnancy on fetal bone growth and collagen metabolism.
During normal prenatal development, fetal vitamin C concentrations in plasma are higher than maternal levels. One study demonstrated an approximately two-fold higher vitamin C level in newborns at birth than in their mothers , implying that a high vitamin C status is of particular importance in the fetus. However, our subjects did not meet the optimal vitamin C range. The mean vitamin C intake in our subjects was 134.3 mg/d, and about 36.2% and 51.2% of pregnant women had vitamin C intakes below the EAR and recommended nutrient intake, respectively (data not shown). The vitamin C intake status in our pregnant women is higher than that reported in China , Iran , England , New Zealand , and Brazil , but lower than that documented in Thailand  and the USA .
The main vitamin C sources were fruits (40.3%) and vegetables (46.5%) (data not shown), but the correlation between fruit and vegetable/vitamin C with fetal and infant growth was partly inconsistent. Notably, in the biparietal diameter, a significant inverse relationship between fruit and vegetable intake and a low growth rate (<25th percentile) was observed, whereas mothers with a relatively lower vitamin C intake did not have a higher risk of low biparietal diameter. It may be that fruits and vegetables have played a role in other ways, such as provide unexpected bioactive components that are not vitamin C. The high number of bioactive components are found in fruits and vegetables, such as vitamins, minerals, phytosterols, phenolic compounds , carotenoids, and fiber. Also, fruit and vegetable intakes may be a pivotal part of a healthy lifestyle and dietary pattern and may be associated with other health factors besides the variables we have adjusted [35, 36].
Our study had some limitations. First, we did not measure plasma vitamin C concentrations, which are considered to be a more objective means to assess vitamin C status than dietary vitamin C intake reports alone, due to the limited amounts of samples for other measurements (e.g., environmental heavy metals and toxins). Second, the maternal dietary intake data from a single 24-h recall may not be sufficient to determine typical daily intake, due to a probable and considerable intra-individual variability in food and nutritional intake. However, the well-trained dietitians applied standard protocols to assist the subjects’ recollection of their daily diet and thereby minimize any potential bias. Regarding intra-individual variability of nutrient intake, as part of the 4th Korean National Health and Nutrition Examination Survey in 2009 , the intra-individual variation in nutrient intake measured by a single 24-h dietary recall was compared to an original 1-d dietary interview. Regarding total energy and, particularly, vitamin C intake, comparable values were obtained from each interview. Besides, the seasonality of food can be a problem for 24-h dietary recall. To minimize its undesired effect, the season could be added as a covariate in our analysis. However, there was no difference between fruits and vegetables and vitamin C intake according to the season (data not shown). This observation seems to be because the timing of the pregnant women’s recruitment, that is, the period of the dietary survey was evenly distributed. Therefore, we decided that it was not necessary to include the season as a covariate. In addition, a Greek study investigating the impact of seasonality on the mean nutrient intake of children and adolescents reported that the intake of vitamin C and the macronutrient contribution to the total energy intake, estimated by two 24-h recalls of different seasons, were statistically similar between the two periods (spring/summer and autumn/winter) . Evaluation of potential genotypes that can affect oxidative stress metabolism, as well as the enzymatic activity for heterozygotes and assessment of some specific oxidative markers, might also have improved the reliability of our results.
The strengths of our study include the sample size and the research strategy, considering the strictly-controlled prospective birth cohort design of the MOCEH study in Korea. Also, this research collected reliable data from medical records. Furthermore, an adjustment for crucial confounders having the potential to affect the correlation between maternal vitamin C intake and birth outcomes of newborns were considered, in detail, for more precise analysis.
Fetal growth is an important determinant of health and disease throughout a human’s lifetime. Interestingly, a prospective birth cohort study in Brazil noted a positive association of birth length with blood pressure at 11 years of age, whereas birth weight and blood pressure were not associated . It exemplifies the importance of the relationship between maternal vitamin C and infant birth length as modifiers that could affect children later in life.