Infant body composition is a critical determinant of long-term health. Higher adiposity in infancy is associated with an increased risk of metabolic disorders, including obesity, type 2 diabetes, and cardiovascular diseases later in life (1, 2). Different maternal factors (e.g., obesity, gestational weight gain, socioeconomic characteristics) have been identified as influencing infant adiposity (1, 3). Lifestyle and nutrition interventions may play a key role in improving maternal metabolic profiles and reducing the risk of obesity in the next generation (4). Specifically, an optimal diet quality during pregnancy, as well as high adherence to a DASH diet or a Mediterranean dietary pattern has been associated with a better glycemic status and a reduce size and adiposity at birth (5–8).

Dietary assessment has shifted from only measuring the quantity of energy and nutrient intake to more qualitative aspects (type of foods, grade of processing). In addition, the recent study of chrononutrition has documented a relationship between nutrient intake, meal timing patterns, and circadian rhythm with metabolic health (9). Current dietary patterns and meal timing may particularly affect pregnancy, considering it is a period of high metabolic vulnerability due to hormonal and physiological adaptations (10–12). Changes in hormones such as leptin, ghrelin, and insulin, when paired with disrupted circadian rhythms, could lead to impaired hunger signals, increased appetite, and a preference for energy-dense foods (13), contributing to suboptimal dietary patterns and consequently impact fetal growth and adiposity. Evidence about meal timing and chrononutrition behaviors in pregnancy is scarce. Some studies (cross-sectional or observational) have reported an association between nighttime eating, high energy/carbohydrate intake later in the day, and metabolic alterations in pregnant women (14). Generating evidence in this area is imperative as it could facilitate the implementation of simple strategies targeting sleeping and eating schedules for pregnant women.

In non-pregnant adults, disturbances in biological rhythms can alter homeostasis and metabolic processes which have been linked to chronic conditions such as obesity, diabetes, and cardiovascular disease (9, 15). Some chronotypes, particularly the evening type, have been associated with suboptimal glycemic control, higher risk of mortality, and even higher risk of eating disorders (9). The mechanisms explaining the intricate connection between the timing of food consumption throughout the day and regulating the biological clock still need to be clarified.

This study aimed to evaluate if certain chrononutrition behaviors in pregnancy are associated with infant adiposity at 6 months of life.

Out of the 502 women enrolled in the OBESO cohort, 44 were excluded due to the presence of chronic or uncontrolled diseases (20 with type 2 diabetes, 13 with hypertension, and 11 with uncontrolled hypothyroidism). Additionally, 183 women were eliminated as they did not attend at least three prenatal visits. Eighty-six women had incomplete dietary information, and one of them was excluded because of a nighttime shift. No woman exhibited an implausible energy intake (<500 or >3,500 kcal/d).

Furthermore, 88 infants were eliminated from the analysis because they lacked FM measurement (non-attendance at visit, age below 6 months, not measured for FM) or birthweight. A total of 100 mother-infant pairs were included in the final analysis.

This is one of the first prospective studies that describe chrononutrition behaviors during pregnancy (meal skipping, morning latency, and nighttime eating) and shows some influence on maternal nutrient consumption (protein, total fat, saturated and monounsaturated fat, omega-3 fatty acids) and infant adiposity. Our findings suggest a potential link between maternal nighttime eating during pregnancy and higher infant adiposity at 6 months.

Around 30% of our population engaged in nighttime eating, a prevalence that falls within the range of 15–45% documented in other reports on pregnant women (12, 25, 26). In our study, maternal nighttime eating was associated with a higher infant adiposity. This could be related to circadian rhythm disruption by nighttime eating. Glucose sensitivity typically peaks during daylight, as insulin release responds to circadian cycles, with pancreatic β-cells having maximum activity around midday. This suggests that food intake might be more efficiently metabolized during the day, as postprandial glycemic digressions are more significant in the evening (15, 27). The liver, crucial in nutrient metabolism, is strongly connected to feeding patterns and can alter the phase of liver clock genes. Consuming food out of daylight may increase plasmatic non-esterified fatty acids and reduce insulin sensitivity (15, 27). Among chronotypes, the evening type is associated with suboptimal glycemic control (9); moreover, people with diabetes with night eating syndrome are more susceptible to obesity, suboptimal HbA1c values, and diabetic complications (28). In the context of pregnancy, nighttime eating was associated with higher maternal glucose values (25, 29) and inversely associated with early-phase insulin secretion (29) in two different cohorts of Asian and African American pregnant women. Turkish pregnant women classified with night eating syndrome also presented greater HOMA-IR (Homeostasis model assessment of insulin resistance), insulin, total cholesterol, and high-density lipoprotein cholesterol (30). Maternal hyperglycemia during pregnancy predisposes the offspring to a higher risk of obesity or higher FM at birth and in childhood (31, 32). A recent systematic review observed that interventions to improve glycaemia during gestation showed a protective effect on infant adiposity (33). While glucose regulation is one of the metabolic processes most studied about chrononutrition, many other mechanisms may be involved. During pregnancy, it has been documented that some characteristics of chrononutrition could affect the secretion of melatonin and cortisol (12), which influence different metabolic processes such as oxidative stress, inflammation, the renin–angiotensin system, epigenetic regulation, among others (11). Higher cortisol concentrations and melatonin concentrations have been associated with adverse perinatal outcomes (34–36). Gestational cortisol is also linked to infant adiposity during the first 6 months of life (37). There is an intricate interplay between circadian rhythm, nutrition, and metabolic processes during pregnancy, underlying the need for further research in this complex area. Manipulating meal timing may be a potential strategy to manage glucose homeostasis, improve metabolic health, and prevent adverse perinatal outcomes.

Nighttime eating was also related to higher consumption of fat (total, saturated, monounsaturated). Previous studies in non-pregnant population have associated evening-eating patterns with higher consumption of unhealthy foods (ultraprocessed products, sweets, sweetened beverages, high-fat foods, fast foods, alcohol) and lower intake of healthy food groups (fruits, vegetables, fish) (38). In other studies, in pregnant women, nighttime eating was associated with higher fat and lower calcium, iron, and riboflavin intake (25, 26). While a low-quality dietary pattern, a high-fat diet, and higher saturated fat consumption during pregnancy have been related to higher neonatal adiposity (39–42), our models did not reveal an effect of total or specific fat intake on infant FM. Maternal nutrition is one of the main determinants of maternal-fetal well-being (43–45). Nighttime eating in pregnancy has been associated with shorter gestation length (−0.45 weeks 95%CI: −0.75 to −0.16) and a higher risk of preterm birth (OR 2.19, 95%CI: 1.01 to 4.72) in Asian women (46).

Most women in our study reported having three main meals per day, and almost no one skipped main meals during gestation. Other studies have reported a high percentage of pregnant women having at least three meals per day women [95.6% of a sample of Polish women (47) and 75.6% of a U.S.-based cohort (48)]. Meal frequency could represent an essential aspect of prenatal nutrition, as recommended by guidelines advocating three meals (in addition to snacks) to adequately address the nutritional demands of pregnancy (49), to reduce the risk of birth defects, inadequate fetal growth and development, and adverse chronic disorders for the mother-infant binomial (50). Our results showed an association of higher protein intake (and a trend of higher energy and fat intake) in those women not skipping meals. Other authors have revealed a lower protein intake, lower plasma docosahexaenoic and eicosapentaenoic acid and beta-carotene (51), iron deficiency (52), and even undernutrition (low mid-upper arm circumference) in pregnant women who skip meals (52). Furthermore, skipping meals may alter various physiological mechanisms due to lower dietary consumption and different factors related to chrononutrition. Skipping meals could trigger a stress response, elevating prostaglandins, adrenaline, and insulin levels, potentially contributing to adverse perinatal outcomes (48, 53). The study performed by Siega-Riz and cols. Showed that women with fewer meals/day had a relative risk of 1.57 (95%CI: 0.88 to 2.79) of early preterm and 1.22 (95%CI: 0.86 to 1.73) of late preterm delivery, even after adjusting for confounding variables such as energy intake (pregestational BMI and supplement use) (48). A more recent study by Englund-Ögge et al. also found that routine consumption of three main meals was related to a lower preterm risk (hazard ratio: 0.89, 95%CI: 0.80 to 0.98, p = 0.028) (adjusted by fiber as an indicator of a healthy diet) (53). Regular breakfast skipping during pregnancy has also been associated with a higher risk of gestational diabetes (54).

More extended night fasting has been related to lower energy intake (55) in accordance with the negative correlation coefficients between fasting hours and general intake we observed. The length of overnight fasting has also been negatively associated with fasting glucose, independent of energy and macronutrient intake, which could reflect clock-related mechanisms of metabolism regulation (55). Women in our sample reported a fasting period close to 12 h, although it is only an approximation and not a true overnight fasting window. Other authors have reported 9.9 h in an Asian perinatal cohort (55) and 13 h in Brazilian pregnancies (38), reflecting the influence of sociocultural factors on diet. Given the current scarcity of research, there is uncertainty regarding what constitutes a healthy duration of fasting time and the potential effects of usual fasting periods on perinatal outcomes. A broader set of studies have assessed Ramadan fasting and perinatal outcomes, but this type of practice differs from a non-religious habitual dietary behavior. A recent meta-analysis showed that Ramadan-related fasting affected lower GWG during fasting but did not affect weight gain throughout pregnancy, gestational age at birth, preterm birth, birthweight, low birthweight, or other maternal-infant outcomes (56).

Breakfast and dinner latencies represent novel elements scarcely studied within the chrononutrition field. These aspects could be associated with alterations in consumption and metabolic issues related to light/dark periods. Our results show that the shorter time between waking up and having breakfast was associated with higher consumption of omega-3 fatty acids, which may be a random finding. Exploring these timing factors and their association with maternal-infant health is essential.

Despite the potential influence of sleep on various aspects of the diet, we found no associations between sleep duration and energy/nutrient consumption or any of the chrononutrition behaviors evaluated. A study in healthy pregnant women by Van Lee and cols. Found no association between sleep quality or sleep hours and fasting, number of meals, or nighttime eating. The only identified association was that good quality sleep was associated with fewer discretionary calories (beverages, cakes, desserts, snacks) (57).

EBF is widely considered the standard in infant nutrition for the first 6 months of life. While in this specific analysis, EBF was not associated with FM, other studies conducted in this early stage show that EBF is an influential factor contributing to differences in FM (58, 59). Given its substantial impact on infant health, including benefits such as reduced mortality, lower rates of infections, positive effects on neurodevelopment (60), and the protective effect against adult obesity (61), promoting breastfeeding should be a priority.

Maternal obesity is recognized as a factor that predisposes the offspring to alterations in fetal growth (62), with excessive and insufficient growth associated with higher adiposity later in life (63). In our bivariate analysis, we found a lower FM in infants born to mothers with gestational obesity. However, pregestational BMI was not associated with FM in our multiple regression models examining the association between maternal nighttime eating and adiposity.

This study presents some limitations that warrant consideration. Firstly, the absence of a specific dietary assessment method to evaluate chrononutrition behaviors prospectively may have introduced error and different biases in the estimation of our indicators. The fact that we analyzed information from each recall and used the average for computing chrononutrition variables is an indirect assessment, especially regarding the approximated overnight fasting duration used in our analysis. In addition, the study design did not allow us to assess or consider differences in chrononutrition behaviors across pregnancy trimesters. It is possible that the behaviors described in this study are not representative of all stages of pregnancy. Although we were able to assess usual sleeping schedule, the retrospective nature of the questionnaire is not ideal, and may have also introduced error.

Additionally, a more comprehensive evaluation of episodes of night eating could have included the proportion of calories or a macronutrient analysis, providing a more detailed understanding of dietary patterns. It is also possible that women under or overreported energy/macronutrient consumption, potentially confounding some results. The lack of FM measurement at birth is another limitation, as having this variable could have allowed for better adjustment in our models or facilitated a longitudinal analysis. On the other hand, considering the sample size of this study, the estimated power is 73%, which is a reasonable value. However, it may somehow decrease the reliability and precision of our findings. The results of this study may not be applicable for the general population, considering ethnicity, socioeconomic status and that women receiving prenatal care in our hospital are classified as “high-risk” pregnancies. Even though we established strict inclusion criteria in this cohort for reducing variability, the external validity may be low.

This study possesses several strengths. Using data from a prospective cohort enabled the assessment of maternal variables throughout pregnancy and early infancy. This longitudinal approach allowed for considering various influential factors on adiposity in our statistical models. The ability to evaluate dietary consumption longitudinally, and with at least three recalls, allowed a better characterization of chrononutrition behaviors throughout pregnancy. Air-displacement plethysmography is a well-validated method to assess FM in this early stage of life. Our outcome variable included different indicators of adiposity such as a percentage (relative), kilograms (absolute), and FMI (relative to length) to have a broader analysis with better indicators of body composition.

The temporal aspects of feeding should be included when evaluating dietary patterns in pregnant women. Different aspects of chrononutrition, including meal skipping, breakfast latency, and nighttime eating, could affect the intake of nutrients and may indicate different nutrition and metabolic imbalances.

Maternal nighttime eating is associated with higher infant adiposity at 6 months of age, independently of maternal obesity, use of metformin or insulin, energy intake in pregnancy, and exclusive breastfeeding. Chrononutrition behavior modification could represent an innovative and feasible strategy that may be incorporated into nutritional counseling during pregnancy for improving nutrient intake and promoting metabolic health.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics and Research committees—Instituto Nacional de Perinatología (Project No. 3300-11402-01-575-17). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

AR-C: Conceptualization, Data curation, Formal analysis, Visualization, Writing – original draft. BM-C: Data curation, Writing – review & editing. IG-L: Investigation, Writing – review & editing. CR-H: Investigation, Writing – review & editing. ER-M: Conceptualization, Project administration, Writing – review & editing. ES-S: Data curation, Writing – review & editing. GE-G: Conceptualization, Funding acquisition, Project administration, Writing – review & editing. OP-P: Conceptualization, Methodology, Project administration, Supervision, Writing – review & editing.