Timing of meals may be important in managing gestational diabetes for better maternal care.

The circadian system is orchestrated by the suprachiasmatic nucleus (SCN) in the hypothalamus, which synchronizes behavioral and physiological rhythms, including feeding, sleep, and hormone release to the 24 h light–dark cycle [41,42,43]. This central pacemaker is complemented by peripheral clocks located in nearly all tissues, including liver, adipose, pancreas, and placenta. These clocks are entrained not by light, but by behavioral cues like feeding and activity, regulating rhythmic gene expression involved in glucose uptake, insulin secretion, lipid metabolism, and energy balance [44,45]. During pregnancy, circadian coordination becomes more complex. Hormonal shifts, particularly in estrogen, progesterone, and placental hormones, modify sleep architecture, appetite regulation, and insulin sensitivity [46]. Evidence from animal models suggests that pregnancy induces phase shifts in peripheral clock gene expression, possibly to support altered metabolic demands [47,48]. In humans, data are more limited, but there is growing evidence that circadian disruption may impair placental function, alter fetal programming, and reduce maternal insulin sensitivity [46]. BMAL1 (ARNTL; brain and muscle ARNT-like 1) and PER3 (period circadian regulator 3) are core molecular clock components: BMAL1 forms part of the activating transcriptional complex that drives rhythmic gene expression in metabolic tissues, while PER3 participates in the negative-feedback arm that shapes circadian phase and amplitude [49,50]. In peripheral blood of pregnant women with GDM, BMAL1 and PER3 transcript levels are significantly reduced in amplitude compared to controls, correlating with elevated HbA1c and HOMA-IR [51]. These human data underscore the clinical relevance of circadian disruption and the need to align meal timing with maternal molecular rhythms.

Chrononutrition refers to the timing of food intake about the body's circadian clock, recognizing that when one eats can be as important as what or how much is eaten [52]. Key terms in the field of chrononutrition are defined in Box 2 to guide interpretation.

Under normal conditions, metabolic processes like glucose tolerance and insulin secretion follow a daily rhythm: humans are generally most insulin-sensitive in the morning and become more insulin-resistant by evening [53,54]. Eating at biologically inappropriate times (such as during the usual sleep phase at night) can desynchronize internal clocks—the central clock in the brain versus peripheral clocks in metabolic organs—leading to circadian misalignment [52,55]. Such misalignment adversely affects metabolism and has been linked to increased risks of obesity and diabetes [52].

During pregnancy, maternal physiology undergoes circadian adaptations. There are shifts in clock gene expression across gestation that modulate the expression of metabolic genes (e.g., those governing gluconeogenesis and insulin signaling) to meet the energy demands of the growing fetus [47,52,60]. If these finely tuned rhythms are disrupted, a pregnant woman may be predisposed to metabolic dysregulation, reduced insulin sensitivity, and impaired glucose homeostasis. In essence, pregnancy might be a state in which circadian coordination is especially critical for maintaining optimal insulin action and energy balance.

Emerging evidence supports the biological rationale that circadian disruption can impact pregnancy outcomes. Pregnant women engaged in shift work (especially night shifts) experience higher rates of miscarriage, preterm birth, low birth weight, and hypertensive disorders [52,61,62,63,64]. Specifically, with respect to GDM, circadian disruption appears to be a notable risk factor. A large prospective study of over 10,000 first-time pregnancies reported that evening and night shift work during pregnancy was associated with approximately 75% higher odds of developing GDM compared to daytime work [65]. Notably, in that analysis, the link between shift work and GDM was partly mediated by irregular sleep timing, accounting for approximately a quarter of the excess risk [65]. These data underscore how modern lifestyle factors, from late-night eating to variable sleep schedules, can misalign circadian rhythms and potentially impair glucose regulation in pregnancy.

Observational studies over the past decade have increasingly pointed to the importance of not just what t women eat during pregnancy, but when n they eat it. Chronotype, an individual's preference for morning or evening activity, may play a role in metabolic health during pregnancy. In a cohort of 305 women with GDM in Brazil, only 7% were identified as having an "evening" chronotype, yet this small subgroup had disproportionately worse outcomes [11]. GDM patients with an evening preference had significantly poorer sleep quality and higher rates of insomnia and fatigue, and importantly, they were far more likely to develop preeclampsia and to have their newborn require NICU admission than women with a morning chronotype. In fact, evening-type GDM mothers had an adjusted odds ratio ~4 for neonatal intensive care unit (NICU) need and ~3 for preeclampsia compared to morning-type mothers. The investigators noted that eveningness was an independent risk factor even after controlling for baseline hypertension and sleep quality, highlighting a potential intrinsic circadian vulnerability [11]. One interpretation is that women with an evening chronotype tend to keep later hours (later meals and later sleep), misaligning their food intake and hormone cycles with the endogenous circadian insulin sensitivity peak. This misalignment could exacerbate the metabolic stress of pregnancy and GDM. It also mirrors findings in non-pregnant populations where "night owls" have higher rates of metabolic syndrome and diabetes [65,66,67], presumably due to behaviors like late eating and reduced sleep that accompany an evening orientation.

Meal timing patterns themselves have been linked to GDM risk. Perhaps the clearest example is breakfast consumption, which anchors the daily feeding–fast cycle. A large Japanese prospective study found that women who frequently skipped breakfast had higher odds of developing gestational diabetes. In over 84,000 pregnancies, those who ate breakfast only 0–2 days per week (vs. daily eaters) showed a 21% increased odds of GDM (adjusted OR 1.21, 95% CI 1.05–1.41) [68]. Even moderate infrequency (eating breakfast 3–4 days per week) trended toward an elevated risk. The authors concluded that consuming breakfast less than three days a week before and during early pregnancy was associated with a significantly higher incidence [68]. Skipping breakfast may lead to greater hunger and larger meals later in the day, which, coupled with the circadian drop in insulin sensitivity that occurs in the evening, creates a metabolic disadvantage. It may also shorten the overall overnight fasting interval and decrease the normal morning rise in insulin action, both of which can worsen glycemic control. This epidemiologic insight reinforces the clinical adage that "breaking the fast" each morning is metabolically beneficial, especially for expecting mothers, by aligning nutrient intake with the body's daytime metabolic preparedness.

Conversely, late-night eating and irregular meal schedules have been associated with poorer glycemic measures in pregnancy. A cross-sectional study within a Singaporean pregnancy cohort of 1061 pregnant women examined 24 h dietary patterns of glucose levels at 26–28 weeks' gestation [69]. Women who prolonged their eating into late-night hours (thereby having a shorter overnight fasting window) tended to have higher glucose levels. In contrast, those who fasted for longer durations overnight had better glycemic profiles. Specifically, each additional hour of nightly fasting (between 7 PM and 7 AM) was associated with a small but significant reduction in fasting glucose by ~0.03 mmol/L [69].

Meanwhile, a greater number of eating episodes per day (grazing or frequent snacks) was linked to higher 2 h postprandial glucose levels (each extra eating episode corresponded to a 0.15 mmol/L increase in 2 h glucose) [69]. In adjusted models, longer uninterrupted nighttime fasts predicted lower fasting sugar, and fewer daily meals predicted lower post-meal sugar, suggesting that a concentrated, aligned eating pattern (three meals in a defined window) might be metabolically advantageous in pregnancy. These findings are in line with chrononutrition principles: continuous grazing or midnight snacking imposes a metabolic strain, whereas time-aligned eating (with sufficient fasting overnight) allows the circadian metabolic machinery to reset and function optimally for the next day [69]. For pregnant women, adopting regular meal times like including a hearty breakfast and avoiding late dinners, could improve glycemic responses. This is a practical discovery.

Importantly, studies focusing on women already diagnosed with GDM show that chrononutrition factors can distinguish better versus worse glycemic control. In a prospective analysis of 208 Israeli women with gestational diabetes, meal timing and composition were strongly associated with the mother's glucose levels and even infant outcomes [70]. Women who habitually ate a late breakfast (significantly delaying their first daily meal) had more difficulty achieving glucose targets, with over twice the risk of suboptimal glycemic control compared to those who ate earlier in the day [70]. Moreover, a higher proportion of each day's calories (particularly carbohydrates) consumed in the evening hours was correlated with worse maternal glycemic control (RR for poor control ~1.2 per incremental increase in evening carb intake) [70]. These results make sense given that eating a carb-rich meal at night (when insulin sensitivity is lowest) will tend to produce larger glucose excursions. Intriguingly, Messika and colleagues also noted effects on fetal growth: in GDM mothers, greater carbohydrate intake in both the morning and evening was associated with giving birth to a larger infant (neonatal birth weight >85th percentile) [70]. One might expect evening intake to matter (as excess nighttime calories could promote fetal overnutrition when maternal glucose is less controlled), but the morning carb effect suggests a more complex relationship, possibly that a very high carb load at any one time of day can overwhelm glycemic regulation in GDM.

Together, these findings emphasize that both intrinsic circadian biology (chronotype) and modifiable behaviors (meal timing, fasting intervals) contribute meaningfully to maternal glucose dynamics. As with the general population, pregnancy appears to amplify the consequences of circadian misalignment.

A growing number of trials are translating these insights into testable interventions. One promising strategy is to intentionally align meal timing with the body's circadian glucose metabolism to improve glycemic control in GDM. For example, a recent randomized controlled trial tested the impact of a chrononutrition-focused intervention in women with gestational diabetes [13]. In this trial, 103 women with GDM were randomized either to standard care alone or to an intervention combining tailored meal timing guidance plus sleep hygiene counseling on top of standard GDM care. The chrononutrition arm encouraged participants to consume more of their calories earlier in the day, avoid heavy late-evening meals, and maintain consistent sleep–wake times. The results were encouraging: the intervention significantly improved maternal glycemic control, reducing the proportion of women with suboptimal glucose readings (defined as <80% of self-monitored values within target range) by a substantial margin. After adjusting for confounders, the risk of poor glycemic control was ~72% lower in the chrononutrition + sleep group compared to controls. Notably, investigators observed that lowering carbohydrate intake in the evening was the key factor mediating this improvement—the intervention group curbed nighttime carbs and subsequently had much better glucose profiles [13]. This RCT did not find a significant difference in newborn outcomes such as incidence of large-for-gestational-age infants but it was relatively small and perhaps not powered for neonatal endpoints. What it did demonstrate is the feasibility and efficacy of integrating circadian-friendly eating patterns into GDM care to enhance maternal glycemic control. In essence, it provides proof-of-concept that lifestyle interventions targeting when mothers eat (and sleep) can complement the traditional focus on diet quality and quantity.

A second trial by Murugesan et al. (2025) focused on intra-meal sequencing. Women with GDM (n = 27) were guided to consume fiber-rich foods first, then protein, then carbohydrates for 10 days following a baseline period [40]. This behavioral intervention, combined with mobile health support, resulted in meaningful glycemic improvements [35]. The simplicity and scalability of this approach, requiring no change in total food content or caloric intake, suggest that food sequencing could be a low-burden, high-impact tool for refining dietary management in GDM, particularly in digitally supported care models.

Other translational research is exploring circadian interventions in pregnancy. For instance, addressing sleep disturbances, which frequently coexist with circadian misalignment, may improve metabolism. Small trials of sleep extension or improved sleep timing in women with GDM have hinted at better fasting glucose and lower insulin resistance [71]. These studies recognize that sleep and meal timing are interlinked pillars of circadian health; a synchronized approach (encouraging earlier dinners and adequate overnight sleep) could synergistically improve outcomes. A consolidated summary of these human chrononutrition studies in GDM is provided in Table 1.

Because insulin sensitivity peaks in the morning and declines by evening, concentrating carbohydrate intake earlier in the day can improve glycemic control. In the Israeli CGM-based trial, women assigned to a chrononutrition intervention consumed most of their carbohydrates before 3 PM and reduced evening carbs; this shift accounted for the 72% lower risk of suboptimal glucose readings compared with standard care [71,72].

Extending the nightly fasting interval allows endogenous metabolic pathways to reset. Data from a Singapore cohort showed that each additional hour of overnight fasting (7 PM–7 AM) correlated with a 0.03 mmol/L reduction in fasting glucose, while shorter fasts and frequent snacking were linked to higher postprandial levels [70]. A pragmatic target of a 10–12 h fast from last meal to first balances feasibility with metabolic benefit.

Altering within-meal order provides another low-burden intervention. In a 15-day CGM pilot, participants ate vegetables first, then protein, and carbohydrates last; this simple reordering reduced postprandial excursions by 20–30% without changing total caloric intake [77]. Real-world cases illustrate how chrononutrition can be applied to improve glycemic outcomes in GDM (see Box 3).

Practical, clinician-facing chrononutrition strategies and the level of supporting evidence are summarized in Table 2.

For detailed clinician-oriented food group and timing guidance (including stage applicability for GDM), see (appended before the References). Table A1

Molecular stratification has revealed heterogeneity in GDM pathophysiology, allowing more precise identification of chrono-sensitive phenotypes. Multi-omics technologies, including transcriptomics, metabolomics, and microbiomics, have unraveled time-dependent biological signatures: Genomic data (maternal and fetal) can identify variants affecting glucose metabolism or circadian regulation, informing risk stratification [85,86]. Epigenetic marks may reveal in utero exposure, while transcriptomics/proteomics (e.g., in placenta) can show active pathways [87,88]. Metabolomics (blood/urine) captures real-time metabolic state, including glucose, lipids and hormone rhythms [89]. The gut microbiome is also important: diet-induced changes in microbial taxa and metabolites (e.g., short-chain fatty acids) influence insulin sensitivity and may mediate chrononutrition effects [90]. These insights suggest that traditional GDM classifications, based solely on glucose levels, fail to account for underlying chrono-biological diversity. Stratifying patients using temporal omics markers could allow clinicians to tailor interventions with greater specificity.

In addition to molecular approaches, simple clinic-based predictors may aid early triage. For example, combining the 50 g glucose challenge test (GCT) with ultrasound-measured abdominal subcutaneous fat thickness (ASFT) at 24–28 weeks has been shown to predict GDM risk with good sensitivity and specificity [91]. Integrating such anthropometric–biochemical markers with meal timing strategies enables a pragmatic, tiered pathway: women flagged as high-risk could be prioritized for early chrononutrition counseling and short-term CGM phenotyping, with intensified, tailored meal timing interventions if CGM shows nocturnal hyperglycemia or large postprandial excursions [91].

Continuous glucose monitors (CGMs) and wearable trackers form the bridge between circadian behavior and molecular profiling. CGMs provide minute-by-minute glucose readings across the day, revealing patterns (fasting, postprandial peaks) that standardized tests often miss [92]. In pregnancy, CGM studies have shown that even before formal GDM diagnosis, women who will develop GDM have higher mean glucose and more time above normal range [92]. This continuous data allows fine-tuning of meal timing: for instance, if late-night snacks consistently spike glucose, algorithms can advise earlier or smaller dinners. Similarly, wearable devices (actigraphy watches, smartphone apps) record sleep onset/offset and light exposure, quantifying chronotype and circadian alignment. Poor sleep quality and late meals have been linked to worse glycemia and larger birth weight in GDM [79]. Mobile apps can log food intake, macronutrients and meal times, which combined with CGM create a precise diary of "chrononutrition." Although most current GDM apps simply record inputs or give generic advice [93], the vision is an intelligent platform: an AI-enabled app that collects data from CGM and sensors, analyzes an individual's circadian-metabolic profile, and then generates dynamic recommendations (e.g., "shift carbs to breakfast," "increase overnight fasting"). Notably, recent work developed an ML model predicting postprandial glucose response in GDM using CGM, diet logs, and microbiome data, finding that including microbiota features significantly improved prediction accuracy over carbohydrate counting alone [90]. Such tools demonstrate that integrating wearables and omics through AI can power precision nutrition.

An integrated chrono-omics pipeline requires advanced analytics. Machine learning algorithms can sift through genomic variants, metabolite levels, microbiome profiles and daily routines to identify predictors of glycemic excursions or dietary success. Some studies have already applied ML to GDM risk prediction, emphasizing the need to tailor models to specific populations [94]. In practice, incoming patient data (genome, lab tests, CGM traces, meal timings) would feed a trained AI model that outputs a personalized meal plan: for example, earlier breakfast time and reduced evening carbohydrates if the patient's pattern shows nocturnal hyperglycemia and late chronotype. The plan might also include probiotic or prebiotic foods to modify a dysbiotic microbiome, or micronutrient suggestions based on genomics (e.g., vitamin D dosing if polymorphisms affect its metabolism). Most importantly, the system operates in a feedback loop: real-time monitoring (CGM, sleep tracker) validates the impact of the recommendations, allowing iterative adjustments.

Implementation and limitations. While the potential is high, practical deployment faces hurdles. Multi-omics assays and AI analytics are still largely research tools, not routine clinical tests. Ensuring data quality (accurate food logging, adequate biospecimen collection) is challenging. Existing mHealth apps for GDM are mostly basic—very few incorporate AI-driven decision support [94]—highlighting a technology gap. Moreover, AI models trained on one demographic may not generalize; indeed, studies show GDM risk predictors vary by ethnicity and underscore the need for diverse training datasets [94]. Feasibility is constrained by cost, data privacy concerns, and the need for user-friendly interfaces. Any recommendations must be validated: randomized trials are needed to prove that a chrono-omics strategy improves outcomes beyond standard care. As Liu et al. (2025) note, large, ethnically diverse cohorts and rigorous methodology are essential before integrating omics into routine pregnancy care [95]. In summary, an integrated chrono-omics framework, linking SCN-driven rhythms with genomics, metabolomics and real-time monitoring, promises to personalize nutritional therapy for GDM. Realizing this vision will require multi-disciplinary collaboration. To illustrate how chrono-omics can be applied in clinical care, we present a real-world vignette integrating molecular profiling, digital monitoring, and behavioral coaching (see Box 4).