Nutrition in Pregnancy May Shape Metabolic Health After Delivery

A trio of studies presented at the American Diabetes Association 2026 Scientific Sessions has reframed the conversation of diet during pregnancy, pointing to diet quality, not just quantity, as a meaningful lever for managing gestational glycemia and postpartum metabolic risk. This is a conversation that has long been viewed through the narrow lens of weight gain and fetal growth.

Across hundreds of pregnancies, researchers of 3 oral presentations found that higher intake of fiber, nonstarchy vegetables, and plant protein were independently associated with lower continuous glucose monitor (CGM) readings,¹ while lower-carbohydrate diets in women with gestational diabetes improved glycemic control but raised micronutrient concerns.² Perhaps most strikingly, women randomly assigned to a higher-complex carbohydrate diet during pregnancy still showed measurably lower postpartum glucose responses 2 months after delivery, suggesting that what a pregnant woman eats may matter long after the birth.³

Together, these findings challenge prevailing assumptions about optimal gestational nutrition and open new questions about how prenatal dietary interventions might be designed to protect both mother and child over the long term.

Diet Quality, Quantity Shape Glycemia in Pregnancy¹

Erin LeBlanc, MD, MPH, of the Kaiser Center for Health Research, presented “Dietary Quality Factors Associated With CGM-Derived Gestational Glycemia,” which encompassed findings from a substudy of GO MOMs (Glycemic Observation and Metabolic Outcomes in Mothers and Offspring), a large, 9-center US study of glycemia across pregnancy.⁴ The total study population of GO MOMs is 2150 participants. LeBlanc discussed how diet quality relates to CGM-derived gestational glycemia and neonatal birthweight.

The analysis included 542 singleton pregnancies without preexisting diabetes. Diet quality was assessed using up to 5 Automated Self-Administered 24-Hour Dietary Recalls (ASA-24), collected at 2 gestational windows: up to 3 at 10 to 20 weeks and up to 2 at 24 to 28 weeks. The primary glycemic outcome was mean CGM glucose at 24 to 28 weeks, and a principal secondary outcome was neonatal birthweight for gestational age. The cohort skewed toward poor baseline diet quality, with a median Healthy Eating Index (HEI) score of 52.5. HEI is scored on a scale of 0 to 100, with higher total scores indicating a diet more aligned with key recommendations.⁵

Higher HEI scores, lower added sugar consumption, and greater intake of nonstarchy vegetables, fiber, and plant protein were independently associated with lower mean CGM glucose. LeBlanc noted that “1 cup higher intake of nonstarchy vegetables was associated with a 2 mg/dL lower mean glucose.” In contrast, animal protein, whole grains, and fat subtypes (ie, saturated, polyunsaturated, and monounsaturated fats) showed no significant associations with glycemia after correction for multiple comparisons.

No dietary components were associated with birthweight, adjusted for gestational age. LeBlanc acknowledged this may reflect confounding by glycemic treatment: “Those with high glucose levels, we don’t know how much treatment could have confounded the data.”

Sensitivity analyses excluding participants with gestational diabetes at the time of CGM data collection did not materially change results. LeBlanc and her team did not find significant associations between dietary factors and gestational diabetes diagnosis itself, which she cited as rationale for focusing on continuous glycemic outcomes. She concluded that future intervention trials should test whether “increasing specifically intake of fiber and reducing intake of added sugar" can meaningfully lower gestational glycemia. Additional GO MOMs results were previewed for the conference's poster session.

Glycemic Gains, Nutritional Trade-Offs With Low-Carb Diets²

Claire Meek, MBChB, MRCP, FRCPath, PhD, of the University of Leicester, presented a secondary analysis of the DiGest trial (Dietary Intervention in Gestational Diabetes), in which she examined whether habitual low-carbohydrate intake before dietary intervention was associated with altered nutrient profiles, glycemia, and pregnancy outcomes in women with gestational diabetes. Her presentation was titled “Low-carbohydrate Diets, Nutritional Adequacy, Glycaemia, and Perinatal Outcomes in Women With Gestational Diabetes: A Secondary Analysis of the DiGest Trial.”

Her analysis drew on data from 215 women with gestational diabetes who completed masked continuous glucose monitoring and a 24-hour dietary recall at baseline, prior to trial randomization. Participants were grouped by carbohydrate intake: standard (> 175 g/day), low (120-175 g/day), or very low (< 120 g/day). A third of the cohort fell into each category, a distribution Meek noted aligned with her clinical observations that many women with gestational diabetes self-restrict carbohydrates before any formal guidance.

Women eating very low carbohydrate diets showed a 7% reduction in continuous glucose monitoring time-above-range, a statistically significant finding after adjustment. However, the glycemic benefit came with nutritional cost. As Meek put it, “If you eat a bit less carb, you’re probably going to eat a bit more fat and protein,” and in this cohort, lower carbohydrate intake was also associated with reduced dietary fiber, vitamin C, B vitamins, calcium, and iron.

On perinatal outcomes, very low carbohydrate intake was associated with lower birthweight, but only in the context of concomitant energy restriction. Meek noted, “If you’re making all those extra calories up with delicious fat and protein, you’re not going to see the benefit here upon birthweight.” No significant differences in large- or small-for-gestational-age rates were observed.

Audience discussion raised concerns about ketone measurement, which was absent from DiGest due to COVID-era recruitment constraints; fetal brain carbohydrate requirements in the third trimester; and potential long-term cardiovascular risk from higher saturated fat intake. Meek acknowledged these as priorities for her team’s next randomized trial, which will directly compare calorie restriction vs carbohydrate restriction in gestational diabetes. A follow-up DiGest study is planned.⁶

Higher-Carb Diet Linked to Lower Postpartum Glucose Levels³

Synneva Hagen-Lillevik, PhD, MS, RD, of University of Colorado Denver-Anschutz Medical Campus, presented postpartum findings from the landmark CHOICE (Choosing Healthy Options in Carbohydrate Energy) randomized controlled trial (NCT02244814), in the discussion, “Women With Gestational Diabetes (GDM) on a Higher–Complex Carbohydrate (CHO) Diet Show Lower Postpartum OGTT Glucose.” Her analysis compared a conventional lower-carbohydrate diet to a higher-complex carbohydrate diet in women with diet-managed gestational diabetes.

The CHOICE trial provided all food to participants for 7 to 8 weeks. The conventional diet consisted of 40% carbohydrate, 45% fat, and 15% protein, while the CHOICE diet inverted the macronutrient balance to 60% carbohydrate and 25% fat, for approximately 300 grams of carbohydrate and 60 grams of fat per day. Both diets were isoenergetic, emphasized complex low-glycemic carbohydrate sources, limited simple sugars, and contained similar amounts of fiber. Participants completed oral glucose tolerance tests (OGTTs) at baseline (~30-31 weeks), after 5 weeks on their assigned diet (~36-37 weeks), and at 2 months postpartum, at which point the dietary intervention had ended and participants had resumed their usual eating habits.

During pregnancy, maternal glucose management was excellent in both groups, with CGM time-in-range above 90% for each arm. After 5 weeks on the study diets, women on CHOICE showed approximately 7% lower glucose area under the curve (AUC) on the OGTT. Infant adiposity at 7 to 10 days and 2 months post partum showed no between-group differences.

The headline postpartum finding was that women previously randomized to CHOICE had approximately 6% lower glucose AUC at 2 months postpartum vs the conventional diet group, with no differences in insulin AUC. However, self-reported dietary intake in the 3 days before the postpartum OGTT did not differ between the groups. As Hagen-Lillevik noted, “The exact metabolic mechanism is unknown and remains under our investigation. One consideration is the potential for progressive beta cell dysfunction after a higher carbohydrate diet.”

Questions from the audience probed whether detailed dietary records might reveal underlying differences in antioxidant or micronutrient intake, and Hagen-Lillevik indicated that “quality of carbohydrates” may warrant deeper exploration in future analyses.

References

1. LeBlanc, E, Gray EL, Andrei AC, et al. Dietary quality factors associated with CGM-derived gestational glycemia. Presented at: American Diabetes Association 2026 Scientific Sessions; June 5-8, 2026; New Orleans, LA. Oral presentation 1182.
2. Ahmad E, Dib S, Griep LO, Kisinski L, Meek C. Low-carbohydrate diets, nutritional adequacy, glycaemia, and perinatal outcomes in women with gestational diabetes. a secondary analysis of the DiGest trial. Presented at: American Diabetes Association 2026 Scientific Sessions; June 5-8, 2026; New Orleans, LA. Oral presentation 1183.
3. Hagen-Lillevik S, Waldman MR, Bosma GN, et al. Women with gestational diabetes (GDM) on a higher–complex carbohydrate (CHO) diet show lower postpartum OGTT glucose. Presented at: American Diabetes Association 2026 Scientific Sessions; June 5-8, 2026; New Orleans, LA. Oral presentation 1184.
4. About the study. GO MOMs. Accessed June 7, 2026. https://www.gomomsstudy.org/about-the-study/
5. How the HEI is scored. Food and Nutrition Administration. Updated April 21, 2025. Accessed June 7, 2026. https://www.fns.usda.gov/cnpp/how-hei-scored
6. Jones D, De Lucia Rolfe E, Rennie KL, et al. Antenatal determinants of childhood obesity in high-risk offspring: protocol for the DiGest follow-up study. Nutrients. 2021;13(4):1156. doi:10.3390/nu13041156