Periconceptional Maternal Diet Characterized by High Glycemic Loading Is Associated with Offspring Behavior in NEST

Abstract

Maternal periconceptional diets have known associations with proper offspring neurodevelopment. Mechanisms for such associations include improper energy/nutrient balances between mother and fetus, as well as altered offspring epigenetics during development due to maternal nutrient and inflammatory status. Using a comprehensive food frequency questionnaire and assessing offspring temperament with the Infant-Toddler Social and Emotional Assessment (n = 325, mean age = 13.9 months), we sought to test whether a maternal periconceptional diet characterized by high glycemic loading (MGL) would affect offspring temperament using adjusted ordinal regression. After limiting false discovery to 10%, offspring born to mothers in tertile 3 of glycemic loading (referent = tertile 1) were more likely to be in the next tertile of anxiety [OR (95% CI) = 4.51 (1.88-11.07)] and inhibition-related behaviors [OR (95% CI) = 3.42 (1.49-7.96)]. Male offspring were more likely to exhibit impulsive [OR (95% CI) = 5.55 (1.76-18.33)], anxiety [OR (95% CI) = 4.41 (1.33-15.30)], sleep dysregulation [OR (95% CI) = 4.14 (1.34-13.16)], empathy [6.68 (1.95-24.40)], and maladaptive behaviors [OR (95% CI) = 9.86 (2.81-37.18)], while females were more likely to exhibit increased anxiety-related behaviors [OR (95% CI) = 15.02 (3.14-84.27)]. These associations persisted when concurrently modeled with the maternal-Mediterranean dietary pattern. In a subset (n = 142), we also found MGL associated with increased mean methylation of the imprint control region of SGCE/PEG10. In conclusion, these findings highlight the importance of maternal dietary patterns on offspring neurodevelopment, offering avenues for prevention options for mothers.

Keywords: ADHD attention-deficit disorder; autism spectrum disorder; child behavior disorders; cord blood methylation; epigenetics; glycemic index; glycemic loading; imprinted genes; maternal diet; neurodevelopment.

Figure 1

MGL and child temperament. For a given tertile of MGL compared to tertile 1 (referent), the odds ratio (95% confidence interval) represents the risk of being in a higher tertile of behavioral outcome. Unadjusted (gray) and adjusted (blue) odds ratios (95% confidence intervals) are plotted. Estimates were adjusted for breastfeeding at least 3 months, age of child at behavioral assessment, maternal fiber intake, total calories, folate, education, diabetes, obesity, smoking, and age, as well as child parity, premature birth, weight, race, and child sex. * p < 0.05; ^† Benjamini–Hochberg FDR < 0.10.

Figure 2

Sex-specific associations of MGL (T2 vs. T1) and offspring temperament. For a given tertile of MGL, compared to tertile 1 (referent), the odds ratio (95% confidence interval) represents the risk of being in a higher tertile of behavioral outcome. Males (gray) and females (magenta) odds ratios (95% confidence intervals) are plotted. Estimates were adjusted as before stratified by sex. * p < 0.05.

Figure 3

Sex-specific associations of MGL (T3 vs. T1) and offspring temperament. Sex-specific associations with MGL (tertile 3 vs. tertile 1) and offspring behavioral outcomes. For a given tertile of MGL compared to tertile 1 (referent), the odds ratio (95% confidence interval) represents the risk of being in a higher tertile of behavioral outcome. Males (gray) and females (magenta) odds ratios (95% confidence intervals) are plotted. Estimates were adjusted as before stratified by sex. * p < 0.05; ^† Benjamini–Hochberg FDR < 0.10.

Figure 4

Sex-specific associations of MGL and offspring imprint control region methylation. Sex-specific associations of ICRs and MGL diet. Tertiles of MGL were assessed for associations with mean methylation values at known imprint control regions. For example, each tertile of MGL was associated with a 1.38% increase in mean methylation at the SGCE/PEG10 ICR. Estimates were adjusted for maternal fiber intake, total energy intake, self-reported race, smoking status, diabetes status, folate intake, age, obesity status, education, as well as paternal age and child birth weight, preterm status, and parity. * p < 0.05.

Figure 5

Concurrent modeling of MGL and Mediterranean dietary adherence on offspring temperament. For the 3rd tertile of MGL compared to the 1st (referent), the odds ratio (95% confidence interval) represents the risk of being in a higher tertile of behavioral outcome. Odds ratios (95%CI) are plotted for glycemic loading (GL; orange) and Mediterranean diet (MD; green). Estimates were adjusted for breastfeeding at least 3 months, age of child at behavioral assessment, maternal fiber intake, total calories, folate, education, diabetes, obesity, smoking, race, and age, as well as child parity, premature birth, weight, and child sex. * p < 0.05; ^± Previously reported significant when modeled without MGL as a concurrent predictor.