Periconceptional diet quality is associated with gestational diabetes risk and glucose concentrations among nulliparous gravidas

Background: Gestational diabetes mellitus (GDM) and elevated glucose concentrations below the threshold for GDM diagnosis have been associated with adverse pregnancy and offspring outcomes. Dietary interventions initiated during pregnancy have demonstrated inconsistent beneficial effects. Limited data exist regarding the effects of periconceptional diet on gestational glycemia. Objective: To evaluate independent associations between periconceptional diet quality with GDM frequency and glucose concentrations from GDM screening and diagnostic tests among nulliparous gravidas. Design: This is a secondary analysis of N=7997 participants from the NuMoM2b multicenter, prospective, observational cohort study of first pregnancies. The Alternative Healthy Eating Index (AHEI)-2010 was computed from food frequency questionnaires completed in early pregnancy (6-13 weeks), reporting usual dietary intake over the preceding 3 months. GDM screening was performed either by non-fasting 1-hour 50g glucose load (N=6845), followed by 3-hour 100g glucose tolerance test (GTT) for those with raised glucose concentrations (N=1116; at risk for GDM), or by a single 2-hour 75g GTT (N=569; all GDM risk levels). Logistic and linear regression were used to estimate the associations between the AHEI-2010 score with odds of GDM, having raised blood glucose on the 1-hour screening test, and continuous glucose concentrations on screening and diagnostic tests. All models were adjusted for a priori covariates: maternal age, race/ethnicity, early-pregnancy body mass index, smoking habits, rate of gestational weight gain, energy intake, nausea and vomiting in early pregnancy, study site. Results: Poorer periconceptional diet quality was observed among participants who were younger, with higher BMI, lower income levels, and of non-Hispanic Black or Hispanic ethnicity. The GDM rate was 4%. Each 1-point increase in AHEI-2010 score was associated with a 1% decrease in the odds of being diagnosed with GDM (beta=-0.015, p=0.022, OR=0.986, 95% CI 0.973 to 0.998). Diet quality was inversely associated with each post glucose load concentration on the non-fasting screening test and the 2-hour and 3-hour GTT. Conclusion: Poor periconceptional diet quality is independently associated with an increased risk of GDM and with minor elevations in serum glucose concentrations on GDM screening and diagnostic tests, in a diverse cohort of nulliparas. Periconception intervention studies targeting diet quality are warranted.