Periconceptional glycemic index and the risk of gestational diabetes mellitus and a large-for-gestational-age infant.

OBJECTIVES

We determined whether the glycemic index (GI), a metric of dietary carbohydrate impact on blood glucose, in early pregnancy was associated with the risk of gestational diabetes mellitus (GDM) or a large-for-gestational-age (LGA) infant among nulliparous individuals.

METHODS

This is a secondary analysis of individuals without pregestational diabetes enrolled in the prospective United States-based Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-To-Be multisite cohort. The exposure was average periconceptual GI calculated from the Block Food Frequency Questionnaire. GI scores were categorized in quartiles from Q1 (best, reference) to Q4 (worst). Outcomes were diagnosis of GDM based on third-trimester screening and LGA birth per sex- and gestational-age standardized birthweight ≥90th percentile. Secondarily, we assessed for effect modification by dietary quality and neighborhood-level food access.

RESULTS

Among 7742 assessed participants, the median GI was 49.8 (IQR: 47.4, 52.1), and 94.9% had a low or favorable GI score (≤55). The highest quartile of GI was not associated with GDM (aRR: 1.17; 95% CI: 0.85-1.62), but Q3 was associated with an increased risk of GDM versus Q1 (Q3 vs. Q1 5.0% vs 3.9%: aRR: 1.46, 95% CI 1.07-1.99). GI was not associated with an LGA infant. There was no evidence of effect modification by dietary quality or food access (P ≥ 0.05 for both).

CONCLUSIONS

In a prospective cohort of nulliparous individuals with predominantly low GI in early pregnancy, the GI was not consistently associated with GDM or LGA birth. Further studies are needed to better understand the mechanisms by which diet-based glycemia may affect pregnancy outcomes.