Ultra-Processed Food Consumption and Oxidative Stress in a U.S. Pregnancy Cohort

Abstract

In recent decades, consumption of ultra-processed foods (UPFs) has increased, especially in high-income countries, with UPFs now accounting for over half of caloric intake in the U.S. UPF consumption is associated with many adverse health outcomes from obesity and cardiovascular disease in general populations to gestational diabetes and hypertensive disorders during pregnancy. Pregnancy represents a critical window of susceptibility in which maternal diet (including consumption of UPFs) can influence long-term maternal-child health outcomes through mechanisms including epigenetic modification and fetal programming. UPF consumption may also increase oxidative stress (OS), which is already naturally elevated during pregnancy due to placental development and increased metabolic demand. This is important because OS has been implicated in preeclampsia, gestational diabetes, and preterm birth as well as changes in fetal growth. The relationship between gestational UPF consumption and OS in pregnancy remains largely unknown, with only one prior study conducted in Mexico. As a result, this study is the first to examine the associations between gestational UPF consumption and urinary OS biomarkers [malondialdehyde (MDA, a marker of lipid peroxidation) and 8-hydroxy-2’-deoxyguanosine (8-OHdG, a marker of oxidative DNA damage)] in a U.S. cohort. Among participants in the UPSIDE pregnancy cohort (Rochester, NY; n=269), the percentage of caloric intake attributed to UPFs (%UPF) was determined from 24-hour dietary recalls in mid-late pregnancy via the NOVA classification system. Specific gravity-adjusted urinary MDA and 8-OHdG were measured in trimesters 2 and 3 using HPLC-MS-MS. Unadjusted, Adjusted, and Extended linear regression models were constructed following a systematic covariate selection process. In this cohort, on average, 55.73% of caloric intake was due to UPF consumption and the average Healthy Eating Index score was 53.7 out of 100, reflecting suboptimal diet quality. Contrary to our hypothesis, in the Adjusted models, UPF consumption was inversely associated with MDA in trimester 3 (-0.40% MDA per percent increase in UPF consumption; 95%CI: -0.70%, -0.096%; p=0.01). Associations with 8-OHdG in trimester 3 were directionally similar, but non-significant, while trimester 2 biomarkers showed no meaningful associations with UPF consumption. These findings, which are consistent with the only other pregnancy study on this topic, suggest a more complex relationship between UPF consumption and OS than expected, potentially reflecting the importance of fatty acid substrate availability in the generation of MDA as an OS biomarker. Nonetheless, high UPF consumption and poor diet quality observed in this cohort reinforce the need for improved access to healthier foods in U.S. pregnant women and for prenatal counseling regarding diet and nutrition. Additional research into the mechanisms, pathways, and health outcomes of UPF consumption during pregnancy is warranted.