Association Between Dietary Soy Isoflavones Intake and the Risk of Hyperemesis Gravidarum: A Cross-Sectional Study in Chinese Pregnant Women

Abstract

(1) Background: Diet plays a crucial role in the intake of phytoestrogens, which are closely related to the pathogenesis of some pregnancy complications. However, no studies have explored the potential association between soy isoflavones, a type of phytoestrogen, and the risk of hyperemesis gravidarum (HG). This study aims to investigate the correlation between dietary intake of soy isoflavones and the risk of HG. (2) Methods: As part of the China Birth Cohort Study (CBCS), 2418 pregnant Chinese women (mean age: 31.2 ± 3.4 years) were enrolled between April 2021 and September 2022. Dietary intake was evaluated using a validated 108-item semi-quantitative food frequency questionnaire, with soy isoflavones intake estimated based on five food groups. HG was defined as a condition characterized by a pregnancy-specific vomiting score (PUQE) ≥ 13, weight loss of ≥5% due to severe nausea and vomiting before 16 weeks of gestation, inability to eat or drink normally, significant limitations in daily activities due to severe nausea or vomiting, or the need for hospitalization caused by the condition. The association between soy isoflavones intake and HG was analyzed using binary logistic regression and restricted cubic spline regression. (3) Results: Among all participants, 212 women (8.8%) were diagnosed with HG. The dietary intake of soy isoflavones was 14.56 (IQR: 9.89, 25.36) mg/d. After full adjustment for confounding factors, the results indicated that individuals with the highest dietary intake of soy isoflavones had a lower risk of developing HG (OR: 0.56, 95% CI: 0.36, 0.88. P_trend = 0.012). (4) Conclusions: Higher dietary intake of soy isoflavones is associated with a reduced risk of HG. We advocate for a dietary approach to the management of HG that prioritizes the intake of legume-rich foods, particularly those abundant in soy isoflavones.

Keywords: dietary soy isoflavones; hyperemesis gravidarum; legumes; soy isoflavones; soybeans.

Figure 1

A multivariable-adjusted restricted cubic spline model was used to examine the association between dietary soy isoflavones intake and HG. Knots were positioned at the 25th, 50th, and 75th percentiles of the dietary soy isoflavones distribution. The solid red line represents the estimated odds ratio, while the red-shaded area indicates the 95% confidence interval. The binary logistic regression was adjusted for age, annual household income, educational level, occupation, physical activity, pre-pregnancy BMI, gestational week, parity, total energy intake, smoking, alcohol consumption, the use of nutritional supplements, intake of meats, fish and seafood, fruits and vegetables. BMI: body mass index. CI: confidence interval. HG: hyperemesis gravidarum.

Figure 2

Subgroup analyses, adjusted for multiple variables, were conducted to explore the association between dietary soy isoflavones intake and HG. Given the linear nature of this association, dietary soy isoflavones were treated as a continuous variable (per standard deviation increase) in the subgroup analysis. The left section presents the variables, grouping categories, and sample sizes for each subgroup, along with the corresponding ORs and 95% CIs. A dashed vertical line represents an OR of 1, while squares denote the point estimates of the ORs, with horizontal lines illustrating the 95% CIs. Black indicates non-significant differences, whereas red highlights statistically significant differences. Due to the limited number of smokers and alcohol consumers in the sample, subgroup analysis for these variables were not performed. The binary logistic regression model was adjusted for age, gestational week, parity, total energy intake, physical activity, pre-pregnancy BMI, annual household income, educational level, occupation, smoking, alcohol consumption, the use of nutritional supplements, intake of meats, fish and seafood, fruits and vegetables. BMI: body mass index. OR: odds ratio. CI: confidence interval.