Table 1.
Demographic and clinical characteristics of pregnant individuals with and without inflammatory bowel disease (IBD) included in the study and recruited between 2019 and 2022.
Fig 1.
Gut and vaginal inflammatory markers and vaginal microbiome diversity in pregnant individuals with and without IBD.
(A) Fecal calprotectin levels. (B) Vaginal cytokine gene expression. (C,D) Vaginal microbial alpha diversity using Shannon (C) and Simpson (D) indexes. (E) Principal Coordinates Analysis (PCoA) showing vaginal microbiota beta, PERMANOVA was used for beta diversity analysis based on Aitchison distance to calculate P-values and R². (F) Vaginal Community State Types (CSTs) proportions compared using Fisher’s Exact Test. Differences in sample size (N) across panels are due to missing data for specific variables.
Fig 2.
Correlation between vaginal microbiota diversity and vaginal IL-6 and TNF-α expression.
(A, B) Vaginal IL-6 expression levels by Shannon (A) and Simpson (B) diversity indexes. (C, D) Vaginal TNF-α expression levels by Shannon (C) and Simpson (D) diversity indexes. P-values for alpha diversity were derived from linear models. The blue line indicates the fitted regression model, with the shaded area representing the 95% confidence interval.
Fig 3.
Association of vaginal microbiota with dietary components.
(A, B) HEI-2015 Added sugar score associated with Shannon (A) and Simpson (B) diversity indexes. (C, D) HEI-2015 Dairy scores associated with Shannon (C) and Simpson (D) diversity indexes. P-values for alpha diversity were derived from linear models. (E) Principal Coordinates Analysis (PCoA) showing vaginal microbiota beta diversity by HEI-2015-Added Sugar score (color gradient bar). PERMANOVA was used for beta diversity analysis based on Aitchison distance to calculate P-values and R². (F) Volcano plot showing significant associations between six microbial taxa and HEI-2015 Added sugar score (MaAsLin2). Significant taxa are highlighted in red based on a false discovery rate (FDR)-adjusted P-value threshold of 0.10, with taxa showing negative estimation coefficients (left) indicating negative correlations and positive coefficients (right) indicating positive correlations.
Fig 4.
Association between vaginal Community State Types (CSTs) and diet.
CST-I (L. crispatus-dominated) and CST-III (L. iners-dominated) comparison for (A) HEI-2015 scores (dietary quality), (B) HEI-2015 Total vegetables scores, and (C) HEI-2015 Added sugar scores. P-values correspond to post hoc pairwise comparisons between CSTs, conducted after detecting significant overall differences across all CSTs using ANOVA or Kruskal-Wallis tests (see S6 Table). Arrows indicate the direction of actual consumption for the added sugar and total vegetables to facilitate interpretation.
Fig 5.
Associations between HEI-2015 dietary components and vaginal cytokine expression in individuals with and without IBD.
Heatmap showing the significant correlations between HEI-2015 dietary components and the expression of three vaginal cytokines by health status (HC and IBD). Coefficients from linear models indicate the strength and direction of the associations, with positive (in blue) and negative (in red) values denoting direct and inverse correlations, respectively.