Low-Diversity Microbiota in Apical Periodontitis and High Blood Pressure Are Signatures of the Severity of Apical Lesions in Humans

Abstract

(1) Background: In developed countries, the prevalence of apical periodontitis (AP) varies from 20% to 50% for reasons that could be associated with the apical periodontitis microbiota ecology. (2) Methods: We performed a clinical study in the Odontology department of Toulouse hospital in France, to sequence the 16S rRNA gene of AP microbiota and collect clinical parameters from 94 patients. Forty-four patients were characterized with a PAI (periapical index of AP severity) score lower or equal to 3, while the others had superior scores (n = 50). (3) Results: The low diversity of granuloma microbiota is associated with the highest severity (PAI = 5) of periapical lesions (Odds Ratio 4.592, IC 95% [1.6329; 14.0728]; p = 0.001; notably, a lower relative abundance of Burkholderiaceae and a higher relative abundance of Pseudomonas and Prevotella). We also identified that high blood pressure (HBP) is associated with the increase in PAI scores. (4) Conclusions: Our data show that a low diversity of bacterial ecology of the AP is associated with severe PAI scores, suggesting a causal mechanism. Furthermore, a second risk factor was blood pressure associated with the severity of apical periodontitis.

Keywords: apical periodontitis; granuloma; high blood pressure; oral diseases; oral microbiome; periapical index.

Figure 1

A principal component analysis (PCA) was generated predicting that the main systemic clinical factor responsible for the dispersion of subjects was high blood pressure (HBP). (A) PCA between the severity of apical periodontitis (AP) (following the PAI scores), and oral and systemic clinical parameters. (B) The statistical analysis for the contribution of each oral and systemic parameter for each dimension.

Figure 2

Microbiota profiles from granuloma split into low- (LowDiv, n = 43) and high-diversity (HighDiv, n = 51) profiles. (A) Granuloma microbiota composition at family level for low (upper panel) and high (lower panel) diversity. Only the top 15 most abundant families are displayed; the remaining families are aggregated in the “other” category. (B) Alpha diversity analyses at OTU level confirm low- and high-diversity microbiome profiles in granuloma. Boxplots show the distribution of bacterial richness (observed OTUs and diversity (Shannon and Simpson) values of granuloma microbiota between low- and high-diversity profiles. Statistically significant differences were determined using a Wilcoxon test (*** p value ≤ 0.001, ** p value ≤ 0.01).

Figure 3

Microbiota profiles from granuloma in PAI ≤ 3 group (n = 44) show an increase in alpha diversity associated with the family Burkholderiaceae. (A) Linear discriminant analysis (LDA) effect size (LEfSe) cladogram of pairwise analysis for 16S rDNA sequence analysis of granuloma samples (the cladogram shows the taxonomic levels represented by rings with phyla at the innermost ring and genera at the outermost ring, and each circle is a member within that level) between two groups’–PAI ≤ 3 and PAI > 3—histogram for relative abundance (%) for taxonomic groups identified by single letters in cladogram. (Data as mean±SEM; * p < 0.05, ** p < 0.01,**** p < 0.0001, unpaired Mann–Whitney test and LDA score 2.0 for cladograms). (B) Relative abundance of the Burkholderiaceae family per sample in PAI ≤ 3 and PAI > 3 groups.

Figure 4

Microbiota profiles of high-diversity samples from granuloma split into the two groups: subjects with high blood pressure (HBP+) and without high blood pressure (HBP−). (A) Granuloma microbiota composition at family level for high diversity split into the two groups, HBP+ and HBP−. Only the top 15 most abundant families are displayed; the remaining families are aggregated in the “other” category. (B) Linear discriminant analysis (LDA) effect size (LEfSe) cladogram of pairwise analysis for 16S rDNA sequence analysis of granuloma samples (the cladogram shows the taxonomic levels represented by rings with phyla at the innermost ring and genera at the outermost ring, and each circle is a member within that level) between the two groups, HBP+ and HBP− The histogram for relative abundance (%) for taxonomic groups identified by single letters in cladogram. (Data as mean ± SEM; * p < 0.05, ** p < 0.01, **** p < 0.0001, unpaired Mann–Whitney test and LDA score 2.0 for cladograms). (C) Alpha diversity analyses at OTU level confirm HBP+ and HBP− microbiome profiles in granuloma. Boxplots show the distribution of bacterial richness (observed OTUs and diversity (Shannon and Simpson) values of granuloma microbiota HBP+ and HBP− diversity profiles. Statistically significant differences were determined using a Wilcoxon test (* p < 0.05).

Figure 4

Microbiota profiles of high-diversity samples from granuloma split into the two groups: subjects with high blood pressure (HBP+) and without high blood pressure (HBP−). (A) Granuloma microbiota composition at family level for high diversity split into the two groups, HBP+ and HBP−. Only the top 15 most abundant families are displayed; the remaining families are aggregated in the “other” category. (B) Linear discriminant analysis (LDA) effect size (LEfSe) cladogram of pairwise analysis for 16S rDNA sequence analysis of granuloma samples (the cladogram shows the taxonomic levels represented by rings with phyla at the innermost ring and genera at the outermost ring, and each circle is a member within that level) between the two groups, HBP+ and HBP− The histogram for relative abundance (%) for taxonomic groups identified by single letters in cladogram. (Data as mean ± SEM; * p < 0.05, ** p < 0.01, **** p < 0.0001, unpaired Mann–Whitney test and LDA score 2.0 for cladograms). (C) Alpha diversity analyses at OTU level confirm HBP+ and HBP− microbiome profiles in granuloma. Boxplots show the distribution of bacterial richness (observed OTUs and diversity (Shannon and Simpson) values of granuloma microbiota HBP+ and HBP− diversity profiles. Statistically significant differences were determined using a Wilcoxon test (* p < 0.05).