Respiratory microbiota diversity and composition in recurrent protracted bacterial bronchitis: a cross-sectional study

Abstract

Introduction: Recurrent protracted bacterial bronchitis (RPBB) is a significant risk factor for bronchiectasis in children, characterized by multiple episodes of protracted bacterial bronchitis (PBB) annually. With an increasing global incidence, a detailed understanding of RPBB's pathophysiology is essential, particularly regarding the role of lung microbiota.

Methods: This cross-sectional study recruited 39 children from Jinhua Maternal and Child Health Hospital between January 2021 and December 2022, including 18 with PBB, 11 with RPBB, and 10 as controls. Bronchoscopy with bronchoalveolar lavage (BAL) was performed to collect lung microbiota samples, which were analyzed using 16S rDNA sequencing. Microbial diversity and composition differences among groups were assessed using alpha and beta diversity metrics, PERMANOVA, and Linear Discriminant Analysis Effect Size (LEfSe), with statistical significance set at P < 0.05.

Results: RPBB patients exhibited a distinct lung microbiota composition compared to controls, characterized by an increased abundance of pathogens such as Acinetobacter and Mycoplasma, alongside a reduction in beneficial genera like Streptococcus and Granulicatella. The RPBB group also demonstrated greater overall microbiota diversity, indicating dysbiosis that may contribute to disease severity and persistent respiratory symptoms.

Conclusion: This study revealed significant alterations in the lung microbiota of children with RPBB, suggesting that microbial imbalance could play a crucial role in disease pathogenesis. These findings highlight the importance of targeted prevention and therapeutic strategies aimed at restoring microbiota balance to improve pediatric respiratory health.

Keywords: children; dysbiosis; lung microbiota; microbial imbalance; recurrent protracted bacterial bronchitis.

Figure 1

(A) Rarefaction curves for the three groups, illustrating species richness and sequencing depth. The curves plateau, indicating sufficient sequencing depth to capture the microbial diversity. (B) Shannon-Wiener diversity curves for the three groups, showing species diversity within each group. The curves level off with increasing sequencing depth, further confirming sequencing adequacy.

Figure 2

Venn diagram depicting the shared and unique bacterial species among the three groups. The RPBB and CG groups shared 633 species, with 553 unique to RPBB and 65 to CG. RPBB and PBB shared 963 species, while 223 were unique to RPBB and 262 to PBB. These findings suggest a substantial overlap in microbiota composition, with distinct species contributing to RPBB and PBB.

Figure 3

Alpha diversity indices (ACE, Chao1, Shannon, Observed species, Pielou, and Simpson) comparing species richness and evenness across the three groups. The RPBB group showed higher species richness values compared to the PBB and CG groups, while the CG group demonstrated greater species evenness than both RPBB and PBB groups. Each index is represented as a box plot to facilitate comparison of the observed trends.

Figure 4

Beta diversity among the three groups as demonstrated through Hierarchical Clustering Analysis. (A) Cluster tree showing distinct separation between CG and RPBB, indicating significant differences in microbial composition. (B) Cluster tree illustrating the relationship between RPBB and PBB, where RPBB clusters within PBB, suggesting a microbiota shift potentially linked to disease progression.

Figure 5

PCoA analysis illustrating beta diversity across the three groups. (A) The RPBB and CG groups formed distinct clusters, reflecting significant compositional differences in lung microbiota (P < 0.05). (B) The RPBB and PBB groups showed partial overlap, indicating some similarities, but overall compositional differences were not statistically significant (P > 0.05).

Figure 6

NMDS analysis depicting microbial community differences among the three groups. A stress value < 0.2 indicates a reliable representation of beta diversity. (A) RPBB and CG showed distinct clustering patterns, suggesting significant differences in microbiota composition. (B) RPBB and PBB exhibited partial overlap, reflecting both shared and unique microbiota characteristics.

Figure 7

ANOSIM plots demonstrating beta diversity among the three groups. (A) ANOSIM analysis showed significant differences in microbial community structure between RPBB and CG (R = 0.487, P = 0.002), with greater variability in RPBB. (B) The differences between RPBB and PBB were not statistically significant (P > 0.05), suggesting similar microbial community structures.

Figure 8

LEfSe analysis identifying differentially abundant taxa in the BALF samples. (A) Cladogram highlighting key bacterial taxa with significant differences between RPBB and CG groups. Red nodes represent taxa enriched in RPBB, while green nodes indicate those more abundant in CG. Yellow nodes represent taxa without significant differences. (B) Linear Discriminant Analysis (LDA) scores demonstrating statistically significant bacterial differences between RPBB and CG, with an LDA score threshold of 2.0 (P < 0.05). Taxonomic assignments follow the IJSEM 2021 taxonomy updates.

Figure 9

Bar diagram illustrating microbial community composition at the phylum, genus, and species levels. (A) At the phylum level, RPBB had a significantly lower abundance of Bacillota compared to CG (P < 0.05), while Pseudomonadota were significantly enriched (P < 0.05), indicating a shift in microbial balance. (B, C) At the genus and species levels, RPBB showed significantly increased levels of gram-negative bacteria, including Acinetobacter, Pseudomonas, and Hemophilus, while beneficial genera such as Streptococcus were reduced (P < 0.05). Taxonomic assignments reflect the reclassification by Oren and Garrity (2021).

Figure 10

ROC analysis assessing the diagnostic potential of bacterial genera with significantly altered abundances in RPBB. (A) ROC curves for bacteria enriched in RPBB, including Acinetobacter (AUC = 0.79, P = 0.024), Mycoplasma (AUC = 0.86, P = 0.006), and Enterobacter (AUC = 0.77, P = 0.038). (B) ROC curves for bacterial genera with significantly lower abundances in RPBB, such as Streptococcus (AUC = 0.96, P < 0.001) and Granulicatella (AUC = 0.964, P < 0.001), which exhibited high sensitivity and specificity as potential diagnostic markers.