Changes in the pharyngeal and nasal microbiota in pediatric patients with adenotonsillar hypertrophy

Abstract

The present study aimed to investigate the pharyngeal and nasal microbiota composition in children with adenotonsillar hypertrophy (AH) and assess longitudinal alterations in both microbiota after a probiotic oral spray treatment. A cohort of 57 AH patients were enrolled and randomly assigned to the probiotic and placebo groups for a 5-month treatment course. Pharyngeal and nasal swabs were collected before and after treatment and analyzed by 16S rRNA-based metataxonomics and axenic cultures for pathobiont identification. 16S rRNA sequences from pharyngeal and nasal swabs of 65 healthy children (HC) were used as microbiota reference profiles. We found that the pharyngeal and nasal microbiota of AH children were similar. When compared to HC, we observed an increase of the genera Rothia, Granulicatella, Streptococcus, Neisseria, and Haemophilus, as well as a reduction of Corynebacterium, Pseudomonas, Acinetobacter, and Moraxella in both microbiota of AH patients. After probiotic treatment, we confirmed the absence of adverse effects and a reduction of upper respiratory tract infections (URTI). Moreover, the composition of pharyngeal microbiota was positively influenced by the reduction of potential pathobionts, like Haemophilus spp., with an increase of beneficial microbial metabolic pathways. Finally, the probiotic reduced the abundance of the pathobionts Streptococcus mitis and Gemella haemolysans in relation to AH severity. In conclusion, our results highlight the alterations of the pharyngeal and nasal microbiota associated with AH. Moreover, probiotic administration conferred protection against URTI and reduced the presence of potential pathobionts in patients with AH.

Importance: Adenotonsillar hypertrophy (AH) is considered the main cause of breathing disorders during sleep in children. AH patients, after significant morbidity and often multiple courses of antibiotics, often proceed to tonsillectomy and/or adenoidectomy. Given the potential risks associated with these procedures, there is a growing interest in the use of nonsurgical adjuvant therapies, such as probiotics, that could potentially reduce their need for surgical intervention. In this study, we investigated the pharyngeal and nasal microbiota in patients with AH compared with healthy children. Furthermore, we tested the effects of probiotic spray administration on both disease symptoms and microbiota profiles, to evaluate the possible use of this microbial therapy as an adjuvant for AH patients.

Keywords: adenotonsillar hypertrophy; nasal microbiota; pharyngeal microbiota; probiotics; upper airway pathobionts.

Fig 1

Study workflow.

Fig 2

Histograms of clinical features of AH cohorts. (A) Medications administered during the entire follow-up time. (B) Drugs administered during the entire follow-up time. (C) Number of otitis and tonsillitis episodes registered during the entire follow-up time. (D) Diseases registered during the follow-up period.

Fig 3

Pharyngeal and nasal microbial ecosystems in AH and HC. Alpha-diversity box plots of pharyngeal and nasal microbiota based on Shannon and Chao1 indexes (A and B, F and G). P-values obtained by the Mann–Whitney test are reported. Beta-diversity Principal Coordinates Analysis (PCoA) plots of pharyngeal and nasal microbiota, performed by Bray–Curtis and unweighted UniFrac algorithms (C and D, H and I). P-values obtained by the PERMANOVA test are reported. LEfSe cladograms of pharyngeal and nasal samples (E and J). Taxa and nodes highlighted in green and red were significantly more abundant in AH patients and HC, respectively.

Fig 4

Differential network analysis. Pearson’s correlation test performed on the 80 most highly expressed genera in both pharyngeal and nasal microbiota, for AH and HC. The bacterial taxa resulted from the comparison of pharyngeal and nasal profiles by Fisher’s z-test (FDR corrected P-value ≥0.05). Nodes represent taxa, color edges identify connected components, and edge thickness indicates the rho absolute value.

Fig 5

Pharyngeal (upper panel) and nasal (lower panel) microbial ecosystems of AH compared with HC. Alpha-diversity box plots of pharyngeal (A and B) and nasal (F and G) microbiota based on Shannon and Chao1 indexes. P-values obtained by the Mann–Whitney test are reported. Beta-diversity PCoA plots of pharyngeal (C and D) and nasal (H and I) microbiota, performed by Bray–Curtis and unweighted UniFrac algorithms. P-values obtained by the PERMANOVA test are reported. LEfSe cladograms of pharyngeal (E) and nasal (J) samples. Taxa and nodes highlighted in green and red were significantly more abundant in AH patients and HC, respectively.

Fig 6

Probiotic effects on pharyngeal and nasal microbiota. Alpha-diversity box plots of pharyngeal (A and B) and nasal (G and H) microbiota based on Shannon and Chao1 indexes of probiotic-treated (ORO) and placebo (PLB) groups and before (T₀) and after (T₁) treatments. P-values obtained by the Mann–Whitney test are reported. Beta-diversity PCoA plots of pharyngeal (C and D) and nasal (I and J) microbiota, performed by Bray–Curtis and unweighted UniFrac algorithms of ORO and PLB groups and before (T₀) and after (T₁) treatment. The PERMANOVA test was applied to beta-diversity matrices. LEfSe cladograms of pharyngeal (E and F) and nasal (K and L) samples of ORO and PLB groups, respectively. Taxa and nodes highlighted in red and green were significantly more abundant before (T₀) and after (T₁) treatments, respectively.

Fig 7

PICRUSt2 functional prediction. Predicted metabolic pathways statistically associated with before (T₀) and after (T₁) probiotic treatment in pharyngeal (A) and nasal (B) microbiota. Red bars represent pathways increased at T₀; green bars represent pathways increased at T₁.

Fig 8

Treatment effects on pharyngeal and nasal pathobionts. Histograms of the pathobiont growth differences after treatment administration in the pharynx (A) and in the nose (B). In the horizontal axis are reported absolute values of growth differences between T₀ and T₁ in colony-forming units per milliliter.

Fig 9

Treatment effects on pharyngeal and nasal bacteria in respect to AH grading. (Upper panel) Histograms of CFU of bacteria resulted in the statistically significant grouping of patients by AH grading. In y-axis, CFU; in x-axis, AH grading. P-values were calculated by the Wilcoxon test. (Lower panel) LEfSe cladograms of pharyngeal and nasal microbiota obtained by grouping patients on the basis of AH grading.