Alterations of Gut Microbiome Composition and Function Pre- and Post-Adenotonsillectomy in Children with Obstructive Sleep Apnea

Abstract

Importance

Obstructive sleep apnea (OSA) in children is linked with alterations in the gut microbiome. The influence of adenotonsillectomy (AT), a primary intervention for OSA, on gut microbiota dynamics relative to disease severity remains to be elucidated.

Objective

This study aimed to investigate the impact of OSA severity and AT on the gut microbiome in pediatric patients.

Design

A prospective observational study.

Setting

Tertiary referral center.

Participants

A cohort of 55 pediatric patients treated with AT for OSA.

Intervention

Total tonsillectomy and adenoidectomy procedures.

Main Outcome Measures

Comprehensive evaluations included in-laboratory polysomnography and 16S rRNA gut microbiome profiling at baseline, and again at 3rd and 12thmonth following surgery.

Results

Initial findings showed uniform α-diversity across different severities of OSA, while β-diversity was significantly elevated in the severe OSA subgroup. Certain gut microbiota taxa (Lachnospiraceae NK4A136 group, Ruminococcaceae UCG-002, Ruminococcaceae UCG-014, Alloprevotella, Christensenellaceae R-7 group, Ruminococcaceae UCG-005, Lactobacillus murinus, and Prevotella nigrescens) were found to inversely correlate with the apnea-hypopnea index (AHI). Significant post-AT improvements in AHI and other polysomnographic metrics were observed. Notably, AHI changes post-AT were positively associated with microbial α-diversity (species richness), β-diversity, and specific bacterial taxa (Enterobacter, Parasutterella, Akkermansia, Roseburia, and Bacteroides plebeius DSM 17135), but negatively with other taxa (Fusicatenibacter, Bifidobacterium, UBA1819, Ruminococcus gnavus group, Bifidobacterium longum subsp. Longum, and Parabacteroides distasonis) and specific metabolic pathways (purine metabolism, transcription factors, and type II diabetes mellitus). The postoperative patterns of α- and β-diversity mirrored baseline values.

Conclusions and Relevance

This study documents significant changes in the gut microbiome of pediatric patients after AT, including variations in α- and β-diversities, bacterial communities, and inferred metabolic functions. These changes suggest a potential association between the surgical intervention and microbiome alterations, although further studies are necessary to discern the specific contributions of AT amidst possible confounding factors such as antibiotic use.

Graphical abstract

Keywords

adenotonsillectomy apnea-hypopnea index functional prediction gut microbiome obstructive sleep apnea

Background

Obstructive sleep apnea (OSA), afflicting 1% to 5% of children globally,¹ primarily arises from adenotonsillar hypertrophy and obesity.² OSA manifests as snoring, hypopnea, and apnea due to upper airway collapse, leading to a cascade of consequences: hypoxia, hypercapnia, autonomic fluctuations, sleep fragmentation, and notably, gut dysbiosis.^3,4 OSA is implicated in a range of adverse outcomes, including neurocognitive deficits, systemic inflammation, metabolic anomalies, hypertension, cardiovascular complications, and behavioral disorders.^2,5,6

Beyond its fundamental roles in digestion and immunity, the gut is often acclaimed as a “second brain,”⁷ a concept derived from its complex bidirectional communication with the brain through immune signaling, tryptophan metabolism, the vagus nerve, and the enteric nervous system.⁸ Alterations in the gut microbiota are associated with various childhood disorders including anxiety, autism, attention deficit hyperactivity disorder, obesity, and OSA.^4,7,9 Dysbiosis in the gut microbiome can manifest in 3 principal forms: a reduction in overall microbial diversity, the loss of beneficial microbes, and the overgrowth of potentially-harmful microbes.¹⁰ Specifically, in children with OSA, increases in Faecalibacterium and Roseburia and decreases in certain Ruminococcaceae members have been observed in comparison with healthy counterparts.⁴ Conversely, adult patients with OSA have shown elevated levels of Bifidobacterium but reductions in Faecalibacterium, Megamonas, Ruminococcaceae, Clostridiales, and Alistipes relative to individuals without OSA.¹¹ These findings highlight the critical need to further explore the roles, mechanisms, and therapeutic potential of the gut microbiome,¹² particularly in pediatric contexts, to better understand its broad implications for health management and its long-term effects on disease.

Adenotonsillectomy (AT) is a widely-recognized treatment for pediatric OSA, known to reduce symptoms such as systemic inflammation, autonomic dysfunction, and hypertension.^13-15 This surgery is typically performed under general anesthesia, with airway security achieved via endotracheal tube intubation or laryngeal mask airway.¹⁶ Despite its benefits, the effects of AT on gut microbiota dysbiosis are less understood, representing a growing field of study. Our research aimed to explore the diversity of the gut microbiota, both within individuals (α-diversity) and across individuals (β-diversity), investigating gut microbial taxonomy and predictive function relative to OSA severity, and assess changes post-AT. This study seeks to deepen our understanding of the intricate links between pediatric OSA and gut microbiome dynamics.

Methods

Participants

This study was a prospective observational study involving 66 pediatric patients preoperative (preOP) diagnosed with OSA who underwent surgical treatment at the Department of Otorhinolaryngology—Head and Neck Surgery, Linkou Main Branch, Chang Gung Memorial Hospital, from March 2017 to January 2019.¹⁷ Eligible participants were children aged 5 to 12 years with an apnea-hypopnea index (AHI) of ≥5.0 events/hour or an AHI of ≥2.0 events/hour accompanied by conditions such as elevated blood pressure, daytime sleepiness, learning difficulties, growth failure, or enuresis. Exclusions were made for specific medical conditions such as craniofacial, neuromuscular, or chronic inflammatory disorders, recent acute infection or inflammatory episodes, and chronic or acute gastroenteritis.¹⁷ To avoid the influence of antibiotics on microbiome analysis and inflammatory marker examinations, stool collection was postponed for at least 2 weeks after any acute episode requiring antibiotic treatment.^17,18 The dietary profile was evaluated using the Short Food Frequency Questionnaire.¹⁹ Follow-up assessments for polysomnography, dietary profiling, and gut microbiome analysis were conducted at 3rd month (postOP3) and 12th month (postOP12) postsurgery. Details of the adenotonsillectomy procedure were documented separately.²⁰

Polysomnography

The study assessed various factors indicative of OSA severity, including AHI, apnea index (AI), mean peripheral blood oxygen (SpO₂), and minimum SpO₂. The in-lab polysomnography procedures adhered to the guidelines of the 2012 American Academy of Sleep Medicine Manual.²¹ Disease severity was classified as “mild” for AHI ≥2 and <5 events/hour, “moderate” for AHI ≥5 and <10 events/hour, and “severe” for AHI ≥10 events/hour. An outcome was considered “cured” if the obstructive AHI was below 2.0 events/hour and the obstructive AI was below 1.0 events/hour.²²

16S rRNA Amplicon Sequencing of Gut Microbiome

Parents collected stool samples from participants in the morning, immediately snap-freezing them before storage at −80°C pending analysis. Genomic DNA extraction from these samples employed a fecal DNA isolation kit (Mo Bio Laboratories, Carlsbad, CA, USA), with DNA quantity and quality subsequently verified using a NanoPhotometer P360 (Implen Gmbh, Munich, Germany).

Amplification targeted the V3-V4 regions of the bacterial 16S rRNA gene using composite primers (341F: 5’-CCTACGGGNGGCWGCAG-3’ and 806R: 5’-GACTACHVGGGTATCTAATCC-3’), in line with the strategy that leveraging these hypervariable regions could improve resolution.²³ Adhering to Illumina’s 16S Metagenomic Sequencing Library Preparation guidelines, PCR employed 12.5 ng of the extracted DNA as a template in conjunction with KAPA HiFi HotStart ReadyMix (Roche Diagnostics Corporation, Indianapolis, IN, USA), under specific conditions: initial denaturation at 95°C for 3 minutes, followed by 25 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds, extension at 72°C for 30 seconds, and a final extension at 72°C for 5 minutes, before holding at 4°C. The integrity of PCR products was verified on a 1.5% agarose gel, selecting for samples with a distinct primary band approximately 500 base pairs in length. Following purification with AMPure XP beads, a secondary PCR introduced Nextera XT Index Kit elements, including dual indices and Illumina sequencing adapters. Indexed PCR product quality was assessed using a Qubit 4.0 Fluorometer (Thermo Fisher Scientific Inc., Waltham, MA USA) and the Qsep100™ system, with equilibrated PCR products pooled to create the sequencing library. This library underwent sequencing on the Illumina MiSeq platform, generating paired 300-base pair reads.

For data processing, paired-end reads were assembled with FLASH (version 1.2.11). Reads with a quality score below 20 were removed using the QIIME (version 1.9.1) pipeline. The sequences that remained were scrutinized for chimeras with UCHIME. Operational taxonomic units (OTUs) were delineated at 97% sequence similarity with USEARCH (version 7.0.1090),²⁴ with each OTU represented by a selected sequence. This representative sequence was annotated, with the annotation extended to all sequences within the OTU. Taxonomy annotation was carried out with the Silva database (Release 123),²⁵ while PyNAST (version 1.2) aligned multiple sequences against the Silva core set database.

Outcome Variables

The study focused on 2 primary outcome variables: the severity of OSA, quantified by the AHI, and the diversity of the gut microbiota, assessed through α-diversity and β-diversity measures. α-diversity metrics, including Chao 1 and observed species for species richness and Shannon and Simpson indices for overall community diversity, were derived from OTU and genera frequency in sequencing data.²⁶ β-diversity was assessed through Bray-Curtis dissimilarity, capturing variations in community composition across samples.²⁷ Secondary outcomes included body mass index (BMI), additional polysomnographic parameters, and various aspects of the gut microbiome, specifically its taxonomy and functional characteristics.

Statistical Analysis

Statistical analysis was conducted recognizing the non-normal distribution of most variables, as verified by the Shapiro-Wilk test. Consequently, nonparametric methods were applied throughout. Between-group differences for continuous variables were assessed using the Mann-Whitney U test or the Kruskal-Wallis test, supplemented by post hoc adjustments (Bonferroni correction or least significant difference) where necessary. Categorical variables were compared using the chi-squared test. Within-group changes for continuous variables were evaluated using the related-samples Wilcoxon signed-rank test, while categorical variables were analyzed with the McNemar test.

The UpSet method presented unique intersections of OTUs in a matrix format based on subgroups and time points. Bacterial community differences were assessed using the analysis of similarity. Based on 16S rRNA datasets and the Kyoto Encyclopedia of Genes and Genomes database,²⁸ microbial community functional capabilities were predicted using Tax4Fun.²⁹ Discrepancies in taxa and metagenome function predictions were evaluated using the linear discriminant analysis (LDA) effect size (LEfSe) method, incorporating the Benjamini-Hochberg procedure to manage the false discovery rate. This procedure also adjusted for multiple comparisons, with statistical significance set at 2-tailed P < .05 or q < 0.05, as appropriate.

To understand the relationship between AHI and gut microbiome diversity, both preOP and postOP, utilizing Spearman correlation and repeated measures correlation analyses, which gauged the within-individual association for paired measures taken on multiple occasions.³⁰ The significance threshold was set at a 2-tailed P < .05. All statistical analyses were carried out using the R software (version 4.3.1) and the microeco package (v1.8.0),³¹ supplemented by GraphPad Prism version 10 (GraphPad Software, Inc, Boston, MA, USA).

Results

The study commenced with 66 pediatric participants, with the final analysis including 55 children, representing 83% of the initially-assessed group (Supplemental Figure S1). This cohort consisted of 13 girls (24%) and 42 boys (76%), with a median age of 7 years (interquartile range: 6-10 years).

PreOP assessments showed significant differences in AHI, AI, mean SpO₂, and minimum SpO₂ across different OSA severity levels (Table 1). No significant differences were observed in age, sex, BMI, obesity rates, or food frequency (Supplemental Table S1) among these groups.

Table 1.

Clinical Characteristics of Children with Obstructive Sleep Apnea Before and After Adenotonsillectomy.

Characteristics	PreOP severity of obstructive sleep apnea			PreOP	PostOP3	PostOP12	PostOP12 outcome
	Mild	Moderate	Severe	PreOP	PostOP3	PostOP12	Cured	Uncured
	(n = 19)	(n = 10)	(n = 26)	(n = 55)	(n = 55)	(n = 55)	(n = 28)	(n = 27)
Age (y)	6 (6-10)	7 (6-10)	8 (5-9)	7 (6-10)^e,f	7 (6-10)^e,g	8 (7-11)^f,g	7 (6-8)^d	10 (8-11)^d
Sex (girl/boy)	5/14	1/9	7/19	13/42	13/42	13/42	10/18	3/27
BMI (kg/m²)	15.7 (15.0-19.9)	19.1 (13.6-22.1)	19.5 (15.9-25.2)	18.1 (15.3-23.0)	18.3 (15.8-24.1)	20.4 (16.0-25.1)	18.0 (15.2-23.5)	21.2 (17.2-26.3)
Obesity (%)^h	32%	60%	62%	51%	44%	47%	43%	56%
AHI (events/h)	4.1 (3.0-4.5)^a,b	6.9 (5.8-9.4)^a,c	23.6 (17.4-36.3)^b,c	9.6 (4.4-22.8)^e,f	2.4 (1.4-3.7)^e,g	2.9 (1.8-5.5)^f,g	1.9 (1.2-2.4)^d	5.7 (3.6-9.0)^d
AI (events/h)	1.1 (0.2-1.9)^b	1.5 (0.6-4.4)	3.4 (1.7-12.5)^b	1.8 (0.8-4.4)^e,f	0.8 (0.3-1.7)^e	1.1 (0.2-1.6)^f	0.7 (0.2-1.5)	1.2 (0.4-1.8)
Mean SpO₂ (%)	98 (97-98)^b	98 (97-98)	97 (96-97)^b	97 (97-98)^e,f	98 (98-98)^e	98 (97-98)^f	98 (97-98)	98 (97-98)
Minimum SpO₂ (%)	92 (91-94)^b	90 (88-91)	84 (82-89)^b	89 (84-92)^e,f	92 (89-94)^e	91 (89-94)^f	93 (91-94)^d	90 (87-91)^d

Data are expressed as median (interquartile range) or number (%). Between-group comparisons used the Kruskal-Wallis test with Bonferroni correction for continuous variables or the chi-squared test for categorical variables. Within-group comparisons utilized the related-samples Wilcoxon signed-rank test for continuous variables.

Abbreviations: AHI, apnea-hypopnea index; AI, apnea index; BMI, body mass index; postOP3, 3 months postoperation; postOP12, 12 months postoperation; preOP, preoperative; SpO₂, pulse oxygen saturation.

Significance at P < .05: preOP mild versus moderate OSA.

Significance at P < .05: preOP mild versus severe OSA.

Significance at P < .05: preOP moderate versus severe OSA.

Significance at P < .05: postOP12 cured versus uncured.

Significance at P < .05: preOP versus postOP3.

Significance at P < .05: preOP versus postOP12.

Significance at P < .05: postOP3 versus postOP12.

Obesity was defined by BMI z-score ≥1.645.

Post-AT evaluations revealed a substantial decrease in AHI at both 3rd and 12th month, with postOP12 AHI values higher than those at postOP3. AI, mean SpO₂, and minimum SpO₂ also improved significantly at postOP3, with stabilization by postOP12. There were no significant changes in BMI or obesity rates following AT. The frequency of full-fat milk consumption increased significantly at postOP3 but reverted to preoperative levels by 12 months.

The cure rates at 3rd and 12th month post-AT were 64% and 51%, respectively. Participants who were considered cured were notably younger, had lower AHI values, and higher minimum SpO₂ levels than those who did not achieve cure status. Other polysomnographic parameters and food frequency metrics did not show significant changes over time.

The sequencing depth was deemed adequate as indicated by the plateau in the rarefaction curves (Supplemental Figure S2). α-diversity indices did not show significant differences across OSA severities (Figure 1A). β-diversity assessment demonstrated a closer clustering within the mild OSA subgroup than within the severe subgroup (P < .001) (Figure 2A).

Figure 1.

α-diversity metrics of gut microbiota. (A) preOP α-diversity metrics are similar across mild, moderate, and severe obstructive sleep apnea severity subgroups. (B) A noticeable decline in bacterial richness at 3rd month (postOP3) and 12th month (postOP12) postoperative compared with preOP levels is depicted by the Chao1 and observed species indices. (C) PostOP12 α-diversity metrics do not differ significantly between the cured and uncured subgroups. The data are presented as median values with interquartile ranges. ***P < .001. preOP, preoperative; postOP3, 3 months postoperation; postOP12, 12 months postoperation.

Figure 2.

β-diversity metrics of gut microbiota. (A) Box plot and principal coordinates analysis using the Bray-Curtis distance at preOP reveal less variation within the mild obstructive sleep apnea subgroup compared with the severe subgroup. (B) β-diversity assessments show significant decreases in variability over time from preOP to 3 months (postOP3) and 12 months (postOP12) postoperative periods. (C) At postOP12, box plot and principal coordinates analysis indicate reduced variability in the cured subgroup relative to the uncured subgroup. The data are presented as median values with interquartile ranges. *P <.05 and ≥.01; ***P < .001. preOP, preoperative; postOP3, 3 months postoperation; postOP12, 12 months postoperation.

LEfSe analysis with an LDA score threshold of 3 identified 8 unique marker microbes distinctly related to varying OSA severities, with the mild subgroup displaying 4 specific microbes and the moderate subgroup presenting a different set of 4 (Supplemental Figure S3). Of the 24 most prevalent genera examined, a subset revealed significant correlations with clinical metrics. Notably, 6 genera, specifically the Lachnospiraceae NK4A136 group, 3 Ruminococcaceae members (UCG-002, UCG-005, and UCG-014), Alloprevotella, and the Christensenellaceae R-7 group, were significantly associated with the AHI (Figure 3A). Additionally, of the 5 most prominent species examined, 3 including Lactobacillus murinus and Prevotella nigrescens, exhibited notable correlations with clinical metrics, particularly with the AHI (Figure 3B). While Tax4Fun analysis revealed that LEfSe testing with a LDA score of 1 did not find differences in functional pathways among the 3 OSA severity groups, it did identify disparities in 11 functional pathways contrasting the mild subgroup with the moderate/severe subgroups in preOP circumstances (Supplemental Figure S4A).

Figure 3.

Association of microbial taxa with clinical metrics preadenotonsillectomy. (A) A hierarchical clustering heatmap represents the correlations between dominant genera and various clinical metrics based on Spearman correlation analyses, with statistically significant associations marked. (B) Another heatmap illustrates the relationships between key microbial species and clinical parameters, highlighting the complexity of their interactions. The color intensity indicates the strength and direction of the correlation. Significance levels are denoted as follows: *P < .05; **P < .01.

While Chao1 and observed species indices for α-diversity showed a significant decline from preOP to postOP3 and postOP12, Shannon and Simpson indices indicated no significant variation over these time points (Figure 1B). β-diversity analysis revealed dynamic changes in the bacterial community structure throughout the different stages (Figure 2B).

LEfSe analysis, using a LDA score threshold of 3, identified 62 unique marker microbes across 3 distinct time points (Figure 4). Repeated measures correlation analyses further revealed that post-AT decreases in the AHI were linked to reduced relative abundances of Enterobacter, Parasutterella, Akkermansia, Roseburia, and Bacteroides plebeius DSM 17135, while there were increases in Fusicatenibacter, Bifidobacterium, UBA1819, Ruminococcus gnavus group, Bifidobacterium longum subsp. longum, and Parabacteroides distasonis, indicating significant changes in the composition of the gut microbiota (Supplemental Figure S5; Table 2). Furthermore, analysis of the functional pathways of the gut microbiota showed 46 distinct shifts post-AT, which varied across different time points (Supplemental Figure S4B). An inverse relationship emerged between AHI and pathways involved in purine metabolism, transcription factors, and type II diabetes mellitus, as demonstrated in the repeated measures correlation analysis (Figure 5), suggesting these pathways may have clinical significance in the context of OSA.

Figure 4.

Linear discriminant analysis (LDA) together with effect size measurement (LEfSe) of gut microbiota with an LDA score threshold of 3. A cladogram identifies 62 taxa with differential abundances across the time points of preOP, 3 months postoperative (postOP3), and 12 months postoperative (postOP12). This included 28 marker microbes dominant at the preOP stage, 12 microbes at postOP3, and 22 microbes at postOP12. preOP, preoperative; postOP3, 3 months postoperation; postOP12, 12 months postoperation.

Table 2.

Associations Between Apnea-Hypopnea Index and Gut Microbiome Composition in Children with Obstructive Sleep Apnea.

Variables	Repeated measures correlation analysis		Potential role in host health	Evidence in pathophysiology or sleep-related disorder
Variables	Coefficient	P-value	Potential role in host health	Evidence in pathophysiology or sleep-related disorder
Enterobacter	0.22	.02	Harmful; opportunistic pathogen⁴⁶	Increased by sleep deprivation⁵³
Parasutterella	0.43	<.001	Harmful; linked to Crohn’s disease⁴⁷	Increased by sleep deprivation⁵³
Akkermansia	0.23	.01	Beneficial; guards against metabolic disorders⁴⁸	Influenced by chronic intermittent hypoxia and sleep fragmentation⁵¹
Roseburia	0.35	<.001	Beneficial; protects against metabolic disturbance⁴⁹	Abundance increased by insomnia⁵²
Bacteroides plebeius DSM 17135	0.20	.04	Beneficial; protects against muscle atrophy in chronic kidney disease⁵⁰	Linked to autism behaviors⁵⁴
Fusicatenibacter	−0.22	.02	Beneficial; its decrease associated with Parkinson’s progression⁵⁵	Abundance increased by mild intermittent hypoxia⁶⁰
Bifidobacterium	−0.26	<.01	Beneficial; improves gut barrier and lowers lipopolysaccharide levels⁵⁶	Decrease by chronic hypoxia⁶⁰
UBA1819	−0.22	.02	Beneficial; mitigates hepatic inflammation⁵⁷	Reduced in obesity and increased carotid thickness⁶¹
Ruminococcus gnavus group	−0.22	.02	Harmful; involved in ulcerative colitis⁶⁴	Reduction linked to less sleep and fatty acid synthesis⁶⁵
Bifidobacterium longum subsp. longum	−0.19	.04	Beneficial; boosts hepatic antioxidant capacity and gut balance⁵⁸	Enhances sleep and reduces stress⁶²
Parabacteroides distasonis	−0.20	.04	Beneficial; produces compounds that ameliorate obesity and metabolic dysfunction⁵⁹	Improves sleep after disturbances⁶³

Figure 5.

Repeated measures correlation analyses of the AHI with specific functional pathways: (A) purine metabolism, (B) transcription factors, and (C) type II diabetes mellitus pathways. This figure illustrates intra-individual correlations between AHI and specific functional pathway predictions across 3 evaluation points for multiple participants. AHI, apnea-hypopnea index.

Additionally, α-diversity indices displayed no significant variations between the 2 subgroups (Figure 1C). β-diversity analysis showed a more homogenous bacterial community in the cured subgroup compared to the uncured subgroup (Figure 2C). Using an LDA score threshold of 3, LEfSe analysis identified 4 marker microbes distinctively associated with the postOP12 cure status, with Lactococcus singularly characterizing the cured subgroup and 3 other microbes (Clostridiaceae-1, Sutterella, Clostridium sensu stricto-1) marking the uncured subgroup (Supplemental Figure S6). Additionally, while an LDA score threshold of 1 in LEfSe analysis did not reveal differential functional pathways between the cured and uncured subgroups, 2 pathways—vascular endothelial growth factor signaling pathway and pantothenate and coenzyme A biosynthesis—were distinctively associated with the mild subgroup compared with the moderate/severe subgroup postOP12 (Supplemental Figure S4C). However, these pathways were not significantly related to AHI using the repeated measures correlation analysis.

Discussion

In the current investigation, we meticulously probed the nuances of OSA patterns in pediatric cohorts, delineating the effects of AT on polysomnography-based sleep metrics and concurrent shifts in the gut microbiome. Initial preOP analyses illuminated distinct disparities in polysomnographic metrics, stratified by OSA severity. Sequential postOP3 and postOP12 evaluations manifested significant ameliorations, albeit a notable resurgence in AHI between these checkpoints was observed, mirroring findings from preceding research.³² Our findings of varying cure rates align with geographic differences reported in other studies, with higher recovery in US cohorts³³ and lower in Japanese groups.³⁴ Additionally, outcomes were better in younger children,³⁵ though only a minority of our participants were under 7. These results highlight the influence of age, demographic factors, and follow-up durations on treatment outcomes, underscoring the need for ongoing research.

Considering the potential effects of dietary habits and obesity on the gut microbiome, our study highlights that dietary habits and obesity rates remained consistent before and 12 months after surgery across different OSA severity groups, as reported previously.³⁶ This consistency suggests that the observed changes in the gut microbiome are likely due to shifts in OSA severity rather than alterations in dietary intake or obesity status.

Our pediatric OSA study presents stable α-diversity across all severities of the condition, aligning with past research,^3,11 and indicates a trend of greater β -diversity with increasing OSA severity. In our cohort, children showed higher Shannon index when compared to healthy control benchmarks, derived from analysis of the V3-V4 regions of the bacterial 16S rRNA gene.³⁷ This work also uncovers a notable correlation between postoperative changes in the AHI and variations in gut microbiota, which diverges from some literature suggesting reduced α-diversity in OSA-afflicted individuals compared with their healthy counterparts.^38,39 The longitudinal nature of this study, highlighting time-linked changes and dose-response effects, underscores the need for a nuanced exploration of the relationship between gut microbiota diversity and the severity of OSA.

While the overall microbial diversity appeared consistent across OSA severity gradients, our investigation revealed distinct variations in bacterial compositions at both genus and species levels. Consolidating existing literature, our findings validated known bacterial links with OSA and spotlighted novel microbial players possibly influencing the disease process. For instance, the attenuated levels of Lachnospiraceae NK4A136 group and Alloprevotella, previously reported in patients with OSA ,^39,40 might exacerbate cognitive deficits via perturbations in gut environment or blood-brain barrier dynamics.^40,41 Likewise, diminished levels of bacterial taxa like Ruminococcaceae UCG-002 might establish a nexus between chronic insomnia and cardiometabolic diseases.⁴² Likewise, reductions in Ruminococcaceae UCG-014, Christensenellaceae R-7 group, Ruminococcaceae UCG-005, and Lactobacillus murinus under hypoxic conditions could compromise serum immunity, attenuate inflammatory reactions, and impact antioxidant capacities, digestive function, and lipid metabolism, as suggested by animal research.^43,44 In contrast, a decline in Prevotella nigrescens might temper T-helper type 17-mediated mucosal inflammations, which are frequently associated with metabolic aberrations and generalized inflammatory states.⁴⁵ Thus, these particular bacterial taxa may play a pivotal role in the clinical manifestations and complications associated with OSA.

The correlations between postOP changes in the AHI, AT, and alterations in gut microbiota composition provide valuable insights into the complex interactions among OSA, therapeutic interventions, and gut health. The association of AHI reduction post-AT with a decrease in certain potentially-harmful bacterial taxa, such as Enterobacter and Parasutterella, suggests the beneficial impacts of AT on gut microbiota (Table 2).^46,47 However, the reduction in relative abundances of beneficial taxa such as Akkermansia, Roseburia, and Bacteroides plebeius DSM 17135 postsurgery raises questions about the full spectrum of AT’s effects on the gut microbiome.^48-50 Considering Akkermansia’s increase in animal models of OSA subjected to chronic intermittent hypoxia and sleep fragmentation,⁵¹ its postsurgical decrease may reflect improved gut health and potentially a mitigation of OSA severity. The links of bacterial taxa such as Roseburia with insomnia,⁵² Enterobacter, and Parasutterella with acute sleep deprivation,⁵³ and Bacteroides plebeius DSM 17135 with autism-related behavioral manifestations⁵⁴ further expand our understanding of the gut-brain axis, emphasizing the need for more comprehensive research in this domain.

On the other hand, observed negative correlation between reduced AHI levels and the proliferation of beneficial bacterial taxa such as Fusicatenibacter, Bifidobacterium, UBA1819, Bifidobacterium longum subsp. longum, and Parabacteroides distasonis highlights the potential positive impacts of treating OSA on gut health.^55-59 This aligns with findings from animal OSA models that showed a decline in Bifidobacterium levels with chronic intermittent hypoxia exposure.⁶⁰ Additionally, lower levels of UBA1819 have been associated with obesity risks and increased carotid intima-media thickness, both of which are potential comorbidities of OSA in pediatric populations.⁶¹ Dietary interventions, such as supplementation with Bifidobacterium longum subsp. longum, have been shown to improve sleep quality and reduce psychological stress in healthy subjects,⁶² while diets enriched in prebiotics that boost Parabacteroides distasonis levels have been linked to better sleep parameters following sleep disturbances.⁶³

Conversely, an increase in the potentially-harmful taxon Ruminococcus gnavus group tempers the overall beneficial effect of AT on gut health,⁶⁴ emphasizing the nuanced and complex impact of AT on the gut microbiota. Interestingly, decreased concentrations of Ruminococcus gnavus group were associated with shorter sleep durations and reduced short-chain fatty acid production.⁶⁵ Furthermore, the finding that overweight or obese men exhibit elevated Fusicatenibacter levels under conditions of mild intermittent hypoxia suggests the influence of body weight and potentially other lifestyle factors on gut health.⁶⁰ These findings highlight adenotonsillectomy’s role not only in reducing OSA severity but also in altering the gut microbiota, potentially affecting behavior and weight status in pediatric patients. Emerging evidence supports the potential benefits of pre- and probiotic supplementation for ameliorating OSA-induced dysbiosis, as observed in animal studies.⁶⁶ Furthermore, microbiota transfer therapy, which has shown promise in managing symptoms of autism spectrum disorder,⁶⁷ could potentially be adapted to mitigate behavioral symptoms associated with OSA. Thus, targeting the gut microbiome presents a promising avenue for comprehensive OSA management, meriting further research.

The observed upregulation of predictive functions related to purine metabolism, transcription factors, and type II diabetes mellitus pathways in children with higher OSA severity preOP, and their persistence post-AT despite reduced AHI, implies a fundamental connection with OSA. These pathways remained consistently regulated across various OSA severities at postOP12, indicating a potential inherent association with OSA that persists beyond surgical intervention. Possible explanations for this phenomenon include an adaptive microbiome response to surgery,⁶⁸ microbial resilience,⁶⁹ or underlying genetic or metabolic predispositions.⁷⁰ However, these findings are based on predictive analysis from Tax4Fun, not direct metabolomic data, which could provide a more accurate representation of microbiome function.⁷¹ Future research should consider direct metabolomic studies to better understand these relationships.

Moreover, the potential long-term consequences of these functional changes, particularly their contribution to predispositions toward conditions such as type II diabetes mellitus, necessitate further investigation. For instance, data from the European Sleep Apnea Database revealing a significantly-lower prevalence of type II diabetes mellitus in OSA patients who underwent tonsillectomy compared with those who did not⁷² highlight the need for a more nuanced understanding of how treatments for OSA can impact broader health outcomes beyond the immediate effects on sleep and respiratory function.

Our study has notable limitations that warrant consideration. First, the absence of a healthy control group limits our ability to distinguish the specific effects of AT on the gut microbiome from normal variations. Second, the small sample size constrains the generalizability of our findings. Future research, ideally incorporating a larger and more diverse cohort along with a case-control study design, is crucial to validate and expand these initial results. Moreover, our data collection on antibiotic and medication uses both pre- and postsurgery was incomplete, which introduces significant confounding variables given their potential to significantly influence microbiome dynamics.⁷³ While our protocol included a standard 5-day postoperative antibiotic regimen, guidelines and recent studies suggest that extending the delay in collecting stool samples to at least 1-month postantibiotic treatment could better ensure microbiome stability.^73,74 Despite our efforts to mitigate this by delaying stool collection for 2 weeks postantibiotic use, this remains a significant limitation. Additionally, the study’s focus solely on a pediatric Taiwanese population may not capture variations that could emerge in different ethnic or age groups, potentially affecting the gut microbiome differently. These factors underscore the need for more comprehensive future studies to confirm and expand upon our findings. Moreover, emerging applications of artificial intelligence in pediatric OSA treatment may offer novel insights and improvements, including in the area of gut microbiome management.⁷⁵

Conclusion

Our study elucidates the complex interplay between pediatric OSA, AT, and gut microbiome alterations. The findings from this research stress the significance of adopting a comprehensive perspective in the treatment of OSA, showcasing the promising potential of gut microbiome-targeted interventions as a novel strategy for addressing pediatric OSA and associated metabolic conditions.

Supplemental Material

sj-docx-1-ohn-10.1177_19160216241293070 – Supplemental material for Alterations of Gut Microbiome Composition and Function Pre- and Post-Adenotonsillectomy in Children with Obstructive Sleep Apnea

Supplemental material, sj-docx-1-ohn-10.1177_19160216241293070 for Alterations of Gut Microbiome Composition and Function Pre- and Post-Adenotonsillectomy in Children with Obstructive Sleep Apnea by Hai-Hua Chuang, Li-Ang Lee, Li-Pang Chuang, Hsueh-Yu Li, Yu-Shu Huang, Shih-Hsuan Chou, Guo-She Lee, Terry B. J. Kuo, Cheryl C. H. Yang and Chung-Guei Huang in Journal of Otolaryngology - Head & Neck Surgery

Footnotes

Acknowledgements

We thank Ruo-Chi Wang and Chung-Fang Hsiao (Department of Otorhinolaryngology, Head and Neck Surgery, Chang Gung Memorial Hospital, Linkou Main Branch, Taoyuan City, Taiwan, ROC) for their technical assistance. Dr. Chung-Guei Huang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Authors’ Note

This paper was presented at the Annual Meeting of the American Academy of Otolaryngology—Head and Neck Surgery, Miami, FL, September 28 to October 1, 2024.

Author Contributions

H.H.C.: conception and design, acquisition of data, analysis and interpretation of data, drafting the article. L.A.L. and C.G.H.: conception and design, acquisition of data, analysis and interpretation of data, drafting the article, revising the manuscript critically for important intellectual content. L.P.C. and H.Y.L., T.B.J.K., and C.C.H.Y.: conception and design, and revising the manuscript critically for important intellectual content. Y.S.H. and G.S.L.: revising the manuscript critically for important intellectual content. S.H.C.: conception and design, analysis and interpretation of data, revising it critically for important intellectual content. All authors read and approved the final manuscript.

Availability of Data and Materials

Data will be made available upon reasonable request to the corresponding author.

Consent for Publication

Not applicable.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Science and Technology Council, Taiwan [grant number 109-2314-B-182-083-MY3 (to LAL)] and the Chang Gung Medical Foundation, Taiwan [grant numbers CMRPG3F1091-3 (to LAL), CORPG3K0242, CORPG3L0481 (to CGH)].

Ethics Approval and Consent to Participate

This prospective observation study was approved by the Institutional Review Board of Chang Gung Medical Foundation (201507279A3). Parents, as well as participants aged 6 and above, provided written informed consent.

ORCID iDs

Guo-She Lee

Chung-Guei Huang

Supplemental Material

Additional supporting information is available in the online version of the article.

References

Marcus

Brooks

Draper

, et al Diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics. 2012;130(3):e714-e755.

Lee

LA.

Sleep-disordered breathing in children. Chang Gung Med J. 2009;32(3):247-257.

Zhang

Wang

Guan

Yin

Metabolomics and microbiome profiling as biomarkers in obstructive sleep apnoea: a comprehensive review. Eur Respir Rev. 2021;30(160):200220.

Cai

Chen

Shen

Lyv

Gut microbiota dysbiosis in 4- to 6-year-old children with obstructive sleep apnea-hypopnea syndrome. Pediatr Pulmonol. 2022;57(9):2012-2022.

Chuang

Hsu

Wang

, et al Hypertension in children with obstructive sleep apnea syndrome-Age, weight status, and disease severity. Int J Environ Res Public Health. 2021;18(18):9602.

Lin

Chiu

, et al Inflammatory biomarkers of neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios in 563 severe OSA patients before and after surgery. J Otolaryngol Head Neck Surg. 2023;52(1):49.

Gershon

MD.

The enteric nervous system: a second brain. Hosp Pract (1995). 1999;34(7):31-32, 35-38, 41-42 passim.

Cryan

O’Riordan

Cowan

CSM

, et al The microbiota-gut-brain axis. Physiol Rev. 2019;99(4):1877-2013.

Checa-Ros

Jerez-Calero

Molina-Carballo

Campoy

Munoz-Hoyos

Current evidence on the role of the gut microbiome in ADHD pathophysiology and therapeutic implications. Nutrients. 2021;13(1):249.

10.

DeGruttola

Low

Mizoguchi

Current understanding of dysbiosis in disease in human and animal models. Inflamm Bowel Dis. 2016;22(5):1137-1150.

11.

Liu

, et al Gut microbiota in obstructive sleep apnea-hypopnea syndrome: disease-related dysbiosis and metabolic comorbidities. Clin Sci (Lond). 2019;133(7):905-917.

12.

Shin

Preidis

Shulman

Kashyap

PC.

The gut microbiome in adult and pediatric functional gastrointestinal disorders. Clin Gastroenterol Hepatol. 2019;17(2):256-274.

13.

Huang

Hsu

Chuang

, et al Adenotonsillectomy-related changes in systemic inflammation among children with obstructive sleep apnea. J Chin Med Assoc. 2023;86(6):596-605.

14.

Lee

Chuang

Hsieh

, et al Using sleep heart rate variability to investigate the sleep quality in children with obstructive sleep apnea. Front Public Health. 2023;11:1103085.

15.

Hsieh

Chuang

Hsin

, et al Effect of preoperative weight status and disease presentation on postoperative elevated blood pressure after childhood adenotonsillectomy. Otolaryngol Head Neck Surg. 2023;168(5):1197-1208.

16.

Khoury

Zabihi-Pour

Davidson

, et al The safety of the laryngeal mask airway in adenotonsillectomy: a systematic review and meta-analysis. J Otolaryngol Head Neck Surg. 2024;53:19160216241263851.

17.

Chuang

Hsu

Chuang

, et al Different associations between tonsil microbiome, chronic tonsillitis, and intermittent hypoxemia among obstructive sleep apnea children of different weight status: a pilot case-control study. J Pers Med. 2021;11(6):486.

18.

Chuang

Huang

Chuang

, et al Relationships among and predictive values of obesity, inflammation markers, and disease severity in pediatric patients with obstructive sleep apnea before and after adenotonsillectomy. J Clin Med. 2020;9(2):579.

19.

Lillegaard

Overby

Andersen

LF.

Evaluation of a short food frequency questionnaire used among Norwegian children. Food Nutr Res. 2012;56:6399.

20.

Lee

Lin

, et al Severity of childhood obstructive sleep apnea and hypertension improved after adenotonsillectomy. Otolaryngol Head Neck Surg. 2015;152(3):553-560.

21.

Berry

Budhiraja

Gottlieb

, et al Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8(5):597-619.

22.

Marcus

Moore

Rosen

, et al A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med. 2013;368(25):2366-2376.

23.

Mukherjee

Beall

Griffen

Leys

EJ.

High-resolution ISR amplicon sequencing reveals personalized oral microbiome. Microbiome. 2018;6(1):153.

24.

Edgar

RC.

UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat Methods. 2013;10(10):996-998.

25.

Quast

Pruesse

Yilmaz

, et al The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41(Database issue):D590-D596.

26.

Schloss

Gevers

Westcott

SL.

Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS One. 2011;6(12):e27310.

27.

Mumby

PJ.

Beta and habitat diversity in marine systems: a new approach to measurement, scaling and interpretation. Oecologia. 2001;128(2):274-280.

28.

Ogata

Goto

Sato

Fujibuchi

Bono

Kanehisa

KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 1999;27(1):29-34.

29.

Asshauer

Wemheuer

Daniel

Meinicke

Tax4Fun: predicting functional profiles from metagenomic 16S rRNA data. Bioinformatics. 2015;31(17):2882-2884.

30.

Bakdash

Marusich

LR.

Repeated measures correlation. Front Psychol. 2017;8:456.

31.

Liu

Cui

Yao

microeco: an R package for data mining in microbial community ecology. FEMS Microbiol Ecol. 2021;97(2).

32.

Huang

Guilleminault

Lee

Lin

Hwang

FM.

Treatment outcomes of adenotonsillectomy for children with obstructive sleep apnea: a prospective longitudinal study. Sleep. 2014;37(1):71-76.

33.

Marcus

Moore

Rosen

, et al A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med. 2013;368(25):2366-2376.

34.

Arima

Koike

Fujinaga

, et al Normalization of breathing with adenotonsillectomy in Japanese pediatric OSA. Auris Nasus Larynx. 2019;46(5):758-763.

35.

Chen

Yin

G-P

Effectiveness and safety of (adeno) tonsillectomy for pediatric obstructive sleep apnea in different age groups: a systematic review and meta-analysis. Sleep Med Rev. 2023;69:101782.

36.

Chuang

Lin

Hsu

, et al Dietary profile of pediatric obstructive sleep apnea patients, effects of routine educational counseling, and predictors for outcomes. Front Public Health. 2023;11:1160647.

37.

Wang

Yang

Chou

, et al Gut microbiota and dietary patterns in children with attention-deficit/hyperactivity disorder. Eur Child Adolesc Psychiatry. 2020;29(3):287-297.

38.

Valentini

Evangelisti

Arpinelli

, et al Gut microbiota composition in children with obstructive sleep apnoea syndrome: a pilot study. Sleep Med. 2020;76:140-147.

39.

Zhang

Chen

Shen

Chen

Sleep apnea is associated with the increase of certain genera of Ruminococcaceae and Lachnospiraceae in the gut microbiome of hypertensive patients. Expert Rev Respir Med. 2022;16(11-12):1247-1256.

40.

Shao

, et al Obstructive sleep apnea is related to alterations in fecal microbiome and impaired intestinal barrier function. Sci Rep. 2023;13(1):778.

41.

Zheng

Wang

, et al A flower-like brain targeted selenium nanocluster lowers the chlorogenic acid dose for ameliorating cognitive impairment in APP/PS1 mice. J Agric Food Chem. 2023;71(6):2883-2897.

42.

Jiang

Zhuo

, et al The gut microbiota-bile acid axis links the positive association between chronic insomnia and cardiometabolic diseases. Nat Commun. 2022;13(1):3002.

43.

Fan

Cui

Wang

, et al Rumen microbiota of tibetan sheep (ovis aries) adaptation to extremely cold season on the Qinghai-Tibetan Plateau. Front Vet Sci. 2021;8:673822.

44.

Zhang

Chen

Wang

, et al Microbiota-derived tryptophan metabolites indole-3-lactic acid is associated with intestinal ischemia/reperfusion injury via positive regulation of YAP and Nrf2. J Transl Med. 2023;21(1):264.

45.

Larsen

JM.

The immune response to Prevotella bacteria in chronic inflammatory disease. Immunology. 2017;151(4):363-374.

46.

Shao

Forster

Tsaliki

, et al Stunted microbiota and opportunistic pathogen colonization in caesarean-section birth. Nature. 2019;574(7776):117-121.

47.

Chiodini

Dowd

Chamberlin

Galandiuk

Davis

Glassing

Microbial population differentials between mucosal and submucosal intestinal tissues in advanced Crohn’s disease of the ileum. PLoS One. 2015;10(7):e0134382.

48.

Hasani

Ebrahimzadeh

Hemmati

Khabbaz

Hasani

Gholizadeh

The role of Akkermansia muciniphila in obesity, diabetes and atherosclerosis. J Med Microbiol. 2021;70(10).

49.

Neyrinck

Possemiers

Verstraete

De Backer

Cani

Delzenne

NM.

Dietary modulation of clostridial cluster XIVa gut bacteria (Roseburia spp.) by chitin-glucan fiber improves host metabolic alterations induced by high-fat diet in mice. J Nutr Biochem. 2012;23(1):51-59.

50.

Pei

Zhu

Yang

, et al Bacteroides plebeius improves muscle wasting in chronic kidney disease by modulating the gut-renal muscle axis. J Cell Mol Med. 2022;26(24):6066-6078.

51.

Wang

Zou

, et al Effects of Chronic intermittent hypoxia and chronic sleep fragmentation on gut microbiome, serum metabolome, liver and adipose tissue morphology. Front Endocrinol (Lausanne). 2022;13:820939.

52.

Yang

Xiao

Zeng

, et al Tianwang buxin granules influence the intestinal flora in perimenopausal insomnia. Biomed Res Int. 2021;2021:9979511.

53.

Yang

Huang

Yang

YSH

Tung

YT.

Acute sleep deprivation exacerbates systemic inflammation and psychiatry disorders through gut microbiota dysbiosis and disruption of circadian rhythms. Microbiol Res. 2023;268:127292.

54.

Chen

Lin

Chien

Tung

Gau

SS.

Altered gut microbiota correlates with behavioral problems but not gastrointestinal symptoms in individuals with autism. Brain Behav Immun. 2022;106:161-178.

55.

Nishiwaki

Ito

Hamaguchi

, et al Short chain fatty acids-producing and mucin-degrading intestinal bacteria predict the progression of early Parkinson’s disease. NPJ Parkinsons Dis. 2022;8(1):65.

56.

Pinzone

Celesia

Di Rosa

Cacopardo

Nunnari

Microbial translocation in chronic liver diseases. Int J Microbiol. 2012;2012:694629.

57.

Xia

Shi

Zhou

, et al Milled flaxseed-added diets ameliorated hepatic inflammation by reducing gene expression of TLR4/NF-κB pathway and altered gut microbiota in STZ-induced type 1 diabetic mice. Food Sci Hum Wellness. 2022;11(1):32-40.

58.

Dong

Gai

Han

Bifidobacterium longum subsp. longum BL21 ameliorates alcoholic liver disease in mice through enhancement of the hepatic antioxidant capacity and modulation of the gut microbiota. J Appl Microbiol. 2023;134(11):lxad251.

59.

Wang

Liao

Zhou

, et al Parabacteroides distasonis alleviates obesity and metabolic dysfunctions via production of succinate and secondary bile acids. Cell Rep. 2019;26(1):222-235 e5.

60.

Zhang

Luo

Niu

, et al Chronic intermittent hypoxia induces gut microbial dysbiosis and infers metabolic dysfunction in mice. Sleep Med. 2022;91:84-92.

61.

Sun

Jin

Yang

, et al Association between gut microbiota and obesity combined with high carotid intima-media thickness among Chinese children. Asia Pac J Clin Nutr. 2023;32(1):93-105.

62.

Boehme

Remond-Derbez

Lerond

, et al Bifidobacterium longum subsp. longum reduces perceived psychological stress in healthy adults: an exploratory clinical trial. Nutrients. 2023;15(14):3122.

63.

Bowers

Summa

Thompson

, et al A prebiotic diet alters the fecal microbiome and improves sleep in response to sleep disruption in rats. Front Neurosci. 2022;16:889211.

64.

Chen

Tan

Yang

Tang

Bai

Tian

Fermentation patterns of prebiotics fructooligosaccharides-SCFA esters inoculated with fecal microbiota from ulcerative colitis patients. Food Chem Toxicol. 2023;180:114009.

65.

Shimizu

Yamamura

Yokoi

, et al Shorter sleep time relates to lower human defensin 5 secretion and compositional disturbance of the intestinal microbiota accompanied by decreased short-chain fatty acid production. Gut Microbes. 2023;15(1):2190306.

66.

Badran

Mashaqi

Gozal

The gut microbiome as a target for adjuvant therapy in obstructive sleep apnea. Expert Opin Ther Targets. 2020;24(12):1263-1282.

67.

Kang

Adams

Gregory

, et al Microbiota transfer therapy alters gut ecosystem and improves gastrointestinal and autism symptoms: an open-label study. Microbiome. 2017;5(1):10.

68.

Saleeb-Mousa

Nathanael

Coney

Kalla

Brain

Holmes

AP.

Mechanisms of atrial fibrillation in obstructive sleep apnoea. Cells. 2023;12(12).

69.

Fassarella

Blaak

Penders

Nauta

Smidt

Zoetendal

EG.

Gut microbiome stability and resilience: elucidating the response to perturbations in order to modulate gut health. Gut. 2021;70(3):595-605.

70.

Rajkov

Zdravkovic

Ninic

, et al Upregulation of peripheral blood mononuclear cells resistin gene expression in severe obstructive sleep apnea and obstructive sleep apnea with coexisting type 2 diabetes mellitus. Sleep Breath. 2023;27(5):2031-2039.

71.

Han

Van Treuren

Fischer

, et al A metabolomics pipeline for the mechanistic interrogation of the gut microbiome. Nature. 2021;595(7867):415-420.

72.

Riha

Kotoulas

Pataka

, et al Obstructive sleep apnoea in adult patients post-tonsillectomy. Sleep Med. 2021;78:189-192.

73.

Schwartz

Langdon

Dantas

Understanding the impact of antibiotic perturbation on the human microbiome. Genome Med. 2020;12(1):82.

74.

Vatanen

Franzosa

Schwager

, et al The human gut microbiome in early-onset type 1 diabetes from the TEDDY study. Nature. 2018;562(7728):589-594.

75.

Brennan

Kirby

SD.

The role of artificial intelligence in the treatment of obstructive sleep apnea. J Otolaryngol Head Neck Surg. 2023;52(1):7.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.61 MB