Introduction
Idiopathic pulmonary fibrosis (IPF) and chronic obstructive pulmonary disease (COPD) are progressive lung diseases that lead to irreversible respiratory decline.1 COPD is significantly more prevalent, affecting an estimated 212 million people globally and contributing to over 3 million deaths annually, making it one of the leading causes of respiratory morbidity.2 In contrast, IPF is a rarer condition, with a global annual incidence of 5–20 cases per 100,000 people, primarily affecting men over 50.3 While COPD is characterized by airflow limitation and chronic inflammation,4 IPF is marked by fibrosis of lung tissue, resulting in progressive respiratory dysfunction, severe breathing difficulties, and eventual respiratory failure.5 Both diseases share overlapping risk factors such as smoking, environmental exposures, and aging,6 yet these factors appear to direct individuals toward different pathological outcomes. In addition to environmental risks, recent evidence emphasizes the role of genetic predisposition in fibrotic lung diseases, such as familial pulmonary fibrosis.7 This highlights the importance of exploring genetic and environmental interactions that may differentiate COPD and IPF pathogenesis. Investigating both diseases together allows for a deeper understanding of potential shared or divergent pathways in their pathogenesis, which may uncover novel insights into targeted prevention and treatment strategies.
Recent studies have highlighted the potential role of saliva microbiota in respiratory diseases, as it contributes to the lung microbiota through microaspiration, influencing immune responses, inflammation, and epithelial integrity, which are critical in the pathogenesis of lung diseases.8 In COPD, changes in saliva microbiota composition are characterized by increased bacterial richness and diversity, with genera like Veillonella, Rothia, and Actinomyces frequently identified. These alterations are linked to disease progression and exacerbations through inflammatory pathways, as evidenced by associations between specific microbiota profiles and elevated salivary inflammatory markers, which correlate negatively with lung function.9 In IPF, the oral microbiota significantly contributes to lung microbial composition, with 32.84% of lung microbiota genes traced to the oral cavity. Enriched genera like Streptococcus, Pseudobutyrivibrio, and Anaerorhabdus are associated with microbial translocation, biofilm formation, antibiotic resistance, and metabolic changes, potentially promoting lung injury, fibrosis, and reduced microbial diversity, highlighting the oral-lung axis in IPF etiology.10 Repeated aspiration of oral or gastric contents due to gastroesophageal reflux may further contribute to lung inflammation and microbial imbalance in IPF.11 These findings suggest that oral microbial changes may influence disease development and warrant investigation to determine their causal role in pathogenesis.
Mendelian Randomization (MR) is a robust genetic epidemiology method that utilizes genetic variants, typically single nucleotide polymorphisms (SNPs) identified through Genome-Wide Association Studies (GWAS), as instrumental variables (IVs) to assess the causal relationship between an exposure and an outcome.12 By minimizing confounding and reverse causality, MR offers a powerful tool to identify causal links when randomized controlled trials may be impractical or unethical.13 This study employs a two-sample MR approach to investigate the genetically causal relationships between saliva microbiota abundance and IPF or COPD, aiming to identify potential shared and disease-specific microbial influences on pathogenesis. The findings may provide novel insights into the microbial mechanisms underlying these diseases, facilitating the understanding of their progression and enabling early risk stratification for targeted interventions.
Materials and Methods
Study Design
The methodology followed the STROBE-MR statement guidelines and employed a two-sample MR framework to assess bidirectional causal relationships between saliva microbiota abundance and respiratory diseases. Forward MR was used to estimate the effects of microbial taxa on COPD and IPF, while reverse MR tested whether genetic liability to COPD or IPF influenced microbial abundance. IVs were selected based on the three key principles of MR analysis: relevance, independence, and exclusion restriction.14 MR Steiger directionality testing was performed to validate the inferred causal direction. GWAS summary statistics for exposures and outcomes were obtained from publicly available datasets. Sensitivity analyses included heterogeneity testing, horizontal pleiotropy assessment, and outlier correction. A schematic overview of the workflow is shown in Figure 1.
|
Figure 1 Study flowchart for the Mendelian randomization (MR) analysis of saliva microbiota and respiratory diseases. This flowchart summarizes the bidirectional MR framework used to investigate potential causal relationships between salivary microbiota abundance and chronic obstructive pulmonary disease (COPD) or idiopathic pulmonary fibrosis (IPF). GWAS summary statistics for 44 saliva microbiota traits were screened, with 43 traits meeting the inclusion criteria for instrumental variable (IV) selection. Forward MR used the inverse variance weighted (IVW) method as the primary analysis. Sensitivity analyses included MR-PRESSO for outlier correction, Cochran’s Q test, MR-Egger intercept, and leave-one-out analysis. Steiger filtering was applied to confirm causal directionality. Reverse MR was performed using genome-wide significant SNPs from COPD and IPF GWAS to evaluate potential feedback effects on salivary microbiota composition.
|
Ethics Statement
All data used in this study were obtained from publicly available datasets containing de-identified human genetic or microbiome data. In accordance with Article 32, Items 1 and 2 of the Measures for Ethical Review of Life Science and Medical Research Involving Human Subjects (promulgated February 18, 2023, China), research using publicly available or legally obtained, de-identified human data is exempt from ethics review and informed consent. Therefore, no further institutional review board approval was required.
Data Sources
GWAS data on IPF were obtained from a published study via https://github.com/genomicsITER/PFgenetics/tree/master. The study analyzed individuals of European ancestry, including 2668 IPF cases and 8591 controls in the discovery phase.15 GWAS data for COPD were sourced from the FINNGEN R10 release (https://r9.risteys.finngen.fi/endpoints/J10_COPD), comprising 18,266 cases, predominantly of Finnish ancestry. Oral microbiome data were obtained from a study conducted on 610 unrelated adults of Danish ancestry from the ADDITION-PRO cohort. The study utilized 16S rRNA gene sequencing to investigate the salivary microbiota and performed GWAS to identify host genetic variants associated with oral bacterial traits. A total of 44 oral microbiome measurements were analyzed, including 43 univariate bacterial features (spanning taxonomic groups from phylum to species levels) and one multivariate bacterial community diversity metric.16 To our knowledge, there is no sample overlap between the exposure (oral microbiota) and outcome (COPD and IPF) GWAS datasets, as these were derived from distinct cohorts (ADDITION-PRO and FinnGen/multicenter studies, respectively). The information on the datasets is summarized in Table S1.
IV Selection
For forward MR analysis, IVs were selected from GWAS summary statistics of saliva microbiota traits. An initial threshold of genome-wide significance (P < 5×10−8) yielded insufficient variants for most taxa; therefore, a relaxed threshold of P < 5×10−6 was adopted, consistent with prior MR studies on gut microbiota, inflammatory factors, and related traits.17 To ensure adequate instrument strength and model stability, a minimum of 3 independent SNPs per exposure was required.18 For exposures with fewer than 3 SNPs at this threshold, the criterion was further relaxed to P < 1 × 10−5, as previously reported.19–21 In contrast, for reverse MR analysis, instruments for COPD and IPF were selected using the standard genome-wide significance threshold (P < 5×10−8) without relaxation.
To guarantee sufficient genetic variability, only SNPs with a minor allele frequency (MAF) greater than 0.01 were included.22 Linkage disequilibrium (LD) effects were minimized by excluding SNPs with high LD (R2 < 0.001) within a 10,000 kb window.23 For cases where the selected IV was absent from the outcome summary data, proxy SNPs with strong LD (R2 > 0.8) were identified and substituted to maintain integrity.24 Furthermore, the F-statistic was calculated for each SNP to assess instrument strength and prevent weak instrument bias. The formula F = R² × (N – 2) / (1 – R²) was used. R² represents the proportion of variance in the exposure explained by the SNP. Only SNPs with an F-statistic > 10 were included to ensure the reliability of the IVs and their ability to capture the causal effect of the exposure on the outcome.25
MR Analysis
In this study, the inverse variance weighted (IVW) random effects method was the primary approach for assessing the causal relationship between exposure and outcomes, providing odds ratios (ORs) with 95% confidence intervals (CIs). The IVW method calculates a weighted average effect size using each SNP’s inverse variance.26 To ensure robustness, MR-Egger, weighted median, and weighted mode methods were also applied. MR-Egger accounts for pleiotropy by including an intercept term,18 while the weighted median method provides reliable estimates if at least 50% of IVs are valid.27 The weighted mode method identifies the causal effect as the mode of the effect estimates, weighted by their precision, and is robust even when most IVs are invalid, provided that the largest subset of valid instruments produces consistent estimates.28 Associations with IVW P < 0.05 were considered statistically significant. P values were adjusted for multiple comparisons using the Benjamini–Hochberg false discovery rate (FDR) method.
Sensitivity Analysis
Sensitivity analyses were performed to evaluate the robustness of causal estimates and detect potential violations of MR assumptions. Heterogeneity among instrument-specific estimates was assessed using Cochran’s Q test under the IVW model. A P-value > 0.05 indicated low heterogeneity, suggesting consistency across SNP effects.29 Horizontal pleiotropy was evaluated using the MR-Egger intercept. A non-significant deviation from zero indicated the absence of directional pleiotropy.18 MR-PRESSO identified and corrected for outlier variants contributing to pleiotropic bias.30 Robustness was further assessed by leave-one-out (LOO) analysis, which sequentially removed individual SNPs to evaluate their influence on the overall causal estimates.31 To ensure correct causal direction, the Steiger directionality test was applied, and SNPs explaining greater variance in the outcome than in the exposure (R2_outcome > R2_exposure) were excluded prior to MR. All analyses were performed using the “TwoSampleMR” package. Diagnostic plots (scatter, forest, funnel) were used for visualization.
Results
IV Selection in Forward MR
For IV selection, out of 44 saliva microbiota abundance datasets, 43 were included in the final analysis. Beta diversity of salivary microbiota (GCST90429842) was excluded due to insufficient information. IVs were primarily selected using the threshold of P < 5×10−6. For unknown Streptococcus species (ASV0003, GCST90429825), species parvula (GCST90429829), genus Alloprevotella (GCST90429822), genus Streptococcus (GCST90429813), and unknown Rothia species (ASV0012, GCST90429834), fewer than 3 SNPs met this criterion. Therefore, the threshold was relaxed to P < 1×10−5. Unavailable SNPs in the abundance data for unknown Schaalia species (ASV0017), unknown Rothia species (ASV0016), and unknown Neisseria species (ASV0004) for IPF, and unknown Rothia species (ASV0016) and species parvula for COPD, were substituted with proxy SNPs, including rs73057773 for rs7807974, rs67487314 for rs61026851, rs2614710 for rs67577153, rs34837414 for rs7247650, rs4860383 for rs113621445, and rs9571821 for rs9564412. The number of IVs per dataset ranged from 3 to 13, with mean F-statistic between 20.53 and 27.10, confirming instrument strength (Table S2).
Forward MR Analysis
MR analysis using the IVW method revealed a significant inverse association between parvula abundance and COPD risk (OR = 0.9546, 95% CI: 0.9270–0.9831, P = 0.002). Higher abundances of class Bacilli (OR = 0.8447, 95% CI: 0.7402–0.9639, P = 0.0122) and genus Porphyromonas (OR = 0.8398, 95% CI: 0.7224–0.9764, P = 0.0231) were significantly associated with reduced risk of IPF (Table 1).
 |
Table 1 Significant Associations Between Saliva Microbiota Abundance and Respiratory Diseases in Forward and Reverse Mendelian Randomization Analyses (IVW Method)
|
MR-PRESSO detected outliers in three associations: genus Fusobacterium with IPF (1 outlier), genus Rothia with COPD (1 outlier), and unknown Rothia species (ASV0012) with COPD (2 outliers) (Table S3). After removing outliers, a significant inverse association emerged between Fusobacterium and IPF (IVW OR = 0.9069, 95% CI: 0.8338–0.9865, P = 0.0227; Table 1). Other associations remained null. Notably, after FDR correction, only the association between parvula and COPD remained significant (adjusted P = 0.019), reinforcing its robustness. Full MR results, including post-correction estimates, are shown in Table S4. Scatter and forest plots illustrated the SNP-specific effects (Figures 2 and 3). These findings provide novel evidence linking specific salivary microbial taxa to COPD and IPF risk, suggesting disease-specific microbial contributions.
 |
Figure 2 Scatter plots show genetic associations between saliva microbiota abundance and respiratory disease using different MR methods. Regression lines for IVW, MR-Egger, weighted median, and weighted mode methods are included when applicable. (A) Saliva microbiota abundance of species parvula and COPD risk. (B) Saliva microbiota abundance of class Bacilli and IPF risk. (C) Saliva microbiota abundance of genus Porphyromonas and IPF risk. (D) Saliva microbiota abundance of genus Fusobacterium and IPF risk. (E) COPD and saliva microbiota abundance of species periodonticum.
|
 |
Figure 3 Forest plots demonstrate genetic associations between saliva microbiota abundance and respiratory disease. Forest plots summarize genetic associations between saliva microbiota abundance and respiratory disease risks, with odds ratios (ORs) and 95% confidence intervals (CIs). Each panel shows associations for different taxa, including individual SNP effects and the overall effect size derived from MR analyses. (A) parvula and COPD risk. (B) Bacilli and IPF risk. (C) Porphyromonas and IPF risk. (D) Fusobacterium and IPF risk. (E) COPD and saliva abundance of periodonticum.
|
Sensitivity and Directionality Analysis in Forward MR
Sensitivity analysis confirmed the robustness of our findings. No evidence of heterogeneity (Cochran’s Q P = 0.7748 for parvula–COPD, P = 0.7179 for Bacilli–IPF, P = 0.2210 for Porphyromonas–IPF, and P = 0.3145 for Fusobacterium–IPF) or horizontal pleiotropy (MR-Egger intercept P = 0.7327, 0.4638, 0.7139, and 0.4938, respectively) was detected in significant associations (Table S5). Funnel plots showed symmetrical SNP distributions (Figure 4). LOO analysis confirmed that no individual SNP disproportionately influenced the results (Figure 5). MR-PRESSO distortion tests supported the validity of the Bacilli–IPF association (OR = 0.8447, 95% CI: 0.7732–0.9228, P = 0.0334). No significant pleiotropy or distortion effects were detected for the significant associations, indicating that horizontal pleiotropy is unlikely to influence these findings (Table S3). To ensure correct causal direction, SNPs explaining more variance in the outcome than in the exposure were excluded prior to MR (Table S6), and all retained associations passed the Steiger directionality test (Table 2), supporting causality from microbiota to disease.
 |
Table 2 Results of Steiger Directionality Testing for Salivary Microbiota Traits and Respiratory Disease Outcomes
|
 |
Figure 4 Funnel plots assess horizontal pleiotropy for the association between saliva microbiota abundance and respiratory disease. (A) parvula and COPD risk. (B) Bacilli and IPF risk. (C) Porphyromonas and IPF risk. (D) Fusobacterium and IPF risk. (E) COPD and saliva abundance of periodonticum.
|
 |
Figure 5 Leave-one-out (LOO) analysis for the association between saliva microbiota abundance and respiratory disease. (A) parvula and COPD risk. (B) Bacilli and IPF risk. (C) Porphyromonas and IPF risk. (D) Fusobacterium and IPF risk. (E) COPD and saliva abundance of periodonticum.
|
Reverse MR
For reverse MR analysis, genetic instruments were selected from GWAS summary data of COPD and IPF at genome-wide significance (P < 5×10−8). For COPD, 20 independent SNPs were initially identified. After removing 5 SNPs unavailable in the microbial GWAS datasets (rs538515410, rs60892124, rs9271399, rs28929474, rs141669463) and 2 additional SNPs due to palindromic structure or weak strength (rs1095705, rs6874581), a total of 13 IVs were retained. Similarly, 15 SNPs were selected for IPF, among which three (rs78238620, rs35705950, rs41308092) were not available in the outcome datasets, yielding 12 final IVs. The mean F-statistics were 51.53 (range: 29.89–330.93) for COPD and 115.52 (range: 36.96–927.06) for IPF, confirming sufficient instrument strength. COPD showed a potential positive association with species periodonticum abundance (IVW OR = 1.5446, 95% CI: 1.0170–2.3460, P = 0.041, adjusted P = 0.979) (Table 1). The reverse MR results are provided in Table S7.
Sensitivity analysis revealed heterogeneity (IVW Q statistic P < 0.05) or evidence of directional pleiotropy (MR-Egger P < 0.05) in several associations, including COPD with family Pasteurellaceae (Q P = 0.032), genus Haemophilus (Q P = 0.016), genus Leptotrichia (Q P = 0.023), genus schaalia (Egger P = 0.049), and unknown Schaalia species (ASV0017) (Egger P = 0.048), as well as IPF with unknown Streptococcus species (ASV0003) (Q P = 0.037) (Table S8). MR-PRESSO identified outliers in the associations of COPD with Haemophilus, Leptotrichia, and Pasteurellaceae (Table S9). After outlier removal, these associations remained non-significant (Table S7). LOO identified rs16969968 as an influential SNP in association with Schaalia (genus and ASV0017). Its exclusion eliminated potential pleiotropy and heterogeneity without altering the null results (Table S7). No horizontal pleiotropy or heterogeneity was detected after correction, except for COPD–Haemophilus (P = 0.03) (Table S9).
Discussion
This bidirectional MR study explored potential causal links between salivary microbiota composition and two chronic respiratory diseases, COPD and IPF. In forward MR, higher abundances of species parvula were associated with a lower risk of COPD, while class Bacilli, genus Porphyromonas, and genus Fusobacterium were inversely associated with IPF. These associations showed no evidence of horizontal pleiotropy or heterogeneity, and all passed the Steiger directionality test, suggesting a possible causal effect of specific microbial taxa on respiratory disease risk. In contrast, reverse MR identified an association between COPD and increased abundance of species periodonticum. While no pleiotropy or heterogeneity was detected for this pair, several other reverse MR associations showed sensitivity to instrument outliers, limiting interpretability. Taken together, these findings suggest microbiota-to-disease directionality is more plausible than the reverse, providing preliminary evidence for salivary microbial involvement in the pathogenesis of COPD and IPF.
After multiple testing corrections, the inverse associations of Bacilli, Porphyromonas, and Fusobacterium with IPF did not remain statistically significant (all adjusted P > 0.05), suggesting limited robustness. In contrast, parvula exhibited a significant negative association with COPD risk (P = 0.002; adjusted P = 0.019), indicating a potentially protective role. However, this finding appears to contrast with previous observational and experimental evidence implicating Veillonella parvula in COPD pathogenesis. V. parvula isolated from the saliva of COPD frequent exacerbators has been shown to impair epithelial barrier integrity, increase cytotoxicity, and activate IL-1β/NF-κB signaling in bronchial epithelial cells.32 It has also been identified as a dominant species enriched in the lower airways of COPD patients, positively correlated with neutrophilic inflammation and reduced lung function.33 As a predominant subgingival species, V. parvula can translocate to the lower respiratory tract via microaspiration of saliva, particularly in individuals with periodontal disease or impaired mucociliary clearance.9 This apparent discrepancy may reflect differences in taxonomic resolution, microbial niche, or study design. The MR framework infers lifetime genetic predisposition to salivary parvula abundance, which may not directly reflect the local effects of transient airway colonization or infection. Additionally, salivary parvula may serve as a proxy for broader microbial community features or immune homeostasis that modulate COPD risk. Future studies integrating strain-level metagenomics, mucosal immunity, and longitudinal sampling are needed to clarify the context-dependent role of parvula in respiratory health.
IPF is characterized by dysfunction of alveolar epithelial cells, leading to impaired barrier integrity, persistent inflammation, and a profibrotic microenvironment that drives fibroblast activation and irreversible lung remodeling.34 Emerging evidence suggests that microbial communities may influence these processes. In our analysis, Bacilli, Porphyromonas, and Fusobacterium showed nominal inverse associations with IPF risk, although none remained significant after multiple testing corrections. These taxa are common components of the oral and respiratory microbiota and have been implicated in maintaining mucosal and immune balance and microbial diversity,35 both of which are considered protective against chronic inflammation and epithelial injury in IPF.36 Preclinical studies suggest that members of the class Bacilli, such as Lactobacillus, can enhance epithelial barrier function and modulate immune responses via the production of lactic acid and bacteriocins.37 These metabolites may suppress pro-inflammatory cytokines and reduce tissue injury, potentially mitigating fibrotic progression. For example, Lactobacillus have been shown to stabilize epithelial monolayers and suppress inflammatory signaling in vitro.38 These properties may counteract the inflammation and tissue remodeling observed in fibrosis.
The genus Porphyromonas, commonly associated with periodontal disease, produces short-chain fatty acids such as butyrate,39 which regulate immune responses by increasing regulatory T cells and reducing pro-inflammatory cytokines.40 Butyrate has also been shown to inhibit TGF-β1-induced myofibroblast differentiation and enhance mitochondrial function, thereby mitigating fibrotic progression.41 These mechanisms align with the observed nominal inverse association between Porphyromonas abundance and IPF risk, suggesting a possible protective role in maintaining mucosal homeostasis. Fusobacterium, a facultative anaerobe frequently detected in the lower respiratory tract,42 has been linked to altered lung microbiota in IPF.10 However, our findings suggest a nominal negative association between salivary Fusobacterium and IPF risk. This may indicate that oral Fusobacterium contributes to microbial stability and barrier defense. For example, Fusobacterium nucleatum is known to support biofilm structure and induce antimicrobial peptides and chemokines that modulate host responses.43 These observations raise the question of whether specific oral commensals exert context-dependent effects—protective in the oral niche but potentially pathogenic upon translocation. One hypothesis is that higher oral abundance of these genera may reflect a more balanced or resilient microbiome state, indirectly influencing systemic or mucosal immune tone relevant to lung disease. Collectively, these findings provide preliminary evidence for the potentially protective roles of specific oral taxa in fibrotic lung disease. Given the exploratory nature of our analysis, further mechanistic and longitudinal studies are needed to clarify their functional relevance and therapeutic potential.
This study establishes potential genetically causal links between specific saliva microbiota and COPD and IPF. However, several limitations should be acknowledged. The reliance on GWAS data from predominantly European populations may restrict the generalizability. Due to limited variant availability in current oral microbiota GWAS datasets based on a relatively small sample size (n = 610), we applied a relaxed significance threshold for IV selection, which enabled broader analysis but introduced weak instrument bias. Larger, ancestry-diverse GWAS are needed to improve instrument strength. Although meta-GWAS or pooled datasets could address this, no suitable resources are currently available. Similarly, triangulation using gut or nasal microbiota is constrained by niche-specific microbial differences and the lack of harmonized cross-site data. At present, replication using independent microbiota GWAS or observational cohorts is not feasible due to the limited availability of comparable salivary microbiota datasets with genetic data. Although oral microbiome data exist for East Asian populations, they were not used for replication because of ancestry differences that may introduce bias. This limits the generalizability and reinforces the exploratory nature of the findings. Some microbial traits, such as ASV0012, remain taxonomically ambiguous. However, re-annotation was not feasible due to the absence of full-length 16S sequences, raw FASTQ files, or shotgun metagenomic data in the original dataset. Additionally, the use of salivary taxa as proxies for respiratory exposure warrants further validation. Moreover, the reliance on broad taxonomic categories may obscure the functional roles of individual microbial species. Addressing these limitations through diverse cohorts, expanded GWAS datasets, advanced microbial characterization, and functional assessments will provide deeper insights into the role of saliva microbiota in respiratory diseases.
Conclusion
This study provides exploratory evidence for genetically inferred associations between specific salivary microbiota and the risk of COPD and IPF, offering new insights into potential microbiome–host interactions in chronic respiratory disease. Increased abundance of species parvula was significantly associated with reduced COPD risk, while Bacilli, Porphyromonas, and Fusobacterium showed nominal inverse associations with IPF. Reverse MR provided limited evidence for disease-to-microbiota effects, further supporting a directional influence of oral microbes on disease susceptibility. These findings suggest the potential of salivary microbiota as biomarkers or modulators of chronic lung disease, warranting further validation in diverse populations and functional studies to clarify their mechanistic relevance.
Data Sharing Statement
All data generated or analyzed during this study are included in this published article and its supplementary information files.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Funding
There is no funding to report.
Disclosure
The authors declare that they have no competing interests.
References
1. Spagnolo P, Semenzato U. Revealing the pathogenic and ageing-related mechanisms of the enigmatic idiopathic pulmonary fibrosis (and chronic obstructive pulmonary disease). Curr Opin Pulm Med. 2022;28(4):296–302. doi:10.1097/MCP.0000000000000876
2. Alemayohu MA, Zanolin ME, Cazzoletti L, et al. Burden and risk factors of chronic obstructive pulmonary disease in Sub-Saharan African countries, 1990–2019: a systematic analysis for the global burden of disease study 2019. EClinicalMedicine. 2023;64. doi:10.1016/j.eclinm.2023.102215
3. Harari S, Davì M, Biffi A, et al. Epidemiology of idiopathic pulmonary fibrosis: a population-based study in primary care. Int Emerg Med. 2020;15(3):437–445. doi:10.1007/s11739-019-02195-0
4. Polosukhin VV, Gutor SS, Du R-H, et al. Small airway determinants of airflow limitation in chronic obstructive pulmonary disease. Thorax. 2021;76(11):1079–1088. doi:10.1136/thoraxjnl-2020-216037
5. Barratt SL, Creamer A, Hayton C, Chaudhuri N. Idiopathic pulmonary fibrosis (IPF): an overview. J Clin Med. 2018;7(8):201. doi:10.3390/jcm7080201
6. Beghé B, Cerri S, Fabbri LM, Marchioni A. COPD, pulmonary fibrosis and ILAs in aging smokers: the paradox of striking different responses to the major risk factors. Int J Mol Sci. 2021;22(17):9292. doi:10.3390/ijms22179292
7. Monteleone G, Sip ISG, Bergantini L, et al. The management of familial pulmonary fibrosis in different medical settings: where does that leave us? An Italian nationwide survey. Sarcoidosis Vasculitis Diffuse Lung Dis. 2024;41(3):e2024047. doi:10.36141/svdld.v41i3.15744
8. Mammen MJ, Scannapieco FA, Sethi S. Oral‐lung microbiome interactions in lung diseases. Periodontology 2000. 2020;83(1):234–241. doi:10.1111/prd.12301
9. Lin M, Li X, Wang J, et al. Saliva microbiome changes in patients with periodontitis with and without chronic obstructive pulmonary disease. Front Cell Infect Microbiol. 2020;10:124. doi:10.3389/fcimb.2020.00124
10. Tong X, Su F, Xu X, et al. Alterations to the lung microbiome in idiopathic pulmonary fibrosis patients. Front Cell Infect Microbiol. 2019;9:149. doi:10.3389/fcimb.2019.00149
11. Ruaro B, Pozzan R, Confalonieri P, et al. Gastroesophageal reflux disease in idiopathic pulmonary fibrosis: viewer or actor? To treat or not to treat? Pharmaceuticals. 2022;15(8):1033. doi:10.3390/ph15081033
12. Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925–1926. doi:10.1001/jama.2017.17219
13. Pierce BL, Ahsan H, VanderWeele TJ. Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. Int J Epidemiol. 2011;40(3):740–752. doi:10.1093/ije/dyq151
14. Skrivankova VW, Richmond RC, Woolf BA, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE-MR statement. JAMA. 2021;326(16):1614–1621. doi:10.1001/jama.2021.18236
15. Allen RJ, Guillen-Guio B, Oldham JM, et al. Genome-wide association study of susceptibility to idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2020;201(5):564–574. doi:10.1164/rccm.201905-1017OC
16. Stankevic E, Kern T, Borisevich D, et al. Genome-wide association study identifies host genetic variants influencing oral microbiota diversity and metabolic health. Sci Rep. 2024;14(1):14738. doi:10.1038/s41598-024-65538-8
17. Ji D, Chen W-Z, Zhang L, Zhang Z-H, Chen L-J. Gut microbiota, circulating cytokines and dementia: a Mendelian randomization study. J Neuroinflammation. 2024;21(1):2. doi:10.1186/s12974-023-02999-0
18. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-egger method. Eur j epidemiol. 2017;32:377–389. doi:10.1007/s10654-017-0255-x
19. Huang Y, Kang Z, He Y, Qiu Y, Song Y, Liu W. Association between gut microbiota and common overlapping gastrointestinal disorders: a bidirectional two-sample Mendelian randomization study. Front Microbiol. 2024;15:1343564. doi:10.3389/fmicb.2024.1343564
20. Li N, Wang H, Pei H, et al. Genus_Ruminococcus and order_Burkholderiales affect osteoporosis by regulating the microbiota-gut-bone axis. Front Microbiol. 2024;15:1373013. doi:10.3389/fmicb.2024.1373013
21. Ren Y, Zhang Y, Cheng Y, Qin H, Zhao H. Genetic liability of gut microbiota for idiopathic pulmonary fibrosis and lung function: a two-sample Mendelian randomization study. Front Cell Infect Microbiol. 2024;14:1348685. doi:10.3389/fcimb.2024.1348685
22. Klaric L, Gisby JS, Papadaki A, et al. Mendelian randomisation identifies alternative splicing of the FAS death receptor as a mediator of severe COVID-19. medRxiv. 2021. doi:10.1101/2021.04.01.21254789
23. Cheng Q, Yang Y, Shi X, et al. MR-LDP: a two-sample Mendelian randomization for GWAS summary statistics accounting for linkage disequilibrium and horizontal pleiotropy. NAR Genom Bioinform. 2020;2(2):lqaa028. doi:10.1093/nargab/lqaa028
24. Sanderson E, Glymour MM, Holmes MV, et al. Mendelian randomization. Nat Rev Meth Prim. 2022;2. doi:10.1038/s43586-021-00092-5
25. Brion MJ, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol. 2013;42(5):1497–1501. doi:10.1093/ije/dyt179
26. Lin Z, Deng Y, Pan W. Combining the strengths of inverse-variance weighting and egger regression in Mendelian randomization using a mixture of regressions model. PLoS Genetics. 2021;17(11):e1009922. doi:10.1371/journal.pgen.1009922
27. Scholz F-W. Weighted median regression estimates. Ann Stat. 1978;6(3):603–609. doi:10.1214/aos/1176344204
28. Burgess S, Foley CN, Allara E, Staley JR, Howson JM. A robust and efficient method for Mendelian randomization with hundreds of genetic variants. Nat Commun. 2020;11(1):376. doi:10.1038/s41467-019-14156-4
29. Pereira TV, Patsopoulos NA, Salanti G, Ioannidis JP. Critical interpretation of Cochran’s Q test depends on power and prior assumptions about heterogeneity. Res Synth Method. 2010;1(2):149–161. doi:10.1002/jrsm.13
30. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–698. doi:10.1038/s41588-018-0099-7
31. Rasooly D, Patel CJ. Conducting a reproducible Mendelian randomization analysis using the R analytic statistical environment. Curr Protoc Hum Genet. 2019;101(1):e82. doi:10.1002/cphg.82
32. Ke L, Chen L, Yaling Y, Can G, Jun L, Chuan Z. Investigation on the pathological mechanism of frequent exacerbators with chronic obstructive pulmonary disease based on the characteristics of respiratory flora. Front Med. 2022;8:816802. doi:10.3389/fmed.2021.816802
33. Gao J, Yang Y, Xiang X, et al. Human genetic associations of the airway microbiome in chronic obstructive pulmonary disease. Respir Res. 2024;25(1):165. doi:10.1186/s12931-024-02805-2
34. Selman M, Pardo A. The leading role of epithelial cells in the pathogenesis of idiopathic pulmonary fibrosis. Cell Signalling. 2020;66:109482. doi:10.1016/j.cellsig.2019.109482
35. Pathak JL, Yan Y, Zhang Q, Wang L, Ge L. The role of oral microbiome in respiratory health and diseases. Respir Med. 2021;185:106475. doi:10.1016/j.rmed.2021.106475
36. Fabbrizzi A, Nannini G, Lavorini F, Tomassetti S, Amedei A. Microbiota and IPF: hidden and detected relationships. Sarcoidosis Vasculitis Diffuse Lung Dis. 2021;38(3):e2021028. doi:10.36141/svdld.v38i3.11365
37. Mokoena MP, Omatola CA, Olaniran AO. Applications of lactic acid bacteria and their bacteriocins against food spoilage microorganisms and foodborne pathogens. Molecules. 2021;26(22):7055. doi:10.3390/molecules26227055
38. Wang J, Ji H, Wang S, et al. Probiotic Lactobacillus plantarum promotes intestinal barrier function by strengthening the epithelium and modulating gut microbiota. Front Microbiol. 2018;9:1953. doi:10.3389/fmicb.2018.01953
39. Anand S, Kaur H, Mande SS. Comparative in silico analysis of butyrate production pathways in gut commensals and pathogens. Front Microbiol. 2016;7:1945. doi:10.3389/fmicb.2016.01945
40. Zhou L, Zhang M, Wang Y, et al. Faecalibacterium prausnitzii produces butyrate to maintain Th17/Treg balance and to ameliorate colorectal colitis by inhibiting histone deacetylase 1. Inflamm Bowel Dis. 2018;24(9):1926–1940. doi:10.1093/ibd/izy182
41. Lee HY, Nam S, Kim MJ, Kim SJ, Back SH, Yoo HJ. Butyrate prevents TGF-β1-induced alveolar myofibroblast differentiation and modulates energy metabolism. Metabolites. 2021;11(5):258. doi:10.3390/metabo11050258
42. Robinson A, Wilde J, Allen-Vercoe E. Fusobacteria: physiology, form, and function. In: Colorectal Neoplasia and the Colorectal Microbiome. Elsevier; 2020:95–134.
43. Groeger S, Zhou Y, Ruf S, Meyle J. Pathogenic mechanisms of Fusobacterium nucleatum on oral epithelial cells. Front Oral Health. 2022;3:831607. doi:10.3389/froh.2022.831607