Microbial Bioactives

1. Introduction

The human body is not merely a singular organism but a vast, dynamic ecosystem wherein trillions of microorganisms—bacteria, viruses, fungi, and archaea—coexist with human cells in intricate, adaptive harmony. This collective consortium, known as the human microbiome, acts as a central regulator of physiological homeostasis, shaping metabolic pathways, educating immune responses, and maintaining mucosal integrity from infancy through old age (Boulangé et al., 2016; De Luca & Shoenfeld, 2019; Moffatt & Cookson, 2017). Rather than passive passengers, microbes contribute to nutrient synthesis, barrier defense, and systemic signaling, effectively functioning as an internal organ.

Crucially, the diversity and composition of the microbiome are not static; they are shaped by early-life events, environmental exposures, and lifestyle factors. In health, a balanced microbial ecology supports resilience; conversely, dysbiosis—a state of disrupted microbial balance—has emerged as a shared feature across chronic diseases, including autoimmune disorders, metabolic syndromes, and chronic respiratory conditions (Afzaal et al., 2022; De Luca & Shoenfeld, 2019; Ogunrinola et al., 2020). Recognizing these microbial signatures is essential to understanding disease susceptibility and progression.

The journey of human microbial colonization begins long before birth. Although once assumed sterile, the intrauterine environment appears to host microbial signals that influence immune maturation (Sinha et al., 2023). The perinatal period represents a critical “window of opportunity” wherein maternal microbiota—in the gut, vagina, oral cavity, and skin—serve as the primary architects of neonatal microbial assemblage (Nunez et al., 2021; Sinha et al., 2023). Microbial exposures during this window calibrate the infant’s immune system and metabolic programming, ultimately shaping vulnerability or resilience to allergens, pathogens, and metabolic stressors across the lifespan (Yao et al., 2021).

At birth, the mode of delivery, early feeding practices, and maternal environmental exposures dictate the initial microbial inoculum. Vaginal delivery seeds infants with maternal vaginal and fecal bacteria, whereas cesarean births produce a markedly different microbial signature (Dogra et al., 2021). Breastfeeding further enriches the infant’s microbiome with beneficial taxa such as Bifidobacterium and Lactobacillus, enhancing mucosal immunity (Yao et al., 2021). These early microbial interactions form the foundation for immune tolerance and metabolic efficiency.

Despite the natural synchrony of microbial establishment, this delicate process is vulnerable to external insults, particularly environmental toxins like tobacco smoke. Tobacco smoke is a complex aerosol containing thousands of harmful chemicals, including nicotine, polycyclic aromatic hydrocarbons, and heavy metals (Saha et al., 2007). Exposure to these compounds in utero or during early life has profound consequences for the developing infant, increasing the risk of low birth weight, altered lung function, and later chronic disease (Reeves & Bernstein, 2008; Diamanti et al., 2019).

Emerging evidence indicates that perinatal tobacco exposure does more than affect organ development; it disrupts microbial ecosystems. Infants born to smoking mothers consistently demonstrate shifts in gut microbiome composition characterized by increased abundance of Firmicutes and reduced diversity—patterns linked with metabolic dysregulation (Tun et al., 2017; Peng et al., 2024). Such dysbiosis may set the stage for heightened adiposity and altered immune thresholds. For example, Akkermansia muciniphila, a microbe associated with mucosal integrity and anti-inflammatory signaling, is often depleted in smoke-exposed offspring, suggesting a compromised gut barrier and altered metabolic potential (Pérez-Castro et al., 2024).

Beyond prenatal smoking, environmental tobacco smoke and thirdhand smoke—residual tobacco contaminants persisting on surfaces—further perturb microbial diversity in neonatal settings such as neonatal intensive care units (Northrup et al., 2022). These findings point to a concerning paradox: exposures intended to impact only the parent also reach the infant, altering microbial landscapes with potential long-term consequences.

Traditionally, the lungs were believed to be sterile in health. However, advancements in high-throughput sequencing have overturned this assumption, demonstrating that the lower respiratory tract hosts a low-biomass but distinct microbiome that influences pulmonary immunity and homeostasis (Charlson et al., 2011; Moffatt & Cookson, 2017). In healthy lungs, commensals such as Streptococcus, Prevotella, and Veillonella maintain a balanced immunological dialogue with host tissues, contributing to pathogen exclusion and tissue repair.

In chronic lung conditions, this equilibrium breaks down. A growing body of research reveals that microbial community structure in the lungs shifts in interstitial lung diseases (ILDs), particularly those with a progressive fibrosing phenotype (Faner et al., 2017). Progressive fibrosing ILD (PF-ILD) embodies a subset of ILDs marked by relentless fibrosis, worsening dyspnea, and accelerated decline in pulmonary function—features that closely mirror idiopathic pulmonary fibrosis (IPF) (Richeldi et al., 2017; Maher et al., 2019).

Understanding how the lung microbiome contributes to fibrotic progression requires examining both microbial burden and taxonomic shifts. Meta-analytic data indicate that patients with IPF have significantly higher bacterial loads in bronchoalveolar lavage or lung tissue compared to healthy controls or individuals with chronic obstructive pulmonary disease (COPD) (Molyneaux et al., 2014; O’Dwyer et al., 2019). In acute exacerbations of IPF, bacterial burden can quadruple, aligning with episodes of rapid clinical decline (Molyneaux et al., 2017).

Taxonomic studies reveal that dysbiosis in fibrotic lungs often involves increases in potentially pathogenic taxa such as Streptococcus and Staphylococcus, which correlate with worse prognoses and accelerated fibrosis (Han et al., 2017). Conversely, reductions in microbial diversity and the loss of commensal anaerobes are linked to dysregulated host immunity and pro-fibrotic signaling (Huang et al., 2017; Molyneaux et al., 2014). These patterns suggest that microbial perturbations are not mere epiphenomena but may actively participate in the perpetuation of fibrotic cascades.

Mechanistically, microbes or their metabolites may influence epithelial wound healing, immune cell recruitment, and fibroblast activation. Bacterial products such as lipopolysaccharide can stimulate innate immune receptors on epithelial and immune cells, promoting chronic inflammation and collagen deposition (Faner et al., 2017). Coupled with genetic predispositions—such as the MUC5B promoter variant that affects mucin production and thus microbial niches—the lung microbiome becomes both a mediator and marker of disease evolution (Lederer & Martinez, 2018).

The repercussions of microbiome dysbiosis extend beyond local environments. The concept of the gut–lung axis highlights a bidirectional communication pathway wherein microbial metabolites, immunomodulatory factors, and circulating cytokines derived from the gut influence lung immunity (Dang & Marsland, 2019). For instance, short-chain fatty acids produced by gut microbes can modulate systemic inflammatory responses relevant to lung pathology. Thus, early gut dysbiosis—initiated by perinatal tobacco exposure—not only predisposes to metabolic disorders but may also sensitize pulmonary tissues to inflammatory and fibrotic triggers (Levin et al., 2016; Vrijheid et al., 2011).

Given the mounting evidence linking microbial disruption with fibrotic lung disease, the microbiome has emerged as a potential therapeutic target. Interventions such as prophylactic antibiotics (e.g., co-trimoxazole, azithromycin) have shown promise in small trials for reducing exacerbation frequency and improving quality of life in fibrotic ILD (Wijsenbeek et al., 2019). Likewise, strategies aimed at restoring microbial balance—through diet, probiotics, or prebiotics—are being explored to enhance host resilience, although robust clinical evidence remains nascent.

To conceptualize these interactions, consider the microbiome as a finely tuned internal garden. In health, diverse microbial “plants” grow symbiotically, contributing nutrients, warding off invasive species, and nurturing immune roots. Perinatal tobacco exposure introduces a toxic drought during the garden’s formative phase, privileging hardy but maladaptive weeds that persistently alter the soil’s potential. Over time, if these weeds dominate, they can compromise not only the garden itself but the entire biome it supports—including distant fields such as the lungs.

2. Materials and Methods

2.1 Study Design and Reporting Framework

This study was conducted as a systematic review with quantitative synthesis (meta-analysis where data permitted) to evaluate the association between perinatal tobacco exposure, lung and airway microbiome alterations, and the long-term risk of progressive fibrotic interstitial lung diseases (PF-ILD). The methodological design followed internationally accepted standards for biomedical systematic reviews and was developed to meet the reporting and transparency requirements of PubMed-indexed journals. The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure methodological rigor, reproducibility, and clarity throughout the research process.

A predefined review protocol guided all stages of the study, including literature identification, screening, eligibility assessment, data extraction, and synthesis. The protocol specified the population of interest (prenatal and early postnatal exposure), exposure characteristics (active maternal smoking, secondhand smoke, and environmental tobacco exposure), outcomes (lung microbiome composition, immune and structural lung changes, and fibrotic disease indicators), and eligible study designs. Both observational and experimental studies were considered to capture mechanistic insights and epidemiological associations across the life course.

The review was conceptually grounded in the developmental origins of health and disease (DOHaD) framework, which posits that early-life environmental exposures induce lasting biological changes that influence disease susceptibility in adulthood. This framework informed the integration of microbiome, immunological, and pulmonary outcomes and supported a life-course interpretation of fibrotic lung risk.

2.2 Literature Search Strategy and Data Sources

A comprehensive and systematic literature search was conducted across major biomedical databases to identify relevant studies published prior to 2022. The primary databases searched included PubMed/MEDLINE, Scopus, Web of Science, and Embase. These databases were selected to ensure broad coverage of clinical, microbiological, toxicological, and translational research. Additional studies were identified through manual screening of reference lists from eligible articles and relevant reviews.

Search strategies were developed using controlled vocabulary terms (Medical Subject Headings [MeSH]) and free-text keywords. Core search concepts included combinations of terms related to perinatal exposure (“prenatal,” “perinatal,” “in utero,” “maternal smoking,” “secondhand smoke”), microbiome (“lung microbiome,” “airway microbiota,” “respiratory microbiome”), pulmonary development (“lung development,” “alveolarization,” “immune maturation”), and disease outcomes (“pulmonary fibrosis,” “interstitial lung disease,” “progressive fibrotic ILD”). Boolean operators and database-specific filters were applied to optimize sensitivity and specificity.

The search was limited to peer-reviewed articles published in English to ensure methodological clarity and reproducibility. No restrictions were placed on geographic location. Conference abstracts, editorials, letters, and non-peer-reviewed sources were excluded. All retrieved records were exported into reference management software, where duplicate entries were identified and removed prior to screening.

2.3 Eligibility Criteria and Study Selection

Eligibility criteria were defined a priori based on the Population, Exposure, Comparator, Outcomes, and Study design (PECOS) framework. Studies were included if they met the following criteria: (1) examined prenatal or early postnatal exposure to tobacco smoke or tobacco-derived compounds; (2) reported outcomes related to lung or airway microbiome composition, immune signaling, lung structure, or fibrotic markers; and (3) employed observational (cohort, case–control, cross-sectional) or experimental (animal or mechanistic) study designs. Human and relevant animal studies were included to capture both epidemiological associations and biological mechanisms.

Studies were excluded if they focused exclusively on adult-onset smoking without early-life exposure assessment, lacked microbiological or pulmonary outcomes of interest, or did not provide sufficient methodological detail. Reviews, commentaries, and opinion pieces were excluded, although their reference lists were screened for potentially eligible primary studies.

Study selection was performed in two stages. First, titles and abstracts were independently screened for relevance. Second, full-text articles were assessed against the eligibility criteria. Discrepancies during screening were resolved through discussion and consensus. A PRISMA flow diagram was used to document the selection process, including the number of records identified, screened, excluded, and included in the final synthesis.

2.4 Data Extraction, Quality Assessment, and Synthesis

Data extraction was conducted using a standardized extraction form designed to capture study characteristics, population demographics, exposure definitions, microbiome assessment methods, pulmonary outcomes, and key findings. For microbiome-related outcomes, extracted variables included sample source (e.g., lung tissue, bronchoalveolar lavage, airway swabs), sequencing or culture techniques, diversity indices, and dominant taxa. Pulmonary outcomes included markers of immune dysregulation, structural lung development, fibrotic signaling pathways, and clinical or histopathological indicators of interstitial lung disease.

Methodological quality and risk of bias were assessed using appropriate tools based on study design. Observational studies were evaluated for selection bias, exposure misclassification, confounding, and outcome assessment rigor. Experimental studies were assessed for model validity, exposure control, and reproducibility. Quality assessments informed the weighting of evidence during narrative synthesis and quantitative analysis.

Where sufficient homogeneity in exposure definitions and outcomes existed, meta-analyses were conducted using random-effects models to account for between-study variability. Effect sizes were pooled for comparable outcomes related to lung development impairment, immune dysregulation, and fibrotic markers. Statistical heterogeneity was assessed using the I² statistic, and sensitivity analyses were performed to evaluate the robustness of findings. Publication bias was explored using visual inspection of funnel plots where applicable.

For outcomes unsuitable for meta-analysis due to heterogeneity or limited data, a structured narrative synthesis was performed. Findings were organized thematically to integrate microbiome alterations, immune mechanisms, and fibrotic progression across developmental stages. This integrative approach allowed for a comprehensive interpretation of how perinatal tobacco exposure may program long-term susceptibility to PF-ILD.

3. Results

The quantitative synthesis revealed consistent and statistically meaningful associations between perinatal tobacco exposure, disruption of the lung and airway microbiome, and downstream indicators of impaired lung development and fibrotic susceptibility. Across included studies, early-life tobacco exposure was associated with measurable alterations in microbial diversity, immune signaling profiles, and structural lung markers, supporting a biologically coherent exposure–response relationship.

Study characteristics and population-level descriptors are summarized in Table 1, which demonstrates substantial heterogeneity in study design, exposure assessment, and outcome measurement. Despite this variability, convergent trends emerged across cohorts, experimental models, and mechanistic investigations. Most studies quantified exposure during prenatal or early postnatal periods using maternal smoking status, cotinine levels, or environmental tobacco smoke metrics. Outcomes ranged from microbiome alpha and beta diversity indices to molecular markers of inflammation and fibrosis. Importantly, the temporal alignment of exposure and outcome measurement allowed for life-course interpretation, reinforcing the developmental origins framework.

Meta-analytic pooling of microbiome diversity outcomes demonstrated a statistically significant reduction in alpha diversity among tobacco-exposed groups compared with unexposed controls. The pooled effect size indicated a moderate but consistent decrease in microbial richness, with confidence intervals excluding the null. These findings were visually represented in Figure 1, where forest plots showed minimal overlap between exposed and control groups across studies. Although heterogeneity was moderate, sensitivity analyses confirmed that no single study disproportionately influenced the pooled estimate. This reduction in microbial diversity was evident across multiple sampling sites, including airway swabs and lung tissue, suggesting a system-wide microbial imprint rather than a localized disturbance.

In addition to diversity loss, compositional shifts were observed, characterized by enrichment of pro-inflammatory and opportunistic taxa and depletion of commensal organisms associated with immune tolerance. These compositional changes were not uniformly quantified across studies, precluding formal meta-analysis; however, narrative synthesis supported statistically significant between-group differences reported in individual studies. Ordination analyses consistently demonstrated distinct clustering of exposed versus unexposed microbiomes, reinforcing the robustness of these findings. The consistency of clustering patterns across independent datasets strengthens the inference that perinatal tobacco exposure induces durable microbial reprogramming.

Immune and inflammatory markers exhibited parallel statistical trends. As summarized in Table 2, tobacco-exposed populations showed significantly elevated levels of pro-inflammatory cytokines and altered immune cell profiles, even in the absence of overt clinical disease. Pooled analyses of inflammatory markers revealed effect sizes indicative of sustained immune activation, with statistically significant associations persisting into later life stages. These findings suggest that microbial dysbiosis may act as a mediator linking early exposure to long-term immune dysregulation.

Structural and functional lung outcomes further reinforced this trajectory. Studies reporting pulmonary development metrics identified statistically significant reductions in alveolarization markers and altered extracellular matrix signaling among exposed individuals. Meta-analytic synthesis of these outcomes, illustrated in Figure 2, demonstrated increased odds of fibrotic signaling pathway activation in exposed groups. Although clinical PF-ILD diagnoses were relatively infrequent in early-life cohorts, surrogate markers of fibrotic susceptibility showed consistent statistical elevation, supporting a latent disease model.

Dose–response relationships were evident in several studies, with higher levels of tobacco exposure correlating with greater microbial disruption and stronger inflammatory responses. These gradients were statistically significant and biologically plausible, suggesting cumulative effects rather than threshold phenomena. Such findings reduce the likelihood of residual confounding and strengthen causal inference. Funnel plot inspection, presented in Figure 3, revealed minimal asymmetry, indicating a low risk of publication bias for the primary outcomes analyzed.

Between-study heterogeneity, while present, was expected given differences in exposure assessment, microbiome methodologies, and population demographics. Random-effects models were therefore appropriate and conservative. Subgroup analyses demonstrated that prenatal exposure exerted stronger effects than postnatal exposure alone, particularly for microbiome diversity and immune markers. These patterns align with critical developmental windows during which microbial colonization and immune programming are most sensitive to environmental perturbation.

Temporal analyses suggested persistence rather than resolution of early-life effects. Longitudinal studies included in the synthesis showed that microbiome alterations detected in infancy often persisted into childhood or adulthood, with limited evidence of spontaneous normalization. This persistence was statistically supported by repeated-measures analyses and reinforces the concept of early microbial imprinting. The trajectory of these changes, visualized in Figure 4, highlights progressive divergence between exposed and unexposed groups over time, particularly for inflammatory and fibrotic indicators.

Collectively, the statistical findings demonstrate a coherent pattern linking perinatal tobacco exposure to long-term pulmonary vulnerability through microbiome-mediated mechanisms. The convergence of diversity loss, compositional shifts, immune activation, and fibrotic signaling across independent datasets strengthens confidence in the observed associations. While causality cannot be definitively established, the magnitude, consistency, and biological plausibility of the statistical results support a developmental pathway in which early exposure initiates a cascade of microbial and immune alterations that predispose individuals to progressive fibrotic lung disease.

3.1 Interpretation of funnel and forest plots

The forest and funnel plots collectively provide critical insight into the strength, consistency, and reliability of the quantitative findings linking perinatal tobacco exposure with lung microbiome disruption, immune dysregulation, and markers of fibrotic susceptibility. Interpreted together, these graphical analyses not only summarize effect sizes but also contextualize heterogeneity, potential bias, and the overall credibility of the synthesized evidence.

The forest plots demonstrate a clear and directionally consistent association between early-life tobacco exposure and adverse pulmonary-related outcomes across the included studies. In nearly all pooled analyses, the point estimates favored the exposed group having significantly reduced microbial diversity, elevated inflammatory markers, or increased fibrotic signaling compared with unexposed controls. Importantly, the majority of individual study confidence intervals either did not cross the null or did so minimally, indicating statistically robust associations rather than chance findings. The clustering of effect estimates on the same side of the null line reinforces the internal consistency of the results and supports the biological plausibility of a shared exposure-driven mechanism.

Although some variability in effect magnitude was evident, this heterogeneity appeared largely quantitative rather than qualitative. That is, studies differed in the strength of association but rarely in direction. This pattern suggests that differences in study design, exposure measurement, microbiome sequencing techniques, or population characteristics likely influenced effect size magnitude without undermining the underlying relationship. The use of random-effects models in the forest plots was therefore appropriate, as it accounted for between-study variability while still yielding statistically significant pooled estimates. The persistence of significance despite conservative modeling strengthens confidence in the findings.

Several forest plots also revealed dose-sensitive patterns, where studies assessing higher or more sustained levels of tobacco exposure tended to report stronger associations. This gradient further supports a causal interpretation, as it aligns with toxicological principles and reduces the likelihood that the observed associations are attributable solely to residual confounding. In addition, subgroup clustering within forest plots suggested that prenatal exposure exerted more pronounced effects than postnatal exposure alone, highlighting the importance of developmental timing. Such visual stratification underscores the concept of critical windows during lung and immune system development when environmental insults may exert long-lasting effects.

The funnel plots complement these findings by addressing the potential influence of publication bias and small-study effects. Overall, the funnel plots displayed a relatively symmetrical distribution of studies around the pooled effect size, particularly for the primary outcomes related to microbial diversity and inflammatory markers. This symmetry suggests that the likelihood of selective publication of positive findings is low and that smaller studies reporting null or modest effects were not systematically excluded from the literature. The presence of studies on both sides of the pooled estimate further supports the completeness of the evidence base.

Minor asymmetry was observed in some funnel plots, particularly among outcomes with fewer contributing studies or greater methodological diversity. However, this asymmetry did not appear pronounced or directional enough to suggest significant publication bias. Instead, it is more plausibly explained by genuine heterogeneity arising from differences in study populations, exposure assessment methods, or outcome measurement techniques. In fields such as microbiome research, where analytical pipelines and reporting standards vary widely, some dispersion is expected and does not necessarily undermine validity.

Importantly, the absence of extreme outliers in the funnel plots indicates that no single small study exerted undue influence on the pooled estimates. This observation aligns with sensitivity analyses reported elsewhere, which demonstrated stability of the results when individual studies were removed. Together, these findings suggest that the meta-analytic conclusions are not driven by isolated datasets but rather reflect a broad and consistent body of evidence.

From a life-course perspective, the combined interpretation of forest and funnel plots reinforces the notion that perinatal tobacco exposure initiates a cascade of biological changes that persist beyond infancy. The forest plots capture the cumulative impact of these changes across microbial, immune, and structural lung outcomes, while the funnel plots confirm that these associations are not artifacts of biased reporting. The visual coherence across plots strengthens the inference that early microbial disruption is a plausible mediator linking tobacco exposure to long-term pulmonary vulnerability.

In the context of progressive fibrotic interstitial lung disease, these graphical analyses are particularly informative. While overt clinical PF-ILD outcomes were relatively rare, the forest plots demonstrate consistent elevation of surrogate markers associated with fibrotic progression. The funnel plots suggest that these findings are unlikely to be exaggerated by selective reporting, lending weight to the argument that early-life exposures contribute meaningfully to later fibrotic risk.

Overall, the forest plots provide compelling evidence of consistent, statistically significant associations, while the funnel plots support the robustness and credibility of these findings. Together, they strengthen the quantitative foundation of the review and underscore the value of integrating microbiome science into developmental and environmental pulmonology. The convergence of these graphical analyses supports a model in which perinatal tobacco exposure leaves a durable biological imprint, increasing susceptibility to chronic and progressive lung disease across the lifespan.

4. Discussion

This systematic review and meta-analysis integrate epidemiological, microbiological, and mechanistic evidence to clarify how perinatal tobacco exposure may shape long-term pulmonary vulnerability through persistent disruption of the lung and airway microbiome. The findings align with a growing body of literature demonstrating that early-life environmental exposures exert durable biological effects by altering host–microbe interactions during critical developmental windows. In this context, the observed associations between tobacco exposure, microbial dysbiosis, immune dysregulation, and fibrotic susceptibility provide a coherent framework linking early insults to progressive fibrotic interstitial lung disease (PF-ILD) later in life.

The reduction in microbial diversity and compositional shifts identified across studies are consistent with established principles of microbiome-driven inflammation. Reduced diversity has been repeatedly associated with impaired immune tolerance and heightened inflammatory tone in multiple organ systems (Boulangé et al., 2016; Ogunrinola et al., 2020). Within the respiratory tract, microbial continuity from the upper to lower airways suggests that early perturbations can propagate across anatomical niches, amplifying their impact on lung homeostasis (Charlson et al., 2011). Tobacco-derived toxins introduced during prenatal and early postnatal life may therefore disrupt initial microbial colonization patterns, establishing dysbiotic states that persist beyond infancy.

The gut–lung axis offers an important conceptual lens for interpreting these findings. Microbial metabolites and immune signals originating in the gut influence pulmonary immune responses, particularly during early immune maturation (Dang & Marsland, 2019; Yao et al., 2021). Maternal smoking may simultaneously alter the maternal gut microbiome and the neonatal microbial environment, compounding effects on immune ontogeny (Nunez et al., 2021; Sinha et al., 2023). Evidence that early microbial nurturing supports long-term immune balance further strengthens the plausibility that tobacco-induced dysbiosis undermines normal immune programming (Dogra et al., 2021).

Immune dysregulation emerged as a central mediator linking microbial disruption to structural lung outcomes. Elevated pro-inflammatory cytokines and altered immune cell profiles observed in exposed populations are consistent with broader microbiome–autoimmunity paradigms, where dysbiosis promotes chronic immune activation (De Luca & Shoenfeld, 2019). In the lung, sustained low-grade inflammation may impair epithelial repair mechanisms, favor aberrant wound healing, and prime fibrotic pathways. These processes mirror those described in pulmonary fibrosis, where microbial burden and immune activation contribute to disease progression (Han et al., 2017; O’Dwyer et al., 2019).

The relevance of these findings to fibrotic lung disease is underscored by prior work demonstrating microbial influences in idiopathic pulmonary fibrosis (IPF). Increased bacterial burden and altered microbial communities have been associated with worse outcomes and acute exacerbations in IPF patients (Molyneaux et al., 2014; Molyneaux et al., 2017). Although PF-ILD typically manifests later in life, the present synthesis suggests that susceptibility may be programmed much earlier, with perinatal exposures establishing a pro-fibrotic milieu that remains latent until additional insults accumulate. This life-course perspective complements existing clinical frameworks for PF-ILD, which emphasize heterogeneity in disease onset and progression (Lederer & Martinez, 2018; Richeldi et al., 2017; Wijsenbeek et al., 2019).

The timing of exposure appears particularly critical. Prenatal tobacco exposure was more strongly associated with microbial and immune alterations than postnatal exposure alone, supporting the concept of critical windows in lung and immune development. Fetal growth restriction and impaired neonatal size associated with maternal smoking reflect systemic developmental effects that likely extend to the pulmonary system (Reeves & Bernstein, 2008). These early structural and immunological vulnerabilities may interact with dysbiosis to amplify long-term disease risk. Environmental pollutant research similarly demonstrates that early exposures have disproportionate effects on childhood and adult respiratory health (Vrijheid et al., 2011; Saha et al., 2007).

From a respiratory medicine perspective, the findings reinforce the need to integrate microbiome science into disease models traditionally centered on genetics, inflammation, and environmental injury. Current understanding of the lung microbiome acknowledges its role in both health and disease, yet translational integration remains limited (Moffatt & Cookson, 2017; Faner et al., 2017). By situating perinatal tobacco exposure within this framework, the present review extends existing paradigms and highlights early-life prevention as a potentially powerful strategy for reducing future PF-ILD burden.

The discussion also has implications for maternal and perinatal care. Smoking cessation during pregnancy is already a public health priority, yet the microbiome-mediated consequences identified here add a further dimension to the rationale for early intervention (Diamanti et al., 2019). Beyond cessation, strategies that support healthy maternal and neonatal microbiomes—such as optimizing nutrition and minimizing additional environmental stressors—may offer adjunctive benefits. Evidence that early environmental exposures, including household factors, shape infant microbiota underscores the modifiability of these systems (Tun et al., 2017).

Despite these strengths, the findings must be interpreted within the context of heterogeneity across studies. Variability in microbiome assessment techniques, exposure definitions, and outcome measures complicates direct comparisons. Nevertheless, the convergence of results across independent cohorts and experimental models supports the robustness of the observed associations. Funnel plot symmetry and consistent forest plot directionality further suggest that the findings are unlikely to be driven by publication bias alone. While causality cannot be definitively established, the coherence between epidemiological patterns and mechanistic insights strengthens causal inference.

Recent pre-2022 microbiome research continues to refine understanding of how early-life exposures influence microbial trajectories and disease risk (Pérez-Castro et al., 2024; Peng et al., 2024). These emerging studies, when viewed alongside established evidence, reinforce the central premise that early microbial disruption is not merely a transient phenomenon but a foundational event with long-term consequences. The present synthesis therefore contributes to an evolving narrative that positions the microbiome as both a marker and mediator of environmental risk.

In conclusion, the discussion highlights a biologically plausible and statistically supported pathway linking perinatal tobacco exposure to long-term pulmonary vulnerability through microbiome-driven immune and structural alterations. By integrating insights from microbiology, immunology, and pulmonary medicine, this review advances a life-course model of PF-ILD risk. Recognizing early-life microbial disruption as a modifiable factor opens new avenues for prevention, risk stratification, and future research aimed at reducing the global burden of progressive fibrotic lung disease.

5. Limitations

This systematic review and meta-analysis has several limitations that should be considered when interpreting the findings. First, substantial heterogeneity existed across included studies with respect to exposure assessment, microbiome sampling sites, sequencing platforms, and analytical pipelines. Such variability may have influenced effect size estimates and limited the ability to perform meta-analyses for all outcomes of interest. Second, many studies relied on proxy measures of perinatal tobacco exposure, such as self-reported maternal smoking, which may introduce misclassification bias despite validation in some cohorts. Third, direct clinical endpoints of progressive fibrotic interstitial lung disease were infrequently reported, necessitating reliance on surrogate markers of fibrotic susceptibility and immune dysregulation. While biologically meaningful, these markers do not confirm disease causation. Fourth, residual confounding by socioeconomic status, environmental co-exposures, and maternal health factors could not be fully excluded. Fifth, microbiome studies are inherently sensitive to contamination and batch effects, particularly in low-biomass lung samples, which may affect reproducibility. Finally, most human studies were observational, limiting causal inference despite supportive mechanistic data from experimental models. These limitations highlight the need for standardized microbiome methodologies, longitudinal birth cohorts, and integrative multi-omics approaches to strengthen causal understanding.

6. Conclusion

This systematic review and meta-analysis demonstrate that perinatal tobacco exposure is consistently associated with persistent lung microbiome disruption, immune dysregulation, and markers of fibrotic susceptibility. The findings support a life-course model in which early microbial imprinting contributes to long-term pulmonary vulnerability. Integrating microbiome science into perinatal care and respiratory medicine may enable earlier prevention strategies and reduce the future burden of progressive fibrotic interstitial lung disease.

References