The periodontal–female reproductive axis and infertility: a scoping review of epidemiological and mechanistic evidence Authors: Isidora Mujica, María José Bendek, Andrés Celis, Angela FV Tarilonte, Aurea Simon-Soro, José A Arraztoa, Manuel B. Donoso, Sebastián E. Illanes, Lara J. Monteiro, Alejandra Chaparro Published: 2026 ABSTRACT BackgroundPeriodontitis is a prevalent chronic inflammatory disease with recognized systemic implications. Its potential link to female reproductive disorders, including infertility, endometriosis, and recurrent implantation failure (RIF), has attracted growing attention, though the evidence remains heterogeneous and largely observational. This scoping review aimed to map available evidence on this relationship, identify shared biological mechanisms, and highlight gaps warranting future investigation.MethodsFollowing PRISMA-ScR guidelines, four databases (PubMed/MEDLINE, Scopus, Cochrane Library (CENTRAL) and Web of Science) were searched with no date restriction, supplemented by grey literature. Two independent reviewers screened records using the PCC framework; 22 studies were included.ResultsEight epidemiological studies showed heterogeneous findings; prospective data suggest a possible modest reduction in fecundability associated with periodontitis, though effect sizes vary, confidence intervals are wide, and exposures were largely self-reported. Serological evidence is consistent with an association between periodontal pathogen exposure and unexplained infertility. Mechanistic studies (in vitro and animal models) suggest plausible biological pathways, but human clinical validation is lacking.ConclusionsCurrent evidence suggests a plausible, though unconfirmed, association between periodontal disease and female reproductive disorders. Methodological heterogeneity and the absence of interventional data preclude causal conclusions. Future research should prioritize standardized exposure assessment and randomized trials evaluating periodontal treatment before assisted reproductive technologies. Background Periodontitis, the most severe form of periodontal disease, is a chronic inflammatory disease affecting the supporting structures of the teeth with an estimated global prevalence of approximately 11% among adults [1]. Beyond its oral consequences (periodontal tissue inflammation, clinical attachment loss, and alveolar bone resorption), periodontitis is associated with systemic comorbidities including type 2 diabetes, cardiovascular disease, Alzheimer’s disease, and adverse pregnancy outcomes [2]. This systemic effect is principally attributed to chronic low-grade systemic inflammation, immune dysregulation, and the hematogenous translocation of periodontal pathogens and their byproducts into the bloodstream [3–5]. Infertility, defined as the inability to conceive after 12 months of unprotected intercourse, affects 8–12% of couples worldwide [6]. While ovulatory dysfunction, tubal obstruction, and male factors are well-established causes, systemic inflammatory processes are increasingly recognized as potential modifiable contributors to female reproductive dysfunction [7]. Recurrent implantation failure (RIF), typically defined as failure to achieve a clinical pregnancy after three or more in-vitro fertilization (IVF) transfers of high-quality embryos, is a particularly challenging condition involving both embryonic and endometrial factors, with immune dysregulation playing a central role [7]. Endometriosis, a chronic condition, affecting approximately 5–10% of reproductive-age women, is a major cause of infertility characterized by inflammation, scar tissue formation, and a range of debilitating symptoms including pelvic pain, painful menstruation and infertility, and shares immunological and oxidative stress pathways with periodontitis [8, 9]. Our research group has contributed to further foundational evidence [10, 11] linking periodontal disease to adverse pregnancy outcomes [12, 13]. Notably, our group has identified P. gingivalis, F. nucleatum, and T. denticola DNA in placental tissue from hypertensive women [14], and established gingival crevicular fluid (GCF) biomarkers, (including placental alkaline phosphatase (PLAP), soluble fms-like tyrosine kinase-1 (sFlt-1), interleukin-6 (IL-6), and metalloproteinase-8 (MMP-8), as early predictors of preeclampsia and gestational diabetes mellitus [15, 16]. Building on this evidence, we hypothesize that analogous inflammatory and microbial pathways may operate at earlier reproductive stages, influencing ovarian function, endometrial receptivity, and embryo implantation. The proposed mechanistic framework involves two interconnected pathways: (1) chronic systemic inflammation driven by periodontal cytokines (IL-1β, IL-6, TNF-α) that may disrupt hormonal signaling and endometrial immune homeostasis [5, 17, 18]; and (2) direct hematogenous translocation of periodontal pathogens or their biologically active byproducts (including bacterial derived extracellular vesicles (EVs) and outer membrane vesicles (OMVs)) to the reproductive tract [19–22]. It should be noted that this mechanistic framework draws substantially on in vitro and animal experimental evidence and on biological plausibility; direct clinical evidence in women with reproductive disorders remains limited and heterogeneous. To address the interdisciplinary complexity of periodontal medicine with reproductive outcomes, mediated by periodontal pathogens or their byproducts and immune activation, this scoping review aims to map the evidence linking periodontal disease with female reproductive disorders. Our analysis focuses on infertility, fecundability, endometriosis, and RIF, considering both natural and ART-mediated conception. Methods Protocol and registration This scoping review was conducted following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines to ensure transparency and reproducibility [23]. The protocol was registered in the Open Science Framework (registration number: 6ge9s. Protocol available at: https://osf.io/6ge9s/overview?view_only=fe8d597b2102476f9ffabc805ced50b9). Eligibility criteria Eligibility was defined using the PCC (Population, Concept, Context) framework. The population considered women of reproductive age, including those with infertility, RIF, or undergoing ART/IVF; in vitro studies and animal models, were also eligible. The concept included periodontal disease, periodontal pathogens (P. gingivalis, F. nucleatum, T. denticola), and pathogen-derived extracellular vesicles (OMVs, EVs). The context included infertility, fecundability, endometriosis, RIF, IVF outcomes, and endometrial receptivity. Eligible designs included observational, experimental, in vitro, and animal studies. Studies focused exclusively on late obstetric outcomes or evaluating male infertility only, were excluded. No date restriction was applied; English and Spanish publications were eligible. Search strategy and information sources A comprehensive electronic search was conducted in PubMed/MEDLINE, Scopus, Cochrane Library (CENTRAL), and Web of Science (Core Collection). Grey literature was searched via Google Scholar and OSF Preprints/medRxiv. The search strategy combined two Boolean blocks using MeSH and free-text terms: Block A (periodontal disease, oral microbiota, periodontal pathogens, outer membrane vesicles, extracellular vesicles) AND Block B (female infertility, endometriosis, recurrent implantation failure, IVF, fecundability, endometrial receptivity). Full search strings are provided as a supplementary appendix. The last search was executed on April 30th 2026. Selection of sources of evidence Records were managed in Rayyan for de-duplication and screening. Two reviewers (IM and AC) independently screened titles and abstracts, then full texts, against the PCC criteria. Disagreements were resolved by a third reviewer (ACh). Studies with mixed-sex samples were included when female-specific subgroup data could be extracted separately. Data charting Key data was extracted using a standardized form, including study design, population, periodontal exposure definition, reproductive outcome, effect estimates with 95% CI, microbial detection and controlled confounders. Extraction was performed independently by IM and AC, with discrepancies resolved by consensus. Critical appraisal of individual sources of evidence In accordance with the PRISMA-ScR [23] guidelines for scoping reviews focused on mapping evidence, a formal assessment of the risk of bias for each individual study was not performed. This decision is justified by the exploratory nature of the review, which aims to synthesize complex biological mechanisms and identify knowledge gaps across diverse study designs rather than establishing a definitive pooled effect size. Synthesis of results A narrative synthesis was employed. Studies were organized into four thematic domains: (A) periodontal disease and female infertility/fecundability; (B) Periodontal disease and IVF outcomes and RIF; (C) Periodontal diseases and endometriosis; and (D) Mechanistic evidence: in vitro and animal studies. Within each domain, evidence was stratified by design (observational/Mendelian randomization; in vitro/animal). No meta-analytic pooling was performed. Results Study selection A total of 287 records were identified through electronic database searching (PubMed: 68; Scopus: 34; Cochrane: 0; Web of Science: 185), and 13 additional records were identified through grey literature and manual screening. After the removal of 102 duplicates, 185 records underwent title and abstract screening. Of these, 162 were excluded based on population (n = 5), study design (n = 64), or conceptual misalignment (n = 93). For the database-seeking arm, 23 reports were sought for retrieval, of which 22 were successfully assessed for eligibility (one report was not accessible). Nine reports were excluded at this stage due to the absence of an oral health component (n = 4) or the lack of a periodontal component (n = 5). Simultaneously, 13 reports from other sources were assessed, with 4 excluded (duplicates: n = 3; survey without periodontal outcomes: n = 1). Ultimately, 22 studies met the inclusion criteria and were included in this scoping review (Databases: n = 13; Other sources: n = 9), as illustrated in the PRISMA 2020 flow diagram (Fig. 1). Inter-rater agreement at title/abstract screening showed a Cohen’s κ of 0.9. Fig. 1Prisma Flow Diagram for Scoping Review. PRISMA 2020 flow diagram summarizing the study selection process. The diagram illustrates the systematic search and screening process. A total of 287 records were identified via electronic databases and 13 through other methods. After removing duplicates and screening 185 records by title and abstract, 23 reports were assessed for eligibility. The final synthesis included 22 studies (13 from databases and 9 from other sources). Abbreviations: WoS: Web of Science; n: number of records. This diagram was generated using the PRISMA2020 R package [24] Prisma Flow Diagram for Scoping Review. PRISMA 2020 flow diagram summarizing the study selection process. The diagram illustrates the systematic search and screening process. A total of 287 records were identified via electronic databases and 13 through other methods. After removing duplicates and screening 185 records by title and abstract, 23 reports were assessed for eligibility. The final synthesis included 22 studies (13 from databases and 9 from other sources). Abbreviations: WoS: Web of Science; n: number of records. This diagram was generated using the PRISMA2020 R package [24] Characteristics of included studies The 22 included studies were published between 2005 and 2025, across twelve countries (USA, China, Croatia, Portugal, Turkey, Japan, Australia, Finland, Germany, Argentina, South Korea, New Zealand). Study designs included seven cross-sectional case-control studies, four prospective cohort studies, one bidirectional Mendelian randomization (MR) study, one cross-sectional population-based analysis (NHANES), one pilot observational study, four in vitro studies using human cell lines, one in vitro study using primary human decidual cells, and one animal model study. No eligible randomized controlled trials were identified. One systematic review met all PCC criteria except the design criterion; its constituent primary studies were retrieved and screened independently; two met the PCC criteria and are included as primary studies. Detailed characteristics are presented in Table 1. Table 1Description of the included studiesAuthorYearCountryDesign n Age (years)DisorderPeriodontal exposureKey outcome/effect estimateDomain A - Periodontal disease and female infertility/fecundability (n = 8) Bond et al. [25]2021USAPC2,764 women21–45Infertility / FecundabilitySelf-reported periodontitis (diagnosis, treatment, tooth mobility)FR = 0.89 (95% CI: 0.75–1.06) for diagnosis; FR = 0.79 (95% CI: 0.67–0.94) for treatment; FR = 0.71 (95% CI: 0.44–1.16) for tooth mobility. Possible dose-response across severity parameters. Huo et al. [26]2021ChinaPC3,915 women< 25 to ≥ 40Fecundability / TTPHalitosis and gingival bleeding (proxy for oral dysbiosis)FR = 0.83 (95% CI: 0.69–1.00) for halitosis + gingival bleeding. Adjusted for age, BMI, lifestyle factors. Sumilin et al. [27]2022CroatiaCC100 (50 cases, 50 controls)Mean 36.1 (cases) / 37.0 (controls)Idiopathic infertilityaMMP-8 salivary levels (subclinical oral inflammation)OR = 1.03 (95% CI: 0.99–1.07; p < 0.05). Elevated aMMP-8 in infertile women without clinically detectable periodontitis, suggesting subclinical oral inflammatory activity. Machado et al. [28]2020PortugalCC36 (18 cases, 18 controls)Mean 33.9 (cases) / 34.1 (controls)Idiopathic infertilityPeriodontitis (clinical examination)OR = 1.75 (95% CI: 0.40–7.71). Higher prevalence of periodontitis in fertility clinic group (6/18 vs. 4/18 controls). Non-significant, likely due to limited sample size. Telatar et al. [29]2021TurkeyCC100 (50 cases, 50 controls)21–39Unexplained infertilityPeriodontitis (PISA, PESA, BOP, CAL, PD)Infertile women showed significantly higher PISA (p = 0.002) and PESA (p = 0.005). Higher rates of moderate (18% vs. 8%) and severe periodontitis (16% vs. 6%). Adjusted OR = 0.24 (95% CI: 0.10–0.58; p = 0.001) for infertility on odds of periodontitis; direction reversed in multivariate model. Kamei-Nagata et al. [30]2025JapanCC146 (76 naturally pregnant, 70 unexplained infertility)Mean 33 / 35Unexplained infertilitySerum IgG against 7 periodontal pathogen strains (no clinical exam)P. gingivalis W83: OR = 3.31 (95% CI: 1.30–11.17; p = 0.029). P. intermedia ATCC25611: OR = 4.50 (95% CI: 1.55–15.08; p = 0.009). P. gingivalis FDC381: inverse association OR = 0.55 (95% CI: 0.31–0.86; p = 0.020). Reflects prior/current exposure, not active clinical disease. Hart et al. [31]2012AustraliaPC(sub-analysis RCT - SMILE)1,956 womenNRInfertility / TTCPeriodontitis (clinical assessment, 2nd trimester)OR = 2.88 (95% CI: 1.62–5.12; p < 0.001) for TTC > 12 months. Effect significant in non-White women; not significant in White women. Retrospective assessment during pregnancy; risk of pregnancy gingivitis misclassification. Bond et al. [32]2023USAPC (PRESTO)3,444 women21–45Spontaneous abortionSelf-reported periodontitis (diagnosis, treatment, tooth mobility)HR = 0.97 (95% CI: 0.76–1.23) for diagnosis; HR = 1.01 (95% CI: 0.79–1.27) for treatment; HR = 1.38 (95% CI: 0.88–2.14) for tooth mobility. Positive association between tooth mobility and SAB risk; bias analysis suggests attenuation toward null. Domain B – Periodontal Disease and IVF outcomes and RIF ( n = 1) Zhang et al. [33]2025JapanCC80 (40 CE, 40 non-CE; all RIF)Mean 32.9RIF / Chronic endometritisEndometrial microbiota (16 S rRNA sequencing)Prevotella and Streptococcus associated with non-pregnancy; Phyllobacterium, Gardnerella, Enterococcus with miscarriage. Lactobacillus, Achromobacter positively associated with clinical pregnancy. Domain C - Periodontal disease and endometriosis ( n = 3) Kavoussi et al. [34]2009USACross-sectional (NHANES)4,136 women18–50EndometriosisPeriodontitis / gingivitis (clinical NHANES data)Adjusted OR = 1.57 (95% CI: 1.06–2.33) for periodontitis and endometriosis. Largest population-based dataset in this review. Cross-sectional; temporal direction cannot be established. Jin et al. [35]2024FinlandBMRGWAS summary data (8,828 IVW-equivalent)≥ 18EndometriosisPeriodontitis (genetic instruments - GWAS)Periodontitis → pelvic peritoneal endometriosis: OR = 1.079 (95% CI: 1.016–1.146; p = 0.014). Reverse (endometriosis → periodontitis): OR = 1.394 (95% CI: 1.006–1.930; p = 0.046). Bidirectional association supported; modest effect sizes; caution in causal interpretation. Hicks et al. [36]2025AustraliaPC64 (9 controls, 24 no-endometriosis, 21 confirmed endometriosis)Mean 31.5 / 35.8 / 35.9EndometriosisOral, vaginal, stool microbiome (16 S rRNA)Fusobacterium enriched in oral samples of moderate-to-severe endometriosis vs. mild/minimal disease. Oral dysbiosis pattern; not direct periodontal assessment. No clinical periodontal examination performed.Domain D - Mechanistic evidence: in vitro and animal studies (n = 10) Hirohata et al. [37]2017JapanIn vitro (HTR-8/SVneo)Cell line—Trophoblast invasion / ImplantationP. gingivalis culture supernatant (PG-sup) — soluble factorsDose-dependent suppression of trophoblast invasion: 85% inhibition at 0.01% PG-sup; 47% at 0.1% PG-sup. No direct cytotoxicity. Disruption of spheroid formation and microvilli structure. Mechanism: soluble factors, not direct bacterial contact. Komine-Aizawa et al. [38]2015JapanIn vitro (HTR-8/SVneo)Cell line—Trophoblast invasion / ImplantationP. gingivalis LPS (PGLPS) ± nicotinePGLPS alone: no effect on invasion. PGLPS + nicotine: significant inhibition of invasion (p < 0.01) via p38 MAPK pathway. Suppression of MMP-2/MMP-9. Smoking as a synergistic factor-relevant confounding disclosure. Riewe et al. [39]2010USAIn vitro (microarray; HTR-8)Cell line—Trophoblast signaling / ImplantationP. gingivalis ATCC 33,277 (MOI 200)≥ 2,045 genes differentially expressed (p < 0.001). MEK3-p38-Max MAPK pathway upregulated (confirmed by immunoblot). IL-1β and IL-8 elevated at 2 h and 24 h post-infection. Establishes global transcriptomic impact of P. gingivalis on trophoblast function. Ren et al. [40]2016ChinaIn vitro (HTR8/SVneo)Cell line—Trophoblast inflammation / ApoptosisP. gingivalis ATCC 33,277 (MOI 50–200)IL-8 and IFN-γ secretion MOI-dependent. Apoptosis confirmed by caspase-3 activation (Annexin V/PI flow cytometry). ERK1/2 and p38 MAPK primary mediators. PGLPS and heat-killed P. gingivalis also induced IL-8 production. Keelan et al. [41]2010Australia/NZIn vitro (primary decidual cells)Primary human decidual cells (n = 5 placentas)—Decidual inflammation / Endometrial receptivityP. gingivalis, F. nucleatum, A. actinomycetemcomitans, E. coli LPS (dose-response)All bacteria elicited decidual inflammatory response. Rank order of potency: Ec > Aa> Fn > Pg. 18 genes significantly upregulated (including TNF-α, CXCR1, IL-1β, IL-6R). P. gingivalis LPS structurally atypical — weakest TLR4 agonist; potential TLR4 antagonism documented. Tavarna et al. [42]2022USAAnimal model (Wistar/SD rats)34–40 dams/group (GD11 and GD18)—Spiral artery remodeling / Early pregnancyP. gingivalis A7UF — rat periodontitis modelGD11: reduced spiral arterial lumen (p < 0.02); increased VSMC retention (p < 0.003); decreased EVT invasion (29% vs. 68% controls; p ≤ 0.002). IL-18 + uNK cells reduced (p < 0.001); Htra1 overexpressed. By GD18: IL-18/Htra1 normalized; TNF + T cells reduced (p ≤ 0.048). Both trophoblast-independent and dependent phases of remodeling disrupted. Einenkel et al. [43]2024GermanyIn vitro (THP-1 macrophages + HTR-8/SVneo)Cell line (macrophages + trophoblasts)—Trophoblast tube formation / ImplantationInactivated F. nucleatum (bacteria: cell ratio 0.1 and 1)Low F. nucleatum (ratio 0.1): increased HIF-1α (p < 0.05) and VEGF-A secretion in M2c macrophages; supported trophoblast tube formation. High F. nucleatum or M1 macrophages: impaired trophoblast migration and tube formation. Dose-dependent bidirectional effect on implantation-related functions. Lara et al. [44]2023ArgentinaIn vitro (HTR-8/SVneo) + animal model (C57BL/6 mice)Cell line / pregnant mice—Trophoblast metabolism / Placentation / Adverse pregnancy outcomeP. gingivalis OMVs (1 µg/mL in vitro; 5 µg/kg i.p. in vivo)In vitro: OMVs internalized via clathrin/caveolin-mediated endocytosis. Reduced glucose uptake (2-NBDG), lactate release, GLUT1/GLUT3 expression, and total ROS; mitochondrial activity sustained → metabolic resting state. Migration (wound healing) and invasion (Matrigel) significantly inhibited; MMP-9 and IL-6 decreased; TIMP1/2 increased. In vivo: OMVs reached implantation sites within 6 h. Reduced placental and fetal weight (p < 0.05); decreased placental GLUT1 expression; altered labyrinth layer histology. Identifies OMV-driven metabolic reprogramming as a mechanistic link between periodontitis and placental insufficiency. Lee et al. [45]2021South KoreaCC (microbiome analysis of EVs in peritoneal fluid)90 women (45 ovarian endometrioma, 45 surgical controls)36.2 ± 1.3 (EM) / 39.4 ± 1.1 (controls)Endometriosis (stages III–IV) / Peritoneal microbiome dysbiosisPeritoneal fluid EVs (microbiota-derived; 16 S rDNA V3–V4 NGS; NTA for size validation)Significant differences in microbiome composition of peritoneal fluid EVs between endometriosis and controls at order, family, and genus levels (beta diversity; Bray–Curtis p < 0.001). Increased in endometriosis: Acinetobacter (p = 0.002), Pseudomonas (p = 0.010), Streptococcus (p = 0.018), Enhydrobacter (p = 0.030), Moraxellaceae (p < 0.001). Decreased: Propionibacterium (p = 0.011), Actinomyces (p = 0.023), Rothia (p = 0.019), Actinobacteria (phylum; p = 0.048). Alpha diversity not significantly different. Supports bacterial contamination hypothesis and role of peritoneal dysbiosis in endometriosis pathogenesis. Bartruff et al. [46]2005USAIn vitro (HGF + HUVEC)Cell lines (HGF, HUVEC)—Periodontal tissue cell proliferation / AngiogenesisP. gingivalis ATCC 33,277 OMVs (0–40 µg/mL protein; NH2SO4 precipitation + ultracentrifugation)Dose-dependent inhibition of ³H-thymidine incorporation in HGF (IC50 ≈9 µg/mL) and HUVEC (IC50 ≈4.5 µg/mL); endothelial cells twice as susceptible as fibroblasts. Capillary tube formation by HUVEC on ECM suppressed ∼80% at 5 µg/mL OMV at 18 h. Protease inhibitor cocktail did not reverse effects; heat-inactivation (90 °C, 5–10 min) significantly reduced suppression, indicating a protein factor. LPS at equivalent concentrations did not suppress tube formation. OMV affected endothelial cell migration/aggregation but not initial attachment. Supports OMV-mediated inhibition of revascularization as a contributor to chronic periodontal tissue destruction.Abbreviations: SR Systematic review, OR Odds ratio, TTC Time taken to conceive, HR Hazard ratio, CC Cross-sectional case-control study, TTP Time to pregnancy, FR Fecundability ratio, RIF Recurrent implantation failure, PC Prospective cohort study, BMR Bidirectional two-sample Mendelian randomization, EM Endometriosis, Pg Porphyromonas gingivalis. Pi: Prevotella intermedia, OMV Outer membrane vesicle[s], EV Extracellular vesicle[s], HGF Human gingival fibroblasts, HUVEC Human umbilical vein endothelial cells, ECM (extracellular matrix), IC50: half-maximal inhibitory concentration, 2-NBDG: 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose; fluorescent glucose analogue, ROS Reactive oxygen species, GLUT1/GLUT3 Glucose transporters 1 and 3, NGS Next-generation sequencing, NTA Nanoparticle tracking analysis, PF Peritoneal fluid, i.p intraperitoneal, LPS Lipopolysaccharide Table 1Description of the included studiesAuthorYearCountryDesign n Age (years)DisorderPeriodontal exposureKey outcome/effect estimateDomain A - Periodontal disease and female infertility/fecundability (n = 8) Bond et al. [25]2021USAPC2,764 women21–45Infertility / FecundabilitySelf-reported periodontitis (diagnosis, treatment, tooth mobility)FR = 0.89 (95% CI: 0.75–1.06) for diagnosis; FR = 0.79 (95% CI: 0.67–0.94) for treatment; FR = 0.71 (95% CI: 0.44–1.16) for tooth mobility. Possible dose-response across severity parameters. Huo et al. [26]2021ChinaPC3,915 women< 25 to ≥ 40Fecundability / TTPHalitosis and gingival bleeding (proxy for oral dysbiosis)FR = 0.83 (95% CI: 0.69–1.00) for halitosis + gingival bleeding. Adjusted for age, BMI, lifestyle factors. Sumilin et al. [27]2022CroatiaCC100 (50 cases, 50 controls)Mean 36.1 (cases) / 37.0 (controls)Idiopathic infertilityaMMP-8 salivary levels (subclinical oral inflammation)OR = 1.03 (95% CI: 0.99–1.07; p < 0.05). Elevated aMMP-8 in infertile women without clinically detectable periodontitis, suggesting subclinical oral inflammatory activity. Machado et al. [28]2020PortugalCC36 (18 cases, 18 controls)Mean 33.9 (cases) / 34.1 (controls)Idiopathic infertilityPeriodontitis (clinical examination)OR = 1.75 (95% CI: 0.40–7.71). Higher prevalence of periodontitis in fertility clinic group (6/18 vs. 4/18 controls). Non-significant, likely due to limited sample size. Telatar et al. [29]2021TurkeyCC100 (50 cases, 50 controls)21–39Unexplained infertilityPeriodontitis (PISA, PESA, BOP, CAL, PD)Infertile women showed significantly higher PISA (p = 0.002) and PESA (p = 0.005). Higher rates of moderate (18% vs. 8%) and severe periodontitis (16% vs. 6%). Adjusted OR = 0.24 (95% CI: 0.10–0.58; p = 0.001) for infertility on odds of periodontitis; direction reversed in multivariate model. Kamei-Nagata et al. [30]2025JapanCC146 (76 naturally pregnant, 70 unexplained infertility)Mean 33 / 35Unexplained infertilitySerum IgG against 7 periodontal pathogen strains (no clinical exam)P. gingivalis W83: OR = 3.31 (95% CI: 1.30–11.17; p = 0.029). P. intermedia ATCC25611: OR = 4.50 (95% CI: 1.55–15.08; p = 0.009). P. gingivalis FDC381: inverse association OR = 0.55 (95% CI: 0.31–0.86; p = 0.020). Reflects prior/current exposure, not active clinical disease. Hart et al. [31]2012AustraliaPC(sub-analysis RCT - SMILE)1,956 womenNRInfertility / TTCPeriodontitis (clinical assessment, 2nd trimester)OR = 2.88 (95% CI: 1.62–5.12; p < 0.001) for TTC > 12 months. Effect significant in non-White women; not significant in White women. Retrospective assessment during pregnancy; risk of pregnancy gingivitis misclassification. Bond et al. [32]2023USAPC (PRESTO)3,444 women21–45Spontaneous abortionSelf-reported periodontitis (diagnosis, treatment, tooth mobility)HR = 0.97 (95% CI: 0.76–1.23) for diagnosis; HR = 1.01 (95% CI: 0.79–1.27) for treatment; HR = 1.38 (95% CI: 0.88–2.14) for tooth mobility. Positive association between tooth mobility and SAB risk; bias analysis suggests attenuation toward null. Domain B – Periodontal Disease and IVF outcomes and RIF ( n = 1) Zhang et al. [33]2025JapanCC80 (40 CE, 40 non-CE; all RIF)Mean 32.9RIF / Chronic endometritisEndometrial microbiota (16 S rRNA sequencing)Prevotella and Streptococcus associated with non-pregnancy; Phyllobacterium, Gardnerella, Enterococcus with miscarriage. Lactobacillus, Achromobacter positively associated with clinical pregnancy. Domain C - Periodontal disease and endometriosis ( n = 3) Kavoussi et al. [34]2009USACross-sectional (NHANES)4,136 women18–50EndometriosisPeriodontitis / gingivitis (clinical NHANES data)Adjusted OR = 1.57 (95% CI: 1.06–2.33) for periodontitis and endometriosis. Largest population-based dataset in this review. Cross-sectional; temporal direction cannot be established. Jin et al. [35]2024FinlandBMRGWAS summary data (8,828 IVW-equivalent)≥ 18EndometriosisPeriodontitis (genetic instruments - GWAS)Periodontitis → pelvic peritoneal endometriosis: OR = 1.079 (95% CI: 1.016–1.146; p = 0.014). Reverse (endometriosis → periodontitis): OR = 1.394 (95% CI: 1.006–1.930; p = 0.046). Bidirectional association supported; modest effect sizes; caution in causal interpretation. Hicks et al. [36]2025AustraliaPC64 (9 controls, 24 no-endometriosis, 21 confirmed endometriosis)Mean 31.5 / 35.8 / 35.9EndometriosisOral, vaginal, stool microbiome (16 S rRNA)Fusobacterium enriched in oral samples of moderate-to-severe endometriosis vs. mild/minimal disease. Oral dysbiosis pattern; not direct periodontal assessment. No clinical periodontal examination performed.Domain D - Mechanistic evidence: in vitro and animal studies (n = 10) Hirohata et al. [37]2017JapanIn vitro (HTR-8/SVneo)Cell line—Trophoblast invasion / ImplantationP. gingivalis culture supernatant (PG-sup) — soluble factorsDose-dependent suppression of trophoblast invasion: 85% inhibition at 0.01% PG-sup; 47% at 0.1% PG-sup. No direct cytotoxicity. Disruption of spheroid formation and microvilli structure. Mechanism: soluble factors, not direct bacterial contact. Komine-Aizawa et al. [38]2015JapanIn vitro (HTR-8/SVneo)Cell line—Trophoblast invasion / ImplantationP. gingivalis LPS (PGLPS) ± nicotinePGLPS alone: no effect on invasion. PGLPS + nicotine: significant inhibition of invasion (p < 0.01) via p38 MAPK pathway. Suppression of MMP-2/MMP-9. Smoking as a synergistic factor-relevant confounding disclosure. Riewe et al. [39]2010USAIn vitro (microarray; HTR-8)Cell line—Trophoblast signaling / ImplantationP. gingivalis ATCC 33,277 (MOI 200)≥ 2,045 genes differentially expressed (p < 0.001). MEK3-p38-Max MAPK pathway upregulated (confirmed by immunoblot). IL-1β and IL-8 elevated at 2 h and 24 h post-infection. Establishes global transcriptomic impact of P. gingivalis on trophoblast function. Ren et al. [40]2016ChinaIn vitro (HTR8/SVneo)Cell line—Trophoblast inflammation / ApoptosisP. gingivalis ATCC 33,277 (MOI 50–200)IL-8 and IFN-γ secretion MOI-dependent. Apoptosis confirmed by caspase-3 activation (Annexin V/PI flow cytometry). ERK1/2 and p38 MAPK primary mediators. PGLPS and heat-killed P. gingivalis also induced IL-8 production. Keelan et al. [41]2010Australia/NZIn vitro (primary decidual cells)Primary human decidual cells (n = 5 placentas)—Decidual inflammation / Endometrial receptivityP. gingivalis, F. nucleatum, A. actinomycetemcomitans, E. coli LPS (dose-response)All bacteria elicited decidual inflammatory response. Rank order of potency: Ec > Aa> Fn > Pg. 18 genes significantly upregulated (including TNF-α, CXCR1, IL-1β, IL-6R). P. gingivalis LPS structurally atypical — weakest TLR4 agonist; potential TLR4 antagonism documented. Tavarna et al. [42]2022USAAnimal model (Wistar/SD rats)34–40 dams/group (GD11 and GD18)—Spiral artery remodeling / Early pregnancyP. gingivalis A7UF — rat periodontitis modelGD11: reduced spiral arterial lumen (p < 0.02); increased VSMC retention (p < 0.003); decreased EVT invasion (29% vs. 68% controls; p ≤ 0.002). IL-18 + uNK cells reduced (p < 0.001); Htra1 overexpressed. By GD18: IL-18/Htra1 normalized; TNF + T cells reduced (p ≤ 0.048). Both trophoblast-independent and dependent phases of remodeling disrupted. Einenkel et al. [43]2024GermanyIn vitro (THP-1 macrophages + HTR-8/SVneo)Cell line (macrophages + trophoblasts)—Trophoblast tube formation / ImplantationInactivated F. nucleatum (bacteria: cell ratio 0.1 and 1)Low F. nucleatum (ratio 0.1): increased HIF-1α (p < 0.05) and VEGF-A secretion in M2c macrophages; supported trophoblast tube formation. High F. nucleatum or M1 macrophages: impaired trophoblast migration and tube formation. Dose-dependent bidirectional effect on implantation-related functions. Lara et al. [44]2023ArgentinaIn vitro (HTR-8/SVneo) + animal model (C57BL/6 mice)Cell line / pregnant mice—Trophoblast metabolism / Placentation / Adverse pregnancy outcomeP. gingivalis OMVs (1 µg/mL in vitro; 5 µg/kg i.p. in vivo)In vitro: OMVs internalized via clathrin/caveolin-mediated endocytosis. Reduced glucose uptake (2-NBDG), lactate release, GLUT1/GLUT3 expression, and total ROS; mitochondrial activity sustained → metabolic resting state. Migration (wound healing) and invasion (Matrigel) significantly inhibited; MMP-9 and IL-6 decreased; TIMP1/2 increased. In vivo: OMVs reached implantation sites within 6 h. Reduced placental and fetal weight (p < 0.05); decreased placental GLUT1 expression; altered labyrinth layer histology. Identifies OMV-driven metabolic reprogramming as a mechanistic link between periodontitis and placental insufficiency. Lee et al. [45]2021South KoreaCC (microbiome analysis of EVs in peritoneal fluid)90 women (45 ovarian endometrioma, 45 surgical controls)36.2 ± 1.3 (EM) / 39.4 ± 1.1 (controls)Endometriosis (stages III–IV) / Peritoneal microbiome dysbiosisPeritoneal fluid EVs (microbiota-derived; 16 S rDNA V3–V4 NGS; NTA for size validation)Significant differences in microbiome composition of peritoneal fluid EVs between endometriosis and controls at order, family, and genus levels (beta diversity; Bray–Curtis p < 0.001). Increased in endometriosis: Acinetobacter (p = 0.002), Pseudomonas (p = 0.010), Streptococcus (p = 0.018), Enhydrobacter (p = 0.030), Moraxellaceae (p < 0.001). Decreased: Propionibacterium (p = 0.011), Actinomyces (p = 0.023), Rothia (p = 0.019), Actinobacteria (phylum; p = 0.048). Alpha diversity not significantly different. Supports bacterial contamination hypothesis and role of peritoneal dysbiosis in endometriosis pathogenesis. Bartruff et al. [46]2005USAIn vitro (HGF + HUVEC)Cell lines (HGF, HUVEC)—Periodontal tissue cell proliferation / AngiogenesisP. gingivalis ATCC 33,277 OMVs (0–40 µg/mL protein; NH2SO4 precipitation + ultracentrifugation)Dose-dependent inhibition of ³H-thymidine incorporation in HGF (IC50 ≈9 µg/mL) and HUVEC (IC50 ≈4.5 µg/mL); endothelial cells twice as susceptible as fibroblasts. Capillary tube formation by HUVEC on ECM suppressed ∼80% at 5 µg/mL OMV at 18 h. Protease inhibitor cocktail did not reverse effects; heat-inactivation (90 °C, 5–10 min) significantly reduced suppression, indicating a protein factor. LPS at equivalent concentrations did not suppress tube formation. OMV affected endothelial cell migration/aggregation but not initial attachment. Supports OMV-mediated inhibition of revascularization as a contributor to chronic periodontal tissue destruction.Abbreviations: SR Systematic review, OR Odds ratio, TTC Time taken to conceive, HR Hazard ratio, CC Cross-sectional case-control study, TTP Time to pregnancy, FR Fecundability ratio, RIF Recurrent implantation failure, PC Prospective cohort study, BMR Bidirectional two-sample Mendelian randomization, EM Endometriosis, Pg Porphyromonas gingivalis. Pi: Prevotella intermedia, OMV Outer membrane vesicle[s], EV Extracellular vesicle[s], HGF Human gingival fibroblasts, HUVEC Human umbilical vein endothelial cells, ECM (extracellular matrix), IC50: half-maximal inhibitory concentration, 2-NBDG: 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose; fluorescent glucose analogue, ROS Reactive oxygen species, GLUT1/GLUT3 Glucose transporters 1 and 3, NGS Next-generation sequencing, NTA Nanoparticle tracking analysis, PF Peritoneal fluid, i.p intraperitoneal, LPS Lipopolysaccharide Description of the included studies PC (sub-analysis RCT - SMILE) Abbreviations: SR Systematic review, OR Odds ratio, TTC Time taken to conceive, HR Hazard ratio, CC Cross-sectional case-control study, TTP Time to pregnancy, FR Fecundability ratio, RIF Recurrent implantation failure, PC Prospective cohort study, BMR Bidirectional two-sample Mendelian randomization, EM Endometriosis, Pg Porphyromonas gingivalis. Pi: Prevotella intermedia, OMV Outer membrane vesicle[s], EV Extracellular vesicle[s], HGF Human gingival fibroblasts, HUVEC Human umbilical vein endothelial cells, ECM (extracellular matrix), IC50: half-maximal inhibitory concentration, 2-NBDG: 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose; fluorescent glucose analogue, ROS Reactive oxygen species, GLUT1/GLUT3 Glucose transporters 1 and 3, NGS Next-generation sequencing, NTA Nanoparticle tracking analysis, PF Peritoneal fluid, i.p intraperitoneal, LPS Lipopolysaccharide Studies were organized into four evidence categories: (A) periodontal disease and female infertility/fecundability (n = 8 studies); (B) Periodontal Disease IVF outcomes and recurrent implantation failure (n = 1); (C) periodontal disease and endometriosis (n = 3); and (D) mechanistic in vitro and animal studies (n = 10). Confounders reported across epidemiological studies included age, BMI, smoking status, socioeconomic factors, and systemic conditions, though their handling was heterogeneous across studies and is detailed below. Periodontal disease and female infertility/fecundability Eight studies evaluated the link between periodontal disease or oral dysbiosis and female infertility/fecundability. Despite reported positive associations, the evidence remains inconclusive due to significant heterogeneity in study design and exposure assessment. Two prospective cohort studies were retrieved. Bond et al., (2021) [25], evaluated 2,764 female pregnancy planners and reported a fecundability ratio (FR) of 0.79 (95% CI: 0.67–0.94) for women reporting a history of periodontitis treatment versus none, suggesting an approximately 21% reduction in per-cycle conception probability. A possible dose-response was observed across severity proxies (diagnosis: FR = 0.89; treatment: FR = 0.79; tooth mobility: FR = 0.71), though wide confidence intervals (particularly for tooth mobility) limit precision. Key confounders, including race/ethnicity, BMI, smoking, diabetes, endometriosis, polycystic ovarian syndrome, bacterial vaginosis and vitamin supplementation; were controlled through directed acyclic graph (DAG)-guided adjustment. Huo et al. [26], reported an FR of 0.83 (95% CI: 0.69–1.00) for halitosis combined with gingival bleeding as parameters for oral dysbiosis in 3,915 Chinese women attempting conception, after adjustment for age, BMI, and lifestyle factors. In both studies, the use of indirect oral health parameters limits pathological specificity. Among case-control studies, Kamei-Nagata et al. [30] suggests a potential link between serological evidence of periodontal pathogen exposure and infertility. In 146 women, serum IgG titers against P. gingivalis W83 (OR = 3.31; 95% CI: 1.30–11.17; p = 0.029) and P. intermedia ATCC25611 (OR = 4.50; 95% CI: 1.55–15.08; p = 0.009) were independently associated with unexplained infertility in multivariate logistic regression. P. gingivalis FDC381 showed an inverse association (OR = 0.55; 95% CI: 0.31–0.86; p = 0.020), indicating strain-specific effects. Importantly, no clinical periodontal examination was performed and thus, findings reflect antibody-mediated exposure rather than active disease status. Telatar et al. [29], found significantly higher periodontal inflamed surface area (PISA) in 50 women with unexplained infertility versus 50 controls (p = 0.002). However, the adjusted OR was 0.24 (95% CI: 0.10–0.58), suggesting that after controlling for confounders, infertility was unexpectedly associated with lower odds of severe periodontitis. The authors attribute this direction reversal to residual confounding by oral hygiene behaviors and bleeding on probing, a finding that underscores the methodological complexities of observational studies in this field and warrants a cautious interpretation. Machado et al. [28], conducted a pilot case–control study comparing the periodontal status of women seeking fertility treatment against age-matched fertile controls. Although periodontitis prevalence was higher in the clinical group (33.3% vs. 22.2%), the resulting OR of 1.75 (95% CI: 0.40–7.71) was not statistically significant. This lack of significance, marked by a wide confidence interval, is likely attributable to the limited sample size (n = 36). Sumilin et al. [27] documented elevated salivary active-matrix metalloproteinase-8 in infertile women without clinically detectable periodontitis (OR = 1.03; 95% CI: 0.99–1.07), suggesting subclinical oral inflammatory activity may be relevant. A study by Hart et al. [31] expands the scope of this domain (sub-analysis of the SMILE randomized controlled trial, n = 1,956) reported OR = 2.88 (95% CI: 1.62–5.12) for periodontitis and time to conception > 12 months; the effect was significant only in non-White women, highlighting potential effect modification by ancestry or socioeconomic context. In the analysis of spontaneous abortion (SAB) data, Bond et al. [32] (2023) extended the PRESTO cohort to evaluate preconception periodontitis and SAB risk in 3,444 women (826 SABs). Hazard ratios for periodontitis diagnosis (HR = 0.97; 95% CI: 0.76–1.23) and treatment (HR = 1.01; 95% CI: 0.79–1.27) were null, while tooth mobility showed a positive association (HR = 1.38; 95% CI: 0.88–2.14). Quantitative bias analysis indicated attenuation toward the null due to self-report misclassification, with the bias-adjusted estimate for tooth mobility reaching HR = 2.00 (95% simulation interval: 0.99–6.41). Residual confounding and the self-reported exposure measurement remain key limitations. Periodontal disease and IVF and RIF outcomes None of the included studies directly investigated the oral microbiome composition in relation to IVF or RIF. However, Zhang et al., [33] investigated the relationship between endometrial microbiota, chronic endometritis (CE), and reproductive outcomes in women with RIF. Using canonical correlation analysis, the study demonstrated that CE was significantly associated with adverse reproductive outcomes and distinct microbial signatures. Specifically, Phyllobacterium, Gardnerella, Enterococcus, and Pseudomonas were correlated with miscarriage, while Prevotella, Streptococcus, and Romboutsia were linked to non-pregnancy. In contrast, genera such as Achromobacter, Proteobacteria, Lactobacillus, and Acinetobacter were positively associated with clinical pregnancy. Periodontal disease and endometriosis A 2024 study by Jin et al., [35] applied a bidirectional two-sample MR using GWAS summary data. The analysis supported a causal effect of periodontitis on pelvic peritoneal endometriosis (OR = 1.079; 95% CI: 1.016–1.146; p = 0.014), Of note, the reverse analysis did not support a causal effect of endometriosis on periodontitis. While MR reduces confounding by measured and unmeasured environmental factors, the modest effect sizes and the possibility of genetic pleiotropy warrant caution. The MR finding is best interpreted as providing suggestive causal evidence consistent with shared inflammatory pathways rather than establishing a definitive causal direction. A 2025 prospective cohort study by Hicks et al. [47] identified enrichment of Fusobacterium in oral samples from women with moderate-to-severe endometriosis compared to minimal-to-mild disease (n = 64). However, no clinical periodontal examination was performed; and oral Fusobacterium enrichment does not necessarily establish periodontitis as the exposure. Results should be considered hypothesis-generating. Mechanistic evidence: In vitro and animal studies Ten mechanistic studies were included, comprising seven in vitro studies using human trophoblast cell lines (HTR-8/SVneo), one in vitro study using primary human decidual cells, one rodent model, and one in vitro macrophage-trophoblast co-culture. Although these studies provide biological plausibility for the epidemiological associations described above cannot be directly extrapolated to clinical human outcomes without independent clinical validation. Effects on trophoblast function (in vitro) Hirohata et al. [37], showed that P. gingivalis culture supernatant dose-dependently suppressed trophoblast invasion by 85% (0.01% v/v) and 47% (0.1% v/v) without direct cytotoxicity, implicating secreted soluble factors rather than the bacterium itself. Riewe et al. [39] showed that P. gingivalis infection of HTR-8 cells differentially regulated over 2,000 genes (p < 0.001), with statistically significant upregulation of the MEK3-p38-Max MAPK pathway (confirmed by immunoblot and phosphorylation assays) and induction of IL-1β and IL-8 at 2 h and 24 h post-infection. Results by Ren et al. [40], showed that ERK1/2 and p38 MAPK pathways mediate P. gingivalis-induced inflammation and apoptosis in trophoblasts; while pharmacological inhibition of ERK1/2 partially reversed both inflammatory cytokine release (IL-8, IFN-γ) and apoptotic cell death. Collectively, these three studies, originating from independent research groups in Japan, the USA, and China; provide a body of evidence regarding the impact of P. gingivalis on trophoblast function. Their findings collectively indicate a plausible involvement of MAPK-dependent pathways in the disruption of both invasion and inflammatory homeostasis. Komine-Aizawa et al. [38] observed that P. gingivalis LPS alone does not impair trophoblast invasion at tested concentrations; co-administration with nicotine resulted in significant inhibition (p < 0.01) via p38 MAPK, alongside the suppression of both MMP-2 and MMP-9. These findings suggest that smoking may act as a synergistic risk modifier within the proposed periodontal-trophoblast axis. This interaction underscores the importance of smoking stratification in clinical studies, as failing to account for this synergy could lead to an inaccurate estimation of the isolated periodontal contribution. Effects on decidual inflammation (in vitro, primary cells) Keelan et al. [41] evaluated the inflammatory response of primary human decidual cells to four periodontal pathogens and their LPS preparations. All bacteria elicited measurable TNF-α and IL-6 production, with potency ranked E. coli > A. actinomycetemcomitans > F. nucleatum > P. gingivalis. Notably, P. gingivalis LPS was the weakest stimulator of decidual inflammation, consistent with its atypical lipid A structure conferring TLR4 antagonism properties rather than canonical TLR4 agonism. Of 112 inflammation-associated genes profiled, 18 were significantly upregulated by at least one LPS preparation, including CXCR1 (IL-8 receptor), TNF-α, RANTES (CCL5), and IL-1β. These results propose that periodontal pathogen-derived products are capable of triggering decidual inflammation at concentrations plausibly achievable during bacteremia, while also establishing that the magnitude of response is bacterium and concentration-dependent. Emerging experimental evidence suggests that OMVs from P. gingivalis and F. nucleatum can translocate into the systemic circulation, and reach the endometrial compartment [21,22,76]. These vesicles could represent an additional messenger bridge between periodontal disease and reproductive dysfunction, complementing the direct bacterial and cytokine-mediated pathways described above. However, no study retrieved by the systematic search conducted for this review directly characterized periodontal pathogen derived EVs or OMV exposure as a primary exposure variable in relation to female reproductive outcomes. This represents a significant gap in the current evidence base and a priority area for future research. Lara et al. [44] observed that P. gingivalis OMVs are internalized by human trophoblast cells, altering their metabolism, reducing glycolytic activity, downregulating glucose transporter expression (GLUT1 and GLUT3), and suppressing lactate production, driving trophoblasts into a metabolically quiescent state. These findings are corroborated by in vivo experiments by Lee et al. [45] showing that mice exposed to OMVs exhibit reduced placental and fetal weights, along with altered placental architecture. These findings resonate with the early mechanistic work of Bartruff et al. [46] which proposed that OMVs exert inhibitory effects on the growth and viability of host cells, suggesting a conserved strategy of cellular disruption by these pathogens. The proposed hypothesis linking these mechanisms is illustrated in Fig. 2. Fig. 2Proposed Periodontal–Reproductive Axis: Potential systemic translocation of bacterial-derived EVs and disruption of the hypothalamic–pituitary–ovarian (HPO) axis. This figure illustrates the bidirectional relationship between periodontal health and the female reproductive system. Sex hormones modulate the oral microbiome and periodontal tissue response, particularly during high-hormone states such as pregnancy or controlled ovarian hyperstimulation in IVF. Conversely, bacterial-derived EVs (OMVs), major virulence factors and pro-inflammatory mediators (e.g., IL-1β, IL-6, TNF-α) released from inflamed tissues translocate into the systemic circulation. These systemic factors can disrupt the HPO axis, impairing folliculogenesis and ovulation, while also reaching the endometrial compartment to downregulate essential adhesion molecules and interfere with the decidualization process. The resulting chronic low-grade inflammation establishes a hostile environment that compromises both oocyte maturation and embryo implantation. Abbreviations: HPO: Hypothalamic–Pituitary–Ovarian; EVs: Extracellular Vesicles; IVF: In Vitro Fertilization; IL: Interleukin; TNF-α: Tumor Necrosis Factor-alpha. Created with Biorender© Proposed Periodontal–Reproductive Axis: Potential systemic translocation of bacterial-derived EVs and disruption of the hypothalamic–pituitary–ovarian (HPO) axis. This figure illustrates the bidirectional relationship between periodontal health and the female reproductive system. Sex hormones modulate the oral microbiome and periodontal tissue response, particularly during high-hormone states such as pregnancy or controlled ovarian hyperstimulation in IVF. Conversely, bacterial-derived EVs (OMVs), major virulence factors and pro-inflammatory mediators (e.g., IL-1β, IL-6, TNF-α) released from inflamed tissues translocate into the systemic circulation. These systemic factors can disrupt the HPO axis, impairing folliculogenesis and ovulation, while also reaching the endometrial compartment to downregulate essential adhesion molecules and interfere with the decidualization process. The resulting chronic low-grade inflammation establishes a hostile environment that compromises both oocyte maturation and embryo implantation. Abbreviations: HPO: Hypothalamic–Pituitary–Ovarian; EVs: Extracellular Vesicles; IVF: In Vitro Fertilization; IL: Interleukin; TNF-α: Tumor Necrosis Factor-alpha. Created with Biorender© Spiral artery remodeling (animal model) Tavarna et al. [42] is the only animal study in this review and provides the most direct in vivo evidence linking periodontal infection to early implantation-related processes. In a rat model of periodontitis, P. gingivalis-infected dams at gestation day 11 exhibited significantly reduced spiral arterial lumen size (p < 0.02), increased vascular smooth muscle cell retention (p < 0.003), and markedly reduced endovascular trophoblast invasion (29% vs. 68% in controls; p ≤ 0.002). Uterine natural killer (uNK) cells were elevated (14.9% vs. 3.9%; p ≤ 0.022) and IL-18 + uNK cells were reduced (p < 0.001), while Htra1 (a marker of oxidative stress and trophoblast migration inhibitor) was overexpressed. While these molecular alterations largely normalized by gestation day 18, the study points towards a potential window of vulnerability during early placentation, though cross-species extrapolation to human outcomes warrants caution and independent clinical validation. Macrophage-mediated trophoblast effects Einenkel et al. [43] explored the bidirectional effect of low-abundance F. nucleatum on macrophage-trophoblast interactions, utilizing trophoblast-macrophage co-cultures. Low bacterial concentration (ratio 0.1) in tolerogenic M2c macrophages significantly increased HIF-1α expression and VEGF-A secretion (p < 0.05), supporting trophoblast tube formation. Results indicate a threshold-dependent effect of the uterine microbiome on trophoblast activity. High bacterial density and M1-polarized inflammatory environments were found to impair migration and function. In contrast, low-level bacterial presence was not associated with such disruptions, suggesting that dysbiotic overgrowth, rather than mere bacterial presence, is the primary factor affecting implantation-critical processes. Discussion This scoping review maps evidence from 19 studies on the possible relationship between periodontal disease and female reproductive disorders. Although a biologically plausible association could be proposed (mediated through systemic inflammation, immune dysregulation and microbial translocation as shown in Table 2), the evidence base is heterogeneous, largely observational, and insufficient to establish causation. Table 2Molecular mediators and proposed pathophysiological mechanisms linking periodontal disease to endometrial dysfunction and reproductive failureMolecule/ComponentPrimary SourceMechanism of Action/Cellular TargetProposed Effects on DecidualizationProposed Effects on ImplantationProposed Effects on EndometriosisPro-inflammatory Cytokines (IL−1β, IL − 6, TNF − α)Perio-EVs; Systemic circulationDownregulation of adhesion molecules; disruption of the HPO axis.Interference with stromal differentiation; reduced expression of PRL and IGFBP1.Downregulation of Integrinαvβ3 and L−selectin; impaired receptivity window.Exacerbation of the systemic pro-inflammatory environment and disease phenotype.LPS and GingipainsPg OMVsActivation of TLR4/NF − κB pathways.Polarization of decidual macrophages toward a pro-inflammatory M1 phenotype.Suppression of LIF and IL − 10; promotion of a hostile implantation site.Triggering of local inflammatory cascades within the pelvic cavity.Mitochondrial ComponentsM1 macrophage-derived EVsDisruption of bioenergetic reprogramming.Impaired mitochondrial dynamics and stromal cell differentiation.Reduced embryo acceptance due to secondary disruption of LIF expression.—Metabolic Inhibitors (via GLUT1/3)Pg OMVsInhibition of glucose transporters and glycolytic activity.—Induction of metabolic quiescence in trophoblasts; inhibited migration and invasion.—Chemokines (IL − 8)Pg OMVsUpregulation of pro-angiogenic and invasive signaling.——Potent induction of inflammatory response and increased invasive potential in lesions.Growth Factors (TGF − β) Fn Induction of phenotypic transitions.——Fibroblast-to-myofibroblast transition; facilitated lesion establishment and migration.This table summarizes the primary bioactive molecules and bacterial components originating from the oral cavity that interfere with reproductive homeostasis. The proposed mechanisms range from the disruption of the hypothalamic-pituitary-ovarian (HPO) axis to specific cellular transformations in the decidua and ectopic endometrial lesions, including the downregulation of essential adhesion molecules, metabolic reprogramming of trophoblast cells, and the induction of a pro-inflammatory M1 macrophage phenotype within the decidua. Abbreviations: EnSC Endometrial Stromal Cells, EVs Extracellular Vesicles, HPO Hypothalamic–Pituitary–Ovarian; Pg Porphyromonas gingivalis; OMVs Outer Membrane Vesicles, LPS Lipopolysaccharide, RIF Recurrent Implantation Failure, IL Interleukin, TNF-α Tumor Necrosis Factor-α, LIF Leukemia Inhibitory Factor, PRL Prolactin, IGFBP1 Insulin-like Growth Factor-Binding Protein 1, TGF-β Transforming Growth Factor-β, Fn Fusobacterium nucleatum Table 2Molecular mediators and proposed pathophysiological mechanisms linking periodontal disease to endometrial dysfunction and reproductive failureMolecule/ComponentPrimary SourceMechanism of Action/Cellular TargetProposed Effects on DecidualizationProposed Effects on ImplantationProposed Effects on EndometriosisPro-inflammatory Cytokines (IL−1β, IL − 6, TNF − α)Perio-EVs; Systemic circulationDownregulation of adhesion molecules; disruption of the HPO axis.Interference with stromal differentiation; reduced expression of PRL and IGFBP1.Downregulation of Integrinαvβ3 and L−selectin; impaired receptivity window.Exacerbation of the systemic pro-inflammatory environment and disease phenotype.LPS and GingipainsPg OMVsActivation of TLR4/NF − κB pathways.Polarization of decidual macrophages toward a pro-inflammatory M1 phenotype.Suppression of LIF and IL − 10; promotion of a hostile implantation site.Triggering of local inflammatory cascades within the pelvic cavity.Mitochondrial ComponentsM1 macrophage-derived EVsDisruption of bioenergetic reprogramming.Impaired mitochondrial dynamics and stromal cell differentiation.Reduced embryo acceptance due to secondary disruption of LIF expression.—Metabolic Inhibitors (via GLUT1/3)Pg OMVsInhibition of glucose transporters and glycolytic activity.—Induction of metabolic quiescence in trophoblasts; inhibited migration and invasion.—Chemokines (IL − 8)Pg OMVsUpregulation of pro-angiogenic and invasive signaling.——Potent induction of inflammatory response and increased invasive potential in lesions.Growth Factors (TGF − β) Fn Induction of phenotypic transitions.——Fibroblast-to-myofibroblast transition; facilitated lesion establishment and migration.This table summarizes the primary bioactive molecules and bacterial components originating from the oral cavity that interfere with reproductive homeostasis. The proposed mechanisms range from the disruption of the hypothalamic-pituitary-ovarian (HPO) axis to specific cellular transformations in the decidua and ectopic endometrial lesions, including the downregulation of essential adhesion molecules, metabolic reprogramming of trophoblast cells, and the induction of a pro-inflammatory M1 macrophage phenotype within the decidua. Abbreviations: EnSC Endometrial Stromal Cells, EVs Extracellular Vesicles, HPO Hypothalamic–Pituitary–Ovarian; Pg Porphyromonas gingivalis; OMVs Outer Membrane Vesicles, LPS Lipopolysaccharide, RIF Recurrent Implantation Failure, IL Interleukin, TNF-α Tumor Necrosis Factor-α, LIF Leukemia Inhibitory Factor, PRL Prolactin, IGFBP1 Insulin-like Growth Factor-Binding Protein 1, TGF-β Transforming Growth Factor-β, Fn Fusobacterium nucleatum Molecular mediators and proposed pathophysiological mechanisms linking periodontal disease to endometrial dysfunction and reproductive failure This table summarizes the primary bioactive molecules and bacterial components originating from the oral cavity that interfere with reproductive homeostasis. The proposed mechanisms range from the disruption of the hypothalamic-pituitary-ovarian (HPO) axis to specific cellular transformations in the decidua and ectopic endometrial lesions, including the downregulation of essential adhesion molecules, metabolic reprogramming of trophoblast cells, and the induction of a pro-inflammatory M1 macrophage phenotype within the decidua. Abbreviations: EnSC Endometrial Stromal Cells, EVs Extracellular Vesicles, HPO Hypothalamic–Pituitary–Ovarian; Pg Porphyromonas gingivalis; OMVs Outer Membrane Vesicles, LPS Lipopolysaccharide, RIF Recurrent Implantation Failure, IL Interleukin, TNF-α Tumor Necrosis Factor-α, LIF Leukemia Inhibitory Factor, PRL Prolactin, IGFBP1 Insulin-like Growth Factor-Binding Protein 1, TGF-β Transforming Growth Factor-β, Fn Fusobacterium nucleatum Sex hormones, periodontal pathogens, and systemic inflammation Sex steroid hormones modulate immune, metabolic, and microbial dynamics throughout the female reproductive lifespan [48]. In the oral cavity, steroid hormone receptors are expressed in the periodontium, oral mucosa, and salivary glands [49], and estrogen and progesterone directly alter gingival vascular permeability, subgingival biofilm composition, and the local immune response [48, 50]. It is therefore plausible that disruptions to this endocrine balance may not only contribute to systemic inflammatory alterations but also exacerbate local inflammatory processes within the oral cavity [51]. This is particularly evident in patients undergoing ART, where controlled ovarian hyperstimulation (COH) may exacerbate periodontal inflammation through increased estrogen and progesterone levels, upregulating progesterone-receptor expression in gingival tissues, providing a nutrient-rich environment that facilitates the proliferation of anaerobic pathobionts such as P. gingivalis and P.intermedia [52]. This is contextually relevant to the included study population, as several studies recruited women undergoing fertility treatment. Conversely, systemic inflammatory mediators of periodontal origin (including IL-6, IL-8, TNF-α, and C-reactive protein) can interfere with hypothalamic-pituitary-ovarian (HPO) axis signaling and alter granulosa cell responsiveness to gonadotropins [53]. It is important to note that this mechanistic pathway is grounded in background biological knowledge and plausible inference from the included studies, not in direct evidence from this review’s included studies, none of which measured HPO axis hormones or ovarian reserve as outcomes. The hormonal-periodontal relationship is therefore best understood as a bidirectional, plausibility-level framework requiring dedicated prospective investigation, and a proposed relation is further illustrated in Fig. 3. Fig. 3Proposed mechanism linking periodontal inflammation and impaired decidualization Proposed mechanism linking periodontal inflammation and impaired decidualization Evidence from the included study by Kamei-Nagata et al. [30] provides serological support for this axis: elevated IgG titers against P. gingivalis W83 (OR = 3.31; 95% CI: 1.30–11.17) and P. intermedia ATCC25611 (OR = 4.50; 95% CI: 1.55–15.08) were independently associated with unexplained infertility in multivariate logistic regression. However, this study lacked a clinical periodontal examination and measured antibody-mediated pathogen exposure rather than active disease. Furthermore, strain-level specificity, including an inverse association with P. gingivalis FDC381 (OR = 0.55), precludes generalization across all periodontal pathogens and warrants cautious interpretation. Mechanistic pathways: periodontal pathogens and embryo implantation Convergent in vitro evidence from three independent research groups proposes that P. gingivalis suppresses trophoblast invasion via soluble secreted factors [37], activates MEK3-p38 and ERK1/2 MAPK signaling with global transcriptomic consequences across > 2,000 genes [39], and induces IL-8, IFN-γ, and caspase-3-mediated apoptosis in trophoblast cell lines [40]. Komine-Aizawa et al. [38], demonstrated that the combination of P. gingivalis LPS and nicotine significantly inhibits trophoblast invasion via p38 MAPK, establishing smoking as a possible synergistic risk modifier with direct methodological implications: effect estimates from studies that do not stratify by smoking status may conflate the isolated periodontal contribution with a smoking-periodontal interaction effect. Building on these observations, Keelan et al. [41] showed that all four periodontal pathogens tested elicit measurable inflammatory responses in primary human decidual cells. The weaker TLR-4 agonism of P. gingivalis LPS (attributable to its atypical lipid A structure) challenges mechanistic models that treat it as a potent direct TLR-4 activator. Tavarna et al. [42] provides the only animal-model evidence in this review, showing disrupted spiral artery remodeling, reduced endovascular trophoblast invasion, and altered uterine NK cell populations in P. gingivalis-infected rat dams at gestation day 11. Einenkel et al. [43] adds dose-dependency nuance: low F. nucleatum loads in tolerogenic M2c macrophages supported trophoblast tube formation via HIF-1α/VEGF-A upregulation, while high loads or pro-inflammatory M1 polarization impaired it, suggesting a threshold-dependent relationship between dysbiotic bacterial burden and implantation success. The periodontal–endometriosis axis: microbial dissemination and tissue alterations Endometriosis is a multifactorial disease with established genetic, hormonal, and immune determinants; any periodontal contribution, if causal, is likely one among multiple interacting factors rather than a primary driver. The three included studies in this domain are insufficient to determine whether a causal relationship exists, what its direction is, or what its magnitude might be in the presence of adequate confounder control [54]. Evidence for a periodontal-endometriosis relationship across the three included studies (Kavoussi et al., [34]; Jin et al. [35], 2024; Hicks et al. [47]., 2025) is suggestive but design-limited. The NHANES cross-sectional analysis by Kavoussi et al. [34] provides the largest population estimate (adjusted OR = 1.57; 95% CI: 1.06–2.33, n = 4,136) but cannot establish temporal precedence. The Mendelian randomization by Jin et al. [35] proposes bidirectionality offered by shared inflammatory pathways, with modest effect sizes. The oral Fusobacterium enrichment in moderate-to-severe endometriosis reported by Hicks et al. [47] is microbiologically intriguing but was measured without clinical periodontal examination. Furthermore, oral dysbiosis is not equivalent to periodontitis as a clinical diagnosis. Supporting these findings, it has been previously proposed that F. nucleatum may promote endometriotic lesion establishment through TGF-β-mediated myofibroblast transition in a murine model [36]. Likewise, IL-8, a chemokine implicated in inflammation and angiogenesis, is known to be upregulated in endometriotic lesions and associated with disease progression [55]. Notably, P. gingivalis OMVs were shown to upregulate the expression of IL-8 in human umbilical vein endothelial cells and were more potent in inducing an inflammatory response than the bacterial cells themselves [56]. A proposed pathophysiological framework integrating these mechanistic pathways is presented in Fig. 4. Fig. 4Proposed Pathophysiological Framework for the Oral-Endometriosis biological plausibility. OMVs. PAMPS and DAMPs, derived from dysbiotic subgingival biofilms and enriched in virulence factors from Porphyromonas gingivalis and other key periodontal pathogens, can disseminate through the bloodstream and reach the uterine microenvironment. Under physiological conditions, endometrial stromal cells undergo decidualization in response to progesterone signaling within a tightly regulated inflammatory milieu dominated by M2 macrophages, an essential process for successful embryo implantation. In contrast, exposure to bacterial derived EVs promotes internalization by endometrial mesenchymal stem cells and macrophages, triggering a persistent pro-inflammatory response, predominant M1 macrophage polarization, and disrupted decidualization, leading to implantation to impaired implantation capacity or ultimately to implantation failure. Abbreviations: OMVs: Outer Membrane Vesicles, PAMPs: pathogen-Associated Molecular Patterns, TGF-β: Transforming Growth Factor-beta, IL-8: Interleukin-8, Pg: Porphyromonas gingivalis, Created with Biorender©. This diagram illustrates the hematogenous dissemination of oral pathogens –specifically Porphyromonas gingivalis and Fusobacterium nucleatum– and their outer membrane vesicles (OMVs) from the oral cavity to the pelvic environment. A Fusobacterium nucleatum infects fibroblast cells of endometriosis lesions, inducing a TGF beta-mediated phenotypic transition to myofibroblasts expressing TAGLN gained the ability to proliferate, adhere, and migrate, potentially promoting endometrial cell survival at ectopic sites. B Porphyromonas gingivalis-derived OMVs upregulate the expression of chemokines in endothelial and stromal cells, promoting a localized hyper-inflammatory state and angiogenesis that drives disease progression. Collectively, these processes bridge the gap between periodontal infection and gynecological disease through shared microbial signatures and systemic inflammatory pathways, contributing to the complex inflammatory phenotype characteristic of endometriosis. Abbreviations: OMVs: Outer Membrane Vesicles, TGF-β: Transforming Growth Factor-beta, IL-8: Interleukin-8, Pg: Porphyromonas gingivalis, Fn: Fusobacterium nucleatum. Created with Biorender© Proposed Pathophysiological Framework for the Oral-Endometriosis biological plausibility. OMVs. PAMPS and DAMPs, derived from dysbiotic subgingival biofilms and enriched in virulence factors from Porphyromonas gingivalis and other key periodontal pathogens, can disseminate through the bloodstream and reach the uterine microenvironment. Under physiological conditions, endometrial stromal cells undergo decidualization in response to progesterone signaling within a tightly regulated inflammatory milieu dominated by M2 macrophages, an essential process for successful embryo implantation. In contrast, exposure to bacterial derived EVs promotes internalization by endometrial mesenchymal stem cells and macrophages, triggering a persistent pro-inflammatory response, predominant M1 macrophage polarization, and disrupted decidualization, leading to implantation to impaired implantation capacity or ultimately to implantation failure. Abbreviations: OMVs: Outer Membrane Vesicles, PAMPs: pathogen-Associated Molecular Patterns, TGF-β: Transforming Growth Factor-beta, IL-8: Interleukin-8, Pg: Porphyromonas gingivalis, Created with Biorender©. This diagram illustrates the hematogenous dissemination of oral pathogens –specifically Porphyromonas gingivalis and Fusobacterium nucleatum– and their outer membrane vesicles (OMVs) from the oral cavity to the pelvic environment. A Fusobacterium nucleatum infects fibroblast cells of endometriosis lesions, inducing a TGF beta-mediated phenotypic transition to myofibroblasts expressing TAGLN gained the ability to proliferate, adhere, and migrate, potentially promoting endometrial cell survival at ectopic sites. B Porphyromonas gingivalis-derived OMVs upregulate the expression of chemokines in endothelial and stromal cells, promoting a localized hyper-inflammatory state and angiogenesis that drives disease progression. Collectively, these processes bridge the gap between periodontal infection and gynecological disease through shared microbial signatures and systemic inflammatory pathways, contributing to the complex inflammatory phenotype characteristic of endometriosis. Abbreviations: OMVs: Outer Membrane Vesicles, TGF-β: Transforming Growth Factor-beta, IL-8: Interleukin-8, Pg: Porphyromonas gingivalis, Fn: Fusobacterium nucleatum. Created with Biorender© Limitations Several limitations constrain the interpretability of these findings. As a scoping review, a formal risk-of-bias assessment was not performed, and the evidence base spans study designs with substantially different susceptibility to bias. Exposure definition was inconsistent: periodontal disease was operationalized variously as: self-reported diagnosis, clinical indices (CAL, PD, BOP, PISA), salivary biomarkers (aMMP-8), serum IgG titers against specific pathogen strains, or oral microbiome composition; variables that are not directly comparable and impede pooling of effect estimates. Residual confounding by shared inflammatory risk factors cannot be excluded and may account for a portion of the observed associations. Several studies were also underpowered, with confidence intervals wide enough to include null or inverse effects. Direct measurement of periodontal pathogen-derived vesicles as an exposure variable was also absent from the included studies. Additionally, a disproportionate reliance on studies from the same research group represents a further limitation. Two of the included infertility/fecundability studies [25, 32] derive from the PRESTO cohort (Boston University), and while both are methodologically rigorous, their findings reflect a predominantly White, high-income, North American pregnancy planning population. This concentration of evidence from a single cohort limits independence of findings and constrains generalizability; independent replication across sociodemographic and geographical diverse populations remains a research priority. Conclusions This scoping review maps evidence from 22 studies and identifies a possible, though not yet established, biological relationship between periodontal disease and female reproductive disorders. The available epidemiological evidence is suggestive but methodologically limited. Prospective data shows only modest effects on fecundability, while mechanistic evidence from in vitro and animal models remains preliminary. Causal conclusions are therefore premature, and future research should prioritize prospective studies with standardized periodontal assessments, with a multidisciplinary approach, to determine whether this relation is clinically meaningful.