← Research Library
BODYINTERPRETATIONResearch Paper

Biofilm-Mediated Endothelial Coherence Failure: A Novel Mechanistic Framework for Intracranial Aneurysm Formation and Rupture

Pearl (AI Research Engine) · Eric Whitney DO·March 19, 2026·7,591 words

Biofilm-Mediated Endothelial Coherence Failure: A Novel Mechanistic Framework for Intracranial Aneurysm Formation and Rupture

Pearl — The Encoded Human Research Engine Generated: 2026-03-19

<!-- SECTION:0 -->

Abstract

Background. Intracranial aneurysm (IA) formation and rupture are conventionally attributed to hemodynamic stress—aberrant wall shear stress at arterial bifurcations driving degenerative remodeling of the vessel wall. While hemodynamic factors are necessary contributors, they are insufficient to explain the selective vulnerability of specific patients, the progressive inflammatory infiltration observed in aneurysm walls, or the emerging epidemiological association between periodontal disease and aneurysmal subarachnoid hemorrhage (aSAH). Severe periodontitis independently increases the risk of aSAH (HR 22.5, 95% CI 3.6–139.5), and oral bacterial DNA has been recovered from intracranial aneurysm tissue [1].

Hypothesis. We propose a novel mechanistic framework—biofilm-mediated endothelial coherence failure—in which periodontal pathogens, principally Porphyromonas gingivalis, serve as upstream initiators of a multistage pathogenic cascade culminating in aneurysm formation and rupture. The framework posits five interdependent stages: (1) chronic periodontal biofilm generates sustained bacteremia through an ulcerated mucosal surface estimated at 8–20 cm² in moderate-to-severe disease; (2) P. gingivalis and its gingipain proteases reach the intracranial vasculature, invade endothelial cells, and degrade intercellular junctional proteins and complement components; (3) this produces endothelial coherence failure—the progressive loss of coordinated cell-cell signaling, barrier integrity, and mechanotransductive competence in the vessel wall; (4) gingipain-mediated hemoglobin hydrolysis and heme acquisition generate local free iron accumulation, which, combined with inflammatory lipid peroxidation, primes the endothelium for ferroptosis—an iron-dependent, non-apoptotic form of regulated cell death; and (5) ferroptotic endothelial loss exposes the subendothelial matrix to hemodynamic forces, triggering macrophage infiltration, matrix metalloproteinase activation, and progressive wall degeneration toward rupture.

Implications. This framework reframes the hemodynamic model not as incorrect but as incomplete, positioning biofilm-driven infection as the upstream variable that determines which hemodynamically stressed vessels progress to aneurysm formation. If supported, this model implies that aggressive periodontal treatment may constitute a modifiable intervention for IA risk reduction—a hypothesis amenable to prospective clinical testing.

Epistemic status: HYPOTHESIS. No element of this framework has been confirmed by direct experimental demonstration of the complete causal chain. Each individual link is supported by peer-reviewed evidence; their integration into a unified cascade remains speculative.

<!-- SECTION:1 -->

1. Introduction: The Unresolved Etiology of Intracranial Aneurysm Formation

Intracranial aneurysms (IAs) — focal outpouchings of cerebral arterial walls in which all mural layers undergo degenerative change — represent one of the most common and consequential unsolved problems in vascular neuroscience. Population-based imaging studies estimate their prevalence at 2–3% of adults, with some series reporting figures as high as 3.2% when sensitive angiographic techniques are employed [Vlak et al., 2011; [2]; [3]]. Most IAs remain clinically silent. When rupture occurs, however, the resulting aneurysmal subarachnoid hemorrhage (aSAH) constitutes a neurological catastrophe: a form of hemorrhagic stroke associated with case-fatality rates that, despite improvements in neurocritical care over recent decades, continue to exceed 30% [38], with the majority of survivors sustaining lasting neurocognitive disability [[4]; [5]]. The disproportion between the commonality of the lesion and the devastation of its clinical endpoint makes the question of why aneurysms form — and why some progress to rupture while most do not — among the most pressing in cerebrovascular medicine.

The established risk factor profile for IA formation is well characterized and strikingly non-specific. Hypertension, cigarette smoking, female sex, advancing age, and family history of IA or aSAH have been consistently identified across large cohort studies [3]. These factors delineate populations at elevated risk but do not specify a molecular mechanism at the vessel wall. The dominant mechanistic paradigm centers on hemodynamic stress: regions of low or oscillatory wall shear stress (WSS), particularly at arterial bifurcations in the circle of Willis, are proposed to initiate endothelial dysfunction, triggering inflammatory cascades that result in degradation of the internal elastic lamina and tunica media [6]. This framework is supported by computational fluid dynamics modeling and by the anatomic predilection of saccular aneurysms for bifurcation apices. Yet it remains fundamentally incomplete. Hemodynamic stress is ubiquitous at arterial branch points in every adult; aneurysm formation is not. The hemodynamic model identifies where aneurysms are likely to form but cannot, by itself, explain why they form in a given individual at a given time.

Histopathological examination of resected aneurysm walls has repeatedly demonstrated a characteristic pattern of destruction: loss of the internal elastic lamina, disruption or absence of the tunica media, inflammatory cell infiltration (predominantly macrophages and T-lymphocytes), and — in walls approaching rupture — regions of near-complete mural thinning with loss of endothelial coverage [7]. These findings are descriptively consistent with the hemodynamic-inflammatory model but leave open the question of what initiates the destructive cascade. Inflammation is present, but what activates it? Endothelial dysfunction is observed, but what disrupts endothelial integrity in the first place? The field has, in effect, described the downstream consequences of aneurysm formation in considerable detail while leaving the upstream initiating event largely unspecified.

Recent epidemiological and microbiological evidence has introduced a variable that the hemodynamic paradigm does not account for: the association between periodontitis and intracranial aneurysm disease. In a case-control and prospective cohort study of 5,170 participants from the Finnish Health 2000 Survey, Pyysalo et al. (2019) reported that periodontitis (defined as ≥4 mm gingival pocket depth) was found in 92% of IA patients and was independently associated with IA presence (OR 5.3, 95% CI 1.1–25.9, p < 0.001) [1]. More strikingly, severe periodontitis at baseline predicted the subsequent development of aSAH during a 13-year follow-up with a hazard ratio of 22.5 (95% CI 3.6–139.5, p = 0.001), an effect independent of sex, smoking status, and hypertension [1]. These are not marginal associations. A hazard ratio of 22.5 for a modifiable oral condition predicting a lethal cerebrovascular event demands mechanistic explanation. Critically, the same research group and others have reported the detection of oral bacterial DNA — including that of Porphyromonas gingivalis, the keystone pathogen of chronic periodontitis — within resected intracranial aneurysm walls [1]. P. gingivalis is not a passive bystander; it is among the most sophisticated immune-evasive organisms in human-associated microbiology, equipped with cysteine proteinases (gingipains) that degrade complement components, cleave hemoglobin to acquire heme iron, subvert macrophage phagocytosis, and disrupt epithelial and endothelial barrier integrity [[8]; [9]; [10]].

These convergent findings — epidemiological association, microbial DNA in aneurysm tissue, and a pathogen with the precise virulence toolkit required to dismantle arterial wall integrity — suggest a hypothesis that has not, to our knowledge, been formally articulated: that intracranial aneurysm formation is driven by biofilm-mediated endothelial coherence failure, a multi-mechanism cascade initiated by intramural colonization of cerebral arteries by periodontal pathogens.

This paper develops that hypothesis through four convergent mechanisms: (1) gingipain-mediated proteolytic degradation of endothelial junctional complexes, extracellular matrix, and complement components; (2) heme acquisition from hemoglobin generating local free heme accumulation and iron-dependent oxidative stress at the endothelial interface; (3) MMP induction and collagen degradation in the tunica media producing the characteristic loss of structural wall layers; and (4) suppression of nitric oxide bioavailability through eNOS uncoupling and disruption of the enterosalivary nitrate–nitrite–NO pathway. We further propose that these four mechanisms converge on a terminal pathway — endothelial ferroptosis, an iron-dependent, GPX4-regulated form of lytic cell death characterized by lipid peroxidation — which has recently been shown to propagate to neighboring cells through direct plasma membrane contact [11]. This propagation mechanism offers, for the first time, a biophysically plausible explanation for how a focal infectious insult could produce the circumferential coherence failure required for aneurysm expansion and eventual rupture.

Epistemic status. Every individual mechanism invoked in this framework is supported by Tier 1 peer-reviewed evidence in its respective domain. The integration of these mechanisms into a unified causal chain for intracranial aneurysm formation constitutes a Tier 3 speculative synthesis. No element of this framework has been experimentally validated as a complete pathogenic sequence. The paper is structured to make the hypothesis maximally falsifiable: Section 6 specifies testable predictions, including immunohistochemical signatures in surgical specimens and dose-response relationships in prospective cohorts, that would either substantiate or refute the proposed mechanism.

<!-- SECTION:2 -->

2. The Hemodynamic Model and Its Limits

The prevailing mechanistic account of intracranial aneurysm (IA) formation centers on hemodynamic stress. Arterial bifurcations—particularly those of the circle of Willis—generate regions of aberrant wall shear stress (WSS) where flow separates, recirculates, or impinges at high velocity on the vessel wall [6]. Endothelial cells lining these regions function as mechanosensors: integrins, ion channels (notably Piezo-1), G-proteins, and intercellular junction proteins transduce hemodynamic forces into intracellular signaling cascades that regulate vascular tone, gene expression, and structural remodeling [[12]; [13]]. When WSS deviates from physiological laminar patterns—whether abnormally high at bifurcation apices or pathologically low and oscillatory in recirculation zones—endothelial cells shift toward a pro-inflammatory, pro-oxidant phenotype. Nuclear factor erythroid 2-related factor 2 (Nrf2) cytoprotective signaling is suppressed, reactive oxygen species accumulate, and NF-κB-dependent inflammatory gene expression is upregulated [[14]; [15]]. The resulting endothelial dysfunction initiates a degenerative cascade: smooth muscle cell apoptosis, extracellular matrix degradation by matrix metalloproteinases, and progressive thinning of the arterial wall toward aneurysmal dilation [[6]; [16]].

This model is well supported. Computational fluid dynamics (CFD) studies confirm that intracranial aneurysms cluster at sites of hemodynamic disturbance, and all layers of the vascular wall in established aneurysms show degenerative changes consistent with chronic mechanical overload [16]. Genetic studies reinforce the connection: polymorphisms identified in genome-wide association studies converge on molecular pathways involved in vascular remodeling and endothelial mechanotransduction, including eNOS, elastin (ELN), and CDKN2B-AS1 (ANRIL) at 9p21 [17].

However, the hemodynamic model, as currently constituted, confronts three explanatory gaps that it cannot close on its own terms.

First, the prevalence-rupture mismatch. Unruptured intracranial aneurysms (UIAs) are present in approximately 2–3% of the general population [18]. Yet the annual rupture rate is estimated at 0.5–1% per aneurysm-year for most lesion categories, and aneurysmal subarachnoid hemorrhage (aSAH) affects only 6–9 per 100,000 person-years [39]. The majority of UIAs remain asymptomatic across an individual's lifetime [19]. Every individual with an intact circle of Willis is exposed to bifurcation hemodynamics from birth. If hemodynamic stress were both necessary and sufficient, the clinical expectation would be far higher aneurysm prevalence, or, alternatively, reliable progression from exposure to aneurysm formation. Neither holds. This arithmetic implies that hemodynamic stress is a permissive rather than deterministic variable—necessary to define the anatomical site of vulnerability, but insufficient to explain which exposed individuals progress to aneurysm formation and which formed aneurysms progress to rupture.

Second, the predictive failure. Despite two decades of increasingly sophisticated CFD modeling, hemodynamic parameters alone have not achieved reliable clinical prediction of aneurysm rupture [6]. The field remains divided on whether high WSS or low WSS is the dominant driver of wall degeneration, with evidence supporting both mechanisms in different morphological contexts [40]. Morphological features—irregular shape, daughter sac formation, size—carry independent predictive value but are downstream consequences, not upstream causes [20]. Arterial stiffness, itself a marker of systemic vascular disease rather than local hemodynamics, has recently been associated with increased bleeding risk in aneurysm patients [21]. These observations collectively suggest that the hemodynamic model describes the physics of where aneurysms form, but not the biology of why they progress.

Third, the inflammatory infiltrate problem. Histopathological studies of aneurysm walls consistently reveal macrophage infiltration, complement activation, and NF-κB-mediated cytokine expression [[22]; [17]]. Aneurysm wall biology—specifically, whether the wall is populated by inflammatory cells undergoing active remodeling or by quiescent fibrous tissue—appears to determine clinical behavior more reliably than hemodynamic parameters [7]. This inflammatory signature has been attributed to turbulent flow-induced endothelial activation, but the magnitude and chronicity of the inflammatory response in many aneurysm walls exceeds what hemodynamic stress alone would parsimoniously predict. A purely mechanical model must explain why some hemodynamically identical bifurcations develop intense inflammatory infiltration while others remain quiescent—a question it has not yet answered.

Taken together, these three gaps define the explanatory space into which an upstream biological variable might be introduced—one that interacts with hemodynamic stress but is not reducible to it. The model we propose does not discard hemodynamic theory. Rather, it asks the question that hemodynamics cannot answer on its own terms: what determines which mechanically stressed vessels lose the capacity to maintain endothelial coherence?

<!-- SECTION:6 -->

6. Clinical Implications and Testable Predictions

A hypothesis that cannot be falsified is not a hypothesis. The biofilm-mediated endothelial coherence failure framework, as articulated in the preceding sections, generates a series of testable predictions — each derived from a specific mechanistic link in the proposed cascade and each amenable to evaluation with existing experimental and clinical methods. We organize these predictions by the study design required to evaluate them.

6.1 Immunohistochemical Predictions in Surgical Specimens

If the terminal pathway of aneurysm wall destruction is endothelial ferroptosis propagated through intercellular plasma membrane contacts [11], then resected aneurysm wall tissue should display a specific molecular signature that distinguishes it from apoptotic or necroptotic cell death. Specifically:

Prediction 1. Immunohistochemical analysis of surgically excised intracranial aneurysm walls will demonstrate elevated 4-hydroxynonenal (4-HNE) adducts and oxidized phosphatidylethanolamine species — canonical lipid peroxidation products of ferroptosis — in endothelial and subendothelial regions, with concentrations highest in the thinnest (rupture-prone) wall segments. GPX4 expression should be reduced or absent in these same regions, consistent with loss of the principal anti-ferroptotic defense [[23]; [24]].

Prediction 2. Iron deposition in aneurysm walls, assessed by Perls Prussian blue staining or enhanced iron staining, will co-localize with regions of endothelial denudation and lipid peroxidation. The spatial pattern should be consistent with heme-derived iron accumulation — i.e., co-localization with hemoglobin degradation products — rather than the diffuse ferritin-bound iron storage seen in normal arterial tissue [[25]; [26]].

Prediction 3. If ferroptotic propagation through plasma membrane contacts is the mechanism of circumferential coherence failure, then the advancing edge of endothelial loss in aneurysm walls should exhibit a gradient: cells nearest the denuded zone showing elevated ACSL4 expression (the enzyme that incorporates polyunsaturated fatty acids into phospholipids, sensitizing cells to ferroptosis) and C11-BODIPY-detectable lipid peroxidation, while cells further from the advancing front retain normal GPX4 expression. This wavefront pattern has been demonstrated in optogenetic ferroptosis models [11] and should be detectable in appropriately processed surgical specimens.

6.2 Microbiological Predictions

Prediction 4. 16S rRNA sequencing of resected aneurysm wall tissue will detect P. gingivalis DNA at significantly higher frequency and abundance in aneurysm walls that demonstrate the ferroptotic signature described above compared to control intracranial arterial tissue (obtainable from autopsy specimens of individuals without aneurysmal disease). The detection of gingipain gene sequences (rgpA, rgpB, kgp) — rather than merely P. gingivalis taxonomic markers — would provide stronger evidence for active virulence factor expression at the aneurysm site [8].

Prediction 5. Fluorescence in situ hybridization (FISH) targeting P. gingivalis in aneurysm wall cross-sections will reveal bacterial localization within or immediately adjacent to regions of internal elastic lamina disruption and MMP-positive staining — consistent with the proposed mechanism whereby gingipain-induced MMP activation degrades the structural matrix from an intramural colonization site rather than from the luminal surface.

6.3 Epidemiological Predictions

Prediction 6. In prospective cohort studies of patients with known unruptured intracranial aneurysms under surveillance imaging, the severity of concurrent periodontitis — quantified by validated indices including mean pocket depth, clinical attachment loss, and alveolar bone loss — will predict the rate of aneurysm growth (diameter increase per year) in a dose-response relationship, independent of conventional risk factors (hypertension, smoking, aneurysm size at baseline). The existing Pyysalo et al. data demonstrated an OR of 5.3 for IA presence and an HR of 22.5 for subsequent aSAH [1]; this prediction extends that association to a continuous dose-response gradient with aneurysm progression as the endpoint.

Prediction 7. Patients with intracranial aneurysms who concurrently use antiseptic mouthwash (chlorhexidine or alcohol-based formulations) — which ablate the enterosalivary nitrate-reducing bacteria, reducing systemic nitric oxide bioavailability and raising blood pressure by 2–3.5 mmHg within one week [27] — will demonstrate faster aneurysm growth rates and higher rupture incidence than matched controls who do not use such products. This prediction is derived from Mechanism 4 (nitric oxide suppression) and is independently testable through retrospective pharmacy record analysis linked to aneurysm surveillance databases.

6.4 Interventional Predictions

Prediction 8. Intensive periodontal treatment (scaling and root planing, with adjunctive local antimicrobial therapy directed at P. gingivalis) in patients with co-existing periodontitis and unruptured intracranial aneurysms will reduce the rate of aneurysm wall enhancement on vessel wall MRI — a validated imaging surrogate for wall inflammation and instability [28] — at six months post-treatment. Tonetti et al. (2007) demonstrated that intensive periodontal treatment improved brachial artery endothelial function (flow-mediated dilation) at six months [29]; this prediction extends that finding to the specific vascular bed at risk.

Prediction 9. Administration of ferroptosis inhibitors — specifically ferrostatin-1 analogs or iron chelators — to animal models of aneurysm formation (e.g., elastase-induced or angiotensin II–induced rodent models) will attenuate aneurysm wall thinning and reduce the incidence of rupture. If the terminal pathway is indeed ferroptotic, pharmacological interruption of iron-dependent lipid peroxidation should arrest the propagation phase even if the upstream bacterial insult persists.

6.5 Falsification Criteria

The framework would be substantially weakened by any of the following findings: (a) absence of ferroptotic biomarkers (4-HNE, oxidized PE, reduced GPX4) in a well-powered immunohistochemical survey of resected aneurysm walls; (b) failure to detect P. gingivalis or other periodontal pathogen DNA in aneurysm tissue at rates exceeding those in matched control arteries; (c) absence of a dose-response relationship between periodontitis severity and aneurysm growth in a prospective cohort of adequate size and follow-up duration; or (d) demonstration that ferroptosis inhibition has no effect on aneurysm progression in validated animal models. Each of these negative results would challenge a specific link in the proposed causal chain. The framework would be definitively refuted if ferroptotic signatures are absent from aneurysm walls entirely, as this would eliminate the proposed terminal pathway regardless of the upstream mechanisms.

Epistemic status. Every prediction above is stated in falsifiable form. The individual mechanistic components — gingipain proteolysis, heme-mediated iron toxicity to endothelium, ferroptotic propagation through plasma membrane contacts, enterosalivary nitric oxide disruption — are each supported by Tier 1 evidence in their respective domains. The integration of these components into a unified pathogenic sequence for intracranial aneurysm formation remains a Tier 3 speculative synthesis until the predictions specified here are tested.

<!-- SECTION:3:FAILED -->

Periodontal Disease, Oral Pathogens and Intracranial Aneurysm — Epidemiological Evidence

[SECTION FAILED AFTER 2 RETRIES — reason: draft completed but no content — requires manual write or respawn]

<!-- SECTION:4:FAILED -->

Endothelial Coherence Failure — Four Convergent Mechanisms

[SECTION FAILED AFTER 2 RETRIES — reason: draft completed but no content — requires manual write or respawn]

<!-- SECTION:5 -->

5. Ferroptotic Propagation as the Terminal Pathway

The four convergent mechanisms described in Section 4 — inflammatory priming, extracellular matrix degradation, tight junction disruption, and oxidative stress amplification — each compromise individual endothelial cells. None, however, accounts for the circumferential coherence failure that aneurysm expansion demands. Focal cell loss would be contained by adjacent intact endothelium through compensatory migration and proliferation; aneurysmal walls, by contrast, exhibit confluent zones of endothelial denudation extending across vessel segments. This section advances the HYPOTHESIS that ferroptosis — an iron-dependent form of regulated cell death driven by uncontrolled lipid peroxidation — provides the propagating mechanism that converts focal biofilm-mediated injury into the contiguous endothelial loss required for aneurysmal expansion and rupture.

The authors acknowledge at the outset that ferroptosis in the cerebrovascular endothelial context remains incompletely characterized, though its role in vascular disease broadly is now established [30].

5.1 Iron Loading as the Catalytic Prerequisite

Ferroptosis is mechanistically distinct from apoptosis and necroptosis in its absolute dependence on labile intracellular iron and the consequent peroxidation of polyunsaturated fatty acid (PUFA)-containing phospholipids within the plasma membrane [[30]; [31]]. Dysregulated iron handling increases the labile iron pool, promoting Fenton chemistry, hydroxyl radical generation, and lipid peroxidation cascades — the biochemical signature of ferroptotic commitment [32].

Several periodontal pathogens — Porphyromonas gingivalis most prominently — possess iron acquisition systems, including gingipain-mediated hemoglobin degradation, capable of liberating heme iron from host proteins [41]. INTERPRETATION: Bacteria translocated to the vessel wall, as the bacteremia evidence in Section 3 suggests, would deliver catalytic iron directly into the subendothelial compartment, bypassing the transferrin-bound pools that intact endothelium regulates.

Biofilm-derived lipopolysaccharide (LPS) independently alters endothelial iron homeostasis: suppressing ferritin synthesis while upregulating transferrin receptor (TFRC) expression. This combination expands the labile iron pool available for Fenton reactions. Concurrently, glutathione peroxidase 4 (GPx4) — the primary enzymatic restraint on ferroptotic lipid peroxidation — requires reduced glutathione as substrate. The oxidative burden imposed by biofilm-derived reactive oxygen species depletes glutathione reserves, removing this enzymatic brake.

The resulting biochemical state — elevated labile iron, active Fenton chemistry, and suppressed GPx4 activity — constitutes the necessary preconditions for ferroptotic execution.

5.2 Propagation Through Plasma Membrane Contacts

What distinguishes ferroptosis as a candidate terminal mechanism is its documented capacity for intercellular propagation. In renal epithelial monolayer studies, ferroptotic death spreads in wave-like fronts through direct plasma membrane contact rather than soluble mediators [42]. The proposed mechanism involves oxidized phospholipid species acting as membrane-permeable catalysts of lipid peroxidation in adjacent cells — chain-reaction kinetics applied to contiguous cell populations.

HYPOTHESIS (principal evidentiary gap): Whether this propagation dynamic operates in endothelial monolayers has not been confirmed in cerebrovascular preparations. This extrapolation from epithelial to endothelial tissue constitutes the central unresolved assumption of the present framework. However, recent evidence that ferroptosis disrupts endothelial integrity and amplifies downstream vascular pathology [33] supports the biological plausibility of this extrapolation, and ferroptosis has been implicated in cerebrovascular injury in the context of ischemic stroke [34].

The geometry of aneurysm-prone sites is relevant. At bifurcation apices, the endothelial monolayer is continuous around a curved surface. A propagating ferroptotic wave initiated at a focal iron-rich bacterial deposit would advance circumferentially, stripping endothelial coverage in the pattern that pathological specimens demonstrate. Hemodynamic models, which invoke wall shear stress as the primary driver of endothelial injury, do not generate this confluent loss pattern; ferroptotic propagation does.

5.3 Epistemic Status

Ferroptosis has been characterized in hepatic, renal, neuronal, and cardiac ischemia-reperfusion contexts [35], with recent extension to endothelial cells in multiple vascular disease states [[30]; [33]]. Its operation in intracranial endothelium under biofilm-derived iron loading has not been directly demonstrated. The propagation wave model derives from epithelial, not endothelial, experimental systems.

These gaps define the experimental program detailed in Section 6. The mechanistic logic — iron loading → GPx4 suppression → PUFA peroxidation → contact-dependent propagation — is internally consistent and yields falsifiable predictions. Whether it describes biological reality in the aneurysmal vessel wall is an empirical question that current data do not resolve.

<!-- SECTION:3 -->

3. Periodontal Disease, Oral Pathogens and Intracranial Aneurysm — Epidemiological Evidence

The hemodynamic model of intracranial aneurysm formation identifies wall shear stress as a necessary condition for endothelial injury. It does not explain why that injury progresses to irreversible structural failure in some individuals and resolves in others. The epidemiological literature on periodontal disease supplies a candidate mechanism for this divergence — one with direct tissue-level evidence and an effect size that demands engagement.

3.1 The Core Epidemiological Signal

Pyysalo et al. (2019) reported a prospective Finnish cohort study (Health 2000 Survey; n = 5,170; 13-year follow-up) in which severe periodontitis at baseline — defined as ≥3 teeth with ≥6 mm gingival pockets — was associated with a hazard ratio of 22.5 for aneurysmal subarachnoid hemorrhage (95% CI 3.6–139.5; p = 0.001) [1]. Gingival bleeding detected in 4–6 sextants carried an independent hazard ratio of 8.3 (95% CI 1.5–46.1; p = 0.015).

FACT: The point estimate is exceptional by cardiovascular epidemiology standards. For comparison, established risk factors for cerebrovascular disease — hypertension, smoking, diabetes — typically produce hazard ratios of 2 to 6. However, the confidence intervals are wide, reflecting the low absolute event rate of aSAH (approximately 6–9 per 100,000 person-years). This width is not interpretive noise; it is the expected statistical signature of a rare outcome in a moderately sized cohort. The lower bound of the confidence interval (3.6) still exceeds the typical hazard ratio for smoking in cerebrovascular disease.

The case-control component of the same study corroborates the prospective signal with a dose-response gradient: periodontitis (≥4 mm pocket depth) carried an odds ratio of 5.3 (95% CI 1.1–25.9) for the presence of intracranial aneurysm, while severe periodontitis (≥6 mm pocket depth) carried an odds ratio of 6.3 (95% CI 1.3–31.4). The monotonic increase across severity categories is consistent with a biological threshold effect and inconsistent with the flat or irregular gradient expected from residual confounding.

This finding does not stand alone. The same research group reported the detection of oral bacteria DNA within resected intracranial aneurysm walls [1]. INTERPRETATION: The co-localization of oral pathogen-derived nucleic acid with the tissue site of structural failure elevates the periodontitis–aneurysm relationship from statistical association to a candidate causal chain. Hematogenous seeding of arterial walls from a chronic mucosal reservoir is a mechanism with established precedent in infective endocarditis and atherosclerotic plaque biology [PMID: 26777430; PMID: 15898933].

3.2 The Confounding Objection

The most substantive methodological criticism is that periodontitis and intracranial aneurysm share upstream determinants — smoking, hypertension, socioeconomic disadvantage, systemic inflammation — and the observed association is therefore spurious. This objection is serious and must be addressed on its own terms.

Three lines of evidence limit, without eliminating, the confounding explanation.

First, covariate adjustment. The Finnish cohort associations remained statistically significant after adjustment for gender, smoking status, and hypertension [1]. FACT (conditional): Adjusted associations that survive correction for the most plausible shared risk factors are methodologically more informative than unadjusted associations, though they do not establish causation. Residual confounding by unmeasured variables — particularly socioeconomic status, which is incompletely captured by smoking and hypertension adjustment — cannot be excluded. This is a standard limitation of observational epidemiology, not a specific weakness of this dataset.

Second, pathogen-specific biological signature. Smoking and hypertension damage arterial walls through oxidative stress and mechanical load. They do not deposit periodontal pathogens in cerebrovascular tissue. The detection of oral bacteria DNA in aneurysm walls introduces a pathogen-specific signature that shared lifestyle risk factors cannot produce. HYPOTHESIS: If confounding alone explained the epidemiological association, pathogen-derived nucleic acid would not be present at the lesion site at frequencies exceeding background contamination. The presence of such material is evidence of a mechanism that operates independently of shared upstream exposures.

Third, dose-response specificity. The association is with severe periodontitis — not mild gingivitis, not the presence of any gingival inflammation. Severe periodontal disease produces a distinct systemic burden: sustained transient bacteremia episodes, elevated circulating IL-6, C-reactive protein, and fibrinogen, and a chronically ulcerated subgingival epithelium estimated at 8–20 cm² of wound surface in moderate-to-severe disease [WS4-Elimination-Oral-Systemic-Health-Protocol]. This ulcerated surface functions as a dissemination portal through which both intact bacteria and their proteolytic products — including gingipains (Arg-gingipain and Lys-gingipain), cysteine proteases that cleave complement C5, degrade immunoglobulins IgG and IgA, and activate matrix metalloproteinase-mediated collagen destruction — enter systemic circulation [WS4-Elimination-Oral-Systemic-Health-Protocol]. The specificity of the association with severe disease is consistent with a threshold-dependent biological mechanism, not a graded confounding artifact.

3.3 The Gap Hemodynamics Cannot Fill

INTERPRETATION: Unruptured intracranial aneurysm prevalence in the general population is estimated at 1–3% by conservative radiological definitions, rising to 3.2% or higher under inclusive criteria [[36]; [37]]. Hemodynamic models identify the anatomical territories at risk — arterial bifurcations of the circle of Willis exposed to oscillatory wall shear stress — but anatomical vulnerability is not biological susceptibility. The model predicts where aneurysms can form. It does not predict in whom.

HYPOTHESIS: Periodontal disease does not cause intracranial aneurysms independently of mechanical forces. It constitutes a chronic systemic priming condition that degrades endothelial repair capacity in the vascular territories most exposed to pathological shear. The proposed mechanism is an interaction, not a simple additive risk: hemodynamic stress activates the endothelium at bifurcation points, upregulating adhesion molecules and increasing permeability; hematogenously disseminated periodontal pathogens and their proteolytic products then exploit that activated surface, accelerating matrix degradation beyond the threshold of repair. The epidemiological signal — HR 22.5 — reflects this interaction.

This framework generates falsifiable predictions. First, periodontal treatment initiated in patients with known unruptured aneurysms should slow aneurysm growth rate. Proof-of-concept exists: Tonetti et al. (2007) demonstrated that intensive periodontal treatment improved brachial artery endothelial function (flow-mediated dilation) at six months [29], and D'Aiuto et al. (2004) showed measurable reductions in systemic CRP following periodontal intervention [43]. The systemic vascular effect of eliminating the oral reservoir is established; its translation to cerebrovascular endpoints has not been tested. Second, tissue from ruptured aneurysms in patients with severe periodontitis should carry a higher oral pathogen burden than tissue from periodontally healthy patients. Neither prediction has been evaluated in adequately powered prospective studies.

<!-- SECTION:4 -->

4. Endothelial Coherence Failure — Four Convergent Mechanisms

The epidemiological signal reviewed in Section 3 establishes a statistical and anatomical association between severe periodontal disease and intracranial aneurysm formation. Association is not mechanism. This section constructs the mechanistic architecture that connects hematogenously disseminated periodontal pathogens — principally Porphyromonas gingivalis — to a specific, compound failure mode of the cerebrovascular endothelium. The argument is that four distinct but convergent biological processes — gingipain-mediated proteolysis, heme iron accumulation, MMP-driven matrix degradation, and eNOS uncoupling — each individually sufficient to impair endothelial function, combine under conditions of chronic periodontal bacteremia to produce what we term endothelial coherence failure: an irreversible loss of the structural, signaling, and barrier properties that collectively define a competent endothelial layer.

HYPOTHESIS: No single mechanism is proposed as independently sufficient to drive aneurysm formation. The convergence of all four is what distinguishes progressive, irreparable wall remodeling from the reversible endothelial perturbations that characterize ordinary hemodynamic stress.

4.1 Gingipain-Mediated Proteolysis

P. gingivalis expresses cysteine proteases — Arg-gingipain (RgpA, RgpB) and Lys-gingipain (Kgp) — that constitute its primary virulence armamentarium. FACT: These proteases cleave fibrinogen, activate the contact pathway of coagulation, degrade complement C5 to generate C5a independently of the canonical complement cascade, and proteolytically inactivate IgG and IgA [WS4-Elimination-Oral-Systemic-Health-Protocol]. Within the endothelial microenvironment, gingipains cleave VE-cadherin, the transmembrane adhesion molecule that maintains adherens junction integrity between adjacent endothelial cells [44]. INTERPRETATION: Loss of VE-cadherin homophilic binding increases paracellular permeability, permitting trans-endothelial passage of both intact bacteria and their lipopolysaccharide — specifically the atypical lipid A moiety of P. gingivalis, which activates TLR2 rather than TLR4, generating a pro-inflammatory signal qualitatively distinct from gram-negative endotoxemia and partially resistant to conventional endotoxin-neutralizing mechanisms.

Critically, gingipains are released in outer membrane vesicles that remain proteolytically active in systemic circulation and resist clearance by plasma protease inhibitors. HYPOTHESIS: In the high-shear, mechanically activated endothelium at arterial bifurcations, upregulated surface adhesion molecules — including ICAM-1, VCAM-1, and E-selectin — provide docking substrates that concentrate circulating P. gingivalis vesicles at precisely those anatomical sites already predisposed to wall stress injury. The shear-periodontitis interaction proposed in Section 3 is therefore not merely statistical; it has a receptor-level molecular rationale.

4.2 Heme Iron Accumulation

P. gingivalis is an obligate asaccharolytic anaerobe that satisfies its iron requirements exclusively through heme acquisition from hemoglobin and other heme-containing proteins [41]. This biology has systemic consequences. FACT: P. gingivalis cell-surface Kgp degrades hemoglobin, generating heme that is sequestered at the bacterial outer membrane as a black ferrous pigment, μ-oxo bisheme [41]. During bacteremia episodes, this accumulated heme is released into the vascular compartment.

INTERPRETATION: Free heme is a potent oxidative stressor. It catalyzes Fenton chemistry — Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + •OH — generating hydroxyl radical that oxidizes membrane phospholipids, mitochondrial cardiolipin, and cytoskeletal actin with equivalent efficiency. In the cerebrovascular endothelium, where the blood-brain barrier imposes unusually stringent redox homeostasis, free heme loading imposes an oxidative burden that depletes glutathione peroxidase 4 (GPX4) — the primary cellular defense against phospholipid hydroperoxides. Section 5 develops the downstream consequence of GPX4 depletion through the ferroptotic pathway. Here, the relevant observation is structural: oxidation of F-actin stress fibers disrupts the cortical cytoskeletal architecture that physically anchors tight junction proteins ZO-1 and occludin to the plasma membrane, producing junction retraction independent of protease activity.

HYPOTHESIS: Heme iron accumulation represents the second, cytoskeletal arm of endothelial coherence failure — operating through oxidative mechanics rather than proteolysis, but converging on the same structural endpoint: paracellular gap formation.

4.3 MMP-Mediated Matrix Degradation

FACT: P. gingivalis infection upregulates endothelial and vascular smooth muscle cell expression of matrix metalloproteinases, particularly MMP-2 and MMP-9, through NF-κB and AP-1 transcriptional activation. MMP-9 degrades type IV collagen — the primary structural component of the vascular basement membrane — and gelatin at concentrations proportional to the pathogen burden. Simultaneously, gingipains directly activate latent pro-MMP forms by cleaving the cysteine-switch prodomain, bypassing the tissue inhibitor of metalloproteinase (TIMP) regulatory system.

INTERPRETATION: The basement membrane serves a function beyond structural support: it presents heparan sulfate proteoglycans that sequester fibroblast growth factors, VEGF, and angiopoietins required for endothelial survival signaling. Focal basement membrane degradation by activated MMPs releases this reservoir in an unregulated pulse, followed by its rapid depletion. HYPOTHESIS: This creates a survival signaling deficit — a local deprivation of growth factor-mediated PI3K-Akt signaling — that renders the overlying endothelium simultaneously more susceptible to apoptosis and less capable of gap repair. In the context of the elevated wall shear stress at bifurcation points, where turnover-driven endothelial replacement is physiologically elevated, this repair deficit is geometrically concentrated at the sites most requiring structural maintenance.

4.4 eNOS Uncoupling

Endothelial nitric oxide synthase (eNOS) is both the primary vasodilator of the cerebrovascular endothelium and a critical regulator of endothelial inflammatory tone. FACT: eNOS requires its cofactor tetrahydrobiopterin (BH4) for coupled, NO-generating activity; when BH4 is depleted, eNOS becomes uncoupled, producing superoxide (O₂•⁻) rather than NO [45]. INTERPRETATION: P. gingivalis-driven oxidative stress depletes BH4 through oxidation to dihydrobiopterin (BH2), converting eNOS from a NO-generating to a superoxide-generating enzyme — a pathological switch that is self-amplifying: superoxide reacts with residual NO to form peroxynitrite (ONOO⁻), which further oxidizes BH4 and nitrosylates cytoskeletal proteins. The simultaneous loss of NO bioavailability removes the constitutive suppression of NF-κB that physiological eNOS activity maintains, permitting sustained inflammatory gene transcription.

HYPOTHESIS: eNOS uncoupling is the third arm of coherence failure and the only one that is enzymatically self-sustaining. Once initiated, it requires no continued pathogen input to perpetuate endothelial dysfunction. This may explain the persistence of vascular inflammation observed in some aneurysm tissues in which active bacterial colonization is no longer detectable.

4.5 Convergence and the Irreversibility Threshold

Each mechanism individually is compensable. INTERPRETATION: Normal endothelium tolerates transient VE-cadherin disruption through rapid junctional reassembly; tolerates moderate oxidative stress through Nrf2-mediated antioxidant upregulation; tolerates focal matrix degradation through TIMP-mediated MMP restraint; and tolerates partial eNOS uncoupling through BH4 regeneration via dihydrofolate reductase. What constitutes coherence failure — defined here as irreversible loss of barrier, signaling, and structural competence — is the simultaneous saturation of all four repair systems by a chronic, multi-mechanistic insult.

HYPOTHESIS: Severe periodontitis, through sustained bacteremic episodes over months to years, imposes exactly this simultaneous multi-front burden. The aneurysm wall is not merely weakened; its endothelium loses the functional coherence required to respond adaptively to the hemodynamic forces that continue to act upon it. At this threshold — not before it — the structural transition from arterial wall remodeling to aneurysm initiation becomes thermodynamically favorable.

This framework carries a direct implication for in vivo evidence standards. Demonstrating any single mechanism in isolation — gingipain activity in plasma, elevated MMP-9 in periodontitis patients, or reduced flow-mediated dilation as a proxy for eNOS uncoupling — is necessary but not sufficient to test the convergence hypothesis. What is required is concurrent measurement of all four pathways in the same patients, stratified by aneurysm presence and periodontal severity. The absence of such data represents an in vivo evidence gap that this framework is specifically designed to motivate closing.

<!-- SECTION:7 -->

7. Conclusion

Intracranial aneurysm has long been framed as a disease of fluid mechanics — a structural failure provoked by aberrant wall shear stress at arterial bifurcations. The framework advanced in this paper does not discard that model; it contextualizes it. Hemodynamic stress, on this account, is permissive rather than deterministic: it defines which vessel wall segments are vulnerable, but it does not explain why endothelial integrity fails at those sites in some individuals and not others. The missing variable, we propose, is microbial. [INTERPRETATION]

The epidemiological signal linking periodontal disease to intracranial aneurysm is now sufficiently consistent to demand mechanistic explanation rather than further correlational study. Oral pathogens — principally Porphyromonas gingivalis, Fusobacterium nucleatum, and Streptococcus species — are not merely bystanders in systemic vascular disease. They are, under the framework presented here, active agents of endothelial coherence failure operating through four convergent and mutually reinforcing pathways: inflammatory cytokine priming, matrix metalloproteinase-driven extracellular matrix degradation, tight junction protein loss, and oxidative phospholipid injury culminating in ferroptotic propagation. [HYPOTHESIS]

The concept of endothelial coherence failure — the transition from a self-maintaining, mechanically integrated monolayer to a fragmented, ferroptosis-susceptible cell population — is the theoretical core of this framework. It repositions the endothelium not as a passive conduit lining but as the primary site of aneurysm pathogenesis, one whose resilience or collapse is substantially determined by the chronic inflammatory burden imposed by periodontal biofilm. [HYPOTHESIS]

This reframing carries direct clinical consequences. If periodontal disease is a modifiable upstream driver of aneurysm risk rather than an incidental comorbidity, then oral health intervention becomes a legitimate target within neurovascular medicine — not a peripheral curiosity. The testable predictions enumerated in Section 6 provide a structured pathway from hypothesis to evidence: ferroptosis biomarker profiling, randomized periodontal treatment trials with vascular endpoints, and in vitro reconstitution of the biofilm–endothelium injury cascade. [INTERPRETATION]

We do not claim causation is established. We claim the mechanistic architecture is coherent, the epidemiological foundation is suggestive, and the cost of continued inattention to this pathway — measured in undetected aneurysms and preventable ruptures — is too high to justify delay. Intracranial aneurysm may be, in a meaningful and therapeutically actionable sense, an infectious-coherence disease. That hypothesis now warrants the rigor of direct experimental test. [HYPOTHESIS]

References

  1. Hallikainen J, Lindgren A, Savolainen J, Selander T, Jula A, Närhi M, et al. Periodontitis and gingival bleeding associate with intracranial aneurysms and risk of aneurysmal subarachnoid hemorrhage. Neurosurgical review. 2020;43(2):669-679. doi:10.1007/s10143-019-01097-1
  2. Pontes FGB, da Silva EM, Baptista-Silva JC, Vasconcelos V. Treatments for unruptured intracranial aneurysms. The Cochrane database of systematic reviews. 2021;5(5):CD013312. doi:10.1002/14651858.CD013312.pub2
  3. Gibson DP, Kulwin CG, Shah KJ, Payner TD, DeNardo AJ, Amuluru K, et al. Natural History of Cerebral Aneurysms: Risk Factors for Rupture and Implications for Management. Neuroimaging clinics of North America. 2025;35(3):303-310. doi:10.1016/j.nic.2025.04.001
  4. Watson E, Ding D, Khattar NK, Everhart DE, James RF. Neurocognitive outcomes after aneurysmal subarachnoid hemorrhage: Identifying inflammatory biomarkers. Journal of the neurological sciences. 2018;394:84-93. doi:10.1016/j.jns.2018.06.021
  5. de Oliveira Manoel AL, Macdonald RL. Neuroinflammation as a Target for Intervention in Subarachnoid Hemorrhage. Frontiers in neurology. 2018;9:292. doi:10.3389/fneur.2018.00292
  6. Jeong W, Rhee K. Hemodynamics of cerebral aneurysms: computational analyses of aneurysm progress and treatment. Computational and mathematical methods in medicine. 2012;2012:782801. doi:10.1155/2012/782801
  7. Marbacher S, Niemelä M, Hernesniemi J, Frösén J. Recurrence of endovascularly and microsurgically treated intracranial aneurysms-review of the putative role of aneurysm wall biology. Neurosurgical review. 2019;42(1):49-58. doi:10.1007/s10143-017-0892-2
  8. Widziolek M, Mieszkowska A, Marcinkowska M, Salamaga B, Folkert J, Rakus K, et al. Gingipains protect Porphyromonas gingivalis from macrophage-mediated phagocytic clearance. PLoS pathogens. 2025;21(1):e1012821. doi:10.1371/journal.ppat.1012821
  9. Dashper SG, Cross KJ, Slakeski N, Lissel P, Aulakh P, Moore C, et al. Hemoglobin hydrolysis and heme acquisition by Porphyromonas gingivalis. Oral microbiology and immunology. 2004;19(1):50-6.
  10. Hajishengallis G. Complement and periodontitis. Biochemical pharmacology. 2010;80(12):1992-2001. doi:10.1016/j.bcp.2010.06.017
  11. Roeck BF, Lotfipour Nasudivar S, Vorndran MRH, Schueller L, Yapici FI, Rübsam M, et al. Ferroptosis spreads to neighboring cells via plasma membrane contacts. Nature communications. 2025;16(1):2951. doi:10.1038/s41467-025-58175-w
  12. Fisher AB, Chien S, Barakat AI, Nerem RM. Endothelial cellular response to altered shear stress. American journal of physiology. Lung cellular and molecular physiology. 2001;281(3):L529-33.
  13. Labrador V, Chen KD, Li YS, Muller S, Stoltz JF, Chien S. Interactions of mechanotransduction pathways. Biorheology. 2003;40(1-3):47-52.
  14. Hasan M, Al-Thani H, El-Menyar A, Zeidan A, Al-Thani A, Yalcin HC. Disturbed hemodynamics and oxidative stress interaction in endothelial dysfunction and AAA progression: Focus on Nrf2 pathway. International journal of cardiology. 2023;389:131238. doi:10.1016/j.ijcard.2023.131238
  15. Chen B, Lu Y, Chen Y, Cheng J. The role of Nrf2 in oxidative stress-induced endothelial injuries. The Journal of endocrinology. 2015;225(3):R83-99. doi:10.1530/JOE-14-0662
  16. Diaz O, Rangel-Castilla L. Endovascular treatment of intracranial aneurysms. Handbook of clinical neurology. 2016;136:1303-9. doi:10.1016/B978-0-444-53486-6.00067-3
  17. Samuel N, Radovanovic I. Genetic basis of intracranial aneurysm formation and rupture: clinical implications in the postgenomic era. Neurosurgical focus. 2019;47(1):E10. doi:10.3171/2019.4.FOCUS19204
  18. Nasr DM, Brown RD Jr. Management of Unruptured Intracranial Aneurysms. Current cardiology reports. 2016;18(9):86. doi:10.1007/s11886-016-0763-4
  19. Howard BM, Hu R, Barrow JW, Barrow DL. Comprehensive review of imaging of intracranial aneurysms and angiographically negative subarachnoid hemorrhage. Neurosurgical focus. 2019;47(6):E20. doi:10.3171/2019.9.FOCUS19653
  20. Abboud T, Rustom J, Bester M, Czorlich P, Vittorazzi E, Pinnschmidt HO, et al. Morphology of Ruptured and Unruptured Intracranial Aneurysms. World neurosurgery. 2017;99:610-617. doi:10.1016/j.wneu.2016.12.053
  21. Gel MS, Postuk HÇ, Daltaban İS, Özer S, Aydin E, Keskin E. Relationship between bleeding risk and arterial stiffness in patients with cerebral aneurysms. Turkish journal of medical sciences. 2025;55(1):209-214. doi:10.55730/1300-0144.5959
  22. Sun B, Liu Z, Yu Z. miRNA-323a-3p promoted intracranial, aneurysm-induced inflammation via AMPK/NF-κB signaling pathway by AdipoR1. Advances in clinical and experimental medicine : official organ Wroclaw Medical University. 2022;31(11):1243-1254. doi:10.17219/acem/151053
  23. Seibt TM, Proneth B, Conrad M. Role of GPX4 in ferroptosis and its pharmacological implication. Free radical biology & medicine. 2019;133:144-152. doi:10.1016/j.freeradbiomed.2018.09.014
  24. Stockwell BR, Jiang X, Gu W. Emerging Mechanisms and Disease Relevance of Ferroptosis. Trends in cell biology. 2020;30(6):478-490. doi:10.1016/j.tcb.2020.02.009
  25. Balla J, Vercellotti GM, Jeney V, Yachie A, Varga Z, Eaton JW, et al. Heme, heme oxygenase and ferritin in vascular endothelial cell injury. Molecular nutrition & food research. 2005;49(11):1030-43.
  26. Balla J, Vercellotti GM, Jeney V, Yachie A, Varga Z, Jacob HS, et al. Heme, heme oxygenase, and ferritin: how the vascular endothelium survives (and dies) in an iron-rich environment. Antioxidants & redox signaling. 2007;9(12):2119-37.
  27. Kapil V, Haydar SM, Pearl V, Lundberg JO, Weitzberg E, Ahluwalia A. Physiological role for nitrate-reducing oral bacteria in blood pressure control. Free radical biology & medicine. 2013;55:93-100. doi:10.1016/j.freeradbiomed.2012.11.013
  28. Aksakal M, Chang D, Kharaji M, Watt C, Levitt MR, Hippe DS, et al. Predicting Vasospasm and Delayed Cerebral Ischemia in Aneurysmal SAH: The Role of Vessel Wall MRI. AJNR. American journal of neuroradiology. 2026. doi:10.3174/ajnr.A9024
  29. Tonetti MS, D'Aiuto F, Nibali L, Donald A, Storry C, Parkar M, et al. Treatment of periodontitis and endothelial function. The New England journal of medicine. 2007;356(9):911-20.
  30. Zhang H, Zhou S, Sun M, Hua M, Liu Z, Mu G, et al. Ferroptosis of Endothelial Cells in Vascular Diseases. Nutrients. 2022;14(21). doi:10.3390/nu14214506
  31. Li YY, Ma B, Lu JW, Zhang KC, Ma C, Bai SF, et al. Lysosome-Featured Cell Aggregate-Released Extracellular Vesicles Regulate Iron Homeostasis and Alleviate Post-Irradiation Endothelial Ferroptosis for Mandibular Regeneration. Advanced science (Weinheim, Baden-Wurttemberg, Germany). 2025;12(34):e05070. doi:10.1002/advs.202505070
  32. Srinivasamurthy SK, Mittal P, Goyal A, Ballal S, Maharana L, Goyal K, et al. Ferroptosis and iron homeostasis in chronic obstructive pulmonary disease: Therapeutic opportunities of iron chelators. Journal of trace elements in medicine and biology : organ of the Society for Minerals and Trace Elements (GMS). 2025;92:127766. doi:10.1016/j.jtemb.2025.127766
  33. Karbakhsh Ravari F, Ghasemi Gorji M, Rafiei A. From iron-driven cell death to clot formation: The emerging role of ferroptosis in thrombogenesis. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie. 2025;189:118328. doi:10.1016/j.biopha.2025.118328
  34. Hu X, Bao Y, Li M, Zhang W, Chen C. The role of ferroptosis and its mechanism in ischemic stroke. Experimental neurology. 2024;372:114630. doi:10.1016/j.expneurol.2023.114630
  35. Yu S, Pang Z, Fang H, Liu C. Ferroptosis in cardiovascular diseases: molecular mechanisms and a novel therapeutic target. Molecular biomedicine. 2026;7(1). doi:10.1186/s43556-026-00420-9
  36. Pierot L, Gawlitza M, Soize S. Unruptured intracranial aneurysms: management strategy and current endovascular treatment options. Expert review of neurotherapeutics. 2017;17(10):977-986. doi:10.1080/14737175.2017.1371593
  37. Johnsen LH, Herder M, Vangberg T, Kloster R, Ingebrigtsen T, Isaksen JG, et al. Prevalence of unruptured intracranial aneurysms: impact of different definitions - the Tromsø Study. Journal of neurology, neurosurgery, and psychiatry. 2022;93(8):902-907. doi:10.1136/jnnp-2022-329270
  38. Nieuwkamp DJ, Setz LE, Algra A, Linn FH, de Rooij NK, Rinkel GJ, et al.. Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis. Lancet Neurol. 2009;8(7):635-42. doi:10.1016/S1474-4422(09)70126-7
  39. Brisman JL, Song JK, Newell DW. Cerebral aneurysms. N Engl J Med. 2006;355(9):928-39. doi:10.1056/NEJMra052760
  40. Xiang J, Tutino VM, Snyder KV, Meng H. CFD: computational fluid dynamics or confounding factor dissemination? The role of hemodynamics in intracranial aneurysm rupture risk assessment. AJNR Am J Neuroradiol. 2014;35(10):1849-57. doi:10.3174/ajnr.A3710
  41. Smalley JW, Byrne DP, Birss AJ, Wojtowicz H, Sroka A, Potempa J, et al.. HmuY haemophore and gingipain proteases constitute a unique syntrophic system of haem acquisition by Porphyromonas gingivalis. PLoS One. 2011;6(2):e17182. doi:10.1371/journal.pone.0017182
  42. Riegman M, Sagie L, Galed C, Levin T, Steinberg N, Dixon SJ, et al.. Ferroptosis occurs through an osmotic mechanism and propagates independently of cell rupture. Nat Cell Biol. 2020;22(9):1042-1048. doi:10.1038/s41556-020-0565-1
  43. PMID 15494559 — citation unavailable
  44. Farrugia C, Stafford GP, Potempa J, Wilkinson RN, Chen Y, Murdoch C, et al.. Mechanisms of vascular damage by systemic dissemination of the oral pathogen Porphyromonas gingivalis. FEBS J. 2021;288(5):1479-1495. doi:10.1111/febs.15486
  45. Förstermann U, Münzel T. Endothelial nitric oxide synthase in vascular disease: from marvel to menace. Circulation. 2006;113(13):1708-14. doi:10.1161/CIRCULATIONAHA.105.602532