Skip to main content

Inflammatory Memory of the Healed Extraction Site: A Mechanistic Framework and Translational Protocol for Testing Whether Prior Periodontitis Conditions the Biological Response to Implant Placement

Paper type
Mechanistic research framework and translational study protocol
Status
Original Work · Working Paper
Contents
  1. 1Introduction
  2. 2Background and Current Scientific Evidence
  3. 3Research Gap
  4. 4Research Objective and Central Hypothesis
  5. 5Scientific and Theoretical Framework
  6. 6Methodology
  7. 7Analytical Development
  8. 8Discussion
  9. 9Clinical Significance
  10. 10Limitations
  11. 11Future Validation
  12. 12Conclusion

Abstract

Background. A history of periodontitis is a consistent risk indicator for peri-implantitis and implant loss. The association is usually attributed to persistent pathogens, shared host susceptibility, and reduced bone volume. Research in innate immunity has identified a further possibility: inflammatory exposure leaves durable epigenetic and metabolic changes in myeloid progenitors, epithelial stem cells, and stromal cells that alter their response to later stimuli. Experimental periodontitis induces such training in bone marrow progenitors, and gingival fibroblasts from periodontitis tissue show features of trained immunity in vitro.

Objective. To state a testable hypothesis about whether a healed extraction site retains a memory of prior periodontal inflammation that modifies the response to implant placement, and to design experiments that separate local memory from central memory and from non-memory explanations.

Framework. Three candidate reservoirs are distinguished: centrally trained myeloid progenitors, locally persisting stromal and epithelial cells, and tissue-resident immune cells. Implant osteotomy is treated as a second stimulus that recalls the memory. The response is modeled as a baseline multiplied by local and central priming terms that decay with time since resolution.

Proposed methods. A murine sequence of ligature-induced periodontitis, resolution, extraction, healing, and implant placement, with split-mouth controls, bone marrow chimeras, and graded healing intervals. A human cohort with a two-by-two structure (patient history by site history) in which tissue normally discarded at osteotomy is profiled for chromatin accessibility, histone marks, and response to restimulation, and early healing is followed clinically.

Central hypothesis. Sites of prior periodontitis retain a locally encoded hyperresponsive state that persists beyond clinical healing, amplifies the inflammatory response to implant placement, and decays with time.

Conclusions. The hypothesis is biologically plausible and unproven. Inflammatory memory can be adaptive, and the direction of any effect on osseointegration is uncertain. No experimental results are reported.

Keywords: trained immunity; inflammatory memory; periodontitis; tooth extraction; osseointegration; epigenetics; gingival fibroblasts; myelopoiesis

1. Introduction

Patients who lose teeth to periodontitis and receive implants do less well than patients who lose teeth for other reasons. In a recent meta-analysis, a history of periodontitis was associated with a relative risk of 4.09 (95% CI 1.93–8.58) for incident peri-implantitis and with greater marginal bone loss (weighted mean difference 0.75 mm; 95% CI 0.18–1.31) [1]. A 10-year prospective cohort reported implant survival of 90.5% with a history of chronic periodontitis and 96.5% without, and peri-implantitis incidence of 28.6% against 5.8% [2]. Another 10-year cohort found survival declining from periodontally healthy to moderately and severely compromised patients [3]. Earlier meta-analyses and the 2017 World Workshop reached the same qualitative conclusion [4–7].

The standard explanations are reasonable. Periodontal pathogens persist at remaining teeth and colonize implants. The host traits that permitted periodontitis, whether genetic, behavioral, or systemic, are still present. Sites that lost teeth to periodontitis have less bone. Each explanation implies that the extraction site itself, once healed, is biologically neutral: the periodontal ligament is gone, the socket has filled with bone, the mucosa has closed, and what remains is tissue with no record of what happened there.

Work in immunology over the past decade gives reason to question that assumption. Innate immune cells and their progenitors undergo long-lasting functional reprogramming after inflammatory exposure, a phenomenon termed trained immunity and mediated by epigenetic and metabolic rewiring [8, 9]. The reprogramming can be imprinted on hematopoietic progenitors in the bone marrow and thereby outlast the short life of circulating myeloid cells [10–12]. Memory is not confined to immune cells. Epidermal stem cells retain accessible chromatin at inflammation-induced loci after the inflammation has resolved and respond faster to a later insult [13, 14]; airway epithelial progenitors from patients with chronic allergic inflammation retain a transcriptional memory in culture [15].

Periodontitis has entered this literature directly. In mice, ligature-induced periodontitis produced trained myelopoiesis in bone marrow progenitors through interleukin-1 receptor signaling, and marrow transplanted from mice with periodontitis worsened arthritis in recipients [16]. Training of osteoclast precursors promoted inflammatory bone loss in experimental periodontitis and arthritis [17]. Human gingival fibroblasts stimulated with lipopolysaccharide from Porphyromonas gingivalis, and fibroblasts isolated from inflamed periodontitis tissue, showed metabolic and epigenetic features of trained immunity [18]. Reviews have drawn out the implications for comorbidities and for the persistence of periodontal disease [19–22].

None of this work concerns implants. A PubMed search combining osseointegration with trained immunity or immune memory returned one record, which was unrelated to periodontitis (Appendix A). The question this manuscript poses is therefore open as far as the search could determine: does a site that has healed after extraction of a periodontally diseased tooth carry a memory of that disease, and does the memory change how the site responds when an implant is placed? This manuscript develops the hypothesis, identifies what would distinguish it from the standard explanations, and describes a translational program to test it.

2. Background and Current Scientific Evidence

Trained immunity and its central reservoir

Trained immunity is defined operationally: after a primary stimulus and a return to baseline, cells mount an altered, usually enhanced, response to an unrelated secondary stimulus [9]. The mechanism involves histone modifications at promoters and enhancers of inflammatory genes, together with shifts in cellular metabolism [8]. Because monocytes and neutrophils are short-lived, durable training requires a longer-lived reservoir, and this was located in hematopoietic stem and progenitor cells: β-glucan and BCG each reprogram progenitors toward enhanced myelopoiesis with lasting functional consequences [10, 11]. Progenitors are now regarded as integrating hubs through which inflammation in one tissue alters responses in another, adaptively or maladaptively [12].

For periodontitis, the central study showed that experimental disease induces this state and that it is transmissible by bone marrow transplantation, establishing that the memory resides at least partly in the marrow [16]. Training also biases progenitors toward osteoclastogenesis [17]. An exploratory human study examined monocyte reprogramming in the context of periodontitis [23]. A central reservoir is, by its nature, not site-specific: it would affect any implant in a patient with a periodontitis history, whether the site lost its tooth to periodontitis, caries, or trauma.

Memory in structural cells

Structural cells are long-lived and stay where the inflammation occurred, which makes them candidates for a site-specific reservoir. In skin, epithelial stem cells that have experienced inflammation maintain chromatin accessibility at key stress-response genes and close wounds faster on rechallenge; the effect persisted for months and depended on an inflammasome component [13]. The establishment, maintenance, and recall of this memory have been dissected at the level of transcription factors and chromatin [14]. In human airway disease, memory of type 2 cytokine exposure was retained by basal progenitors in culture [15].

In the oral mucosa, a single-cell atlas found epithelial and stromal populations with inflammatory signatures and linked exaggerated stromal responsiveness to neutrophil recruitment in periodontitis [24]. The direct demonstration that gingival fibroblasts can be trained, and that fibroblasts from diseased tissue resemble trained cells epigenetically, is recent and comes from in vitro work [18]. Epigenetic alterations in periodontitis have been reviewed repeatedly, with emphasis on methodological obstacles such as cell-type heterogeneity and small samples [25–28]. A review devoted to epigenetic memory in periodontal healing describes it as a plausible mechanism for disease persistence and variable healing and characterizes clinical applications as investigational [22].

What persists after extraction

Extraction removes the tooth, the periodontal ligament, and, in periodontitis, much of the pocket epithelium and granulation tissue. It does not remove the gingival connective tissue at the socket margin, the alveolar bone and its marrow, or the vasculature. The socket heals through a coagulum, granulation tissue, provisional matrix, and woven bone, followed by remodeling and substantial dimensional loss [29, 30]. In humans the process is highly variable between individuals [31], and a minority of sockets heal erratically, 4.24% of sites in one tomographic series [32].

Recent studies have begun to characterize sockets that heal in an inflammatory environment. Single-cell transcriptomics has been applied to granulation tissue from human inflammatory alveolar sockets [33]. In mice with periodontitis, socket healing is delayed, and macrophage polarization has been implicated [34, 35]. These studies examine healing in the presence of, or immediately after, active inflammation. They do not address what the tissue retains once healing is clinically complete, which is when a delayed implant is placed.

Three populations in the healed site could carry memory. Mucosal fibroblasts and keratinocytes that bordered the diseased pocket migrate over the socket and persist. Stromal and osteoprogenitor cells of the socket walls and adjacent marrow, which were exposed to the inflammatory milieu of the lesion, populate the new bone. Resident macrophages, including those associated with bone surfaces [36], may self-renew locally or be replaced from trained marrow progenitors. A fourth population, tissue-resident memory lymphocytes, belongs to adaptive immunity and is outside the trained-immunity concept, but would produce a similar site-specific effect and must be considered in the design.

Osseointegration as an immune-modulated process

Osseointegration has been reinterpreted as a foreign body equilibrium in which the immune system shapes both the establishment and the later breakdown of the bone–implant interface [37, 38]. Macrophages are positioned at the interface from the first days and influence the balance between bone formation and resorption [36]. Early healing passes through an inflammatory phase whose timely resolution precedes bone formation [39, 40]. If the cells that mount this early response are primed, the amplitude or duration of the inflammatory phase could change. Whether that would harm or help integration is not obvious, a point taken up again below.

Summary

Trained immunity is established as a phenomenon; its induction by experimental periodontitis in marrow progenitors has been shown in mice; training of gingival fibroblasts has been shown in vitro. The persistence of any such memory in a healed extraction site has not been examined, and neither has its effect on implant healing. The clinical association between periodontitis history and implant complications is robust and has several accepted explanations, none of which has been tested against a memory hypothesis.

3. Research Gap

The gap can be stated as four unanswered questions. Is there a detectable epigenetic or functional signature of prior periodontitis in the tissues of a clinically healed extraction site? If so, in which cell populations does it reside, and is it local, central, or both? How long does it last? And does it alter the early response to implant placement and, through that, the course of osseointegration?

Answering them requires designs that the existing clinical literature does not offer. Cohort studies compare patients with and without a periodontitis history; they cannot separate a property of the site from a property of the patient, or either from the microbial environment. A memory hypothesis becomes testable only when site history and patient history are varied independently and when the tissue is examined directly.

4. Research Objective and Central Hypothesis

The objective is to determine whether healed extraction sites retain an inflammatory memory of prior periodontitis and whether that memory conditions the response to implant placement.

The central hypothesis is that sites from which a tooth was extracted because of periodontitis retain, after clinical healing, a locally encoded hyperresponsive state in stromal and resident myeloid cells, detectable as increased chromatin accessibility and activating histone marks at inflammatory loci and as an amplified cytokine response to restimulation; that implant osteotomy recalls this state, producing a larger or longer early inflammatory response than at control sites; and that the effect diminishes as the interval between extraction and implant placement lengthens.

Two auxiliary hypotheses partition the effect. A central component, attributable to trained myelopoiesis, is present at all sites in an individual with a periodontitis history. A local component is present only at sites of prior disease. The two are separable by design.

The hypothesis would be refuted by the absence of any site-level difference in chromatin state or restimulation response when patient-level factors are held constant, or by a difference that does not relate to any healing outcome. A finding that memory is present but accelerates healing would refute the maladaptive form of the hypothesis while confirming that memory exists, and would be of equal interest.

5. Scientific and Theoretical Framework

Reservoirs, recall, and competing explanations

Table 1 sets out the candidate mechanisms and the observations that would distinguish them. The memory hypothesis is one of several explanations for the clinical association, and the design must allow each of the others to account for the data if it can.

Table 1. Candidate explanations for poorer implant outcomes after periodontitis and their distinguishing predictions.
ExplanationLevelPredicts a difference between diseased and non-diseased extraction sites in the same patient?Predicts a difference between patients at non-diseased sites?Predicts decay with healing time?
Local structural-cell memorySiteYesNoYes
Local resident immune memorySiteYesNoYes, possibly slower
Central trained myelopoiesisPatientNoYesYes, over months, if periodontitis is controlled
Persistent microbial reservoirPatient, modified by proximityWeakly, by proximity to remaining diseased teethYesNo, while reservoir persists
Shared host susceptibilityPatientNoYesNo
Reduced bone volume or qualitySiteYesNoNo

Two site-level explanations remain confounded in a simple within-patient comparison: local memory and local anatomy. They are separated by measuring anatomy and adjusting for it, and by the decay prediction: anatomical deficits do not recover with a longer healing interval, whereas memory should fade.

A minimal quantitative model

Let Y be the amplitude of the early inflammatory response to implant placement, measured for example as a composite of wound-fluid cytokines over the first two weeks. We propose

Y=Y_0 (1+m_C(t_C)) (1+m_L(t_L)) e^ε  (1)

where Y_0 is the response at a naive site in a naive host, m_C is central priming as a function of time t_C since control of periodontal inflammation, m_L is local priming as a function of time t_L since extraction, and ε is residual variation. Each priming term is assumed to decay:

m_j(t)=m_(j,0) e^(−t/τ_j),  j∈{C,L}  (2)

with initial magnitude m_(j,0) and time constant τ_j (weeks). Taking logarithms gives an additive model in which the two components are estimable from a two-by-two design, and in which the time constants are estimable from variation in healing interval. Exponential decay is the simplest assumption consistent with dilution of marks through cell division and turnover; the skin data suggest persistence over months [13], and the true form is unknown.

The multiplicative structure embodies a specific claim: central and local priming act on the same response and compound. An alternative is that they are redundant, each saturating the response. The interaction term in the two-by-two analysis tests this.

From early inflammation to osseointegration

The link from Y to integration is the least certain part of the framework. One line of reasoning holds that a larger or more prolonged inflammatory phase delays the transition to bone formation, promotes osteoclastogenesis at the interface, particularly if osteoclast precursors are themselves trained [17], and so reduces early bone-to-implant contact. A second line holds that memory evolved because it is useful: primed epithelium closes wounds faster [13], and a brisker early response might clear debris and bacteria more effectively. The framework does not assume the first. It treats the direction and size of the effect on bone as outcomes to be measured, with soft-tissue closure and bone formation assessed separately, because memory could plausibly help one and harm the other.

6. Methodology

Murine program

Model. Adult mice of a standard inbred strain. Periodontitis is induced by silk ligature around a maxillary molar by an optimized protocol [41, 42]. The contralateral molar is unligated. After a fixed period of disease, ligatures are removed and both molars are extracted. Sockets heal for a prespecified interval, after which titanium implants are placed in both healed sites, adapting published murine implant methods [43].

Experiment M1, existence and locality. Within-animal comparison of ligated and unligated sides addresses local memory with central factors held constant. A second group of never-ligated animals provides naive sites in naive hosts; comparison of unligated sides in ligated animals with naive animals addresses the central component. Together these form the two-by-two structure of equation (1).

Experiment M2, decay. Healing intervals of 4, 8, and 16 weeks between extraction and implant placement estimate τ_L and test the prediction that the local effect diminishes.

Experiment M3, central versus local reservoir. Bone marrow chimeras: irradiated naive recipients reconstituted with marrow from periodontitis donors, and periodontitis animals reconstituted with naive marrow after disease resolution, following the transplantation logic used to demonstrate central training [16]. If the site effect survives replacement of the marrow, the reservoir is radioresistant and local. If it transfers with marrow, it is central.

Experiment M4, mechanism. Blockade of interleukin-1 receptor signaling during the disease phase, which prevented central training in the earlier work [16], tests whether the same pathway is required for local memory.

Outcomes. At days 3 and 7 after implant placement: flow cytometry of peri-implant tissue for myeloid and lymphoid populations; cytokine protein and transcript levels. At 2 and 4 weeks: micro-computed tomography for peri-implant bone volume fraction; histomorphometry for bone-to-implant contact; osteoclast counts by tartrate-resistant acid phosphatase staining. In separate animals killed at the time implants would have been placed: sorted fibroblasts, endothelial cells, and myeloid cells from the healed site and marrow progenitors, profiled by assay for transposase-accessible chromatin [44] and by chromatin profiling for H3K4me3 and H3K27ac; and ex vivo restimulation of sorted or cultured cells with a toll-like receptor agonist, with cytokine output as the functional readout that defines training [9].

Controls. Sham ligature placement and immediate removal controls for mechanical trauma. Anatomical controls: socket bone volume and density at the time of implant placement, measured by micro-computed tomography and entered as covariates. Antibiotic-treated groups address a persistent local microbial difference between sides.

Design standards. Randomization of side, blinding of outcome assessment, prespecified primary outcome (bone-to-implant contact at 2 weeks for M1), and reporting to ARRIVE 2.0 [45]. Sample size is set from a pilot estimate of the within-animal standard deviation of the paired difference; none is assumed here.

Human cohort

Design. Prospective observational cohort with a nested within-patient comparison, reported to STROBE [46].

Groups. The two-by-two structure is realized as follows. Patients with a documented history of periodontitis, treated and in supportive care, who are to receive implants at healed sites: sites where the tooth was extracted for periodontitis (P+S+) and, in the same or other patients, sites where the tooth was extracted for caries, fracture, or endodontic failure with a healthy periodontium at that tooth (P+S−). Patients with no history of periodontitis receiving implants at healed sites (P−S−). The P−S+ cell is empty by definition, so the design estimates the central effect from P+S− versus P−S− and the local effect from P+S+ versus P+S−, with the within-patient subset giving the cleanest estimate of the latter.

Eligibility. Adults; delayed placement at least 12 weeks after extraction; no immediate implants; no augmentation at the study site before or at placement; non-smokers or smoking recorded and stratified; HbA1c recorded; no systemic anti-inflammatory or immunomodulatory therapy. The reason for extraction must be documented contemporaneously, with radiographs.

Tissue. At placement, the mucosal tissue removed by punch or flap trimming and the bone removed by the first drill or a narrow trephine are collected. These tissues are otherwise discarded, so collection adds no surgical step beyond careful retrieval. Peripheral blood is drawn for monocyte isolation.

Molecular endpoints. Chromatin accessibility and histone marks in sorted populations where yield allows, or single-nucleus multiome profiling to resolve cell types within small samples, which addresses the heterogeneity problem identified in the periodontal epigenetics literature [28]. Explant-derived fibroblasts and blood monocytes are rested and restimulated; cytokine output relative to unstimulated baseline is the functional measure of training.

Clinical endpoints. Wound-fluid or peri-implant crevicular fluid cytokines at days 3, 7, and 14; soft-tissue healing index; implant stability by resonance frequency analysis at placement and at 2, 4, 8, and 12 weeks [47]; marginal bone level at loading and at 12 months; early failure.

Covariates. Time since extraction; time since completion of active periodontal therapy; current periodontal status by the 2017 classification; full-mouth plaque and bleeding scores; distance from the site to the nearest tooth with residual pocketing; site bone volume and density from the planning scan; subgingival microbiome at adjacent teeth.

Sample size. The primary within-patient contrast is the paired difference in the restimulation response between P+S+ and P+S− sites. For a standardized paired difference of 0.5, which is an assumption chosen as the smallest effect likely to matter and not an estimate, 34 patients with both site types give 80% power at two-sided α of 0.05. Recruitment of such patients will be slow, and a multicenter design is anticipated. Between-patient contrasts require larger numbers and will be sized after variance estimates from the first 30 participants.

Analysis. Linear mixed models on log-transformed responses with patient as a random effect, implementing equation (1); time since extraction enters as a continuous modifier of the site effect to estimate decay. Mediation of any effect on stability trajectory or marginal bone through the early inflammatory response is examined with the caution appropriate to observational mediation analysis [48]. Multiple testing in the epigenomic analyses is controlled by false discovery rate, and a prespecified set of inflammatory loci identified in the murine experiments is tested first.

Bias and confounding

The reason for extraction is the exposure and is vulnerable to misclassification; it is taken from records made at the time, and cases of combined periodontal–endodontic pathology are excluded. Sites lost to periodontitis are likely to have less bone, so anatomical covariates are essential and a sensitivity analysis is restricted to sites matched on bone volume. Proximity to residual pockets could produce a site-level microbial difference that mimics local memory; the distance covariate and adjacent-tooth microbiome address this. Operators are not blinded to site history, but laboratory personnel and outcome assessors are.

Ethics

Animal work requires institutional animal care and use committee approval. The human study requires institutional review board approval and written consent, including consent for genomic and epigenomic analysis and data sharing under controlled access. Collection is limited to tissue that would be discarded and to a venous blood sample.

7. Analytical Development

Identification of local and central components

Taking logarithms of equation (1) and indexing patient history by P∈{0,1} and site history by S∈{0,1},

logY=β_0+β_CP+β_LS+β_(CL) PS+ε  (3)

with β_C=log(1+m_C) and β_L=log(1+m_L). Because S=1 occurs only when P=1, β_L and β_(CL) are not separately identified in humans; the within-patient contrast estimates their sum, which is the local effect in a centrally primed host. In mice, where local disease can be induced unilaterally, the ligated side and the unligated side of the same animal share P=1, and naive animals supply P=0; the chimera experiment supplies the missing cell by placing a diseased site in a host with naive marrow. Only the animal program can therefore test whether local memory operates without central priming.

Estimating decay

Substituting equation (2), the local log-effect at healing interval t_L is log​(1+m_(L,0) e^(−t_L/τ_L)), which for small m_(L,0) is approximately m_(L,0) e^(−t_L/τ_L). Three healing intervals in mice allow a two-parameter fit with one degree of freedom for lack of fit. In the human cohort, healing interval is not assigned, and it will be associated with the reason for delay; the decay estimate there is exploratory. A finding of no decay over the intervals studied would not distinguish very durable memory from an anatomical effect, which is why the anatomical covariates and the molecular signature are both required: memory predicts that the chromatin signature and the functional effect fade together.

What counts as a memory signature

Three criteria, taken together, would constitute evidence of local memory. Differential accessibility or histone marking at inflammatory loci between P+S+ and P+S− sites in the same cell type, in the absence of differential expression at baseline, which is the defining pattern of poised memory in epithelial stem cells [14]. An enhanced response to restimulation in cells from P+S+ sites after a rest period in culture long enough to exclude carry-over of active inflammation. And persistence of both through at least two passages or a defined number of population doublings. Any one alone is insufficient: accessibility differences can reflect cell-composition differences, and enhanced responses can reflect residual activation.

Predicted patterns

If local structural memory exists and is maladaptive, the ligated side in M1 shows higher early cytokine levels, more osteoclasts, and lower bone-to-implant contact at 2 weeks than the unligated side, with the difference shrinking across M2 intervals and persisting in chimeras with naive marrow. If only central training matters, the two sides of ligated animals are equal and both differ from naive animals, and the effect transfers with marrow. If memory is adaptive for soft tissue, mucosal closure is faster on the ligated side regardless of the bone outcome. If none of these patterns appears while anatomical covariates explain the side difference, the memory hypothesis is not supported.

8. Discussion

The hypothesis offers a mechanism for a well-documented clinical association, and it is stated in a form that existing data cannot evaluate. That is its main value and its main weakness. The association between periodontitis history and implant complications is fully consistent with the conventional explanations, and nothing reviewed here shows that memory contributes. The case for investigating it rests on plausibility from adjacent fields and on the availability of a decisive experiment.

Several considerations argue against a large effect. The cells most directly exposed to the periodontal lesion, those of the pocket wall and periodontal ligament, are removed or lost at extraction. Socket healing replaces much of the local tissue through proliferation, and epigenetic marks can be diluted by division. The interval between extraction and delayed implant placement is typically months. And in the best-studied structural-cell system, memory improved the response to injury [13]. A defensible prior is that any local effect is small, transient, and possibly beneficial for soft tissue.

Other considerations argue for taking it seriously. Fibroblasts from periodontitis lesions retained a trained phenotype in culture [18], which implies stability through at least several divisions. The stromal compartment of the oral mucosa is an active participant in neutrophil recruitment [24]. Trained osteoclast precursors are a demonstrated route from innate memory to bone loss [17]. And the central component, at least, has direct experimental support [16]; even if local memory proves negligible, a central effect on peri-implant healing would be a new and clinically relevant finding, and the proposed design detects it.

The distinction between local and central memory has practical weight. A central effect would be modifiable, if at all, by systemic control of periodontal inflammation and by time, and would apply to every implant in the patient. A local effect would bear on the timing of implant placement at specific sites and on local pretreatment. The human design cannot fully separate the two, as the identification analysis shows, and conclusions about locality will lean on the animal experiments.

Adaptive immune memory deserves more attention than this framework gives it. Tissue-resident memory T cells persist in barrier tissues after inflammation and could confer a site-specific recall response indistinguishable, at the level of clinical outcome, from structural-cell memory. The flow cytometric and single-cell endpoints will detect them, and lymphocyte-deficient mice or depletion strategies could be added if the initial experiments show a local effect.

A methodological caution concerns the murine implant model. Mouse maxillary implants are technically demanding, failure rates are not negligible even in control groups [43], and the scale of the socket limits tissue yield. A larger species with naturally occurring or ligature-induced periodontitis would be closer to the clinical situation but would sacrifice the genetic and chimera tools on which the mechanistic experiments depend.

9. Clinical Significance

If a local, decaying memory were demonstrated and shown to impair integration, it would provide a biological rationale for the interval between extraction and implant placement at periodontitis sites that goes beyond bone fill, and it would motivate studies of whether the interval can be shortened by intervention. If a central effect were demonstrated, the emphasis would shift to the completeness and duration of periodontal disease control before implant therapy, already a clinical recommendation on other grounds [7, 49], and to the possibility that systemic inflammatory status at placement is a modifiable risk factor. If memory proved beneficial for soft-tissue healing, that too would be informative for flap management and timing.

A molecular signature of memory in tissue discarded at osteotomy could in principle become a site-level risk marker. That prospect is remote. It would require the signature to be reproducible, to predict outcomes independently of clinical risk indicators, and to be measurable at reasonable cost, and none of these has been examined.

No change in clinical practice is warranted by this manuscript.

10. Limitations

The hypothesis is extrapolated from skin, airway, marrow, and in vitro gingival fibroblast data; none of the supporting studies examined a healed extraction site. The quantitative model is a convenience for design and analysis, and its exponential decay and multiplicative structure are assumptions. In humans the exposure (reason for extraction) is not assigned and is confounded with site anatomy and microbial proximity; adjustment may be incomplete. The empty P−S+ cell prevents full separation of local and central effects in patients. Tissue obtained at osteotomy is small and heterogeneous, and epigenomic profiling of sorted populations may not be feasible from single sites. Ligature-induced periodontitis in mice is acute and may not reproduce the chronicity that could be necessary for durable local imprinting. The healing intervals feasible in mice may be too short relative to the time constants relevant in humans. The link from early inflammatory amplitude to long-term implant outcome is assumed for design purposes and is itself unproven. The supporting literature was assessed from abstracts and index records.

11. Future Validation

The murine experiments M1 and M2 are the logical first step, because a clear negative result would spare the considerable effort of the human cohort. If a local effect is found, replication in a second laboratory and a second periodontitis model is needed before mechanistic experiments proceed. In humans, a feasibility phase should establish tissue yield, the reliability of the restimulation assay across operators, and the recruitment rate of patients with both site types. If the cohort supports the hypothesis, the next study would test an intervention, the simplest being randomized variation of the healing interval within clinically accepted limits, with the memory signature and early healing as outcomes. Independent confirmation of any epigenomic signature in a separate cohort is a precondition for considering it a marker.

12. Conclusion

Whether a healed extraction site remembers the disease that cost the tooth has not been examined. Innate immune training by periodontitis is established in marrow progenitors of mice, trained phenotypes have been described in gingival fibroblasts, and structural-cell memory is well documented in other barrier tissues. These findings make it plausible that prior periodontitis leaves a local or central imprint that is recalled when an implant is placed. They do not show that it does, and the effect, if present, could favor healing as readily as hinder it. The experiments proposed here separate memory from the conventional explanations and local memory from central. Until they are done, inflammatory memory of the extraction site is a hypothesis.

Cite this paper

Jain S, Gupta A. Inflammatory Memory of the Healed Extraction Site: A Mechanistic Framework and Translational Protocol for Testing Whether Prior Periodontitis Conditions the Biological Response to Implant Placement. Working paper. Center for Implant Dentistry; 2026. Available from: https://bayareaimplantdentistry.com/clinician-research-and-publications/extraction-socket-inflammatory-memory

References

  1. Serroni M, Borgnakke WS, Romano L, Balice G, Paolantonio M, Saleh MHA, et al. History of periodontitis as a risk factor for implant failure and incidence of peri-implantitis: A systematic review, meta-analysis, and trial sequential analysis of prospective cohort studies. Clin Implant Dent Relat Res. 2024;26(3):482-508. doi:10.1111/cid.13330 (opens in a new tab). PMID: 38720611 (opens in a new tab).
  2. Karoussis IK, Salvi GE, Heitz-Mayfield LJ, Brägger U, Hämmerle CH, Lang NP. Long-term implant prognosis in patients with and without a history of chronic periodontitis: a 10-year prospective cohort study of the ITI Dental Implant System. Clin Oral Implants Res. 2003;14(3):329-39. doi:10.1034/j.1600-0501.000.00934.x (opens in a new tab). PMID: 12755783 (opens in a new tab).
  3. Roccuzzo M, De Angelis N, Bonino L, Aglietta M. Ten-year results of a three-arm prospective cohort study on implants in periodontally compromised patients. Part 1: implant loss and radiographic bone loss. Clin Oral Implants Res. 2010;21(5):490-6. doi:10.1111/j.1600-0501.2009.01886.x (opens in a new tab). PMID: 20337668 (opens in a new tab).
  4. Sgolastra F, Petrucci A, Severino M, Gatto R, Monaco A. Periodontitis, implant loss and peri-implantitis. A meta-analysis. Clin Oral Implants Res. 2015;26(4):e8-e16. doi:10.1111/clr.12319 (opens in a new tab). PMID: 24382358 (opens in a new tab).
  5. Ramanauskaite A, Baseviciene N, Wang HL, Tözüm TF. Effect of history of periodontitis on implant success: meta-analysis and systematic review. Implant Dent. 2014;23(6):687-96. doi:10.1097/ID.0000000000000156 (opens in a new tab). PMID: 25343317 (opens in a new tab).
  6. Heitz-Mayfield LJ. Peri-implant diseases: diagnosis and risk indicators. J Clin Periodontol. 2008;35(8 Suppl):292-304. doi:10.1111/j.1600-051X.2008.01275.x (opens in a new tab). PMID: 18724857 (opens in a new tab).
  7. Schwarz F, Derks J, Monje A, Wang HL. Peri-implantitis. J Periodontol. 2018;89 Suppl 1:S267-S290. doi:10.1002/JPER.16-0350 (opens in a new tab). PMID: 29926957 (opens in a new tab).
  8. Netea MG, Joosten LA, Latz E, Mills KH, Natoli G, Stunnenberg HG, et al. Trained immunity: A program of innate immune memory in health and disease. Science. 2016;352(6284):aaf1098. doi:10.1126/science.aaf1098 (opens in a new tab). PMID: 27102489 (opens in a new tab).
  9. Netea MG, Domínguez-Andrés J, Barreiro LB, Chavakis T, Divangahi M, Fuchs E, et al. Defining trained immunity and its role in health and disease. Nat Rev Immunol. 2020;20(6):375-388. doi:10.1038/s41577-020-0285-6 (opens in a new tab). PMID: 32132681 (opens in a new tab).
  10. Mitroulis I, Ruppova K, Wang B, Chen LS, Grzybek M, Grinenko T, et al. Modulation of Myelopoiesis Progenitors Is an Integral Component of Trained Immunity. Cell. 2018;172(1-2):147-161.e12. doi:10.1016/j.cell.2017.11.034 (opens in a new tab). PMID: 29328910 (opens in a new tab).
  11. Kaufmann E, Sanz J, Dunn JL, Khan N, Mendonça LE, Pacis A, et al. BCG Educates Hematopoietic Stem Cells to Generate Protective Innate Immunity against Tuberculosis. Cell. 2018;172(1-2):176-190.e19. doi:10.1016/j.cell.2017.12.031 (opens in a new tab). PMID: 29328912 (opens in a new tab).
  12. Chavakis T, Mitroulis I, Hajishengallis G. Hematopoietic progenitor cells as integrative hubs for adaptation to and fine-tuning of inflammation. Nat Immunol. 2019;20(7):802-811. doi:10.1038/s41590-019-0402-5 (opens in a new tab). PMID: 31213716 (opens in a new tab).
  13. Naik S, Larsen SB, Gomez NC, Alaverdyan K, Sendoel A, Yuan S, et al. Inflammatory memory sensitizes skin epithelial stem cells to tissue damage. Nature. 2017;550(7677):475-480. doi:10.1038/nature24271 (opens in a new tab). PMID: 29045388 (opens in a new tab).
  14. Larsen SB, Cowley CJ, Sajjath SM, Barrows D, Yang Y, Carroll TS, et al. Establishment, maintenance, and recall of inflammatory memory. Cell Stem Cell. 2021;28(10):1758-1774.e8. doi:10.1016/j.stem.2021.07.001 (opens in a new tab). PMID: 34320411 (opens in a new tab).
  15. Ordovas-Montanes J, Dwyer DF, Nyquist SK, Buchheit KM, Vukovic M, Deb C, et al. Allergic inflammatory memory in human respiratory epithelial progenitor cells. Nature. 2018;560(7720):649-654. doi:10.1038/s41586-018-0449-8 (opens in a new tab). PMID: 30135581 (opens in a new tab).
  16. Li X, Wang H, Yu X, Saha G, Kalafati L, Ioannidis C, et al. Maladaptive innate immune training of myelopoiesis links inflammatory comorbidities. Cell. 2022;185(10):1709-1727.e18. doi:10.1016/j.cell.2022.03.043 (opens in a new tab). PMID: 35483374 (opens in a new tab).
  17. Haacke N, Wang H, Yan S, Barovic M, Li X, Nagai K, et al. Innate immune training of osteoclastogenesis promotes inflammatory bone loss in mice. Dev Cell. 2025;60(13):1854-1870.e6. doi:10.1016/j.devcel.2025.02.001 (opens in a new tab). PMID: 40020679 (opens in a new tab).
  18. Liu J, Tian H, Ju J, Nie F, Yin Q, Zhao J, et al. Porphyromonas gingivalis-Lipopolysaccharide Induced Gingival Fibroblasts Trained Immunity Sustains Inflammation in Periodontitis. J Periodontal Res. 2025;60(11):1156-1167. doi:10.1111/jre.13372 (opens in a new tab). PMID: 39665166 (opens in a new tab).
  19. Hajishengallis G, Li X, Divaris K, Chavakis T. Maladaptive trained immunity and clonal hematopoiesis as potential mechanistic links between periodontitis and inflammatory comorbidities. Periodontol 2000. 2022;89(1):215-230. doi:10.1111/prd.12421 (opens in a new tab). PMID: 35244943 (opens in a new tab).
  20. Hajishengallis G, Chavakis T. Inflammatory memory and comorbidities. Clin Transl Med. 2022;12(7):e984. doi:10.1002/ctm2.984 (opens in a new tab). PMID: 35839318 (opens in a new tab).
  21. Hajishengallis G, Netea MG, Chavakis T. Trained immunity in chronic inflammatory diseases and cancer. Nat Rev Immunol. 2025;25(7):497-514. doi:10.1038/s41577-025-01132-x (opens in a new tab). PMID: 39891000 (opens in a new tab).
  22. Bıçakçıoğlu HA, Çolak G. Epigenetic memory in periodontal healing: mechanisms, evidence, and emerging therapeutic perspectives. Odontology. 2026. doi:10.1007/s10266-026-01417-0 (opens in a new tab). PMID: 42166030 (opens in a new tab).
  23. Noz MP, Plachokova AS, Smeets EMM, Aarntzen EHJG, Bekkering S, Vart P, et al. An Explorative Study on Monocyte Reprogramming in the Context of Periodontitis In Vitro and In Vivo. Front Immunol. 2021;12:695227. doi:10.3389/fimmu.2021.695227 (opens in a new tab). PMID: 34484192 (opens in a new tab).
  24. Williams DW, Greenwell-Wild T, Brenchley L, Dutzan N, Overmiller A, Sawaya AP, et al. Human oral mucosa cell atlas reveals a stromal-neutrophil axis regulating tissue immunity. Cell. 2021;184(15):4090-4104.e15. doi:10.1016/j.cell.2021.05.013 (opens in a new tab). PMID: 34129837 (opens in a new tab).
  25. Barros SP, Offenbacher S. Modifiable risk factors in periodontal disease: epigenetic regulation of gene expression in the inflammatory response. Periodontol 2000. 2014;64(1):95-110. doi:10.1111/prd.12000 (opens in a new tab). PMID: 24320958 (opens in a new tab).
  26. Larsson L, Castilho RM, Giannobile WV. Epigenetics and its role in periodontal diseases: a state-of-the-art review. J Periodontol. 2015;86(4):556-68. doi:10.1902/jop.2014.140559 (opens in a new tab). PMID: 25415244 (opens in a new tab).
  27. Larsson L, Kavanagh NM, Nguyen TVN, Castilho RM, Berglundh T, Giannobile WV. Influence of epigenetics on periodontitis and peri-implantitis pathogenesis. Periodontol 2000. 2022;90(1):125-137. doi:10.1111/prd.12453 (opens in a new tab). PMID: 35913702 (opens in a new tab).
  28. Dommisch H, Schäfer AS, Kuzmanova D, Michaud DS. Epigenetic Regulation in the Pathogenesis of Periodontitis. J Periodontal Res. 2025. doi:10.1111/jre.70036 (opens in a new tab). PMID: 40951997 (opens in a new tab).
  29. Cardaropoli G, Araújo M, Lindhe J. Dynamics of bone tissue formation in tooth extraction sites. An experimental study in dogs. J Clin Periodontol. 2003;30(9):809-18. doi:10.1034/j.1600-051x.2003.00366.x (opens in a new tab). PMID: 12956657 (opens in a new tab).
  30. Araújo MG, Lindhe J. Dimensional ridge alterations following tooth extraction. An experimental study in the dog. J Clin Periodontol. 2005;32(2):212-8. doi:10.1111/j.1600-051X.2005.00642.x (opens in a new tab). PMID: 15691354 (opens in a new tab).
  31. Trombelli L, Farina R, Marzola A, Bozzi L, Liljenberg B, Lindhe J. Modeling and remodeling of human extraction sockets. J Clin Periodontol. 2008;35(7):630-9. doi:10.1111/j.1600-051X.2008.01246.x (opens in a new tab). PMID: 18498382 (opens in a new tab).
  32. Kim JH, Susin C, Min JH, Suh HY, Sang EJ, Ku Y, et al. Extraction sockets: erratic healing impeding factors. J Clin Periodontol. 2014;41(1):80-5. doi:10.1111/jcpe.12173 (opens in a new tab). PMID: 24117498 (opens in a new tab).
  33. Li Y, Xue Y, Zhang Y, Hu KJ. [Deciphering the fate of granulation tissue in the human inflammatory alveolar microenvironment using single-cell RNA sequencing]. Zhonghua Kou Qiang Yi Xue Za Zhi. 2024;59(8):804-809. doi:10.3760/cma.j.cn112144-20240430-00176 (opens in a new tab). PMID: 39036912 (opens in a new tab).
  34. Zhang N, Ji S, He Z, Wu D. Lpar3-mediated macrophage polarization promotes inflammatory tooth extraction socket healing and alveolar bone regeneration. Regen Ther. 2026;32:101116. doi:10.1016/j.reth.2026.101116 (opens in a new tab). PMID: 42059006 (opens in a new tab).
  35. Bae JE, Hwang SM, Aryal YP, Kim TY, Sohn WJ, An SY, et al. Effects of erythropoietin on osteoblast in the tooth extraction socket in mice periodontitis model. Front Physiol. 2022;13:987625. doi:10.3389/fphys.2022.987625 (opens in a new tab). PMID: 36277197 (opens in a new tab).
  36. Miron RJ, Bosshardt DD. OsteoMacs: Key players around bone biomaterials. Biomaterials. 2016;82:1-19. doi:10.1016/j.biomaterials.2015.12.017 (opens in a new tab). PMID: 26735169 (opens in a new tab).
  37. Trindade R, Albrektsson T, Tengvall P, Wennerberg A. Foreign Body Reaction to Biomaterials: On Mechanisms for Buildup and Breakdown of Osseointegration. Clin Implant Dent Relat Res. 2016;18(1):192-203. doi:10.1111/cid.12274 (opens in a new tab). PMID: 25257971 (opens in a new tab).
  38. Albrektsson T, Dahlin C, Jemt T, Sennerby L, Turri A, Wennerberg A. Is marginal bone loss around oral implants the result of a provoked foreign body reaction?. Clin Implant Dent Relat Res. 2014;16(2):155-65. doi:10.1111/cid.12142 (opens in a new tab). PMID: 24004092 (opens in a new tab).
  39. Terheyden H, Lang NP, Bierbaum S, Stadlinger B. Osseointegration–communication of cells. Clin Oral Implants Res. 2012;23(10):1127-35. doi:10.1111/j.1600-0501.2011.02327.x (opens in a new tab). PMID: 22092345 (opens in a new tab).
  40. Davies JE. Understanding peri-implant endosseous healing. J Dent Educ. 2003;67(8):932-49. PMID: 12959168 (opens in a new tab).
  41. Abe T, Hajishengallis G. Optimization of the ligature-induced periodontitis model in mice. J Immunol Methods. 2013;394(1-2):49-54. doi:10.1016/j.jim.2013.05.002 (opens in a new tab). PMID: 23672778 (opens in a new tab).
  42. Graves DT, Fine D, Teng YT, Van Dyke TE, Hajishengallis G. The use of rodent models to investigate host-bacteria interactions related to periodontal diseases. J Clin Periodontol. 2008;35(2):89-105. doi:10.1111/j.1600-051X.2007.01172.x (opens in a new tab). PMID: 18199146 (opens in a new tab).
  43. Pirih FQ, Hiyari S, Barroso AD, Jorge AC, Perussolo J, Atti E, et al. Ligature-induced peri-implantitis in mice. J Periodontal Res. 2015;50(4):519-24. doi:10.1111/jre.12234 (opens in a new tab). PMID: 25244403 (opens in a new tab).
  44. Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods. 2013;10(12):1213-8. doi:10.1038/nmeth.2688 (opens in a new tab). PMID: 24097267 (opens in a new tab).
  45. Percie du Sert N, Hurst V, Ahluwalia A, Alam S, Avey MT, Baker M, et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol. 2020;18(7):e3000410. doi:10.1371/journal.pbio.3000410 (opens in a new tab). PMID: 32663219 (opens in a new tab).
  46. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-7. doi:10.1016/S0140-6736(07)61602-X (opens in a new tab). PMID: 18064739 (opens in a new tab).
  47. Sennerby L, Meredith N. Implant stability measurements using resonance frequency analysis: biological and biomechanical aspects and clinical implications. Periodontol 2000. 2008;47:51-66. doi:10.1111/j.1600-0757.2008.00267.x (opens in a new tab). PMID: 18412573 (opens in a new tab).
  48. MacKinnon DP, Fairchild AJ, Fritz MS. Mediation analysis. Annu Rev Psychol. 2007;58:593-614. doi:10.1146/annurev.psych.58.110405.085542 (opens in a new tab). PMID: 16968208 (opens in a new tab).
  49. Herrera D, Berglundh T, Schwarz F, Chapple I, Jepsen S, Sculean A, et al. Prevention and treatment of peri-implant diseases-The EFP S3 level clinical practice guideline. J Clin Periodontol. 2023;50 Suppl 26:4-76. doi:10.1111/jcpe.13823 (opens in a new tab). PMID: 37271498 (opens in a new tab).