Why Won’t My Infection Clear? Understanding Biofilm in Chronic Lyme and Mold Illness

There is a pattern we see so often it has become a diagnostic clue in its own right.

A patient starts treatment — antibiotics, herbs, antifungals, it hardly matters which. Within a few weeks they feel meaningfully better. Then, somewhere between week four and week ten, the improvement stalls. They hold for a while. Then they slide. By the time the next course starts they are close to where they began, and the next round of improvement is a little shallower and a little shorter than the last.

“Every time I treat it, it comes back. What am I missing?”

Something in that patient is surviving treatment that susceptibility testing says should have died. And one of the best-characterised reasons for that discrepancy in all of microbiology is biofilm.

Here is where this article is going to be different from most of what you will read on the subject. Biofilm is real, rigorously studied, and genuinely explains a great deal about chronic infection. It is also the single most abused concept in the functional medicine supplement market, where an enormous amount of laboratory data about dishes and catheters has been quietly relabelled as evidence about people. Both of those things are true at once, and a patient deciding how to spend their money and their energy deserves to know which parts are which.

So this piece does three things. It explains what a biofilm is and why it defeats both antibiotics and your immune system — that part is solid science. It sorts through the agents we and others use against it, and tells you honestly what each one has behind it. And it explains how we actually sequence biofilm work at Tree of Light Health, including the situations in which we deliberately do not use it.

What a Biofilm Actually Is

Bacteria have two fundamentally different ways of living. The one every microbiology course teaches is the planktonic state: free-floating, individual, dividing rapidly, and — importantly — the state in which laboratories test antibiotic susceptibility. The other is the biofilm state: organisms living as a dense community, held together and to their surroundings by a self-manufactured matrix.

That matrix is the key to everything that follows. It is called the extracellular polymeric substance, or EPS, and it is not simply slime. It is a structured material built from polysaccharides, extracellular DNA, proteins and lipids, and it typically accounts for the large majority of the biofilm’s dry mass — often cited as around 90% — with the bacterial cells themselves a minority component.^1,2 One detail worth knowing, because it determines which agents could plausibly work: extracellular proteins and DNA are not minor constituents. In many organisms proteins exceed polysaccharides, and extracellular DNA functions as an electrostatic net holding the whole structure together.^1,3

Inside that matrix, the community behaves differently from the same organisms free-floating. Gene expression changes. Growth slows dramatically. Gradients form. It is, functionally, a different organism with the same genome.

Two corrections worth making up front

The 80% figure you have read is not what you think it is. Almost every article on this subject opens with a claim that biofilms are responsible for 65% or 80% of human bacterial infections, usually attributed to the NIH. We traced it. It appears in NIH grant programme announcements going back to 1998 — and in every instance it appears with no citation attached to it.⁵ It is a decades-old, unsourced funding-announcement estimate that has been laundered into the scientific literature by repetition, and it is still being repeated uncritically in 2026 reviews. We are not going to use it. What is well quantified is prevalence in specific infection types: biofilm is found in roughly 68–78% of chronic wounds across meta-analysed studies,⁶ and in 68.8% of colorectal tumours in one 2025 series.⁷ Those are real numbers with real methods behind them.

The picture in your head is probably wrong. The textbook diagram — bacteria attach to a surface, build mushroom-shaped towers, then disperse — comes from one organism (Pseudomonas aeruginosa) grown in one apparatus (a laboratory flow cell). The field has now largely moved on. In actual human infections, biofilms are usually small, free-floating aggregates rather than surface-attached films, and attachment to a surface is not required for bacteria to behave as a biofilm at all.^8,9

This matters clinically and it matters for how you read marketing material. An illustration showing a slimy layer coating your intestinal wall or lining your arteries is showing you a laboratory image, not what a biofilm in a human body typically looks like. The real thing is more like scattered clusters of bacteria in a protective gel, distributed through tissue.

Biofilm is not a coating you scrape off. It is a way of living that bacteria switch into.

Why Biofilms Defeat Antibiotics

Before anything else, a distinction that almost every article on this subject gets wrong, and that changes how you should think about the whole problem.

So why does a susceptible organism survive a drug it is susceptible to? Several reasons, which are usually presented as equally important and are not.

The biggest one: oxygen and nutrient gradients

This is the mechanism to understand, because it explains most of the effect.

A biofilm consumes oxygen and nutrients from the outside in. The cells at the surface get plenty; the cells deeper in get progressively less. In a landmark study of Pseudomonas biofilms averaging 210 micrometres thick, oxygen penetrated only about 50 micrometres — leaving roughly three quarters of the biofilm anoxic. Active protein synthesis was confined to a band about 30 micrometres deep at the surface. The rest of the community was alive but metabolically idling, growing at roughly one-fiftieth the rate of the same organism in free culture.¹¹

Now recall how most antibiotics kill. Beta-lactams interfere with building a cell wall. Aminoglycosides and macrolides interfere with making protein. Fluoroquinolones interfere with replicating DNA. Every one of these requires the bacterium to be doing something. A cell that has stopped building, synthesising and dividing is not resistant to the drug — it is simply not engaged in the process the drug interrupts.

In that same study, oxygen limitation alone accounted for at least 70% of the biofilm’s protection, for every antibiotic tested. That is the core of it.

The one everybody cites, which matters less than you think

The popular explanation is that the matrix physically blocks antibiotics from getting in. This is partly true and largely beside the point.

Measured penetration times for many antibiotics are remarkably short — vancomycin in about half a minute, daptomycin in a minute and a half.¹² Tetracycline has been shown to reach every cell in an E. coli biofilm within ten minutes with no loss of viability whatsoever. Penetration is genuinely reduced for some drug classes, particularly beta-lactams and vancomycin against staphylococcal biofilm, while amikacin and ciprofloxacin pass through unimpeded.¹³ And a 2025 methodological paper showed that older measurements systematically overestimated penetration, so the effect may be larger than the field thought.¹⁴

But the decisive observation is this: even antibiotics that fully penetrate a biofilm still fail to kill the cells inside it. Diffusion cannot be the main explanation for something that persists after diffusion is complete.

This matters practically, because the entire consumer logic of biofilm supplements — “break the wall so the antimicrobial can get in” — rests on the mechanism that the evidence supports least.

The rest of the picture

• Persister cells. A small subpopulation of deeply dormant, phenotypically tolerant cells that survive antibiotic exposure and regrow afterwards. Real, and important in Lyme specifically — but note that even careful reviewers describe the link between biofilm dormancy and classical persisters as probable rather than established.^12,15
• Efflux and stress responses. Certain efflux systems are upregulated under the low-oxygen conditions inside biofilm, and extracellular DNA binds magnesium, triggering signalling cascades that reduce aminoglycoside susceptibility.¹⁰
• Horizontal gene transfer. Genuine resistance genes move between organisms more readily inside a biofilm than outside it.¹⁶ Worth knowing — but this is a public health mechanism, not the reason your biofilm survived your antibiotic course. Popular articles blur the two constantly.

How much harder is it, in numbers?

The honest answer is: it varies enormously, and anyone giving you a single figure is simplifying. In the foundational dataset, eradicating biofilm bacteria required concentrations 100- to more than 1,000-fold higher than killing the same organisms free-floating — but with striking exceptions, notably tobramycin and amikacin against Pseudomonas, which stayed effective.¹⁷ Across the wider literature, tolerance factors range from 1.0 — meaning no difference at all — to over 2,000, depending on the organism and drug.¹²

The defensible statement is that eradicating biofilm bacteria can require 10- to 1,000-fold higher drug concentrations than killing the same organism free-floating, and for some drug-organism pairs there is no difference whatsoever. Which is another way of saying that standard susceptibility testing — performed on free-floating organisms — can be a poor guide to what will happen in a patient with a biofilm infection.

Why Biofilms Defeat the Immune System

If antibiotic tolerance were the whole story, biofilm would be a treatment problem. It is worse than that. Biofilm is also an immune problem, and this is where it starts to explain how a patient can feel chronically inflamed and chronically infected at the same time, for years.

Frustrated phagocytosis

Staphylococcus aureus biofilms actively prevent macrophage phagocytosis and blunt the inflammatory response they provoke.¹⁸ Neutrophils arriving at a Pseudomonas biofilm settle onto its surface, become immobilised, and degranulate without ingesting anything — releasing their full destructive payload into the surrounding tissue while killing nothing.¹⁹ The phrase for this in the literature is “frustrated phagocytosis,” and it is an unusually apt description of what a chronic infection patient experiences: maximum inflammatory cost, minimum microbial benefit.

Complement evasion

Biofilm-grown Staphylococcus epidermidis triggers more complement activation than free-floating cells while showing less complement and antibody deposited on its surface, and survives neutrophil killing better — even when pre-opsonised.²⁰ Streptococcus pneumoniae in biofilm shows impaired C3b deposition, reduced binding of C-reactive protein and C1q, and increased recruitment of factor H, which switches off the alternative pathway.²¹ The biofilm does not hide from complement. It absorbs the activation and deflects the consequences.

The matrix disarms the immune system’s own weapons

A 2026 study provides the cleanest example of this yet. Neutrophils release DNA nets studded with antimicrobial histones. Staphylococcus aureus biofilm extracellular DNA electrostatically sequesters histone H3 and neutralises it — and adding salt to disrupt the charge interaction abolishes the effect, confirming the mechanism.²² The bacteria’s own extracellular DNA is being used to disarm the host’s antimicrobial proteins.

Quorum Sensing, Dispersal, and a Risk Nobody Mentions

Bacteria in a biofilm coordinate. They release small signalling molecules, sense their concentration, and when the population density crosses a threshold they switch gene expression collectively — turning on matrix production, virulence factors, or the machinery of dispersal.²⁴ This is quorum sensing, and it is one of the genuinely elegant findings in modern microbiology.

It is also a word that appears on a great many supplement labels. So: quorum sensing inhibitors are a real and active research field, and no quorum-sensing inhibitor is an approved human therapeutic. Products invoking the concept are borrowing the vocabulary, not the evidence.

Dispersal is an active decision, not an accident

Biofilms do not simply erode. When conditions call for it — nutrient depletion, nitric oxide signalling, falling levels of the second messenger cyclic-di-GMP — the community actively releases cells back into circulation to colonise elsewhere.²⁵

And here is the finding that changes how a careful clinician approaches this whole subject. Dispersed cells are not simply planktonic cells again. They are a distinct third phenotype, with an altered transcriptome, altered physiology, and increased virulence compared with both biofilm and ordinary free-floating bacteria.^25,26

What this does and does not mean for you

Let us be careful, because this could easily be over-read in either direction.

These are animal and laboratory findings, in acute wound and device models, with organisms and agents that are not what an integrative practitioner is typically using. Nobody has demonstrated that a patient taking an oral enzyme capsule develops bacteraemia, and we are not claiming that.

But the principle is sound and it is well established: releasing organisms from a biofilm is not an inherently safe act, and the released organisms may be more dangerous than the ones you started with. It provides the first real mechanistic explanation for something clinicians in this field observe constantly — that biofilm work can make patients acutely and genuinely worse, in a way that is not simply “detox.”

And it yields a clear clinical rule, which we follow without exception: never disperse without cover. Biofilm agents belong alongside an active antimicrobial strategy and open drainage, not on their own and not before the patient is prepared. Dispersal without something waiting to meet the released organisms is the one configuration the evidence actively warns against.

Opening a biofilm is only useful if something is waiting on the other side.

Biofilm in Lyme Disease — The Contested Part

Everything above is mainstream microbiology, uncontroversial and well replicated. This section is different, and we are going to mark the boundary clearly, because you will not find that done often.

What has actually been shown

In 2012, Eva Sapi’s group at the University of New Haven demonstrated that Borrelia burgdorferi in culture forms dense aggregates on a wide range of surfaces, and that those aggregates — but not individual spirochetes — carry several hallmarks of biofilm: a protective mucopolysaccharide layer, abundant extracellular DNA, calcium deposits, and complex, continuously rearranging architecture on atomic force microscopy.²⁹

In 2016, the same group examined six archived human skin biopsies of borrelial lymphocytoma, some dating to the 1970s, and reported aggregates in the tissue displaying the same biofilm markers, with internal channels and surface protrusions visible by microscopy.³⁰

An earlier paper from the group found that while doxycycline reduced free spirochetes by roughly 90%, it doubled the number of round-body forms — and that across five agents tested, biofilm-like colonies were reduced by only 30–55%.³¹

The independent corroboration

This is not exclusively the work of one laboratory, which is worth knowing. An Austrian group independently reported cutaneous Borrelia biofilm in a patient with therapy-resistant disease in 2022.³² More substantially, an Italian and Slovenian collaboration in 2025 characterised biofilms formed by B. afzelii and B. garinii isolated from patients with erythema migrans, measuring them at 35–45 micrometres thick with biomass correlating to extracellular DNA content — and quantified the consequence: planktonic MICs of 0.125–0.5 µg/mL against biofilm inhibitory concentrations of 2 µg/mL for amoxicillin, 16 for ceftriaxone and 32 for doxycycline. Up to a 64-fold difference.³³

What we do with that

We treat Borrelia biofilm as a well-motivated working hypothesis rather than an established fact. It is one reason we lean on botanicals with independent activity against dormant and aggregated forms — a subject we cover in detail in our companion article, which herbs actually work for chronic Lyme. It informs how we sequence treatment and it makes sense of clinical patterns we observe — particularly the relapsing course described in our article on the terrain-based approach to chronic Lyme. It is not the load-bearing premise of a treatment plan, and we do not tell patients that biofilm is definitely why they are not better, because nobody can currently know that.

One further caution. A 2025 paper proposing combined Candida and Borrelia biofilms in Lyme disease is circulating and being cited clinically. It is explicitly a hypothesis paper with no original data, in a journal that does not peer-review for validity, and its own authors acknowledge no direct experimental evidence. It is an idea, not a finding.

Biofilm in Mold Illness, Sinuses and CIRS

This is the territory where our patients live, and it is also where the gap between what is claimed and what is documented is widest. So it gets the longest treatment in this article, and the boundaries are marked as we go.

Which moulds actually form biofilm?

Start here, because the answer is not what the mold-illness literature implies.

What the Aspergillus matrix is made of

Where the research does exist, it is detailed. The A. fumigatus biofilm matrix contains galactosaminogalactan — the principal adhesin, which also masks beta-glucan from immune recognition — along with galactomannan, alpha-1,3-glucan, melanin, hydrophobins and, in some models, extracellular DNA.^37,82 Treating mature Aspergillus biofilm with DNase improves susceptibility to both caspofungin and amphotericin B, which tells you how structurally load-bearing that extracellular DNA is.

One nuance that matters: the matrix is not a fixed recipe. Alpha-1,3-glucan and melanin appear in aspergilloma matrix but not consistently in invasive disease, and extracellular DNA appears in some laboratory models and not others. “The Aspergillus biofilm matrix” is several different things depending on where it grew.

The Candida finding that explains a great deal

This is also why the raw numbers are so stark: against established Candida biofilm, fluconazole minimum inhibitory concentrations exceed 1,024 µg/mL, versus 0.25–16 µg/mL for the same organisms free-floating — while amphotericin B loses only 2- to 16-fold activity.³⁶

The same fungal polysaccharide does something else remarkable: it shields bacteria. Staphylococcus aureus embedded in a Candida biofilm survived 800 µg/mL of vancomycin at far higher rates than in monoculture, and digesting the beta-1,3-glucan abolished the protection.¹¹⁵ A fungal sugar protecting a bacterium from a glycopeptide antibiotic is a good illustration of why polymicrobial biofilm is harder than the sum of its parts.

Fungal biofilm in the sinuses — three questions people blur together

Does fungal biofilm exist in the sinuses? Yes, and this is genuine human tissue evidence. In 50 chronic rhinosinusitis patients and 10 controls examined by fluorescence in situ hybridisation with confocal microscopy, 36 of 50 CRS patients had biofilm and none of the 10 controls did. Of those 36, eleven had fungal biofilm — and seven of those eleven also had S. aureus biofilm.⁷⁵ That co-occurrence is real, and mechanistically interesting: fungal spores alone fail to establish biofilm because mucociliary clearance removes them, whereas co-inoculation with S. aureus creates a permissive inflammatory niche that lets fungi adhere.³⁸

Is fungal biofilm the cause of chronic rhinosinusitis? The mainstream answer is no, and the reason is the next section. Fungi in the sinus are best characterised as colonising and immunologically provocative rather than primarily infectious.⁸⁹

Is there fungal biofilm in the sinuses of mold-exposed patients who don’t have sinus disease? Unstudied. We searched deliberately for this and found no study that imaged or characterised fungal biofilm in a defined, water-damaged-building-exposed, non-invasive, chronically ill cohort. The findings above come from surgical CRS populations — patients sick enough to be having sinus surgery. Extending them to mold-exposed chronic fatigue patients is an inference by analogy, not a demonstrated finding. It is not disproven either. It simply has not been tested.

MARCoNS — where we part company with the protocol

We treat MARCoNS clinically, we run the swabs, and we have written a full article on why it so often will not clear. So this is not dismissal. But our readers deserve the state of the evidence.

Coagulase-negative staphylococci are textbook biofilm formers — they are the archetypal device-infection organism. That part is not in question. What is unestablished is the rest of the chain. No study has confirmed that nasal MARCoNS in these patients exists as a biofilm. The one peer-reviewed study of MARCoNS clearance is retrospective, single-site, uncontrolled, performed no speciation or toxin profiling, made no biofilm confirmation whatsoever, and its own authors explicitly decline to claim causality.³⁹ And there is a base-rate problem that is difficult to get around: methicillin-resistant coagulase-negative staphylococci are found in the noses of roughly one in six healthy people.⁴⁰

We continue to address MARCoNS in the right clinical context because we see patients improve when it clears. We also think honesty requires saying that the mechanism is assumed rather than demonstrated, and that a positive swab on its own does not distinguish a sick patient from a well one. The S. aureus–fungal synergy finding above is, we suspect, a more productive way to think about the sinus in these patients than MARCoNS alone.

What about antifungals?

The question we are asked most by mold patients is whether they should be on itraconazole or another systemic antifungal, and specifically whether it will address biofilm.

The short answer is that no antifungal has been shown to disrupt biofilm. No licensed antifungal has a matrix-disrupting mechanism at all — the agents that genuinely degrade fungal matrix are experimental enzymes not available as human therapeutics. And in the one human condition where fungal sinus biofilm has actually been demonstrated on tissue, a Cochrane review of eight randomised trials in 490 adults found no benefit from topical or systemic antifungals on symptoms or quality of life, with more nasal irritation than placebo.⁸⁰

That is not the same as saying antifungals have no place — we use them in selected patients, for defined indications, with monitoring. It is saying that biofilm disruption is not the reason to use them, and that the antifungal question in mold illness is a large and genuinely contested subject in its own right. We are writing that one separately.

Biofilm in the Gut

Gut biofilm is where most patients first encounter this subject, usually through a protocol promising to strip a layer of pathogenic slime off the intestinal wall. The real picture is more interesting and considerably more nuanced.

The evidence that gut biofilm matters

The strongest human data come from endoscopy. In a study screening more than 1,400 patients across two European centres, visible mucosal biofilms were found in 57% of patients with irritable bowel syndrome and 34% with ulcerative colitis, against 6% of controls. Escherichia coli and Ruminococcus gnavus dominated, and bile acids accumulated within the biofilms — offering a plausible mechanism for the diarrhoea.⁴¹

Separately, colonic mucosal biofilms have been repeatedly associated with colorectal neoplasia, including 68.8% of tumours in a 2025 series with specific enrichment of Fusobacterium nucleatum.^7,42

So: gut biofilm is real, it is visible endoscopically, and it is associated with disease. That is a meaningful finding and it validates the instinct that something structural is going on in these patients.

Two findings that complicate the simple story

First, in a screening colonoscopy cohort of more than two thousand participants, mucosal biofilm turned out to be common in people without disease, and the authors concluded that biofilms alone may not drive early polyp development.⁴³ Second, and more pointedly, mucosal biofilms taken from healthy donors proved carcinogenic in a mouse model.⁴⁴ Biofilm presence and pathology are not the same thing.

Can You Test for Biofilm?

No. And this deserves stating bluntly, because a great deal is sold on the implication that you can.

A 2026 review in the Journal of Clinical Microbiology puts it directly: standardised measures for diagnosing bacterial biofilm have yet to be established, and minimum inhibitory concentrations cannot be used to predict treatment outcomes in biofilm-associated infection. Biofilm-specific assays exist — minimum biofilm inhibitory and eradication concentrations — but they are not standardised for routine clinical use and lack regulatory validation.⁴⁷

What does work requires a physical specimen from the infected site: sonication of explanted hardware (78% sensitivity, 91% specificity for prosthetic joint infection), tissue culture, and electron or confocal microscopy on tissue. None of which applies to a patient sitting in an office wondering whether they have biofilm.

The best available framework is a set of six clinical criteria — organisms surface-associated or aggregated, infection localised, recalcitrant to antibiotics despite favourable susceptibility testing, culture-negative despite clinical suspicion, ineffective host clearance — and their authors state plainly that this distinction “is not a simple determination that can be evaluated by the tests and examinations usually employed in medical diagnostic work-ups.”⁴⁸ It is an inference, not a measurement.

The one exception worth knowing about

There is a partial exception, and since many of our patients are already running the test, it is worth explaining precisely.

MicrobiologyDx, the laboratory that processes the deep nasal MARCoNS swab, offers biofilm testing as an add-on on cultures that have already come back MARCoNS-positive — currently around $100, with roughly a week’s turnaround, reported as present or absent. It has to be specifically requested; it is not part of the standard culture report. Their stated rationale is the standard one: biofilm-producing organisms may require several-fold higher antimicrobial concentrations than organisms that do not produce biofilm.

So if you want a biofilm result on paper, that is the one place in this whole territory you can get one. Two caveats, stated plainly.

First, the laboratory publishes no method and no validation data for this assay — no sensitivity, specificity or reproducibility figures. Second, and more important conceptually: the standard way to answer “does this isolate form biofilm” is to grow the cultured organism in a microtitre plate and see whether it produces a matrix. That tests whether the organism can form biofilm under laboratory conditions. It does not demonstrate that biofilm was present in the patient’s sinus. Those are different claims, and the distinction matters when a result is being used to justify months of treatment.

MicrogenDX, by contrast, does not report biofilm at all — its platform identifies organisms and their relative abundance, which is a different and also useful thing.

Fry Laboratories and the Limits of Looking

No account of biofilm in chronic illness is complete without Dr. Stephen Fry, and we want to handle this carefully, because he occupies an unusual position: genuinely influential, genuinely early, and genuinely criticised.

What he deserves credit for

Fry, a physician and microbiologist, founded Fry Laboratories in Scottsdale in 2007 and built it around a question mainstream medicine was not asking at the time: are there organisms in chronically ill patients that conventional testing does not find, living in forms conventional testing does not look for? He pushed biofilm into the chronic-illness conversation years before most clinicians in this space had heard the word, at a point when mainstream medicine confined the concept to prosthetic devices and cystic fibrosis.

He also did real work. His laboratory published in peer-reviewed, indexed journals — including a 2017 paper in Molecular and Cellular Probes reporting polymicrobial communities in explanted vascular filters and atheroma debris,⁹⁸ which is a legitimate contribution. He holds a granted patent on protozoal detection methods, and his lab developed enough genuine high-throughput sequencing capability to partner with Beckman Coulter Life Sciences in 2021. Patients tested there were not dealing with someone uninterested in evidence.

What happened with the organism

In 2011 Fry described what he presented as a novel biofilm-forming protozoan — FL1953, later named Protomyxzoa rheumatica — reported in patients with chronic fatigue, fibromyalgia, MS, ALS and rheumatoid arthritis. The formal description appeared in 2014 as a series of six single-patient case reports.⁹⁹

Why this happened, and why it matters beyond one laboratory

The same soil fungus recurs across the lab’s own published work — in vascular filters, in atheroma, and in 11 of 12 prostate cancer specimens.^98,100 Finding one environmental organism in over 90% of unrelated specimens from unrelated organs is the classic signature of reagent and environmental DNA contamination. That is not a hostile interpretation; the 2022 paper’s own limitations section records that confirmatory PCR for these fungi failed to validate the sequencing results.

This is a known, general problem, not a personal failing. DNA extraction kits and PCR reagents carry their own microbial DNA — the so-called “kitome” — and in low-biomass samples that background dominates the signal.⁹⁷ Blood is about the lowest-biomass specimen in clinical medicine, which makes it maximally vulnerable. The paper establishing this appeared in 2014, the same year as Fry’s case series, and it changed the standard every such study is now held to.

Fungal colonies and biofilm in blood

The claim that chronically ill patients harbour fungal colonies or biofilm circulating in their blood is one we are asked about often, so let us be direct about the evidence.

The best study of the phenomenon took blood from 30 healthy people and examined the “bacteria-like structures” visible on darkfield microscopy using electron microscopy, proteomics, lipid analysis and 16S sequencing. They turned out to be membrane vesicles and protein aggregates containing ordinary blood proteins — albumin, fetuin-A, apolipoprotein-A1 — arising from normal blood cell breakdown during handling. And critically, the accompanying bacterial DNA signal appeared in the negative controls too.⁹⁵ The microscopy artefact and the sequencing artefact arise together.

Separately, deep sequencing of nearly 10,000 healthy people found no core blood microbiome at all; the signals detected were consistent with contamination and transient translocation, not a resident community.⁹⁶

There is a technical point here worth understanding, because it applies to any image you are shown. The fluorescent stains used in this kind of work — Hoechst and ethidium bromide — bind any double-stranded DNA. They cannot distinguish microbial DNA from your own nuclear or mitochondrial DNA, from platelet and white-cell debris, or from the extracellular DNA that neutrophils release in webs when they are activated. A glowing web-like structure in a blood film is exactly what host extracellular DNA looks like. Calcofluor white, used for fungi, binds chitin and cellulose — so it also stains plant material, cotton fibres and assorted debris. A fluorescent image cannot identify an organism without something else confirming it.

Real fungaemia, incidentally, is a well-characterised and serious condition — but it occurs in ICU patients, transplant recipients, people with central lines and the profoundly immunosuppressed, it presents acutely with sepsis, and it carries 30–40% mortality. There is no described entity of indolent, occult fungal colonisation of the blood in well-appearing ambulatory patients.

What he actually recommends — since patients ask

We looked, because it is a reasonable question and the answer is not widely written down. There is no published Fry protocol; everything below comes from interviews and conference lectures rather than any peer-reviewed source.

His documented approach is pharmaceutical first: tetracyclines as first line, then azithromycin, then hydroxychloroquine combined with a tetracycline, with ivermectin (at fractional doses twice weekly), metronidazole and, later in his practice, antifungals including terbinafine. Supportively he uses heat — sauna, hot tub, steam — vitamin D, high-dose vitamin C, curcumin and berberine.

Two things stand out, and both are worth knowing.

The first is his ultra-low-fat, whole-food plant-based diet, which he names as the McDougall diet and describes as his most powerful single tool, on the explicit theory that the organism and its biofilm depend on lipids. He reports that most patients starting it have a Herxheimer reaction. We should be straightforward that the general principle does not hold up: the published biofilm literature runs the other way, where medium-chain fatty acids and monoglycerides — lauric acid and glycerol monolaurate, the very molecule in AquaLaurin — are anti-biofilm, and a fatty acid, cis-2-decenoic acid, is the canonical biofilm dispersal signal. There is no published evidence that dietary fat restriction reduces biofilm burden in any organism or any model.

The second is that his one biofilm-specific intervention is a subtraction: he tells patients to stop magnesium supplements, on the reasoning that extracellular DNA is the structural rebar of a biofilm matrix and magnesium stabilises it. The mechanism is genuinely grounded — divalent cations do stabilise the eDNA lattice, and adding calcium, magnesium or iron blocks EDTA-mediated biofilm killing in vitro. The leap to oral magnesium supplements is not supported: serum ionised magnesium is tightly regulated, and oral dosing does not produce the local millimolar excursions used in those experiments. We do not withhold magnesium on this basis.

Notably, he does not recommend any of the biofilm agents this article is mostly about — no enzymes, no EDTA, no NAC, no lactoferrin, no bismuth, and no fibrinolytics. The lumbrokinase and “fibrin cage” framing that circulates alongside his name in patient forums is not his; his stated model of biofilm structure is an eDNA lattice, not fibrin.

We should add one thing in fairness: we found no regulatory action, licensing sanction, or fraud case against Fry or his laboratory, despite the rumours that circulate. We looked. The scientific criticism is substantial; the regulatory criticism appears to be unfounded, and we will not repeat it.

The Problem Nobody Mentions: Concentration

This is the most important section in the article, and the one you will almost never see addressed. If you understand it, you can evaluate any biofilm product claim yourself.

Every in-vitro biofilm study reports the concentration at which the agent worked. Those concentrations are startling once you start reading them:

• N-acetylcysteine disrupts biofilm at 1–50 milligrams per millilitre — and has an oral bioavailability of roughly 4–11%.
• Glycerol monolaurate kills mature biofilm at 500 micrograms per millilitre — with no human pharmacokinetic data at all.
• Cistus incanus reduces established biofilm at 5–60 milligrams per millilitre — that is 0.5–6% by weight.
• Grape seed proanthocyanidins work from 0.5 up to 250 milligrams per millilitre.

Now ask where in a human body those concentrations are achievable.

The answer is consistent, and it is the hinge of this entire subject: the mouth, the sinus cavity if you irrigate it, the gut lumen, an open wound, and the inside of a catheter. Those are compartments where you can place a substance directly and let it sit at high concentration.

They are not achievable in plasma. Not in joint fluid. Not in the central nervous system. Not in connective tissue. Not, in other words, in the places where a disseminated chronic infection actually lives.

Three further points follow from this, and they recur throughout the table below.

Inhibiting formation is not the same as dissolving what exists. This is the most common conflation in the field. Lactoferrin, serrapeptase, grape seed extract and Cistus all have good data showing they stop biofilm from forming. Very few agents have data showing they dismantle a mature one — a far harder task, and the one that actually applies to a patient who has been ill for years.

Mechanistic plausibility is being reported as mechanism. The story that NAC’s thiol group reduces disulfide bonds in matrix proteins is chemically reasonable and repeated everywhere. According to the most recent dedicated review, it has never been demonstrated in a mature biofilm. It is an inference from chemistry, not an observation.

A biofilm disruptor is not automatically benign. More on this below, but briefly: berberine at the low concentrations actually achievable in the body may inhibit biofilm dispersal rather than promote it; NAC antagonises some antibiotics; and nattokinase measurably alters human coagulation.

AquaLaurin and Glycerol Monolaurate

AquaLaurin is the agent we reach for most often, and it is worth explaining why — and also being precise about which parts of the case are established and which are not.

The molecule

The active ingredient is glycerol monolaurate, or GML: a monoglyceride formed from glycerol and lauric acid. It is not an exotic compound. It is one of the principal antimicrobial constituents of human breast milk, present at roughly 3,000 µg/mL in human milk against about 150 in bovine milk and none in infant formula — and when GML is stripped from human milk its antibacterial activity disappears, returning when GML is added back.⁶⁷ That provenance is not sentimentality; it is the reason the tolerability profile is what it is. This is a molecule mammalian infants consume in quantity from birth.

The microbiology is genuinely impressive, and unusually so for a natural product:

There is good antifungal data too. Monolaurin significantly reduced viable counts in mature 24-hour Candida albicans biofilms, comparably to fluconazole controls⁶⁸ — which, given the fluconazole tolerance figures earlier in this article, is a meaningful result.

The limitations, stated plainly

Three, and they are worth knowing before you spend money.

It is Gram-positive selective under physiological conditions. Pseudomonas and the Enterobacteriaceae, including E. coli, resist GML at neutral pH. They become susceptible at acidic pH or when EDTA is added.⁴⁹ So GML is not a universal antimicrobial, and if the organism you care about is a Gram-negative, this may not be your agent.

There is no human pharmacokinetic data at all. A systematic review of the literature on monolaurin as a dietary supplement found 190 papers and not one human clinical trial, with the only human studies being topical.⁶⁹ Whether oral GML reaches the systemic circulation intact is unknown — and there is a specific pharmacological reason for scepticism: monoglycerides are the normal end-products of fat digestion, absorbed by enterocytes and re-esterified into triglycerides, while pancreatic and intestinal lipases hydrolyse monolaurin back to lauric acid and glycerol. The reasonable expectation is that oral GML acts primarily in the gut lumen and oropharynx.

Anti-Borrelia claims are preliminary. The only source we could find is a 2022 conference abstract — not a peer-reviewed paper, no methods, and it concerns antimicrobial activity rather than biofilm specifically. Treat any claim that monolaurin kills Lyme biofilm as unsupported for now.

The product, and what is and is not known about it

AquaLaurin is a water-soluble nano-formulation of GML with added phytochemicals, produced by Northstar Biosciences. Making GML water-soluble is a real and non-trivial formulation achievement — GML is a lipid, and lipids do not readily disperse in water, which historically limited how it could be delivered.

The manufacturer describes enhanced bioavailability and a slow-release lyotropic crystalline structure, and offers delivery by nebuliser, nasal spray and oral solution with different absorption fractions for each route. Some of the promotional material also invokes structured water and vibrational modelling. We want to be straightforward: the formulation data are proprietary and unpublished, and we cannot point you to peer-reviewed evidence for the specific bioavailability figures or for the structured-water mechanism. They may well be accurate. They have not been independently demonstrated, and we do not repeat them to patients as established fact.

What we can say, and do say, is this. The active molecule has among the best established-biofilm data of anything in this category. It is found naturally in human milk. It does not select for resistance. It is remarkably well tolerated — we have not seen a serious adverse reaction to it. And in our hands it is gentle in a way that matters enormously for this patient population, who so often cannot tolerate the more aggressive options. That combination is why it is our first choice, and it is a judgement based on mechanism, safety and clinical observation rather than on trial evidence, which does not exist.

Cistus incanus — The One We Talk About Most

If AquaLaurin is our workhorse, Cistus is the agent we recommend most widely, because the risk-benefit arithmetic is close to unbeatable: it is a food-grade herbal tea, it costs very little, it is exceptionally well tolerated, and it has more relevant biofilm data behind it than most prescription options. We have written separately on why Cistus incanus has become such a useful tool in chronic illness, and everything there still stands.

Here is what the biofilm evidence actually consists of — and one paper that deserves to be much better known.

The paper most people miss

The supporting evidence, and its boundaries

Cistus creticus inhibits Streptococcus mutans biofilm formation and shows low minimum inhibitory concentrations against oral anaerobes.⁵¹ A systematic review of herbal interventions against multispecies oral biofilm rated cistus tea among the most bactericidal agents studied, with reductions up to 66%.^52,70 And in the closest thing to human data anywhere in this article, studies using intraoral splints worn by volunteers found cistus tea at 10–20 mg/mL reduced bacterial colonisation on the dental pellicle.⁵²

Separately, Cistus extract potently blocks attachment of enveloped viruses by binding their envelope proteins, with polyphenols confirmed as the active principle and no resistance emerging over 24 weeks of exposure.⁷¹ That is antiviral activity, not biofilm activity — the two get conflated constantly — but it is relevant to the viral coinfection burden that so many of our Lyme and CIRS patients carry.

Now the boundary. Every one of the effective concentrations above — 5 to 60 milligrams per millilitre — is a mouthwash or tea concentration. The digestion study tested oral bioaccessibility, meaning persistence in the mouth, not systemic absorption. Cistus has genuine established-biofilm data, and that data is confined to surfaces you can bathe directly. There is no human outcome data for Cistus in Lyme disease or CIRS.

How we use it

As tea, first and foremost — one to two teaspoons of dried herb steeped ten minutes, two to three cups daily. This is not us being old-fashioned. Tea is the form that puts the polyphenols into sustained, high-concentration contact with the oral and gastrointestinal mucosa, which is precisely where the evidence says they work. Capsules do not do this.

We also use cooled tea as an oral rinse and a nasal rinse, which is arguably the single best-supported application of anything discussed in this article. For patients who prefer a standardised preparation or want it alongside the tea, we stock Cistus incanus by BioPure.

One practical point: Cistus is tannin-rich, and tannins bind minerals and some medications. Separate it from supplements and prescriptions by about two hours.

Swish it, gargle it, rinse with it, drink it. That is where the evidence lives.

Your Antimicrobial Herbs Are Already Doing Biofilm Work

Here is something that gets lost when biofilm is treated as a separate product category: several of the herbs used as antimicrobials in chronic infection have anti-biofilm activity built in. They are rarely sold as biofilm agents, but the laboratory data are there — and, importantly, they are not there for all of them.

We went through the published record herb by herb, because patients deserve to know which of their bottles are doing double duty and which are not.

Essential oils — the strongest natural-product biofilm literature there is

Essential oils have the largest in-vitro anti-biofilm literature of any natural product class — larger than the enzymes, larger than the chelators. Oregano, cinnamon bark and clove bud oils sterilised stationary-phase cultures at 0.05–0.1% and dissolved biofilm-like aggregates; a follow-up found garlic oil and cinnamaldehyde active at 0.05%.^65,66

The catch is the one that runs through this whole article. Those concentrations are reachable in the gut lumen and on a surface, not in plasma. Essential oils are a luminal tool, which is exactly why they belong in gut protocols rather than being asked to treat tissue infection.

We cover the full botanical protocol — which herbs, which form, and the dry herb to tincture to liposomal progression — in our companion article, which herbs actually work for chronic Lyme. The short version for present purposes: if you are already running a properly sequenced herbal protocol, you are not starting from zero on biofilm.

The best biofilm agent is often one you are already taking for something else.

Gut Biofilm: NAC, Enzymes and Bismuth-Thiols

Biofilm work divides naturally by compartment, and the gut is the one where oral agents can genuinely reach therapeutic concentration. It is therefore where most of the realistic opportunity lies — and it is worth treating as a distinct project from systemic work rather than lumping them together.

NAC — our default for the gut

N-acetylcysteine has the best human evidence of anything in this article, and it is still not strong evidence. That is the honest summary of the field.

The striking result is in refractory Helicobacter pylori. In forty patients who had all failed four or more previous eradication attempts, a short NAC pre-treatment before culture-guided antibiotics raised eradication from 20% to 65%. Notably — and this is rare in the biofilm literature — the investigators actually looked: biofilm persisted on gastric biopsy only in the treatment failures.⁵⁴

The counterweight is a Cochrane review of eight trials in 559 participants which concluded it is uncertain whether NAC improves eradication, rating the evidence very low certainty on methodological grounds.⁵⁵ One striking trial in a highly selected population, not robustly replicated.

Two mechanistic points that should shape how you use it. NAC is most active below pH 3.3, and its efficacy falls substantially at physiological pH — which is a strong argument that its plausible biofilm action is in the stomach, gut lumen, sinus and wound rather than systemically. And oral bioavailability is only 4–11%, so systemic concentrations come nowhere near the milligram-per-millilitre range at which it works in vitro. NAC is an excellent gut and mucosal agent. It is not a systemic biofilm drug, and the widely repeated claim that its thiol group reduces disulfide bonds in matrix proteins has, according to the most recent dedicated review, never actually been demonstrated in a mature biofilm.⁵³

We use NAC 600 mg at one to two capsules daily for gut biofilm work, which is the dose range used in the clinical studies. It is also a glutathione precursor, so it is doing useful work on the detoxification side simultaneously — a genuine two-for-one in a population that needs both.

Bismuth-thiols — the best chemistry, awkwardly delivered

Bismuth-thiol complexes have among the strongest anti-biofilm credentials of any chemistry in existence. A pharmaceutical bismuth-thiol has completed randomised, double-blind, placebo-controlled Phase 1b and Phase 2 trials in infected diabetic foot ulcers with positive results.⁵⁶ This is not fringe science.

Biofilm Phase-2 Advanced is an intelligent attempt to bring that chemistry to the gut: 200 mg bismuth subnitrate plus 300 mg alpha lipoic acid — a bismuth salt and a thiol, co-administered in the hope that they associate in the lumen — with black cumin seed, which has independent in-vitro anti-biofilm activity. The reasoning is good.

Proteolytic enzymes — the honest ranking

Enzymes are the most-used and least-evidenced category in integrative biofilm practice, and the internal ranking is not what most people assume.

Nattokinase has the best data — better than serrapeptase. In a rat intravascular catheter model it produced 85–90% biofilm biomass reduction, and combined with rifampicin achieved complete eradication within catheters. The mechanism is specific and physically sensible: device-associated staph biofilm is built largely on a scaffold of host fibrin, and fibrinolytics dissolve fibrin.⁵⁷ Note what that implies, though — it does not automatically generalise to gut, sinus or tissue biofilms, which are not built on host fibrin.

Serrapeptase inhibits biofilm formation in vitro at low concentrations, but the key study did not test established biofilm and did not test antibiotic combination.⁵⁸ A systematic review of twenty-four studies concluded the evidence is insufficient to support its use, and noted that human oral bioavailability data are simply absent from the literature — the absorption evidence people cite is from rats.⁷² Rare eosinophilic pneumonitis has been reported.

Lumbrokinase is the clearest case of the field running ahead of the data. We searched specifically and found no primary peer-reviewed study of lumbrokinase against any bacterial biofilm, and none against Borrelia. It has a real fibrinolytic and cardiovascular literature — that is a different claim. We stock Lumbroxym and we use it, but for hypercoagulability and microcirculation, which is where its evidence actually is. We also use Theraxym as a systemic proteolytic enzyme where the indication is inflammation and fibrin rather than biofilm specifically. We think patients are entitled to know which rationale they are buying.

How We Use Biofilm Work in Practice

All of the above is background. Here is what actually happens in the office, and the sequencing matters more than the product selection.

Seven More Worth Knowing About

The table above covers what we use most. These are the ones patients ask about, plus one that deserves far more attention than it gets.

1. Baking soda — the cheapest thing in this article, and it works

For oral biofilm specifically, plain sodium bicarbonate has better human evidence than most of the supplements in this field. A systematic review of 21 studies and 43 comparisons found baking soda dentifrice significantly outperformed controls on plaque removal after a single brushing and reduced gingival bleeding.¹²²

Three mechanisms, and it is worth not conflating them: it is a low-hardness abrasive whose crystals fracture on contact, so it removes biofilm without the enamel wear penalty of silica; it neutralises acid within the plaque biofilm; and there is direct in-vitro evidence of physical biofilm disruption independent of brushing. Be honest about the size of the effect though — these are small absolute differences, and baking soda beats no dentifrice more clearly than it beats other dentifrices over time. Most of the studies were industry-funded.

Worth adding: the one direct test of bicarbonate nasal irrigation against biofilm found only a small reduction, and adding xylitol to it made no difference. So this is an oral-cavity recommendation, not a sinus one.

2. InterFase and InterFase Plus — what is actually in them

These enzyme blends have been a staple of gut biofilm protocols for years, including in our own practice, and the formulation logic is sound: InterFase is a proprietary blend of enzymes aimed at matrix polysaccharides and proteins — glucoamylase, chitosanase, cellulase, hemicellulase, beta-glucanase, a protease/peptidase complex, lysozyme and enteric-coated serratia peptidase. InterFase Plus adds disodium EDTA, which is the meaningful difference: a chelator to strip the calcium and magnesium cross-linking the matrix, alongside enzymes to degrade it.

We still use them, in defined courses, away from food. We simply do not pretend the evidence is stronger than it is.

3. EDTA suppositories — a reasonable workaround, with an honest caveat

Oral EDTA is poorly absorbed, which is why we sometimes use rectal EDTA suppositories instead — the reasoning being that rectal administration bypasses first-pass metabolism and the gut absorption problem.

Two things patients should know. The pharmacokinetic evidence for rectal EDTA absorption is thin — essentially one small animal study — and there is no published evidence whatsoever for EDTA suppositories against biofilm specifically. Even the manufacturers do not claim it. When we use them it is on chelation reasoning, not biofilm reasoning, and we say so.

One safety point that genuinely matters: calcium disodium EDTA and disodium EDTA are not interchangeable. Disodium EDTA given intravenously has caused fatal hypocalcaemia. Suppository products use the calcium disodium form for this reason, and this is not a place to improvise.

4. For the sinuses — and a correction on what is in the bottle

We use BioFilm Clear nasal spray, which brings EDTA, xylitol and a botanical together in one delivery to the compartment where those concentrations are actually achievable.

A correction worth making, because we have said it wrongly ourselves: the botanical is grapefruit seed extract, not grape seed extract. Those are different products with different evidence, and grapefruit seed extract has a specific problem — commercial preparations have repeatedly been found to owe their antimicrobial activity to synthetic preservative contaminants rather than to anything in the fruit. The original formula also contains colloidal silver at 30 ppm; a silver-free version (Clear-X) is available, which we generally prefer for nasal use. The EDTA salt form is not disclosed on either label.

Not to be confused with BEG spray — the compounded prescription combining mupirocin, disodium EDTA at 1% and gentamicin — which is a different product with a specified EDTA concentration, used in MARCoNS treatment.

5. Mimosa pudica — what the evidence does and does not say

Mimosa pudica seed, the basis of CellCore’s Para 1, is widely described as removing gut biofilm. We use it and patients often feel it does something. Here is what we could actually verify.

The seed genuinely forms a viscous gel — that part is well characterised, as a galactomannan mucilage studied as a pharmaceutical binder and hydrogel base. What does not exist is any study of Mimosa pudica seed against gut biofilm. Not one. The single anti-biofilm paper on the plant used an aqueous aerial extract as a dental mouthrinse against Streptococcus mutans, and chlorhexidine outperformed it. The anthelmintic data are from leaf extracts tested on earthworms — a cheap screening model in which nearly every crude plant extract “works.” There is no human clinical data of any kind. Worth noting: CellCore’s own product page does not use the words “biofilm” or “parasite” — those come from retailers and practitioners, not the manufacturer.

Practical cautions that follow from the gel mechanism rather than from any Mimosa-specific study: separate it from all medications by about two hours, take it with adequate fluid, and be careful in anyone with strictures, adhesions, prior bowel surgery, gastroparesis or opioid-related slow transit.

6. Hyperbaric oxygen — the one that should be getting more attention

7. Modified citrus pectin, galectin-3 and the “survival paradox”

Several of our patients cannot tolerate the usual binders. Charcoal constipates them, clay is too drying, cholestyramine is intolerable. For those patients we often use modified citrus pectin (PectaSol) — citrus pectin broken into short fragments small enough to be absorbed. It is gentle, it is easy to take, and in our experience patients who react to everything else generally do well on it. Tree of Light Health sells PectaSol, so read what follows with that in mind.

The interest in it comes from the work of Dr. Isaac Eliaz, an integrative physician in California who has published most of the human research on modified citrus pectin and wrote the book The Survival Paradox.¹³¹ His argument, in short: the body’s survival response — inflammation, fibrosis, walling things off — is protective in the short term and destructive when it never switches off. He places a single molecule at the center of that stuck response: galectin-3.

The hypothesis we find interesting. A persistent biofilm is, among other things, a persistent immune stimulus the body cannot clear. Dr. Robert Naviaux has described a related idea from the mitochondrial side: the cell danger response, in which cells that have met a threat stay in a defensive metabolic state long after the threat is handled, and healing stalls at an incomplete stage.¹³³ Eliaz’s survival paradox describes much the same trap one level up, in the immune and fibrotic response.

Could a biofilm the body cannot resolve be one of the things holding a patient in that survival state?

It is a reasonable question and it fits what we see clinically: patients who improve slowly and incompletely, with inflammation that will not settle, until something the body had walled off is finally addressed. But we want to be clear that it is a hypothesis, not a finding. Nobody has measured galectin-3 in a biofilm-associated infection, nobody has shown that clearing biofilm lowers it, and nobody has tested modified citrus pectin in these patients.

Where we land. We use modified citrus pectin as a gentle binder for people who cannot tolerate the others, at about 5 g one to three times daily, and for its possible effect on the inflammatory and fibrotic side of chronic illness. Its metal-binding data are small and uncontrolled — increased urinary arsenic and cadmium over a few days in healthy volunteers, and lower blood lead in hospitalised children^128,129 — which is enough to make it a reasonable gentle option alongside proper treatment, and not enough to call it a chelator in the medical sense. It is not a biofilm agent, and we do not present it as one.

One distinction we want to draw carefully, since we also run ozone therapy. Ozone is not hyperbaric oxygen, and it does not have biofilm evidence. Topical and dental ozone has in-vitro anti-biofilm data — where it is consistently inferior to plain sodium hypochlorite — and systemic ozone has none at all. The mechanism above requires sustained restoration of oxygen tension inside an aggregate, which is a partial-pressure phenomenon; a bolus of oxidant into blood is largely consumed within seconds. We value EBOO ozone for its oxidative and immune effects, and we do not claim it disrupts biofilm.

Two others deserve a mention for completeness. Rifampin has the strongest human anti-biofilm evidence of any antibiotic, in prosthetic joint and device infection — always in combination, never alone, because resistance emerges fast. And nitric oxide at very low concentrations is the cleanest proof that biofilms can be deliberately switched off: in a small randomised trial, inhaled nitric oxide significantly reduced Pseudomonas biofilm aggregates in cystic fibrosis patients.¹²⁵ The dispersal window is picomolar to low micromolar, bactericidal is above a millimolar, and above about 2.5 millimolar it promotes biofilm — so this is emphatically not a supplement-shelf intervention, and there is currently no product that lets a clinician do it.

One Biofilm We Have Not Covered Here: The Mouth

There is a large, chronic, bleeding biofilm reservoir with daily access to the bloodstream that most chronic illness protocols never mention: the subgingival plaque below the gumline, and in particular the cluster of organisms periodontal researchers call the red complex — Porphyromonas gingivalis, Tannerella forsythia and Treponema denticola.

We are keenly aware of it, we look for it, and we work alongside biological dentistry on it. It also deserves more room than a subsection, so it has its own article: Are Red-Complex Bacteria Really a Problem?, with two companions on salivary testing and on supporting gum health.

Three things are worth saying now.

• Mechanical disruption by a dental professional is the evidence-based backbone, and it is not optional. This is the one biofilm in the body a clinician can physically reach and physically remove. Everything else in this article is a workaround for not being able to do that.
• The systemic associations are real but frequently oversold. The strongest is diabetes, where treating periodontitis produces a clinically meaningful drop in HbA1c across randomised trials. Cardiovascular and rheumatoid associations are genuine but unproven as causation. And the cautionary tale is preterm birth: observational studies link it consistently to gum disease, yet randomised treatment trials show no reduction.
• It frequently sits alongside the interference fields we already treat. The same patients often have cavitations, failed root canals and old dental foci — which we address with neural therapy, and which are a recurring reason a systemic protocol stalls.

If you are working through a chronic infection protocol and have not had your periodontal status properly assessed, that is likely the highest-yield unaddressed biofilm you have.

Frequently Asked Questions