Evidence

Epithelial barrier damage

The first stage of the model is damage to the epithelial barrier. Although that initial damaging event for myself was chemical, it doesn't have to be for others. It can come from infection, such as COVID-19, a respiratory virus that first infects the lining of the nose and airways, or from an inherited weakness in connective tissue that leaves the structural integrity more vulnerable. Because the damage is non-specific, the illness can begin in many different ways and look different from one person to the next, which is part of why these conditions are so heterogeneous.

So what evidence is there that epithelial barrier damage is occurring, and what further testing could we do to prove and categorize the type of damage further? Photographing barrier damage in my own airway would be the most concrete way to prove this and it could be done, but only with an invasive biopsy. That would not be a standard clinical test, which is why I haven't had it done. We do know however that environmental exposures can cause this type of change, as it has been shown directly in laboratory airway epithelium directly exposed to formaldehyde (E1).

What viral infections do to the epithelial barrier

We also now have evidence of this in people with Long COVID. Baalbaki et al. (2026) studied nasal epithelial cells from Long COVID patients in the Dutch P4O2 cohort, alongside people who had recovered from COVID-19 and healthy controls. In the subgroup of Long COVID patients that this test was conducted on, the nasal epithelium showed impaired barrier function in all patients, with reduced levels of the tight-junction proteins ZO-1 and occludin, and were more sensitive to viral triggers. Barrier repair was impaired despite faster wound closure, and it produced more inflammatory signaling, including IL-1β, which in turn activated immune cells in the blood. The authors conclude that nasal epithelial dysfunction in this subgroup may contribute to the wider immune dysregulation seen in Long COVID.

So chemical exposures are one way to damage this barrier, viral infections are another, and now SARS-CoC-2 infections have been studied very closely. In a human chip model built from epithelial cells, endothelial cells and immune cells, SARS-CoV-2 infected the epithelial layer efficiently, reaching high viral loads and provoking a strong inflammatory and interferon response (Deinhardt-Emmer et al., 2021).

Twenty-eight hours after infection that epithelium was visibly coming apart. Electron microscopy showed dead cells and cell remnants lying on the layer, with the plasma membrane disrupted and the microvilli lost. Lactate dehydrogenase, which a cell releases when its membrane ruptures, rose significantly on the epithelial side. Over the same window the barrier itself began leaking: permeability was unaffected at eight hours and significantly increased at twenty-eight.

What happened underneath is the part I keep returning to. The endothelial layer was never productively infected — no viral components were visible inside endothelial cells at eight, twenty-eight or forty hours — and yet by forty hours its barrier function was failing too. The paper puts that down to cytokines released by the infected epithelium beside it, and reports viral particles crossing the disturbed barrier. The authors note this matches what was found at autopsy in people who died of COVID-19: the endothelial layers of the alveolar capillaries still intact, the epithelial tissue seriously damaged (Deinhardt-Emmer et al., 2021b).

That is the shape of the argument this whole site makes, arrived at from a completely different direction: damage the epithelial barrier, and things that were never meant to get through start getting through.

The gap in Long COVID research

We already know an acute respiratory infection can strip away this barrier. E2 shows it happening within a day, and that part isn't controversial. What I find strange is how little of the Long COVID research effort has followed it. The field has concentrated on viral persistence, immune dysregulation and microclots. Barrier damage is not absent from that literature — but it is not being treated as a central mechanism either.

The evidence that the damage persists is there to be picked up. In one study, 100% of tested Long COVID patients showed impaired nasal epithelial barrier function, including near-total loss of Occludin, one of the core proteins holding the tight junctions between epithelial cells together, compared with healthy controls (Baalbaki et al., 2026). In hamsters, upper respiratory tract tissue stayed inflamed and epithelial cells kept dying for months after the infectious virus itself was gone (Liu et al., 2026) — damage outlasting the trigger.

Barrier integrity has also been measured as a predictor of how badly the acute illness goes. In 328 COVID-19 patients against 49 controls, severe cases carried significantly more circulating bacterial DNA, and a panel of proteins tracking bacterial translocation separated the severe cases out (Yazici et al., 2023). That study measures the gut barrier rather than the airway, and I include it for what it shows about barrier integrity being treated as a determinant of outcome, not as evidence about the tissue this site is concerned with.

Epithelial Barrier Theory

Damage to the epithelial barrier has already been implicated in a wide range of diseases, most firmly allergies and asthma, but also eczema and inflammatory bowel conditions. What has barely been considered is its role in complex chronic illnesses like ME/CFS and Long COVID. The researcher who has done the most work on this is Cezmi Akdis, at the Swiss Institute of Allergy and Asthma Research. His Epithelial Barrier Theory proposes that barrier damage plays a general role in many illnesses: toxic substances introduced by industrialization and modern lifestyles disrupt the epithelial barrier of the skin, the upper and lower airways and the gut, causing microbial dysbiosis and an inflammatory immune response that can initiate or aggravate more than 70 chronic inflammatory diseases (Epithelial Barrier Theory; Akdis, 2021). The model on this site is my own, but it builds on that foundation.

The exposures it names are the ones this site is about, and the airway work is specific. Laundry detergent, diluted 1 in 25,000, disrupts the tight junction barrier of primary human bronchial epithelial cells — and so does the residue left behind after rinsing (Wang et al., 2019). PM2.5 lowers transepithelial resistance and cuts claudin-1, occludin and ZO-1 in human nasal epithelium, and pretreatment with a steroid does not restore the barrier (Xian et al., 2020). Microplastic particles cross a human lung epithelial model, and nylon fibers and cleaned polyethylene disrupt that barrier and trigger IL-6 release without killing the cells (Donkers et al., 2022).

Where my argument departs from theirs is the disease list. The 2024 survey of the theory works through roughly 68 conditions across skin, airway, digestive, neuropsychiatric, autoimmune, metabolic and ocular categories (Sun et al., 2024). ME/CFS is not among them. Neither is fibromyalgia, and neither is multiple chemical sensitivity — none of the three is mentioned anywhere in that paper. Long COVID is discussed, but only as a possibility, and it appears in none of the paper's tables; severe COVID-19, by contrast, gets a table row and a firm statement that it fits the theory. So the mechanism is established and published. The connection is well established for allergies and asthma. It has not been made for ME/CFS or multiple chemical sensitivity, and that is the part I am arguing for.

Three findings that don't fit neatly under skin or vasculature, but that I think belong in the same barrier-damage picture.

A

B

Ocular mucus shedding. In 2025, mucus began coming off the surface of my eye in small clumps, repeatedly, for several months — not a one-off irritation. I had to remove clumps from the lower lid several times a day at the worst of it. It has not recurred since. I read a single shedding event like this as a permanent change to the eye's surface rather than a repeating condition, and I've since been diagnosed with corneal neuropathic pain, with neuropathy, neuromas and Langerhans-cell infiltrates found on corneal exam.

The eye's surface is protected by the same kind of barrier as the airway: a mucus layer, built from mucin proteins, that keeps irritants off the epithelium beneath it. Huitsing et al. (2024), working with Nancy Klimas's group, proposed that this layer may be compromised in ME/CFS. They point out that roughly 70% of ME/CFS patients report sicca symptoms (dry eyes and mouth) and 75–80% an irritant rhinitis with increased mucin production. In a pilot genetic dataset of 383 ME/CFS patients, they also report variants in the mucin genes MUC16, MUC19 and MUC22 in most of them. Their argument is that a weakened mucus layer leaves the epithelium exposed to repeated irritation, which could drive the kind of chronic low-grade inflammation seen in ME/CFS. What I saw in my own eye (E4) fits that picture from the other direction. Months of mucus shedding from the eye's surface were followed by a diagnosis of corneal neuropathy, with neuromas and immune-cell infiltrates in the cornea: the outcome you would expect if the protective layer failed and the tissue underneath was left exposed. The paper is a hypothesis, not a measurement. It doesn't examine anyone's eyes, and the effect of those gene variants is unknown. But it is the closest published account of what happened to mine.

Crimson crescents. This is a recognized clinical sign in ME/CFS — crimson discoloration along the anterior pharyngeal pillars, seen in roughly 80% of patients at one CFS clinic, without accompanying pharyngitis (Cunha, 1992) (E5). Its cause has never been established. I think two things combine to produce them. The first is breathing pattern: when the brain is short of oxygen, breathing shifts toward the mouth, which takes in more air with less effort than breathing through the nose. That sends more unfiltered, unhumidified air across the back of the throat. The second is barrier damage. In tissue whose barrier is already compromised, irritation concentrates where friction from airflow is greatest. As air turns down into the throat it meets the most resistance against the pillars at the sides, the way an aircraft wing's leading edge takes the most force. Inhaled VOCs and particles would irritate and deplete tissue exactly there. That would explain why the discoloration forms crescents at the sides rather than spreading evenly across the throat. It is my hypothesis; the cause has never been established.

Tissue depletion

Tissue depletion is Stage 2 of the model. The airway lining is constantly repaired by its own stem cells, the basal cells, which divide to replace cells that are damaged or killed. Every inhaled exposure that kills lining cells spends some of that repair capacity, and my hypothesis is that repeated injury eventually uses it up faster than it can recover, leaving the barrier thinner, slower to heal and easier to breach. Because the central airways take the heaviest share of everything we breathe in, I would expect that depletion to be concentrated in the central chest. The evidence below is what led me there: what I felt, what other patients describe, and what my own imaging shows.

Since many of the studies I would love to see done in Complex Chronic Illness haven't been done, I have to fill in the gaps with the only information source more trustworthy than peer-reviewed scientific studies: primary observation. While my own lived experience has been a fantastic (and horrible) lens through which to view these illnesses, I have to add the disclaimer that the sample size is obviously small. However, I've compared my experiences with the stories of enough people in the community who have the same diagnoses, and seen the same patterns, to justify a broader relevance. I might even speculate that the relevance here across the community is significantly larger than that of much better-known comorbidities, including CCI, MCAS, IBS and SFN.

In my own case, after my decline in 2021, I started noticing this "crack" forming in my chest. I first noticed it when I drank something cold: it felt as if I had spilled ice water down my chest, and I could turn the sensation on and off just by tightening or relaxing my chest muscles. Earlier, I could also temporarily tolerate environments with bad air quality by tightening those same muscles. My interpretation is that compressing the tissue closed gaps that exposures were otherwise getting through. I have wondered whether this is partly why so many people with these illnesses slouch: not only because of spinal laxity, but because compressing the chest may close tissue gaps in the airway. It sounds weird, but it's the best way I can describe it. It was something unusual that I had never felt before in my life, and I knew there had to be some clinical meaning behind it. As time went on, I could tell this exposed area in my chest was growing. Noticing it took careful attention to subtle sensations of blood flow, very mild burning, weakness when breathing, and more. I found that other members of the community, even some with very different onsets from mine, described some of these bizarre symptoms exactly the way I would (see the patient testimony below).

Own observation

My thymus

Publicly available healthy control

“Normal” thymus from internet

The two scans are shown at their own scale, not matched to each other: they were taken at different fields of view, and only the left-hand scan carries a scale bar.

“Anyone else have an empty feeling deep in the middle of their chest. Like something is missing? Like a large area just isn’t there anymore?” Patient testimonyApril 14, 2025

The chest CT scan I had in 2017 shows the region (E7) where I have always felt the root of my illness lies, and it sits right beside the central airways, where I would expect the effects of repeated airway injury to be greatest. Although the radiologist reviewing the scan deemed it normal, the area where the thymus sits in the retrosternal space looks, in my opinion, very compressed relative to healthy scans. Because I'm convinced that damage to this area is at the heart of my illness, I keep coming back to this scan; I know for certain that something is not right there. I think compression here could affect the heart and great vessels next to it. The area is densely innervated by the vagus nerve and sits adjacent to the airway and to the vessels that carry blood out of the heart.

Entry of abrasive nanoparticles and connective tissue failure

A — Rats exposed to titanium dioxide (heart muscle)

B — Long COVID, one day after post-exertional malaise (skeletal muscle)

Published research A

Own observation

B

Stage 3 of this model is the separation of tissue by abrasive nanoparticles, which causes blood vessel leakage and perforation of collagen bundles. Because dermal manifestations are the easiest to track and image without an invasive biopsy, I frequently took photographs of the changes that followed major exposures and health changes. This is yet another reason to acknowledge that these illnesses are systemic: doing so allows us to acquire cheap and concrete data sets. After the painted room, the changes to my skin were rapid and shocking: my skin became pale, striated, slow to heal and almost unrecognizable compared with what I had seen every day for the prior 30 years of my life.

Petechiae. The red dots shown (E10) are not a rash — they're not raised, not itchy, and only triggered by mechanical means. I believe what we are seeing here is blood. Titanium dioxide nanomaterials have been shown to cause endothelial cell leakiness by disrupting VE-cadherin, and they are able to penetrate cell junctions because of their small size. This results in blood vessel leakage. Under normal circumstances of daily activities, I never saw them. They were only brought on by two things: excessive vasodilation from heat (hot showers), or the physical pressure of lying on my arms while I was asleep on my side, which is the mechanical-stress reading of post-exertional malaise set out on the Model page. No other types of exposures triggered them. Diet didn't trigger them. Anti-inflammatories and allergy medications didn't get rid of them. The mechanical nature of their presence, in my view, necessitates strong consideration of the mechanical nature of this disease.

Ligament laxity. Signs of this particulate exposure show up in deeper tissue as well. Once the fibers within collagen and elastin bundles separate and lose their alignment, interstitial pockets enlarge and become places for waste and contaminants to collect. That separation is also my working explanation for the ligament laxity described on the Personal Journey page — acquired structural damage to connective tissue from environmental causes, not genetic ones (E12). This goes against all current medical knowledge on the subject of connective tissue. If a patient is suspected of having a connective tissue disease, they are given either an antibody test for autoimmune disease or a genetic test to identify genetic connective tissue disorders. There is no clinical test for acquired connective tissue disease from environmental causes, nor is there even any test for directly observing it. TEM of collagen structure was once used more widely in diagnosis, but it is now rarely used clinically and has largely given way to genetic testing, which is cheaper, quicker, and less open to varying interpretations.

Endothelial leakiness. Setyawati et al. (2013) exposed human blood-vessel cells to titanium dioxide nanoparticles 23.5 nm across and found the cell layer became leaky within 30 minutes, at concentrations as low as 10 µM. Titanium dioxide particles about 30 times larger had no effect. What stands out is how it happened. The leakiness began before the cells had taken up any meaningful number of particles, and blocking uptake didn't prevent it. The authors propose that, simply because they are so small, the nanoparticles migrate into the adherens junctions between endothelial cells and disrupt them directly. They showed the particles binding VE-cadherin, the protein that holds neighboring cells together, and that VE-cadherin was then broken down. In mice, a small subcutaneous dose increased dye leakage from blood vessels (E9A). This is the kind of mechanical, particle-driven damage stage 3 of my model describes: no toxicity or immune response is needed, only particles small enough to get between cells and pry them apart. The authors note that their mouse experiments used artificial delivery routes, and that reaching these doses in people would take extreme, long-term exposure across possibly damaged barriers. That is exactly the situation this model describes.

The same paper argues this is mechanical rather than oxidative: reactive oxygen species only rose significantly 24 hours after exposure, well after the leakiness itself appeared, across every concentration tested (Setyawati et al., 2013).

This same junction protein is not only vulnerable to titanium dioxide. SARS-CoV-2 has been shown to disrupt VE-cadherin directly, which is one of the reasons I think a virus and a nanoparticle can produce the same downstream barrier failure by different routes (E3).

Muscle fiber separation. The same exposure produced striations deep in the skin (E11) shortly afterward. Titanium dioxide has been shown to cause separation of muscle fibers through disruption of junction proteins. In 2024 a muscle-biopsy study of people with Long COVID found strikingly similar changes after a crash induced by exercise (E8B): large areas of necrotic fibers, small atrophic fibers and infiltration by immune cells, worsening between a baseline biopsy and one taken the day after post-exertional malaise (Appelman et al., 2024). Healthy controls showed some necrosis after the same exercise too, and the paper did not find the difference between groups statistically significant; what stands out is how much more common it was in the Long COVID group after the crash. The resemblance to titanium dioxide damage in animal tissue (E8A) is what led me to propose, in February 2024, that nanoparticles help drive post-exertional malaise. The comparison has clear limits: A is rat heart muscle after a controlled dose, B is human skeletal muscle, and the Long COVID study did not measure titanium dioxide or any other particle. It shows the same pattern of damage, not the same cause. Testing muscle biopsies like these for nanoparticles is the obvious next step.

Capillaroscopy. This capillaroscopy (E9B) is my own. I bought a microscope and examined my own nailbed with it, several times a day, over a period of six months to a year. That repeated-measures approach is the difference from a single clinical snapshot: it is the same patient, imaged many times, rather than one image taken once. The enlarged capillaries and microhemorrhages I see are not constant — only a small fraction are affected at any one look — but across that many looks, they are consistently there.

Those dark spots (E9B) are the finding I find most interesting. They could be hemosiderin from vessel damage, or they could be the microclots themselves. They are slightly larger than the microclots seen in blood testing, which would fit if clots continue to grow before becoming visible in the nailbed.

Two more physical signs from the same period, covered in more detail on the Personal Journey page: new fingernail changes appearing with each major step-down, and a sudden joint hypermobility with no injury behind it.

Where does injected titanium dioxide actually go? One of the older studies in the reference list injected it into animals and then measured where it ended up — the highest concentration by a wide margin was in the abdominal lymph nodes, well above any other organ tested. That is consistent with what I'd expect if lymph nodes are where clearance bottlenecks: material that cannot be filtered out efficiently accumulates exactly where filtering is supposed to happen.

Histology & Imaging

Nerve biopsy. The standard clinical practice for small fiber neuropathy (SFN) testing is highly flawed. Standard neuropathy tests count the nerve density in a skin punch biopsy. Density is a relative measure, not an absolute one: if the surrounding tissue structure is damaged and the total volume of skin shrinks along with the nerve endings, the density stays consistent, so the absolute count can fall while the relative density stays the same and reads as normal. Even then, the standard clinical biopsy will only tell you whether a nerve is dead — it says nothing about nerve compression, myelin damage, improper migration, or mitochondrial swelling. In other words, the standard clinical test is a measure of nerve death, not of damage or dysfunction. So I decided to take matters into my own hands, and in 2023 I had a 3 mm punch biopsy taken 3 cm above the medial side of my left elbow and imaged with transmission electron microscopy. I wanted to see what these peripheral nerves looked like down to the ultrastructural level, which would show the types of damage and dysfunction that the standard SFN tests would not, in addition to histological changes in other cell types.

Corneal imaging. In 2019, confocal microscopy of my corneas showed neuropathy: neuromas, areas of reduced nerve density, and inflammatory dendritic (Langerhans) cells clustered along the nerves (E13), and I was diagnosed with neuropathic corneal pain. Yet a standard small fiber neuropathy skin biopsy near my ankle came back normal. That is further proof that our current neuropathy testing is flawed. Beyond the problems with density measurement I've already described, there is a simpler one: if neuropathy is patchy, or confined to specific tissue types or locations, a single biopsy taken near the ankle will miss it entirely.

It took me four and a half years, but I was finally able to get the TEM imaging I had always wanted through clinical means. However, I ran into a problem: a lack of control images. Even if I found some abnormalities, I had no idea whether they were medically relevant, because virtually everyone has some degree of histological variation in their samples. I needed input from someone who sits at a microscope and looks at these types of images day in and day out. I sent these images to dozens of nerve morphologists to try to find someone who could give input. Several confirmed the abnormality independently. Two of their assessments:

There is some sort of breakdown of the Schwann cell cytoplasm. The dark material in the cytoplasm is from breakdown. It is abnormal. Dr Mary Bartlett BungeUniversity of Miami

The white clumps with black dots are most likely glycogen depositions. The inclusions you see are in myelinating Schwann cells. They are not normally found in Schwann cells, but can be found in some pathological conditions. Dr Angelo QuattriniOspedale San Raffaele

On the collagen (E15): fibril diameter and fibril counts were normal (I counted and measured them myself using ImageJ software), but the bundles were separated and disorganized. A synthesis defect would be expected to change the fibrils themselves. Normal fibrils in disordered bundles read to me as mechanical disruption after the fact.

A second, independent observation

I had actually seen this exact pattern once before, years earlier, and never made it public until now. While working in ME/CFS research — before I had this illness myself — I looked at a small, limited number of tissue samples from other ME/CFS patients. Collagen diameter, calcium deposits, and signs of degradation all came back normal. The one abnormality, in every sample, was the same one I would later find in my own biopsy: collagen bundles that should be held tightly together were separated and distorted in direction, more so than in the healthy samples I compared them against. I want to be precise about what this is and isn't — it's my own reading of a small unpublished sample, not a peer-reviewed case-control study, and it predates my own diagnosis, so it isn't self-referential in the way my own biopsy is. But it is a second, independent observation of the same specific abnormality, in different patients, made before I had any stake in finding it.

Three things stand out to me. Glycogen deposits and Nissl substance in mass, pressing on nerves. Pockets of demyelination that do not look like MS or dense degeneration, and look more like puncture or tearing. And glycogen deposits in keratinocytes, the cells that form the epithelial barrier.

The peripheral nervous system does have a normal feature that produces myelin pockets — Schmidt-Lanterman incisures, which are stable, funnel-shaped openings that appear in every healthy nerve. I want to be upfront that these could just be that. But in my own view the openings I see do not look stable or uniformly organized the way incisures typically do, which is why I read them as mechanical damage rather than a normal structural feature. That is my interpretation, not a settled morphological finding.

The dense particles in the third image are the least settled finding here (E16). Size, shape and composition all affect how a particle interacts with a given cell type, and I have not been able to get compositional analysis done. I am not claiming these are titanium dioxide. I am saying they warrant testing.

Entry and lodging of fibers and polymers

Stage 4 of the model is about substances that are able to move through the bloodstream, but deposit and accumulate in the body: fibers and polymers from plastics, paint, adhesives, mulch and other substances with high adhesive properties and/or irregular shapes. Every day we breathe in fibers shed by bed sheets, carpets, adhesives, paint and plastics, which float through indoor air in large numbers. Research has already shown substances like microplastics can accumulate in the brain from inhalational exposures (Amato-Lourenço et al., 2024; Chen et al., 2025), and we know this is a phenomenon that affects everyone.

Plastics in the brain. The strongest published evidence that plastics accumulate in the human brain comes from Nihart et al. (2025) in Nature Medicine. They measured microplastics in liver, kidney and frontal-cortex tissue from people who had died, using pyrolysis gas chromatography–mass spectrometry. The brain held far more plastic than the other organs: a median of about 4,900 µg per gram in 2024 samples, against roughly 430 µg/g in liver and 400 µg/g in kidney. Brain concentrations rose significantly between 2016 and 2024, and about three-quarters of the plastic in brain tissue was polyethylene. Brains from people with dementia held considerably more (a median of about 26,000 µg/g), with notable deposits in the walls of the brain's blood vessels and in immune cells.

The authors stress, though, that the findings are associative and don't establish that plastics cause disease. The particles they imaged were mostly nanoscale shards about 100–200 nm long, not long fibers. They measured plastics, not paint, adhesive or other polymer sources, and they didn't measure perfusion, intracranial pressure or cognition. So the study supports the first step, that plastics reach and accumulate in the brain, including in vessel walls. The clogging, and what it does to perfusion and drainage, is my model's interpretation.

Lymphatic vessels. Dr Raymond Perrin, an osteopath who has spent decades examining people with ME/CFS, describes visibly distended lymphatic vessels on the upper chest as one of the illness's physical signs. He attributes them to lymph backing up behind impaired drainage. If lymphatic vessels and nodes are distended, they are distended with something. The usual assumption is fluid or cell debris. I think it would take material that is thick, sticky or rigid to physically obstruct them for years, and fibers such as plastics and wood pulp are the kind of material I would look for. That isn't far-fetched: lymph nodes are known to trap insoluble particles the body can't break down. Tattoo pigment, including titanium dioxide nanoparticles, accumulates in the nearby nodes of tattooed people (Schreiver et al., 2017), and silicone from ruptured implants does the same (Zambacos et al., 2013). With plastics now found in almost every human tissue examined, including the brain, it is worth checking what these vessels and nodes contain.

When this sign was tested, though, in Hives et al., two examiners almost never agreed on whether it was present (κ = 0.03), and it did not improve diagnostic accuracy. The two most useful signs were thoracic spine changes and a tender celiac plexus. No one has analyzed what is inside these vessels or nodes in ME/CFS, so the material is my hypothesis. The test is straightforward: polymer analysis (pyrolysis GC-MS, or Raman/FTIR microscopy) of lymph node tissue from people with ME/CFS compared with controls.

Patients could see that something was wrong with their lymphatics years before research took an interest: engorged vessels like these are visible to the naked eye (E17). New sensing and imaging tools are welcome, but each measures one narrow thing. They can't tell us whether lymphatic vessels are compressed, engorged, malformed, clogged, stretched, distorted or leaking, or what they contain. Only tissue samples examined under the microscope can answer those questions. Sensors are good for deciding where to look; they can't replace looking. And as long as research focuses only on viruses and biological pathways, a physical blockage like this will never be found.

Systemic effects

Microclots. Amyloid-fibrin microclots have become a major focus in Long COVID and ME/CFS research (E18). My question is whether nanoparticles such as titanium dioxide could be nucleating them. No one has tested that directly, but each step has published support.

The first step is binding. Fibrinogen, the protein that forms clots, binds titanium dioxide strongly. Both fibrinogen and albumin adsorb irreversibly to titanium dioxide nanoparticles, even against electrostatic repulsion, probably changing shape or orientation as they do (Marucco et al., 2013). Fibrinogen adsorbed to titanium dioxide gradually loses part of its α-helical structure (Yongli et al., 1999), and how it unfolds depends on the particle's crystal phase, surface charge (Zhao et al., 2014) and even crystal facet (Struczyńska et al., 2023). A 2021 University of Akron honors research project found fibrinogen bound with similar affinity to titanium dioxide nanotubes and nanoparticles of different shapes, while albumin bound poorly (Miller, 2021).

The second step is nucleation. Amyloid forms slowly until a seed appears, then quickly. Titanium dioxide nanoparticles shorten that seeding step for β-amyloid (Wu et al., 2008), and stimulate the aggregation and fibrillation of α-synuclein (Slekiene et al., 2022), the proteins behind Alzheimer's and Parkinson's plaques. The proposed mechanism is that the particle surface concentrates the protein and helps it change shape. A single particle carrying many partly unfolded fibrinogen molecules is the same arrangement, applied to a different protein.

Titanium dioxide also acts on the cells involved in clotting. It raises calcium in human platelets and triggers their activation, aggregation and procoagulant activity (Bian et al., 2024), and it makes red blood cells expose phosphatidylserine, a surface that accelerates clotting (Bian et al., 2021).

What hasn't been shown: no study I know of has tested whether titanium dioxide nucleates fibrin amyloid, or looked for titanium inside microclots. The nucleation work is on β-amyloid and α-synuclein, not fibrin. And some binding studies find that tightly adsorbed fibrinogen can hide its platelet-binding sites, which would reduce clotting rather than drive it (Zhao et al., 2014; Struczyńska et al., 2023). So this is a hypothesis with plausible steps. The direct test, looking for particles at the center of microclots, is what's missing.

Own observation

Other results flagged low

A My urine organic acid panel: markers also measured in panel B (top), then every other result flagged low

From published research

Control 4 days 1 month 2 months

B Rat urine NMR, titanium dioxide

• Citrate: lower than control by 2 months in TiO₂-exposed rats; below range in mine (112, range 143–765)
• Succinate: lower than control by 2 months in TiO₂-exposed rats; below range in mine (0, range 2–12)
• α-Ketoglutarate: lower than control by 2 months in TiO₂-exposed rats; 0 in mine (the lab range has no lower limit, so it isn't flagged)

Organic acids. My organic acid panel came back chronically low across the board: succinic, methylsuccinic, glutaric, malic, adipic, pyruvic, isocitric and citric all below reference range, several at zero (E19). That isn't only a general sign of the low metabolic activity already described in ME/CFS. It also resembles the urinary profile of rats exposed to titanium dioxide nanoparticles. Valentini et al. (2019) injected rats with titanium dioxide nanoparticles and analyzed their urine by NMR spectroscopy at four days, one month and two months. Citrate, α-ketoglutarate and succinate decreased — at the two highest doses by four days, and at the lowest dose, the 0.5 g/kg group shown in E19, progressively over one and two months — alongside rises in markers of oxidative stress and liver and kidney strain. Citrate and succinate are both low on my panel. Bu et al. (2010), dosing rats orally, found pyruvate reduced in serum, as it is on my panel, although in that study urinary citrate rose rather than fell, so the published picture isn't uniform. Those rats were dosed by injection or by mouth, not by inhalation, and an NMR urine profile in rats isn't the same test as a human organic acid panel, so this is a resemblance in direction, not a measured match. It is still the closest published metabolic data I've found to what my own testing showed.

Skin fragility. I took this photograph (E20) in 2023, the day after the skin biopsy for my electron microscopy (E14). The biopsy site had been covered with an ordinary adhesive bandage for 24 hours. When I took it off, the skin where the adhesive had touched came away with it. That isn't normal for me. I've worn hundreds of bandages over my life, often for much longer than a day, and never had a problem.

This is the photograph I show people who tell me this illness is a problem of neural sensitization. Nothing about a sensitized nervous system removes skin, and no psychological process does either. This is an acquired chemical vulnerability with a visible physical result, from a product I had used my whole life without incident.

Air quality and pain. This one isn't my own data. Jones, Haskin and Younger (2024) followed 58 women with ME/CFS who rated their pain and fatigue every day for an average of 62 days. The researchers then matched those ratings against local weather and air quality records collected after the study ended, so participants didn't know what was being tested. Of the 14 weather and air quality measures examined, seven were significantly related to pain or fatigue.

The strongest link was between air quality and pain: on days with a worse Air Quality Index, pain was higher (p < 0.001). Pain was also higher on days with more particulate matter and lower wind speed. Fatigue did not track the overall Air Quality Index, but it was higher on days with more particulates and lower barometric pressure. Not every pollutant pointed the same way: pain was higher on days with lower carbon monoxide, sulfur dioxide and ozone. The authors describe these effects as small, apart from the link between the Air Quality Index and pain. In the figure (E21), the pain curve rises in the "unhealthy for sensitive groups" range, but the uncertainty band around it is wide.

Other patients describe the same experiences in their own words (E22).

Time and Exposure Order Influences

This illness isn't a fixed state. It progresses, and the effects depend on time. Damage accumulates, and each exposure lands on tissue already changed by the ones before it, so the same exposure can do more harm later in the illness than it would have at the start. In my case the decline wasn't gradual. It came in step-downs: sharp drops in function after particular exposures that never fully recovered. Some of those step-downs left visible marks. My nails changed three times over six years (E23): Terry's nails and clubbing in 2017, ridging and opaqueness in 2021, and cuticle loss in 2023. Each change appeared with a step-down and persisted afterward. Taken together, they record a disease that moved forward in stages rather than holding still.

References

These are the published studies the model draws on. They are real and independent of me; the synthesis and the application to ME/CFS are my own.

1. Ren B, Wu Q, Muskhelishvili L, et al. Evaluating the sub-acute toxicity of formaldehyde fumes in an in vitro human airway epithelial tissue model. Int J Mol Sci. 2022;23(5):2593. ↩
2. Deinhardt-Emmer S, Böttcher S, Häring C, et al. SARS-CoV-2 causes severe epithelial inflammation and barrier dysfunction. J Virol. 2021;95(10):e00110-21. ↩ ↩ ↩
3. Baalbaki N, van Egmond D, Jaeger P, et al. Barrier dysfunction in nasal epithelium contributes to persistent inflammation in long COVID. J Allergy Clin Immunol. 2026;157(1):203–16. ↩ ↩
4. Deinhardt-Emmer S, Wittschieber D, Sanft J, et al. Early postmortem mapping of SARS-CoV-2 RNA in patients with COVID-19 and the correlation with tissue damage. eLife. 2021;10:e60361. ↩
5. Liu F, Xia Y, Lee ACY, et al. Prolonged dysregulation and pathological changes in the upper respiratory tract of SARS-CoV-2 infected hamsters. npj Viruses. 2026;4:15. ↩
6. Yazici D, Cagan E, Tan G, et al. Disrupted epithelial permeability as a predictor of severe COVID-19 development. Allergy. 2023;78(10):2644–58. ↩
7. Epithelial Barrier Theory. epithelialbarriertheory.com. Swiss Institute of Allergy and Asthma Research (SIAF), University of Zurich. ↩
8. Akdis CA. Does the epithelial barrier hypothesis explain the increase in allergy, autoimmunity and other chronic conditions? Nat Rev Immunol. 2021;21(11):739–51. ↩
9. Wang M, Tan G, Eljaszewicz A, et al. Laundry detergents and detergent residue after rinsing directly disrupt tight junction barrier integrity in human bronchial epithelial cells. J Allergy Clin Immunol. 2019;143(5):1892–1903. ↩
10. Xian M, Ma S, Wang K, et al. Particulate matter 2.5 causes deficiency in barrier integrity in human nasal epithelial cells. Allergy Asthma Immunol Res. 2020;12(1):56–71. ↩
11. Donkers JM, Höppener EM, Grigoriev I, et al. Advanced epithelial lung and gut barrier models demonstrate passage of microplastic particles. Microplastics and Nanoplastics. 2022;2(1):6. ↩
12. Sun N, Ogulur I, Mitamura Y, et al. The epithelial barrier theory and its associated diseases. Allergy. 2024;79(12):3192–3237. ↩
13. Huitsing K, Tritsch T, Carrera Arias FJ, et al. The potential role of ocular and otolaryngological mucus proteins in myalgic encephalomyelitis/chronic fatigue syndrome. Mol Med. 2024;30(1):1. ↩
14. Cunha BA. Crimson crescents — a possible association with the chronic fatigue syndrome. Ann Intern Med. 1992;116(4):347. ↩
15. Loebel M, Grabowski P, Heidecke H, et al. Antibodies to β adrenergic and muscarinic cholinergic receptors in patients with Chronic Fatigue Syndrome. Brain Behav Immun. 2016;52:32–9. ↩
16. Liu HR, Zhao RR, Zhi JM, Wu BW, Fu ML. Screening of serum autoantibodies to cardiac β1-adrenoceptors and M2-muscarinic acetylcholine receptors in 408 healthy subjects of varying ages. Autoimmunity. 1999;29(1):43–51. ↩
17. Markousis-Mavrogenis G, Minich WB, Al-Mubarak AA, et al. Clinical and prognostic associations of autoantibodies recognizing adrenergic/muscarinic receptors in patients with heart failure. Cardiovasc Res. 2023;119(8):1690–705. ↩
18. Cabral-Marques O, Riemekasten G. Functional autoantibodies targeting G protein-coupled receptors in rheumatic diseases. Nat Rev Rheumatol. 2017;13(11):648–56. ↩
19. Scherbaum I, Heidecke H, Bunte K, Peters U, Beikler T, Boege F. Autoantibodies against M5-muscarinic and beta1-adrenergic receptors in periodontitis patients. Aging (Albany NY). 2020;12(16):16609–20. ↩
20. Hall J, Bourne KM, Vernino S, et al. Detection of G protein-coupled receptor autoantibodies in postural orthostatic tachycardia syndrome using standard methodology. Circulation. 2022;146(8):613–22. ↩
21. Sfera A, Osorio C, Zapata Martín del Campo CM, et al. Endothelial senescence and chronic fatigue syndrome, a COVID-19 based hypothesis. Front Cell Neurosci. 2021;15:673217. ↩
22. El-Bestawy EM, Tolba AM. Effects of titanium dioxide nanoparticles on the myocardium of the adult albino rats and the protective role of β-carotene. J Mol Histol. 2020;51(5):485–501. ↩
23. Appelman B, Charlton BT, Goulding RP, et al. Muscle abnormalities worsen after post-exertional malaise in long COVID. Nat Commun. 2024;15:17. ↩ ↩
24. Setyawati MI, Tay CY, Chia SL, et al. Titanium dioxide nanomaterials cause endothelial cell leakiness by disrupting the homophilic interaction of VE-cadherin. Nat Commun. 2013;4:1673. ↩ ↩ ↩ ↩
25. Huggins CB, Froehlich JP. High concentration of injected titanium dioxide in abdominal lymph nodes. J Exp Med. 1966;124(6):1099–106. ↩
26. Amato-Lourenço LF, et al. Microplastics in the olfactory bulb of the human brain. JAMA Netw Open. 2024;7(9). ↩
27. Chen L, Liu Y, Li H, et al. Size-dependent pulmonary toxicity and whole-body distribution of inhaled micro/nanoplastic particles in male mice from chronic exposure. Environ Sci Technol. 2025;59(14):6993–7003. ↩
28. Nihart AJ, Garcia MA, El Hayek E, et al. Bioaccumulation of microplastics in decedent human brains. Nat Med. 2025;31(4):1114–19. ↩
29. Schreiver I, Hesse B, Seim C, et al. Synchrotron-based ν-XRF mapping and μ-FTIR microscopy enable to look into the fate and effects of tattoo pigments in human skin. Sci Rep. 2017;7:11395. ↩
30. Zambacos GJ, Molnar C, Mandrekas AD. Silicone lymphadenopathy after breast augmentation: case reports, review of the literature, and current thoughts. Aesthetic Plast Surg. 2013;37(2):278–89. ↩
31. Hives L, Bradley A, Richards J, et al. Can physical assessment techniques aid diagnosis in people with chronic fatigue syndrome/myalgic encephalomyelitis? A diagnostic accuracy study. BMJ Open. 2017;7(11):e017521. ↩
32. Marucco A, Fenoglio I, Turci F, Fubini B. Interaction of fibrinogen and albumin with titanium dioxide nanoparticles of different crystalline phases. J Phys Conf Ser. 2013;429:012014. ↩
33. Yongli C, Xiufang Z, Yandao G, Nanming Z, Tingying Z, Xinqi S. Conformational changes of fibrinogen adsorption onto hydroxyapatite and titanium oxide nanoparticles. J Colloid Interface Sci. 1999;214(1):38–45. ↩
34. Zhao AS, Zhou S, Wang Y, Chen J, Ye CR, Huang N. Molecular interaction of fibrinogen with thermally modified titanium dioxide nanoparticles. RSC Adv. 2014;4(76):40428–34. ↩ ↩
35. Struczyńska M, Firkowska-Boden I, Levandovsky N, Henschler R, Kassir N, Jandt KD. How crystallographic orientation-induced fibrinogen conformation affects platelet adhesion and activation on TiO₂. Adv Healthc Mater. 2023;12(13):2202508. ↩ ↩
36. Miller M. Titanium dioxide nanoparticles binding to human blood plasma proteins. Williams Honors College, Honors Research Projects 1361. University of Akron; 2021. Undergraduate honors research project, not peer-reviewed. ↩
37. Wu WH, Sun X, Yu YP, et al. TiO₂ nanoparticles promote β-amyloid fibrillation in vitro. Biochem Biophys Res Commun. 2008;373(2):315–18. ↩
38. Slekiene N, Snitka V, Bruzaite I, Ramanavicius A. Influence of TiO₂ and ZnO nanoparticles on α-synuclein and β-amyloid aggregation and formation of protein fibrils. Materials. 2022;15(21):7664. ↩
39. Bian Y, Jin Q, He J, et al. Biomedical application of TiO₂NPs can cause arterial thrombotic risks through triggering procoagulant activity, activation and aggregation of platelets. Cell Biol Toxicol. 2024;40(1):67. ↩
40. Bian Y, Chung HY, Bae ON, Lim KM, Chung JH, Pi J. Titanium dioxide nanoparticles enhance thrombosis through triggering the phosphatidylserine exposure and procoagulant activation of red blood cells. Part Fibre Toxicol. 2021;18:28. ↩
41. Valentini X, Rugira P, Frau A, et al. Hepatic and renal toxicity induced by TiO₂ nanoparticles in rats: a morphological and metabonomic study. J Toxicol. 2019;2019:5767012. ↩ ↩
42. Bu Q, Yan G, Deng P, et al. NMR-based metabonomic study of the sub-acute toxicity of titanium dioxide nanoparticles in rats after oral administration. Nanotechnology. 2010;21(12):125105. ↩
43. Jones CL, Haskin O, Younger J. Association Between Chronic Pain and Fatigue Severity with Weather and Air Pollution Among Females with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). Int J Environ Res Public Health. 2024;21(12):1560. ↩ ↩