Mold Illness and CIRS: Diagnosis, Evidence, and Treatment Sequencing

A Symptom Picture That Overlaps With Everything Else

The symptom picture, fatigue, brain fog, word-finding trouble, headaches, joint pain, sinus congestion, static shocks, temperature dysregulation, and a tendency to feel worse in damp or poorly ventilated spaces, overlaps heavily with depression, chronic fatigue syndrome, fibromyalgia, hypothyroidism, and perimenopause. That overlap is exactly why this gets missed for years: the symptoms are real, but a standard metabolic panel, CBC, and TSH will usually look unremarkable, so the workup stalls unless someone knows to look further.

The standard first step is Visual Contrast Sensitivity (VCS) testing, a cheap, fast screening tool. It wasn't originally developed as a CIRS-specific test, it was built to detect subtle visual cortex impairment from neurotoxic exposures more broadly, but it's abnormal in a majority of CIRS patients and gives a reason to move forward with more expensive labs rather than order the full panel on every fatigued patient who walks in.

From there, the core biomarker panel is C4a (a complement activation marker, often dramatically elevated), MMP-9 (a tissue-degrading enzyme tied to blood-brain barrier permeability, elevated), TGF-beta1 (often elevated), VEGF (dysregulated, sometimes high, sometimes paradoxically low), and MSH and VIP themselves (both often suppressed, since they're the regulatory neuropeptides being consumed by the ongoing inflammatory drive). This can be paired with environmental confirmation when the exposure history is unclear: ERMI dust sampling for the building itself (a value above 2 suggests elevated mold burden), HERTSMI-2 as a faster five-species screen (above 11 flags a high-risk environment), and creatinine-corrected first-morning urine mycotoxin testing to confirm actual body burden rather than just environmental presence. HLA-DR genotyping rounds out the picture as supporting context, per the caveat above.

Each Step Gated by the Biomarker Before It

The order matters here as much as the individual steps, because each step is gated by whether the biomarker it depends on has actually normalized. The sequence starts with removal from the exposure, since nothing downstream works reliably while the trigger is ongoing. Next comes binding and eliminating the toxin itself, along with support for the detoxification pathways that clear it, glutathione repletion is central here, since mycotoxins directly deplete glutathione stores. If there's confirmed sinonasal fungal colonization, antifungal therapy gets layered in at this stage as well.

Binding works by intercepting the toxin in the gut before it's reabsorbed. Mycotoxins are excreted into bile and then substantially reabsorbed in the intestine rather than leaving the body, so a binder that captures the toxin in the gut lumen redirects it into stool instead of sending it back into circulation. Cholestyramine has the strongest data behind that underlying mechanism. It's FDA-approved as a bile acid sequestrant, and a controlled human trial in workers exposed to the pesticide chlordecone, a different lipophilic toxin that recirculates through the same bile pathway, found it increased fecal excretion of the toxin sevenfold and roughly halved its elimination half-life from blood and fat (Cohn et al., 1978). That trial wasn't done in mycotoxin-exposed patients, so cholestyramine's use in CIRS specifically is an extrapolation from that mechanism plus Shoemaker's case series, not a mycotoxin-specific human trial. Welchol (colesevelam) is a gentler alternative when cholestyramine isn't tolerated, though its binding capacity for these specific toxins is less well characterized.

Chlorella has the most directly relevant data of the non-prescription binders. A 2025 study found that Chlorella sorokiniana reduced levels of three major mycotoxins, deoxynivalenol, ochratoxin A, and fumonisin B1, both in vitro and in an animal model (Yamaguchi et al., 2025), a meaningfully different evidence tier than the chlordecone rat data most chlorella-for-mold references actually cite. It's still animal and in vitro work rather than a human trial, so it's supportive, not proven, for mycotoxin binding in people. Other non-prescription binders in common use, activated charcoal, bentonite or zeolite clay, and Saccharomyces boulardii, work on the same adsorption principle but don't have mycotoxin-specific data behind them at all, human or animal.

Dry and infrared sauna support elimination through sweat and can help with overall detox load. They're an adjunct to binding rather than a substitute for it, since sweat isn't a major clearance route for mycotoxins the way it is for some other toxin classes.

Only after that does the neuroinflammatory and immune-regulation phase begin, following the biomarkers in sequence: C4a first, then MMP-9, then VEGF. The reasoning is mechanistic, not procedural for its own sake: MMP-9 elevation reflects active blood-brain barrier permeability and ongoing tissue-degrading enzyme activity, so any immunomodulatory therapy introduced while that barrier is still compromised risks amplifying the inflammatory cascade rather than resolving it.

What This Diagnosis Doesn't Yet Have

CIRS is not currently recognized in the ICD-10-CM and isn't endorsed by mainstream allergy, immunology, or infectious disease societies. Most of the foundational literature comes from one research group, and two of three authors of the most comprehensive independent treatment review disclosed paid expert-witness work in CIRS litigation, worth knowing when weighing the evidence base's independence. The widely cited "25% of the population" HLA-DR susceptibility figure is Dr. Shoemaker's own analysis of registry data, not an independently replicated population study, and should be treated as a clinical heuristic rather than a fixed number. No large, multi-center, independently replicated randomized trial has yet established that the full sequential protocol outperforms other approaches to unexplained fatigue and inflammation.