Immune and Neuroimmune Probes
1 Cimetidine
Cimetidine uniquely blocks H2 receptors on CD8+ suppressor T cells, removing an inhibitory signal and enhancing cell-mediated immunity. Also inhibits CYP450 (1A2, 2D6, 3A4). Famotidine lacks these properties — a differential that cleanly isolates cimetidine’s immunomodulatory mechanism. Cimetidine response defines the VIM (viral-immune-metabolic) phenotype.
1.1 If cimetidine works
Cimetidine blocked H2 receptors on suppressor T cells → suppressor signal removed → T-cell function enhanced → immune surveillance improved → symptoms improved.
- Certainty
- Low to Medium — H2 receptor on CD8+ suppressor T cells is well-characterized in immunology; cimetidine’s T-cell effect is documented in oncology. But no ME/CFS-specific study measured T-cell function before/after.
- Does NOT tell us
- why T-cells were suppressed — adaptive immune regulation, viral immune evasion, or primary immunodeficiency.
- Level of action
- Partial root cause — corrects T-cell dysfunction amplifier.
Enhanced T-cell function improved immune control of latent herpesviruses. CYP450 inhibition may potentiate co-administered antivirals.
- Certainty
- Low — cimetidine has no direct antiviral activity; inference is indirect (immune enhancement → viral control).
- Level of action
- Partial root cause.
Famotidine blocked H2 (acid suppression only) → no benefit. Cimetidine blocked H2 + enhanced T-cells → benefit. The differential isolates cimetidine’s unique immunomodulatory property.
- Certainty
- Medium-High — pharmacological fact: famotidine lacks immunomodulatory and CYP450 effects.
- Level of action
- Classificatory — confirms VIM phenotype.
1.2 What a positive response does NOT reveal
A positive cimetidine response confirms that T-cell-mediated immune dysfunction was present but does not reveal why the T cells were suppressed — adaptive immune regulation, viral immune evasion, or primary immunodeficiency all remain possible. It does not directly demonstrate viral reactivation (cimetidine has no antiviral activity; the viral inference is indirect). Because benefit may partly reflect CYP450 inhibition altering the metabolism of co-administered drugs, and because no ME/CFS-specific study has measured T-cell function before and after treatment, a pharmacological immunomodulatory effect cannot be fully separated from confounders or placebo.
1.3 If cimetidine does NOT work
- T-cell suppression not a dominant mechanism — autoantibody, mast cell, or metabolic dysfunction may predominate.
- Suppressor T-cell blockade insufficient — T cells may be too exhausted (PD-1+, TIM-3+) to respond even when the H2 brake is removed.
- CYP450 drug interaction context — benefit or harm may reflect altered metabolism of co-administered drugs.
1.4 Key caveat
Non-response does not exclude T-cell dysfunction. T cells may be too exhausted (PD-1+, TIM-3+) to respond even when the H2 brake is removed, so the mechanism can still be present while cimetidine fails to correct it. Non-response is weaker evidence than response — it makes T-cell-mediated pathology less likely but does not rule it out.
1.5 How cimetidine combines with other medications
- Cimetidine works + LDN does not → VIM phenotype distinct from neuroinflammatory phenotype.
- Cimetidine works + valacyclovir works → complementary immune-mediated + pharmacological viral suppression.
- Cimetidine works + pyridostigmine works → viral-immune-autonomic phenotype.
1.6 Compendium
The full pharmacodiagnostic entry — including mechanism-exclusion logic, dose-specific side-effect diagnostic patterns, combination diagnostics, and worsening risk profiles — is at Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe (sec-12, H2 Antagonists entry).
2 Immunoglobulins (IVIG) and Rituximab
IVIG neutralizes circulating autoantibodies and modulates Fc receptors. Rituximab eliminates CD20+ B cells (autoantibody-producing precursors). Both probe whether humoral autoimmunity is contributing. Both are Category C (energy-demanding), extremely expensive, and require infusion center access.
Patient-reported signal. In the largest treatment-outcome survey to date (Eckey et al. 2025, \(n = 3{,}925\)), IVIG/subcutaneous IgG was among the most positively rated treatments overall (NAS 58.2%) and, notably, had its highest response in the multisystemic symptom cluster (73.3% positive, alongside manual lymphatic drainage at 73.9%) — the cluster with the highest symptom/comorbidity burden (Eckey et al. 2025). (Severity: survey was severity-stratified; the multisystemic cluster had the lowest capacity, 39.1% ± 17.6%.) This is patient-reported, unblinded evidence, not RCT-level, but it supports prioritizing an immune-targeting probe in multisystemic patients and is consistent with the humoral-autoimmunity hypothesis above. (Evidence class: patient-reported survey, single study.)
2.1 If IVIG or rituximab works
IVIG neutralized autoantibodies / rituximab eliminated B cells → symptoms improved → autoantibodies were pathogenic.
- Certainty
- Low to Medium — IVIG: uncontrolled series; rituximab: Phase III negative overall but subset signal in autoantibody-positive patients.
- Does NOT tell us
- which specific autoantibody is pathogenic (anti-β2AR, anti-M3R, anti-TRPM3, etc.).
- Level of action
- Partial root cause — both reduce autoantibody effects but do not eliminate the long-lived plasma cells (CD20−) that produce them. This is why rituximab responses are often transient.
Response then relapse as B cells repopulate → disease is B-cell-dependent but long-lived plasma cells maintain autoantibody production independently.
- Certainty
- Medium — Phase III rituximab data + plasma cell sanctuary model.
- Level of action
- Partial root cause — eliminates one autoantibody source (short-lived plasmablasts), spares another (long-lived plasma cells).
2.2 What a positive response does NOT reveal
A positive response confirms that humoral autoimmunity was contributing but does not identify which specific autoantibody is pathogenic (anti-β2AR, anti-M3R, anti-TRPM3, and others remain indistinguishable from the treatment response alone). It does not reveal what triggered the autoantibody production (post-infectious, molecular mimicry, or primary autoimmunity), nor does it locate the persistent source: because both drugs spare long-lived CD20− plasma cells, a response followed by relapse cannot by itself prove whether short-lived plasmablasts or long-lived plasma cells dominate ongoing production. With IVIG uncontrolled and rituximab negative in its Phase III overall, an individual response cannot be cleanly separated from placebo or natural fluctuation.
2.3 If IVIG/rituximab does NOT work
- Autoantibody-mediated pathology not dominant — rituximab Phase III was negative overall.
- Relevant autoantibody-producing cells not targeted — long-lived plasma cells (CD20−) spared; IVIG neutralization insufficient.
- Autoantibodies present but not pathogenic — detected autoantibodies may be biomarkers, not drivers.
2.4 Key caveat
Non-response does not exclude autoantibody-mediated pathology. Both drugs spare long-lived CD20− plasma cells, so autoantibodies driven from that sanctuary can persist even when the treatment is working as designed on B cells and circulating antibody. Non-response is weaker evidence than response — it makes humoral autoimmunity less likely to be dominant but does not rule it out, particularly given the energy-demanding, expensive nature of these therapies that limits adequate trials. For the limits of what a rituximab null can conclude about autoimmune origin specifically — including why four autoimmune pathways survive the null — see What Pharmacodiagnostics Can and Cannot Rule Out: Origin Hypotheses.
2.5 How IVIG and Rituximab combine with other medications
- IVIG/rituximab work + antivirals work → autoantibody-driven pathology coexisting with active viral replication; suggests a combined mechanism in which immune dysregulation both produces autoantibodies and permits viral reactivation.
- Rituximab works + relapse on B-cell repopulation → short-lived plasmablasts are sufficient to sustain disease; long-lived plasma cells are not the dominant source, and maintenance B-cell depletion (rather than plasma-cell-directed therapy) may hold benefit.
- IVIG works transiently + rituximab works longer → distinguishes plasmablast-driven from plasma-cell-driven production: transient IVIG benefit reflects neutralization of circulating antibody, while durable rituximab benefit points to plasmablast-dependent supply.
- Neither IVIG nor rituximab works → non-humoral pathology dominates; look toward neuroinflammatory, metabolic, autonomic, or viral mechanisms instead.
2.6 Compendium
The full pharmacodiagnostic entries — including mechanism-exclusion logic, dose-specific side-effect diagnostic patterns, combination diagnostics, and worsening risk profiles — are at Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe (sec-12, IVIG entry and Rituximab entry).
3 Low-Dose Aripiprazole (LDA)
LDA (1–2 mg/day) is a partial agonist at D2/D3 and 5-HT1A, antagonist at 5-HT2A. At microdose levels (≤1 mg/day), the pharmacologically dominant action is presynaptic D2/D3 partial agonism — insufficient for antipsychotic occupancy but adequate for microglial D2 receptor engagement. Carries significant risks: akathisia, serotonin syndrome, CYP2D6 variable exposure, and prediabetes/metabolic syndrome (Aripiprazole-Associated Prediabetes and Metabolic Syndrome Risk).
LDA acts at three levels relevant to ME/CFS: (1) microglial D2/D3 — suppressing pro-inflammatory cytokine release, raising the activation threshold of primed microglia; (2) mesocorticolimbic D2/D3 — restoring tonic dopamine signalling in prefrontal and hippocampal circuits; (3) 5-HT1A autoreceptors — reducing raphe serotonergic tone with downstream ANS stabilisation. The balance of these actions shifts with dose: microglial modulation dominates at ≤1 mg, postsynaptic D2 and 5-HT2A effects emerge above 1.5 mg.
3.1 If LDA works
Dopamine D2/D3 receptors are expressed on microglia. D2-like agonism suppresses pro-inflammatory cytokine release (TNF-α, IL-1β, IL-6), nitric oxide production, and reactive oxygen species from activated microglia. If LDA raises the microglial activation threshold, exertion signals (ATP, ROS, heat-shock proteins) that normally trigger a post-exertional neuroinflammatory cascade no longer cross into symptomatic territory. The drug acts as a threshold modulator, not a fuel replacement or linear anti-inflammatory. The U-shaped dose-response — benefit at 0.2–2 mg but harm at higher antipsychotic doses — is a partial agonist inverted-U: net agonism at low occupancy inverts to antagonist dominance as occupancy exceeds 50% (Sun et al. 2020). Crosby et al. (2021) reported PEM improvement at doses of 0.2–2 mg in a pilot study (n=25) (Crosby, Kalantar, and DeRisi 2021).
Synergy with mast-cell stabilisers. Mast cells release histamine, tryptase, and PGD2, which activate microglia. Antihistamines (H1, H2) and leukotriene receptor antagonists reduce the baseline pressure on microglia. LDA raises the activation threshold. The two act on the same cell population at different nodes — pressure vs. gate — and the combination may create a wider operational window than either alone.
Synergy with LDN. LDN blocks TLR4 on microglia — a separate receptor from D2 on the same cell. D2 agonism + TLR4 antagonism are complementary, not redundant: they suppress microglial activation through different signalling pathways converging on the same cytokine-output machinery. A patient may be a full LDN non-responder while LDA is effective, because TLR4 antagonism alone does not raise the activation threshold — it only removes one input among many.
- Certainty
- Low — Crosby et al. (2021) pilot data only (n=25, no control group). Microglial D2 expression and D2-agonist suppression of microglial cytokines are well-established in neuroinflammatory models, but no study has measured microglial D2 tone before/after ≤1 mg aripiprazole in any clinical population. The non-linear threshold model is mechanistically coherent with D2 partial agonism pharmacology but prospectively unvalidated.
- Does NOT tell us
- whether microglial stabilisation, dopamine tone restoration, or 5-HT1A ANS stabilisation is the dominant mechanism — all three are plausibly involved; whether the mast-cell/LDA synergy is additive or multiplicative; whether PEM protection persists beyond the first weeks (receptor adaptation may alter the net effect).
- Action
- LDA at 0.2–2 mg/day is a rational probe for microglial-mediated PEM. Combined with mast-cell stabilisers, it tests whether microglial activation threshold is the rate-limiting node in PEM generation. Do not increase dose if PEM protection is already complete — higher doses shift into postsynaptic D2/5-HT2A range and may reverse benefit via U-shaped dose-response.
- Level of action
- Partial root cause — LDA raises the microglial activation threshold without removing the trigger that primed the microglia (autoantibodies, infection history, metabolic stress).
LDA’s partial agonism at presynaptic D2 autoreceptors and postsynaptic D2/D3 receptors restores tonic dopamine tone in prefrontal and hippocampal circuits. CSF catecholamine reduction is documented (Walitt et al. 2024) and stimulant surveys report 77.1% brain fog improvement (Eckey et al. 2025), supporting a dopaminergic component to cognitive symptoms. Crosby et al. (2021) reported cognitive improvement at doses of 0.2–2 mg in a pilot study (n=25) (Crosby, Kalantar, and DeRisi 2021).
Domain dissociation is a testable diagnostic principle: if LDA fully resolves cognitive symptoms but leaves other symptoms untouched, the protected domains share a dopaminergic/microglial pathway while the resistant domain operates through a different mechanism (e.g., thalamo-cortical GABAergic filtering, vascular dysregulation). This domain dissociation is itself informative — it identifies which symptoms share a pathway and which do not, without requiring a biomarker.
- Certainty
- Low to Medium — CSF catecholamine reduction is documented (Walitt et al. 2024), stimulant surveys 77.1% brain fog improvement (Eckey et al. 2025), but LDA specifically has zero controlled ME/CFS trials beyond the Crosby pilot.
- Does NOT tell us
- whether cognitive benefit is dopaminergic, serotonergic (5-HT1A), or from improved sleep architecture; whether the deficit is primary or secondary to neuroinflammation.
- Action
- cognitive/dopaminergic support is indicated. Domain dissociation between protected cognition and persistent non-cognitive symptoms means additional targets (GABAergic, vascular) should be probed separately — do not increase LDA dose chasing symptoms in a dopamine-resistant domain.
- Level of action
- Symptom management for cognition; partial root cause if acting through microglial stabilisation upstream of cognitive circuits.
LDA is a 5-HT1A partial agonist. 5-HT1A activation in raphe nuclei reduces serotonergic tone. Downstream effects include autonomic nervous system stabilisation — potentially reducing the post-exertional sympathetic-parasympathetic crash that may contribute to PEM. This mechanism cannot be isolated clinically without a selective 5-HT1A probe, so its contribution relative to D2-mediated effects is unknown.
- Certainty
- Low — no way to distinguish from dopamine-mediated benefit with LDA alone. Selective 5-HT1A agonists (buspirone) could theoretically disentangle, but have not been studied in ME/CFS.
- Does NOT tell us
- whether serotonergic ANS stabilisation is a meaningful contributor vs. epiphenomenal; whether benefit is from pre- or post-synaptic 5-HT1A activation.
- Action
- If 5-HT1A contribution is suspected, a buspirone trial could isolate serotonergic effects without dopamine receptor engagement.
- Level of action
- Symptom management — modulates ANS output without correcting the underlying dysautonomia trigger.
3.2 What a positive response does NOT reveal
A positive LDA response cannot distinguish whether the benefit is microglial (D2 → cytokine suppression), dopaminergic (mesocorticolimbic tone restoration), serotonergic (5-HT1A → ANS stabilisation), or a combination of all three — LDA acts at all targets simultaneously. It does not identify the upstream cause of microglial priming (autoantibodies, infection history, metabolic stress remain indistinguishable from LDA alone). It does not reveal which D2 population is responsible — microglial D2 and neuronal D2 are pharmacologically identical. Because no controlled ME/CFS trials of LDA exist, a genuine pharmacological effect cannot be separated from placebo or natural fluctuation.
3.3 If LDA does NOT work
- Microglial activation threshold may not be D2-responsive — other microglial activation pathways (P2X7, TREM2, C5aR) dominate.
- Dopamine deficit not the primary cognitive mechanism — glymphatic failure, direct cytokine effects, cerebral hypoperfusion, or neuronal mitochondrial failure do not respond to dopamine.
- Wrong dose — partial agonist net effect is dose-dependent and varies by regional dopamine tone. U-shaped dose-response curve means both too little and too much can fail.
- Dopamine receptors downregulated — chronic inflammation reduces D2 expression.
- Metabolic constraint limits neurotransmitter release even with receptor stimulation.
- CYP2D6 poor metabolizer status → drug accumulation → postsynaptic D2 antagonism dominates → paradoxical worsening or no net benefit.
3.4 How LDA combines with other medications
- LDA works + mast-cell stabilisers work → microglial activation is the central node: mast cells supply the baseline pressure (reduced by antihistamines) and D2 agonism raises the threshold (provided by LDA). Combined response is stronger evidence for microglial PEM generation than either alone.
- LDA works + LDN does not → microglial D2 pathway is the dominant activation route; TLR4 antagonism is insufficient alone. Does NOT imply neuroinflammation is absent — LDN may be blocking the wrong receptor on the same cell population.
- Dose-resolved cross-drug pattern — LDN full-range tested (ceiling signal only, no therapeutic window at any dose) + LDA with sharp dose threshold (PEM protection at 1 mg, absent at 0.5 mg): This pattern is the strongest available evidence that D2 is the rate-limiting microglial receptor and TLR4 is excluded as a therapeutic target. LDN’s ceiling signal (sickness at 4.5 mg) proves the microglial TLR4 population is present and drug-accessible — the drug engaged its receptor and produced a biological response. But the hormetic collapse occurred before any therapeutic window appeared: benefit never preceded the ceiling, at any dose tested with adequate duration. LDA’s sharp threshold (0.5→1.0 mg PEM protection) independently confirms the same microglial population is D2-responsive and that D2-mediated activation threshold modulation is the dominant therapeutic mechanism. The combination — LDN gave a ceiling but no window, LDA gave a window bounded by a sharp threshold — means: (a) microglial involvement is confirmed by two independent receptor probes on the same cell type; (b) D2 is the therapeutically accessible receptor, not TLR4; (c) the upstream activator of this microglial population is NOT primarily TLR4 ligand (LPS, HMGB1, HSPs), consistent with non-TLR4 damage-associated signals (ATP→P2X7, RNA→TLR7/8, complement→C5aR) or HSAT2-mediated ETS/PU.1 reprogramming that lowers the microglial activation threshold through a non-TLR4 pathway. This pattern does NOT mean TLR4 is absent — the ceiling proves it is present and functional. It means TLR4 blockade cannot produce therapeutic benefit because the hormetic collapse occurs before the therapeutic window opens, while D2 agonism raises the activation threshold without a hormetic ceiling. The two receptors sit on the same cell but produce opposite dose-response profiles in this patient: one has a ceiling with no window, the other has a window with no ceiling (up to the partial-agonist inversion point). This asymmetry IS the diagnostic signal.
- LDA works + LDN works → neuroinflammation confirmed; multi-receptor microglial involvement (TLR4 + D2). Treat neuroinflammation as upstream driver via multiple complementary targets.
- LDA works + specific symptom domain persists → domain dissociation: the persistent symptom operates through a non-dopaminergic mechanism (e.g., thalamo-cortical GABAergic filtering). Do not increase LDA — probe the resistant domain with a different drug class.
- LDA works + methylphenidate worsens → receptor-level support needed; stimulant side effects (noradrenergic POTS, +7% REE) intolerable.
3.5 Key caveat
Non-response is weaker evidence than response. LDA non-response does not exclude dopamine deficiency, microglial involvement, or cognitive-circuit pathology — D2 receptors may be downregulated from chronic inflammation, the dose may be outside the therapeutic window (U-shaped dose-response), or metabolic constraint may limit neurotransmitter release even with receptor stimulation.
3.6 Compendium
The full pharmacodiagnostic entry — including dose-specific side-effect patterns (akathisia at micro-dose, persistent akathisia after discontinuation, receptor-state consolidation vs. microglial triggering), combination diagnostics with LDN, and worsening risk profiles (metabolic, akathisia, orthostatic) — is at Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe (sec-12, Aripiprazole entry). The Gerlier 4-pathway dopaminergic convergence cascade (Steps Z1–Z4) is at Chapter Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe, sec-07. Side effects classified by the five diagnostic patterns (Patterns 1–5) are detailed at Chapter Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe, sec-10.
4 Low-Dose Naltrexone (LDN)
LDN blocks TLR4 receptors on microglia (reducing neuroinflammation), restores TRPM3 calcium channel function in vitro, and triggers compensatory endorphin upregulation. Does not increase metabolic rate — can be tried even in severe patients.
4.1 Why Higher Dose Is Not Necessarily Better — The Hormetic Dose-Response Window]
Origin: Kevin Lee (personal communication, July 2026) — the clinical observation that even if LDN works via TLR4/TRPM3, this does not explain why higher dose is not always better and why some patients do better at lower doses.
If LDN were a simple receptor-level drug (more TLR4 blockade = more benefit), the therapeutic logic would push toward 4.5 mg for everyone — maximal target engagement within the LDN window. This is not what is observed clinically. Some patients do better at 0.5–1.5 mg than at 3.0–4.5 mg, and dose escalation can extinguish a response that was present at a lower dose. (Severity applicability: all — optimal dose is mechanism-dependent, not severity-dependent; start lower in severe/very severe patients for safety, but the dose-response curve shape is predicted to be independent of severity.) (Dara et al. 2023) (Bruun-Plesner et al. 2020) (Calabrese and Kozumbo 2021) (Toljan and Vrooman 2018)
LDN’s dose-response is better understood through three superimposed mechanisms with different concentration optima:
TLR4 hormetic window (low-dose priming): Partial TLR4 blockade at 0.5–1.5 mg may trigger compensatory anti-inflammatory priming via Nrf2-mediated transcriptional upregulation — a hormetic response in which the mild stress of receptor blockade activates the cell’s own anti-inflammatory programmes (M1→M2 phenotype shift, metabolic switch from glycolysis to OXPHOS in microglia). At higher doses (3.0–4.5 mg), TLR4 blockade removes the basal TLR4 tone needed to sustain this compensatory response — the hormetic benefit collapses, not because too little receptor is blocked, but because too much is. (Kučić et al. 2021) (Calabrese and Kozumbo 2021) (Kwilasz et al. 2021)
Opioid receptor compensatory upregulation (mid-dose plateau): Brief overnight opioid blockade triggers compensatory endorphin production — this mechanism has an individual ceiling set by endogenous opioid precursor expression. Beyond that ceiling, increasing LDN dose adds nothing to endorphin tone. (Boyadjieva et al. 2004) (Kreek 1992)
Why the overnight blockade window matters — GPCR resensitization kinetics: LDN’s 4–6 hour blockade window (bedtime dosing, daytime washout) functions as a de facto pulsed regimen. This is not a convenience convention — it tracks the conserved GPCR resensitization cycle: activation → desensitization → internalization → endosomal dephosphorylation → recycling to the membrane surface in a drug-naive state (Costa-Neto and Parreiras-E-Silva 2025) (Gupta, Mohan, and Naga Prasad 2018) (Kliewer, Reinscheid, and Schulz 2017). If LDN were taken continuously (multiple daily doses), the opioid and TLR4 receptors would remain in the internalized, desensitized pool — the resensitization interval that the 16–20 hour daytime washout provides would be filled, and the compensatory endorphin upregulation that requires a drug-free resensitized receptor surface would not occur. LDN’s therapeutic effect depends on the recovery from blockade as much as on the blockade itself — a time-dependent dimension that single-mechanism “LDN blocks TLR4” models do not capture. This principle generalizes: any drug whose therapeutic mechanism depends on a compensatory response triggered by transient receptor perturbation will lose effectiveness with continuous dosing. The 4–6 hour blockade window is empirical (it is what clinicians use), but the GPCR resensitization rationale for why the window matters is inferential — no study has compared pulsed vs continuous LDN dosing in any condition. (Hodding, Jann, and Ackerman 1980) (withdrawal risk counterpart: the same drug-free interval that enables resensitization could trigger withdrawal if opioid dependence has developed)
Dose-window overlap and individual variation: Each of LDN’s four targets (TLR4, TRPM3, opioid receptors, orexin disinhibition) has a distinct concentration-response curve. Some patients’ optimal dose is determined by the TRPM3 window (closer to 3.0–4.5 mg), others by the TLR4 hormetic window (closer to 0.5–1.5 mg), and still others by endorphin upregulation kinetics. The clinical observation that “the right dose varies” is not noise — it is the predictable consequence of a multi-target drug with non-overlapping dose optima. (Toljan and Vrooman 2018) (Dara et al. 2023)
Falsifiable prediction: A prospective within-range dose-response trial (0.5, 1.5, 3.0, 4.5 mg, n≥30 per arm, crossover, 8 weeks per dose) in ME/CFS should reveal non-monotonic individual response curves — most patients will not show monotonic improvement across the full dose range. Falsified if the group-average dose-response is strictly monotonic AND individual curves do not show inversions. Currently contradicted by zero data: no within-range dose-response trial has been conducted. (Bruun-Plesner et al. 2020) (ED50=3.88 mg in fibromyalgia with wide interindividual variation)
Certainty: 0.30 — no within-range dose-response trial exists in any condition; the hormetic model is mechanistically grounded ((Calabrese and Kozumbo 2021) cert 0.72, (Calabrese and Giordano 2021) cert 0.65) but untested for LDN specifically. Evidence is from: single-dose fibromyalgia trial ((Bruun-Plesner et al. 2020), ED50 estimation), microglial M1→M2 dose-dependence in vitro ((Kučić et al. 2021)), opioid receptor feedback models ((Boyadjieva et al. 2004), (Kreek 1992)), and review-level biphasic characterization ((Dara et al. 2023), (Toljan and Vrooman 2018)). The absence of direct evidence is itself informative — no one has asked the question in a design that could answer it. The LIFT trial ((Meadows et al. 2025)) uses a single LDN dose; it cannot address this.
Consequence: For clinicians and patients, the hormetic model provides a mechanistic rationale for dose-finding patience: start at 0.5 mg, titrate slowly, and recognise that “more” is not always “better.” If a lower dose works better than a higher dose, this is not a failed titration — it identifies that the patient’s therapeutic mechanism is the TLR4 hormetic window or endorphin ceiling, not TRPM3 restoration. For researchers, the absence of a within-range dose-response study is the single largest evidence gap in the LDN literature — a four-arm crossover dose-response trial in ME/CFS would cost an estimated $ 200,000–$ 400,000 and answer a question affecting every LDN prescription written in any chronic illness.
4.2 If LDN works
LDN blocked TLR4 → microglia calmed → inflammatory cytokines (IL-1β, TNF-α) decreased → symptoms improved.
- Certainty
- Low to Medium — LDN’s TLR4 antagonism is established in vitro (Younger, Parkitny, and McLain 2014); neuroinflammation is documented in ME/CFS (Hundreds of Blood Biomarkers Distinguish ME/CFS, Independent of Inactivity). But no direct evidence that clinical benefit is TLR4-mediated.
- Does NOT tell us
- what is causing the neuroinflammation — autoantibodies (Oxidative Stress Sensing Polymorphisms as Safe Mode Predisposition), viral components, mitochondrial debris, or vagal stress signals.
- Action
- anti-neuroinflammatory strategies become relevant.
- Level of action
- Partial root cause — LDN calms microglia without removing the trigger.
LDN restores TRPM3-mediated calcium flux in NK cells in vitro (Cabanas et al. 2018). TRPM3 is essential for immune killing, neurotransmitter release, and vascular tone. TRPM3 dysfunction is the most replicated ion channel finding in ME/CFS (Cabanas et al. 2021).
- Certainty
- Low to Medium — the central link (LDN restores TRPM3 in living humans, producing clinical benefit) has never been demonstrated.
- Does NOT tell us
- whether TRPM3 dysfunction is primary or secondary to GPCR autoantibody-driven PIP2 depletion (PIP2 Exhaustion Links GPCR Autoantibodies to Multi-Channel Dysfunction).
- Action
- TRPM3-targeting strategies (pregnenolone sulfate, PIP2 maintenance) become relevant.
- Level of action
- Potential root cause — TRPM3 is a trigger-capable root cause (GPCR Autoantibody Cascade as Trigger-Capable Root Cause); LDN partially restores its function.
LDN reduces hypothalamic microglial activation → less PGE2/TNF-α suppressing orexin neurons → increased orexin → improved wakefulness and cognition. Orexin falls into an intermediate “gray zone” in ME/CFS CSF (Chronic Orexin Suppression as a Central Driver of ME/CFS Fatigue and Sleep-Wake Instability).
- Certainty
- Low — no study has measured orexin before/after LDN in ME/CFS.
- Level of action
- Symptom management.
Brief overnight opioid blockade triggers compensatory endorphin upregulation.
- Certainty
- Medium for pain, Low for other symptoms.
- Level of action
- Symptom management.
4.3 Dose-Dependent Differential Inference — The Dose-Response Curve Shape Is Diagnostic
LDN’s four mechanisms have non-overlapping dose optima (hormetic dose response): TLR4/Nrf2 hormetic priming peaks at 0.5–1.5 mg, opioid compensatory upregulation plateaus at 1.5–3.0 mg, TRPM3 restoration may require 3.0–4.5 mg, and orexin disinhibition tracks the TLR4 dose-response. Response at which dose therefore discriminates which mechanism is rate-limiting — information a single fixed-dose trial cannot provide.
Partial TLR4 blockade primes Nrf2-mediated M1→M2 microglial shift; higher doses remove the basal TLR4 tone sustaining the priming signal and the benefit collapses (Calabrese and Kozumbo 2021) (Kučić et al. 2021). Window width estimates Nrf2 transcriptional reserve — benefit lost by 1.5 mg = narrow reserve; benefit persisting to 3.0 mg = broad reserve.
- Certainty
- Low — hormetic model mechanistically grounded but no within-range dose-response trial exists.
- Does NOT tell us
- what is activating TLR4 (autoantibodies, viral components, mitochondrial debris); whether Nrf2 reserve is itself impaired.
- Action
- maintain the low dose — escalation is counterproductive; response to other Nrf2 activators (sulforaphane, lithium, melatonin) is predicted by window width.
- Level of action
- Partial root cause — identifies neuroinflammation as TLR4-driven and microglia as Nrf2-responsive.
Compensatory endorphin upregulation has a ceiling set by POMC/proenkephalin expression (Boyadjieva et al. 2004) (Kreek 1992). Plateau without high-dose worsening excludes TLR4 hormetic dominance, which would worsen at 4.5 mg.
- Certainty
- Low to Medium — endorphin rebound documented; ceiling inference untested.
- Does NOT tell us
- whether pain/mood benefit masks ongoing disease progression.
- Action
- titrate to the plateau point and no higher; treat the underlying pain driver directly.
- Level of action
- Symptom management.
TRPM3 restoration may require higher concentrations; orexin disinhibition tracks hypothalamic TLR4 dose-response, which may differ from cortical populations. No TRPM3 dose-response data exist — all in vitro work used a single concentration (Cabanas et al. 2018).
- Certainty
- Low — the TRPM3 dose-response is the single largest evidence gap in the LDN literature.
- Does NOT tell us
- which of TRPM3 vs. orexin is responsible — wakefulness/cognition-dominant benefit suggests orexin; multi-system benefit (immune, vascular, neurotransmitter) suggests TRPM3.
- Action
- TRPM3 function measurement when clinically available; pregnenolone sulfate as adjunct.
- Level of action
- Potential root cause if TRPM3 (GPCR Autoantibody Cascade as Trigger-Capable Root Cause); symptom management if orexin.
TLR4/Nrf2 hormetic window at low dose plus a second mechanism (likely TRPM3) at higher dose. The patient has two rate-limiting mechanisms addressable by the same drug at different doses.
- Certainty
- Low — predicted by the multi-target model; never tested.
- Does NOT tell us
- whether combining low-dose LDN with a second TRPM3-targeting agent captures both optima simultaneously.
- Action
- choose the dose targeting the dominant symptom domain; consider mechanism-specific adjuncts rather than a compromise dose.
- Level of action
- Mixed — one mechanism may be root-cause-level (TRPM3), the other partial (TLR4).
The patient has probed the entire dose range and each band returned a specific null: (1) the top of the TLR4 hormetic band (3.0–4.5 mg) produced systemic sickness — confirming the TLR4 hormetic ceiling was crossed and Nrf2 reserve is narrow; (2) the upper range below the ceiling (3.0–4.0 mg) produced no benefit — excluding TRPM3/orexin as therapeutic mechanisms; (3) lower doses (0.5–1.5 mg, 1.5–3.0 mg) were either not trialled or produced no therapeutic signal despite adequate duration — excluding TLR4/Nrf2 hormetic priming and endorphin compensation. The combination of ceiling signal (sickness at 4.5 mg) + therapeutic null across all bands means: the microglial population IS present and TLR4-responsive (the ceiling proves receptor engagement), but TLR4 blockade does not produce a therapeutic window — the anti-inflammatory benefit never appears before the hormetic collapse. This is distinguishable from pure non-response (no signal at any dose), which could reflect insufficient trial duration or drug access failure rather than true mechanism absence. A ceiling signal without a therapeutic window is a stronger exclusion than simple null response.
- Certainty
- Low — no prospective dose-response trial validates the ceiling-only pattern as a mechanism exclusion. The inference that hormetic crossing without prior benefit excludes TLR4 as a therapeutic target is mechanistically coherent but unvalidated.
- Does NOT tell us
- whether lower doses (0.5–3.0 mg) were actually trialled for adequate duration (>=4 weeks each) — if not, the exclusion of TLR4/Nrf2 and endorphin mechanisms is premature. The ceiling signal at 4.5 mg only proves receptor engagement; it does not prove the lower bands were adequately tested.
- Action
- If lower bands were not adequately tested: trial 0.5 mg × 4 weeks, then 1.5 mg × 4 weeks before concluding TLR4/Nrf2 non-involvement. If all bands were adequately tested: the therapeutic target is NOT TLR4. Probe D2-mediated microglial activation (LDA), mast-cell→microglial pathway (ketotifen, cromolyn), or non-microglial mechanisms entirely. The ceiling signal + null therapeutic window strongly predicts LDA response — the same microglial population that LDN’s TLR4 blockade engaged but could not produce benefit from is likely D2-responsive.
- Level of action
- Diagnostic — the ceiling-only pattern is a specific null that narrows the differential to non-TLR4 microglial pathways.
4.4 What a positive response does NOT reveal
Which of the four mechanisms is responsible at a single fixed dose (LDN targets all simultaneously) — however, the dose-response curve shape discriminates: benefit lost on escalation implicates the TLR4/Nrf2 hormetic window; plateau without loss implicates opioid compensation; high-dose-only benefit implicates TRPM3 or orexin (see dose-dependent differential inference above). Whether response is pharmacological or placebo (no ME/CFS-specific RCT vs. placebo); the upstream cause of the dysfunction.
4.5 If LDN does NOT work
- Neuroinflammation may not be dominant — or LDN may fail to reach brain microglia (BBB permeability, pharmacokinetics).
- TRPM3 dysfunction may be absent or not LDN-responsive — or LDN may restore NK cell TRPM3 without affecting neuronal/vascular TRPM3.
- Dose or duration insufficient — no ME/CFS-specific dose-response studies exist. The failure may be a dose too high, not too low: if the therapeutic mechanism is the TLR4/Nrf2 hormetic window (0.5–1.5 mg), a trial conducted only at 3.0–4.5 mg sits above the therapeutic window and guarantees apparent non-response even in a true responder (hormetic dose response). Non-response cannot be concluded unless the low-dose window has also been tested.
- LDN may simply not work better than placebo in ME/CFS. No large RCT exists. If LDN = placebo, every inference above collapses.
- Paradoxical reactors worsen on LDN (Medication Sensitivity Phenotypes) — their adverse response is itself diagnostic.
- Critical caveat for LDN non-response with subsequent LDA response: TLR4 antagonism and D2 partial agonism are complementary targets on the same microglial population. LDN non-response does not mean neuroinflammation is absent — it may mean TLR4 is not the rate-limiting input; D2 agonism can still suppress microglial cytokine output through a different signalling pathway. If LDA works dramatically after LDN non-response, this is evidence FOR microglial involvement, not against it.
4.6 Side Effects per Dose Zone — What Side Effects at Each Dose Reveal
Side effects are not uniform across doses — they track the mechanism engaged at that concentration. A side effect at one dose and not another is dose-dependent diagnostic information, not random intolerance.
Systemic sickness (malaise, flu-like, nausea) appearing at a specific dose and absent below it:
This is a within-range hormetic crossing. LDN’s TLR4 blockade at low doses (0.5–1.5 mg) triggers Nrf2-mediated M1→M2 microglial shift — the hormetic anti-inflammatory programme. At higher doses, TLR4 blockade removes the basal TLR4 tone that sustains this priming. When the threshold is crossed, the M2 programme collapses → microglia revert to M1 → cytokine surge (IL-1β, TNF-α) → systemic sickness. The dose at which sickness appears identifies the patient’s hormetic ceiling, which is also the threshold of their Nrf2 transcriptional reserve.
The patient crosses the TLR4 blockade threshold between their tolerated dose and the sickness dose — basal TLR4 tone falls below what is needed to sustain Nrf2-mediated M2 priming. The microglial anti-inflammatory programme collapses and systemic cytokine release produces flu-like symptoms. This is exactly the pattern the hormetic model predicts: benefit is lost when blockade crosses the threshold, and loss presents as active sickness, not just neutral return-to-baseline.
- Certainty
- Low to Medium — mechanistically grounded in the hormetic TLR4/Nrf2 model (Calabrese and Kozumbo 2021) (Kučić et al. 2021) but no clinical study has measured microglial cytokines during LDN dose escalation. Patient self-report of dose-specific sickness is the only available signal.
- Does NOT tell us
- what is driving TLR4 activation — the trigger that loads the TLR4 signalling pathway and makes microglial M1 the default state; whether Nrf2 reserve is intrinsically impaired or merely narrow against a strong TLR4 signal.
- Action
- stay at or below the highest tolerated dose — escalation past the ceiling is counterproductive and actively harmful. The inversion point is the diagnostic signal: the dose just below sickness identifies the patient’s therapeutic ceiling. Narrow window → Nrf2 reserve limited; predict narrow windows for other Nrf2 activators (sulforaphane, lithium, melatonin). Broad window (sickness only at 4.5 mg) → Nrf2 reserve adequate but TLR4 signal strong — anti-inflammatory adjuncts (PEA, luteolin) may widen the window.
- Level of action
- Partial root cause — identifies Nrf2 reserve as the limiting factor for anti-inflammatory benefit.
Dose-specific nausea without systemic sickness — gut TLR4 blockade:
Enteric glial cells and enteric neurons express TLR4. At a dose that suppresses gut TLR4 without crossing the CNS hormetic threshold → local GI effects (nausea, altered motility, gut discomfort) without systemic cytokine surge. This is distinct from CNS-mediated systemic sickness — the gut is the primary site of TLR4 antagonism at this dose in this patient, either because of first-pass metabolism concentrating naltrexone in the gut wall, or because gut TLR4 tone is higher than CNS.
- Certainty
- Low — enteric TLR4 blockade by LDN has not been studied in humans. Inference from gut-brain axis pharmacology.
- Does NOT tell us
- whether gut permeability (elevated LPS) is driving gut TLR4 activation; whether CNS benefit occurs at a different dose.
- Action
- split the dose — bedtime CNS-targeted dose + morning gut-targeted dose at the tolerated level; or use compounded transdermal LDN to bypass gut first-pass. If nausea persists on transdermal → CNS TLR4 threshold crossed, not gut-localized.
- Level of action
- Symptom management.
Sedation at low dose (0.5–1.5 mg):
Opioid receptor blockade suppresses orexin neurons that are already deficient — consistent with the orexin gray-zone hypothesis (Chronic Orexin Suppression as a Central Driver of ME/CFS Fatigue and Sleep-Wake Instability). Sedation typically resolves as compensatory endorphin upregulation occurs over 2–4 weeks. Persistent sedation → orexin deficit is structural, not opioid-suppressed.
- Certainty
- Medium for orexin mechanism — orexin’s role in LDN sedation is inferred; no study has measured orexin during LDN initiation.
- Does NOT tell us
- whether sedation will resolve with continued dosing; whether other orexin-suppressing drugs (antihistamines, benzodiazepines) are being taken concurrently.
- Action
- start at 0.25 mg if sedation at 0.5 mg; if sedation persists at 0.25 mg → LDN may not be tolerated. If sedation resolves after 4 weeks → endorphin upregulation has compensated; this confirms functional orexin reserve.
- Level of action
- Symptom management — identifies orexin status but does not treat it.
Vivid dreams / nightmares at any dose:
Opioid agonists suppress REM sleep; opioid antagonism disinhibits it → REM rebound with vivid dreams. Confirms that endogenous opioid tone was present and active — the opioid system is functional.
- Certainty
- Medium — REM disinhibition by opioid antagonism is well-established in sleep pharmacology.
- Does NOT tell us
- whether the vivid dreams will subside with continued dosing (endorphin compensation restabilizing REM).
- Action
- switch to morning dosing if sleep disruption is unacceptable — concedes endorphin upregulation benefit but preserves TLR4/TRPM3 mechanisms. If vivid dreams persist despite morning dosing → unrelated to opioid rebound; consider concurrent sleep pathology.
- Level of action
- Symptom management — identifies opioid tone as load-bearing for sleep architecture.
Dysphoria or depression at any dose — the paradoxical reactor phenotype:
Opioid receptor blockade unmasks an endogenous opioid deficit — the system that was maintaining mood stability through opioid tone collapses when even partial blockade is applied. This is a permanent trait: the patient’s mood regulation depends on opioid signalling that LDN interrupts.
- Certainty
- High for the trait — paradoxical dysphoria on opioid antagonists is a well-documented clinical phenomenon.
- Does NOT tell us
- which opioid receptor subtype is responsible (mu vs. delta vs. kappa); whether the deficit is pre-existing or induced.
- Action
- stop LDN immediately. The patient should permanently avoid all opioid-modulating drugs: naltrexone, naloxone, buprenorphine, tramadol, tapentadol. This is not a slow-titration issue — the reaction is pharmacodynamic, not dose-rate-dependent.
- Level of action
- Safety — paradoxical reactor identification protects against iatrogenic harm from future opioid-modulating medications.
No side effects at any dose up to 4.5 mg:
Neither TLR4 blockade (no GI/CNS sickness) nor opioid blockade (no sedation, no dysphoria, no vivid dreams) produces perceptible effects. This is not the same as therapeutic non-response — the patient may still benefit via TRPM3 restoration or endorphin upregulation. But the absence of side effects is itself diagnostic: it means the TLR4 hormetic ceiling has not been reached (broad Nrf2 reserve), the orexin system has adequate reserve, and opioid tone is not load-bearing for mood or sleep.
- Certainty
- Low — absence of side effects cannot confirm receptor function; it only confirms no receptor-level crisis was triggered.
- Does NOT tell us
- whether higher doses (>4.5 mg) would produce effects that the current range did not.
- Action
- titrate to max tolerated or max clinical dose (4.5 mg). If no benefit at 4.5 mg + no side effects → none of LDN’s four mechanisms are rate-limiting.
- Level of action
- Diagnostic — silence across all four receptor systems is a strong null signal.
Warning — dose-too-high non-response masquerading as non-response:
A patient trialled only at 3.0–4.5 mg who experiences no benefit AND no side effects may be a true non-responder. But a patient who experiences systemic sickness at 4.5 mg and was never trialled at lower doses is a missed responder — their therapeutic mechanism is the TLR4/Nrf2 hormetic window (0.5–1.5 mg), and the 4.5 mg dose sits above their therapeutic range. Non-response cannot be concluded unless the low-dose window (0.5–1.5 mg) has also been tested. This error (starting too high, concluding drug doesn’t work, and stopping) is common enough to merit explicit documentation in the differential diagnostic algorithm.
4.7 Key caveat
non-response is weaker evidence than response. LDN non-response does not exclude immune involvement — autoantibody, B-cell, mast-cell, and T-cell mechanisms are not directly targeted. LDN non-response especially does not exclude microglial involvement if LDA (D2) works — the two drugs target different receptors on the same cell.
4.8 How LDN combines with other medications
- LDN + pyridostigmine both work → neuroinflammation → autonomic dysfunction (inflammatory-to-autonomic pathway).
- LDN works + valacyclovir does not → neuroinflammation without active viral replication (post-infectious mechanism).
- LDN works + antihistamines do not → neuroinflammation from non-mast-cell sources (autoantibodies, microglial priming).
- LDN does not work + cimetidine works → VIM phenotype (Domain 6 Evidence Limitations — No ME/CFS RCT Exists): T-cell dysfunction, not TLR4/TRPM3.
- LDN does not work + LDA works → microglial involvement confirmed through D2 pathway despite TLR4 non-response. Complementary targets (TLR4 + D2) on the same cell population; combination may be synergistic even when LDN appears ineffective alone — LDN may reduce one input while LDA raises the threshold on all inputs simultaneously.
- LDN does not work + mast-cell stabilisers work → mast-cell → microglial pathway confirmed, but TLR4 is not the dominant microglial receptor for mast-cell-derived signals in this patient.
4.9 Compendium
The full pharmacodiagnostic entry — including mechanism-exclusion logic (TLR4/TRPM3/endorphin/orexin), dose-specific side-effect diagnostic patterns (SE0–SE5), combination diagnostics with aripiprazole and midodrine, and worsening risk profiles — is at Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe (sec-12, LDN entry). The M1–M4 cascade tracing LDN’s multi-target dose-response through each mechanism is at Chapter Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe, sec-08. Side effects classified by the five diagnostic patterns (Patterns 1–5) are detailed at Chapter Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe, sec-10.
5 Pyridostigmine (Mestinon)
Pyridostigmine inhibits acetylcholinesterase, prolonging acetylcholine at autonomic ganglia, the neuromuscular junction, and vagal immune synapses. Probes preload failure, cholinergic anti-inflammatory pathway, and residual autonomic nerve function. One of the strongest evidence bases: Joseph 2022 iCPET RCT (n=45) showed objective hemodynamic improvement (Joseph et al. 2022), Squires 2023 long-term follow-up (n=37) showed sustained benefit (Squires, Al-Zayer, and Systrom 2023).
5.1 If pyridostigmine works
Pyridostigmine increased acetylcholine at autonomic ganglia → improved venous return → increased cardiac filling → increased oxygen delivery → improved exercise capacity. The Joseph 2022 RCT demonstrated this objectively: +13.3 mL/min peak VO₂ with pyridostigmine vs. −40.2 mL/min worsening with placebo.
- Certainty
- Medium — double-blind RCT with invasive hemodynamic measurements.
- Does NOT tell us
- whether preload failure is from autoantibody-mediated vascular dysfunction, hypovolemia, or connective tissue laxity.
- Action
- volume expansion (fludrocortisone, sodium) and compression become particularly relevant.
- Level of action
- Partial root cause — pyridostigmine compensates for impaired ganglionic transmission without addressing what caused it.
Pyridostigmine increases acetylcholine, activating α7 nicotinic receptors on macrophages and T cells — vagally-mediated immune modulation. PISCO COVID-19 RCT (n=188): pyridostigmine halved 28-day mortality (HR 0.47) with no change in systemic cytokines (Fragoso-Saavedra et al. 2022) — suggesting tissue-level immune modulation.
- Certainty
- Low — evidence from COVID-19, not ME/CFS.
- Level of action
- Partial root cause.
Pyridostigmine potentiates existing acetylcholine — it cannot create a signal where nerves have degenerated. Okamoto et al. (2025) showed preserved baroreflex function predicts response (Okamoto, Walsh, et al. 2025).
- Certainty
- Medium.
- Action
- if pyridostigmine works, autonomic nerves are functional — treating the upstream cause could restore normal autonomic tone, potentially eliminating the need for ongoing pyridostigmine.
- Level of action
- Symptom management.
5.2 What a positive response does NOT reveal
A positive pyridostigmine response confirms that impaired ganglionic transmission and preload failure were contributing, but it does not reveal what caused them — autoantibody-mediated vascular dysfunction, hypovolemia, and connective tissue laxity remain indistinguishable from the drug response alone. It shows that residual autonomic nerve function was preserved (pyridostigmine can only potentiate existing acetylcholine, not replace degenerated nerves) but does not localize the upstream driver. The cholinergic anti-inflammatory contribution is inferred from COVID-19 data, not ME/CFS, so its role in any given patient remains uncertain.
5.3 If pyridostigmine does NOT work
- Severe autonomic failure with insufficient residual nerve function.
- Preload failure may not be dominant — hypovolemia (fludrocortisone-responsive), excessive venous pooling (compression-responsive), or hyperadrenergic POTS (ivabradine-responsive) may predominate.
- GI side effects limiting dose — ME/CFS patients with pre-existing dysmotility or MCAS may not tolerate doses sufficient for autonomic benefit.
- Wrong autonomic target — hyperadrenergic POTS patients may worsen from combined sympathetic/parasympathetic enhancement.
5.4 Key caveat
Non-response does not exclude autonomic dysfunction. Pyridostigmine can only potentiate existing acetylcholine — where autonomic nerves have degenerated (severe autonomic failure with insufficient residual nerve function), the mechanism may be present but beyond what the drug can restore. Dose-limiting GI side effects can also prevent reaching a therapeutic level, so non-response may reflect intolerance rather than absence of preload failure.
5.5 How pyridostigmine combines with other medications
- Pyridostigmine + LDN both work → neuroinflammation → autonomic dysfunction (inflammatory-to-autonomic cascade).
- Pyridostigmine + fludrocortisone/midodrine → multi-mechanism preload failure (ganglionic + volume + vasoconstriction). Meta-analysis: combined therapy −6.74 mmHg systolic drop (p < 0.05) (Pavic et al. 2025).
- Pyridostigmine does not work + ivabradine works → insufficient ganglionic reserve; excessive sinus node automaticity or inadequate parasympathetic restraint predominates.
5.6 Compendium
The full pharmacodiagnostic entry — including mechanism-exclusion logic, dose-specific side-effect diagnostic patterns, combination diagnostics, and worsening risk profiles — is at Mechanistic Cascade Tracing: From Hypothesis to Clinical Probe (sec-12, Pyridostigmine entry).