Low Brain Norepinephrine Levels in ME/CFS and Long COVID Pt. II: A ChatGPT Based Treatment Program

by Cort Johnson | Nov 16, 2025 | Autonomic Nervous System, Brain, Brainstem, Clonidine, CoQ10, D-Ribose, GLP-1, GLP-1 agonist, Guanfacine, Homepage, Ivabradine, Kaiser, L-carntine/Acetylcarnitine, Locus coeruleus, Mast cell activation syndrome, Mestinon, Neuroplasticity, Neuroplasticity, Norepinephrine, Pacing, Prefrontal Cortex, Propanolol, Rapamycin, Research, Stimulants, Urolithin A | 163 comments

UPDATE

Health Rising will have an AI use page which explains how it’s being used

Getting Up to Speed with AI – To be honest, I was floored by how controversial this ChatGPT brainstorming treatment post was. The fact that it happened in a post with pretty moderate treatment suggestions only underscored how hot a button topic AI is.

With this post I stepped into a minefield I didn’t know existed. That’s my mistake – I’ve been tooling along so well with AI search engines both personally and with blogs that I honestly missed the controversy.

Health Rising’s Quickie Summer Donation Drive is On!

Keeping up with the latest research in ME/CFS, long COVID, fibromyalgia, and allied diseases. Exploring new treatment possibilities. Learning how others have recovered. All in as thoroughly and comprehensively as we can.

Please support Health Rising during our quickie summer donation drive. Our goal is to raise $15,000.

Find out more here.

So I will do what should have been done earlier: take a deep dive into the pros and cons of ChatGPT, ensure it’s used as safely as possible, produce a page outlining how that will be done, and, from time to time, demonstrate how it responds to queries. Thanks to the support of a Health Rising supporter, last week I also enrolled in a class on AI.

Note that the norepinephrine brain finding, while intriguing, is based on one unpublished study.

The low norepinephrine in the brain finding potentially opens a new window into what’s happening in diseases like ME/CFS and long COVID.

The idea that an energy production problem is causing both the sympathetic (fight/flight) and parasympathetic (rest/digest) branches of the autonomic nervous system to underperform makes intuitive sense, given what we know about these diseases.

Norepinephrine also plays a major role in core brain issues such as alertness, arousal, and the stress response that face ME/CFS and long COVID patients. Additionally, findings in other diseases suggest that the chronic sympathetic nervous system activation seen in ME/CFS can deplete brain norepinephrine levels and reduce vesicle filling.

Some provisos do exist, however. While a respected researcher oversaw the study, it has not been published or validated. While the ATP hypothesis makes sense, other issues, such as damage to the transporter, could apply. Plus, other things are going on in ME/CFS and long COVID.

While reduced mitochondrial ATP production makes sense, we don’t know that it is the problem.

The next steps for Dr. Goldstein would be to assess the NE finding pharmacologically by determining if drugs such as LDOPA, Droxidopa, amphetamines (methylphenidate, dextroamphetamine), Ritalin, MDMA (ecstasy), Dexedrine, some oxetine drugs (atomoxetine, reboxetine), desipramine, Bupropion (Wellbutrin), and yohimbine that can boost NE levels.

Most of these drugs temporarily boost NE levels by blocking NE uptake or increasing NE release from vesicles.

They’re probably not long-term solutions, though. Tyrosine and LDOPA, for instance, seem like good candidates, as both tyrosine and LDOPA are part of the NE production chain (Tyrosine → L-DOPA → Dopamine → (VMAT2) → Vesicles → Dopamine → NE (via DBH)), and can increase NE levels. Increasing NE levels, though, doesn’t address the core problem – the inability to get NE into the vesicles – which may be related to reduced ATP levels.

THE GIST

While ChatGPT does not believe that neuroplasticity can refill NE stores, by reducing the threat response, it can lay promote healing when the system able to do that.

The low norepinephrine-in-the-brain problem potentially opens a new window on what’s happening in diseases like ME/CFS and long COVID.
It suggests that both the fight/flight and rest/digest systems have been hit hard by problems with energy production in these diseases.
This makes sense given what we know, but note that the study is unpublished and its results have not been validated.
This blog mostly consists of treatment protocols produced by ChatGPT 5.1 Pro. ChatGPT and other AI search engines’ ability to quickly access and integrate vast amounts of information means they can detect patterns and, at times, produce answers that humans might miss.
They’re also not prey to the numerous biases (“availability”, confirmation bias, premature closure, etc.) that humans are.
On the other hand, AI engines lack the training, experience, and hands-on knowledge that doctors and other practitioners have regarding their patients ‘ illnesses.
AI-generated treatment protocols are best seen as brainstorming sessions that you can take to your doctor to review.
Because it’s not possible to post the protocols in the GIST you’ll have to access the paper to check them out.

Support Health Rising and Keep the Information Flowing!

Health Rising is not a 501 c (3) non-profit

While these drugs/supplements can and do help some people, they can also make things worse by producing unused NE, which then gets stored in the cell and, when it breaks down, produces free radicals like hydrogen peroxide, activates the microglia, and suppresses ATP production to boot (!). In the end, unused NE creates more “noise” in the system.

A ChatGPT Treatment Approach: Pro’s and Cons

The blog was largely the result of asking ChatGPT questions. Think of it as a brainstorming sesssion.

This long blog -which is basically a series of possible treatment protocols – is built out of a series of questions posed to ChatGPT 5.1 Pro.

(Given its length it might be easier to print it out using the print feature on the lower left bottom of the page)

Why use ChatGPT to develop a treatment approach? Because it can quickly scour far more information than the human brain can integrate and come up with logic-based interpretations and answers.

It has its limitations, though. Because AI engines are limited to the data available to them, i.e., the data in studies, clinical trials, blogs, anecdotal reports, etc., they also miss a lot.

They can also “hallucinate”; that is, provide citations that aren’t real (I haven’t found that yet), draw premature conclusions, etc.

They also can’t access, the data that an experienced physician carries around in his/her head. Gut instincts – which can encapsulate knowledge doctors are not aware they possess – are out. AI engines didn’t spend a decade or so in college preparing to become doctors. They can’t assess you in an office and don’t know your history.

For all these reasons, ChatGPT-based treatment protocols should be viewed as preliminary, brainstorming sessions that need to be interpreted by a doctor.

Of course, doctors, being human, have their own inherent potential biases to deal with. Many people with these diseases have probably come across “anchoring,” which occurs when a doctor latches onto the first explanation that comes to mind (“anxiety” ) and then sticks to it.

Doctors – particularly busy doctors – may display “premature closure” and stop their diagnostic search too early, or exhibit “confirmation bias” where they only take in information that confirms what they already believe. “Availability” occurs when, after doctors make a diagnosis, they’re more likely to repeat it in subsequent patients. Doctors may have unrecognized biases, such as being more likely to diagnose women with psychological disorders.

While ChatGPT lacks an experiential component and is not entirely trustworthy, it can present organized plans and options that doctors might not be aware of, and which could help inform their treatment plans.

This blog presents a variety of approaches, which, should you wish to investigate them further, you could bring to your doctor for review. These ChatGPT search results were generally in line with what many ME/CFS and long COVID practitioners employ but it did at times package them in ways I hadn’t seen before.

ChatGPT 5.1’s Treatment Protocols to Increase Norepinephrine Levels in the Brains of People with ME/CFS and Long COVID

ChatGPT wants to increase ATP levels and improve NE vesicle filling in the locus coeruleus neurons in the brain without raising NE levels in the bloodstream and sparking a hyperadrenergic reaction. While some of its suggestions might increase sympathetic nervous system activity in the body slightly, they are specifically chosen because they don’t strongly activate it outside the brain.

The Core Program

This program gets considerably more layered and complex as we go on.

Support ATP supply specifically in neurons.

Creatine monohydrate – 3–5g/day – this improves phosphocreatine buffering and supports the transporter (V-ATPase), which transports dopamine into the vesicles where it gets turned into norepinephrine.
Take CoQ10 + Riboflavin + NAD⁺ precursors – improve Complex I → electron flow → ATP output in LC neurons.
Take L-Tyrosine – only if ATP is sufficient. Otherwise, tyrosine increases dopamine levels in the cell without improving vesicle loading, which worsens anxiety/jitteriness.

One of ChatGPT’s goals is to stabilize the norepinephrine-producing neurons in the brain.

Make Sure NE isn’t being lost.

Vitamin C – ChatGPT’s second step is making sure that the vesicles aren’t leaking by adding Vitamin C (250–750 mg with breakfast, not high megadosing). NE synthesis also requires ascorbate as electron donor.

Prevent neuron over-firing and under release; i.e. stabilize the locus coeruleus neurons and improve signal quality.

Low-dose guanfacine – Carefully use α2A agonists (very-low-dose guanfacine (0.1–0.3 mg at night) provide better control, reduce the “noise” in the system, and neuron over-firing/ under-release); (Note ChatGPT does not recommended extended release guanfacine here (it does in other places) but you should not break up extended release tablets to get low doses).
Clonidine – reduces sympathetic tone body-wide taking pressure off the system, sometimes cognitively stabilizing.

SUBSETS

ChatGPT recommended different protocols if you are more active in the morning or the afternoon

Since I feel noticeably worse in the morning and typically have a lot of trouble getting going but often feel increased energy in the late afternoon, I asked ChatGPT about that. The program produced different protocols based on energy peaks.

If you feel worse in the morning

Noting that I have a hard time getting going in the morning, I asked ChatGPT if that changed things, and it did…

Take a low dose of Guanfacine (0.1 mg ) or clonidine (0.025-0.5mg) at bedtime plus 250-500mg Vitamin C in the evening to support dopamine-NE conversion, and 3-5 g of creatine monohydrate with dinner (to increase ATP reserve).
Upon awakening, take 100-200 mg CoQ10 (Ubiquinol preferred) to improve ATP output, 100 mg Riboflavin (B2) (cofactor for CoQ10), and 250-500 mg L-Tyrosine (if you do not experience anxiety), and do light/slow breathing for 3 minutes. (Do not take Tyrosine first thing alone, as you need to increase ATP levels first.) Eat a small, protein/fat-rich meal and drink some electrolytes. Do not take stimulants or exercise (both use up NE) for the first 60-90 minutes.
At midday, take 250-500 mg of vitamin C and a small protein snack. If you’re taking Mestinon, take it now. Stop caffeine by 1:00–2:00 PM, and consistently turn your lights out at about the same time at night.
Then see if it helps. If it does, your morning brain fog will lift, and you’ll feel like you can start moving a bit better over 1-3 days. Mental clarity, better stress tolerance, and less wired-and-tired feelings begin to show up from 4 days to 2 weeks. After two weeks, orthostatic intolerance and sensory tolerance problems begin to gradually improve.

If you feel worse in the afternoon

Your early morning NE stores are OK, but they become depleted over the day. In this case, you skip the guanfacine/clonidine/creatine monohydrate in the evening and instead take creatine monohydrate (3-5 g) and the others (100-200 mg CoQ10 (Ubiquinol preferred), 100 mg Riboflavin (B2) (cofactor for CoQ10), and 250 mg Vit C in the morning.
One to two hours before you experience your usual afternoon slump, take 150-350 mg tyrosine, 250-500 mg vit. C, and electrolytes (sodium+magnesium) to reduce LC stress by dehydration.
If you feel wired and unfocused, take a microdose of guanfacine (.05-.1mg) or L-theanine (100-200 mg) mid-afternoon.
In the evening, have a protein-rich, complex-carb dinner and wind down.
Then, see if over 2-4 days, your afternoon crash fades; over 1-2 weeks, your clarity in the afternoon improves; and over 3-6 weeks your ability to engage in sustained attention is enhanced.

Getting Up: A Suggested Early Morning Sequence (Total: 20–30 minutes). This was individualized for me. My normal wake up time (unfortunately) iis around 5 am).

🕠 Daily Schedule — Wake 5:00 AM

Time	Action	Dose / Instructions	Mechanistic Purpose
5:00 AM (Wake)	Sit up slowly in bed, no phone yet	Dim light / bedside lamp only	Prevents premature LC/NE spike; maintains low “noise floor”
5:02 AM	Open blinds or step outside for sky light	5–15 min natural light to eyes (not bright LEDs)	Activates ipRGC → SCN → TMN histamine pathway → primary cortical wake signal
5:10 AM	Hydrate with electrolytes	250–500 mL water + 300–600 mg sodium + 100–200 mg potassium (LMNT, LMNT-split, or pinch of salt + nuun)	Raises brainstem arousal permissiveness without increasing systemic NE
5:12–5:16 AM	Micro-movement	Slow pacing in home, calf raises, neck/shoulder rolls (very low intensity)	Proprioceptive input → orexin neuron activation → supports wakefulness without adrenaline
5:15–5:20 AM	Take Modafinil	12.5 mg (¼–⅛ tablet) with warm tea/warm water; optional small toast/peanut butter	Now modafinil reinforces orexin–histamine state (instead of replacing it), preventing head pressure / crash
5:20–5:50 AM	No cognitive load yet	No email / news / planning. Light household wandering, showering, or dressing only.	Allows PFC network connectivity to stabilize before cognitive demand
5:50 AM onward	Begin first cognitive or organizational task	Choose something structured but not emotionally loaded	Prevents PFC from “overheating” early in the day

Night Before (Stabilization Layer)

Time	Action	Dose	Role
8:30–10:30 PM	Guanfacine ER*	0.5 mg nightly	Increases prefrontal cortex functioning, reduces morning neural static, and lowers metabolic cost of cognition the next day

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

After 4–7 Days

If mornings are clearer but task initiation still resists:

Add	Timing	Dose
Bupropion XL	With breakfast (after modafinil)	37.5–75 mg/day

Do not add earlier — we first stabilize brain-state, then add drive.

Signs the Protocol Is Working

You will not feel “stimulated.” Instead, expect:

Domain	Change	When
Waking ability	Less internal fog, easier to stand and start moving	Days 1–3
Attention	Smoother, less friction shifting into tasks	Days 3–7
Fatigue	Morning fatigue window shortens; afternoon not worsened	Week 1–2
Crash risk	Less “cost” after focusing	Week 2+

Indications that the Dose Is Too High

(We adjust downward, not upward.)

Pressure behind eyes / forehead → modafinil dose too high → go back to 12.5 mg
Midday crash (~1–3 PM) → modafinil timing too early → shift modafinil 10–20 min later
Sleep becomes too light → move guanfacine earlier in the evening (8 pm instead of 10 pm).

Other Mitochondrial Enhancers

ChatGPT’s list of mitochondrial enhancers was rather short, so I asked it whether others (urolithin A, aceylcarnitines, methylene blue, oxaloacetate, D-ribose, or rapamycin) could help with the NE-filling problems in the vesicles.

Because anything that raises neuronal ATP could indirectly help, urolithin A, aceylcarnitine (250–500 mg once–twice daily), oxaloacetate, and methylene blue (.5 – 2 mg/kg daily) might be able to help. If rapamycin is dosed too aggressively, it could dampen catecholaminergic output and add metabolic friction.

D-Ribose, though, hit the sweet spot. An older supplement introduced by Dr. Teitelbaum decades ago to the ME/CFS/FM field, ChatGPT believes D-ribose “directly targets the core energetic bottleneck that underlies the low norepinephrine (NE) vesicle-filling issue in ME/CFS, long COVID, and related autonomic disorders.”

ChatGPT suggested starting with 1-2 grams once daily and building to 5 grams 2-3 times a day, taken with food or after light exertion. Combine it with magnesium, CoQ10, carnitine, and B2/B3. Note that it can temporarily lower blood sugar. D-Ribose will be added in the final protocol.

Mast Cells

Since mast cell degranulation can inhibit NE synthesis, mast cell treatments are part of the protocol.

Mast cell activation can impair NE filling in several ways. Because mast cells co-occur with sympathetic nervous system neurons across the body, stabilizing mast cells can be a “foundational” part of improving NE level. Mast cells reduce ATP, produce free radicals that can damage the transporter, drain NAD+ and promote firing of the neurons in the locus coeruleus.

Reducing mast cell activation doesn’t directly fill the vesicles, but it does remove brakes on the system, which makes it difficult for filling to occur.

Again, we see different protocols for those who feel worse off in the morning or later in the day.

Worse in the Morning

MORNING (Prime ATP levels and stabilize the mast cells Mast Cells)

• CoQ10 50–100 mg / D-ribose 3–5 g / Riboflavin 20 mg/ Cetirizine or loratadine (H1) / Famotidine (H2) / Magnesium glycinate 100 mg

Rationale: Ribose + CoQ10 = ATP → vesicle refilling improves. H1/H2 block = ↓ nitric oxide + ↓ histamine → mitochondria run more cleanly. Magnesium stabilizes mast cells AND autonomic tone

MID-MORNING (If you feel fatigued / foggy)

Optional: • tiny caffeine (20–40 mg) (if your ATP levels are OK) / • small salt/electrolytes• L-tyrosine (100–300 mg) if tolerated.

Rationale: Low-dose caffeine only works if ATP supply is stabilized. Tyrosine raises dopamine for vesicle filling. Electrolytes help perfusion.

MIDDAY (Autonomic stability)

• Cromolyn sodium 30 min before lunch, Mestinon 30 mg with meal; continue hydration / electrolytes

Improves perfusion → stabilizes LC activity.

AFTERNOON (Refuel vesicles)

• D-ribose 3–5 g, CoQ10 50 mg, H1/H2 if symptoms flare, very light movement / horizontal rest cycles

Rationale: Replenishing ATP here prevents evening crash and “empty vesicle firing”.

EVENING

• Ketotifen (if tolerated) OR quercetin/luteolin, Magnesium glycinate, Taurine, or glycine

Rationale – provides better sleep depth → LC resting → improved vesicle refilling overnight.

Other Drugs / Supplements

Caffeine and Stimulants

Many of us use caffeine or other stimulants to get a boost, and caffeine may help some people with exercise.

Can Caffeine Help with Exercise in Fibromyalgia, ME/CFS and Long COVID?

ChatGPT asserts that while caffeine and most stimulants provide a short-term boost in NE output, using them risks depleting NE levels in the vesicles later, worsening fatigue, and causing a crash.

Using caffeine comes with a cost. There are ways to make caffeine use safer, though.

Stimulants fall into the same group. While they vary in their effects, in general, they increase locus coeruleus firing frequency, which, according to this scenario, we don’t want to happen.

Adderall and Vyvanse empties the vesicle of NE, which makes them more of a no-no. Note, though, that a small Vyvanse ME/CFS study was pretty darn successful.

Stimulating Results in Vyvanse Chronic Fatigue Syndrome (ME/CFS) Treatment Trial

Methylphenidate/modafinil – which blocks the reuptake of NE – preserves NE vesicle content “somewhat”. These agents appear to be safer, particularly if they are paired with mitochondrial support, a key part of Jon Kaiser’s approach. Kaiser has found that combining methylphenidate with supportive nutrients is quite helpful. and helps avoid stimulant based crashes. His small, open-label clinical trial worked in Gulf War Syndrome but failed in ME/CFS largely because of high placebo effects.

Synergizing Health? The K Pax Chronic Fatigue Syndrome Study Results are In…

Atomoxetine inhibits NE uptake in the prefrontal cortex, potentially raising NE levels there. Because MAO inhibitors reduce the breakdown of NE (and dopamine), they can be helpful if ATP levels are sufficient. Otherwise, they may result in the spillover of NE into the bloodstream.

Safer Caffeine Use? Since I often use caffeine to get going in the morning, I rather desperately asked ChatGPT to suggest ways to more safely use it in these diseases.

ChatGPT suggested waiting for 90 minutes after waking up and then taking small amounts (20–50 mg (¼–½ cup coffee or 1–2 oz espresso) or cold brew (smoother pH, slower absorption), which blocks adenosine (the sleep chemical) partially without producing a massive vesicle release.

Or use green tea/matcha, which has L-theanine and ECCG and promotes α-wave relaxation, slower uptake, and a gentler sympathetic effect.

Fifteen to 30 minutes before the caffeine, though, Chat GPT suggests plumping up the mitochondria with CoQ10/Ubiquinol (ubiquinol is best) (50 mg), riboflavin (10–20 mg), and magnesium glycinate (100 mg), staying hydrated, and including electrolytes/salt if POTS-prone. Add nicotinamide riboside or acetyl-L-carnitine if fatigue-dominant, and L-theanine (100 mg) or taurine (250 mg) if anxiety-prone. Every now and then, do light movement (a brief walk or a breathing exercise) to increase cerebral perfusion.

If you crash, take 500 ml water + electrolytes, 200 mg magnesium glycinate, 100 mg CoQ10 or ubiquinol, 10 min light breathing / vagal stimulation (humming, slow exhale) to help re-acidify vesicles and restore parasympathetic balance.

Adenosine Reset – Take 1-2 days off every week to allow for an adenosine reset. I recently went on probably a 2 month-plus caffeine binge in which I took it every day in the morning/early afternoon. My sleep and energy steadily declined. I took a 2-day caffeine break and felt much better.

• Use micro-doses: 20–50 mg (¼–½ cup coffee or 1–2 oz espresso).

• Stop at ≤ 100 mg total/day unless specifically tolerant.

Lower doses block adenosine partially—enough for focus—without forcing massive vesicle release.

GLP-1 Agonists

What about the talk of the town – GLP-1 agonists? ChatGPT stated that the effect these new drugs have on NE filling in the brain is extremely speculative. Their ability to reduce inflammation in the brain, support the mitochondria, reduce oxidative stress, and protect nerves looks promising. Their ability to increase heart rates, alter autonomic nervous system functioning, and engage locus coeruleus circuits – thus increasing their workload – on the other hand, could cause problems. We need more information.

Using Brain Norepinephrine-Enhancing Drugs

NE production occurs in the small locus coeruleus nucleus in the brainstem and spreads across the entire brain. (Image from Bruce Blass, Wikimedia Commons)

The problem with central nervous system NE-enhancing drugs is that they can all raise NE levels in the body as well. Some brain-enhancing NE drugs, called selective norepinephrine reuptake inhibitors (NRIs), though, preferentially increase NE levels in the brain while only minimally impacting the body. These are the ones ChatGPT believes might be helpful but only in the later stages of the protocol. They all work in different ways.

Atomoxetine (Strattera), reboxetine, and viloxazine ER are used in ADHD to increase prefrontal cortex signaling. Bupropion (Wellbutrin) is a dopamine enhancer (reuptake inhibitor) that can be cognitively energizing without producing sympathetic nervous system (SNS) arousal. Modafinil/Armodafinil increases wakefulness and both NE and dopamine without activating the SNS as well.

Notice, below, how many of these drugs increase NE levels in the prefrontal cortex (PFC) – the home of executive control. The prefrontal cortex, with its ability to rein in and calm down the limbic system and improve autonomic nervous system functioning and emotional lability, is a part of the brain we dearly want to enhance.

ChatGPT’s suggestions for NE Enhancing Drugs in ME/CFS and Long COVID (only use when ATP levels are good)

First-Line Drug to Increase Attention + without Causing More Fatigue

Bupropion XL (Wellbutrin)

Mechanism: Mild DA/NE reuptake inhibition in PFC; minimal peripheral NE
Effect Profile: Improved initiation, drive, focus, sometimes energy, low anxiety risk.

Low-dose titration (ME/CFS / dysautonomia protocol):

Week	Dose	Notes
1	37.5 mg/day (¼ of 150 mg tablet or 75 mg SR split)	Evaluate overstimulation / HR
2	75 mg/day	Most benefits appear here
3–6	100–150 mg/day max	Avoid going above 150 mg in fatigue disorders

What to watch:

Monitor HR changes week-to-week, not day-to-day
If mild HR ↑: add magnesium glycinate ~300–400 mg nightly
If HR > 12–18 bpm over baseline → slow taper down one step.

Why this could work:

Bupropion increases task initiation without invoking fight-or-flight NE load. Works well with low-dose guanfacine.

To Improve Attention + Cognitive Stability / Low Sympathetic Tone

Atomoxetine (Strattera) (NET inhibitor, central > peripheral) – It’s too activating at normal doses, but at very low doses it selectively improves NE signal-to-noise in the prefrontal cortex

Dose	Effect
5–10 mg/day	Improved working memory, reduced distractibility, no HR rise
18 mg/day	Max typical dose before sympathetic risks emerge (avoid going beyond unless monitored)

IF tachycardia / POTS is present, then add low-dose guanfacine ER.

Guanfacine ER (Intuniv) (Selective α2A agonist in prefrontal cortex)

Enhances NE neural network stability
Reduces noise, increases sustained attention
Protects against cognitive crash post-task.

Dose

Time	Dose	Purpose
Night	0.5–1.0 mg ER*	Sleep depth, reduces next-day “brain static”
Daytime (if needed)	0.5 mg IR	Smooths cognitive stamina without fatigue

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Key Benefit: Does not raise NE → no sympathetic cost.. Works extremely well combined with bupropion or low-dose atomoxetine.

If Fatigue is the Primary Problem

Modafinil or Armodafinil (Wakefulness without NE spike)

Promotes orexin → histamine → cortical activation
Does not rely on NE for wakefulness.

Microdosing protocol:

Week	Dose
1	12.5–25 mg AM
2	25–50 mg AM (stop here if effective)

Avoid 100+ mg — that’s where head pressure / burnout appears in ME/CFS phenotypes.

Combinations of NE Brain-Enhancing Drugs

Different drugs are better at helping with different aspects of cognition. ATP levels should be increased first.

Note how common guanfacine ER is added. Guanfacine improves prefrontal cortex functioning – a big deal – and does not create problems with POTS or tachycardia. (Should be ok for people with hyperadrenergic POTS).

Best Combinations to Improve Attention and Fatigue (w/out excessively activating the sympathetic nervous system)

Combo A: (Most common, gentle, stable)

Bupropion 75–150 mg/day
Guanfacine ER 0.5–1 mg nightly*

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Improves motivation + cognitive clarity without raising HR.

Combo B: (If too much mental stimulation and high distractibility is a problem )

Atomoxetine 5–10 mg/day
Guanfacine ER 0.5–1 mg nightly*

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Strong signal-to-noise enhancement in PFC means less “noise”; i.e. anxiety is present, plus minimal systemic activation of the sympathetic nervous system.

Combo C (If fatigue, particularly morning fatigue and problems with attention, are primary)

Modafinil 12.5–50 mg AM
Guanfacine ER 0.5 mg PM”

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Improves wakefulness/alertness without sympathetic strain, plus stabilizes the prefrontal cortex stabilization. Does not raise NE levels in the body (periphery).

Reducing Anxiety Using this NE-Enhancing Protocol

Too much “noise” in the system is contributing to anxiety in these diseases.

With regard to the NE problem, ChatGPT believes that the key with anxiety is:

Very low, steady cortical NE tone (PFC)
Avoid peripheral adrenergic activation (which worsens fatigue + autonomic load)
Start below psychiatric dosing and titrate slowly.

To reduce anxiety while maintaining the attention + fatigue benefits, shift the balance of norepinephrine signaling, not suppress it entirely. Three levers are adjusted.

1) Strengthen the Prefrontal Cortex – (when strong enough, the prefrontal cortex keeps the highly reactive limbic system under control.) The simplest and most effective move is to increase Guanfacine ER slightly

Time	New Dose	Expected Effect
Night	0.5 → 1.0 mg ER*	Decreases amygdala reactivity, reduces “mental noise,” improves calm focus

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Why this works:

Guanfacine enhances top-down regulation of limbic signals. It doesn’t sedate or blunt — it prevents anxiety signals from amplifying and makes thoughts quieter and emotional reactions less sticky, without reducing alertness.

Important: Increase after 7–14 days on 0.5 mg, not sooner.

2) Add Low-Dose Propranolol Only As Needed for situational anxiety (not for daily use).

Dose	Timing	Effect
5–10 mg Propranolol IR	Take 30 minutes before an event / stressor	Reduces physical anxiety signals (heart tightness, jitter, anticipation tension)

Mechanism: Reduces the anxiety feedback loop to the brain. Does not reduce attention or cognition at these doses.

3) Optional Layer if Anxiety Is Thought-Driven (Not Physical) Try Buspirone (Buspar) — but only ultra-low dose

Dose	Frequency	Purpose
2.5–5 mg 1–2x/day	Every day or only on anxious days	Tones serotonin 5-HT1A calming system without sedation or fatigue

Unlike SSRIs/SNRIs, buspirone does not blunt affect or create apathy and it works synergistically with guanfacine:

Guanfacine → reduces amygdala signal gain
Buspirone → reduces interoceptive threat interpretation.

Together, they are calming without flattening.

4) (Optional, Highly Effective) — Magnesium L-Threonate at night to improve sleep depth and reduces limbic over-reactivity.

Form	Dose	Timing
Magnesium L-Threonate	1,000–2,000 mg	With guanfacine, 60–90 min before sleep

Do not use magnesium citrate (too GI stimulating) or magnesium oxide (poor absorption). L-Threonate crosses the blood–brain barrier → increases synaptic calm.

Matching to Your Anxiety Pattern

Which describes your anxiety best:

Pattern	Description	Best Move
Mind racing / intrusive worry	Thoughts won’t settle	The problem is increased cortical noise – increase guanfacine ER to 1.0 mg nightly to bulk up the prefrontal cortex and calm the amygdala/limbic system and reduce default mode network activity. Key note: the mind gets quieter, not dumber (as with some other drugs).
Physical tension / heart feeling “tight”	Body anxiety > thoughts	Add propranolol as-needed
Global anxious mood / anticipatory dread	Emotional tone feels “tilted”	Add low-dose buspirone
Anxiety only when fatigued	Brain loses control → overwhelm	Keep guanfacine constant; adjust modafinil timing, not dose

📅 Week-by-Week Progression Plan (Mind Racing Variant)

Week	Guanfacine ER Dose	Morning Modafinil	Expected Change
1	0.5 mg nightly*	12.5–25 mg AM	Attention improves; mornings easier
2	Increase to 1.0 mg nightly	Keep modafinil same	Mind racing decreases, emotional reactivity softens
3–4	Stay at 1.0 mg nightly	Adjust modafinil timing (not dose)	Cognitive endurance rises; thoughts feel “linear”, not scattered

*Note that guanfacine ER is not available in 0.5 mg doses, and breaking up an extended-release tablet is not advisable. You can, however, get immediate-release tablets that can be split.

Do not raise modafinil above 25 mg while adjusting guanfacine.

The clarity comes from top-down stabilization, not stimulation.

🧩 Optional — If Mind Racing Peaks During Quiet or Bedtime

Add one of these only if needed after week 3:

Option	Dose	When	Purpose
Magnesium L-Threonate	1000–2000 mg	60–90 min before bed	Eases mental overactivity; improves slow-wave sleep
Buspirone	2.5–5 mg	Late afternoon or bedtime	Reduces anticipatory worry, not physical tension

Both do not blunt cognition.. Both synergize with guanfacine.

🧭 How to Know It’s Working. You will notice:

Thoughts still occur, but they don’t grip
You can stop thinking when you choose
You can leave a thought unfinished without tension
Emotional tone feels less driven, more placed.

This is executive function reasserting control.

Try the 8–20 Second “Cognitive Quieting” Technique

This is not meditation, distraction, or thought suppression. It works by shifting the prefrontal cortex from verbal sequencing mode → sensory precision mode. When the mind starts racing:

Step 1 — Say (Internally):

“Label Once.”

This means: briefly name what’s happening, one time only:

“Thinking fast”
“Planning”
“Future worry”
“Reviewing”
“Evaluating”.

Do not explain. Do not analyze. Just name. Once.

This single labeling step activates the dorsolateral PFC, which is the region that guanfacine strengthens.

Step 2 — Shift to Sensory Precision

Immediately place 100% of attention into one physical sensory detail, for one breath:

Examples:

The feeling of the air at the nostrils
The sensation of fingers touching each other
The shape of the tongue resting in the mouth
The exact weight of your body on the chair or bed.

Focus not broadly, but microscopically.

This shifts activity to PFC Layer III recurrent circuits, the same circuits stabilized by α2A activation.

Step 3 — Lengthen the Exhale

Inhale 4 seconds — Exhale 6–8 seconds, through the nose.

The long exhale:

Stimulates nucleus tractus solitarius
Activates vagal efferents
Reduces amygdala firing volume within 6–12 seconds.

This is not relaxation breathing — it is circuit recalibration.

Key takeaway. You are not fighting the racing thoughts. You are removing the neural fuel source that keeps them looping.

What You Should Feel: Within 8–20 seconds

Thoughts slow without force
The “pressure to finish a thought” disappears
Emotion decouples from the thinking stream
Awareness becomes wider, quieter, more stable.

This is the signature of PFC control reasserting itself.

Drugs that Can Relieve Pressure on the Sympathetic Nervous System But Do Not Raise Brain NE Levels

Drugs that take pressure off the sympathetic nervous system can be helpful by relieving pressure on the NE producing neurons in the locus coeruleus neurons. Not only does that give the neurons more time to build up NE, but it relieves the “noise” in the system, which itself is driving the system a little crazy.

Mestinon

Mestinon is one of several drugs that can reduce pressure on the sympathetic nervous system.

Mestinon strengthens both sympathetic and parasympathetic nervous system activity by inhibiting the breakdown of acetylcholine. By doing so, it indirectly helps with the NE vesicle-loading problem by reducing pressure on the sympathetic nervous system. Mestinon improves how the blood vessels to respond to NE , and by doing so, tightens up the veins, reduces blood pooling, and the preload problem in ME/CFS and long COVID; in short, it gets more blood to the muscles and brain; thus, in some people, it enhances their ability to exercise.

In effect, it improves how the blood vessels respond to NE, thereby reducing the need for it, allowing the SNS to calm down, and reducing locus coeruleus neuron firing. That reduces NE spillover, therefore helping with hyperadrenergic POTS.

Exertion Enhancer? Mestinon (Again) Moves the Needle on a Key Marker in ME/CFS

Note how the NE problem is being attacked from various angles.

Mestinon is stabilizing
Guanfacine is de-noising
CoQ10 + Creatine are re-powering
Vit C is completing NE synthesis
Tyrosine is optional and only used once stable.

Other Sympathetic Nervous System-Calming Approaches

Beta blockers come with their own set of potential issues, but low-dose propranolol, metoprolol, bisoprolol, and carvedilol can all calm the SNS. Ivabradine (slows heart rate) and midrodrine (improves vascular tone) also calm the SNS. Increasing blood volume (fludrocortisone, electrolyte solutions) and using compression stockings and abdominal binders also take pressure off the SNS.

Neuroplasticity

A main aim of neuroplasticity programs that seek to “retrain” the brain is to calm the fight/flight or sympathetic nervous system. It’s clear that at least some people have extraordinary successes with programs like ANS REWIRE, Amygdala and Insula Retraining Program, and the DNRS. Literally dozens of these programs that have sprung up.

While ChatGPT does not believe that neuroplasticity can refill NE stores, by reducing the threat response, it can promote healing when the system is able to do that.

ChatGPT does not believe these programs, though, directly address the core problem – low norepinephrine levels in the vesicles of the locus coeruleus. Because NE vesicle filling is still low, the participant may feel calmer and less reactive but will still tend experience afternoon crashes, non-restorative sleep, brain fog and exertion intolerance. (While this may be true for many people who are participating in these courses, we also know that some people completely recover.)

The programs can affect the control loops that are keeping the system stuck in “threat-mode”. They do this by calming down the limbic system, reducing the firing of the neurons in the locus coeruleus, reducing mast cell activation, and improving parasympathetic nervous system health.

Reducing threat signals – a key part of neuroplastic programs – allows for cleaner, more efficient NE release. Breathing techniques such as slow diaphragmatic breathing with longer exhalations, micro isometric exercises (handgrip exertion at 30% of capacity), attention training and rewiring how we react to body sensations can all increase parasympathetic nervous system functioning and reduce the arousal that’s helping to keep the locus coeruleus neurons in a state of chronic activation.

Started too early, ChatGPT believes that neuroplasticity protocols can feel fake or forced. Once the NE levels improve, though, and the locus coeruleus calms down and glymphatic clearance improves, then “neuroplasticity gains become stable instead of fragile”. Again, I have heard, from time to time, of very, very ill people who have improved dramatically on these programs.

ChatGPT even predicts the physiological effects these different programs are having vis a vis the NE filling problem.

Pairs well with glymphatic recovery.

Program	Strength	Notes
Gupta Program	Best for limbic threat loop	Works well once LC noise is reduced
DNRS	Best for sensory and autonomic reactivity	Helps reduce mast-cell-triggered flares
Safe & Sound Protocol (SSP)	Best for vagal tone retraining	Strong if sound sensitivity is present
NeurOptimal / EEG neurofeedback	Best for LC firing variability stabilization	Helps shift out of “hypervigilance mode”
Somatic Tracking (Pain Reprocessing Therapy)	Best for interoceptive threat-prediction loop

And it suggests the kind of patient which will not likely benefit. It predicts that people who have

Morning pressure/fog that lifts slowly
Feeling tired-but-wired at night
Visual/sound overwhelm
Temperature sensitivity / adrenaline flares

will probably not do well (???). Once the brain has regained enough flexibility to learn a new state, the neuroplasticity program will stick. ChatGPT provides three signals that it believes can track how well neuroplasticity is moving the brain out of threat monitoring into restoration.

Each should be assessed 30-60 minutes after wakening.

How long does it take for your mind to come online enough to shift attention deliberately? Five to 20 seconds is a good sign.
When you hear a sound, see light, or notice motion do you feel like you react to it, or you absorb it?
If your attention gets pulled (sound, thought, interruption) how much effort does it take to return to what you meant to be doing?

If any of the three improves, even slightly, and not every day, neuroplasticity is taking hold. Progress in neuroplasticity usually looks like this:

“Nothing → nothing → nothing → tiny shift → tiny shift → slightly more → suddenly noticeable”

It asserted that most neuroplasticity work fails not from doing too little — but from doing too much too soon in a nervous system that is already in LC hyperfiring + glymphatic-off + metabolic-deficit mode. “The goal is to train the state, not to push the state.”

The “Glue” – Pacing

ChaptGPT believes that pacing is the “behavioral glue” that lets vesicles refill, keeps catecholamine spillover down, and prevents the post-exertional spiral as this program is put into effect.

By preventing the sympathetic system from going into overdrive and producing lactate spikes, norepinephrine demand lowers, giving the dopamine transporter the window it needs to boost ATP levels and refill the vesicles with NE.

Final Notes

I asked ChatGPT to summarize its approach. Check it out here – ME_CFS_treatment_regimen_framework

Support Health Rising and Keep the Information Flowing!

Health Rising is not a 501 c (3) non-profit

Health Rising’s Quickie Summer Donation Drive is On!

Please support Health Rising during our quickie summer donation drive. Our goal is to raise $15,000.

Find out more here.

163 Comments

Submit a Comment Cancel reply

Skip Pridgen on November 17, 2025 at 4:16 pm

None of this will work. Need to REVERSE the underlying cause.
Reply
- Cort Johnson on November 17, 2025 at 6:46 pm
  
  Do tell 🙂
  Reply
- Roonie on November 18, 2025 at 11:21 am
  
  Im with skip…find the cause
  Reply
  - Brian on November 18, 2025 at 2:24 pm
    
    Until then do what?
    Walk and chewing at the same time is too much?
    
    A nonsensical comment.
    Reply
Mel on November 17, 2025 at 4:24 pm

0. GROUND RULES
• Work with a clinician who understands ME/CFS, POTS/OI, and mast cell activation.

I’m out. I don’t have a doctor who understands any of these, nor will I be able to get one in the near future. Ever since I was diagnosed with fibromyalgia, ME/CFS, and hEDS, I’ve pretty much been on my own, except for my psychiatrist who has worked with me to try everything in his power to help with my sleep issues and depression caused by being someone with a chronic illness who no doctor wants to work with as a patient.
Reply
- Jo on November 17, 2025 at 9:27 pm
  
  Hi Mel,
  Not sure if this might help you with sleep. My GP prescribed me a very low dose of 0.5mg clonazepam, of which I take 1/3 to 1/2 tablet before bed. I have been taking this for many years with no side effects. Please discuss with your doctor first 🙂
  Regards,
  Jo.
  Reply
  - Nancy Nichols on November 29, 2025 at 8:04 am
    
    I am on a great program with Mestinon, adding in midodrine, very controlled exercise pacing, bupropion, now starting ubiquinol, and I am better. Hydrate intensely and also electrolytes daily. I have a very good doctor – neurologist- and I am actually feeling better!!
    Reply
    - Cort Johnson on November 29, 2025 at 9:44 am
      
      Congratulations on a) finding a neurologist who can treat these diseases and b) your progress 🙂
      Reply
    - Shubham on December 11, 2025 at 5:37 am
      
      Can you pls share your whole protocol ? I have low blood sugar issues too but sugar is not healthy for the system. Mestinon does not help me such. My mysthania gravis and all antibodies test are fine except ANA which is positively skewed. Also I have neck arthritis making symptoms worse. I do get a lot of brain fog and nothing seems to be helping. Brain MRI is fine , though hypometabolism in PET scan in pre frontal cortex. I am taking Concerta too but it doesn’t seem to help with pre frontal cortex issues. I am on Seroquel for sleep but I want to wean off it. Pls suggest any thing for sleep
      Reply
- Heather on November 21, 2025 at 5:39 pm
  
  Same. Can’t find doctors that want to help chronic fatigue patients. I just went to an alternative medicine doctor for chronic fatigue, Lyme, Gerd, MCAS etc and he just wants to do food allergy testing. Not cheap and not helping my current situation.
  Reply
  - Shubham on December 11, 2025 at 5:40 am
    
    Food allergy symptoms may help. Get mold testing done though. I also have Lyme 2 bands positive but no one wants to take the long course of antibiotics action and CDC also takes into consideration 5 bands and more as positive.
    Reply
    - Heather on December 12, 2025 at 9:52 pm
      
      I had 5 bands and have been tested three times. No one believes that I have Lyme. My worst symptoms are unbearable neck and back pain with the worst GERD of my life. The acid reflux is so bad I’ve lost a ton of weight because I can hardly eat and react to all foods now (even smells). Anyways, test myself for mold through an OAT test? Or test my house for mold? Thanks in advance!
      Reply
- JR on November 24, 2025 at 8:05 am
  
  I’m glad you’re having a psychiatrist who’s supportive.
  Sharing some experiences what works for me regarding sleep, in case anything of it might help:
  My “basic” protocol for sleep currently is 500mg L-Tryptophane, low dose Trimipramine (ca. 6-7mg) (sometimes also an extra dose of ca. 15 mg trimipramine taken at the time energy runs out during the day to help me rest), ca. 2.5 mg of Melatonin, plus 3.5 mg Nicotine patch, all applied before sleep. (After a period of adjustment, a 3.5 mg dose of nicotine taken at night became calming for me and made sleep more restful, a higher dose however would be too much for me before sleep.)
  Also I try to drink 1 pot of (non-caffeine) tea earlier during the day already, so that I can sleep longer without needing the toilet in the morning. I recommend L-Tryptophane in particular because I think it might also positively influence mood (but if so, best buy a reputable brand because I’ve noticed differences between different German brands of L-Tryptophane).
  
  But right now I’m trying to work with my Fitbit with HRV monitoring to further improve sleep quality: HRV is measured during the night, for me HRV also about correlates with ME/CFS state of fitness (higher HRV is good), and whenever my HRV suddenly jumped up from one day to the other, I looked up in my supplements notebook what supplements I took on that day, and am now experimenting to try and reproduce or narrow down the effect. Some supplements that I suspect of further improving restorative quality of sleep are: L-Theanine; Gingko biloba extract (Tebonin brand in Germany); Ibuprofen; D-Ribose (taken before sleep). Need to experiment further though as for example Tebonin I also suspect of causing alertness before sleep.
  Magnesium threonate (= “Magtein”) seems to prolong sleep but not improve sleep quality for me, so it might still be useful in combination with other supplements.
  In my experience, it’s trial and error to find the right balance of sleep supplements and also go by gut feeling from day to day how much “downers” are required on a specific day (too much, and there might be enhanced dreams or one might wake up too much “dampened”.)
  Out of the above, only Trimipramine is a prescription medication, so, as you probably know well from “being your own doctor”, there are things that can be done for sleep with the help of supplements. And none of the above are considered addictive (not even nicotine patches, thought it must be said there isn’t really any data on long-term use of nicotine patches).
  Trimipramine is an imperfect but important medication for me, as I also use it to treat crashes (15-20mg) in combination with nicotine patches (+/- 7 mg for me in case of crashes).
  A problem that I have not yet solved yet for myself is that Trimipramine (which is an old-school antihistamine too besides being an antidepressant) is prone to drug interactions with other drugs – please carefully check for possible drug interactions before use and watch out for possible drug interactions even if not listed: Though for example antihistamines like Levocetiricine have a slight anti-anxiety and hypnotic effect for me (suggesting there might be some mast cell activation at play), in my own experience I do not feel I can safely combine even the more selective Levocetirizine with Trimipramine any more (I think I could in the past, but getting severe with ME/CFS, as well as an accidental bad drug interaction event in the past that involved antihistamines among other medications seem to have heightened medication sensitivities for me).
  Plus of course some common “sleep hygiene” like avoiding caffeine, using blue light blocker (f.lux) on the screen…or sometimes calling a mental health helpline (Telefonseelsorge in Germany) as needed to relieve stress.
  
  Fitbit is also useful for showing how much deep sleep one gets under current medication.
  
  Last but not least, I am experiencing distinct extra hormonal depression and possibly sleep trouble since the start of menopause troubles, and am trying to treat both hormonally (starting with off-label Slinda (drospirenone) which does actually help to some extent; with a view to upgrading to progesterone later, and possibly try a herbal mix which a relative recommended). I got the prescription from an online gynecologist (since about 2 yrs, it is possible to get online prescriptions via DocCheck App in Germany).
  
  All in all, I guess sleep medications can be very individual – for example I have a friend with chronic pain who takes one prescription drug to go to sleep plus another one to sleep through the night.
  
  Just leaving this here in case anyone finds it helpful for own sleep management.
  Reply
dave ward on November 17, 2025 at 4:32 pm

I recently saw an item on over 60s, re coffee first thing, and not to do it on an empty stomach. Can’t recall why, now, just a mental note.
Reply
Willgb305 on November 17, 2025 at 4:35 pm

I guess we won’t know unless we try – any reduction of symptoms means better quality of life and increased function.
Reply
- Waiting on November 17, 2025 at 6:11 pm
  
  I agree – and in any case, most ME patients I know of try all kinds of things on their own – so many do not have an ME-literate doctor (there are so few) and patients just don’t want to wait around forever for *the* evidence-based treatment.
  
  That said, of course this means there is more risk than if guided by a good doctor. Even then, there is risk because patients can have adverse reactions to anything. So, I would say to proceed with caution even if you have a good doctor – and more so if you don’t.
  
  Also, lots of supplements, prescription medications and other therapies mentioned in this blog have been around & recommended by ME-literate doctors a LONG time – I think the value here is that there is guidance (gleaned by ChatGPT from a vast array of information online) on what to do in different circumstances and what dosing to try.
  
  Sometimes when & how much of something to take makes the difference between treatment success or failure.
  Reply
Gary Reeves on November 17, 2025 at 4:55 pm

This was so interesting that I also got a second opinion:

https://x.com/i/grok/share/xQlkC8rnh8VQBU0mj8jXPE7Xu
Reply
Alicia Butcher Ehrhardt, PhD on November 17, 2025 at 5:39 pm

Even the IDEA of trying a huge array of medications and a protocol horrifies me. I overreact to most medications, and it takes me forever to get over having taken them after they don’t work – and make me sick.

Based on a bunch of algorithms – and stolen fiction/non-fiction* book and internet garbage contents – by a system KNOWN to LIE and MAKE THINGS UP (and apologize like a 5-year-old when caught at it)? *(Including one of my ancient physics papers.)

Just because you can query these things and get an answer?

I’m bad enough already.

And also agree with Skip below that we need to understand the CAUSE first.
Reply
- Cort Johnson on November 17, 2025 at 6:40 pm
  
  Not everyone is like you, Alicia. I know that you’ve had really bad experiences – and I appreciate the caution but not everyone has had these experiences. I, for instance, don’t tend to respond well to medications but I also don’t respond negatively, and many people are able to tolerate and even at times benefit from extensive treatment protocols.
  
  That’s basically what ME experts provide. Since, as you note, we don’t know the cause, we’re left with trial and error.
  
  As I’ve reported and shown before, the queries I put into ChatGPT get almost wholly answered by medical journal articles not “internet garbage”. I have checked dozens and dozens of citations put forth by AI and have yet to find one that wasn’t real or relevant.
  
  AI does best when asked precise questions which is what I use it for. It is at its worst when asked general, open ended questions.
  
  Be forewarned ChatGPT will continue -used responsibly – to inform the blogs.
  Reply
  - Yo on November 17, 2025 at 8:52 pm
    
    Keep in mind a lot medical journal articles are in fact garbage.
    Reply
    - Cort Johnson on November 17, 2025 at 9:05 pm
      
      Where does a statement like that leave us? It leaves me believing that it would be best to find a way to ferret out bad studies and focus on better studies. Other than that, though, it’s simply deflating. Whatever the problem with scientific studies they are still our best access to understanding what’s going on in these diseases.
      
      I am going to keep in mind that studies are often small and limited in scope and can cover only so much. In general, because the results have become more consistent over time, I believe, in general, we’re on the right track.
      
      It may be that in 20 or 50 years people will be amazed at how primitive our understanding of the human body is – I sincerely hope they are! – but this is what we have now.
      Reply
    - Roonie on November 18, 2025 at 11:24 am
      
      👆yes and written by ghost writers. Ghost writing has been going on for DECADES
      
      FOLLOW THE MONEY TRAIN
      $$$$$$$$$$$$$$$$$$$$$$$
      Reply
    - Brian on November 18, 2025 at 2:32 pm
      
      evidence?
      
      None? Yup, it is just your “opinion” of course.
      Reply
    - Roonie on November 20, 2025 at 11:24 am
      
      You,you are correct .they want to keep us confused
      Reply
    - Katrina Berne on November 28, 2025 at 7:06 pm
      
      AI is wonderful in providing information and synthesizing the information it finds. It isn’t good at evaluating the legitimacy of studies and extrapolating from them, and of course it doesn’t take into account a patient’s makeup or medical history, which are highly significant in creating an individual treatment plan. It’s a challenge to use AI for its strengths and to recognize its liabilities. It’s essential to mention when it is the source of information – a signal to take the results with a grain of salt.
      
      In my 4 decades as an ME patient, my reactions to new suggested therapies have changed. I used to try many of the new treatment options that came along. In most cases I had no response. In some cases there were deleterious side effects that caused me to discontinue. A few therapies helped temporarily. Now I react more strongly overall, and more specifically, new drugs are likely to trigger a strong negative reaction that’s much like PEM. I’m not as likely to jump on the bandwagon early. I’ll sit (or lie) back to see what develops.
      
      To those disinclined to trust study results, please recognize that no one study finding provides a complete picture, and some are dead wrong. That is the nature of science advancing in fits and starts.
      
      Cort, as always I appreciate the digging and reporting you do. The answer, or more likely answers, will be uncovered at some point. It requires a great deal of digging and experimenting to get us there.
      Reply
  - Alicia Butcher Ehrhardt on November 17, 2025 at 10:11 pm
    
    Every single thing I’ve ever gotten from AI requires hand checking, and is often either completely wrong or off somehow.
    
    I’m sorry my experience is so different from yours.
    
    People are giving the things credence way too soon, IMO – I’m sorry I am so negative about it, but I am, and will be for a long time.
    
    The user’s INTENTIONS are often excellent – such as this post – and if the things were reliable and accurate they can be more complete than humans, because they don’t get tired, and consider everything. I’m hoping more and more of the ‘Expert Medical Systems’ will be validated.
    
    Let’s compare notes in a couple of years.
    
    I appreciate that you give every possible idea which might help us space and attention – I hope your own quest is fulfilled soon. I just know how badly I react.
    Reply
    - Cort Johnson on November 18, 2025 at 8:30 am
      
      Thanks, I appreciate your reply very much. I was really stung by your first comment. These types of blogs, in particular, take an enormous amount of work.
      
      The reason I use ChatGPT is that my experience with it has been very good. With regard to this blog I was both encouraged (and a little disappointed (lol) that ChatGPT wasn’t more creative. It largely provided protocols that are in line with what ME and long COVID doctors are doing
      
      Here’s what a different AI engine said about the treatments suggested
      
      “This protocol is a thoughtful, evidence-informed framework that aligns well with current research on ME/CFS and Long COVID pathophysiology.”
      
      I will continue to explore these AI engines pros and cons find ways to either get around them (there are practices that can be helpful) and check on them and understand more on where not to rely on their answers.
      Reply
      - Brian on November 18, 2025 at 1:56 pm
        
        AI can be a useful tool for research and clinical professionals. Cort, you are treating it as a qualified medical professional, given you valid advice. It is a lazy, and frankly unethical, approach.
        
        I highly respect many of you posts, but not this one
        Reply
        
        Mari on November 19, 2025 at 11:53 pm
        
        I agree. I was disappointed with this.
        
        Especially since it isn’t true that AI doesn’t have biases. It absolutely does. It has been trained on texts that are full of biases.
        It has recently been shown that it gives people it reads as women different medical advice than people it reads as men e.g. regarding what to do with symptoms that could be a heart attack.
        
        AI can absolutely be useful but it is important the user understands what it is, what it can and cannot do. Ultimately it is a monkey with a type writer, except what comes out sounds plausible. But it isn’t necessarily correct.
        Reply
        
        Cort Johnson on November 20, 2025 at 12:56 pm
        
        Especially since it isn’t true that AI doesn’t have biases. It absolutely does. It has been trained on texts that are full of biases.
        
        This is not a reason, Mari! The reason that it’s not a reason is that whether it’s AI or a researcher or doctor or myself, we are all using these same texts (with their biases) to learn about these diseases. (Removing as much bias as possible is why research articles go through a review process so hopefully the biases are less in medical journal articles – which AI uses with my queries.)
        
        It has recently been shown that it gives people it reads as women different medical advice than people it reads as men e.g. regarding what to do with symptoms that could be a heart attack.
        
        I don’t know what that refers to but the heart attack symptoms for women are different from men. Again, if this is an error, it could probably not come from a peer-reviewed study.
        
        Do you see anything untoward in the suggestions?
        Reply
        
        Brian on November 24, 2025 at 11:35 am
        
        “Do you see anything untoward in the suggestions?”
        
        Yes, providing medical advice without the expertise or qualifications to do so, with patients who you know are suffering, and, are therefore, vulnerable.
        
        APA
        “Due to the unpredictable nature of these technologies, do not use chatbots and wellness apps as a substitute for care from a qualified mental health professional.”
        https://www.apa.org/news/press/releases/2025/11/ai-wellness-apps-mental-health
        Reply
      - Alicia Butcher Ehrhardt on November 18, 2025 at 6:16 pm
        
        So sorry about my tone – you pushed so many buttons I overreacted.
        
        I think the ‘AS’ are not ready for public use, and those who would rely on them mostly don’t know what they’re doing and lack the capacity to do adequate vetting – and without adequate vetting the things are dangerous.
        
        I’m thinking this is an EXAMPLE for a hypothetical case, which should be vetted and supervised by a knowledgeable physician (if you can find one!). Individual cases would have their own data for the inputs.
        
        However, I will not be being a guinea pig – and I have never had the pleasure of a knowledgeable physician seeing me, much less ‘managing’ my treatment for ME/CFS – and I have had a range of clueless but trying to be helpful ones who let me try some of the things recommended over the years, none of which worked (which could also be my intolerance of medications which worked for others but make me ill, which seem to be most of them).
        
        AND I’m afraid too many people, medical and not, will assume the answer is already there, and research will have a hard time being funded further… just as we might have been getting somewhere and this administration gutted research.
        
        We’re desperate, and desperate people often make decisions out of that desperation.
        
        I’m glad you’re optimistic.
        Reply
  - Gijs on November 18, 2025 at 3:58 am
    
    https://www.nytimes.com/2024/10/23/technology/characterai-lawsuit-teen-suicide.html
    Reply
    - Cort Johnson on November 18, 2025 at 8:36 am
      
      Yes, a 14 year old developed an emotional attachment to an AI engine and committed suicide AND the blogs focus solely on technical aspects of these diseases. I don’t know how this fits in the “Is ChatGPT giving us worthy answers to questions about ME/CFS, Long COVID, etc.” question
      Reply
    - Cort Johnson on November 18, 2025 at 1:49 pm
      
      Thanks! I liked it’s conclusion “This protocol is a thoughtful, evidence-informed framework that aligns well with current research on ME/CFS and Long COVID pathophysiology.” 🙂
      Reply
- Matthias on November 18, 2025 at 12:57 am
  
  I have a pretty strong hunch that neuroinflammation is at the heart of these issues.
  So that’s the primary factor that needs addressing.
  Interestingly, though, dopamine quells neuroinflammation whilst neuroinflammation quells dopamine…. So a bit of a feedback loop to address
  Reply
- Brian on November 18, 2025 at 1:53 pm
  
  Cort is conveying very detailed medical advice, which should raise RED flags for anyone.
  Reply
  - Cort Johnson on November 18, 2025 at 2:34 pm
    
    As the blog noted these should be considered treatment SUGGESTIONS that came from a ChatGPT “brainstorming” session. They are best followed up with your doctor. I fail to see how these mostly benign suggestions are a danger to people.
    
    They mostly involve supplements and drugs that are well-known in these communities. I think people in this community can handle suggestions about supplements. With regard to the drugs, you can’t get them without going through your doctor, anyway.
    
    So, what’s the problem?
    Reply
    - Brian on November 18, 2025 at 2:39 pm
      
      Cort,
      Perhaps you should read my comment left at 2:29 prior to Trump-like reactivity/threats.
      Reply
      - Cort Johnson on November 18, 2025 at 3:01 pm
        
        My comment stands. If this program was suggesting dangerous or strange treatments, I would agree with you but they are not. Your stance seems to be that you can’t trust anything coming from these programs and that even if the result make sense it’s unethical to use them.
        
        On the other hand, I feel like people with these diseases should have access to a broad range of treatments – whether they come from ChatGPT, or a doctor or even a patient who has found something that works. Each of these presents pros and cons.
        
        As the blog notes doctors are prone to all sorts of cognitive errors. Patients can only provide an N-1 experience and often lack deep knowledge of this field. AI engines make connections that are not valid or it can even hallucinate at times (I haven’t found that with my queries).
        
        My concern is with plausibility. If a patient wants to do a blog on something that is not plausible or might be harmful I might not want to publish that blog or I would put warnings.
        
        I’ve been studying these diseases for decades and while I lack deep knowledge about biological systems I feel I have a sense about how findings might fit in this field. Since, IMO, these findings fit well, and make sense, I published them.
        
        Nobody has to follow them, and, indeed with regard to the drugs, you can’t follow them without your doctor signing off on them.
        
        In the end I think we should have all the information that reaches a kind of undefined level plausibility i.e. does it make sense given what we know about these diseases. To me, they did.
        
        For all the hullabaloo about this blog, only one person (who noted that you shouldn’t split Guanfacine ER tablets) has cited something that is wrong or off. It’s all about – you used ChatGPT to produce so it must be wrong.
        
        I think maybe it should be the other way around. First explore the blog and then conclude whether the process that led to it worked or not.
        
        Health Rising will have a page on how it’s using AI coming up.
        Reply
        
        Deek Martin on December 2, 2025 at 3:21 pm
        
        I agree with Cort. And we are adults and can make choices. we don’t need censorship.
        Reply
- Brian on November 18, 2025 at 2:29 pm
  
  AI is a tool. While your skepticism is valid, your suspiciousness is not (mostly). Just beware of “garbage-in garbage-out” scenarios and the use of this tool by non-experts. I think Cort means well, but he is not sufficiently trained and licensed to evaluate the validity of the AI output, so I disagree with his presenting it as medical advice. That does not mean I don’t respect his great deal of work compiling information from experts in his various blogs, much of which is MANAGING THE SYMPTOMS… Soling seeking THE CAUSE(S) is foolish at this point; besides it is multi-factorial.
  Reply
  - Cort Johnson on November 18, 2025 at 2:42 pm
    
    I do not have a license to validate AI output (are such things available?) but last week I enrolled in a class on using AI to assess research findings.
    Reply
Lisa on November 17, 2025 at 6:07 pm

First off I want to ask ? Who gets up at 5am ? NOT ME ! Secondly I have tried many of these things. Mucuna Puriens and SNRIs leave me feeling really bad like I’m dying . Creatine monohydrate sorta helped me but I have big kidney stone issues and advised not to take it the ones that help me most on this list is Vit C and CoQ10 and cromolyn sodium .
Reply
- Cort Johnson on November 17, 2025 at 6:43 pm
  
  I’m glad) you don’t wake up at 5 am Lisa but some people with poor sleep (like me), do. (I would love to wake up at 7am!) That was based on an individual prescription provided by ChatGPT for me. I forgot to add that in and I just did.
  Reply
  - Lisa on November 17, 2025 at 11:52 pm
    
    I meant to put an LOL after NOT ME! I don’t get up at 5am !
    I am one of the ones that feel like absolute rubbish in the morning. Sometimes it’s because of insomnia ..but often it’s the postural tachycardia and I must stay in bed for at least an hour after waking and drink lots of water and take my metoprolol. I also forgot to mention that a small amount of Cold coffee helps me but it has to be a small amount or I feel jittery and get palpitations ! I just got prescribed Mestinon and will start trialing it starting next week.
    Reply
    - Cort Johnson on November 18, 2025 at 8:39 am
      
      Suggestion from the blog and from Jon Kaiser MD. Try some mitochondrial enhancing supplements beforehand. I don’t know if he still has them but he was producing caffeine + mitochondrial support pills.
      
      And/or try green tea for a possibly cleaner experience.
      Reply
Jachan on November 17, 2025 at 6:11 pm

Interesting, a LOT to take in. I’m curious as to whether any reason was given for the 5 AM wake up?! Would the same stack (theoretically) work with a later (more “normal” imho) waking time?
Reply
- Cort Johnson on November 17, 2025 at 8:26 pm
  
  My mistake. That’s a normal wake time for me – so I asked ChatGPT to make up a schedule for me. I just added that to the blog.
  Reply
  - Jachan on November 19, 2025 at 9:04 am
    
    Haha- thanks for clarifying! I’d need to ask it how to help me even consider waking up at 5 first 🙂
    Reply
Skip Pridgen MD on November 17, 2025 at 7:19 pm

https://youtu.be/dGbb7UDcoHw?feature=shared

PolyBio Fall Symposium 2025
Go to the 3hr 59 min point and David Putrino PhD will fill everyone in. We take a Dogwood Pharmaceutical IMC-2 and ….well watch it and find out
Reply
- Cort Johnson on November 17, 2025 at 8:27 pm
  
  Let’s hope! I assume that’s for long COVID (?). I don’t think we have an answer for ME/CFS and FM.
  Reply
  - Matthias on November 18, 2025 at 1:08 am
    
    It’s long covid-directed, but if you think that viral reactivation is an important aspect of ME /CFS – I don’t, although many do, and I at least acknowledge it’s *possible* if not likely – then it’s plausible that it could help ME/CFS
    Reply
    - Matthias on November 18, 2025 at 1:17 am
      
      https://batemanhornecenter.org/innovative-trials-research-and-clinical-insights-on-long-covid-therapy/
      
      So if you think herpes virus reactivation is key to ME/CFS, then this could plausibly help in ME/CFS as well as long covid…
      Reply
    - Cort Johnson on November 18, 2025 at 8:50 am
      
      Certainly, it is plausible, and I applaud the multi-viral approach to these diseases, which Dr. Pridgen has brought into the field. In retrospect, why would we think that one antiviral was going to do it?
      
      Unfortunately, while I’ve certainly heard of real success stories and know one person who recovered using it, Dr. Pridgen’s duo-antivlral approach, for whatever reason, (results/funding/?) appears to be at a standstill. I could not find any ongoing clinical trials.
      
      Hopefully, the triple antiviral approach will move mountains. The early, preliminary results – as published in Health Rising – were good.
      Reply
      - Matthias on November 18, 2025 at 11:17 am
        
        I guess we need to hear from Dr Pridgen.
        Of course the triple approach probably isn’t relevant to ME/CFS, given one of the inputs is paxlovid – specific to covid
        Of course, there is every possibility that any benefit achieved is purely down to the antinflammatory med
        Reply
Scotty81 on November 17, 2025 at 7:51 pm

Thanks for your reporting over the many years. It looks like this research (Part I and Part II) describes the patients who have hyper POTS – that is those with BP on the high side. So, I understand that you don’t want to raise the NE level in the peripheral nervous system, as I believe you mentioned, in those cases.

My daughter, however, is opposite to the patient population discussed in the research. She does have PEM, but she has chronically low blood pressure. Her catecholamines were also tested and she has low NE, not a high level. So, does the research say anything about what’s going on in the brain in her case? And does that mean all the treatments that you say are contraindicated are actually OK for those cases?

I can say that my daughter has tried many treatments, including supplements, but could not tolerate them. She did try Mestinon, but she couldn’t tolerate the side effects. She is on Northera, a prescription known to increase NE in the bloodstream. That medication does raise her BP to normal levels, yet she still suffers from fatigue, brain fog, and a host of other symptoms. I point this out to note that not every patient is like the ones studied in the research you mentioned.

I do like your “mechanistic review of what’s going on in the brain from Part I. I think that theory just needs to be expanded to patients who behave opposite to those who you described.
Reply
- Cort Johnson on November 17, 2025 at 8:51 pm
  
  OK. So I asked ChatGPT to try to explain your daughter’s situation. I’m just going to dump the whole answer here. It’s a complex but typical kind of answer I get from this AI search engine. Hopefully it will help you, and others can see the type of answers it gives. Plus I learned that someone named Tamara Carnac has a hypothesis about this situation in ME/CFS. 🙂
  
  I asked “how would an ME/CFS patient with low blood pressure and low NE bloodstream levels who cannot tolerate Mestinon, and is on Northera, to increase NE levels in the bloodstream fit into the low NE vesicle and low NE pathway levels picture? Northera does raise her BP to normal levels, yet she still suffers from fatigue, brain fog, and a host of other symptoms.”
  
  Short version: what you’re describing actually fits the “low NE vesicle / low NE pathway” idea pretty well. Northera is patching the blood-pressure side of the problem by boosting peripheral norepinephrine (NE), but it doesn’t necessarily fix:
  
  how much NE is inside the neurons and vesicles
  
  how well those neurons fire in the brainstem and cortex
  
  how the receptors and transporters are behaving
  
  or the many non-adrenergic problems in ME/CFS
  
  …so it’s very plausible to still have fatigue, brain fog, etc., even with normal BP.
  
  I’ll break that down in human language.
  
  1. Quick context (so we’re talking about the same thing)
  
  ME/CFS is now understood as a multi-system illness: immune, metabolic, autonomic, and neurovascular abnormalities all show up, not just “low energy.”
  PMC
  +1
  
  Recent work and hypotheses suggest a strong role for adrenergic (NE) dysfunction—abnormal catecholamine levels, receptor function, and autonomic responses, not just “too much” or “too little” NE.
  PubMed
  
  Tamara Carnac’s “noradrenergic neuron dysfunction” hypotheses (including the later phosphatidylcholine/insulin signaling paper) propose that, in at least some ME/CFS subtypes, NE inside neurons and vesicles is depleted or dysregulated, while extracellular levels can be normal or even high, depending on subtype.
  Preprints
  +2
  Preprints
  +2
  
  So: “low NE vesicle and low NE pathway levels” = noradrenergic neurons that can’t store, release, or recycle NE properly, especially in key brain and autonomic circuits.
  
  2. What Northera is actually doing vs what it isn’t
  
  Northera (droxidopa) is a prodrug that gets converted to norepinephrine by dopa decarboxylase. It’s approved mainly for neurogenic orthostatic hypotension (nOH) to raise blood pressure.
  BioMed Central
  +1
  
  Key points:
  
  It mainly increases NE in the peripheral nervous system, causing vasoconstriction and a rise in blood pressure.
  northerahcp.com
  +1
  
  Pharmacokinetics: shortish half-life (~1.5 hours), with small, transient rises in plasma NE rather than a continuous, high NE level all day.
  northerahcp.com
  +1
  
  Whether it meaningfully crosses the blood-brain barrier is actually debated: some sources suggest it can, others say its effects are largely peripheral.
  PMC
  +1
  
  So Northera is great at:
  
  “Give the body extra NE so blood vessels constrict and BP doesn’t crash when upright.”
  
  It is not designed to:
  
  Restore NE content in central vesicles
  
  Normalize firing patterns in locus coeruleus and other brain NE pathways
  
  Fix receptor sensitivity, transporter trafficking (NET), or insulin-signaling issues in neurons
  
  Address downstream stuff like immune activation, microclots, mitochondrial dysfunction, etc.
  Preprints
  +1
  
  That’s a big reason why normal BP ≠ “cured ME/CFS”.
  
  3. How your pattern fits the “low NE neuron/vesicle” picture
  
  You mention:
  
  ME/CFS diagnosis
  
  Low BP, low NE in the bloodstream
  
  Cannot tolerate Mestinon (pyridostigmine—different pathway, cholinergic)
  
  Northera raises BP to normal
  
  But fatigue, brain fog, and many other symptoms persist
  
  Mechanistically, that’s very compatible with a “low-output noradrenergic” subtype (like Carnac’s “ME3”-type idea with reduced NE synthesis and low intracellular NE).
  ResearchGate
  +1
  
  Here’s what could be going on (conceptually):
  
  Baseline: low NE production & storage
  
  Noradrenergic neurons (in brainstem, sympathetic ganglia, etc.) aren’t making or storing enough NE in vesicles.
  
  That gives you low plasma NE and low BP at rest/upright.
  
  Northera: bypassing synthesis, but only in the periphery
  
  Droxidopa is converted to NE outside neurons (and possibly in some neurons) and temporarily raises circulating NE and BP, mostly peripherally.
  northerahcp.com
  +1
  
  This improves orthostatic hypotension and may help some OI symptoms, but it doesn’t necessarily refill central vesicle stores or repair dysfunctional NE circuitry in the brain.
  
  Central NE pathways may still be underpowered or mis-signalling
  
  Locus coeruleus and other NE nuclei regulate arousal, attention, working memory, effort preference, and autonomic tone. Emerging ME/CFS work points to central catecholamine pathway dysregulation even when peripheral measures are partly normal.
  Preprints
  +1
  
  If neurons inside the brain still have low vesicular NE, they may fire “weakly,” unreliably, or with the wrong pattern. That maps very nicely to fatigue, brain fog, reduced motivation, sensory overload, etc.
  
  Receptors and transporters may be mis-tuned
  
  Even if you raise extracellular NE with a drug, if adrenergic receptors (α/β) or the NE transporter (NET) are up- or down-regulated, internal signaling could still be wrong.
  
  Carnac’s hypothesis and older NET data suggest some subtypes may have NET dysfunction, shifting NE between inside/outside the neuron in abnormal ways.
  Preprints
  +1
  
  So: a “normal” NE blood level after Northera doesn’t guarantee healthy synaptic NE signaling.
  
  Your experience matches other data on droxidopa in dysautonomia
  
  In POTS / refractory orthostatic intolerance, droxidopa often improves orthostatic symptoms and BP, but quality of life and fatigue/cognitive issues may not fully normalize—exactly what you describe.
  ResearchGate
  +1
  
  In other words, your pattern is not a contradiction; it’s actually what we’d expect if ME/CFS involves deeper noradrenergic neuron dysfunction rather than just “low NE in the blood.”
  
  4. Why brain fog and fatigue can persist even when BP looks “good”
  
  A few additional factors that can keep symptoms going:
  
  Cerebral blood flow (CBF) is its own problem
  
  Multiple studies show marked reductions in cerebral blood flow in ME/CFS during tilt testing—even in patients whose heart rate and BP don’t look dramatically abnormal.
  PubMed
  +2
  PMC
  +2
  
  So, you can have normal arm cuff BP but still low blood flow in parts of the brain → brain fog, dizziness, cognitive slowing.
  
  Endothelial and microvascular issues
  
  ME/CFS research points to endothelial dysfunction and abnormal microclots (fibrinaloid microclots, platelet hyperactivation) which can impair tissue perfusion and oxygen delivery.
  BioMed Central
  +1
  
  Fixing systemic BP doesn’t automatically fix microcirculation at the brain or muscle level.
  
  Immune and metabolic load
  
  Chronic immune activation, altered cytokines, mitochondrial changes, altered lactate handling, etc., are all described in ME/CFS.
  Frontiers
  +1
  
  Those can independently drive fatigue, PEM, and cognitive issues, regardless of NE.
  
  So if we zoom out: Northera is helping one slice of the puzzle (peripheral hypotension) but the rest of the ME/CFS machinery is still there.
  
  5. Where this leaves you in the “model”
  
  Within a “low NE vesicle / low NE pathway” framework, your story fits something like:
  
  Primary phenotype: low NE synthesis / storage → low NE in blood and vesicles, low BP
  
  Therapeutic patch: Northera supplies peripheral NE → BP normalizes, OI somewhat better
  
  Residual disease: central NE circuits, receptor/NET function, microvasculature, immune/metabolic problems remain abnormal → fatigue, brain fog, PEM, etc.
  
  So your experience is actually very consistent with the idea that ME/CFS is not just “low norepinephrine” but a systems-level noradrenergic and neuroimmune disorder, where:
  
  Raising NE in the bloodstream helps the plumbing
  but doesn’t fully repair the wiring or the control software.
  
  6. Practical next steps (non-prescriptive)
  
  I can’t give you personal medical advice or change your meds, but conceptually, people in your situation often work with:
  
  An autonomic/ME-literate clinician to:
  
  Optimize non-pharmacologic OI management (fluid/salt, compression, recumbent exercise, sleep optimization, pacing, etc.)
  
  Carefully review whether Northera dosing schedule, interactions, or side effects are aligned with symptom timing.
  Bateman Horne Center
  +1
  
  Clinicians who are familiar with ME/CFS research and adrenergic theories, so they can interpret your low NE + Northera response in that broader context.
  
  If you’d like, you can tell me a bit about which symptoms improve on Northera (if any) and which stay the same or worsen, and I can help you map those to different parts of the “noradrenergic + circulation + immune” picture in more detail—purely for understanding, not for prescribing.
  Reply
  - Brian on November 18, 2025 at 2:52 pm
    
    More AI results conveyed as medical advice. Apparently, Cort now believes he is a medical expert, but without the accountability, as he does not have a medical license. A highly problematic combination. Beware.
    Reply
    - Cort Johnson on November 18, 2025 at 3:15 pm
      
      I don’t have a medical license for any of this, Brian! I’m not a doctor or a researcher. I don’t have a PhD in biology either.
      
      Don’t worry, though, I can’t prescribe for anyone – so I’m not a danger. In order to get drugs – which I assume are the main concern – you have to get your doctor – who is a medical expert – to agree to give you them.
      Reply
      - dejurgen on November 18, 2025 at 4:25 pm
        
        I appreciate your intend to help, but clearly see Brian’s concerns.
        
        At “Don’t worry, though, I can’t prescribe for anyone – so I’m not a danger. In order to get drugs – which I assume are the main concern – you have to get your doctor – who is a medical expert – to agree to give you them.” I reply:
        
        How many people took Ivermectin during the Covid pandemic? How much of them got and took it without proper licenced medical expert? In the US alone that may well be hundreds of thousands, a very lose estimate of me. How they got it? Probably illegal circuits and veterinarians. (Note that I do not chose to discuss the merrits and dangers of Ivermectin nor it’s legal status here, just that evidently plenty of people got something they were legally not supposed to get their hands on).
        
        Confidence is important. Some people may follow your advice based on the confidence they have in you and find a way without a proper licensed professional.
        
        What you hope to help here, at great effort of yourself as you say, you risk to lose among many readers and the professionals that help you spread word on ME/CFS medical progress. I can’t know how the researchers of the study on brain NE / dopamine (previous blog) will feel about this blog, but as a former researcher I can more then imagine that some researchers who shared in the past with you may reconsider how and when they’ll share information outside official publications.
        
        Part of what you worked for so hard in the past, including reputation among researchers, may or may not be damaged due to this type of blogs. Most likely researchers will be sort of divided on the issue too just like patients. Anyhow, I’d estimate that you are here at risk to achieve less in the long run (lifetime of commitment to ME/CFS, FM, … patients) by doing extra effort then if you did nothing at all. That wouldn’t be economical.
        
        Your blog, your choice. This is not meant to be personal. It’s a risk for all of us with the advent of AI technology, that we feel we are given wings without considering well enough if they can support us mid air. Are it Icarius wings or not?
        Reply
        
        Cort on November 18, 2025 at 8:04 pm
        
        I’m going to stick with my comment. Ivermectin is a controversial veterinary drug that sprang to prominence and was easily obtained on the black market; in other words, its a unicorn. I don’t think that worry really applies to these drugs. Maybe the stimulants to some extent but they’re not new to the ME/CFS field at all and are not suggested unless patients are doing better. (Are people hunting for Guanfacine on the black market?)
        
        Plus the drug dosages recommended are almost uniformly low.
        Reply
        
        Roonie on November 19, 2025 at 4:24 pm
        
        Ivermectin, FYI pulled me out of a very dark hole.
        I became, after my first dose ,itchy all over and a week later realized I was 70% better.
        
        I won’t mention any names but a certain friend of mine has used ivermectin and fenbendozole to “CURE”
        his own stage 4 cancer and now two other people.
        You can go on youtube…there are many, many testimonials.
        There is a dr marik (oncologist)in Edmonton ,Canada that began using it….they pulled his license….they don’t want to “cure” any disease because the customer will be lost.
        You can go on youtube and listen to john campbell interviews with dr marik.
        Now the state of Florida is finally doing trials on the use of ivermectin and fenbendozole to successfully CURE cancer.
        The money train is running out of gas finally after decades of people dying needlessly.
        “RUN FOR THE CURE”
        yeah right.Billions of dollars raised for decades and still not even close.
        Same with all other diseases…millions of dollars poured into all these diseases.
        Reply
        
        Brian on November 24, 2025 at 11:40 am
        
        dejurgen, Well said.
        Reply
    - Cort Johnson on November 18, 2025 at 3:20 pm
      
      I don’t think we are going to agree on this but I want you to know that I did get your concern and I am digging into AI deeper. Soon Health Rising will have a page that clarifies how Health Rising uses AI and how its attempting to obviate the problems inherent in it. As I noted, I’m enrolled in a class on AI. Pretty soon I’ll know much more about it.
      Reply
    - Deek Martin on December 2, 2025 at 3:49 pm
      
      Basically, there are no medical experts in this field.. There are few that claim to be, but they all have different opinions. So what Cort is offering here from ChatGPT Is something that we who are suffering can analyze and make our own decisions about what’s best for us. Are you saying we shouldn’t be offered this analysis?
      Reply
      - Brian on December 2, 2025 at 6:04 pm
        
        Thanks for reminding me why I usually do not to leave comments; individuals with uninformed “opinions” who want to argue. You are clearly unarmed.
        Reply
        
        Deek Martin on December 2, 2025 at 10:59 pm
        
        Brian, I didn’t want to argue, I just wanted to support Cort and appreciate his offering.
        Reply
- Nadine on November 18, 2025 at 3:02 am
  
  Maybe she can try Cerebokan against Brain fog it’s Ginko and helps some
  Reply
Deanne on November 17, 2025 at 9:00 pm

I found this very interesting & informative. I appreciate you sharing it, Cort. As others have said, I too have tried many of these drugs & supplements over the years. However, I didn’t do them as part of a regimen with specific goals in mind. Seems to me rejecting it out of hand without trying it is unwarranted, given the AI is drawing from solid information/research..
That said, some of the medications discussed I cannot tolerate. Wellbutrin gave me hallucinations (never happened before or since), & Metoprolol lowers my heart rate too much (haven’t tried a micro dose tho).

Honestly, just adhering to such a complicated regimen requiring a lot of attention to detail & logging symptoms feels overwhelming to me. I don’t have a knowledgeable doctor, enough focus, or energy. However, I’d be interested in knowing if it makes a noticeable difference for others, maybe a future update?
Reply
Laurence on November 17, 2025 at 11:31 pm

CONGRATULATIONS, CORT,
To my knowledge, this is the most CONFUSING article you have ever written.
Rushing names of all these drugs which I have never heard of coming at me, not explained or identified. I feel as though I have just been ejected from a circular
washing machine.
And Chat GPT? It doesn’t care or know what we can understand. Just rush it all out.
No wonder its “results” can’t be tested in a clinical trial. It presents too many to be tested validly. And I am not an MD or a PhD. Like most of your readers.
Please return to HUMAN references. Or you will lose a reader.
And probably not just me.
Reply
- Cort Johnson on November 18, 2025 at 8:57 am
  
  I anticipated a comment like this would show up. I agree that this article is complex – which is why I suggested that you print it out. I did my best to winnow it down so that it was as manageable as possible. Still, it’s a load, I agree.
  
  What can I say? The diseases are complex and the treatment protocols tend to be quite complex.
  
  “Rushing names of all these drugs which I have never heard of coming at me, not explained or identified. I feel as though I have just been ejected from a circular washing machine”
  
  I got it. Actually, I’ve heard of most of the drugs mentioned but that’s me.
  
  Regarding ChatGPT I urge you and others to go by the results and whether they are plausible or not. The reason I’m using it is that my experience over the past couple of months with it has been very good. Its not giving weird answers, it’s backing its answers with plausible rationales, etc.
  Reply
  - Karen on November 18, 2025 at 11:13 am
    
    Thanks for taking on such complex disorders and trying to provide struggling patients with potential treatment options based on symptoms. The brain is complex, and it is no wonder that there is no one size fits all solution. The best we have are possible treatments based on trial and error. A big hurdle for those of us with non-CFS/ME, FM literate doctors will be trying to convince MDs to gain access to some of the prescription drugs. Other than that, I will continue to try cautious trial and error in order try to function better.
    Reply
Patrick Usher on November 17, 2025 at 11:32 pm

It’s very interesting research, thanks Cort. Perhaps we are so easily stressed because our fight/flight system cannot rise to the challenge?

Do you remember the paper by Scott and Dinan from 1999 which found our adrenal glands are 50% the normal size (they hypothesised as a result of faulty feedback loops from the HPA axis which control adrenal gland renewal)? Again, not a sign of our stress-response system being overactive but of it being diminished which, paradoxically, will make us stressed by everything.
Reply
- Cort Johnson on November 18, 2025 at 9:24 am
  
  Perhaps we are so easily stressed because our fight/flight system cannot rise to the challenge?
  
  What an interesting idea! It makes sense to me that if energy production problems occur across the body why wouldn’t both the PNS and SNS be underpowered?
  
  Time will tell with the energy hypothesis but it does make sense.
  Reply
Jay on November 18, 2025 at 12:34 am

What I’ve always wanted in a treatment is for it to be based on nothing except which words are statistically likely to go together, without regard for factual accuracy. /sarcasm

You’ve presented yourself as someone who communicates facts, and this blog post misleads readers about the capabilities of ChatGPT in a deeply irresponsible way.
Reply
- Cort Johnson on November 18, 2025 at 9:21 am
  
  If you feel that way you can just ignore the blog!
  
  You might notice, however, that these treatment suggestions are pretty similar to what you might get if you went to an ME-literate doctor, and ChatGPT provides rationales for all of them. It’s presenting some weird, off-the-charts suggestions.
  
  As I’ve said several times, ChatGPT returns my queries with references to scientific papers, and when I’ve checked, they’ve all been valid.
  Reply
- Brian on November 24, 2025 at 11:45 am
  
  Jay, as you can see, Cort is clearly too thin-skinned to deal with any, even slightly, critical feedback. It’s sad, really. This is clearly his personality structure, and nothing to do with ME/CFS symptoms (excess fight-flight symptoms? aha…); that’s just an excuse.
  Reply
T on November 18, 2025 at 2:40 am

A speculative disease hypothesis with treatment suggestions from an often hallucinating AI that gathers data from Facebook and Reddit and the like…. This is likely to cause as much harm as help tbf.
Reply
- Cort Johnson on November 18, 2025 at 9:03 am
  
  As I’ve now said – and even demonstrated – quite a few times, the questions I ask these search engines are technical in nature – and it responds in kind – with results, the vast majority of which – come from research papers – which it cites.
  
  I have checked dozens of articles cited by ChatGPT and have yet to find any that are not real.
  
  AI search engines are at their worst when asked general questions and at their best when asked precise questions.
  
  Because the questions I ask are so technical “does ampligen affect TLR4 pathway” or please explain this “”We have developed a novel epigenetic assay (EpiSwitch® Explorer Assay), a bespoke design based on Agilent SurePrint 1 M array, that allows simultaneous screening for 106 of 3D chromosomal” REddit or Facebook or twitter or whatever don’t have anything to offer.
  
  Again, I would judge it by the results in this particular instance not by some idea you have in your head. Do the results in this blog make sense, or do they look like they’re the product of a hallucination?
  
  If you read the blog, I think you’ll find that the results are pretty much in line with what many doctors are doing.
  Reply
Brayden Finch on November 18, 2025 at 2:41 am

Hi Cort,

Thanks for the articles on this. I think the NE theory sounds really promising, and the direction you have given here in Part II is very much appreciated. As someone who uses AI a lot to work through this stuff (and who has spent a lot of time learning to understand LLMs and their strengths and limitations, as it is clear you have), this post has saved me a lot of time and energy and I wouldn’t have gotten nearly as far without it.

For those complaining this protocol won’t work for them, or is overwhelming, it’s been very easy to import this post into my AI app along with my current meds and a history of what I’ve reacted to, to give me a simplified summary, as well as a more personalised and narrowed-down list of recommendations. For me it’s given two supplements to try first, and how to do so cautiously. And as Cort has said, nobody is saying this is a prescriptive protocol, but it’s a great starting point to explore options further and discuss them with my doctor with a rationale in mind, which definitely beats shot-in-the-dark trial and error.

I’m sorry to read some of the outright rude comments you’re receiving here – you’d think people would at least extend a little politeness given your blog is able to be accessed for free.
Reply
- Brian Buckbee on November 18, 2025 at 1:08 pm
  
  I agree about the rude comments. Thanks for making the point.
  Reply
  - Cort Johnson on November 18, 2025 at 2:17 pm
    
    One would hope…a little trust would go a long way as well…These protocols were not out of left field after all. They all pretty much fit in with the way this field is going, but that, for some reason, didn’t matter…
    Reply
  - Waiting on November 18, 2025 at 3:09 pm
    
    So do I – there’s no need for this rudeness.
    
    Would those who are being rude please make your concerns known without the rudeness?
    
    It’s frankly disheartening and stressful to read them.
    Reply
- Brayden Finch on November 22, 2025 at 5:01 am
  
  I should add too – the two supplements AI has suggested I look at first are Coq10 and D-Ribrose, both of which are used widely already and were already on my (very long) list of things to try. All this has done is bumped them up the list to read up on more, and then ask my doctor about. If ChatGPT was spitting out a bunch of untested heavy medications it would be a very different story.
  Reply
  - Cort Johnson on November 22, 2025 at 11:58 am
    
    “If ChatGPT was spitting out a bunch of untested heavy medications it would be a very different story.”
    
    My thoughts as well 🙂
    Reply
Yamila on November 18, 2025 at 2:44 am

And again roping unproven or more like disproven brain retraining bull into this chat GPT treatment course. This is terrain on which I will simply not follow. Super unprofessional
Reply
- Cort Johnson on November 18, 2025 at 9:10 am
  
  Actually, there is plenty of evidence that mind/body practices in general can be helpful in many diseases – and I put neuroplasticity practices in that category.
  
  Besides the hundreds of recovery stories out there may not emanate from a study but they do provide more data points. IMO its actually scientific to gather all the data possible.
  Reply
- Waiting on November 18, 2025 at 10:15 am
  
  Professionals in many, if not all, fields now use AI in their work – it just has to be done responsibly – as it has been here with the source clearly labelled.
  
  Of course there can be pitfalls in AI but it looks like it’s here to stay. Technology continues to move and evolve, and so must we – or get left behind.
  Reply
Brighid Ó Dochartaigh on November 18, 2025 at 3:10 am

Hi, there’s a quite important missing word 😉 could you check it? in your heading near the top , ‘if you feel worse in the afternoon’, in the first bullet point – the last item in the list is missing, the one to take at 250 mg in the morning. Many Thanks!
Reply
- Cort Johnson on November 18, 2025 at 3:26 pm
  
  It was vitamin C!
  Reply
  - Brighid on November 19, 2025 at 1:42 am
    
    Awesome thanks! I thought either vit C or tyrosine 😆 And I meant to say big thanks for collating all this fascinating info! I’m not confident about chstgpt but see your reasoning about its use – I wouldn’t take any of it as gospel for now but it’s excellent hypothesising for things to be researched…
    Reply
    - Cort Johnson on November 19, 2025 at 11:16 am
      
      I like your approach! Which pretty much fits mine. So long as something isn’t dangerous I’m willing to give it a try.
      Reply
      - Eileen on March 31, 2026 at 3:32 pm
        
        Cort,
        Thank you for your hard work and efforts to share your personal research. BTW, may I ask how the protocol is working for you and specifically which suggestions helped and which didn’t help?
        Reply
        
        Cort Johnson on April 1, 2026 at 5:07 pm
        
        Boy, I haven’t really followed the protocol. To be honest, I’m a lousy patient. I do believe that Guanfacine which I’m out of did help a bit and I will sget back on that. It is quite clear to me on the other hand that Mestinon – which I don’t remember is in the protocol or not – is helping. Interestingly, it helped enough that it caused be to crash in a peculiar way. This has happened with things that helped in the past (and its not due to more exertion). I’m building up my dose. It’s really rare that I find things that help.
        Reply
        
        Eileen on April 1, 2026 at 11:30 pm
        
        Thank you for sharing! Yes, Mestinon is in the protocol. Glad to hear it helped and you will continue to titrate your dosing… 2 steps forward, 1 step back, 1 day at a time. Thank you for all you do and for giving so much to our community! Prayers and well wishes for your continued progress!!
        Reply
Herbert Renz-Polster on November 18, 2025 at 3:16 am

This is interesting, and, of course, speculative as all these interventions are built on a certain assumption, namely that the bottleneck of the inadequate NE regulation of the “ME/CFS brain” is related to inadequate vesicle filling in the locus coeruleus due to reduced ATP levels.
While this appears plausible it needs to be kept in mind that there may be other explanations for the messed-up NE regulation, some of which would tie well into other phenomena known about the “ME/CFS brain”. For instance, the “high noise” and inadequate “as needed” handling of NE may be related to inadequate signaling to the locus coeruleus, i.e. to the connectivity issues described in ME/CFS (see, for instance, Barnden´s work which shows that the ME/CFS brain does not adequately get activated when activation is needed, i.e., in response to work load).
Would also be interesting to hear the thoughts of the Aregawi/Goldstein duo about building a whole clinical treatment philosophy on their basic research results, wouldn´t it?
Also: did you have a chance to talk to one of the authors about their research and what they are planning next?
Reply
- Cort Johnson on November 18, 2025 at 9:11 am
  
  I agree. Very early days. I have not tried to get in touch with them but will try. I would love to know how this study got started and what they plan to do next.
  Reply
Nadine on November 18, 2025 at 3:20 am

Hello Cort, I honestly wish you would have tried the protocol before you wrote about it.
Chatgpt is unreliable, in the free version it will make up false quotations (puts up links that claim to be a paper, but it is another one totally unrelated), it will also disguise forum articles as papers, it scouts any social media input and if enough people have written about it, it will claim it’s medical.
Social media is full of false claims and plenty of retraining websites exist and make big assumptions, but I know nobody who got healthy with that.

Chatgpt is a software that will talk to you, and give you stuff to assume you have a human counterpart, but letting it set up a full treatment plan for plenty of drugs that are so far rarely used in ME/CFS is dangerous.

As a fact i’ve stopped telling people that a drug helps, because my sicknes is very much a wave form, and sometimes I’m just better and it almost impossible to tell if a drug is really helpful, or just wasting my money.

Many people online, once they feel a little bit better, however do proclaim in forums that this, or that helped them be healthy and then they quit the forums.

A few months later they are back, but they never post about it.

So internet information is biased and excludes the follow up posts, because people are ashamed to admit they were wrong, or are so much worse they anyhow can’t write.

This will also affect the payed version.

I’ve taken immunoglobuline infusions for 1,5 years now and I can say they help against sfn if continious and against infections.

They don’t stop me/cfs though.

I’ve recently started blood thinner injections because of micro cloths and oddly I immediately felt better (I was pretty bad), but I won’t tell people that until I’ve had half or one year of it, because most of us are variable and I’m a prime example of a wave form.

While time runs out on all of us, getting false hopes up is one of the worst things for the mind, so please next time, at least try that stuff first and evaluate with enough time.

Additional I was onto a possible connection of GAD 65 AAK and Histamin & Glutamate intolerance but I so far could not find enough studies that actually make a connection possible (while to me, from the papers I’ve read it seems to be possibly connected), if you have the pro version, maybe you could look that up and see what it says, or if there are any papers that actually connect this. I only found separate papers for rats, but can’t make the missing connection point.
Reply
- Cort Johnson on November 18, 2025 at 9:15 am
  
  Once again, when I use ChatGPT (I’m using the pro version – I don’t know if that makes any difference) the answers it gives me are almost wholly come from research papers. Everytime I’ve checked citations they are there =- no hallucinations. I don’t remember seeing any social media content.
  
  ” but letting it set up a full treatment plan for plenty of drugs that are so far rarely used in ME/CFS is dangerous.”
  
  Really, not to worry! You would have to get your doctors permission to use these drugs anyway – which is one reason this blog was described as a brainstorming session that you could then, take, if you wanted to, to your doctor, to see what she/he thinks about it.
  
  You do realize you just trashed ChatGPT and then asked me to submit a query to it! 🙂
  
  I’ll see what it says 🙂
  Reply
  - Hugh on May 2, 2026 at 12:21 pm
    
    I don’t know if you realise this Cort, but there is nothing stopping chatgpt from “citing” a paper, giving the impression it has somehow pulled information from that paper, when in fact it may have taken nothing from the paper. It may have taken only aspects of grammar, vocabulary, maybe it did what so many humans do, and “read” a paper, only to believe that the findings were the exact opposite of what the paper actually says.
    Reply
    - Cort Johnson on May 2, 2026 at 3:02 pm
      
      That’s why I regularly check the citations.
      Reply
- Cort Johnson on November 18, 2025 at 9:29 am
  
  Is there a connection between GAD 65 AAK and Histamine & Glutamate intolerance
  
  Here’s ChatGPT’s Answer
  
  Short version:
  There’s no solid, proven, direct medical link between GAD-65 autoantibodies and “histamine intolerance” or “glutamate intolerance.”
  There are some indirect, mechanistic overlaps that can make the whole picture feel connected, but the evidence is mostly theoretical or anecdotal.
  
  Is there a connection between GAD 65 AAK and Histamine & Glutamate intoleranceI’ll break it down in plain language.
  
  1. What GAD-65 autoantibodies actually do
  
  GAD-65 (glutamic acid decarboxylase 65) is an enzyme that converts glutamate → GABA (the main calming neurotransmitter in the brain). It’s found mainly in nerve endings and is responsible for a big chunk of GABA production; when GAD levels drop, glutamate rises and GABA falls.
  PMC
  +1
  
  GAD-65 autoantibodies (GAD65 AAb) are immune-system antibodies that target this enzyme. They’re best known as markers for:
  
  Type 1 diabetes and “latent autoimmune diabetes in adults (LADA)”
  Diabetes Journals
  +1
  
  Several neurological autoimmune syndromes: stiff-person syndrome, cerebellar ataxia, limbic encephalitis, autoimmune epilepsy, etc.
  Frontiers
  +2
  JNNP
  +2
  
  So medically, GAD-65 AAb mainly signal autoimmunity involving pancreatic islet cells and/or the nervous system, not histamine handling.
  
  2. What “histamine intolerance” is (and isn’t)
  
  What’s usually called histamine intolerance is thought to be:
  
  A state where your body’s ability to break down histamine (mostly via the enzyme DAO in the gut, and HNMT in tissues) isn’t enough for the histamine load, leading to symptoms like flushing, headaches, hives, GI upset, etc.
  PMC
  +2
  PMC
  +2
  
  Key points:
  
  It’s not an IgE allergy and not an autoimmune disease by definition.
  
  It’s often linked to low diamine oxidase (DAO) activity and/or gut issues, not to GAD-65.
  PMC
  +1
  
  So mechanistically, histamine intolerance = histamine metabolism + mast-cell activity + gut, whereas GAD-65 AAb = GABA/glutamate balance + autoimmunity.
  
  3. “Glutamate intolerance” and how GAD-65 fits in
  
  “Glutamate intolerance” isn’t a formal diagnosis in mainstream medicine; it’s more of a descriptive term people use when high-glutamate foods (MSG, aged cheese, bone broth, etc.) seem to trigger:
  
  anxiety, agitation
  
  headaches/migraines
  
  insomnia
  
  neurological “overwhelm” or sensory overload
  
  We do know some biology here:
  
  GAD-65 converts excitatory glutamate → inhibitory GABA, helping keep the glutamate/GABA balance in check.
  PMC
  +2
  Wiley Online Library
  +2
  
  When GAD function is reduced (genetically, by disease, or possibly by autoantibodies), you can see less GABA, relatively more glutamate, and more neuronal excitability.
  PMC
  +2
  MDPI
  +2
  
  So in theory:
  
  If GAD-65 autoantibodies significantly impair GAD-65 in the brain, that could push you toward a hyper-glutamate / low-GABA state, which might make you feel more sensitive to glutamate or MSG (more anxious, wired, seizure-prone, etc.).
  
  However:
  
  That mechanism is plausible but not well proven in humans for everyday “glutamate intolerance.”
  
  Most GAD-65 AAb research focuses on frank neurological diseases, not food reactions.
  Frontiers
  +1
  
  So: there could be an indirect link between GAD-65 autoimmunity and feeling “overdriven” by glutamate, but this is not firmly established as “glutamate intolerance” in clinical guidelines.
  
  4. Histamine, GABA, glutamate, and the brain
  
  Histamine is also a brain neurotransmitter. It interacts with the same networks that involve GABA and glutamate:
  
  Histamine neurons in the brainstem project widely and influence arousal, temperature, attention, etc.
  Journal of Neuroscience
  +1
  
  Some GABAergic neurons express GAD65 and can be modulated by histamine signaling; studies show complex interactions between histamine receptors and GABA release.
  PMC
  +1
  
  So at the cellular level:
  
  Histamine ↔ GABA/glutamate systems are interconnected.
  
  Changes in one system can influence the others (e.g., histamine can modulate GABA release and neuronal excitability).
  
  But this is basic neuroscience, not evidence that:
  
  “Having GAD-65 antibodies causes histamine intolerance,”
  
  or
  
  “Histamine intolerance is caused by GAD-65 AAb.”
  
  Medical literature just doesn’t say that.
  
  5. Autoimmunity + mast cells: where people get the idea of a link
  
  There is some overlap between:
  
  Autoimmune diseases (like T1D, thyroid disease, etc.), where GAD-65 antibodies are more common, and
  
  Mast-cell–related conditions, which can involve histamine symptoms.
  
  Reviews of mast cell disorders and autoimmunity note that autoantibodies in general are more common in people with mast-cell–driven diseases, and that some patients collect multiple autoimmune markers.
  Ovid
  +1
  
  So you can have one person with:
  
  GAD-65 antibodies (autoimmunity) and
  
  Histamine-like symptoms / mast cell activation
  
  …but that’s correlation in individuals, not proof that one causes the other. It more likely reflects a general immune dysregulation tendency.
  
  6. So, is there a connection?
  
  Putting it all together:
  
  What we do know
  
  GAD-65 → converts glutamate to GABA; anti-GAD-65 antibodies are markers of autoimmunity (T1D, neurological syndromes).
  Mayo Clinic Laboratories
  +2
  Frontiers
  +2
  
  Histamine intolerance involves histamine metabolism (DAO, HNMT) and mast cells, mainly in the gut and vasculature.
  PMC
  +2
  PMC
  +2
  
  GABA / glutamate / histamine systems interact in the brain, and GAD-65 is part of that network.
  Frontiers
  +2
  PMC
  +2
  
  Autoimmunity and mast-cell–related issues sometimes co-occur in the same people.
  Ovid
  +1
  
  What we don’t have good evidence for
  
  No strong, peer-reviewed studies showing that GAD-65 autoantibodies directly cause histamine intolerance.
  
  No consensus that “glutamate intolerance” is a distinct, GAD-65–mediated medical condition (even though high glutamate can clearly affect brain excitability).
  
  So the most honest answer is:
  
  There’s no established, direct causal link between GAD-65 autoantibodies and histamine or glutamate intolerance in mainstream medicine.
  Any connection is likely indirect and individual, via general immune dysregulation and altered GABA/glutamate balance rather than a one-to-one cause.
  Reply
  - Nadine on November 18, 2025 at 6:35 pm
    
    Thank you very much!
    
    This is as much as I already researched, I was hoping professional could access more papers than I can, as you know loads of papers are behind paywalls and I’m sure AIs have special deals to access them partly, at least one would hope so since you pay for them.
    
    I still stand by the point that ai yet are not as smart as we hope and plenty of faulty things can come up and will come up and I think before for going into details with a treatment, you should have tried it, or talked to somebody who actually tried it (or parts of it) and has some results they can talk about, because all theory is nice, but you have to have at least one proof of concept.
    
    Imagine id take the causal connection of gad 65 (Which I do strongly believe might be influencing histamine and glutamate intolerance, according to what I read from animal studies and reactions in the brain of rats, but have no proof of the missing study that would link it all together more closely).
    
    And then write about it and suggest the treatment plan as well.
    
    I’m all for theories, but going into detail with treatments without having first, or second hand knowledge of using it, I’d not do it.
    
    Again thank you very much for checking the ai for me too!
    Reply
    - Cort Johnson on November 19, 2025 at 11:28 am
      
      If you think about it most of these treatments have already been tried in ME/CFS and are part of many patients protocols. There’s nothing new – other than the timings and combinations – about the supplements, Mestinon, mast cells, stimulants or neuroplasticity. Wellbutrin, atomextine and a few others may be new – but then you have your doctor to query about them.
      Reply
- Waiting on November 18, 2025 at 11:12 am
  
  I actually know 2 ME patients who recovered with brain retraining programs – 1 a decades long patient and another ill about 4 years.
  Reply
Patrice on November 18, 2025 at 8:24 am

Thank you Cort! That must have been exhausting to compile all that information. I am so thankful for your work.

I already take all those supplements and a few more. Most of them I started after reading one of your articles. I have continued to improve and credit much to my supplement regimen. Although, I also have other practices I do that have also helped. Again, many I picked up from your articles.

I’m excited to try the dosage and daily time schedule you recommended. I tend to stay away from prescription meds.

For pacing help I started using the Visible app two years ago. I heard about it from the Workwell foundation. I don’t recall if you’ve mentioned that before. It has really improved my skills and results from pacing.

Thanks again Cort for your work. I have significantly improved the last few years ( I got ill in 2000) from gleaning things to try from your work.
Reply
- Cort Johnson on November 18, 2025 at 1:55 pm
  
  Thanks! These are particularly exhausting blogs to write.
  
  I did write about Visible app but haven’t heard anything about it in years. I’m glad it’s helped.
  
  https://www.healthrising.org/blog/2022/11/19/visible-app-long-covid-chronic-fatigue-syndrome/
  
  I’m currently trying – as soon as I can get it going – the Lumia earpiece – which measures blood flows
  
  https://www.healthrising.org/blog/2023/06/22/stat-earpiece-chronic-fatigue-long-covid-pots/
  
  Glad to hear you’re getting better 🙂
  Reply
Sieglinde on November 18, 2025 at 9:23 am

Hi Cort, I’ve personally observed several clinically relevant connections that are often overlooked by primary care physicians, especially when timely referrals to hematology or endocrinology are missed.

How often are patients actually tested for cortisol? Despite its key role in energy and stress regulation, cortisol is rarely assessed in routine care unless a clear endocrine issue is suspected, leaving many cases undiagnosed.

In my case, I experience significantly low energy in the mornings, which I’ve traced to consistently low cortisol levels, typically ranging between 3.1 and 4.5 µg/dL. These suboptimal levels make it extremely difficult to function early in the day. However, after taking hydrocortisone in the early morning, I notice a gradual improvement in energy, particularly by midday.

Additionally, I have elevated erythrocyte levels, which contribute to increased blood viscosity and impaired circulation. The commonly suggested remedy of increasing water intake has minimal impact in my experience, likely because systemic hydration is not adequately addressed through simple fluid consumption alone.

Another frequently overlooked factor is elevated iron and ferritin levels. Persistent fatigue in the presence of high iron markers can point to underlying issues such as iron overload, chronic inflammation, hepatic stress, or oxidative damage, all of which may be misattributed if metabolic causes are not thoroughly investigated.

While tools like ChatGPT can help uncover less commonly considered insights, as you’ve mentioned, they should complement, not replace, the critical review of peer-reviewed research and clinically validated resources.
Reply
- Roonie on November 18, 2025 at 12:02 pm
  
  Instead of compiling Chatgbt why not wait for the real science from real scientist.
  Heck, ive only been waiting decades so for a couple more years I think I can wait.
  Im very close to recovered on my own.
  Reply
  - Cort Johnson on November 18, 2025 at 12:23 pm
    
    As I’ve stated the citations that I get from ChatGPT check out i.e. they come from real scientists and the conclusions I see are plausible.
    
    Apparently you’re selective when you embrace AI and when you don’t.
    
    “Oliver….I used AI and asked if tetracycline causes fatty liver disease. The answer is “yes”
    My liver tests are fine also but the internal medicine doctor keeps saying its my liver.my ND also states the same.
    
    As stated by the researcher Robert navauix, tetracycline is one of the bad ones listed”
    
    How about producing something positive and telling us how you’re very close to recovering on your own?
    Reply
    - Roonie on November 20, 2025 at 11:11 am
      
      Cort, that was before I watched dr berg’ video on how inaccurate
      AI really is.
      
      I told you how I got better, by taking ivermectin
      Reply
  - Sieglinde on November 18, 2025 at 1:32 pm
    
    Roonie: This is what I intended when I referred to ‘a critical review of peer-reviewed research and clinically validated sources.’
    Reply
Mala on November 18, 2025 at 1:31 pm

Chat GPT and brain retraining. This truly is a new low. Don’t think I’ll ever donate again
Reply
- Cort Johnson on November 18, 2025 at 1:36 pm
  
  I guess my arguments about ChatGPT being helpful fell flat! Thanks for your past support, though.
  Reply
Bebe on November 18, 2025 at 1:43 pm

One of the things that Chat can overlook is how medications are dosed. You cannot split a 1 mg ER Guanfacine tablet to create a .5 mg dosage. My ME specialist was very, very clear that if we want to try Guanfacine, we have to take the entire 1 mg because the medication is not dispersed evenly within the extended release tablet. He also told us that short acting Guanfacine is not a suitable replacement. So it’s 1 mg to start or we have to trial something else.

I worry that this protocol repeatedly mentioning .5 mg ER Guanfacine will lead to people splitting the tablets and having a negative experience with Guanfacine. Guanfacine, taken as directed, has been a wonderful med for my brain fog.
Reply
- Cort Johnson on November 18, 2025 at 2:00 pm
  
  Thanks for mentioning that. ChatGPT did emphasize extended release but it did not say anything about splitting the tablet. This is where doctors come in – which is what the blog was all about – a ChatGPT brainstorming session – which is then informed by doctors (or by your own research). Guanfacine is a prescription medicine so you cannot get it without going through your doctor.
  
  Glad to hear it’s been helpful. I was surprised that it showed up so many times in the treatment suggestions.
  Reply
  - Bebe on November 21, 2025 at 6:48 pm
    
    Thanks for adding the dosing clarification to your post. I frequently come across people who split pills in an effort to take the lowest possible dose, not realizing that not all pills can be split. We’re generally sensitive to meds, so the intention makes perfect sense but the execution can be harmful without proper guidance. And, as you know, not all of us have great medical guidance. A lot of DIY’ing, necessarily, happening with ME.
    
    I’ve been really enjoying this recent series of articles that are building upon each other and helping form a picture of what might be happening in the brain. I reread both the article before and after this one several times. I also recently started a supplement from one of your older posts that I haven’t seen discussed anywhere else and it’s been making a noticeable difference. Funnily enough, most of the comments on that post were fairly negative, saying things like, “This can’t work.” But it’s working for me. Thanks for the effort you put into dissecting these studies and highlighting new possibilities.
    Reply
Brian on November 18, 2025 at 1:51 pm

While this very long blog is impressive in terms of effort and detail provided, it does trouble me that too often Cort crosses the line between medical journalist into providing medical advice (without medical qualifications).
Reply
Scotty81 on November 18, 2025 at 6:29 pm

Cort, thanks for your perseverance in continuing to post on this thread despite some negative feedback. Perhaps it would have helped if you had put a big disclaimer before the AI results saying that 1) you are doing this for the academic exercise to see what ChatGPT has to say and 2) the results are not a substitute for medical advice.

Regardless of whether you used AI or not, I would have hoped that readers would have appreciated that what you post should not be construed as medical advice anyway. So I am not surprised at healthy skepticism and even critical feedback, but I am surprised at the negativity. I agree with other posters that one can provide contrarian views without being rude. Additionally, even if my daughter fit your model perfectly, I certainly wouldn’t have just given her the supplements or asked for the prescriptions without one of her medical specialists weighing in.

All that said, my daughter is already all doing the non pharmacological treatments e.g. water, salt, compression sleeves, recumbent bike, etc. It’s not that she feels better when she does those things, but rather, she feels worse when she doesn’t do those things.

In addition to her other diagnoses, she also has Sjogren’s (a known cause of POTS), autonomic neuropathy, GI issues, and venous insufficiency. However, she could not tolerate Plaquenil (for Sjogren’s) and IVIG, which is being studied for a variety of systemic illnesses, didn’t work, despite a lengthy trial. Also, one of the main ways to regenerate neurons is through exercise. Yet, despite this, she is not seeing any improvement in her vascular tone or neuropathy symptoms.

I do agree with you that something must still be wrong in the periphery since she is still very symptomatic. Northera makes many of her symptoms a bit better, likely due to her BP being more in the normal range. However, “better” for her means she still has many of the classical symptoms that many on this site have. It doesn’t make any of her symptoms worse.

Perhaps there will be some research advances in either the MECFS, Sjogren’s or other fields that will advance or cement our hypotheses about these illnesses which will lead to targeted therapies without intolerable side effects.
Reply
Josh on November 18, 2025 at 7:38 pm

Cort,

It’s an interesting coincidence you post this as I’m about to start Droxidopa. Last year, I took an SNRI for mood and briefly used Straterra as well. Noticed I was having less issues with dysautonomia and while I still had brain fog I was feeling a little more physically energetic. Seeing that meds affecting NE have helped, I brought Droxidopa up with my CFS specialist and they agreed to let me try it out. I’m excited to see how it goes as your post seems to bring some theoretical backing to my personal experience with NE support.

I may give Qelbree a try in the future since I’m stillhaving cognitive issues and stimulants seem to be a bit too much for me. Curious to see what an increase of NE in the prefrontal cortex may do for my symptoms.

I’m sorry the post has been a bit polarizing. But, I’m glad you put it out even if not everyone may be satisfied with it.

Josh
Reply
- Cort Johnson on November 18, 2025 at 8:16 pm
  
  Thanks.
  Reply
- Cort Johnson on November 19, 2025 at 11:24 am
  
  It’s been a rough go for sure. Honestly, I never dreamed this post would spark so many comments. I used ChatGPT in my daily life for all sorts of stuff and of course with trying to understand these very complex topics. My experience has been good and I was pretty much out of the loop with regard to worries about it.
  Reply
Matthias on November 18, 2025 at 11:28 pm

I appreciate your efforts, Cort. And I am someone who is a bit skeptical of AI.
But what you set out isn’t that far removed from what a human, Dr Jay Goldstein, did in treating ME/CFS patients in the 1990s.
As I have said before, he was way ahead of his time in terms of the role of the brain in ME/CFS. And his range of pharmacological approaches, dependent on patient profile, is quite similar to what you set out here.
Reply
- Cort Johnson on November 19, 2025 at 11:20 am
  
  “And his range of pharmacological approaches, dependent on patient profile, is quite similar to what you set out here.”
  
  Thank you!
  Reply
Matthias on November 18, 2025 at 11:32 pm

Cort, some of the above isn’t so far removed from what Cortene was looking to do.
Cortene sounded promising. Has it fallen over?
Reply
- Roonie on November 19, 2025 at 4:50 am
  
  https://youtu.be/iV6osRQSbXE?si=-nqNw3wczExEXU28
  
  Dr Berg proves AI is bunk
  
  Just my 2 cents
  Reply
- Cort Johnson on November 19, 2025 at 11:19 am
  
  Cortene was/is promising but when last I heard – about a year ago – they had gathered some funding they didn’t have enough. Two problems exist: while the hypothesis makes sense, it’s just too new for pharmaceutical companies which are making bucks off tweaking their formulations and have proven drug categories to dig into. The other is ME/CFS – it’s pretty much an automatic turnoff. Starting with ME/CFS rather some other disease (the drug might be able to impact many diseases) may have been a mistake.
  Reply
Leo on November 19, 2025 at 3:54 am

To everyone out there with ME look it’s not your responsibility to feel like you must do complex protocols like this or you don’t deserve to get better, even if such a protocol gives you 10% improvement. Treatment protocols like this exist because the scientific establishment has failed and no one has any clue so they are just shooting into the dark with supplement and generic med stacks hoping something will stick for some people. I know ME experts are trying to help but no one has a clue if it will help. It’s not your fault if you don’t want to jump through a ton of hoops every single day and spend all this money, in other chronic illnesses they aren’t asked to do this
Reply
P Hassack on November 19, 2025 at 7:44 am

What do you mean by fatigue dominant? Do you mean brain fatigue? Do you mean muscle fatigue? Fatigue unqualified doesn’t mean very much.
Reply
- Cort Johnson on November 19, 2025 at 11:14 am
  
  People for whom fatigue in any feature is their main symptom
  Reply
Betty Mekdeci on November 19, 2025 at 11:24 am

I agree with Skip and Roonie. ME/CFS can’t be successfully treated until the cause is identified.

Before we knew that ulcers were caused by H-Pylori which could be treated with an antibiotic, patients were told they were stressed and should eat a bland diet.

Before the cause of AIDS was discovered, all kinds of bizarre things were tried, some quite toxic. With new protocols AIDS patients are now living nearly normal life spans.

Not too long ago, mothers were blamed if their children were autistic. Now there are 29 peer-reviewed studies linking autism and other neurodevelopmental disorders to the maternal use of acetaminophen.

In September, the FDA sent out a “Dear Doctor” letter with this information.

Using an untested technology to suggest treatments for “symptoms” is not helpful and potentially dangerous.
Reply
- Roonie on November 19, 2025 at 12:17 pm
  
  Betty ,let’s not forget aluminum was/is used as an adjuvent in most of those vaccines.
  There is a dr combria(spelling?) that has successfully treated autism using a protocol to remove aluminum from autistic children brains
  Reply
- Matthias on November 20, 2025 at 12:50 am
  
  I am pretty confident that neuroinflammation is key to this illness. And that leads to the issues with dopamine etc. which flows on to key symptoms.
  For me, the key unresolved question is what leads to the chronic neuroinflammation?
  Is it reactivated viruses? This is possible. But I am not totally convinced. I’m interested to see how Polybio’s trials of antivirals goes. If it is successful, this might suggest reactivated viruses play a central role.
  
  The other possibility, which I think more likely, is that some sort of neuroinflammatory switch gets turned on by an initial insult, and never gets turned off. And that the neuroinflammation, while triggered by a virus, is not perpetuated by them. I’m interested to see how the bezisterim trial goes.
  The things that have worked best for me – nicotine, curcumin, fluoxetine – all work on neurotransmitters and inflammation.
  Reply
E. Hale on November 19, 2025 at 3:06 pm

I would like to let you know that I am unsubscribing explicitly because of the use of ChatGPT here.

Generative AI is not logic-based; it is statistics-based. It is pulling what it thinks is the most likely thing to be relevant; it applies no logic beyond ‘does this conform to the statistics’. And it is well-known to ‘hallucinate’- just because you have never noticed it doesn’t mean it doesn’t, or that it hasn’t to you and you’ve missed it.

If I discovered a doctor was using ChatGPT to make a treatment plan for me, I would immediately request to switch doctors, because to me, the time you have to spend fact- and logic-checking everything ChatGPT gives you outweighs its ability to bring up all the data quickly- even ignoring the fact that it may not be bringing up ALL the data, or all the CORRECT data.

In any case, ChatGPT is fed on what data there is- meaning that all the institutional biases are still represented in its datapool. That when it isn’t being fed on copyrighted material without the copyright holders’ consent, which is most of the time.

This is of course ignoring the massive, massive environmental impact AI systems like this have, requring as they do vast amounts of water to cool their server farms, water which is then contaminated enough that it can’t be used for anything else.

I’m sorry, but I have lost trust in your reporting.
Reply
- Cort Johnson on November 20, 2025 at 1:00 pm
  
  I would once again ask you to ask yourself whether the results AI produced look like hallucinations or more like something that is in line with the ME/CFS/Long COVID field – and then judge the results based on that.
  
  In any case, ChatGPT is fed on what data there is- meaning that all the institutional biases are still represented in its datapool.
  
  C’mon – what isn’t? Do you think that peer-reviewed analyses are any different? AI, researchers, doctors, bloggers – we’re ALL USING THE SAME DATA…
  
  If you can’t handle checking out ChatGPT produced treatment options which were clearly described as the result of BRAINSTORMING session that you should go over with your doctor – and which feature supplements and low doses of drugs that are for the most part commonly used in this community
  
  If this is too much for you, or going to your doctor to ask if this or that makes sense is too much for you – then I’m sorry, maybe you shouldn’t be following this blog anyway.
  
  Good-bye!
  Reply
- Waiting on November 20, 2025 at 2:55 pm
  
  Cort has been writing on & for & on behalf of this community for 21 years – and you’re going to unsubscribe based on this 1 article which used AI?? And which he clearly cited and described as a brainstorming session? Really!
  
  He’s received an award for his work and has helped countless patients – and you’re going to stop trusting him?
  
  He’s published thousands of articles – I think you’re going to miss a lot.
  Reply
- Brian on November 24, 2025 at 11:36 am
  
  Well said
  Reply
Alicia on November 20, 2025 at 9:36 am

I am wondering– does the combination of wellbutrin and mestinon work? I have been looking into this because I want to try mestinon for OI, but family really want me to take an antidepressant, too. Wellbutrin (along with other antidepressants) can block nicotinic acetylcholine receptors. And, mestinon needs acetylcholine to work. Would wellbutrin (or other antidepressants that do this) block mestinon from working properly? Could AI be trusted with this query? And, who knows if the antidepressants (including some SSRIs) block enough of these receptors to make a difference?
Reply
- Cort Johnson on November 20, 2025 at 12:51 pm
  
  You can decide. At the least it will provide some new information. Here was it’s take:
  In theory, a nicotinic blocker like Wellbutrin could blunt some acetylcholine-mediated effects.
  In practice, there’s no good clinical evidence that bupropion or similar antidepressants meaningfully “turn off” Mestinon, and they’re often used together. Any effect is likely modest and very individual.
  
  1. What each drug is doing
  
  Mestinon (pyridostigmine)
  
  Inhibits acetylcholinesterase → raises acetylcholine (ACh) at synapses.
  ScienceDirect
  +1
  
  Acts mainly in the periphery (neuromuscular junction and some autonomic synapses); it barely crosses the blood–brain barrier.
  Encyclopedia Britannica
  +1
  
  Wellbutrin (bupropion)
  
  Antidepressant: primarily blocks reuptake of norepinephrine and dopamine.
  NCBI
  
  Also a non-competitive antagonist at several neuronal nicotinic ACh receptors (α4β2, α3β2, α7, etc.) – this is why it helps with smoking cessation.
  ScienceDirect
  +2
  PMC
  +2
  
  There are also data showing bupropion can bind to muscle-type nicotinic receptors (the neuromuscular junction subtype), but potency and functional block at real-world doses look limited.
  PMC
  +1
  
  2. Does blocking nicotinic receptors cancel Mestinon?
  
  Conceptually:
  
  Mestinon → more ACh in the synapse.
  
  Nicotinic antagonist → fewer nicotinic receptors available / lower probability of opening.
  
  If you fully block nicotinic receptors, no amount of extra ACh from Mestinon will help at those receptors. But:
  
  Bupropion is not a complete, global nicotinic blocker. At therapeutic concentrations it partially inhibits certain neuronal nicotinic subtypes, especially in the brain.
  ScienceDirect
  +1
  
  Mestinon’s clinically important effects (MG, POTS) are mostly:
  
  At the neuromuscular junction (muscle-type receptors), and
  
  At peripheral autonomic synapses, including muscarinic ACh receptors (for heart rate, gut motility, etc.).
  ScienceDirect
  +1
  
  Bupropion doesn’t block muscarinic receptors.
  
  Real-world data:
  
  Bupropion is not flagged as a major “avoid” drug for myasthenia gravis, and a pharmacovigilance analysis found no increased reporting of MG worsening with bupropion (reporting odds ratio < 1). Wiley Online Library In MG education/interaction lists, the big concerns are aminoglycosides, fluoroquinolones, magnesium, certain beta-blockers, some antiepileptics, etc. – bupropion isn’t a prominent villain there. PMC +1 Mestinon interaction checkers list many interacting drugs, but bupropion is not highlighted as an interaction that reduces efficacy. Drugs.com +1 So while mechanistically bupropion might shave off some nicotinic signaling in a few places, there’s no strong evidence it meaningfully prevents pyridostigmine from working at the doses we use clinically. 3. What about other antidepressants that hit nicotinic receptors? Several antidepressants have mild nicotinic-modulating actions (this is an active research area in depression). American Chemical Society Publications But the bigger practical issues for someone on Mestinon are usually: Strongly anticholinergic drugs (especially older tricyclics) that block muscarinic receptors and can directly oppose ACh actions, worsen orthostasis, etc. PMC +1 Sedating drugs that lower blood pressure or worsen fatigue, which may counteract what you’re trying to get from POTS/MG management. Modern agents like SSRIs and bupropion are generally considered lower-risk in MG and are commonly used alongside pyridostigmine with careful monitoring. MGteam +1 4. How I’d translate this into “what to do” No, Wellbutrin is not known to outright block Mestinon from working. A subtle interaction is theoretically possible (especially at some neuronal nicotinic sites), but evidence that it causes clinically important loss of Mestinon effect is weak to absent. If you personally notice: “When I added Wellbutrin, Mestinon stopped helping my POTS / MG symptoms,” that’s very valid N=1 data and worth bringing to: Whoever manages your autonomic / MG / POTS care and Whoever prescribes your antidepressant, so they can weigh alternatives (e.g., an SSRI, SNRI, or different NDRI) and maybe adjust Mestinon timing/dose. Because your situation involves complicated autonomic and NE-vesicle stuff, I’d expect your team to think in terms of overall net autonomic balance: bupropion slightly pushes NE/DA up (more sympathetic tone), Mestinon leans parasympathetic/neuromuscular; the question isn’t just “do they block each other,” but “what does the combo do to you.” If you’d like, I can walk through timing strategies (e.g., how far apart to dose bupropion and Mestinon so you can empirically see what each is doing). Here are the resources which I looked up which are used in this query (I didn't look up all of them_ https://www.sciencedirect.com/science/article/abs/pii/S0022356524389050?utm_source=chatgpt.com / https://pmc.ncbi.nlm.nih.gov/articles/PMC2067251/?utm_source=chatgpt.com / https://pmc.ncbi.nlm.nih.gov/articles/PMC2756054/?utm_source=chatgpt.com / https://www.sciencedirect.com/science/article/abs/pii/S0022356524389050?utm_source=chatgpt.com / https://www.sciencedirect.com/topics/neuroscience/pyridostigmine?utm_source=chatgpt.com / https://onlinelibrary.wiley.com/doi/pdf/10.1111/ene.14773?utm_source=chatgpt.com / https://pmc.ncbi.nlm.nih.gov/articles/PMC8038781/?utm_source=chatgpt.com / https://www.drugs.com/drug-interactions/pyridostigmine%2Cmestinon.html?utm_source=chatgpt.com / tps://pubs.acs.org/doi/10.1021/jm401937a?utm_source=chatgpt.com /
  Reply
Carolyn McCubbing on November 20, 2025 at 1:56 pm

Daylight at 5am? Not anywhere in northern Europe or Canada for half of the year
Reply
Susanne on November 21, 2025 at 3:07 am

Thank you Cort for the fantastic work you do, the pieces of the puzzle are falling into place. I am so grateful, I am very well-read but would not have been able to draw the conclusions myself. It is such an overwhelming amount of information as you know.

My son started taking amitriptyline and it made his stomach hurt more so he stopped. A few years later he was put on Cymbalta, also a drug that blocks both serotonine and norepinephrine. I did ask his doctor if it really was the right drug since I had read that if you already hace a heightened sympathetic response, a NE inhibitor can affect an alreday sensitive SNS/PNS.

A few weeks ago, he started feeling worse after taking Cymbalta so he stopped (on his own initiative) he then felt “sharper”.

After reading part I in your NE series I asked chatgpt about my son’s symtoms before and after taking Cymbalta as well as Amitriptyline. Chatgpt then explained as you have done why he might have felt better in the beginning (even though I still had my hesitations about the drug since he had less apetite and was more tired; he would crash in the afternoon if he got up early).

Chatgpt answered that he might have had more NE in the vesicles in the beginning but after a while when running low/empty it would be like a car running on an empty tank; there was no ATP to produce NE. So there would be more “noice” – no wonder!

I’m so grateful to you adding those pieces to the puzzles. Also, it now also makes more sense to me why neuroplasticity programs like GUPTA and DNRS might work or not. I can understand how you feel getting all the negative and rude comments. I was so excited reading your lengthy and detailed protocol and then reading all the comments made me a bit disappointed and sad how people react.

They clearly don’t understand the mechanism’s you are trying to explain. They don’t have to agree with you using AI but I’m with you. How are we to sort in all the research out there otherwise? Chatgpt is a valuable tool that can gather and sort all the information for us.

If most of the people are waiting for the CAUSE to ME/CFS then why are they even reading all the research articles that you are compiling since they end up disappointed and let down it not being the answer they are waiting for. Hurray for you! I will contribute to your work and most certainly make a donation.
Reply
- Cort Johnson on November 21, 2025 at 12:43 pm
  
  Thanks so much, Suzanne, after all the hullaballo, it was very nice to kind of end this discussion on a positive note! Thank you very much. Good luck with your son!
  Reply
  - Claude on November 21, 2025 at 4:43 pm
    
    Félicitation Cort. I.A. est un must et ce ne serais-ce que pour avoir des réponses que la majorité de nos Docs sont incapable de fournir! I.A. a un bassin de d’informations énorme mais nécessite des questions précises comme tu as mentionné. J’aime bien et recommande aussi d’avoir deux applications afin de comparer etc. C’est définitivement un cadeau qui nous est offert pour comprendre et conserver l’espoir d’obtenir certains traitements lorsque la cause de cette maladie complexe sera élucidée! Difficile d’avoir des traitements adéquats tant et aussi longtemps que nous ne connaissons pas la cause! Surtout ne lâche pas et ne te soucie pas trop des commentaires négatifs
    
    Google Translation
    
    Congratulations Cort. AI is a must, if only to get answers that most of our doctors are unable to provide! AI has a huge database of information but requires specific questions, as you mentioned. I like it and also recommend having two apps to compare results, etc. It’s definitely a gift that helps us understand and maintain hope for treatments once the cause of this complex disease is elucidated! It’s difficult to get adequate treatments as long as we don’t know the cause! Above all, don’t give up and don’t worry too much about negative comments.
    Reply
    - Cort Johnson on November 22, 2025 at 12:04 pm
      
      Thanks so much Claude, In the end the concerns about AI will only make this blog stronger 🙂 Thanks for the ideas two compare apps 🙂
      Reply
      - Claude Paquin on November 22, 2025 at 8:24 pm
        
        Juste ajouté que pour avoir fait l’essai d’une panoplie de médicaments alors dans mon cas les anti dépresseurs incorporant des inhibiteurs de la Noradrénaline sont néfastes pour moi et comprennent trop d’effets secondaires! Par conséquent j’ai de l’amélioration avec de l’amantadine 100 mg. 2 fois par jour que je prends depuis plus de 20 ans et j’ai été capable avec ce médicament d’enrayer les fameux effets Pseudo grippaux. Ce médicament est utilisé dans la maladie de parkinson et augmente le taux de dopamine….. si je ne me trompe! En tout cas je suis addic de ce médicament qui augmente mon énergie et me sécurise! Alors comme tu vois je suis complètement à l’inverse ou à l’opposé des recommandations faite dans le reportage ci-dessus! Si je reviens aux Anti-dépresseurs alors la Sertraline et le Luvox sont ceux que je tolère le mieux et comme disait mon Doc. à l’époque….il y a longtemps, ce sont des médicaments utilisés par les personnes âgées! Avec du recul je constate que effectivement notre métaboliste à l’usure de personnes âgées! je poursuis mes recherches depuis 40 ans avec médication (Essai/erreur) etc et présentement je crois avoir détecté une piste intéressante avec laquelle je te reviendrai lorsque je serai parvenu à faire l’essai d’une certaine médication! Ceci dit il faut garder l’espoir puisque c’est la VIE.
        Reply
Matthias on November 21, 2025 at 4:10 pm

I am a little confused.
The NIH study led by Nath found normal norepinephrine, but lower dopamine metabolites, eg. LDOPA
Goldstein was involved in that study.
But now he’s finding low norepinephrine, but normal dopamine.

Confused….
Reply
T Allen on November 22, 2025 at 7:02 am

Just came back to this page after reading the Nov 20th blog to see what all the chaos was about. There is nothing wrong with this blog the way Cort wrote it. There are disclaimers about appropriate topics, AI use and not medical advise. The complaints about complexity are valid but it was explained in a more simple way as well. It was long but fascinating.

I understand that people afraid of AI but there is very little difference from reading through any other source(s) on the Internet, FOR SCIENCE BASED INFO. Comments here are from people who have “heard stories” or used AI for NON SCIENCE purposes. THAT is a whole different ballgame. Totally different, were AI is giving you the answer it thinks you want not answering a specific fact based question. I personally have gotten wrong answers from AI on political or other non-science based questions. I’ve have a couple of “teaching” moments on AI, me teaching it. Science questions on AI you look at the source of the info just like you do when looking up any science topic online. If you don’t trust the source you ask AI for another source.

Should the average person without a medical background be looking at AI for science answers? Yes, IF they know how to do research and NO if they don’t.

Cort knows how to do research and I trust he is checking answers if something sounds “off” or like he did, you run the first AI answer through a second AI and see if they agree.

Many of us have no issues in him using AI. He ID’d it as AI and if you don’t want to read it or trust it that’s OK. But DO NOT distrust Cort because he’s using AI (it’s a lot faster) especially if you aren’t donating to Health Rising! He’s doing all of us a huge service. Keep up the great work Cort!
Reply
- Cort Johnson on November 22, 2025 at 11:57 am
  
  Thanks! 🙂
  Reply
Jim Kepner on November 22, 2025 at 8:10 am

Cort- thanks for braving the reactivity to AI. It seems to be carrying the load of built-up resentment for having to self-manage our illness without competent medical help. I would point out that every “treatment” for ME in the 50 years I’ve had it has been pure speculation! And even less based on logic and data than what you developed with your ChatGPT exploration. I haven’t yet gone through all the protocols you developed in detail.

Maybe part of the reactivity is, like mine, a bit of cognitive overload, but I think you demonstrate a process for inquiring using AI. If nothing else, it shows how to work a hypothesis about brain fog and SNS problems, along with some logical approaches to accessible, testable treatments. Other than noting the speculative nature of any AI chat process, I think it’s an important contribution to the science-based discussions you have here, all of which are speculations on intelligence, artificial or otherwise.
Reply
Matthias on November 22, 2025 at 1:36 pm

My daughter has had long covid or ME/CFS – we aren’t sure which – for more than 2 years.
A few months ago she tried creatine, and we saw significant improvements within a week. Rather than sleeping 14-15 hours a day, she came back to 11-12 hours.
I wonder if this links to the ideas around ATP etc?
I am very ‘evidence based’ – we tried the creatine off the back of a small UK study which reported benefits of creatine in ME/CFS.
The benefits of the creatine fell away a bit. I wonder if this is because she needs to do other things, as described in Cort’s article, such as take tyrosine?
Reply
Marcia Mersky on November 22, 2025 at 3:11 pm

Hi Cort – I’ve been using RTHM Intelligence AI platform and I’m very impressed with AI’s capability to take large amounts of data (many, many, many lab tests) and making sense of them. I have caught some errors, but I expected that. It has made some treatment recommendations that have surprised me but I’m reviewing them with my doctor and figuring out if they make sense IRL. I think AI has a lot to offer complex patients, so I feel enthusiastic and hopeful about how it may move the needle. I also think it’s going to help researchers group patients for better clinical trials. ANyway, I share your enthusiasm and optimism about what it can do for us.
Reply
Sharon on November 22, 2025 at 6:33 pm

Cort, many many thanks for all the work you do, including the research using AI tools. Very grateful for all the information.
I’m wondering if you can run a query for the same type of detailed protocols for “worse in the afternoon” like you have for “worse in the morning” ?
Reply
- Cort Johnson on November 27, 2025 at 4:41 pm
  
  Here you go! You may not want to take all these treatments (Survovexant is expensive, if I remember correctly).
  
  D-Ribose + Caffeine + Guanfacine + Mestinon + Suvorexant + Mast Cell Tx
  Daily Timing Map – Afternoon-Worse Pattern
  
  Big disclaimer: This is only a discussion template. It’s not medical advice or a prescription.
  Doses, exact drugs, and timing must be set by your treating clinician, especially if you have comorbidities or are on multiple meds.
  
  1. Overall logic for “afternoon-worse”
  
  Mornings: You have some energy; you don’t want to blow your whole ATP + NE budget early.
  
  Early–mid afternoon: This is the crash window → main support (D-ribose, mast cell coverage, guanfacine, Mestinon if used) is aimed here.
  
  Evening/night: Focus shifts to mast cell quieting + orexin/LC calming (sedating H1, suvorexant), without “pinging” the system with late caffeine or big sugar loads.
  
  2. MORNING (Wake – Late Morning)
  
  Goal: Gentle ramp-up, partial MC control, modest ATP support, but do not use your “big guns” yet.
  
  2.1 Before breakfast
  
  MC stabilizer (if using cromolyn / similar) – ~20–30 min before food
  
  Why here:
  
  Tames breakfast-triggered gut/MC flares that can “steal” energy and push you closer to an early crash.
  
  Sets a calmer baseline going into the day.
  
  2.2 With breakfast
  
  Non-sedating H1 ± H2 (1st dose)
  
  Goal:
  
  Tone down histamine-driven brain fog, flushing, tachycardia, GI upset that can accumulate into a worse afternoon.
  
  Pattern:
  
  H1: cetirizine/loratadine/fexofenadine-type
  
  H2: famotidine etc., as per your doc.
  
  D-ribose – small/medium dose
  
  Timing: With breakfast or shortly afterward.
  
  Goal:
  
  Gently support ATP so the morning is functional, without giving your biggest boost yet.
  
  Rationale for smaller AM dose:
  
  You want to shift the main energy support toward the afternoon crash window.
  
  2.3 After breakfast (late morning)
  
  Caffeine (if used) – modest dose
  
  Timing: After food + D-ribose (≈30–60 min later).
  
  Goal:
  
  Light lift in alertness and focus during the “better” part of your day.
  
  Guardrails:
  
  Keep dose modest to avoid a hard rebound in the afternoon.
  
  No “stacking” with additional late-morning caffeine.
  
  Mestinon – 1st dose (if prescribed)
  
  Timing: Before your limited morning upright activity (short walk, shower, quick errands).
  
  Goal:
  
  Support circulation and neuromuscular function while you do a small amount of functional activity, without pushing too hard.
  
  Guanfacine – Option A (if your main “noise” is later)
  
  If your worst mental noise / adrenergic symptoms are afternoon, you might skip guanfacine here and concentrate it later (see Section 4).
  
  If mornings already feel hyper / over-amped, your clinician might place a low dose here instead.
  
  3. MIDDAY (Late Morning – Early Afternoon, Pre-Crash)
  
  Goal: Position your main supports so they peak right before the usual crash window.
  
  3.1 Before lunch
  
  MC stabilizer (2nd dose) – ~20–30 min pre-lunch
  
  Goal:
  
  Avoid lunch-triggered MC + GI flares that set up the afternoon crash.
  
  3.2 With lunch
  
  Non-sedating H1 ± H2 (2nd dose, if BID)
  
  Goal:
  
  Make sure histamine coverage will still be strong when the afternoon crash window opens.
  
  Note:
  
  If you’re QD only, your clinician may time the single dose here instead of breakfast.
  
  D-ribose – (if not yet moved to main afternoon dose)
  
  Some people with mild afternoon dips might do a medium dose at lunch and a smaller “top-off” later (see below).
  
  For clearly afternoon-worse, the real star is the early–mid afternoon dose.
  
  4. EARLY–MID AFTERNOON (Crash Window)
  
  Critical window. This is where we focus the “big guns.”
  
  4.1 30–60 minutes before your usual crash
  
  D-ribose – main dose
  
  Timing: 30–60 min before the time you usually start feeling heavy, brain-dead, wiped, or PEM-ish.
  
  Goal:
  
  Top up ATP right as energy demand and vulnerability rise.
  
  Support NE vesicle refill / neuronal energy so the LC and cortex aren’t pushing from an empty tank.
  
  Mestinon – 2nd dose (if BID/TID)
  
  Timing: Also in the lead-in to your main afternoon upright block (errands, PT, appointments, dog walk).
  
  Goal:
  
  Improve orthostatic tolerance and neuromuscular performance for your biggest functional chunk of the day.
  
  Guanfacine – main once-daily dose (Option B)
  
  Timing: Early–mid afternoon, just before your worst mental “noise” and adrenergic storm typically hit.
  
  Goal:
  
  Reduce over-firing LC / cortical “noise” when energy is fragile.
  
  Smooth out agitation, racing thoughts, sensory overload, and adrenergic “spikes.”
  
  4.2 During/after the crash window
  
  Caffeine – generally avoid here
  
  Why:
  
  A late caffeine “rescue” can worsen the sympathetic rebound in the evening and disrupt sleep, even if it briefly helps.
  
  If absolutely necessary, discuss with your clinician whether micro-doses are safer than full doses and how close they can be to bedtime.
  
  5. LATE AFTERNOON / EARLY EVENING
  
  Goal: Contain lingering mast-cell & autonomic noise, but start steering toward rest and sleep.
  
  5.1 Before dinner
  
  MC stabilizer (3rd dose, if TID/QID) – ~20–30 min pre-dinner
  
  Goal:
  
  Prevent dinner-related MC flares that can propagate into a bad night.
  
  Non-sedating H1 ± H2 (3rd dose, if used TID)
  
  Placement: Often late afternoon / early evening if patients have a second histamine wave at this time.
  
  Goal:
  
  Keep histamine load from climbing as you move into the evening.
  
  Leukotriene blocker (e.g., montelukast, if used & tolerated)
  
  Common timing: Often evening, since some people prefer to keep any mood/neuro side-effects away from peak functional hours.
  
  Goal:
  
  Blunt leukotriene-driven inflammation that can interact with MC activation and autonomic stress.
  
  5.2 Post-dinner / early evening
  
  Guanfacine – 2nd dose (if BID)
  
  Timing: Late afternoon or early evening, scheduled to precede your usual second wave of cognitive overload / adrenergic noise.
  
  Goal:
  
  Calm NE “gain” without fully muting cognition, support a smoother glidepath into night.
  
  D-ribose – usually avoid large doses now
  
  If used:
  
  Very small dose, with food, only if you personally experience it as calming / stabilizing rather than activating (this is very individual).
  
  Goal:
  
  Avoid late sugar/energy spikes that interfere with sleep architecture.
  
  6. NIGHT
  
  Goal: Let orexin/LC and mast cell axes go quiet so sleep is deeper and less fragmented.
  
  6.1 Pre-bed
  
  Sedating H1 (if part of your plan)
  
  Timing: 30–60 min before bed, or as directed by your clinician.
  
  Examples: hydroxyzine, doxepin, or equivalent (if appropriate).
  
  Goal:
  
  Reduce night-time histamine spikes → fewer awakenings, less “wired but tired,” less nocturnal tachycardia or itching.
  
  Suvorexant – at bedtime as prescribed
  
  Goal:
  
  Quiet orexin → lower LC drive → help the brain stop “surveillance mode,” which is often stuck ON in ME/CFS / LC / MC-heavy states.
  
  6.2 General “do nots” at night
  
  Caffeine:
  
  None after mid-afternoon if at all possible; even small doses can wreck the delicate LC/mast cell balance you built all day.
  
  Big sugar loads (including large D-ribose doses):
  
  Avoid near bedtime; they can spike arousal and sympathetic tone.
  Reply
  - Sharon on November 28, 2025 at 4:09 pm
    
    Thank you Cort!
    Reply
Chris on November 23, 2025 at 10:28 am

Hi Cort! I am definitely in the “use AI to generate possible treatment protocols” camp. The article is very complex and left my brain spinning, but it can be broken down into bite sized chunks that can be reviewed piece by piece as my brain power allows. It certainly provided a ton of ideas to follow up on! IMO using AI to generate a bunch of ideas for treatment is not any different than doing a google search for treatments. It just can handle more complicated inquiries. Regardless of which way you search, you still need to go look at the research that backs up what is suggested. I mean, any google search for CFS treatments yields lots of garbage that you have to sort through so you can discard the ones that are not backed by at least some actual research. Same is true with chatGPT. So I’m not sure what all the controversy is about. There were complaints about the algorithms that chatGPT uses, but a google search uses the same algorithms. So does a Facebook search. Thank you for the idea of using chatGPT to seek a treatment protocol. I don’t use it much and would not have thought of it – I’m excited to try it and see what ideas it can come up with. Given that all the things my doctor has tried have not helped, and that most of the time it is MY ideas that we try if she thinks they are safe, this will simply provide me with more ideas to discuss with my doctor. Thank you so much for all the work you do for our community!
Reply
- Cort Johnson on November 24, 2025 at 5:41 pm
  
  Thanks Chris!
  
  “There were complaints about the algorithms that chatGPT uses, but a google search uses the same algorithms. So does a Facebook search.”
  Reply
Mia A on November 24, 2025 at 8:25 am

Hi Cort
I appreciate the time you’ve taken to do this, and to do all your work.

Like others, I am very reserved about using LLMs for queries or advice.

LLMs are just algorithms that respond to a ‘prompt’ with statistically probable content. So the LLM sees the content of your prompt and compiles a bunch of words that are commonly used together within the training data set. There is no thinking, no understanding. It’s just stringing words together. It cannot judge anything for itself.

To quote Katie Mack on Bluesky: “Chatbots — LLMs — do not know facts and are not designed to be able to accurately answer factual questions. They are designed to find and mimic patterns of words, probabilistically. When they’re “right” it’s because correct things are often written down, so those patterns are frequent. That’s all.”

The way I’m going to approach this particular post is I’ll review the content and then use it to search for things manually.

Thanks again for everything, and for being so open about your process. It is very rare!
Reply
Brian on November 24, 2025 at 11:30 am

In the adjacent field of professional psychology the American Psychological Association has the following sage advice:
“Due to the unpredictable nature of these technologies, do not use chatbots and wellness apps as a substitute for care from a qualified mental health professional.”
The same applies for the diagnosis and treatment of ME/CFS.

https://www.apa.org/news/press/releases/2025/11/ai-wellness-apps-mental-health
Reply
- Cort Johnson on November 24, 2025 at 5:39 pm
  
  The question was “Do you see anything untoward in the suggestions?”
  Reply
Sam on November 25, 2025 at 7:22 pm

Meditation and breathing have by far been the most effective intervention and I’ve tried many different medications – Nortriptaline, Mestinon, Coq10, Magnesium, B vitamins, TUDCA, etc. I wouldn’t discount mind/body techniques.
Reply
- Cort Johnson on November 26, 2025 at 2:55 pm
  
  Thanks for sharing your experience 🙂
  Reply
Deekran Martin on December 2, 2025 at 2:41 pm

I wonder why LDN isn’t mentioned. Seems to have so many good attributes for CFS.
Reply

Visible At Last? Are Long COVID and ME/CFS Tissue Diseases? The PolyBio 2026 Symposium Pt I

Health Rising’s Quickie Summer 2026 Donation Drive

Low Brain Norepinephrine Levels in ME/CFS and Long COVID Pt. II: A ChatGPT Based Treatment Program

<img decoding="async" class="alignright size-full wp-image-67962" src="https://www.healthrising.org/wp-content/uploads/2023/06/Blast-off-Piggy_Istock-1.gif" alt="" width="350" height="262" />Health Rising’s Quickie Summer Donation Drive is On!

THE GIST

Support Health Rising and Keep the Information Flowing!

Health Rising is not a 501 c (3) non-profit

A ChatGPT Treatment Approach: Pro’s and Cons

ChatGPT 5.1’s Treatment Protocols to Increase Norepinephrine Levels in the Brains of People with ME/CFS and Long COVID

The Core Program

Support ATP supply specifically in neurons.

Make Sure NE isn’t being lost.

Prevent neuron over-firing and under release; i.e. stabilize the locus coeruleus neurons and improve signal quality.

SUBSETS

If you feel worse in the morning

If you feel worse in the afternoon

Other Mitochondrial Enhancers

Mast Cells

Worse in the Morning

Other Drugs / Supplements

Caffeine and Stimulants

GLP-1 Agonists

Using Brain Norepinephrine-Enhancing Drugs

ChatGPT’s suggestions for NE Enhancing Drugs in ME/CFS and Long COVID (only use when ATP levels are good)

First-Line Drug to Increase Attention + without Causing More Fatigue

To Improve Attention + Cognitive Stability / Low Sympathetic Tone

If Fatigue is the Primary Problem

Combinations of NE Brain-Enhancing Drugs

Reducing Anxiety Using this NE-Enhancing Protocol

Mestinon

Other Sympathetic Nervous System-Calming Approaches

Neuroplasticity

The “Glue” – Pacing

Final Notes

Support Health Rising and Keep the Information Flowing!

Health Rising is not a 501 c (3) non-profit

<img decoding="async" class="alignright size-full wp-image-67962" src="https://www.healthrising.org/wp-content/uploads/2023/06/Blast-off-Piggy_Istock-1.gif" alt="" width="350" height="262" />Health Rising’s Quickie Summer Donation Drive is On!

163 Comments

Submit a Comment Cancel reply

Health Rising’s Quickie Summer 2026 Donation Drive

Subscribe

Lives Interrupted

Lives Interrupted Latest Story

Lives Interrupted Stats

StudyME

Stay Up to Date with ME/CFS, Long COVID and Fibromyalgia News

Thank you for signing up!

Pin It on Pinterest

Health Rising’s Quickie Summer Donation Drive is On!

Health Rising’s Quickie Summer Donation Drive is On!