Fibromyalgia, ME/CFS, and the Contested Illnesses

TL;DR. These conditions share a pattern that has served patients badly: severe, disabling symptoms with normal blood tests and normal scans. For most of the twentieth century that combination was read as evidence that nothing was wrong, and patients were told their symptoms were psychological. That framing was wrong, and the correction is now reasonably well established: pain can be generated by a sensitised nervous system without tissue damage, which pain researchers formally recognised in 2017 as a third mechanism of pain alongside injury and nerve damage. What remains genuinely unresolved is the mechanism of ME/CFS, where the cardinal feature is that exertion makes people worse for days afterwards, and where a major guideline reversal in 2021 withdrew a treatment that had been recommended for two decades. This chapter tries to hold two things at once: these illnesses are real and disabling, and the science is genuinely incomplete.

Key takeaways

• Fibromyalgia affects roughly 2 to 4 percent of people. It is not inflammatory, not degenerative, and not imaginary: it is a disorder of pain processing with measurable abnormalities on quantitative sensory testing and brain imaging.
• Nociplastic pain was formally recognised in 2017 as a third mechanism, distinct from nociceptive pain (tissue damage) and neuropathic pain (nerve damage).
• The cardinal feature of ME/CFS is post-exertional malaise: a disproportionate worsening of all symptoms after exertion, typically delayed by 12 to 48 hours and lasting days.
• In 2021 NICE withdrew graded exercise therapy as a treatment for ME/CFS and recommended energy management instead, reversing two decades of guidance after re-examining the evidence.
• Opioids should not be used for fibromyalgia. They do not work for this mechanism and they make central sensitisation worse.
• The two errors run in opposite directions: missing a treatable disease that mimics these conditions, and subjecting someone to endless testing when the diagnosis is already clear.

The idea that makes these conditions comprehensible

In short: Pain can be produced by a sensitised nervous system with nothing wrong at the site that hurts, and this is now a formally recognised mechanism rather than a diagnosis of exclusion.

Pain has traditionally been divided into two mechanisms:

The third category is the important one, and it is not a polite label for "we found nothing." It describes a specific set of measurable changes:

• Lowered pain thresholds on quantitative sensory testing: pressure, heat, and cold produce pain at intensities that others find merely noticeable.
• Amplified temporal summation: repeated identical stimuli produce escalating pain, indicating that the spinal cord is winding up rather than adapting.
• Impaired descending inhibition: the brain's own pain-dampening pathways work less well.
• Altered brain activity on functional imaging, with greater activation in pain-processing regions to the same stimulus, and altered connectivity between them.

The analogy that helps: the problem is with the volume control, not the microphone. That is why the tissue looks normal, why anti-inflammatories do not work, and why drugs that act on nerve signalling do.

Central sensitisation is a normal process running persistently. After any injury, the nervous system temporarily amplifies pain signals to protect the healing area, which is why sunburned skin hurts in a warm shower. In these conditions, that amplification persists and generalises.

Fibromyalgia

In short: Widespread pain, unrefreshing sleep, fatigue, and cognitive difficulty, from a sensitised nervous system, with treatments that work modestly and one that actively harms.

What it is

Fibromyalgia is chronic widespread pain with associated fatigue, sleep disturbance, and cognitive symptoms, affecting roughly 2 to 4 percent of the population.

The diagnostic criteria evolved in an instructive way. The 1990 criteria required pain in specified regions plus tenderness at 11 of 18 defined tender points, which turned out to measure general pain sensitivity rather than anything specific and to be performed inconsistently. The 2010 and 2016 criteria abandoned tender points entirely in favour of two scores: a widespread pain index (how many of 19 body regions have hurt in the past week) and a symptom severity scale (fatigue, waking unrefreshed, cognitive symptoms, and other somatic symptoms). Symptoms must have been present for at least three months.

Crucially, the current criteria state that fibromyalgia can coexist with other diagnoses. It is not a diagnosis of exclusion, and someone with rheumatoid arthritis or lupus can also have fibromyalgia, which matters because escalating immunosuppression will not treat the second problem.

What it feels like

What actually goes wrong

The current model has three components:

1. Predisposition. Heritability is around 50 percent by twin studies, with first-degree relatives at substantially increased risk. Variants affecting neurotransmitter metabolism and ion channels are implicated, none decisively.
2. A trigger. Physical trauma, infection, surgery, or sustained psychological stress precede onset in many, though by no means all, patients.
3. Maintained sensitisation. Once established, poor sleep, inactivity, distress, and fear of movement all sustain and amplify it, forming a self-perpetuating loop.

Two findings complicate the simple central picture and are worth knowing. A substantial proportion of fibromyalgia patients, in some studies around a third to a half, show reduced small nerve fibre density on skin biopsy, suggesting a peripheral contribution in at least a subgroup. And a 2021 mouse study found that transferring antibodies from fibromyalgia patients produced pain sensitivity in the animals, raising the possibility of an autoimmune component in some patients. Neither finding has changed practice yet, and both argue that "fibromyalgia" is likely more than one condition.

Treatment

In short: Modest effects from several approaches, best used together, and one class to avoid.

What does not work, and what harms:

• NSAIDs and paracetamol are largely ineffective, because there is no inflammation to suppress.
• Corticosteroids do not work and cause harm.
• Opioids are contraindicated. They perform poorly for nociplastic pain, and prolonged use causes opioid-induced hyperalgesia, in which the drug increases pain sensitivity, making the underlying problem worse. A substantial proportion of fibromyalgia patients are nonetheless on long-term opioids, and de-prescribing them, carefully and with support, frequently improves both pain and function.

ME/CFS

In short: A distinct condition defined by post-exertional malaise, with no biomarker, no established mechanism, and a treatment guideline that was reversed in 2021.

What it is

Myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS) is a chronic, complex, multisystem disease. The 2015 US Institute of Medicine (now National Academy of Medicine) report proposed diagnostic criteria requiring:

1. Substantial reduction in the ability to engage in pre-illness levels of activity, persisting over six months, accompanied by fatigue that is often profound, is new, is not the result of ongoing exertion, and is not substantially relieved by rest.
2. Post-exertional malaise.
3. Unrefreshing sleep.

Plus at least one of cognitive impairment or orthostatic intolerance (symptoms worsening on standing upright).

Post-exertional malaise is the defining feature

PEM is what distinguishes ME/CFS from ordinary fatigue, and understanding it changes how the condition is managed.

It is a disproportionate worsening of all symptoms following physical, cognitive, or emotional exertion, characteristically delayed by 12 to 48 hours and lasting days to weeks. The exertion needed to trigger it can be trivial: a shower, a conversation, a short walk.

The delay is the clinically important part. Because the crash comes a day or two later, both patients and clinicians fail to connect it with the activity that caused it, which is exactly how people are advised to push through and get progressively worse.

Two-day cardiopulmonary exercise testing offers the closest thing to objective evidence: healthy people and people with other chronic illnesses reproduce their performance on the second day, whereas many ME/CFS patients show a measurable fall in capacity at the anaerobic threshold, which is not seen in deconditioning alone.

The history, and why it matters

This is a condition whose history has directly shaped its care, so it belongs in the clinical account rather than as background.

Outbreaks of a post-infectious fatiguing illness were described through the twentieth century, including at the Royal Free Hospital in London in 1955, which produced the name myalgic encephalomyelitis. In the 1980s a cluster in Nevada led to the term chronic fatigue syndrome, which patients have long objected to as trivialising, and the press label "yuppie flu", which was worse.

Through the 1980s and 1990s the dominant clinical model in several countries held that the illness was perpetuated by unhelpful illness beliefs and deconditioning, and that the treatment was cognitive behavioural therapy to change those beliefs and graded exercise therapy to reverse the deconditioning.

The PACE trial, published from 2011, reported that both improved outcomes. It became the basis for guidelines internationally. It was then subject to sustained methodological criticism: outcome measures were changed after the trial began in ways that lowered the thresholds for "improvement" and "recovery", the trial was unblinded with subjective primary outcomes, and objective measures such as walking distance and employment status showed little change. After a legal action forced release of the underlying data, independent reanalysis found substantially smaller recovery rates than originally reported.

In 2021, NICE published a revised guideline that withdrew graded exercise therapy as a treatment for ME/CFS, stated that people with ME/CFS should not be offered any therapy based on physical activity or exercise as a cure, and recommended energy management ("pacing") within the person's energy limits. It also removed the framing of CBT as a curative treatment, retaining it as supportive help for coping with a chronic illness. The US CDC had already removed graded exercise and CBT from its recommendations in 2017.

This remains contested among clinicians, and it is now the position of the major guideline bodies.

What is known about mechanism

Honestly: not enough. The findings that have some replication:

• Onset follows infection in a large proportion, including Epstein-Barr virus, enteroviruses, Q fever, Giardia, and SARS-CoV-2.
• Immune abnormalities: altered cytokine patterns, particularly in the first few years of illness, and reduced natural killer cell function.
• Metabolic and mitochondrial findings: impaired energy production on cellular testing, reduced oxygen extraction during exercise.
• Autonomic dysfunction: orthostatic intolerance is common, and a subset meet criteria for postural orthostatic tachycardia syndrome (POTS).
• Brain imaging shows differences in several studies, without a consistent replicated finding.

There is no diagnostic biomarker, which is the central practical problem: it means diagnosis is clinical, that patients cannot demonstrate their illness, and that research funding has historically been a fraction of that for comparably disabling conditions.

Severity, which is rarely appreciated

Roughly a quarter of people with ME/CFS are severely or very severely affected: housebound or bedbound, unable to tolerate light, sound, or conversation, and dependent for personal care. This group is largely invisible because they cannot attend clinics or participate in research, and severe ME has been associated with deaths from complications including malnutrition.

Long COVID and the post-infectious pattern

In short: A large post-infectious cohort with overlapping features, which has forced attention onto a question that had been neglected for decades.

Long COVID (Chapter 27) produced, within a few years, a very large population with post-infectious fatigue, exertion intolerance, cognitive difficulty, and autonomic symptoms. A substantial proportion meet ME/CFS criteria, including post-exertional malaise.

This has had two effects. It has directed research funding and scientific attention at post-infectious illness on a scale that ME/CFS never received. And it has produced, in a compressed period, the same sequence ME/CFS patients experienced over decades: initial dismissal, then recognition, then a search for mechanism.

The candidate mechanisms under investigation overlap substantially: viral persistence, autoimmunity, microclotting and endothelial dysfunction, mitochondrial dysfunction, and dysregulation of the autonomic nervous system. None is established.

Related conditions

In short: Three conditions that frequently accompany these and that are each genuinely contested to different degrees.

POTS (postural orthostatic tachycardia syndrome) is a sustained heart rate rise of 30 beats per minute or more within ten minutes of standing, without a fall in blood pressure, with symptoms of light-headedness, palpitations, and fatigue. It is objectively measurable, which distinguishes it from much of this chapter, it frequently follows infection, and it is managed with increased salt and fluid, compression garments, graded recumbent exercise, and drugs such as ivabradine, midodrine, or fludrocortisone.

Hypermobility spectrum disorder and hypermobile Ehlers-Danlos syndrome involve joint hypermobility with pain, fatigue, and frequently POTS and gut symptoms. Hypermobility itself is common and often asymptomatic; hEDS remains defined clinically with no identified gene, and its boundaries are contested.

Mast cell activation syndrome (MCAS) is the most contested. A well-defined form exists with objective criteria (documented mediator elevation during episodes plus response to mast cell treatment). A far broader use of the label, applied to multi-symptom illness without those criteria, has grown rapidly and is not supported by evidence, and it is worth distinguishing the two.

The diagnostic trap, in both directions

In short: Both missing a treatable disease and testing endlessly are common errors, and the balance is the clinical skill.

Error one: labelling too early and missing something treatable. Conditions that mimic these and that are genuinely treatable:

A basic panel covering these is reasonable and sufficient in most cases.

Error two: endless testing after the diagnosis is clear. Once the mimics are excluded, further scans and referrals rarely help and actively harm: they delay treatment, reinforce the search for a structural cause that will not be found, and expose people to incidental findings and their consequences. A positive, explained diagnosis is more useful than continued investigation.

Living with these conditions

In short: The dominant experience is not being believed, and that has measurable clinical consequences.

Surveys consistently find that people with these conditions report being disbelieved by clinicians, employers, families, and benefits systems. The consequences are not merely emotional: they include delayed diagnosis, inappropriate treatment, withdrawal from healthcare, and worse outcomes.

Several features drive it. The symptoms are invisible. The tests are normal. The conditions predominantly affect women, and there is documented evidence that women's pain reports are taken less seriously (Chapter 69). And the historical framing of these illnesses as psychological has left a residue in medical education that is only now being revised.

Two things that demonstrably help, from the patient experience literature:

• Being given a positive diagnosis with a mechanistic explanation, rather than a list of what was excluded. Understanding nociplastic pain or post-exertional malaise changes how people manage.
• Being believed. This sounds soft and it is measurable: therapeutic alliance predicts outcome in chronic pain as strongly as most specific treatments.

What the person can do

In short: Get the mimics excluded once, learn your own limits, and be wary of both extremes of advice.

• Have the treatable mimics excluded once, properly. Then stop chasing tests.
• In fibromyalgia, exercise, starting far below what you think you should manage, and progress by small increments. Most failures come from starting too hard.
• In ME/CFS, learn pacing. Identify your energy envelope, plan activity to stay within it, and rest before you need to rather than after. Heart rate monitoring is used by many patients to stay below the threshold that triggers a crash.
• Know the difference between the two above, because it is the central practical distinction in this chapter: graded increases in activity help fibromyalgia and can harm ME/CFS where post-exertional malaise is present. If activity reliably makes you worse a day or two later, that is information, not failure.
• Treat sleep properly, and get sleep apnoea excluded.
• Avoid long-term opioids for fibromyalgia, and seek help to reduce them if you are on them.
• Address depression and anxiety if present, not because they cause the illness, but because they worsen it and are treatable.
• Be sceptical of expensive private testing and treatments marketed to these patient groups, which are numerous, unregulated, and frequently costly.
• Pace your medical appointments too. For severely affected people, a home visit or a telephone consultation may be the difference between getting care and crashing for a week to obtain it.

What's next

In short: Finding a biomarker, which would resolve most of what is wrong with how these conditions are managed.

• Biomarkers. Work on immune signatures, metabolomics, and objective measures of post-exertional malaise. A validated biomarker would transform diagnosis, research, benefits assessment, and, not least, being believed.
• Long COVID research spillover, which is the largest infusion of funding and scientific attention post-infectious illness has ever received, and which is testing antivirals, immune modulators, and anticoagulants.
• Subtyping. The evidence that fibromyalgia includes a small-fibre subgroup and possibly an autoimmune subgroup suggests these labels cover several conditions, and separating them is the precondition for targeted treatment.
• Better trials in ME/CFS, which after the PACE controversy require careful design, objective outcomes, and involvement of severely affected patients who have historically been excluded.
• Undoing the educational legacy, which is slow, and which determines what most patients actually experience in a consultation.

Sources and notes

Nociplastic pain: IASP terminology adopted 2017; Kosek et al., Pain, 2016. Fibromyalgia prevalence: Queiroz, Current Pain and Headache Reports, 2013, and subsequent population studies. ACR criteria evolution: Wolfe et al., 1990, 2010, and 2016 revisions. Quantitative sensory testing and imaging findings: Clauw, JAMA, 2014. Small fibre findings: Üçeyler et al., Brain, 2013. Antibody transfer study: Goebel et al., Journal of Clinical Investigation, 2021. Fibromyalgia treatment evidence: EULAR recommendations, 2016, and Cochrane reviews of exercise and of individual drugs. Opioid-induced hyperalgesia: standard pain medicine references. ME/CFS diagnostic criteria: Institute of Medicine, Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, 2015. Two-day CPET findings: Snell et al., Physical Therapy, 2013, and subsequent replications. PACE trial: White et al., The Lancet, 2011; criticism and reanalysis: Wilshire et al., BMC Psychology, 2018. NICE guideline NG206, 2021. CDC removal of GET/CBT recommendations, 2017. Severe ME prevalence and invisibility: Pheby and Saffron, and patient organisation surveys. POTS diagnostic criteria: Heart Rhythm Society consensus. MCAS criteria: Valent et al. consensus criteria versus broader clinical usage.

Open questions. Almost everything mechanistic. Whether ME/CFS is one disease or several is unknown. Whether the PACE controversy reflects a flawed trial, a flawed model, or both remains argued among clinicians. Whether long COVID and ME/CFS share a mechanism is unresolved and is currently the most likely route to answering both.

Next: the genetic diseases whose cause is known precisely, which is the opposite situation to this chapter. 👉