Central sensitisation: why pain persists long after tissue has healed
Central sensitisation is the process by which the spinal cord and brain turn up the volume on pain, so that the nervous system keeps generating real pain long after any tissue has healed. It is the main reason a scan can look normal while the hurt is genuine and severe. Crucially, this is a change in how the nervous system processes signals, not a sign that the pain is imagined or that you are exaggerating, and because it is a learned, plastic state it can often be turned back down with the right approach.
Key facts
- In 2017 the IASP formally added nociplastic pain as a third category alongside nociceptive and neuropathic pain, describing pain from altered nervous-system processing without ongoing tissue damage or nerve injury.1
- Fibromyalgia is the prototype of a centralised pain state, and features of central sensitisation also appear in a large subset of chronic low back pain, IBS, chronic headache and chronic pelvic pain.36
- Disc bulges and degeneration on MRI are common in people with no pain at all, which is a large part of why "the scan is normal but it still hurts".2
- Just one night of total sleep loss produces generalised hyperalgesia, impairs the body's own pain-damping system and facilitates pain build-up in healthy volunteers.7
- UK guidance (NICE NG193) recommends not starting opioids, gabapentinoids, paracetamol or NSAIDs for chronic primary pain, and instead points to exercise, psychological therapy, acupuncture and certain antidepressants.9
- Three kinds of pain: nociceptive, neuropathic, nociplastic
- Inside the spinal cord: windup and the dorsal horn
- The brain's dimmer switch: descending modulation and CNS gain
- Why the scan is normal but it still hurts
- Where central sensitisation shows up
- Sleep, stress and fear: what cranks the gain
- What genuinely helps, and what does not
- What to do next
Three kinds of pain: nociceptive, neuropathic, nociplastic
For most of the twentieth century pain was split into two types. Nociceptive pain is the ordinary kind: you sprain an ankle or burn a hand, specialised nerve endings called nociceptors detect the threat, and the amount you hurt roughly tracks the amount of harm. Neuropathic pain comes from actual damage or disease in the nervous system itself, such as a trapped nerve, shingles or diabetic nerve damage, and tends to feel burning, electric or shooting.
The problem is that a large group of people did not fit either box. They had widespread, persistent pain with no tissue damage to explain it and no detectable nerve lesion. In 2016 a group led by Eva Kosek argued that medicine needed a third mechanistic descriptor, and in 2017 the International Association for the Study of Pain (IASP) adopted the term nociplastic pain: pain that arises from altered nociception despite no clear evidence of tissue damage or of a disease or lesion of the somatosensory system.1 A 2021 Lancet review by Fitzcharles, Cohen, Clauw and colleagues set out how this category covers many of the most common and most disabling chronic pain conditions.3
Central sensitisation is the leading mechanism behind nociplastic pain. It is best thought of not as a separate disease but as a change of state in the pain system, one that can ride alongside ordinary nociceptive or neuropathic pain and keep it going.
| Type | What drives it | Typical examples |
|---|---|---|
| Nociceptive | Activation of nociceptors by actual or threatened tissue damage | Sprains, fractures, burns, early osteoarthritis flare |
| Neuropathic | A lesion or disease of the somatosensory nervous system | Sciatica from a trapped nerve, shingles pain, diabetic neuropathy |
| Nociplastic | Altered central processing and amplified pain signalling, no clear damage or lesion | Fibromyalgia, much chronic low back pain, IBS, chronic tension-type headache |
Inside the spinal cord: windup and the dorsal horn
To see how the volume gets turned up, follow a pain signal as it arrives at the spinal cord. Incoming nerve fibres meet a relay station in the back of the cord called the dorsal horn, where the signal is passed on to neurons that carry it up to the brain. This relay is not a passive wire. It can amplify or dampen what passes through, and that gain is adjustable.4
The first clue to how amplification happens is a phenomenon called windup. If the slow-conducting C-fibres are stimulated repeatedly at a steady, low rate, the dorsal-horn neurons fire more and more strongly with each identical pulse, as if the volume knob is being turned up stroke by stroke. This temporal summation depends heavily on the NMDA receptor, a glutamate receptor that becomes progressively easier to activate.4 Windup itself is short-lived and settles when the input stops, so on its own it does not explain months of pain. But the more lasting changes it can trigger, including a long-term increase in synaptic strength and excitability, are what Clifford Woolf described in his landmark work as central sensitisation: a prolonged but reversible rise in the excitability and synaptic efficiency of neurons in the central pain pathways.5
Once that state sets in, two tell-tale features appear. Hyperalgesia means something that should hurt a little hurts a lot. Allodynia means something that should not hurt at all, such as light touch, a waistband or the weight of bedding, becomes painful.5 These are not signs of fresh injury. They are signs that the relay has been left switched to high gain.
The brain's dimmer switch: descending modulation and CNS gain
The dorsal horn is only half the story, because the brain reaches back down and adjusts the relay from above. From the brainstem, descending pathways run to the spinal cord and can either dampen incoming pain (descending inhibition) or amplify it (descending facilitation).4 In a healthy system this is a useful dimmer switch: it is why a footballer barely notices an injury mid-match and why pain can flood in once you stop and rest.
In centralised pain states this dimmer switch is faulty. The brain's pain-damping system becomes weaker while its pain-amplifying drive becomes stronger, so less inhibition and more facilitation reach the cord, leaving the whole system running hot.46 Pain researchers often describe this as raised central nervous system gain. Daniel Clauw's group has shown, using a measure called conditioned pain modulation (where one painful stimulus should normally turn down the pain of another), that this damping response is reduced in people with fibromyalgia and similar conditions, and that brain imaging shows more pain-related activity for the same physical stimulus.6 The hardware is intact, but its settings have drifted toward amplification.
Evidence strength, plainly. Windup, NMDA-driven temporal summation and descending modulation are well established from decades of animal and human laboratory work.45 That these processes fully explain any one person's chronic pain is harder to prove, and the bedside tests used to infer central sensitisation in humans (pressure pain thresholds, temporal summation, conditioned pain modulation) are reliable research tools rather than validated diagnostic tests.6 The mechanism is real; precise measurement in the clinic is still developing.
Why the scan is normal but it still hurts
This is the question that brings most people to read about central sensitisation, and the answer follows directly from the mechanism. If pain is being generated and amplified by the nervous system rather than by ongoing damage, then an X-ray, MRI or blood test that looks at the tissues will often find nothing to match the severity of the pain. The pain is not in the tissue; it is in the processing.
It helps to know that "abnormal" scans are extremely common in people with no pain whatsoever. In a systematic review by Brinjikji and colleagues, disc degeneration was seen on MRI in about 37% of pain-free 20 year olds and over 96% of pain-free 80 year olds, with disc bulges in well over half of older people who felt fine.2 Findings like these are often just the radiological equivalent of grey hair. So a "normal" scan does not mean the pain is invented, and an "abnormal" scan does not prove the disc is the cause. This is exactly why modern guidance steers away from routine imaging for ordinary back pain. We cover the practical side of this in our companion piece on how to actually help chronic pain.
When a scan does matter. Central sensitisation is a diagnosis made after serious causes are excluded, not instead of excluding them. See a doctor promptly, and do not assume it is "just" sensitised pain, if you have new pain with unexplained weight loss, fever, night sweats, a history of cancer, loss of bladder or bowel control, numbness around the genitals or inner thighs, progressive weakness or numbness in the legs, or pain that is much worse at night and unrelieved by rest. These can signal conditions that need urgent assessment.
Where central sensitisation shows up
Central sensitisation is not confined to one body part, which is part of what makes it confusing. The same amplification can show up as muscle pain, gut pain, headache or bladder pain, and the conditions tend to cluster together in the same person.
- Fibromyalgia is the clearest example, widely described as the prototypical central sensitisation state, with widespread tenderness, fatigue, poor sleep and "fibro fog".36
- Chronic low back pain. Not all back pain is centralised, but a recognisable subgroup shows widespread sensitivity, facilitated temporal summation and impaired pain damping without matching tissue damage.8
- Irritable bowel syndrome. IBS is increasingly understood as a disorder of gut-brain interaction in which the gut becomes hypersensitive, so normal amounts of gas or stretch are felt as pain. This visceral hypersensitivity shares the central-sensitisation machinery.10 Our guide on whether it is IBS or SIBO covers the overlap.
- Chronic headache, particularly chronic tension-type and transformed migraine, frequently shows central pain amplification and overlaps with fibromyalgia.3
The practical message is that if you have one of these, you are more likely to have another, and a clinician who recognises the shared mechanism is less likely to keep hunting for a separate structural cause in each region. You can read more about how these systems interact across our health library.
Sleep, stress and fear: what cranks the gain
If central nervous system gain is adjustable, the obvious next question is what turns it up. Three levers stand out, and they are not "psychological" in the dismissive sense, they are biological inputs into the same circuits.
Sleep is the most striking. Poor sleep is not just a consequence of pain, it directly amplifies it. In a controlled experiment, Schuh-Hofer and colleagues found that a single night of total sleep deprivation produced generalised hyperalgesia in healthy volunteers.7 Later work confirmed that sleep loss impairs conditioned pain modulation, the brain's own damping response, and facilitates temporal summation, the windup that drives sensitisation.7 Poor sleep and pain feed each other in a loop, which is why protecting sleep is treated as pain treatment, not an optional extra.
Stress keeps the descending facilitation system primed and sustains a state of nervous-system arousal that overlaps heavily with the wired-but-tired pattern we describe in this piece on the autonomic nervous system. Fear and avoidance form the third lever, captured by the fear-avoidance model of Vlaeyen and Linton: when pain is interpreted as a sign of ongoing damage, it triggers fear, which leads to avoidance of movement, then physical deconditioning, lower mood and hypervigilance, all of which amplify the pain and entrench disability.11 A 2022 meta-analysis by Rogers and colleagues confirmed that pain catastrophising and fear of movement are reliably linked to greater disability, distress and pain intensity.11 Breaking that loop, gently and gradually, is one of the most effective things that can be done.
What genuinely helps, and what does not
Because the problem is amplified processing rather than damaged tissue, the treatments that work best are the ones that retrain the system, not the ones aimed at a tissue that is not the source. UK guidance (NICE NG193, 2021) for chronic primary pain reflects this, and it is worth knowing it has been debated by clinicians who treat severe pain.9
Pain neuroscience education comes first, because simply understanding that hurt does not equal harm reduces the threat value of pain. A 2025 systematic review and meta-analysis found that pain neuroscience education, especially when combined with exercise, reduces pain, disability, fear of movement and catastrophising, though effect sizes are modest and it works best as part of a package rather than alone.12 Graded activity and exercise gradually rebuild confidence and tolerance and gently turn the gain back down. CBT and acceptance and commitment therapy (ACT) target the catastrophising and avoidance that keep the loop running. NICE recommends supervised group exercise, CBT or ACT, and acupuncture as core options.9
Medicines play a smaller, honest role. Amitriptyline, a low-dose tricyclic, and duloxetine, an SNRI, are thought to work partly by boosting the brain's descending pain-damping pathways. The honest numbers matter here. In Cochrane reviews, amitriptyline for neuropathic pain has a number needed to treat (NNT) of around 5 for one extra person to get good relief, with side effects common.13 For fibromyalgia, duloxetine gives substantial pain relief to roughly 1 in 10 people over 4 to 12 weeks, with little evidence of benefit beyond six months.14 These are worth trying for some people, but they help a minority meaningfully and are not cures.
Off-label use, stated plainly. When NICE NG193 recommends amitriptyline, duloxetine, citalopram, fluoxetine, paroxetine or sertraline for chronic primary pain, this is an off-label use of these antidepressants, and most of the supporting evidence comes from women with fibromyalgia.9 Your prescriber should explain this. Do not start, stop or change doses without medical advice, and never stop an antidepressant abruptly.
| Approach | Verdict | Honest framing |
|---|---|---|
| Pain neuroscience education | Helps | Modest effect, best combined with exercise12 |
| Graded exercise / activity | Helps | Core treatment; supervised group programmes recommended9 |
| CBT or ACT | Helps | Targets fear, avoidance and catastrophising911 |
| Sleep treatment | Helps | Sleep loss measurably amplifies pain7 |
| Amitriptyline / duloxetine | Sometimes helps | Off-label; benefits a minority, NNT around 5 to 101314 |
| Opioids | Do not start | No good evidence of benefit in chronic primary pain; real harms9 |
| Gabapentinoids, paracetamol, NSAIDs, benzodiazepines | Do not start | NICE advises against for chronic primary pain9 |
What does not help is as important as what does. NICE advises against starting opioids, gabapentinoids (pregabalin, gabapentin), paracetamol, NSAIDs, benzodiazepines and antipsychotics for chronic primary pain, because the evidence does not show they improve quality of life and they carry real risks of harm and dependence.9 This does not apply to chronic pain with a clear underlying disease, such as inflammatory arthritis, where those medicines may be appropriate. Repeated scans, injections and surgery aimed at incidental findings also tend to disappoint when the real driver is central, not structural. If you are reviewing what you take, our stack builder can help you organise a conversation with your prescriber.
What to do next
The single most validating and useful idea is this: your pain is real, it is produced by a real nervous system doing real signalling, and because that system is plastic, the settings can change. Progress is usually gradual, measured in steadier function and better days rather than a sudden cure, and it comes from steady inputs rather than a single fix.
- "Do you think my pain is mainly nociplastic or centralised, rather than from ongoing damage?"
- "Have we ruled out the serious causes, and do I actually need more scans, or could repeated imaging just find harmless age-related changes?"
- "Can I be referred to a pain management programme, physiotherapy with a graded activity plan, or CBT/ACT?"
- "If we try amitriptyline or duloxetine, what is the realistic chance it helps me, and is this an off-label use?"
- "Can we review any opioids, gabapentinoids or other long-term painkillers that may not be helping?"
References
- Kosek E, et al., 2016. Do we need a third mechanistic descriptor for chronic pain states? Pain. link
- Brinjikji W, et al., 2015. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol. link
- Fitzcharles MA, Cohen SP, Clauw DJ, et al., 2021. Nociplastic pain: towards an understanding of prevalent pain conditions. Lancet. link
- D'Mello R, Dickenson AH, 2008. Spinal cord mechanisms of pain. Br J Anaesth. link
- Woolf CJ, 2011. Central sensitization: implications for the diagnosis and treatment of pain. Pain. link
- Clauw DJ, 2014. Fibromyalgia: a clinical review. JAMA. link
- Schuh-Hofer S, et al., 2013. One night of total sleep deprivation promotes a state of generalized hyperalgesia. Pain. link
- Schuttert I, et al., 2021. The definition, assessment, and prevalence of (human assumed) central sensitisation in patients with chronic low back pain: a systematic review. J Clin Med. link
- National Institute for Health and Care Excellence, 2021. Chronic pain (primary and secondary) in over 16s (NG193). link
- Fikree A, Byrne P, 2021. Management of functional gastrointestinal disorders. Clin Med (Lond). link
- Rogers AH, et al., 2022. A meta-analysis of the associations of elements of the fear-avoidance model of chronic pain. Eur J Pain. link
- Siddall B, et al., 2025. Effectiveness of pain neuroscience education in physical therapy: a systematic review and meta-analysis. Brain Sci. link
- Moore RA, et al., 2015. Amitriptyline for neuropathic pain in adults. Cochrane Database Syst Rev. link
- Lunn MPT, et al., 2014. Duloxetine for treating painful neuropathy, chronic pain or fibromyalgia. Cochrane Database Syst Rev. link
This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.