Low Back Pain and the Nervous System

Low back pain disables more people than any other condition, and most of it is unexplained

Fewer than five percent of low back pain cases trace to a named cause, and the other ninety-five percent take a name that describes the failed search. Nadine Foster and colleagues set that out in The Lancet in 2018.

The low back runs from the lower ribs to the buttock folds. Underneath sit five lumbar vertebrae, the discs between them, and the facet joints that guide how each pair moves. Ligaments limit the movement, and deep and superficial muscles hold the arrangement upright. Nerve roots leave the cord at each level and travel into the leg. Every one of those parts can hurt, and every one of them can be photographed.

Most adults have an episode of low back pain. Most episodes settle within weeks. A minority do not settle, and that minority accounts for the disability, the work loss and nearly all of the cost. Nearly the whole burden of the condition comes from that small chronic group rather than from the enormous number of acute episodes.

The Foster review reports the part that should have stopped the field in its tracks. In the large majority of people with low back pain, the workup names no cause. No fracture, no infection, no tumor, no inflammatory disease, no nerve root under compression. Those specific causes exist and matter enormously, because the treatment for each is urgent and particular. They are the small minority. Everyone else receives a diagnosis of non-specific low back pain.

A condition that disables more people than any other, and whose cause is not found in nineteen cases out of twenty, is a condition described in the wrong terms. The regulation of the region is the term that has been missing.

Non-specific low back pain is named for what the workup rules out

A clinician clears the short list of causes that would change management, finds none of them, and the remaining 90 to 95 percent of cases take the name by default. The label reports the result of the search and says nothing about what produces the pain.

The exclusion list is short, and it is the whole of what the label rules out. Fracture. Infection in the disc or vertebral body. Cancer, whether primary or spread from elsewhere. Inflammatory disease of the sacroiliac joints and spine.

Cauda equina compression, where the nerve bundle at the base of the cord is squeezed and bladder control, sensation in the saddle area and leg power are threatened. A radiculopathy severe and progressive enough to need surgical decompression. Finding any of these changes what happens next, and missing one is a serious error.

Nothing on that list is common. Foster's figure of five percent or fewer covers all of them together. Once the list is cleared, the diagnosis carries no positive information at all. It states that the pain is real, that the dangerous causes have been excluded, and that the search has run out of things to look for.

Naming by absence is a recognizable move in medicine

Medicine has done this before. Essential hypertension means high blood pressure with no cause found, and it covers the overwhelming majority of raised blood pressure. Idiopathic means the same thing in Greek.

Both terms mark a place where a common condition resisted the lesion hunt, and in both cases the missing element turned out to be a regulatory setting rather than a broken part. Blood pressure has a defended value, and the defense can be set wrong without anything being damaged. The page on idiopathic conditions works through the pattern in general terms.

Low back pain sits in exactly that position. The search that comes back empty has been looking for a lesion, and a regulation set wrong leaves no lesion. Sharper imaging will not close that gap, because the search is aimed at the wrong category of thing.

Spinal degeneration is nearly universal and predicts low back pain poorly

The imaging of 3,110 people with no back pain whatsoever shows the same wear the clinic blames for pain. Waldemar Brinjikji, a radiologist, pooled those scans in 2015 and counted how often each finding appeared.

Scans of painful backs do show plenty. Bulging discs, worn facet joints, degeneration, height loss and narrowing turn up constantly in people with low back pain. The experiment that settles what those findings mean scans the people who feel nothing, and Brinjikji's group ran it across the published literature.

The numbers are decisive. Among people with no back pain at all, disc degeneration appeared in 37 percent of 20-year-olds and 96 percent of 80-year-olds. Disc bulges rose from 30 percent to 84 percent across the same age range. Disc protrusions, annular fissures and facet degeneration all followed the same climb. These features were sitting quietly in the spines of thousands of people who had never sought care for their backs.

What the imaging finding does and does not license

A finding that appears in most pain-free spines of a given age cannot, on its own, explain a pain that is specific to one person. The same bulge that a clinic blames for one patient's agony is sitting silently in the next patient's comfortable back. Prevalence that high converts the finding from a diagnosis into a description of age.

Tissue still matters. Discs do herniate onto nerve roots, and when a root is genuinely compressed the leg symptoms follow a mapped pattern and the imaging matches the examination. The pain is produced by a real nervous system doing real work. What the result removes is the assumption that a picture of the parts tells you who hurts. Something other than the visible state of the lumbar spine is deciding that, and it does not appear on the film.

The brain's map of the low back is coarse, and chronic pain blurs it further

Chronic back pain shifts and enlarges the cortical territory for the back, and the size of the shift scales with the years the pain has lasted. Herta Flor recorded that in the sensory cortex in 1997.

The brain holds a map of the body across a strip of cortex, and the map is not drawn to scale. Territory is allocated by how finely a region needs to be resolved rather than by how large it is. The hand and the lips command vast areas.

The trunk and the low back command very little. That is why you can identify a coin by touch alone and cannot tell, with your eyes closed, which of two points a hand's width apart on your back is being pressed. The low back is a large structure reported in broad strokes.

Flor, a psychologist who studies how pain reshapes the brain, compared the sensory maps of people with long-standing back pain against people without it. The patch representing the back had reorganized, expanding and moving into neighboring territory. The blurring was proportional to chronicity. A map already coarse had become coarser.

The motor map blurs with the sensory map

The neighboring strip of cortex issues movement rather than receiving sensation, and it degrades the same way. Siobhan Schabrun, a physiotherapy researcher, mapped the motor representations of the trunk muscles in people with low back pain and asked whether the blurring tracked their suffering. It did. The smudging of the motor map was worse in the people whose pain and disability were worse, in her 2017 report.

The brain's picture of where the low back is and what it feels loses resolution, and the brain's ability to address individual trunk muscles separately loses resolution with it. Distinct muscles that should be commanded independently begin to be commanded as a block. The region is both sensed and moved with less precision than before, and the degradation is measurable in the cortex of people whose lumbar spines look ordinary on film.

A body region can be regulated no more precisely than it is represented

Regulation runs on the report, and the low back files the coarsest report of any high-demand structure in the body. That mismatch is the Unified Model of Tone's reading of why the region became the world's leading site of unexplained pain.

The lumbar spine carries the largest mechanical demand of any segment of the axial skeleton. It bears the weight of the trunk, transmits the force of every step into the pelvis, and does it through a stack of five joints that must stay both stable and mobile. Its representation in the cortex is among the coarsest in the body, far below the hand, the face and the tongue.

Every regulator works from a report. A thermostat controls the room it can measure. A nervous system controls a region to the resolution at which that region reports itself and can be individually addressed. Where the report is fine-grained, corrections are fine-grained.

Where the report arrives in broad strokes, the only corrections available are broad. That is the model's claim about the low back: the mismatch between mechanical demand and representational resolution is the structural reason this region loses its range so readily and recovers it so slowly.

Two consequences follow, and both are observed

The first is that a coarse map has only coarse answers. Faced with a threat it cannot localize, the system protects the whole region at once. It stiffens the entire trunk rather than adjusting one muscle, because addressing one muscle requires a resolution it does not have. Global guarding is the correct response for a system working from a low-resolution report, and it is exactly what a painful back does.

The second is that the loop runs downhill. Guarding reduces the variety of movement, which reduces the variety of the sensory traffic coming back, which further degrades the map. Flor's expanding smudge and Schabrun's merged motor representations are two readings of the same descent. Resolution and regulation fall together.

The claim makes a prediction, and it can be checked against work already published. If the pain is downstream of lost resolution, then treatments that restore the resolution should reduce the pain, without touching the disc, the joint or the drug receptor.

Lorimer Moseley and Herta Flor reviewed exactly that literature in 2012 and found that approaches which normalize the reorganized cortical maps reduce pain. Sharpening the picture of the region moves the pain in the region. That is a result the lesion account has no place to put.

A painful back protects itself by spending its movement options

The nervous system answers a threatened region by guarding it, and the guard outlasts the threat. Paul Hodges spent a career recording what the body does with a part that hurts, and in 2011 he and Kylie Tucker set the pattern out as a general theory.

Their account of adaptation to pain describes a nervous system that redistributes activity within and between muscles when a region is threatened. Some muscles are recruited earlier and harder, others are quieted, and the mechanical behavior of the whole region changes. The trunk is stiffened. Movements are made through fewer routes, more slowly, with less variation between one repetition and the next.

In the short term this is good engineering. A guarded lumbar spine is a protected lumbar spine, and the strategy shields tissue while it heals. Hodges and Tucker name the cost directly: the adaptation carries long-term consequences, because the loading it produces is not the loading the tissues are built for.

Guarding is a loss of options, not a gain in strength

The measure that matters here is variability. A healthy back lifts the same object a slightly different way each time, distributing the work across many combinations of muscle and joint angle. A guarded back lifts it the same way every time, through one stiff pattern. Fewer options means the same tissues take the same load repeatedly, and it leaves fewer answers when the load arrives unexpectedly.

Stiffness reads as safety and behaves as fragility. This is the property the model calls constraint. The constraint page sets out the tension network that holds a spine together, and the slack the nerves and fascia spend when you move. The movement page covers what variability is and why losing it costs a system its range.

The pain system can raise its own gain and leave it raised

Sustained danger signaling leaves the dorsal horn of the spinal cord more excitable, and it stays that way after the input stops. Clifford Woolf's 2011 synthesis gave the change its clinical name, central sensitization.

Danger signals from the back arrive at a gate in the spinal cord. Traffic from nearby nerves and descending traffic from the brain widen it or narrow it. Woolf showed that the gate can be left propped open. After enough sustained input, central sensitization raises the responsiveness of the whole junction, and signals that should never register as pain arrive at a cord that now passes them upward as threats.

Two features of chronic low back pain follow from that single change. Gentle bending begins to hurt, because ordinary mechanical input is being scored as danger. And the pain outlasts any injury, because the change lives in the pain system rather than in the tissue that first triggered it. There is no wound left to find, and there does not need to be one.

Amplification is the property the model calls gain. The gain page and the page on pain hold the general case, including how the descending brake weakens and how it is scored. What matters for the low back is the direction of the error. A region already reported in broad strokes is now reported loudly, and volume has replaced detail.

The brain issues low back pain from its forecast of harm

The brain runs on prediction. It forecasts the state of the body and the world, then works to keep those forecasts from being wrong. Karl Friston set the principle out in 2010.

The free-energy principle is an influential account of brain function, and it changes what a pain is. If the brain is continuously predicting, then pain is one of its predictions: a forecast that this region needs protecting, assembled from incoming signals and from everything the brain already holds about the back.

The forecast has content, and in low back pain the content is usually known. Bending is dangerous. The disc is fragile. This will get worse. A person who has been told their spine is degenerating carries a prediction of harm into every movement, and the prediction is issued before the movement occurs. The trunk braces first and moves second.

The model reads that forecast as a setting of tone. A back braced on an expectation of harm is in a physical state, and the prediction and the state are one thing seen from two sides. That is why altering what the system expects of the back alters what the back does. The general case sits on the prediction page.

Years of low back pain alter the brain regions that do the regulating

Vania Apkarian found 5 to 11 percent less neocortical gray matter in people with chronic back pain than in pain-free controls, on the scale of 10 to 20 years of normal aging. The size of the difference scaled with the years the person had been in pain.

Apkarian, a neuroscientist studying what long-term pain does to the brain, reported reduced gray matter density in the prefrontal cortex and the thalamus. Those are the regions that govern attention, weigh emotional significance and gate the flow of bodily signals into awareness.

His group then watched the brain during the pain itself, following the spontaneous rise and fall of a person's own back pain rather than applying a stimulus. The ongoing pain engaged emotional circuitry and disturbed the default network, the pattern the brain settles into when nothing is being asked of it.

The disturbance was present at rest, when the person was doing nothing at all. A wider review by Apkarian and colleagues in 2011 assembled the same picture across many studies of clinical chronic pain. Structure and function both shift, and the shift concentrates in the circuits of regulation and feeling.

The changes are one finding, not a scatter of lesions

Read as a list, these results look like damage accumulating in unrelated places. Read as regulation, the results collapse into a single observation. A nervous system that has held one region guarded for years, at high gain, on a forecast of harm, has been running a narrowed range for years. The thinned prefrontal cortex, the disturbed resting network and the smudged maps are what a narrowed range looks like in a scanner.

The direction of the arrow is the standard difficulty with observational imaging, and it applies here in full. These studies cannot show whether the brain differences produce the chronic low back pain or follow from it. The model does not rest its case on them, and the prediction it stakes itself on is prospective.

The state that meets the strain decides who stays in pain

A loop converts a passing back pain into a lasting one, and it runs on what happens after the pain arrives rather than on what caused it. Johan Vlaeyen and Steven Linton assembled the evidence for it in 2000.

Their fear-avoidance model describes the sequence. Pain read as evidence of damage produces fear. Fear produces avoidance of the movements that seem to threaten the back. Avoidance produces disuse, deconditioning and more time spent monitoring the region. Disuse and vigilance feed more pain into the loop. People who keep moving and stay unworried tend to recover. People who brace and withdraw tend to entrench.

The time course page holds the general mechanics of entrenchment, including how the spinal cord learns and how a problem is staged from acute to adaptive to entrenched. What matters here is the arithmetic specific to the low back. Almost everyone strains their back. Almost no one has a lasting problem from it. The strain is therefore a poor candidate for the cause of the lasting problem.

The same strain lands on two backs and becomes two different fates, because an input never meets a blank slate. It meets a nervous system already tuned a particular way.

Holding the guard has a running cost

Bruce McEwen named the price of staying mobilized after the danger has passed. His account of allostatic load, set out in 1998, describes the wear that accumulates when stability is bought through constant expensive compensation instead of flexible regulation. A trunk held guarded as standing policy is exactly that purchase, and the bill arrives in fatigue, in stress physiology and in sleep.

Sleep is an input to the back, not a separate complaint

During sleep the brain clears metabolic waste at a rate waking cannot match, with the interstitial space expanding to allow it. Lulu Xie showed that in 2013. A back that keeps a person awake, and a nervous system too braced to settle, block one of the brain's own means of recovery.

That makes sleep a lever rather than a symptom. Restore it and the pain system loses one of the things holding it up. The page on why recovery differs states the general law behind every mixed result in this literature: the same input lands differently on differently organized people.

One network feels the low back and sets the guard that holds it

Bud Craig mapped where reports about the condition of the tissues arrive in the brain, and showed in 2002 that they land in the insula, the same cortex that handles emotion.

Craig's synthesis on interoception argued that the sense of the body's internal condition is a genuine sense, distinct from touch, with its own pathway and its own destination. Warmth, effort, fatigue and ache in the low back all travel that route. The felt back is built in the cortex that also builds mood, which is why the ache of the region and the state of the person are not separable in practice.

The guard is set from the same place. Eugene Benarroch mapped the network that composes autonomic output and named it the central autonomic network in 1993. Threat, emotion and reports from the body converge there and leave as a setting of the sympathetic accelerator and the parasympathetic brake. The circuits that feel the low back and the circuits that mobilize the trunk are the same circuits, reciprocally connected.

How freely a person moves between accelerator and brake can be measured. Julian Thayer and Richard Lane published their model of neurovisceral integration in 2000. Shared prefrontal circuits govern how flexibly the body adapts, and that flexibility can be read in the beat-to-beat variation of the heart.

Reading that variation as one face of tone is the model's interpretation rather than a settled equation, and the page on heart rate variability sets out what a personal number does and does not mean. A system with narrow flexibility stays braced. A braced body keeps its trunk guarded, and that guard is one of the routes low back pain takes to persist.

Quieting low back pain and restoring its regulation are different acts

A drug that damps the cord manages the output in one direction. Restoring tone widens the range the system can move through, and the treatments with the strongest evidence in chronic low back pain all act on the second target.

Nothing here argues against symptomatic relief. A person in an acute pain crisis needs the volume turned down, and lowering it is often the correct and sometimes the only immediate answer. The distinction is between two aims. One overrides a lever and holds the output where it is wanted. The other returns the system's capacity to find its own value, so the pain settles because the regulator has recovered.

Movement is the best-evidenced lever

Jill Hayden and colleagues pooled the whole trial literature in 2021, covering 249 trials in 24,486 people. Exercise therapy reduced chronic low back pain and functional limitation compared with no treatment or usual care, with effects the authors describe as modest. No exercise program changes a disc. What it changes is the variety of movement available, the traffic returning from the region, and the forecast the brain holds about bending.

Retraining the forecast changes the back

Peter Kent led the RESTORE trial, published in 2023, testing cognitive functional therapy against usual care in 492 people with chronic disabling low back pain. The approach works on what a person believes about their spine and coaxes the trunk back into confident movement. It produced large reductions in activity limitation against usual care, and the difference was still present at 52 weeks. Nothing was cut, injected or blocked.

Sharpening the map lowers the pain

The map result is the sharpest test of the resolution claim, because it acts on the representation and nothing else. Moseley and Flor's 2012 review found that approaches which normalize reorganized cortical maps reduce pain. A distorted map can be re-tuned rather than merely muffled, and the pain follows the map.

Major guidance has converged on the same logic. The American College of Physicians, in a 2017 clinical practice guideline led by Amir Qaseem, places non-drug approaches first for low back pain, ahead of reaching for medication. The evidence for each of these approaches is real and modest at once. What unites them is the target. Every one reaches the same coupled regulator.

Bidirectional restoration is the prediction a painkiller cannot copy

One trial design tells an intervention that restores regulation from one that masks a symptom. The Unified Model of Tone stakes itself on a result two groups with chronic low back pain would have to produce by moving in opposite directions.

The claim is bidirectional restoration. An intervention that genuinely restores tone should move a dysregulated system toward the middle from either side. Take autonomic flexibility, read at the heartbeat, or the excitability of the pain pathway, or the discreteness of the trunk motor map. A person whose regulation is too rigid should trend toward more flexibility. A person dysregulated in the other direction should trend the other way. What is restored is the capacity to reach the middle, not any particular value.

A drug behaves the opposite way by design. It pushes one direction, moving the high and the low alike, because it substitutes for the regulator rather than returning it.

Restore the tone and people converge on one center from opposite sides. Mask it and everyone slides the same way. That divergence is the signature, and it is measurable.

State the reading plainly, before the data arrive. The test needs a prospective design with a sham arm matched for contact and attention. The target measure must be taken before the outcome is known, and the direction of the predicted change stated before the data arrive.

Convergence toward the middle from both starting sides is the signature. If the treated group converges no more than the sham group, the input did not restore regulation. If everyone moves the same way regardless of where they started, the input is pushing the output, and half the group ends further from the middle than it started.

The cause of most low back pain is regulation, and regulation leaves no lesion

Non-specific low back pain earns its name because the search runs on the wrong category. There is nothing to biopsy in a smudged map, nothing to resect in an amplified cord, and nothing to cut in a forecast of harm.

Set the findings against each other and they resolve into one account. The pain is non-specific because the fault is in the tuning rather than the parts. Brinjikji's 3,110 pain-free spines show worn discs in people who feel nothing, because structure does not author pain.

The same strain disables one person and passes through another, because an input meets a nervous system already organized a particular way. The pain tracks fear, sleep and long stress, because those are inputs to the regulator. And the treatments that help most act on movement, belief and the cortical map, because those reach the regulator directly.

State the health side and the disease side together, because half of that contrast explains nothing. Tone within its healthy range is health, since the range is what allows a back to guard hard against a genuine threat and then release the guard when the threat has gone.

Tone that drifts outside that range and stays there is what shows up as illness. In the low back it shows up as a trunk held stiff, a cord amplifying, a map blurring and a forecast of harm that no longer expires. Nothing is broken. The range has narrowed, and the narrowed range is the condition.

Where the claim stops

Secondary causes hold throughout. Fracture, infection, cancer, inflammatory spinal disease and a nerve root under genuine compression are real, findable and urgent, and each needs treating on its own terms. Every serious cause must be excluded before the regulatory reading applies, and the exclusion is not optional. What this account addresses is the large remainder, where the workup is clean, the hardware is intact, and the nervous system has lost its range in the region it resolves worst.

How low back pain relates to the rest of the library

Low back pain is the library's clearest case of a region losing resolution and range at once, and each neighboring page carries one part of the mechanism.

Three foundations of tone do the heaviest work here.

• Constraint is the tension network a spine is held by, and the slack the fascia and nerves spend every time you bend, which is what a guarded back stops spending.
• Gain is the volume control behind central sensitization, the reason a light bend can be scored as damage.
• Time course is the axis an episode travels along, from a protective response doing its job to a pattern maintained by its own consequences.

The remaining foundations each carry a piece.

• Input quality is the fidelity of the body's report about itself, and the low back's report is the coarsest of any high-demand region.
• Prediction is the forecasting nervous system, and a braced trunk is a forecast acted on before the movement.
• Set point is the value a system defends, which is why resting trunk stiffness settles higher after an episode than before it.
• Load is the running cost McEwen measured.
• Oscillation is the rhythm that narrows as the guard is held, and coupling is why fear, sleep loss and stiffness arrive together rather than separately.

The condition pages divide the rest.

• Pain is the parent page, holding nociception, the spinal gate, the descending brake and the meaning of nociplastic pain.
• Neck pain is the same argument in a region with the opposite problem, where dense position sense makes the mismatch with the eyes and balance organs the dominant fault.
• Movement covers variability and why losing it costs a system its range.
• Fibromyalgia is what amplification looks like when it stops being regional.
• Insomnia holds the sleep loop that keeps a back sore, and stress and physical symptoms covers the braced physiology that pays for the guard.

Three more pages hold the frame and the instruments.

• Idiopathic conditions is the general case of a diagnosis named by absence.
• Why recovery differs states the law behind every mixed trial in the back pain literature.
• The autonomic nervous system is the anatomy of the accelerator and the brake, and heart rate variability is the measure the bidirectional test would be run on.

Frequently asked