Alf Breig

The claim

Alf Breig proved the spinal cord is a tensioned structure

Alf Breig was a Swedish neurosurgeon working in Stockholm, and he showed that the spinal cord and its membranes are mechanically pre-tensioned. They change length every time the spine changes shape. He set the case out in two books. Biomechanics of the Central Nervous System appeared in 1960 through Almqvist and Wiksell in Stockholm, 183 pages, with a Chicago edition through Year Book Publishers (Breig 1960). Adverse Mechanical Tension in the Central Nervous System followed eighteen years later, in 1978, 264 pages, published jointly by Almqvist and Wiksell and by John Wiley in New York under ISBN 0 471 04137 8 (Breig 1978). The title of that second book is the phrase the field still uses for the problem. Breig named it, and the name stuck because it was accurate.

Ask what that name commits you to. If tension in neural tissue can be adverse, then tension in neural tissue is a real quantity with a normal range and an abnormal one. It can be measured. It can be exceeded. It can be relieved. That single move takes the nervous system out of the category of signaling hardware and puts it into the category of loaded structure, the same category as a tendon, a ligament or a guy wire. Nothing about conduction is suspended by this. The cord still fires. It simply fires inside a mechanical environment that can help it or hurt it. Every argument on this page follows from that.

The evidence

The case was built on fresh cadaver material, not on argument

Breig came to biomechanics from pain research. His earliest indexed papers sit in a different world, an atypical course of paralysis in poliomyelitis reported in Medizinische Klinik in 1953, and a study of cutimuscular pain reflexes in Der Nervenarzt in 1955. By the middle of the nineteen sixties he was cutting spines. His central paper is Breig, Turnbull and Hassler, Effects of mechanical stresses on the spinal cord in cervical spondylosis. A study on fresh cadaver material, published in the Journal of Neurosurgery in 1966, volume 25, pages 45 to 56 (Breig and Turnbull 1966). The method was blunt and effective. Forty two cervical specimens were fixed in different positions, then sectioned and examined.

Fixing the tissue in flexion and in extension mattered more than it sounds. It meant the cord could be inspected in the shapes a living neck actually adopts, rather than in the neutral posture a dissection table encourages. In the same year Breig published with A. F. el-Nadi in Acta Radiologica Diagnosis, volume 4, pages 602 to 624, under the title Biomechanics of the cervical spinal cord. Relief of contact pressure on and overstretching of the spinal cord (Breig and el-Nadi 1966). Read those two titles side by side. One establishes that mechanical stress deforms cord tissue. The other establishes that the deformation can be relieved. Breig was never only describing a problem. He was always hunting for the maneuver that unloaded it.

Flexion

Flexion lengthens the canal, and the cord has to find that length somewhere

The vertebral canal is not a fixed corridor. Bend the neck forward and the canal grows longer along its posterior wall, because the back of every joint in the column opens while the front closes. The cord inside gets no vote. Breig's cadaver work showed the cord and the dentate ligaments lengthening under flexion, with the anteroposterior diameter of the cord falling in histological section. Length goes up, thickness goes down, which is what any material does when you pull on it. Extend the neck and the process runs backward. The cord shortens and thickens, and the nerve roots and dentate ligaments slacken with it. He had already reported the principle in 1964, in a paper titled Stretch displacement of dura and spinal cord in the spinal canal (Breig 1964).

So the cord is not a passenger. It is a participant in every movement the spine makes. That is the fact most anatomy teaching leaves out, and leaving it out changes what posture appears to mean. A head carried forward for eight hours is not only a load on the neck extensors. It is a sustained length change inside the conducting tissue, held there by muscle that never gets to stop working. Breig took this material to the anatomists on their own ground in 1965, publishing in Verhandlungen der Anatomischen Gesellschaft, volume 115, pages 49 to 69, under the title The biomechanics of the spinal cord and its membranes in the spinal canal (Breig 1965).

Slack

The cord adapts by unfolding first and by stretching second

This is where the popular version of Breig goes wrong. The cord gets described as stretching like elastic. His material does not support that picture. Neural tissue carries reserve length in the form of folding, waviness and axial glide, and it spends that reserve before it develops much real tensile strain. The cord slides within the canal. The roots take up their own slack. The whole tract redistributes length toward the region that needs it most, so a movement made at the neck is answered by tissue as far away as the sacrum. Slack is not sloppiness. Slack is the reserve that keeps the system quiet during ordinary movement.

The consequence matters at the bedside. While reserve remains, movement is cheap. Once reserve is gone, tension climbs steeply for very little further movement. That is a non-linear response, and it explains why symptoms in these presentations tend to arrive suddenly at one particular position rather than growing smoothly across the range. The system has a threshold, and thresholds behave nothing like gradients. Breig spent a career locating it. His 1970 paper in the Journal of Biomechanics runs to three pages and carries the whole claim in its title, that overstretching and circumscribed pathological tension in the spinal cord are a basic cause of symptoms in cord disorders. It has been cited more than eighty times (Breig 1970).

The name

Adverse mechanical tension is Breig's own phrase, not a later invention

The term entered the literature as the title of the 1978 book, and it entered from neurosurgery. Phillip Harris reviewed the book for the Journal of Neurology, Neurosurgery, and Psychiatry in July 1979, page 674 (Harris 1979). Manual therapy authors carried the phrase into physiotherapy teaching over the following decade, and it is now often credited to them rather than to him. The credit belongs to Breig. He had been publishing in surgical, radiological and anatomical journals for twenty five years before the phrase reached a physiotherapy classroom, and the idea arrived there second hand.

This matters for how the idea gets defended. Adverse mechanical tension is not a therapy slogan that later acquired a scientific gloss. It started in a laboratory, in cadaver dissection and operative observation, and it traveled outward from there. Anyone who wants to argue with it has to argue with the tissue. That is a strong position and it is worth protecting. Ideas lose their evidence when they lose their address, and this one has an address, a publisher, a page count and a date.

Pincer

Tension and compression are one problem seen from two sides

Breig gave a chapter of the 1978 book to what he called pincer and clamping actions on the spinal cord, pages 61 to 83. The picture is simple once you see it. A bony ridge on the front wall of the canal does little harm to a slack cord. Put that same cord under axial tension and press it against that same ridge, and the tissue is caught between a longitudinal pull and a transverse push. The two loads do not add. They multiply, because tension removes the cord's freedom to deform out of the way of the obstacle. A rope under load is easier to cut across than a loose one. The same principle holds in tissue.

The cadaver work showed which structures pay the bill. The lateral columns and the anterior horns were deformed by mechanical stress produced by spondylotic bars during flexion. That is a specific claim about specific tracts, not a general statement about pressure, and it matches what those patients report in the clinic. Tension alone is often tolerated. Compression alone is often tolerated. Tension and compression at the same segment is the combination that takes conduction down. It also explains why a static image taken in a neutral position can look reassuring while the patient's symptoms are entirely real.

Numbers

Breig described a threshold, and later engineering put a figure on it

In his 1984 paper in Fortschritte der Neurologie-Psychiatrie, volume 52, pages 2 to 5 (Breig 1984), Breig named the injury that starts the cascade. He wrote of intramedullary fissures typically produced by any compressive spinal cord injury which reduces the anteroposterior diameter of the cord by 25 percent or more. A quarter. That is the deformation at which the inside of the cord begins to tear along its own planes, and it is a figure a radiologist can check against a scan. He then followed the mechanics forward in time. Scar forms in the fissure, the axis cylinders are pushed off their normal course, and fibers that were never severed end up bent and stretched into silence.

Later engineering carried the argument forward with different tools. Henderson and colleagues published a finite element analysis of the craniocervical junction in Surgical Neurology International in 2010 (Henderson 2010). Normal craniocervical flexion produced strain near 0.1, roughly ten percent elongation. An abnormal clivo-axial angle in flexion drove strain toward 0.2, and twenty percent is the neighborhood where axons lose conductivity. The margin between ordinary movement and mechanical failure is a single doubling. That is a narrow margin for a structure carrying no pain fibers of its own to warn you, and it is the quantitative form of what Breig had been saying since 1960.

The periphery

The same logic runs out to the nerve roots and the sacral plexus

Breig did not stop at the cord. With O. Marions he published Biomechanics of the lumbosacral nerve roots in Acta Radiologica Diagnosis in 1963, pages 1141 to 1160 (Breig and Marions 1963). Sixteen years later he worked with the British spine researcher J. D. Troup on the straight leg raising test, publishing in Spine in 1979, volume 4, pages 242 to 250 (Breig and Troup 1979). The study paired six cadaver dissections with a clinical series of 442 patients, and it tested what happens when the hip is rotated medially during the raise. Cadaver and clinic inside one paper was unusual then and is unusual now. It is also the right design for a mechanical claim.

One correction belongs here. Breig did not invent the straight leg raise. That test predates him by roughly a century and reaches modern practice through the students of Lasegue. What he and Troup added was the mechanical account of why the test works and a protocol for performing it the same way twice. Medial hip rotation puts extra tension on the sacral plexus, and adding tension deliberately turns a crude yes-or-no test into a graded one. That is the difference between a sign and a measurement. The sentence below is their own wording from the paper.

One tract

Breig treated the brainstem and the cord as a single tissue tract

In 1973 he published a chapter in Clinical Neurosurgery, volume 20, pages 85 to 94, titled Pathological stress in the pons-cord tissue tract and its alleviation by neurosurgical means (Breig 1973). The phrase pons-cord tissue tract is doing real work in that title. It refuses the usual boundary between brainstem and spinal cord, and it treats the whole neuraxis as one continuous mechanical member running from the skull base to the sacrum. Textbooks divide the neuraxis for teaching convenience. Loads do not respect teaching conveniences. A pull applied at one end of a continuous structure is felt at the other.

That framing is why his name still appears at the top of the spine. Craniocervical flexion raises strain in the brainstem, which is precisely the region any clinician would least like to load. The modern literature on craniocervical instability and abnormal clivo-axial angles rests on this continuity, and the Henderson group named Breig directly among the sources of their approach. He had form for this style of thinking elsewhere too. In 1967 he reported in The Lancet that an elongated basilar artery could produce hydrocephalus, describing it as a new clinicoradiological syndrome. Mechanical shape, once again, producing neurological consequence.

The limits

Breig's surgery is where the reading has to slow down

From 1972 onward Breig pursued a surgical program he called spinal cord relaxation. The argument was that releasing tension around a scarred cord could wake conduction in fibers that survived the original injury but stopped firing. He set it out in Paraplegia in 1972, volume 9, pages 173 to 182, in a 1984 letter to The Lancet written with Troup, and in a short 1989 Springer book written with Michel Renard and John R. Silver, Skull Traction and Cervical Cord Injury, 84 pages. His 1984 abstract compares his own results with those obtained by epidural electrical stimulation in the United Kingdom and the United States.

That program did not become standard care. His claims about restoring dormant function in chronic tetraplegia and paraplegia were not widely replicated, and the operation is not part of ordinary spinal practice today. Say that plainly rather than quietly. The biomechanics survived and the surgery did not. The two are separable, and treating the measurements as discredited because the operation failed would be an error in both directions. Dissection and operative technique answer to different standards of proof. His dissections have been reproduced and extended for sixty years. His operation has not.

Breig and the model

Breig built the cord mechanics of the Unified Model of Tone

The Unified Model of Tone rests its whole account of the spinal cord on Breig's finding. The cord is a tensioned structure whose shape is maintained by the dentate ligaments anchoring it within the canal. Raised tension, rather than compression, carries the primary neurophysiological consequence, because even a compressive lesion generates axial tension. That is Breig's pincer chapter restated as a general law. The implication the model draws from it is direct: any input capable of changing cord tension can, in principle, change cord function. The Hodgkin and Huxley account of the action potential explained the signal. Breig supplied the mechanical conditions the signal runs inside, and the model needs both.

The model names the resulting condition Adverse Mechanical Cord Tension, extending the title Breig set down in 1978. The term covers any state of the cord and its soft tissues that interferes with its function and oscillation and predisposes the whole nervous system to facilitation. The state is non-linear and system-wide. A small change in tension at one site can produce a large global effect while a large change at another produces little, depending on the state the system is already in. That behavior is the threshold Breig located in cadaver material, carried into living regulation. Network Spinal Analysis built an entire adjusting system on this condition, and Breig's work is the floor it stands on.

The suspension the model describes is specific. The dura anchors at the foramen magnum, at the bodies of the second and third cervical vertebrae where it blends with the posterior longitudinal ligament, and at the sacrum. Dorsally it adheres to the posterior arches of the atlas and axis. The dentate ligaments, typically twenty-one pairs, run from the pia to the dura and suspend the cord within that sleeve, and they are stronger in the neck than lower down. That anatomy concentrates the mechanical stakes in the upper neck. In post-concussive states and upper-neck trauma, sustained traction on the brainstem through these ligaments can keep the autonomic centers and the vagal nuclei dysregulated long after the original injury has healed.

Two claims are embedded in this account of cord tension, and both can be tested rather than assumed. The first is that mechanical tension changes what the cord transmits, so the mechanical state of neural tissue is itself a computational variable. The test is direct: a controlled mechanical strain applied to neural tissue should change the fidelity, timing, and rate of what passes through it. Transmission that shifts with physiological strain, measured in fidelity, timing, and rate, confirms the claim. The second is that the body regulates this tension rather than merely suffering it. Cord tension is a controlled variable with a set point, sensed through the cord's own state, adjusted through the tissues that suspend it, and reported upward into the autonomic centers. The test is to change autonomic state while holding posture fixed and image the cord for a corresponding change in its tension or length. A cord whose tension or length changes with autonomic state while posture is held fixed confirms the loop.

Hold the boundary where the record puts it. Breig established that the cord is tensioned, that flexion loads it, and that tension multiplied by compression takes conduction down. He never claimed the body defends a cord-tension set point, and he never studied an adjustment. Those two steps belong to the model, built on his dissections and stated with the findings that would establish them. Breig proved that cord tension is real. The model claims that cord tension is regulated. The measurement is his. The set point is the model's.

What the record shows

Breig's case for cord tension in seven dated findings

• 1960. Biomechanics of the Central Nervous System, Almqvist and Wiksell, Stockholm, 183 pages: the first book-length case that the spinal cord and its membranes are mechanically pre-tensioned (Breig 1960).
• 1966. Breig, Turnbull and Hassler fixed 42 fresh cervical cadaver specimens in flexion and extension for the Journal of Neurosurgery, volume 25: the cord lengthens and narrows in flexion, shortens and thickens in extension (Breig and Turnbull 1966).
• 1978. Adverse Mechanical Tension in the Central Nervous System, 264 pages, Almqvist and Wiksell with John Wiley: the phrase the field still uses is the title of Breig's own book (Breig 1978).
• 1979. With J. D. Troup in Spine, volume 4: six cadaver dissections and 442 patients showed that medial hip rotation adds tension to the sacral plexus and grades the straight leg raise (Breig and Troup 1979).
• 25 percent. Breig's 1984 paper named the threshold: flattening the cord's anteroposterior diameter by a quarter or more produces intramedullary fissuring, the injury that starts the scarring cascade (Breig 1984).
• Twenty-one pairs. The typical count of dentate ligaments anchoring the pia to the dura, the suspension Breig identified as the keeper of the cord's shape.
• 0.1 to 0.2. Henderson's 2010 finite element analysis: normal craniocervical flexion strains the cord near ten percent, and an abnormal clivo-axial angle drives strain toward twenty percent, where axons lose conductivity (Henderson 2010).