Central Pain: Zeroing in on the devil thalamus

This is a technical article, but it will not seem technical to someone thinking of surgical lesioning for pain. Since most Central Pain subjects have burning pain, we are always interested in studies of heat pain, also called thermo-algesic pain.


The thalamus sits at the center of the brain, straight back from the eyes. What we call the thalamus is really two structures, one in either side of the brain. (You have heard of left brain and right brain). The thalamus is a hub, where signal from the cord is either blocked, or channeled on up to the gray matter, or cortex, a thin layer which coats the outer brain. The chemicals involved in pain are exceedingly minute. The total amount of core pain chemicals in the body could be fit into a thimble. We are dealing with extremely minute, but potent compounds, something like on the order of black widow venom.

It is hard for the public to imagine that tiny imbalances in pain chemistry can cause agonizing, constant pain, but they can and do. In central pain, the overabundance of pain chemicals feels very much like acids, which in fact many of them are. The study of pain chemistry is to a large extent, the study of organic acids and their effect on a neuron.

Because there is no context of shared human experience, it is necessary to resort to comparisons to explain the central pains. A normal observer can get closest to an idea of central pain if they imagine themselves suspended in a solution of acid. The posterior cord pain is different. It involves shooting electric pain and cramping or tightening. So let us tell the observer to imagine they are being shocked while in acid suspension. It sounds rather dramatic, and is. It also sounds a little like hell, and we can assume the simile may be appropriate. The great neurosurgeon, Dr. Ron Tasker, has said that severe nerve injury pain is the worst pain known to man. Given the duration and the wide body distribution, it is hard to imagine that anything else comes close. It is Nature’s torture and it extracts everything we thought about ourselves, almost.

If the egg shaped thalamus is imagined to be in the shape of the human body, lying on its stomach, the pain centers would be at the outside edge of both thighs. The outside area is the VPL nucleus and just inside it is the VPM. The VPL/VPM area is thought to control pain in body and face. Pain from the anterior cord, or spinothalamic tract, arrives at the VP complex, as does the muscle pain signal coming in from the cerebellum (vermis), which we associate with the posterior cord. Patrick Wall showed that there are really seven spinothalamic tracts, but they are wound together like telephone wire and interlace with other tracts as well, resulting in modulation of the pain signal.

We do not know how many tracts come from the vermis since it has never been studied. (The leading scientist of the vermis, Dr. Carl Saab, has written one of the articles at this site). Constant abnormal firings are observed in VPL/VPM in central pain. These are presumed to be associated with the spontaneous pain of CP, since no one has yet done dynamic testing to study evoked pain, just as no fMRI has yet tested for evoked pain in CP. What is known pertains to the spontaneous burning only.

The scientific community has not gotten around to distinguishing between evoked pain, which involves actual sensory input, and the spontaneous pain which reflects hypersensitization of the pain neurons by chemical pain exciters which are constantly produced. The patterns of firing have a periodicity, or frequency. Normal people do not have these frequencies. Because the frequencies are slow, they are sometimes known as “thalamic oscillations”.

It would be a simple matter to prove pain is present, except that this signal cannot presently be measured without open brain surgery. In the meantime, Central Pain subjects must listen to lectures about how weak they are or some other psychological nonsense. This article pertains to new research suggesting that VPL/VPM is not the whole story. This means that more research is needed before concluding that the ablation of VPL/VPM would be effective for Central Pain.

Cord central pain and stroke central pain are both studied here. Although both are called thalamic pain, that term has lately been more reserved for stroke induced central pain. Here we look at a recent study which indicates that the heat pain aspect of central pain may be traveling OUTSIDE the area in the thalamus traditionally thought to carry all pain. This is similar to the anatomists who claim noxious heat is carried in the reticulothalamic tract, as opposed to the spinothalamic tract. This makes a difference if you have burning dysesthesia and are contemplating deep brain implants to the thalamus.

People with cord injury, and those with stroke in the thalamus, can develop identical central pain, except that strokes are usually unilateral and produce pain on the opposite side of the body. We have not yet had a detailed report of facial pain from stroke, but it is not rare in cord injury. When facial pain does occur in the face, it should be on the same side as the stroke, since the crossing of pain fibers occurs in the lower head, ie at the level supplied by the medulla oblongata. In Jan Pain 05, C Montes, et al (including F. Maguierre, a long time contributor to Central Pain) identified a patient with a lesion in the thalamus who developed Central Pain.

This is a very interesting study because it is the first to sort out lesions in the VPL and the VPM nuclei of the thalamus as pertaining to painful heat, which the authors call thermo-algesic Central Pain.

The authors also show sophistication by differentiating between lemniscal pain (posterior column path of lancinating pain, muscle cramping pain etc) and spinothalamic (heat, pins and needles, and light touch pain) tract pain. (It is not yet known where visceral, or hollow organ, pain is carried, but we are examining the survey carefully and suspect it is the ST tract, despite the cramping quality). These two pain systems are NOT the same thing. The results and conclusions by Montes about heat pain are surprising.

The primary sensory relay to the cortex is the thalamus, which is shaped like an egg or short pontoon. It has long been held that pain is handled on the ventral underside of the thalamus (near the subthalamic nucleus, which is also a pain nucleus). If one looks at the lateral edge of the ventral thalamus, more toward the back than the front, there has been identified the ventral postero lateral nucleus (VPL area for body pain) and the ventro postero medial nucleus (VPM area for facial pain) The VPM and VPL have been thought of as more or less similar, with facial pain being handled just a little medial to body pain. Together they are sometimes called the VP complex meaning the part of the thalamus which handles pain for the body and the face. Those who complete the survey show a clear demarkation between those who feel only body pain below the crossing of pain fibers at the medulla and those who feel both body and facial pain. The latter group generally have injury at C5 or highter. Pain from the face, carried in the trigeminal nerve, Cranial Nerve V, descends into the cervical cord to join the substantia gelatinosa, Rexed Layer II, and then comes back up to the brain with the spintothalamic fibers which are coming up from the body.

Touch sensation on the face does NOT drop down into the cord with the descending tract of Cranial Nerve V, but goes directly to the brain. Many medical students learn ONLY the tract serving TOUCH, since amazingly, pain is not well taught in medical school. Just as paralytics are sometimes told it is impossible for them to feel pain, those with injury to the trigeminal nerve in the neck are often told it is impossible to have pain in the face, because the doctor does not know that pain dips down in a descending tract, unlike touch which goes straight to the brain. It is not impossible to feel this facial pain, it is anatomical. The doctor who claims it cannot happen just hasn’t read the literature.

We are now in the interesting position of finding a patient whose thalamic nucleus for heat pain is destroyed, but who still feels noxious heat pain. Instead of telling that patient heat pain is impossible, we can learn from the patient. This is the best science of all, one which accomodates to reality instead of forcing the patient to cram their symptoms onto our Procrustean bed of ignorance.

Recently, the ventro medial nucleus has been subdivided further by the identification of a region at the rear of it, the posterior ventro medial nucleus, or VMpo. In the patient described by Montes, the lesion was in the VPL and the anterior VPM. By using somatosensory evoked potentials (which measures transmission speed in the posterior columns or leminiscal area) and including measurement of transmission of pain in the front of the cord by CO2 laser evoked cortical response, the authors showed that with a lesion in VPL, there was still pain from noxious heat in this patient.

This appears to indicate that thermo algesia is handled further back, in an as yet unknown nucleus of the thalamus. However, relays might be involved. It is not proven that the sensation of heat in central pain is transmitted in the same area as normal heat. This is also true for ALL the painful sensations, since what is perceived in CP is different from normal sensation. A lesion in VPL and VPM might shift some transmission to a neighboring area, explaining the bizarre quality. Harte’s article in February 2005 pain claims that the medial thalamus is part of the little known pain system which involves serotonin, or 5 hydroxy tryptamine. This system is theoretically inhibitory, at least lower down, so heat pain may be due to loss of inhibition. Heat pain has also been associated with the reticulothalamic tract rather than the spinothalamic tract. If so, the medial thalamus may have connections to the reticulothalamic fibers. All of this knowledge is important in crafting therapy.

A “nucleus” is an area of cells doing the “same” thing, Cells subserving one operation are termed “computationally similar”, while cells in the nucleus which do not perform that operation are known as “non-computational” cells of that nucleus. However, neighbor cells in the brain often have a tiny series of holes by which nutrients are shared. This is necessary because blood vessels do not course through brain cells as they do elsewhere in the body. These tiny holes may serve a function of inserting growth or regulation factors, but sharing cells do not necessarily have the same computational function. Only about one third of cells in a thalamic “nucleus” are actually performing the computational function attributed to that particular nucleus. Neighboring cells not only contribute “nutritionally” to each other, but they also release chemicals into the space around cells, which can be potent influencers since much of cell activity occurs at the surface, on the cell membrane,where channels move chemicals both ways across the cell membrane.

This space where chemicals bathe the external membrane is known as the perineural space. Although it is tempting to think of each neuron as an individual, the brain is all about communication. Each neuron has at least six thousand synapses where other cells add or subtract regarding a particular effect. Any cell in the brain can theoretically communicate with any other cell in the brain. Given the trillions of brain cells, and calculating permutations and combinations of six thousand synapses for each cell, an author in Scientific American declared that there are more potential synapses in the brain than there are atoms in the known universe. Cells must communicate, and there is a tremendous amount of noise going around in the circuits of the brain, particularly in the security circuits of the pain apparatus, which must be kept in readiness.

The miracle of the brain is its ability to sort out true signal and suppress noise. This ability may be impaired in central pain, possibly because chemicals meant to repair damage increase the gain in (hypersensitize) damaged neurons which retain the ability to fire in an abnormal fashion. Firing in damaged neurons often lacks the ability to move chloride to the membrane, due to failed production of KCC2, the protein carrier of chloride. This makes what firing they can do into excitation. Without chloride, inhibition need not apply. the job has been taken by excitation. The pain system was not meant to be excitation only. Inhibition is an essential part of pain, a part which is lost in CP. The steady state condition in nerve injury is spontaneous continual firing without input or stimulus, the dreadful spontaneous burning of CP, which is called dysesthesia. It is an alien monster.

Cooperating cells of the brain must be separated into those performing the same computational activity, and those which only join nutritionally by the tiny holes, or which do not join at all and are just using that route to get somewhere else. Neighboring cells not performing the same computational activity also synapse (physical/chemical junction) with the neurons of a nucleus and synaptic connections virtually control what a cell does.

Excitatory synaptic events seem to predominate in the cord, while inhibitory synaptic results seem to predominate in the brain. The brain must sort out signal from a tremendous deluge of “noise” which comes in from the neurons in the body, which ALL fire all the time, at some level (contrary to the myth that we never use more than five percent of our brain). Chemical neurotransmission is inherently slow compared to the movement of electrons. The body gets around this by keeping pain neurons very close to firing. The neurons approached a firing level and then drops back and then moves close to firing again.

During this process, some of the neurons actually go over the mark and generate an action potential. This action potential sets off a little current. Since there are millions of these action potentials reaching the brain at any one moment, it is a miracle that the brain can derive signal in the midst of all the noise. In medicine, this is done by watching several thousand tracings, and having a computer average them and then sort out any non averaged events. This leads to the tracing we call a somatosensory evoked potential. It is not known how the brain does this, and how it does it so rapidly.

The marvelous inhibitory work of the brain allows us to think of one thing at a time. We only wish it would think of nerve injury pain less often. The thalamus, and the changes in it from nerve injury, are what determines that central pain will be perceived. For this reason, central pain is also called thalamic pain. This makes it tricky to say that an area of the thalamus devoted to normal pain would always be the same cells registering nerve injury pain. It is conceivable that injured cells might recruit nearby cells through the release of chemicals from their injured chromosomes.

It is also possible that non computational neighbors might be injured and release chamicals into the remaining functioning neurons, altering their chemical behavior.

Existing theory indicates that injured neurons have a failed inhibitory apparatus. Growth and repair factors hypersensitize injured neurons by greatly increasing the production of certain ion channels and kinases. Normally inert chemicals are converted to the active form through the action of kinases which attach high energy phosphate bonds. The combination of these events, new ion channels and activation of kinases, leads to failed inhibition and greatly enhanceed excitation, the double whammy of central pain. The work of Devor indicates that growth factors affect not only injured neurons, but also their uninjured neighbor neurons. The result is what we call central pain, or hypersensitization of neurons in the area of damage and in related areas. “Peripheral damage induces central change”–Patrick Wall.

In the cortex (outside gray matter of the brain), we have three known areas for pain. SI, or the postcentral gyrus, locates the pain. The operculum of the parietal lobe identifies the type of pain, and the insular cortex subserves the painfulness of pain. Since the cortex receives its signals from the thalamus, scientists wonder exactly what is going on in the thalamus. Are there discrete areas of the thalamus which subserve mechanical touch pain, joint position pain, thermal pain, etc?

The work by Montes suggests that a patient with a knocked out vento posterior complex can still feel heat pain. This does not prove that thermoalgesia of central pain is carried in the same area of the thalamus as normal heat pain, but it is likely to be true. This is interesting since all pain was traditionally thought to route through the VP complex. If another area is found to signal the painfulness of a thermal pain, lesioning that area might stop burning dysesthesia.

A critique we have of the literature is that it does not differentiate between spontaneous pain and evoked pain, and between the areas actually being stimulated with noxious input and neighboring skin, which is not stimulated but may nevertheless develop “secondary allodynia” (see elsewhere at this site, using search). Unless these investigations are made, the scientist cannot be sure what pain is actually being tested. Lumping spontaneous pain and evoked pain, and lumping anterior cord pain with posterior cord pain is clinically oversimplified investigation.

We need work like that of Montes to help us understand the thalamus. We have to start somewhere, and given the small number who actually study the thalamus, we are thankful for what we can get. We quibble over this point because those completing the survey show that posterior column pain (shooting and muscle pains) can usually be treated with opioids, while spinothalamic anterior cord pain (burning to light touch and hypersensitivity to heat) often cannot be treated with opioids. We would like to know why this is true, and would also like to see scientists speak of posterior and anterior cord pain in different words, as Montes does. Montes’ work also confirms our opinion that different types of pain (heat, touch, chemical, etc) are probably handled in different locations of the thalamus. We are not the first to make this suggestion, but we would like to think we set an example to the scientific world of NOT lumping all pain together.

Very intense examination of the excitatory reticular activating system in the medulla oblongata is presently going on. Cholinergic pain pathways are of interest for their inhibitory inpact as well, as work on the PAG (periaqueductal continues. The major role of the subthalamus is also under consideration. It is past time for thinking all pain must go to VPL/VPM via the spinothalamic tract. The reticulothalamic tract (from cord to medullary reticular activating system) must be considered, and names be given to the pain modalities which traverse it. It is always an encouraging sign to see investigators speak of the nerve injury pain in the plural, as in “the central painS”. For example, if the reticulothalamic tract were found to carry evoked pain, and I had only sponataneous burning, what good would lesioning the reticulothalamic tract, or its attachment at the thalmus, or wherever, do? It is time to get specific. Montes’ work is a step in that direction.

Those who lesion the thalamux must be aware of the symptoms a patient has, AND understand the precise area which carries those particular symptoms, before stereotactic planning of lesions to be placed. It is unscientific to publish articles in the literature which report only that a patient had “central pain” and that they were “benefitted” by the neurosurgical procedure. What pain and what benefit? We must know. We are entitled to know. If a subject has only spontaneous pain, or only evoked pain, it is essential to understand where these pathways travel.

To a neuroscientist, a nucleus is merely a collection of cell bodies devoted to doing the same thing computationally, (as opposed to nutritionally dependent on each other). Montes did confirm the traditional idea that painful touch is carried in the VPL and anterior VPM.

We have taken an intense interest in those whose pain is spontaneous only, evoked only, or is both spontaneous and evoked. Nearly all CP subjects experience burning from EITHER light occlusive touch or from heat. It is presumed that occlusive touch, which prevents air from circulating across the skin, has heating as an evocation. It would be helpful if the neuroscientists would include information on the clinical symptoms, when anatomical conclusions are made. As much as we love the work Montes is doing, it is hoped that the thermal pain can be differentiated into spontaneous and evoked. It is not precisely clear whether the unidentified nucleus would be handling thermal pain which would be painful to normals, or not. A decreased sensation to noxious heat was present in this patient, who experienced painful thermal sensations to some heat.

We are attemping to find a map of the thalamus which we can post so that this article will make more sense. For our purposes we are concerned only with the VPL and VPM nuclei, but would liketo be able to show you where these small areas which cause such big trouble are located. It has not been so long that neurosurgeons ablated this part of the thalamus, and later, stimulated it with electrodes. Results were mixed. It is still dangerous and tricky to do deep brain surgery on the thalamus.

The trend seems to be more toward magnetic stimulation of the motor cortex, even though it is unknown why stimultion of the MOTOR cortex should inhibit pain in the SENSORY cortex. The possible answer may lie in the work of Carl Saab, (see his article at this webstie) which indicates that the vermis, part of the motor cerebellum may be the focus of inhibition of pain through muscles. It is also posssible that the motor cortex in the brain has some linkage with the postcentral gyrus, which is just across a little groove which runs across the brain.

We are very happy to see Maguierre still doing brain research and hope that the younger scientists in Malaga Spain will continue to study central pain. We have posted elsewhere the work of Sergio Canavero in Italy, one of our correspondents, who reported the undesirable creation of painful phantom limbs from motor evoked cortical stimulation, so we are merely noting the practice, not promoting it.