Pain and Regional Cerebral Blood Flow

PET scan results raise many questions about pain.

The Golden Rule:

Administer analgesia unto others as you would have it administered unto yourself.

This article can be read without a knowledge of brain anatomy, but reference to a diagram helps pull it together.

A simplified chart of brain anatomy can be viewed at

Save image and then enlarge the jpeg image to read the legend.

Pain travels up a peripheral nerve to the dorsal root ganglion, enters the cord via the dorsal horn, crosses over, ascends to the thalamus inside the brain, and then on to the primary somatosensory cortex, (SI) on the brain surface. Right? Apparently, Wrong. The SI does not increase cerebral blood flow (which is thought to reflect brain activity) as the thalamus does, when the thalamus responds to cutaneous pain.

Further, we should have been suspicious after the reports of BILATERAL central pain following UNILATERAL lesions in the brain.(See Canavero S, Bonicalzi V Bilateral Central Pain with Unilateral Brain Lesion Eur. Neurol 1999:42:118, See also Kim, J.S. Delayed-onset ipsilateral sensory symptoms in patients with central poststroke pain. Eur Neurol 1998:40:201-206) With all the focus on what was going up toward the brain (orthodromic), there should have been more attention paid to what was coming down (antidromic). In other words, there was a circuit going that traveled BOTH directions. The PET scans are going to have to look at what is going downward in central pain, not just what is coming upward into the brain. Thalamic circuits are not a new idea, just not a popular thing to study.

Cortico-cortico, and cortico-thalamic crosstalk should have been recognized and further investigated before we set up imaging experiments to confirm old ideas that were sure to be disproved.

So far, every attempt to simplify pain has only succeeded in revealing greater complexity. It is something like Gordon Moore (of Intel) who proposed a law that the number of transistors which could be placed on an integrated chip would double every two years (Moore’s “law”). By analogy, the known complexity of pain circuity/chemistry will double every two years (Painonline’s law).

It is clear we must find a cure before threading out every truth. Perhaps someone will find a way to deliver resiniferatoxin to EACH dorsal root gangion to kill all the TRPV1 receptors. Then we can search out the chemistry at our leisure. At the present time, however, theory is so fascinating and so fluid that few have decided to go for the kill. (See below for one possibility from Sergio Canavero).

How primitive are we in our understanding pain anatomy? Well, we don’t really know. We don’t know what we don’t know. If we did know the truth, then and only then, would we know how close or how far from the real truth our science is.

Certain of the fMRI and PET findings are so much in conflict that some have even termed such research “modern day phrenology”, which is hopefully an unjustified assessment. In the opinion of the painonline editorial staff, the researchers are missing the obvious weak point, STUDY DESIGN. We need to get clinicians closer to the magnet and imaging scientists closer to the clinicians. A patient who evokes nerve injury pain to a particular stimulus will disappoint if a different stimulus, or none, is being applied. Knowledge of CP fundamentals should make this obvious.

There is plenty of emotion which modifies brain response to any pain. When researchers understand more about the confounding nature of mental status during pain testing, then they should be able to design better investigations.

For example, many SCI patients have some degree of spasticity. The loud clacking of the MRI magnet startles and annoys them, and this can be distressing. Perhaps the central pain subject is having a terrible time holding still or is about to choke on saliva. Perhaps the cold of the room is burning their skin (dysesthesia).

The brain response to such peripheral matters may have nothing to do with pain, but would almost certainly show up in the study. Using concommittant EEG as a backup is an interesting proposition, but does not eliminate problems stemming from poor study design.

Another example might be the person already in serious spontaneous pain from CP. This person may well respond to an evoking stimulus, but where is the imaging lab prepared to apply various evoking stimuli, such as implements of graded sharpness, temperature probes, occlusive skin covering, etc. ?

A PAIN IMAGING CENTER, set up at say the NIH would be best, where study design could be refined, but there is no money at the present time. There may never be unless the nation finally decides that pain is a priority, similar to the way infectious diseases are considered a priority.

Since more money is spent on pain than any other medical item, it seems reasonable to take pain seriously. However, we tend to take only our own physical pain seriously, not that of others, and not as a general matter for study. Yes, in the future, this attitude will seem to have been Medieval, but this is the climate in which we live.

Under this fiction, which excludes oneself, pain is there and there is not much you can do about it–stop whining. As we all know, “There is no problem which cannot be made worse by whining about it” Severe physical pain should be excluded from this axiom. We suggest a new saying, “There is no axiom about pain so insensitive that it cannot be made worse by implying that those in severe pain are whiners.”

It is conspicuous how silent the “stop whining” crowd become when they personally are laid out on the surgical table for an operative procedure, where pain relief is suddenly, without any discussion at all, a HUMAN RIGHT of the very FIRST order. Period. End of report!

Of course, with cost containment now an issue, maybe they would like to have a less expensive, but certified, personal pain trainer come into the surgical suite and mock them by shouting “stop whining”, in place of their costly anesthesia. Even childbirth today rates an expensive epidural, and we have been assured by parturient central pain sufferers that central pain is much worse than labor pains.

There are those who assume nerve injury pain could not be very severe, but this misinformation can be eliminated easily with a tiny injection of capsaicin, to turn on the TRPV1 receptors in one small area. Then, the magnitude of the identical sensation over a large area of the body becomes, well, inconceivable. There are no skeptics about nerve injury pain among capsaicin testees, yet the burn from capsaicin is modest compared to the real thing, to fully elaborated central pain, the worst pain state known to man. It is also certainly the most relentless.

Given the misunderstandings about the multiplicity of pains, and variance in evoking stimuli in nerve injury, it should therefore come as no surprise that pain imaging studies frequently fail to agree with each other. Study design cannot progress without the inclusion of an expert clinician to weigh the variables in the pain calculus. Since fMRI scientists typically regard the magnet as proprietary to the specialty, they are led into frequent errors in recognizing the different types of pain.

Svensson et al in J Neurophysiol. 1997 Jul;78(1):450-60, shed early light on how the brain handles pain. The researchers used a technique of nuclear medicine known as positron emission tomography or PET. What is looked for is regional cerebral blood flow. This is taken to indicate a reaction or area of processing which reflects brain activity. One must be careful to sort out the emotional reaction centers since these can be expected to be involved in any pain experience.

For quite some time it has been unclear whether emotion and pain can really be separated, but fMRI and PET seem to indicate that they are separable. This information, if correct, undercuts many assumptions about pain from certain behaviorial theorists, who prefer to see all pain as emotional in origin (although it escapes us what could possibly be emotional about holding one’s hand over a Bunsen burner).

There has long been talk of the “painless soldier”, and such, to the point where some practically stopped acknowledging that pain was an actual sensory phenomenon. Pain seems to be more like water flowing, which can be stopped by a dam (distraction) but if continuing, soon overflows the barrier. Anesthesia just happens to be a very high barrier. The pain is blocked, but when the patient awakens, the cutting has stopped, so pain is much less. Yet, pain messages in the brain can be recorded even as the patient is anesthetized.

And, reexamination of the painless soldier usually reveals that once the shock is past they may require large amounts of opiates, even if the bullet has long since passed through. Further, for every soldier who felt no pain, there are thousands who felt plenty of pain. What is happening is that once there has been time for neuroinflammation, for the glia which surround neurons to put out the growth factors which activate the inflammatory chemicals so that the area will be hypersensitized, and therefore not used, the pain has grown exponentially.

The number of glia increases by four, but the pain seems like it has grown to the fourth power. New ion channels, not seen since fetal life, come pouring out of the protein factories in the DNA/RNA. Acids form, and the TRPV1 channels open like floodgates to allow pain currents to pour in. THIS soldier needs morphine, and plenty of it. Once healing occurs, the neuroinflammation, because it is under proper control, subsides, and the body allows the soldier to be up and around.

However, for the poor unfortunate who progresses to permanent nerve injury pain, there will be no relief, the pain will never stop. Eventually, the soldier is not a soldier any more, and in many ways, not human. They cannot whine, they can only giggle nervously as they dangle helplessly between earth and hell. If they are paralyzed, others may defer to the injury; but they never defer to the pain, because everyone EXCEPT the CP sufferer knows that nothing is worse than paralysis.

The deeper truth is that there is something much, much worse, endless severe pain. “Will someone please make them stop whining?”. Yes, right after we get the paralytics up and running around the track, we will be sure to tell them to shut up in the name of decency. Let us summon those who successfully use shaming in the operating room to silence the cries. Hmmm. Where are such people when we need them??? There seems to be a shortage at the moment.

Then, perhaps we might call for pain clerics, instead, to read surgical patients the book of Job. What! no ecclesiastical anesthesia in the operating room either? Then, perhaps we will be compelled to do the hard work of nerve biochemistry, since we cannot keep central pain subjects asleep for the remainder of their lives. A shortcut would save a lot of trouble, but it looks like this problem will require man to follow the “out of Eden” command to subdue the earth.

This really tests your faith, sort of like other theological issues such as mosquitos, Antarctic winters, the impact after falling off the Empire State building, and the jaw muscles of crocodiles. By the way, why doesn’t God deliver turkeys already plucked and roasted to set on the doorstep?–Wait! Please excuse. That sounded a little like whining. It is hard to delineate unjustifiable whining. It depends on whose ox gets gored. (We attempt a little humor here, in respect to Dr. Patrick Wall’s plea not to theologize pain illness, for fear this would lessen the drive for scientific research.)

During WWII, forty times as many of Merck’s 1/4 grain syrettes of morphine were used as there were soldiers in the military, wounded or not. What a lamentable failure of proper pain prevention. Foolishly ignorant of the emotional ways to block out pain, the military included the syrettes in every parachute and in every kit, along with a tourniquet and some sulfa. (see Flags of our Fathers, by James Bradley). A manual on how to psychologically overcome pain would, if you subscribe to some opinions, have been unbreakable, cheaper, and equally effective. May such authors have a bundle of these instructional pain avoidance booklets to bury themselves in during the next root canal, gasoline explosion, or nearby asteroid collision.

Henderson et al in Neuroimage. 2007 Nov 12 presume a “pain neuromatrix” (a theoretical term, not necessarily anatomically correct), by which is meant the ANTERIOR-cingular, insular, somatosensory, and cerebellar cortices; as well as, emotional pain processing areas, which they indicated includes MID-cingulate cortex, dorsolateral prefrontal cortex, hippocampus and cerebellar cortex. Unfortunately Henderson had no way to show whether activity (rCBF) in the named areas was excitatory or inhibitory. At least in the case of the roof structures in the cerebellum (vermis) we know, thanks to Carl Saab, that the action is inhibitory, so it may be presumptive to automatically assign cerebellar pain tracts to the emotional column.

Cerebellar response to central pain was first reported at the Ninth World Congress of the IASP by Carl Saab, who has authored articles here at painonline. His findings were considered so radically different from existing theory, that a storm of objection met him from some individuals. He has been vindicated many times over.

Svensson’s study is a bit more intriguing because it addresses the bilaterality of pain centers to a greater degree. We list below the findings. Remember that pain coming up from the body is CROSSED at the cord and so pain on the right side usually shows up in the left brain. Crossed pain is contralateral. Pain coming up and staying on the same side is ipsilateral (such as in the outermost marginal layer, or Rexed Layer I of the dorsal horn of the cord) With this in mind, one can see that most pain shows up on the opposite side of the brain.

Svensson found regional Cerebral Blood Flow after pain in:

Contralateral anterior insula
Contralateral somatosensory cortex (SII)
inferior parietal lobule (Brodmann area 40)
contralateral lateral prefrontal cortex (Brodmann areas 10/46)
contralateral anterior cingulate cortex

ipsilateral premotor cortex (Brodmann areas 4/6)
Ipsilateral cerebellum.

Astoundingly, the primary somatosensory cortex (SI*) {traditionally considered the main pain area] and the primary motor cortex (MI)[considered of late to be a primary pain area] lit up a little but the difference of pain/nonpain was not statistically significant.

Svensson noted a NEGATIVE correlation between changes in rCBF for thalamus and MI/SI for cutaneous pain. This means that what causes increasing blood flow to thalamus (presumably pain) causes a DECREASED activity in BOTH the primary sensory and primary motor areas. This raises the question whether SI and MI might actually be trying to shut down when the thalamus is overloaded with pain. Given reports of pain benefit from transcranial DC current to the primary motor cortex, SI and MI may actually be acting as pain supressants at a certain point in the pain curve. In other words, exactly the opposite of what they have long been assumed to have been doing, “causing” or “generating” the pain.

Svensson also noted a POSITIVE correlation between blood flow to the thalamus and the anterior insula. This nearly cements the claims by Crick and McHenry of the important place of the insula in pain traversing the thalamus.

More studies are needed, but it looks like we must go back to the drawing board. So many assumptions are in question that we must go back and question everything carefully.

Major assumptions about whether a patient is in pain (eg. if your SI doesn’t light up, you are not in pain) have followed PET and fMRI, but of course it is necessary to actually know where pain TRULY appears before such value judgment should even be attempted. As the studies above show, such hasty conclusions were unjustified, since SI does NOT light up when the thalamus does.

Central Pain is also known as thalamuc pain, although of late, some have used thalamic pain to refer to central pain after stroke. There is no absolute requirement to so limit the term “thalamic pain” since Dejerine and Roussy used “thalamic pain” and “pain of central origin” interchangeably. S. Weir Mitchell also termed it pain of central origin. Riddoch first popularized the shorter term “Central Pain”.

Some now wish to call it Central Pain Syndrome, but the addition of “syndrome” while applicable, stems from the variation seen in discrete etiologies, not merely from a collection of identifiable clinical signs. Dr. Canavero chooses “Central Pain Syndrome” and that is good enough for us. He and he alone has done a good job of describing how the condition differs in say, SCI vs. stroke, traumatic brain injury vs. MS, etc. To our knowledge, no one else has produced such data, although it is essential to grasping the problem.

Canavero reminds us that central pain manifests differently according to the etiology, and quite possibly according to the anatomical location. We like anatomy, it is so much easier to grasp than chemoarchitecture. If you have not yet purchased a copy of Sergio Canavero’s book from Cambridge Press, “Central Pain Syndrome”, you should certainly consider doing so.

Clinical signs also have their drawbacks. Ultimately, how we should subdivide the pain may well turn out to be based on acidity levels or components of neuroinflammation, rather than by clinical signs, which are elusive for any but the most experienced and highly trained expert, and sometimes, even for them. The reason is the lack of a vernacular for Central Pain. Without a vocabulary available, the patient can only go so far and no further.

The clinical picture is probably more reliable now than any imaging–ie. a careful medical history and skilled exam, following the guidelines of Dejerine, Egger, and Roussy on history and exam in pain of central origin (including testing for Mitchell’s delay).

It seems a bit ironic that before Dr. Francis Crick and Dr. Kenneth McHenry linked the “painfulness of pain” (Crick’s term) to the insula, in a short article at painonline; (shortly thereafter Crick alone published the complete article in the journal Pain), pain professionals paid little attention to the insula. Yet, presently, after only a year or two, the insula is considered the lead area in pain.

The imaging teams are doing pain patients a great service and it seems likely that it will shortly be impossible to challenge someone with severe central pain, because imaging will have taken it out of the realm of doubt. It is not really in doubt now, for good professionals, but too many ignore the literature of the past, where the characteristics clinically of central pain were well worked out. The signal quality of burning dysesthesia, which is that it evokes with LIGHT TOUCH, but NOT heavy pressure or hard blows, is hardly even mentioned in the literature today.

Other important features of CP such as Mitchell’s Delay and atopoesthesia are almost unknown. The imaging scientists will overrun the examiners before they make themselves learn the lessons of the past. Then, the position of CP will be secure. This is not the goal of course. A cure is the desired end, but just proving that CP is lurking should lead to much more intense study.

*The “I” in SI is a Roman numeral.

The clinical signs of central pain can be reviewed at the various articles here at this website or by obtaining a copy from the IASP of Pain, Clinical Updates, “Lessons from my Central Pain” Volume X, No. 3 Sept 2002. Requests should go to:

909 NE 43rd St. Suite 306
Seattle Washington
Tel: 206-547-1703

Other excellent publications on pain are available through IASP.