THE ANATOMIC BACKGROUND OF LOW–BACK PAIN
S. ROBERT
S. RAOUL
0. HAMEL
S. PLOTEAU
Neurotraumatology Department, Anatomy Laboratory, Nantes Teaching Hospital
Introduction
For any surgeon of the spine, low–back pain is a day–to–day concern. One of the most common causes for consultation, low–back pain can become chronic and often proves intractable to medical or surgical treatment. A study of its anatomic background should give us a clearer understanding of the mediation involved and thereby enable us to better engineer the treatment to overcome it.
General comments on pain
For pain to occur there must
be first and foremost a transfer from peripheral receptors right
through to the central nervous system.
There are two main types of receptors:
- nociceptors that are essentially
pain carriers and are unequally distributed throughout the body.
They are found in particularly large numbers in the dental pulp,
the cornea, the lachrymal bone bed, where mere touch is painful.
- polymodal receptors that take
their name from the fact that they are only algesiogenic when
the excitation threshold is crossed. The so–called silent afferents
originating in the viscera no doubt belong to this category.
- Fibres: pain mediating fibres
are small calibre, poorly myelinated or unmyelinated (Ad, C).
They follow the posterior roots and, irrespective of where they
originate in the three embryonic layers, are located in the ventral
part of the posterior root before penetrating the grey matter
of the posterior horn. These fibres are poor conductors with
a high excitation threshold.
To systematise the posterior grey horn of the spinal cord, three zones can be identified:
- Apex
- Isthmus
- Base
The apex is exteroceptive in kind. Pain projected there has an ectodermal origin.
The isthmus is proprioceptive in kind. Pain projected there has a mesodermal origin.
The base is interoceptive in kind. Pain projected there has a visceral origin.
 |
DRAWING A = Apex 1 = fibres of exteroceptive origin (skin, integuments) 2 = fibres of proprioceptive origin (muscles, tendons) 3 = fibres of interoceptive origin (viscera) |
|
Figure 1 Pain: Penetration of nociceptive fibres deriving from the 3 embryonic layers |
|
The progression of the pain tracts
spinothalamic tract: this is the deutoneuron of the pain tracts. It originates from the posterior grey hom, where the neurons that compose it decussate on entering to form the spinothalamic tract (Dejerine's crescent). This pathway, that in the classical concept is divided into a ventral spinothalamic (neo–spinothalamic) tract and a dorsal spinothalamic (paleo–spinothalamic) tract, also contains spinomesencephalic fibres that transmit essentially interoceptive pain. In this classical concept the ventral part is associated with coarse touch and the dorsal part is associated with temperature and pain. However, these notions must be revised in the light of more recent data. In fact, the sensory roots originating from the three ectoblastic, mesoblastic and entroblastic layers and terminating respectively in the apex, isthmus and base of the posterior horn, converge towards the neurons in lamina V of REXED's laminae to form the spinothalamic tract. All these fibres are potentially algesiogenic but we shall see later that they have a specific role in pain transmission that needs to be explained.
|
DRAWING 1 = Exteroceptive fibres |
 |
|
Figure 2 Pain: constitution of the tracts |
Pain–fighter path: the fibres penetrating the posterior horn at once encounter the collaterals of the thick A alpha fibres forming the fasciculus gracilis of GOLL and the fasciculus cuneatus of BURDACH. The collaterals of these fibres that transmit epicritic touch and conscious proprioceptive sensibility inhibit the transfer of pain. These thick fibres and their collaterals are high–speed conductors of nervous impulse and have a low excitation threshold. They are consequently well endowed to compete advantageously with the miserable C and Ad pain fibres. This theory goes by the name of GATE CONTROL. Described by WALL and MELZACH, it provides a good explanation of behaviour under pain conditions, in particular that of therapy against pain. Once "through the gate" pain has not finished its route. Several ascending tracts are available to reach the brain stem:
 |
DRAWING 1 Nociceptive exteroceptive fibres |
|
Figure 3 Gate Control (Wall & Melzach) |
The ventral spinothalamic or neo–thalamic tract hangs onto the deutoneurons of the fasciculus gracilis and the fasciculus cuneatus that we noted do not transmit pain. So, the deutoneuron or lemniscus tracts are formed, creating a real sensibility highway. Just as a cyclist can hang onto a motorcycle, the ventral spinothalamic tract takes advantage of the ribbon of REIL (lemniscus medialis) and gets a fast lift up to the thalamus. We shall see later what happens to these neo–spinothalamic fibres.
Not quite so fortunate, two possibilities await the fibres of the dorsal spinothalamic tract: they can plod on, uneventfully, all the way to the thalamus, or they impinge on the reticulated formation and at the expense of multiple synapses and numerous inhibition zones will eventually reach the thalamus, setting off, on the way, the anti–pain batteries of the reticulospinal tracts, that in particular reinforce Gate Control and exert supra–spinal control of pain.
|
DRAWING 1. Thick A fibres |
|
|
Figure 4 Schematic of the brain stem |
thalamus: the majority of impulses converge at the lateral part of the ventroposterior thalamic nucleus. We should also note that pain mediated by the reticular fon–nation and visceral sensibility are projected in the median and intralaminar nuclei. In any case, the thalamus acts as a filter that most of the time will spare the cerebral cortex, that is, consciousness, from having to manage pain since it is taken over automatically by the thalamus, in particular by motor rearrangement. When the pain threshold is too high, the thalamus opens the gates and pain is then transmitted to the cortex.
pain projection in the older cortex and in the hypothalamus: the hypothalamus and the archaeocortex are functionally associated and anatomically close. Emotion and memory use the same anatomic paths that involve both structures. Memory and emotion are an integral part of the history of pain. When pain reaches the temporal cortex and more precisely the hippocampus, it travels backwards along the path of the fornix, penetrates the mammary process and synapses there. From there, the mamillothalamic tract (bundle of VICQ D'AZYR) plunges into the anterior nucleus of the thalamus. A new synapse draws the information into the cingulate or peri–callosal cortex. In this way the memory and emotion circuit is formed, commonly called the multi–neuronal circuit of PAPEZ (hippocampus, mamillary process, thalamus, cingulate cortex, and back to hippocampus.
 |
DRAWING 1. Mamillary process |
|
Figure 5 Pain: Multi–neuronal circuit of PAPEZ Memory–Emotion |
|
What does this anatomic scheme tell us?
When there is emotion there is memory retention, whether the emotion is active or passive. To experience pain, to remember it, is a means of controlling pain should it reoccur. So pain is worth remembering. This memory storage that is probably multi–cortical, is fundamental in the day–to–day experience of pain. Emotion accompanies pain and we need only take the example of the tears of pain that the child does not try to hide, to readily understand their pathway in anatomic terms. From the fomical and mamillo–hippocampal afferents, there is projection on the septal nuclei that are the effector site of the limbic system. From there, by complex paths through either the mammary process or the habenular ganglion, the projections reach the reticular formations of the brain stem and the spinal cord, but also the nuclei of the cranial nerves. Thus, to give only one example, the lacrymo–muco–nasal nucleus described by YAGITA is excited in this way by pain stimuli. It secretes tears that accompany pain. Another example is that of the motor nucleus of the bronchial arch of cranial nerve VII that transforms a serene expression into a sardonic mask via the facial muscles.
projection on the neo–cortex: this occurs in two ways : fibres from the neo–spinothalamic tract project into the postcentral gyrus. In this region the image of a little man is projected part for part on the cortex – the homunculus described by PENFIELD. The upper limb and face are lateral, the lower limb medial, in relation to the hemisphere. This neo–spinothalamic tract that we saw took advantage of the lemniscus medialis to accompany it in good time to the thalamus, provides us with somatotopic information on pain which is the first information received on the occurrence of a pain stimulus. The other fibres are projected essentially in the prefrontal cortex, the cortex of feelings, that will provide a qualitative evaluation of pain. This qualitative evaluation is always preceded in time by the topographical evaluation.
The specificity of low–back pain
What are the components of pain in the lumbar region?
The work of KUSLICH is very valuable in this respect since, under local anaesthesia, surgery on well–informed and well–prepared patients enables us to test the precise painful sites responsible for low–back pain. Practically only the posterior and surface region of the annulus fibrosus and the posterior longitudinal ligament are algesiogenic to stimuli. We can note that the articular facets, the bony structures, the muscular and aponeurotic structures are only slightly algesiogenic.
Innervation of the vertebrae and the disc–corpus–ligament complexes
Somatic innervation: each spinal root of the somatic system has a motor ventral branch and a sensory dorsal branch. The sensory dorsal branch that concerns us has certain particularities at levels L4 and LS since it does not reach the cutaneous area. The roots higher up and more especially L2 take over sensory innervation of the low–back region. Other innervation "blanks" exist: the cutaneous region between the shoulder blades is taken over by the cervical roots. Pain between the shoulder blades is a common feature of cervical spine pathology. We also need to consider that pain projected in the lumbar region is not merely somatic. Somatic pain generated through a sensory root is metarneric and consequently limited to one root path. Low–back pain is diffuse and so does not strictly conforin to these characteristics.
Ortho–sympathetic innervation: only ortho–sympathetic fibres have sensory structures at vegetative system level. The vegetative metamere is organised as follows: with respect to the sensory cells, they leave from the target organ, occasionally pass into the pre–visceral ganglion, follow the latero–vertebral or catenary chain (via the vessels and mediated by the splanchnic nerves), that is a compulsory relay. From there, and via the rami communicantes, the afferent fibres of the ortho–sympathetic system reach a nearby somatic nerve and along with the sensory fibres of this nerve transmit the information at the level of the base of the posterior hornm, as we mentioned previously. This explains the metameric character of ortho–sympathetic pain that uses one or several sensory roots in the neighborhood of the painful region to indicate its topography. This information, even if lacking in precision, gives at least an indication of the position of the painful organ.
Low–back pain, a diffuse, burning pain in bands, commonly bilateral, is a visceral type pain. It can be compared to the spreading chest pain of myocardial infarct, the pain pervading the right iliac fossa in appendicitis, the pain pervading the right hypochondriac zone in biliary colic.
Ortho–sympathetic innervation of the lumber region has been very well described. The fibres of the ortho–sympathetic system (initially, for the motor fibres and near the end of their travel for the sensory fibres) follow the path of the somatic fibres at the level of the latero–vertebral ganglia via the rami communicantes in the latero–vertebral ganglia chain. Proximal neutrons take over motor and sensory vegetative innervation of the para–vertebrat viscera. VON LUSCHKA, as far back as the XlXth century, described the recurrent nerves that he called the sinu–vertebral nerve. Penetrating the intervertebral foramen latero–medially, they fan out on the dorsal face of the annulus fibrosus and in the two layers of the posterior longitudinal ligainent, as shown in more recent studies; They also innervate the meninge, especially its ventral and lateral part. These LUSCHKA fibres reach the lateral vertebral ganglia that converge towards the ortho–sympathetic centres and terminate classically at level L2.
 |
DRAWING 1. Latero–vertebral ganglia chain |
|
Figure 6 Sympathetic system and somatic system projections at L2 level |
|
Consequently, it is this last root that will be the privileged projection site of the viscero–sensory fibres originating from the lumbar region. Considering that L2 also takes over the somatic cutaneous components of lumber region L4, L5, the logic of infiltrating at level L2 instead of the lower metameres, in the medical management of low–back pain, is intellectually satisfying. Such a test of infiltrations has been carried out with success, as reported, for example, by Japanese authors.