Emeritus Professor Nikolai Bogduk

The aim of this study was to determine the subdural haematoma recurrence rate in patients with symptomatic chronic subdural haematomas (CSDH) treated with either burr hole trephination (BHT) or minicraniotomy (MC) and to determine whether a statistically significant difference exists between the two techniques. A retrospective analysis of all consecutive patients with symptomatic CSDH treated with BHT or MC at the John Hunter Hospital Neurosurgery Department between July 2004 and July 2016 was performed. After inclusion/exclusion criteria were applied, 368 patients with 439 CSDHs were eligible. Baseline demographic data was recorded for all patients. Statistical analysis was performed assessing haematoma recurrence as the primary outcome and mortality as a secondary outcome. Three hundred and sixty eight patients were included in the study, with 225 being treated with MC and 143 with BHT. Baseline demographic data was similar between the two groups. The recurrence rate for patients treated with BHT was 0.13 (95% CI 0.08¿0.18), versus 0.18 (95% CI 0.13¿0.23) in the MC group. This difference was not statistically significant. Similarly, there was no statistically significant difference in mortality rates between the two groups. The mortality rate in the BHT group was 0.09 (95% CI 0.05¿0.14) versus 0.09 (95% CI 0.05¿0.13) in the MC group. In our series there was no difference in recurrence rates or mortality rates between the two groups, suggesting MC is an effective alternative to BHT in the management of symptomatic CSDH.

An audit was conducted to determine if epidural injections for back pain and sciatica were practised in accordance with guidelines prescribed by the National Health and Medical Research Council (NHMRC). More than 80% of injections were used for conditions for which they were not indicated, and which the NHMRC considers experimental. Most commonly, epidural injections afforded no benefit to the patients, yet were often repeated. In no instance was informed consent recorded as recommended by the NHMRC. The practice of epidural injections is dissonant both with the evidence-based literature and with recommended guidelines, and squanders both financial and professional resources.

This article carefully itemizes the various anatomic structures that can evoke neck pain, putting in perspective what clinicians know, what they assume, and what they need to understand better about neck pain and pain referred from the neck. The critique of many of the accepted entities in the differential diagnosis of neck pain is crucial to an understanding of the causes of neck pain and an ability to implement appropriate therapies.

Objectives: To describe recent research which elucidates the role of the cervical synovial joints as sources of pain in patients with post-traumatic headache. Findings: Recent research has demonstrated that the cervical synovial joints are innervated and that they can be potent sources of neck pain and headache if stimulated in normal subjects or if injured in patients. Postmortem studies of injured joints have revealed chondral and subchondral fractures, braising/damage of the intra-articular inclusions, hemarthroses and capsular tears or avulsions. These acute lesions constitute the substrate for the development of post-traumatic arthritis and, consequently, chronic post-traumatic headache. Such lesions evade detection by conventional diagnostic techniques. However, the advent of innovative diagnostic approaches, including manipulative assessment, biomechanical analysis and controlled diagnostic injection techniques, has allowed the identification of painful cervical synovial joints in vivo. Using these diagnostic techniques, epidemiological studies have shown that the cervical zygapophysial joints [in particular C2-3] are common sources of post-traumatic headache. Conclusions: These findings support the hypothesis that cervical synovial joint pain is a real and common clinical entity. Controlled diagnostic blocks are the only reliable means whereby this condition can be identified. Controlled blocks provide a criterion standard against which other diagnostic techniques could be calibrated in the future.

The instantaneous centre of rotation has proven to be a useful parameter of vertebral motion. The normal location of instantaneous centres has been determined in cadavers and in normal volunteers for the cervical, thoracic and lumbar spines, and abnormal location of centres has been shown to correlate with spinal pain. However, to date, an instantaneous centre has constituted no more than a convenient mathematical summary of vertebral kinematics. It has defied resolution into biologically meaningful parameters. This study offers a novel model of vertebral motion in which the instantaneous centre of rotation can be shown to be a function of the location of the centre of reaction of a vertebra, and the intrinsic rotation and translation it undergoes. These parameters are strictly linked by equations that determine the location of an axis of rotation. These equations allow aberrations in the location of an axis to be interpreted in terms of the anatomical and pathological factors that affect the centre of reaction of the vertebra and the rotation and translation it undergoes. © 1995, Institution of Mechanical Engineers. All rights reserved.

One hundred seventy-six consecutive patients with chronic low-back pain and no history of previous lumbar surgery were studied to test the clinical criteria of Fairbank et al. and Helbig and Lee for zygapophysial joint pain. All patients underwent a history, examination, and a series of zygapophysial joint injections or blocks of the medial branches of the dorsal ramus with lignocaine. Those patients responding to the first series of blocks were given confirmatory blocks using bupivacaine. None of the clinical features tested was found to be associated with response to the confirmatory block. The Fairbank et al. and Helbig and Lee criteria were shown to be unreliable in distinguishing pain of zygapophysial joint origin from pain of other origins. © 1994 Raven Press, Ltd.

A consecutive series of 100 patients was studied to determine the prevalence of third occipital nerve headache in patients with chronic neck pain (> three months in duration) after whiplash. Seventy one patients complained of headache associated with their neck pain. Headache was the dominant complaint of 40 patients, but was only a secondary problem for the other 31. Each patient with headache underwent double blind, controlled diagnostic blocks of the third occipital nerve. On two separate occasions the nerve was blocked with either lignocaine or bupivacaine, in random order. The diagnosis of third occipital nerve headache was made only if both blocks completely relieved the patient's upper neck pain and headache and the relief lasted longer with bupivacaine. The prevalence of third occipital nerve headache among all 100 whiplash patients was 27% (95% confidence interval (95% CI) 18-36%) and among those with dominant headache the prevalence was as high as 53% (95% CI 37-68%). There were no distinguishing features on history or examination that enabled a definitive diagnosis to be made before the nerve blocks. Those patients with a positive diagnosis, however, were significantly more likely to be tender over the C2-3 zygapophysial joint (p = 0-01). Third occipital nerve headache is a common condition in patients with chronic neck pain and headache after whiplash. Third occipital nerve blocks are essential to make this diagnosis.

Whiplash is a poorly understood problem that attracts accusations of malingering and compensation neurosis. Recent research has revealed a variety of occult lesions that can be responsible for the chronic pain and suffering after whiplash; however, appropriate diagnostic techniques are still either lacking or not widely used. While there are reasonable options for acute management there is no proven therapy for the chronic situation.

Lumbar diskography is not a test for disk herniation. It is a test for intrinsically painful, externally intact disks and so does not compete with computed tomography as a diagnostic instrument. Formal studies have shown that the lumbar disks are innervated and can be a source of pain that has pathomorphologic correlates. The biological basis for lumbar diskography, therefore, resists criticism. Cervical diskography, however, lacks the same biological basis and remains open to criticism. Utility is the weak point of both lumbar and cervical diskography. Although both provide diagnostic information, it remains to be shown that this information makes a difference to outcome when treatment is direct to the ostensibly painful disk. © 1994 American Pain Society.

Background and Objectives. To determine the specificity of cervical medial branch blocks for the diagnosis of cervical zygapophyseal joint pain by ascertaining the disposition of the local anesthetic after injection of the medial branches of the cervical dorsal rami. Methods. Sixteen consecutive patients with chronic neck pain from motor vehicle accidents underwent cervical medial branch blocks. A 22- gauge, 90-mm spinal needle was placed onto the target nerve under image-intensifier guidance. Immediately after each target nerve had been infiltrated with 0.5 ml of local anesthetic, 0.5 ml of contrast medium was injected to map the spread of injectate. Radiographs were recorded to document the pattern of spread. Results. Twenty-five injections of local anesthetic and contrast medium were performed. Contrast medium dispersed in characteristic patterns at all vertebral levels and always incorporated at least 5 mm of the perceived course of the target nerve. There was never any spread to the ventral ramus, beyond the medial fibers of semispinalis capitis or to the adjacent medial branches. No other single structure was consistently within the field of the contrast. Eleven patients obtained complete or definite relief of their pain, which could only be attributed to anesthetization of the zygapophyseal joint innervated by the nerves blocked. Conclusions. Local anesthetic blocks of the cervical medial branches are a specific test for the diagnosis of cervical zygapophyseal joint pain. The local anesthetic always reaches the target nerve and does not affect any other diagnostically important structures. © 1993 American Society of Regional Anesthesia and Pain Medicine.

A null hypothesis was tested to the effect that cervical zygapophyseal joint pain was uncommon. The prospectively recorded responses were reviewed of 313 consecutive- patients with ntfactable neek pa-lr. who underwent provocation discography anc cervical zygapophyseal joint blocks* Symptomatic zygapophyseal joints were encountered in 75% rtf the sample with a possibility that a further 33% suffered zygapophyssal pain but were not appropriately Investigated. The null hypothesis was rejected. Cervical zygapophyseal joint pain is not rare, and is worthy of further consideration not just in research but in clinical practice. © Lippincott-Raven Publishers.

The reliability of a modified protocol for plotting the instantaneous axes of sagittal rotation for the cervical spine was evaluated by measuring the observer differences when the process was performed separately by two observers, and by a single observer, on two separate occasions. Small observer differences were found both for individual steps in the process and for the process as a whole. These differences were substantially less than those found using the conventional technique for plotting the instantaneous axes of rotation. The improvement in the technique was resultant from the use of stricter criteria for recognizing and tracing vertebral landmarks. © 1991.

The reliability of the conventional technique for determining the instantaneous axes of rotation of the cervical motion segments was evaluated by comparing the determinations made by two observers and the same observer on two separate occasions. Different observers found the same mean location for the instantaneous axis at each segment in a test population of 17 normal subjects, but for any given subject the interobserver and intraobserver differences were large, with a relative variation of up to 35%. Thus, while reliable for determining the mean location of the instantaneous axis of rotation in a population, the technique was not reliable for determining the axis in a given subject. Qualitative analysis suggested that the identification of the radiographic images of the vertebrae and their tracing is the greatest source of error for this technique. © 1991.

The lumbar disc serves to sustain compression loads and is subject to tension and shear in forward bending and rotation. Its outer third is innervated and can be a source of pain. The annulus fibrosus may be injured in rotation and flexion of the lumbar spine and may become symptomatic as a ligamentous injury. Compression injuries of the disc are initially asymptomatic but may set in train a degradative process that, in time, leads to internal disc disruption, which becomes symptomatic as a result of chemical or mechanical irritation of nociceptors in the annulus fibrosus. Disc prolapse is but one possible end stage of internal disc disruption and represents the culmination of a series of destructive processes affecting the disc. This condition can be symptomatic while the external appearance of the disc remains normal and before nerve roots are affected in any way.

This article summarises the cardinal concepts of biomechanics relevant to the understanding of the movements of the lumbar spine. It explains in simple terms the definition of movements, stress-strain relationships, stiffness, range of movement, creep and hysteresis and provides the basis for later lectures in this symposium which deal with the biomechanical effects of injury and ageing.

Intervertebral disc herniation is not a common cause of back pain. Rather, the discs can be a source of pain without rupture of herniation. The pathological changes responsible are not those of disc degeneration, as commonly understood, but two specific entities: torsional injury of the annulus fibrosis and compression injuries of the disc. Torsional injuries occur as a result of forceful or excessive rotation combined with flexion and result in circumferential tears in the innervated outer third of the annulus fibrosis. This lesion probably underlies what is commonly interpreted as back pain due to 'ligament strain'. Compression injuries arise as a result of excessive weight-bearing and are initiated by fractures of the vertebral end-plate. Degradation of the nuclear matrix follows, resulting in the condition of internal disc disruption. The nuclear degradation erodes the annulus fibrosis centrifugally and the disc may become symptomatic as a result of chemical and/or mechanical irritation of the nerve endings in the compromised annulus fibrosis. Strong correlations have been established between the reproduction of discogenic pain and the morphology of internal disc disruption as seen on CT discography.

The ability of a manipulative therapist to diagnose symptomatic cervical zygapophysial joint syndromes accurately was evaluated in a series of 20 patients. In 11 patients the presence, or absence, of a symptomatic joint was established by means of radiologically-controlled diagnostic nerve blocks. These patients were assessed by the manipulative therapist, without knowledge of the medical diagnosis. Another nine patients were first seen by the manipulative therapist whose diagnosis was then evaluated by means of diagnostic blocks. The manipulative therapist identified correctly all 15 patients with proven symptomatic zygapophysial joints, and specified correctly the segmental level of the symptomatic joint. None of the five patients with asymptomatic joints was misdiagnosed as having symptomatic zygapophysial joints. Thus, manual diagnosis by a trained manipulative therapist can be as accurate as can radiologically-controlled diagnostic blocks in the diagnosis of cervical zygapophysial syndromes. However, before generalized claims about the reliability of manual diagnosis can be made, further studies of this nature are required to validate intertherapist reliability and the ability of manual techniques to diagnose other spinal pain syndromes.

Lateral radiographs of ten normal Individuals were studied to determine the location of the Instantaneous axis of rotation (IAR) of every lumbar vertebra for the movements, of flexion and extension from the upright position and flexion from the fully extended position; and errors Involved In the technique were quantified to establish confidence limits for the results of the calculations. The distribution of the IARs was found to fall within a small range from the mean location at each level, particularly for the movement of flexion from the extended position. Within-observer and between-observer errors occurred in tracing and superimposing radiographs and marking x and y coordinates. Unacceptably large errors occur when the movement of the joint is less than 5°, and only the IAR for flexion from extension can be plotted with acceptable confidence. This result invalidates the notion that plotting centrodes may be of diagnostic value in recognizing mechanical disorders. The determination of a single extension to flexion IAR may be of more value clinically, to which end this study provides essential normative data. © Lippincott-Raven Publishers.

The lumbar erector spinae consists of two muscles iliocostaiis lumborum and longissimus thoracis each with distinct thoracic and lumbar parts. The thoracic parts consist of tiny muscle bellies with segmental origins from the thorax and long caudal tendons that form the erector spinae aponeurosis. The lumbar fibers arise from the lumbar accessory processes and the L1-4 transverse processes, and insert independently of the erector spinae aponeurosis into the ilium. The intrinsic lumbar fibers of the erector spinae are poorly described in the literature, and the existence of the iliocostaiis lumborum pars lumborum has rarely been recognized even though it constitutes a substantial portion of the total muscle mass acting directly on the lumbar vertebrae. © Lippincott-Raven Publishers.

Prompted by clinical failures of percutaneous radiofrequency neurotomy in the treatment of back pain and neck pain, we performed a study to determine the shape and size of lesions made by radiofrequency electrodes. Experimental lesions were made in egg white and fresh meat at temperatures recommended in clinical practice. The cardinal finding was that lesions do not extend distal to the tip of the electrode. They only extend radially around the electrode tip in the shape of an oblate spheroid, with a maximal effective radius of only 2 mm. Consequently, if electrodes are directed perpendicularly onto a nerve, the nerve may not be encompassed by the lesion generated. Some of the clinical failures of percutaneous medial branch neurotomy ('facet rhizotomy') may be due to this phenomenon. We suggest modified techniques for medial branch neurotomy in which the electrodes are introduced parallel to the target nerve whereupon it is more readily encompassed by the radial spread of the lesion.

The back muscles alone are unable to provide the extensor moment required to lift large weights, and must be aided by another source of anti-flexion moments. It has been postulated that contraction of the abdominal muscles can provide an extension moment by developing tension in the thoracolumbar fascia (TLF). Anatomical studies and a biomechanical analysis, however, reveal that the anti-flexion moment generated in this way is only very small. Too little of the abdominal musculature attaches to the TLF to generate a significant tension in it. Previous calculations of the forces in the TLF have overestimated the tension developed in it because of erroneous assumptions and interpretations of the relevant anatomy. Whatever the role played by the TLF in lifting it must be essentially independent of abdominal mechanisms. © 1987.

One of the putatative causes of headache is osteoarthritis of the C2-3 zygapophysial joint. A technique for blocking the third occipital nerve which innervates this joint was devised and used as a screening procedure for headache mediated by this nerve. Seven out of ten consecutive patients presenting with suspected cervical headache were found to suffer pain mediated by the third occipital nerve and stemming from a C2-3 zygapophysial joint. Because third occipital headache may be indistinguishable clinically from tension or other forms of headache, third occipital nerve blocks are advocated as means of establishing this largely unrecognised diagnosis.

Whereas the symptoms of whiplash injury are frequently poorly understood or misrepresented as due to neurosis, a review of the literature reveals a considerable amount of biomechanical and experimental data that substantiate a diverse organic basis for these symptoms, including disorders commonly not considered or sought for in the investigation of whiplash. Moreover, formal studies fail to substantiate the notion of 'litigation neurosis' being the cause of prolonged symptoms. This review synthesises the anatomy and pathology of whiplash, and the pathophysiology of symptoms, into what can be construed as the organic basis for whiplash. © 1986.

The possible actions of the lumbar multifidus were determined by plotting the points of attachment and orientation of each of its component fascicles on radiographs of 5 cadavers and 21 living subjects. Subsequent analysis revealed that the principal action of multifidus is posterior sagittal rotation (extension without posterior translation) of the lumbar vertebrae. It has no translatory action. Any axial rotation exerted by the lumbar multifidus is only a minor, secondary action which must be coupled with posterior sagittal rotation. This extension balances the flexion moment generated by the abdominal muscles which rotate the trunk. The constancy of the sites of attachment of the multifidus allows each of its fascicles to be plotted accurately on radiographs or computer diagrams which can be used to produce highly detailed analyses or models of the forces exerted by the multifidus on the lumbar spine. © 1986.

Dissection studies revealed that the fibres of the lumbar multifidus are divided by distinct cleavage planes into five bands. Each band arises from a lumbar spinous process, and is innervated unisegmentally. The lumbar multifidus is therefore composed of five myotomes arranged such that the fibres that move a particular segment are innervated by the nerve of that segment. Target points are described that enable electromyography to be performed on paraspinal muscles of known unisegmental innervation. © 1986.

The posterior elements of the vertebral column are innervated by branches of the dorsal rami of the spinal nerves, while the intervertebral discs and related ligaments are innervated by various branches of the ventral rami and sympathetic nervous system. A knowledge of this nerve supply forms the basis for a systematic classification of the possible sources of primary spinal pain, and the basis for several diagnostic techniques that use needles to provoke and anaesthetize putative sources of pain. In particular, the demonstration of a nerve supply to intervertebral discs vindicates the concept that these structures intrinsically may be sources of pain. © 1985, Australian Physiotherapy Association. All rights reserved.

As an explanation for the pathological basis for acute locked back, meniscus entrapment is a concept that is inconsistent with both the anatomy of zygapophysial menisci and the clinical features of acute locked back. Two alternative explanations that better fit the clinical features are postulated: meniscus extrapment and intradiscal nuclear displacement, both of which are amenable to manipulative therapy.

As an explanation for the pathological basis for acute locked back, meniscus entrapment is a concept that is inconsistent with both the anatomy of zygapophysial menisci and the clinical features of acute locked back. Two alternative explanations that better fit the clinical features are postulated: meniscus extrapment and intradiscal nuclear displacement, both of which are amenable to manipulative therapy.

We have examined the responsiveness of the vertebro-basilar circulation of the anesthetized Macaca nemestrina monkey to vasoactive agents infused directly into the artery. Infusion of noradrenaline caused a slight increase in vertebral arterial resistance. This constriction was less than that seen in previous experiments with either the internal or external carotid arteries. In the presence of vasodilatation caused by inhalation of a CO2-rich gas mixture, this constriction became a dilatation. Serotonin was without significant effect on the vertebral arterial bed. Bradykinin, histamine and prostaglandin E1 all produced slight dilatation, with bradykinin being the most potent. In all cases the concentration required to produce an effect on the vasculature was much greater in the vertebral circulation than it is in the internal carotid and extracerebral circulations. We conclude that the intact vertebro-basilar circulation is much less sensitive to vasoactive agents than experiments with isolated segments of these arteries would indicate and that therefore these agents are unlikely to play a significant part in the pathogenesis of vertebro-basilar migraine. © 1984.

Stimulation of the trigeminal nerve or ganglion in the cat caused a frequency-dependent reduction in carotid vascular resistance. Systemic arterial blood pressure (SABP) decreased at low frequencies (0.2 to 5 sec-1) and increased at higher frequencies, thus increasing carotid blood flow at the higher frequencies. The effect on resistance was predominantly ipsilateral and was unaltered by cervical sympathectomy, but was abolished or substantially reduced by section of the trigeminal root proximal to the ganglion. Diminution of carotid vascular resistance was replicated by stimulation of the greater superficial petrosal (GSP) nerve without any change in SABP. Section of the 7th cranial nerve reduced or abolished the response to stimulation of the trigeminal nerve but not that from the GSP nerve. The trigeminal response was prevented by ganglion-blocking drugs in 7 out of 8 cats. The resistance response was unaffected by noradrenergic, cholinergic, serotonergic, and histamine-2 blocking agents. No neural connection could be demonstrated between the GSP and the trigeminal ganglion, and the vascular response to GSP stimulation persisted after trigeminal section. It is concluded that activation of the trigeminal system increases carotid blood flow by a pathway involving the 7th cranial nerve, the GSP and Vidian nerves, and a parasympathetic synapse employing an unconventional transmitter. A varying proportion of the response (greatest in the 3rd division) may be mediated by antidromic activation of trigeminal nerves. These findings may have clinical implications for the vascular changes of migraine and other facial pain.

Any structure innervated by the cervical spinal nerves can be a primary source of neck pain and referred pain to the head, upper limb or chest wall. The most likely sources of chronic pain are the cervical intervertebral discs and the cervical zygapophyseal joints. Disc pain can be diagnosed by provocative disc stimulation; zygapophyseal joint pain can be diagnosed by local anaesthetic blocks of the cervical dorsal rami. Disc pain can be treated by anterior cervical fusion and zygapophyseal pain by percutaneous radiofrequency neurotomy.

The literature describing menisci In the lumbar zyga-pophyseal joints Is reviewed. The only true menisci In these joints are rudimentary fibrous invaginations of the dorsal and ventral capsule. So-called menisci at the superior and inferior poles of the joint are basically fat- filled synovial reflections, some of which contain dense fibrous tissue, which probably arises as a result of mechanical stress. The theory of meniscus entrapment is appraised, but is considered to have been an overstated cause of those forms of ¿acute locked back¿ that responds to manipulation. © 1984 Lippincott-Raven Publishers.

The thoracolumbar fascia was studied by dissection in ten adult human cadavers. The posterior layer of this fascia was found to consist of two laminae. The superficial lamina is formed by the aponeurosis of latissimus dorsi. The deep lamina consists of bands of fibers passing caudolaterally from the midline. Both laminae form a retinaculum over the back muscles, and the deep lamina constitutes a series of accessory posterior ligaments that anchor the L2 to L5 spinous processes to the ilium and resist flexion of the lumbar spine. The function of these ligaments Is enhanced by the contraction of the back muscles and the action of certain, restricted portions of the abdominal muscles. © Lippincott-Raven Publishers.

The lumbar intervertebral discs are innervated posteriorly by the sinuvertebral nerves, but laterally by branches of the ventral rami and grey rami communicantes. The posterior longitudinal ligament is innervated by the sinuvertebral nerves and the anterior longitudinal ligament by branches of the grey rami. Lateral and intermediate branches of the lumbar dorsal rami supply the iliocostalis lumborum and longissimus thoracis, respectively. Medial branches supply the multifidus, intertransversarii mediales, interspinales, interspinous ligament, and the lumbar zygapophysial joints. The distribution of the intrinsic nerves of the lumbar vertebral column systematically identifies those structures that are potential sources of primary low-back pain. © 1983 Harper and Row, Publishers, Inc.

The L 1-4 dorsal rami tend to form three branches, medial, lateral, and intermediate, which are distributed, respectively, to multifidus, iliocostalis, and longissimus. The intertransversarii mediales are innervated by a branch of the dorsal ramus near the origin of the medial branch. The L 4 dorsal ramus regularly forms three branches while at the L 1-3 levels the lateral and intermediate branches may, alternatively, arise from a short common stem. The L5 dorsal ramus is much longer than the others and forms only a medial and intermediate branch. Each lumbar medial branch innervates two adjacent zygapophysial joints and ramifies in multifidus, supplying only those fascicles which arise from the spinous process with the same segmental number as the nerve. The comparative anatomy of the lumbar dorsal rami is discussed and the applied anatomy with respect to 'rhizolysis', facet denervation' and diagnostic paraspinal electromyography is described.

In a study of 82 lumbar zygapophysial joints three types of intra-articular structures were identified. They were adipose tissue pads and fibro-adipose meniscoids, both located at the superior and inferior poles of the joint, and connective tissue rims, located along the dorsal and ventral margins. Every lumbar zygapophysial joint contained at least one of these structures and 47 contained more than one type. Connective tissue rims are short, central projections of the joint capsule and do not enter between the articular surfaces. Adipose tissue pads are covered by synovium and fill the subcapsular space at the superoventral and inferodorsal poles of the joint. Fibro-adipose meniscoids, also covered by synovium, project from the joint capsule at the superior and inferior poles and enter between the articular surfaces. Adipose tissue pads and fibro-adipose meniscoids are probably derived from a common primitive mesenchymal meniscus which primarily differentiates into a fatty structure. The fibrous component of fibro-adipose meniscoids then secondarily develops as a result of compression of the tip of the fatty structure between the articular surfaces. The function of these intra-articular structures is not evident but may be related to protection of the articular processes as they subluxate during flexion and extension. Meniscus entrapment in the lumbar zygapophysial joints has been proposed as a cause of acute locked back, but the present morphological data are inconsistent with this view.

A detailed description of the anatomy of the cervical dorsal rami is provided on the basis of the past literature and independent studies on five cadavers. In particular, the anatomy of the upper cervical dorsal rami and the innervation of the cervical zygapophyseal joints are described. The clinical significance of the cervical dorsal rami is discussed in relation to headache, occipital neuralgia, and neck pain. The surgical anatomy of cervical facet denervation is discussed. The radiologic anatomy of the medial branches of the cervical dorsal rami is described, and target points suitable for diagnostic blocks or facet denervation are illustrated. © 1982 Harper & Row, Publishers, Inc.

The lumbar intervertebral discs are supplied by a variety of nerves. The posterior aspects of the discs and the posterior longitudinal ligament are innervated by the sinuvertebral nerves. The posterolateral aspects of the discs receive branches from adjacent ventral primary rami and from the grey rami communicantes near their junction with the ventral primary rami. The lateral aspects of the discs receive other branches from the rami communicantes. Some rami communicantes cross intervertebral discs and are embedded in the connective tissue of the disc deep to the origin of psoas. Such paradiscal rami are likely to be another source of innervation to the discs. The anterior longitudinal ligament is innervated by current branches of rami communicantes.

The C2 nerve roots and rami were dissected in 5 cadavers to explore the pathogenesis of Neck-Tongue Syndrome. The most likely cause of the simultaneous occurrence of suboccipital pain and ipsilateral numbness of the tongue is an abnormal subluxation of one lateral atlanto-axial joint with impaction of the C2 ventral ramus against the subluxated articular processes.

Rhizolysis has been regarded as a controversial, but safe, procedure for the treatment of back pain. A case of a neuralgia-like pain syndrome resulting from transection of the lateral branches of the lumbar dorsal rami during rhizolysis is reported. The anatomy of the lumbar lateral branches is described, and their vulnerability is discussed. Attention is drawn to the likelihood of this complication with the use of rhizolysis in the vicinity of cutaneous branches of the dorsal rami.

The anatomy of the vertebral nerve was investigated in humans and in monkeys. The effect of stimulation of the vertebral nerve and the cervical sympathetic trunk in the monkey was studied. The vertebral nerves in man and monkey represent a series of deep grey rami communicantes which form intersegmental neural arcades around the vertebral artery between C7 and C3. Above C3 the vertebral artery is accompanied by direct branches from the C1¿3 ventral rami. Electrical stimulation of either the vertebral nerve or the cervical sympathetic trunk had a minimal effect on vertebral blood flow. In contrast, sympathetic stimulation had pronounced effects on carotid flow and resistance. Anatomically and physiologically there are no grounds to support the hypothesis that irritation of the ¿vertebral nerve¿ is the pathogenetic mechanism of cervical migraine. © 1981, International Headache Society. All rights reserved.

Dissections of five human adult cadavers revealed that the C2 spinal ganglion bears a constant relationship to the dorsal aspect of the lateral atlanto-axial joint. Radiologically, the ganglion lies extradurally opposite the midpoint of the silhouette of the lateral atlanto-axial joint space. Needles can be introduced onto this target point using fluoroscopic control and used to perform selective local anesthetic blocks of the C2 spinal nerve. This technique is applicable in cases where it is difficult to decide on clinical grounds whether occipital headaches are due to an upper cervical abnormality or are a symptom of tension headache or common migraine. In particular the technique anesthetizes the otherwise inaccessible articular branches of the median and lateral atlanto-axial joints which may be an occult source of headache. © 1981, International Headache Society. All rights reserved.

The anatomy of the lumbar mamillo-accessory ligament (MAL) was studied by gross dissection in six cadavers. The MAL bridges the mamillary and accessory processes of each lumbar vertebra and encloses the medial branch of the dorsal ramus in an osseofibrous tunnel. The tunnel maintains the proximal course of the medial branch in a constant relationship to bone. This constancy allows for accurate percutaneous techniques to stimulate, anesthetize or destroy the medial branch. The MAL morphologically appears to represent remnants of transversospinal elements in the lumbar region, and is ossified in over 10% of lower lumbar vertebrae. Ossification may interfere with some percutaneous denervation techniques. The MAL may be a site of entrapment of the medial branch and may be a source of low-back pain. ©1981 Harper & Row, Publishers, Inc.

Observations made in the present study of the dorsal lumbar muscles of the cat are at variance with descriptions in the available literature. The morphology of these muscles is described in detail, and a revised interpretation of this morphology is made. There are 5 muscles - multifidus, intertransversarii mediales, lumbococcygeus, iliocostalis lumborum and longissimus lumborum. They are arranged in 5 parallel polysegmental columns and are covered by the dorsal layer of thoracolumbar fascia and the erector spinae aponeurosis. The multifidus consists of fibres connecting mamillary and spinous processes. From each mamillary process 4 sets of fasciculi arise. Each set has constant specific attachments. The intertransversarii mediales are fibres connecting accessory and mamillary process. 3 principal fasciculi arise from each accessory process and insert into particular mamillary processes at more caudal levels. The ilicostalis and longissimus arise from the ilium and from opposite surfaces of an intermuscular septum. The lumbococcygeus arises in the lumbar region from the accessory processes but inserts in the tail. Reasons for the revised nomenclature and interpretation are discussed. A specific definition of the longissimus lumborum and iliocostalis lumborum is made on the basis of their observed morphology and nerve supply. It is argued that the lumbococcygeus should be considered as the lumbar portion of sacrocaudalis dorsalis lateralis rather than part of a common muscle mass.

In the lumbar region the longissimus thoracis and iliocostalis lumborum are separated by the erector spinae aponeurosis and its ventral reflection - the lumbar intermuscular aponeurosis. Lumbar fibres of the longissimus arise from the ilium and the lumbar intermuscular aponeurosis and insert into the accessory processes and proximal ends of the transverse processes of the lumbar vertebrae. Lumbar fibres of iliocostalis insert into the costal elements of the first four lumbar vertebrae. The lumbar insertions of these muscles are homologous to their thoracic insertions. The lumbar intermuscular aponeurosis is homologous to the lumbar intermuscular septum in the dog, cat and monkey. The details of attachment of the lumbar fibres of the erector spinae and of the lumbar intermuscular aponeurosis should be taken into account in biomechanical analyses of the lumbar vertebral column.

Low back pain, referred pain in the lower limbs, and spasm of the back, gluteal, and hamstring muscles are clinical features which can be induced in normal volunteers by stimulating structures which are innervated by the lumbar dorsal rami. Conversely, they can be relieved in certain patients by selective interruption of conduction along dorsal rami. These facts permit the definition of a lumbar dorsal ramus syndrome, which can be distinguished from the intervertebral disc syndrome and other forms of low back pain. The distinguishing feature is that, in lumbar dorsal ramus syndrome, all the clinical features are exclusively mediated by dorsal rami and do not arise from nerve-root compression. The pathophysiology, pathology, and treatment of this syndrome are described. Recognition of this syndrome and its treatment with relatively minor procedures, can obviate the need for major surgery which might otherwise be undertaken.

Percutaneous lumbar medial branch neurotomy is a technique for facet denervation in which the target is specifically the medial branch of the dorsal ramus. The radiology of the technique is illustrated, and the technical aspects of the procedure are described. The accuracy of previous techniques for facet denervation as compared with medial branch neurotomy is reviewed in a comparative analysis of radiographs illustrating the various techniques. It is suggested that the greater accuracy of medial branch neurotomy will permit a more adequate trial of the rationale and efficacy of facet denervation. © 1980 Harper and Row, Publishers, Inc.

Dissections of the dorsal rami of L1-5 were performed in human cadavers, and the course of the dorsal rami, their branches, and the innervation of the zygapophyseal joints in the lumbar region were specifically studied. At the L-1 through L-4 levels, the dorsal rami divide into medial and lateral branches within the intertransverse ligaments. Each medial branch runs across the root of the adjacent superior articular process. At the caudal edge of the process, the branch turns medially beneath the mammillo-accessory ligament. Beneath the mammillo-accessory ligament, medial branches occur that innervate the adjacent zygapophyseal joint, and distal zygapophyseal branches arise at the laminar level to innervate the next lower joint. The L-5 dorsal ramus runs along a groove between the ala of the sacrum and its superior articular process. At the caudal edge of the articular process, the ramus divides into medial and lateral branches, and the medial branch supplies the L5-S1 articulation.

The lumbosacral dorsal rami of the cat were studied by gross dissection. The L1-6 dorsal rami form three discrete branches: lateral, intermediate and medial. The lateral branches supply the iliocostalis lumborum and become cutaneous over the back. The intermediate branches ramify in the longissimus lumborum, and are separated from the lateral branches by the lumbar intermuscular septum. The medial branches supply the multifidus and have a constant branch, the nerve to intertransversarii mediales. The L7 dorsal ramus forms only medial and intermediate branches. The S1 and S2 dorsal rami form three branches, the middle of which form the ascending sacral trunk and accessory ascending sacral trunk. The ascending sacral trunk is derived from S1 and S2, the accessory ascending sacral trunk from S2. Both nerves are the exclusive nerve supply of lumbococcygeus.

Four elements of the nervous system may be involved in the production of the lumbar intervertebral disk syndrome. These are the lumbosacral nerve roots, the spinal nerves, the dorsal rami and the sinuvertebral nerves. Each nerve is associated with a particular group of pathologic conditions which may irritate the nerve and produce symptoms. The anatomy of each nerve determines which particular conditions may irritate it. Moreover, one or both of two mechanisms may be involved in symptom production. The type of nerve irritated determines which mechanism, low back pain and referred lower limb symptoms are produced when afferent fibers from dorsal and ventral rami are stimulated where they pass in common through spinal nerves or nerve roots. In the second mechanism, dorsal rami or sinuvertebral nerves are stimulated. This directly produces low back pain, but referred pain is produced by reflex mechanisms in the spinal cord.