History

The recognition of spinal disorders has a history almost as long as written language, with early Egyptian, Greek, Roman, and Arabic texts making reference to treatment of spinal disorders as early as 1550 BCE. Hippocrates (460 BCE-377 BCE), Celsus (25 BCE-50 BCE), and Galen (131 BCE-199 BCE) developed similar methods of treating spinal fractures that included traction and immobilization.¹ These methods continued to be used and taught by Chinese and Arabic practitioners and are used even today in a modified manner when treating spinal fractures nonoperatively.

Hippocrates considered the possibility of surgical treatment of paralyzed individuals with spinal fractures but did not attempt the intervention. Paulus of Aegina proposed operative repair of spine injuries in the 7th century, but spine surgery was initially undertaken in the 18th century as a means of treating tuberculous paravertebral abscesses. Progress in spine surgery was limited by high rates of postoperative infection until the ideas of Lister and Semmelweis made antisepsis standard during surgery.²

Sciatica, which describes pain originating in the back and radiating into the buttocks and legs, was recognized as being the result of an insult to the sciatic nerve by Cotugno in 1764. Definitive association with the intervertebral disk was established by the breakthrough work of Mixter and Barr in 1934. They published a small study describing symptoms that they postulated were caused by degenerative changes in the intervertebral disk that might be relieved by surgical intervention.³

Rapid advances in the treatment of spinal disease have been made in the last century, in no small part due to the development of Röntgen’s use of x-rays for imaging in 1895. Other imaging modalities, including myelography, computed tomography (CT), and magnetic resonance imaging (MRI), further revolutionized the diagnosis and treatment of spinal disorders. The advent of spinal instrumentation, prosthetics, fiberoptics, and new biotechnologies continues to increase options for the treatment of spinal disease.

Anatomy

The basic anatomy of the lumbar spine is well understood. There are typically five lumbar vertebrae, each composed of a vertebral body anteriorly and a neural arch posteriorly. The neural arch in turn is made up of a posterior spinous process, two lateral transverse processes, and the laminae between them. Between the transverse processes and the vertebral body are the pedicles. The spinal and transverse processes act as attachment points for the deep back muscles. The neural arch surrounds the vertebral canal and protects the spinal cord.

Two superior articular facets and two inferior articular facets are positioned at the point where the pedicles and laminae meet. The superior articular facets form a synovial joint articulation with the inferior articular facets of the vertebrae one level higher. Each pedicle is notched superiorly and inferiorly. These notches in two contiguous vertebrae form the intervertebral foramen, which allows passage of the spinal nerves from the spinal cord.

Vertebral continuity is maintained by the intervertebral disks, the facet joints, and the spinal ligaments. The anterior longitudinal ligament (ALL) runs along the anterior surface of the vertebral bodies and resists extension. The posterior longitudinal ligament (PLL) runs along the posterior surface of the vertebral bodies and resists flexion. This ligament attaches primarily to the annulus fibrosis of the vertebral disks and lacks the mechanical strength of the ALL.

The facet joint is responsible for weight bearing only when the spine is in extension. It is the intervertebral disk that bears the axial load of upright posture. The intervertebral disk is composed of a central nucleus pulposus derived from the primitive notochord and the surrounding fibrous annulus fibrosus. The annulus resists the lateral forces created by compression of the nucleus pulposus during weight bearing.

The recurrent sinuvertebral nerve innervates the disk space area, specifically the PLL and posterior annulus. The lateral and anterior regions of the annulus are supplied by autonomic nerves. Immunohistochemical studies have demonstrated that the outer regions of the annulus are innervated but the inner regions of the annulus and the nucleus are not.⁴ Degenerated lumbar disks show increased innervation and vascularity.⁵ The bony vertebral end plate also receives a similar distribution and density of innervation.⁶

Interosseous tributaries from the lumbar arteries coming directly off the aorta supply the vertebral body and serve the capillary beds of the cartilaginous vertebral end plate. The disk receives most of its nutrition by diffusion from the capillary beds of the cartilaginous end plate.⁷

Pathophysiology

The fibrocartilaginous intervertebral disk is an extremely effective shock absorber that serves to distribute forces equally across the intervertebral end plates; it resists spinal compression while allowing limited movement.

The nucleus pulposus is composed of a proteoglycan and water gel held together by an irregular network of type II collagen and elastin fibers. Aggrecan, the primary proteoglycan of the intervertebral disk, has a high concentration of both keratin sulfate and chondroitin sulfate. Sulfation of these molecules in turn provides a negatively charged molecule with the osmotic properties necessary to draw large quantities of water into the disk.⁸ It is the water content of the disk that allows its shock-absorbing capabilities. When the disk is loaded, water is extruded from the disk while the lateral forces are restrained by the collagen fibers of the annulus fibrosus. When the disk is unloaded, the osmotic gradient between the disk and plasma causes water to return to the disk, thereby preparing it to dissipate load forces again.

As mentioned previously, the disk is dependent on diffusion from the end plate capillary beds for acquiring nutrients and dissipating metabolites. Studies have shown that chronic lack of oxygen to the cells of the nucleus pulposus will cause the cells to become quiescent whereas lack of glucose can kill these cells.⁹ Because the disk is dependent on simple diffusion to meet its metabolic needs, it is very susceptible to metabolic or mechanical damage and has very limited ability to recover.¹⁰ Decreases in end plate permeability, diffusion, and hence metabolic transport are a normal function of aging. In contrast, end plate damage as a consequence of disk degeneration causes an increase in the permeability and diffusion of metabolites. This may be one way in which normal aging changes may be differentiated from pathologic degenerative changes.¹¹

Disk cells are capable of growth and adaptive remodeling to cope with the stresses applied to them.¹² Matrix is synthesized, broken down, and turned over by the actions of matrix metalloproteinases (MMPs). Studies using markers for MMPs have shown MMP levels to be highest during growth and to decline in adulthood. Degenerative disks also have high MMP levels, thus providing a possible therapeutic target.¹³ It has been postulated that keeping MMPs in check may minimize breakdown of the extracellular matrix in degenerative disks.¹⁴

Aging is associated with loss of proteoglycan content in the nucleus and hence water content, which decreases the ability of the disk to effectively and evenly transmit loads to the vertebral end plate.¹⁵ Loss of proteoglycan and water content may be regional, a finding suggesting focal damage and degeneration.¹⁶ At the same time, the fibers of the annulus fibrosus weaken and lose their ability to constrain the lateral forces imposed on the nucleus. It would seem, then, that the incidence of lumbar disk herniation would increase with age. This is true to a point, with the maximum incidence of lumbar disk herniation occurring in individuals 30 to 50 years of age.¹⁷ However, the incidence of lumbar disk herniation decreases after the age of 50.

There is an explanation for this seemingly paradoxical finding. The disk gradually loses its ability to expand over time. In a young person, the disk has a tremendous ability to expand, but the strong annulus fibrosus prevents excessive lateral expansion. As the annulus loses strength and the disk is still able to expand, the possibility of herniation increases during the third to fifth decades of life. After about 50 years of age, the disk begins to significantly lose its ability to expand. As a result, the incidence of disk herniation decreases despite the continued decline in annulus strength.¹⁸

Whether disk degeneration is a process different from the normal aging process or simply the normal aging process proceeding at different rates in different individuals is a controversial issue. Adams and Roughley argue that degeneration and aging are not synonymous and propose the following working definition for disk degeneration. “The process of disk degeneration is an aberrant, cell-mediated response to progressive structural failure. A degenerate disk is one with structural failure combined with accelerated or advanced signs of aging.”¹⁹ What determines whether a disk will undergo this aberrant response is a multifactorial response. Certainly, there appear to be risk factors for the development of disk degeneration and disk herniation, including driving of motor vehicles, sedentary occupations, vibration, previous full-term pregnancy, smoking, physical inactivity, increased body mass, and tall stature.²⁰

If a degenerative disk is one suffering from structural failure, it stands to reason that mechanical loading is ultimately implicated.²¹ The structural disruption then initiates a cascade of biomolecular events that lead to further failure in a positive feedback loop. Degenerative disks display higher levels of MMPs, nitric oxide (NO), interleukin-6 (IL-6), and prostaglandin E₂ (PGE₂).²² Once again, differentiating cause and effect is difficult. Increased hydrostatic loading causes release of NO, which in turn downregulates the synthesis of proteoglycan.²³ A disk with decreased proteoglycan content has decreased water content and thus decreased ability to dissipate applied forces and is more susceptible to degeneration, with further release of NO. It can be seen that this type of cascade can rapidly lead to severe pathology.

The pathophysiology of the radicular pain associated with sciatica has still not been entirely elucidated. There is, of course, an element of mechanical pressure on the nerve root, particularly in the central disk herniations responsible for cauda equina syndrome, but not all pain can be explained by mechanical pressure. It is likely that cell-mediated processes involving NO, MMPs, cytokines, and other inflammatory mediators are involved as well.²⁴

There can be little doubt that genetic factors are also at work.²⁵ Intervertebral disk disease shows a tremendous familial predisposition, with heritability ranging from 34% to 61%, depending on the vertebral level.²⁶ Various alleles have been identified that may play a contributing role, including those for haplotypes of collagen,²⁷ IL-1β,²⁸ and aggrecan triplet-repeat polymorphisms.²⁹ It has even been demonstrated that individuals with a certain IL-6 haplotype are far more likely to be affected with sciatica.³⁰ Research continues in this area and will probably demonstrate links between intervertebral disk disease and many other gene polymorphisms.

Clinical Findings

It is important to keep in mind that disk degeneration is not synonymous with back pain or disk herniation. A study by Paajanen and associates demonstrated that 35% of a normal control group had substantial disk degeneration in at least one level as demonstrated by MRI.³¹ Miller and coworkers found a 90% incidence of disk degeneration after 50 years of age in a series of 600 autopsy specimens obtained from 273 cadavers ranging in age from infancy to 96 years.³² However, when low back pain is present in individuals with changes seen on MRI, as many as 80% of patients are able to accurately localize their pain to the appropriate level when asked to define a horizontal line across their region of greatest pain.³³