Degenerative Disease of the Spine

Nomenclature and Classification

Since the first description of a ‘ruptured disc’ with monoradiculopathy by Mixter and Barr in 1934, the terminology to grade and report degenerative disease of the spine has been controversial and confusing. Some nomenclature systems are based on description of the observed morphology of the disc contour, while others include the pathological, clinical, functional and/or anatomical features. Cross-sectional imaging is based on other definitions and concepts compared with myelography or discography. In 2001 a new nomenclature was proposed by the Combined Task Forces of the North American Spine Society, American Society of Spine Radiology and American Society of Neuroradiology that consists of a classification system for the reporting on imaging studies based on pathology. In this chapter we follow the general classification of disc lesions as proposed by the Combined Task Forces. The general classification as proposed by Milette is given in Table 48.1 .

Age-Related Changes in the Intervertebral Disc

The Combined Task Forces reserved the term ‘normal’ for young discs that are morphologically normal, without signs of disease, trauma or ageing. However, the ‘normal’ appearance of an intervertebral disc is age related due to biochemical and anatomical changes that result in a variable appearance on magnetic resonance imaging (MRI).

In infants and young children , the intervertebral disc is prominent relative to the height of the adjacent vertebral bodies. With increasing age, the relative disc volume decreases. The transition between the intranuclear cleft and the outer portion of the annulus fibrosus is sharp but becomes less distinct with age.

In young adults the disc contour coincides with the margins of the adjacent vertebral end plates. On MRI the normal adult disc has a low to intermediate signal on T ₁ weighted images and a high signal intensity on T ₂ weighted images relative to the bone marrow in the adjacent vertebral bodies. On T ₂ the bright nucleus pulposus is indistinguishable from the inner annulus. The normal adult end plates, the outer annulus fibrosus and the ligamentous structures are relatively hypointense on T ₁ and T ₂ . The outer annulus is visualised on T ₂ and has a low signal intensity due to a high collagen content. In young adults, diurnal changes in T ₂ relaxation are present but these changes disappear after the age of 35 years and that is thought to be a normal aspect of ageing.

In the third decade the intranuclear cleft appears as a horizontal band of decreased signal intensity on T ₂ in the central part of the discs, giving it a bilocular appearance on sagittal images. The intranuclear cleft represents a fibrous transformation of the gelatinous matrix of the inner nucleus pulposus.

In middle-aged and elderly persons there is a gradual signal loss of the intervertebral discs on the T ₂ images until the disc becomes hypointense. The loss of signal correlates to a decrease in water and proteoglycan content and an increase in collagen content. Although the decrease in T ₂ signal of the intervertebral disc is age related, when it is more pronounced it predisposes to degenerative changes such as loss of disc height, disc herniation and annular tears ( Fig. 48.1 ). The highest T ₂ values are seen near the vertebral end plates, and the lower T ₂ values are present in the intranuclear cleft and the peripheral annulus fibrosus due to its fibrous nature.

In general, in the normal ageing disc the height is preserved, the disc margins remain regular and radial annular tears are not a usual consequence of ageing. On the basis of a series of postcontrast MRI studies of the lumbar spine, degeneration and normal ageing have been shown to be two separate processes.

Resnick and Nywayama conclude there are two different processes of degeneration: a first type, which can be considered normal ageing, involving the annulus fibrosus and adjacent ring apophysis (spondylosis deformans) ( Fig. 48.2 ), and a second type, called intervertebral osteochondrosis, affecting the nucleus pulposus and the vertebral end plates, corresponding to a pathological ageing process.

Anterior and lateral marginal osteophytes are considered as normal ageing, whereas end plate changes and reactive bone marrow changes are considered pathological. Large amounts of gas in the central disc space, as seen on conventional radiographs or computed tomography (CT), are indicative for pathological intervertebral osteochondrosis, whereas small amounts of gas near the apophyseal enthesis should be considered as spondylosis deformans ( Fig. 48.3 ).

Degenerative Disc Disease

The prevalence of degenerative disc disease is linearly related to the age. Intervertebral disc degeneration begins already early in life. Many other factors (e.g. biomechanical and the quality of collagen) are also implicated. Degeneration includes changes involving the end plates (sclerosis, defects, Modic changes and osteophytes), as well as disc changes (fibrosis, annular tears, desiccation, loss of height and mucinous degeneration of the annulus).

The relation between low back pain and abnormalities in the lumbar spine is controversial since these findings are often seen in asymptomatic patients on plain radiographs, CT and MRI. Degenerative changes in the disc are already seen in one-third of healthy asymptomatic persons between 21 and 40 years old. The high prevalence of asymptomatic disc ‘degeneration’ must be taken into account when MRI is used for assessment of spinal symptoms.

Although the validity of disc height as an indication for degenerative disc changes is questionable, the loss of height of the intervertebral space is the earliest sign of disc degeneration on plain radiographs. Loss of disc height has been reported in asymptomatic subjects, indicating there is no direct relationship between clinical symptoms and imaging findings. The position of the patient (lying flat or standing) should also be taken into account.

Other signs including sclerosis of the vertebral end plates, osteophytes, vacuum phenomenon and calcification are more reliable indicators of pathology, although they indicate late degenerative changes.

Signal loss on T ₂ is an early indicator of intervertebral disc degeneration on MRI. As described earlier, in normal ageing the decrease in signal intensity on T ₂ should be uniformly distributed over the different levels. If the signal loss is seen only in one or two levels, this should be interpreted as abnormal. This finding is often referred to as ‘a black disc’ and has been applied to describe discogenic pain syndrome (see Fig. 48.1 ). The degenerative process typically starts at the levels with the highest mechanical stress (motion/weight bearing). In the cervical spine levels C5 to C6 and C6 to C7 are most commonly involved and in the lumbar spine the levels L5 to S1 and L4 to L5.

Annular Tears

With ageing, the intervertebral disc becomes more fibrous and less elastic. The degenerative changes are accelerated when the structural integrity of the (posterior) annulus fibrosus is damaged by overload. This will eventually lead to formation of fissures in the annulus fibrosus. In the international literature the term ‘annular tear’ is most widely used, and it is also supported by the Combined Task Forces; however, the terminology ‘annular fissure’ is also used. One should take into account that ‘annular tear’ does not imply this is caused by trauma.

Annular tears can be categorised into concentric, transverse or radial tears:

• Concentric tears are circumferential lesions found in the outer layers of the annulus fibrosus. Like onion rings, they represent the splitting between adjacent layers of the lamellae annulus. They are believed to be posttraumatic from torsion overload injuries.

• Transverse tears or ‘rim lesions’ are horizontal ruptures of Sharpey's fibres near the insertion in the bony ring apophysis. The clinical significance of transverse tears remains unclear, although some authors believe they influence and accelerate degeneration and are associated with discogenic pain. They are believed to be post-traumatic in origin in some cases and are often associated with small osteophytes.

• Radial tears are annular tears permeating from the deep central part of the disc and extending outwards toward the annulus in either the craniocaudal or the transverse plane ( Fig. 48.4 ). Most of these tears do not reach the pain-sensitive outer layers of the annulus. Radial annular tears are associated with disc degeneration and a complete radial tear is necessary for progressive deterioration of the disc.

  

The clinical significance of annular tears remains unclear. Some annular tears can cause low back pain without the presence of modification of the disc contours, also known as discogenic pain. On the other hand, annular tears are often found in asymptomatic patients and can be seen as a part of the ageing process.

On MRI, annular tears can be seen as an area of high signal intensity on T ₂ or as foci of annular enhancement on gadolinium-enhanced T ₁ ( Fig. 48.5 ). On T ₂ the signal intensity is the same as the adjacent cerebrospinal fluid (CSF). Repetitive microtrauma may cause annular tears to enlarge and become inflamed; this can be seen as an area of increased signal intensity. This phenomenon is seen on T ₂ and is known as a high-intensity zone (HIZ). The HIZ is a combination of radial and concentric annular tears which merge in the periphery of the disc. The presence of an HIZ is believed to be related to discogenic pain as it involves the outer, highly innervated layers of the annulus. However, the value of this sign is limited due to a poor sensitivity and a limited positive predictive value. On T ₁ , extradural inflammation is seen as a zone of intermediate signal intensity replacing the fat between the disc and the dural sac; on postcontrast images there is intense enhancement.

In the cervical spine, annular tears, rim lesions and prolapsed disc material are poorly recognised on MRI, even in severely degenerative disc.

In the thoracic spine, herniated disc fragments are often associated with abnormal straight or curvelinear densities in the intervertebral space on CT, also known as the ‘nuclear trail sign’. On MRI this finding may also be associated with a comet tail configuration in the axial plane. This sign indicates advanced disc disruption and degeneration and must be distinguished from an ageing disc that has not failed.

Disc Herniation

Herniation is defined as a displacement of disc material (cartilage, nucleus, fragmented annular tissue and apophyseal bone) beyond the limits of the intervertebral disc space. The definition of the intervertebral disc space is the three-dimensional volume defined by the adjacent vertebral end plates and the outer edges of the vertebral ring apophysis, excluding osteophytes. A break in the vertebral end plates or disruption of the annulus fibrosus is thus necessary for disc displacement to occur. Disc herniations through one or both vertebral end plates are called intervertebral herniations. These herniations are also called Schmorl's nodes and are often surrounded by reactive bone marrow changes ( Fig. 48.6 ). One hypothesis is that this type of herniation is caused by a weak spot in the vertebral end plate caused by regression of the nutrient vascular canals, leaving a scar. When in young individuals a herniation of the nucleus pulposus through the ring apophysis occurs before bony fusion a small segment of the vertebral rim may become isolated. This is called a limbus vertebra and is most commonly found in the lumbar region and less frequently on the midcervical level. They are characterised by a defect in the anterior wall of the vertebra and usually at the anterior superior margin in the lumbar spine and at the anterior inferior margin at the cervical level.

A ‘bulging’ of the disc is defined as a circumferential or generalised disc displacement involving more than 50% of the disc circumference and is not considered as being a disc herniation ( Fig. 48.7 ). The term ‘bulging’ is not correlated with pathology or aetiology but only refers to the morphological characteristics. A bulging can be physiologically seen on the level L5 to S1 and, on midcervical level, can reflect advanced degenerative changes, can be a pseudoimage caused by partial volume effect, can be associated with bone remodelling or occur in ligamentous laxity. There are two types of disc bulging: an asymmetrical type, as frequently seen in scoliosis, or the symmetrical type, with equal displacement of the disc in all directions.

Two types of disc herniations can be differentiated on the basis of the shape of the displaced disc material ( Fig. 48.8 ). A disc herniation is called an ‘extruded disc’ when the base against the disc is smaller than the diameter of the displaced disc material, measured in the same plane, which can be either axial or sagittal ( Fig. 48.9 ). A ‘disc protrusion’ is used when the base of the disc is broader than any other diameter of the displaced disc material ( Fig. 48.10 ). A herniated disc can be focal if <25% of the disc circumference is involved or broad based when 25%–50% of the disc circumference is involved.

When there is no connection between the disc and the displaced disc material, it is called a sequestrated fragment or free fragment and is also described as a disc extrusion ( Fig. 48.11 ). On imaging studies, it is often impossible to determine if continuity exists. Therefore it is more practical to use the term ‘migration’, which signifies displacement of disc material away from the site of extrusion regardless of the continuity ( Fig. 48.12 ).

The Combined Task Forces have maintained the distinction between protrusion and extrusion, which is useful because an extrusion is seldom seen in asymptomatic patients. The term ‘disc extrusion’ is also more acceptable to patients compared with a disc herniation.

A ‘contained’ herniation refers to the displacement of disc material which is covered by the annulus fibrosus. If this cover is absent the herniation is ‘uncontained’. With discography it is possible to distinguish a contained from an uncontained disc herniation and separate a leaking from a non-leaking disc, depending on the displacement of injected contrast agent. On cross-sectional imaging (CT or MRI) it is often impossible to differentiate contained from uncontained disc extrusions.

Communication with clinicians requires an accurate and simple classification of the disc fragments in the vertical and horizontal direction. The Combined Task Forces have opted for a classification based on anatomical boundaries frequently used by surgeons. In the transverse/axial plane the following zones are used ( Fig. 48.13 ):

• central (posterior midline) ( Figs 48.14 and 48.15 )

  
  

• paracentral (right/left central) ( Fig. 48.16 )

  

• right/left subarticular (lateral recess) ( Figs 48.17 and 48.18 )

  
  

• right/left foraminal (neural foramen) (see Fig. 48.10 )

• right/left extraforaminal (outside the neural foramen) ( Fig. 48.19 )

  

• anterior zone (anterior and anterolateral).

And in the vertical plane from top to bottom:

• pedicle level

• infrapedicle level

• disc level

• suprapedicle level.

Spontaneous Regression of Disc Herniation

Spontaneous regression of a lumbar disc herniation is a common finding, although the underlying mechanism remains unclear. Several hypotheses have been proposed: dehydration or shrinkage of the disc, retraction of the disc in the intervertebral space or resorption due to an inflammatory reaction. Free fragment herniation, herniations with peripheral contrast enhancement on T ₁ or high signal intensity on T ₂ are predisposing for spontaneous regression (see Fig. 48.11 ). Disc material exposed to the epidural space appears to resolve more quickly than subligamentous disc herniations. Contrast enhancement of the posterior longitudinal ligament indicates an inflammatory response, whereas areas of enhancement in the epidural space below or above the herniated disc indicate venous congestion. Strong contrast enhancement indicates an ongoing absorption process and can be used to evaluate disc reabsorption.

Spontaneous regression of a disc herniation has also been described in the cervical spine and only rarely in the thoracic spine. In the cervical spine, median or diffuse soft-tissue herniations are more likely to regress compared with focal-type herniations.