Disc space narrowing and osteophytes

Key points

•

Disc space narrowing and osteophyte formation are features of the general spine condition known as intervertebral disc (IVD) degeneration
•

IVD degeneration is an age-related condition
•

IVD degeneration is highly heritable
•

The burden of IVD degeneration predicts the likelihood of low back pain episodes

Introduction

The outgrowth of small spicules of bone—osteophytes—from the endplate of the vertebral body is one of the hallmarks of the bony response intervertebral disc (IVD) degeneration ( Fig. 7.1 ) (see Chapter 6 ). Such degeneration is characterized by activation of matrix metalloproteinases, reduction in the length of proteoglycans attached to types I and II collagen, and consequently a loss of hydrophilicity of the matrix molecules [ ]. The loss of water from the IVD is evident on T2 weighted magnetic resonance (MR) scans as a change from bright white IVD to light gray, then dark gray ( Fig. 7.2 ). In addition, IVD height is lost and this feature is known interchangeably as loss of IVD height and IVD space narrowing, the latter because early imaging using plain radiology film showed a space between vertebrae rather than an image of the IVD itself ( Fig. 7.3 ).

The features of lumbar IVD (osteophyte formation and loss of IVD height on MR scan) are correlated [ ], and both are known to be heritable [ ]; that is, a significant proportion of phenotypic variation is accounted for by genetic factors. In fact, twin studies have shown a surprisingly high level of heritability (60%–80%)—which varies by spine level—for a condition that was long considered to be largely occupational [ ]. In this chapter, we discuss the changes that take place at the cellular level and describe the histology of osteophytes and the IVD as it loses height and degenerates. We describe how the two features are related to one another and their epidemiology. The most important question, however, is their contribution to low back pain, a highly prevalent condition worldwide, with considerable social and economic costs. Low back pain now causes more disability on a global level than any other condition [ ]. Finally, we discuss the rare complications that may arise from osteophyte formation.

Pathogenesis

The primary cause of osteophyte formation at any joint is thought to be joint instability, detected perhaps by proprioceptive receptors. Osteophyte growth is an attempt to stabilize the joint [ ]; they grow out laterally and appear to try and increase the surface area of the joint. In the spine, osteophytes grow horizontally from the margin of the vertebral body. The most likely causes of perceived instability are IVD degeneration or facet joint osteoarthritis (see Chapter 14 ), which is closely related (see later) ( Fig. 7.3 ). It is difficult to perform well-designed studies in humans to work out the sequence of events but longitudinal cohorts are beginning to address this important issue [ ]. The main cause of joint instability is believed to be changes in the IVD triggered by endplate damage (see Chapter 10 ); loss of IVD height occurs with subsequent peripheral annular bulging. IVD cells are lost, and matrix metalloproteinases degrade the proteoglycans are degraded and water escapes, some through fissures formed with degeneration of the annulus fibrosus (AF), resulting in further thinning of the IVD; vertebral endplate sclerosis and osteophyte formation ultimately follow. One study found anterior and lateral traction osteophytes in greater numbers than posterior and posterolateral traction osteophytes [ ]. Rarely, the osteophytes can lead to local pressure effects and, rarely, serious complications (see later).

One large American cohort study, the Johnson County Osteoarthritis Project, has found different radiographic biomarkers associated with facet joint osteoarthritis and IVD space narrowing relate to osteoarthritis at other sites [ ]. For example, there was a close relationship between lumbar spine IVD and hand and knee osteoarthritis, which was not seen at the hip. This highlights the close relationship between IVD, facet joint osteoarthritis and osteoarthritis of the cervical and lumbar spines and peripheral joints, which has been reported in many epidemiological studies [ ]. The relationship between IVD degeneration facet joint osteoarthritis is discussed further in the following.

Radiological features

Visual assessment and subjective evaluation remain the commonest method for classifying osteophytosis and IVD space narrowing into mild, moderate, and severe, using standard imaging techniques. A number of coding methods have been devised for epidemiological study but standardization of the scales used to code these features has been slow despite numerous calls for international agreement [ ]. Different specialties tend to use different scales: for example, surgeons use the Pfirrmann or Schneiderman classifications to characterize the overall degree of degeneration, while epidemiologists are working with more granular scales that code each radiological feature separately [ ]. An international working party was set up (Hong Kong, 2013) to try and address this and set international standards through professional groups, such as the International Society for the Study of the Lumbar Spine. In addition, machine learning methods are used to try and improve image interpretation, particularly the correlation between imaging features and symptoms, by reducing the noise introduced by human error in coding [ ].

Measuring the osteophyte extension beyond the vertebral body margin and commenting on the temporal orientation of the osteophytes are only performed by a few people. Clearly, the imaging tool used has an important influence—plain films only demonstrate osteophytes in a single plane while they may occur throughout the margin of the vertebral endplates, to a variable degree. MR is also not the optimal imaging method for bone but is the best for IVD. Some classify according to location of osteophytes on the vertebral body as anterior, lateral, and posterior osteophytosis ( Fig. 7.4 ).

Epidemiology

Both loss of IVD height and osteophyte formation are clearly age-related traits, and this has been recognized for centuries but the first formal epidemiology studies date from the 1960s [ ]. Both features are observed occasionally in children, have a low prevalence in adolescents and young adults [ ], and become highly prevalent—almost universal—in those over 60 years old [ , ]. Degenerative change in the vertebral column is seen in all ethnic groups but there are some differences, particularly in osteophyte floridity, which may reflect different genetic background. Although there is a correlation between different forms of imaging, differences in prevalence may also reflect the coding methods used to determine the presence or absence of osteophytes and IVD narrowing. The most widely reported method is plain radiography for reasons of cost and availability, and large cohorts have been assembled for clinical studies [ ] and drug trials in osteoporosis. Although the imaging in these studies has been selected for the incidence and prevalence of vertebral fracture, the plain radiographs have lent themselves well to studies of IVD. The superior method of MR imaging is increasingly used because the costs are falling, and collections now exist in the United States, United Kingdom, Hong Kong, and Japan that are vital to performing well-powered studies. Finally, there are reports of a few studies that have examined postmortem specimens of the spine.

In the study of Karasik et al. [ ], a significant association between increased anterior lumbar osteophytes and prevalent abdominal aortic calcification was demonstrated, suggesting a link between IVD and cardiovascular disease. Whether there is truly a pathogenic connection involving calcium metabolism or whether both simply represent diseases of aging remains to be determined but there does seem to be a link [ ] ( Fig. 7.1 ).

Osteophyte epidemiology

In one of the largest, early epidemiology studies of the distribution, determinants, and clinical correlates of vertebral osteophytes in Northern Europeans, the Manchester group examined radiographs from 681 female and 499 male individuals recruited through five general practice registers, inviting them to attend a clinical interview and plain lateral thoracic and lumbar spine radiograph [ ]. The mean age of the recruits was 63 years, and the majority (84% men and 74% women) manifested at least one vertebra considered of grade 1 or higher osteophyte, between T4 and L5. The presence of osteophytes was shown to be associated with increasing body mass index; and in men, there was also an association with heavy physical exercise that was not seen in women. Osteophyte formation has also been shown to be associated with increased bone mineral density (BMD), even after taking body mass index into account, and this remains the case even when considering measures of BMD at the hip, remote from the spine [ ]. In the United States, a study of women undergoing spine radiography to exclude skeletal damage after trauma revealed an age-related increase in osteophytes [ ]. A similar prevalence in women was seen as in the UK study, with almost all subjects over 50 years old having some form of osteophyte [ ]. Of the 840 participants enrolled in the US Johnston County Osteoarthritis Project, the prevalence of osteophytes again was shown to vary by age, with 79% in age group 45–55 rising to 96% in those aged over 75 years. Osteophytes were significantly associated with age, weight, and manual social class with men manifesting significantly more osteophytes than women [ ]. There was also a significant association of spine osteophytes with knee osteoarthritis, which likely reflects the shared genetic predisposition.

Longitudinal radiographic studies, for example the UK Chingford study of women ( www.chingfordstudy.org.uk ), have estimated that anterior osteophytes develop at a rate of approximately 4% per annum in mid-life [ ]. The distribution of osteophytes throughout the spine has been shown by a number of authors to manifest two peaks of high frequency—one in the mid-thoracic region T6–T7 in women or T8 in men, and another one in the mid-lumbar area L3–L4 [ , ]. Longitudinal change has been reported on MR scans in the TwinsUK study, and it is found that the anterior osteophyte score at least doubles over a 10-year follow-up period [ ]. This change was found to be heritable—influenced by genetic factors—in those under 50 years at baseline, reflecting the paradigm that some people are genetically predisposed to be “bone formers” while other women are “bone losers” and go on to develop postmenopausal osteoporosis.

In a postmortem study in Thailand subjects having a mean age of 63 years (range, 15–96) were found to have lumbar osteophytes in 175 specimens (97.2%) [ ]. This very high prevalence likely reflects the more thorough 3D examination compared to plain radiography. The highest frequency was at L4, most were on the vertebral superior and inferior surfaces and articular facet (39.7%, 38.4%, and 22%), respectively. The osteophyte length was correlated with age, and males had significantly longer osteophytes than females.

IVD narrowing epidemiology

Intervertebral disc space narrowing, so called because the IVD isn’t actually visualised on plain film but is a ‘space’, is usually graded on a categorical scale with the help of a coding atlas. It was reported in a large study from Scotland of approximately 500 men and women over 50 years old recruited from primary care. Subjects underwent lateral spine radiographs and were found to have an IVD space narrowing prevalence of 37% at a mean age of 65 years [ ]. In the Johnson County study, prevalence estimates for IVD space narrowing were lower than for osteophytes, at 58% overall, but again the effect of age was clear: 36% prevalence in the 45–54 year age group; rising to 80% in those over 75 years [ ]. However, the sex distribution was different from that seen in osteophytes, with a higher prevalence of IVD space narrowing in women compared to men (60% vs. 54%). This observation is in keeping with the notion that the specific triggers for the individual features of degeneration may not be the same across the sexes.

In large radiographic studies of thousands of elderly Chinese [ ], the IVD space narrowing was examined and also showed an increased prevalence with age, even continuing through the advanced seventh to ninth decades. As in the Caucasian studies, IVD space narrowing was more prevalent in women than men and the difference was greatest in the oldest age group, suggestive of more rapid IVD space loss in women as they age.

Longitudinal change in IVD space narrowing has been reported in the Chingford radiographic study of women as being 3% per annum [ ]. In the MR imaging study in TwinsUK, change in IVD height over a decade of follow-up was not found to be significant ( n = 450 twins) nor was it found to be heritable [ ].

IVD space narrowing is not the only pathology to stimulate the development of osteophytosis. Other causes include trauma, infection, inherent, and occupational—through changes in lumbar IVD morphology associated with prolonged sitting [ , ]. Climate changes have been linked to accelerated IVD degeneration, with greater preponderance in cold weather, presumably due to loss of elasticity of the nucleus pulposus (NP) and theoretically increased vulnerability of the IVDs and hydration (IVDs shrink in dehydrated individuals): evidence from athletes and diurnal variation in IVD height [ ] affecting spine flexibility, intradiscal pressure, and contact compressive forces in the facet joints [ ]. In a study by An et al. utilizing modern imaging techniques, a clear relationship between the severity of IVD and increases in segmental lumbar motion in vivo was identified [ ]. Nutritional factors were also considered as predisposing factors [ ]. Associated conditions include enthesopathy and enthesophyte formation, which are similar but different, and are most commonly caused by inflammatory spine diseases such as ankylosing spondylitis. The Scottish study clearly demonstrated an association between IVD joint space narrowing and osteophytes with endplate sclerosis, which is another feature of IVD degeneration [ ]. Other features of IVD include vertebral edema and fibrosis/sclerosis known as Modic change (see Chapter 11 ) [ ] and endplate damage [ ] are associated with each other as well as IVD degeneration.

Association with facet joint arthropathy

There is a strong association between facet joint arthropathy and vertebral osteophytosis although the co-occurrence of these age-related conditions is a challenge to study because different imaging techniques are needed for IVD degeneration (MR imaging) and facet joints (CT imaging). Whether the anterior elements degenerate first, leading to malalignment of the posterior facet joints, or vice versa, remains a subject of debate [ , ].

It was thought that the IVD degenerates first, then the facet joint, based on one study that has used both forms of imaging to obtain information about both sites [ ]. More recent work on postmortem specimens, however, suggested that the pattern is age related. Young people in their 20s have been found to have facet joint osteophytes in the absence of IVD degeneration [ ]. The prevalence of facet joint arthropathy rises steeply with age such that some facet joint osteoarthritis is pretty much universal in the over 50s [ ]. Men have a higher prevalence of facet joint arthrosis than women [ ]. Right- and left-sided facets are equally affected. The spine level manifesting the highest prevalence of facet joint disease is L4/5.

Influence of osteophytes and IVD space narrowing on low back pain

There is some doubt as to whether osteophytosis in general in the spine gives rise to symptoms of back pain, but various reports suggest that it may sometimes be responsible for certain obscure myelopathies in elderly people. However, there remains a lot of debate because osteophytosis is often seen in asymptomatic individuals. The major debate around osteophytes and IVD space narrowing exists on two fronts. Firstly, whether population studies support the notion that the overall degeneration “burden” of the spine seen on imaging predisposes to episodes of low back pain. And secondly to what extent imaging can provide an accurate diagnosis of the cause of low back pain in any given individual. This chapter will focus on the former question, but it must be recognized that the lack of standardized definition of both low back pain and imaging phenotypes in epidemiological practice hampers progress massively. Mild symptoms of low back pain are so prevalent as to be considered normal, and stringent criteria, for example, including measures of pain severity and disability as in the UK Medical Research Council Neck and Back Pain Questionnaire, are rarely used. Among the first large epidemiological studies published, one from the United Kingdom suggested an association between osteophytes and low back pain in men, with no association detected in females despite a sample of 681 women [ ]. From an epidemiological study of >1100 women in the Hague, Netherlands, a nested case-control sample was drawn of women having/not having previous back pain. The burden of IVD degeneration on plain film was found higher in those reporting previous low back pain. A meta-analysis of studies published in 1999 found no firm evidence for an association between degeneration of the IVD (as determined by radiographic osteophytes, IVD space narrowing, and endplate sclerosis) and low back pain.

In contrast, the North American Johnson county project [ ] found low back symptoms were significantly associated with IVD space narrowing but not osteophytes (or facet joint osteoarthritis). The Japanese ROAD study used the composite Kellgren and Lawrence score combining IVD space narrowing and anterior osteophytes and showed that, over 3 years of follow-up, those having a higher score at baseline were at increased risk for low back pain over the preceding month, in both men and women [ ]. The use of MR imaging may improve the sensitivity to detect these vertebral changes, as well as others including IVD signal intensity and IVD bulge. In one of the largest studies of its time, the TwinsUK study undertook T2-weight MR spine imaging in 1050 unselected twins and found a clear relationship with previously reported episodes of severe and disabling low back pain. The results showed that MR score—a combined measure of the four degenerative traits of the IVD and osteophytes—was the single strongest predictor for episodes of low back pain [ ] although it must be borne in mind that psychological and social factors also contribute [ ].

Low back pain itself is heritable approximately 40% [ , , ]. It has been shown that low back pain and chronic widespread musculoskeletal pain share genetic determinants, suggesting that at least some of the genetic predisposition to low back pain lies among “pain” gene variants rather than “IVD degeneration” genes [ ]. Some researchers are moving away from a spine-based focus to understand low back pain, recognizing that the biopsychosocial model of back pain [ ] may not only inform advances in treatment based on risk stratification [ ] but has recently been shown to be underpinned by genetic factors [ ]. These findings highlight the wealth of evidence demonstrating only a poor correlation between imaging abnormalities and symptoms: if we are to address low back pain adequately in patients then a broader approach, looking beyond simple spine anatomy, is required.

Complications/sequela of osteophytosis and IVD space narrowing

In addition to their possible association with low back pain, there are structural conditions that may arise from the local effects, particularly of osteophyte formation. These may be divided in common and rare. Common complications include the following: