Advertisement

Inflammatory and Infectious Disorders of the Spine

Imaging Approach
  • Marcel WolfEmail author
  • Marc-André Weber
Living reference work entry

Abstract

Spondylarthropathies are a heterogeneous group of chronic inflammatory diseases affecting the spine and sacroiliac joints. They display a negative serostatus for rheumatoid factor, thus termed seronegative spondylarthritides, but are strongly associated with HLA-B27. Spondylarthropathies can be classified in ankylosing spondylitis (Bechterew disease), the most frequent seronegative spondylarthropathy, or as psoriatic spondylarthritis, reactive arthritis (formerly termed Reiter syndrome), enteropathic spondylarthritis (associated with inflammatory bowel disease), and undifferentiated spondylarthritis. As each category does not represent a distinctive entity, clinical, laboratory, and also imaging findings may overlap, clinical neuroradiology may not lead to a definitive diagnosis, but to certain imaging patterns, and possible differential diagnoses.

Noninflammatory conditions with ossifications of the paravertebral ligaments may result in ankylosis of the spine as well. The most frequent entities of noninflammatory ankylosis are diffuse idiopathic skeletal hyperostosis (DISH), and ossification of the posterior longitudinal ligament (OPLL), but exclusive ossification of the flava ligaments (OFL) may occur as well.

Spondylodiscitis is an infectious disorder of the spine found mostly in elderly, immobile, and immune suppressed patients, in particular after trauma or surgery. As clinical symptoms and laboratory tests are unspecific, imaging plays a pivotal role in its diagnosis. In certain cases, image-guided acquisition of tissue for microbiological tests may be useful for diagnosis, targeted antibiotic therapy, or evaluation of suspicious imaging findings. In early stages of the disease, radiography and CT may be normal, or at least unspecific, in case of preexisting degenerative conditions. MRI is the radiological technique of choice; as it shows pathological findings even in early stages of the disease and potential involvement of the paraspinal tissue.

Keywords

Seronegative spondylarthropathy Ankylosing spondylitis Bechterew disease Spondylodiscitis Pyogenic vertebral osteomyelitis Diffuse idiopathic skeletal hyperostosis DISH ossification of the posterior longitudinal ligament OPLL Imaging Radiology MRI 

Abbreviations

CRP

C-reactive protein

DISH

Diffuse idiopathic skeletal hyperostosis

DMARD

Disease modifying anti-rheumatic drug

ESR

Erythrocyte sedimentation rate

HLA

Human leukocyte antigen

NSAID

Non-steroidal anti-inflammatory drug

OFL

Ossification of flava ligament

OPLL

Ossification of the posterior longitudinal ligament

STIR

Short tau inversion recovery

TNF

Tumor necrosis factor

Inflammatory and Infectious Disorders of the Spine

The spine may be involved in various inflammatory diseases. The two major categories of inflammatory disease of the spine are infectious and noninfectious. Infectious disease can be located in the spine itself, or the surrounding structures. Hematogenous seeding is the most common cause of pyogenic and nonpyogenic infection, but direct inoculation is possible as well. Noninfectious inflammatory conditions of the spine are systemic diseases that can be divided into two main categories, rheumatoid disorders including rheumatoid arthritis, and seronegative spondylarthropathies.

Entheses, the bony insertions of ligaments and tendons, are target structures in certain nonpyogenic inflammatory diseases. Spinal enthesitis can be caused by a seronegative spondylarthropathy, like ankylosing spondylitis, and lead to calcifications of the ligaments, and finally ankylosis of the spine and sacroiliac joints.

Early diagnosis is crucial for initiation of adequate anti-inflammatory therapy. As no single specific diagnostic test is available, and even imaging findings may overlap, the diagnosis of seronegative spondylarthropathies is based on clinical, laboratory, and radiological examinations. Thus, clinical neuroradiology alone may not be capable to lead to a definitive diagnosis, but certain imaging patterns may contribute to possible differential diagnoses.

Noninflammatory conditions may result in ossification of spinal ligaments as well, involving the anterior longitudinal ligament in diffuse idiopathic skeletal hyperostosis (DISH), the posterior longitudinal ligament (OPLL), or the flava ligament (OFL), that can be reliably diagnosed by imaging methods. Although they may be an incidental finding in radiological studies, radiologic work-up is mandatory, when symptoms of spinal canal stenosis, and myelopathy (OPLL and OFL), or fractures occur.

Seronegative Spondylarthropaties

Definition of Entity and Clinical Highlights

Spondylarthropathies can be classified in ankylosing spondylitis (Bechterew disease), the most frequent seronegative spondylarthropathy, or as psoriatic spondylarthritis, reactive arthritis (formerly termed Reiter syndrome), enteropathic spondylarthritis (associated with inflammatory bowel disease), and undifferentiated spondylarthritis. Among all entities of seronegative spondylarthropathies, ankylosing spondylitis or Bechterew disease is by far the most frequent one. As imaging findings are similar, and nonspecific for different spondylarthropathies, this chapter will focus on the most frequent seronegative spondylarthropathy. Ankylosing spondylitis or Bechterew disease is an HLA-B27 associated chronic inflammatory disease of the axial skeleton, manifested by back pain, and resulting in ankylosing of the affected joints, with progressive stiffness of the spine and sacroiliac joints. Besides stiffening of the spine and sacroiliac joints, decrease of lumbar and cervical lordosis and accentuation of thoracic kyphosis occurs, which can be compensated to a certain extent by hip extension, but may lead to loss of sagittal balance and loss of straight vision. The progressive rigidity of the spine, and the accompanying osteopenia, predispose for fractures, even caused by minor trauma.

Basic Epidemiology and Demographics

The incidence of Bechterew disease varies among different ethnic groups. The prevalence of Bechterew disease in a given population is correlated with the prevalence of human leukocyte antigen (HLA)-B27 in that group. Approximately 5–6% of HLA-B27-positive individuals suffer from ankylosing spondylitis. The mean prevalence of ankylosing spondylitis per 10,000 is about 23.8 in Europe, 31.9 in North America, 10.2 in Latin America, 16.7 in Asia, and 7.4 in Africa (Dean et al. 2014).

Relatives of an affected patient have an increased risk for ankylosing spondylitis themselves: monozygotic twins – 63%; first-degree relatives – 8.2%; second-degree relatives – 1.0%; third-degree relatives – 0.7%; and parent-child – 7.9% (Brown et al. 2000). The male-female ratio is approximately 2:1 to 3:1.

Pathophysiology

Typical clinical symptoms of ankylosing spondylitis are back pain and progressive stiffening of the spine. The sacroiliac joints are frequently involved as well, even in early stages of the disease. The primary targets of the autoimmune inflammatory processes in ankylosing spondylitis are the entheses, fibrous and fibrocartilaginous tissue, providing anchorage of ligaments and tendons to bones. The enthesitis is accompanied by small erosions of the cortical bone, and reactive subcortical osteosclerosis (osteitis) and bone reabsorption. With progression of the disease, osteoproliferation may result in ossification of ligaments, tendons, joint capsules, and eventually ankylosis. Sacroiliitis may cause erosions and later ossification of the sacroiliac joints. The spinal stiffness is caused by calcified syndesmophytes, ossifications of the paraspinous ligaments, bridging adjacent vertebral bodies. Furthermore, a range of extraskeletal manifestations, as uveitis, inflammatory bowel disease, psoriasis, cardiovascular, or pulmonary disease may occur. The most common extraarticular symptom is unilateral uveitis, affecting 25–40% of patients (Stolwijk et al. 2015). About 50% of cases with acute uveitis are associated with ankylosing spondylitis. As uveitis may be the first symptom requiring medical evaluation, it should alert the treating physician to the possibility of ankylosing spondylitis. Histological examinations of the mucosa of the ileum and colon show ulcerations in about 50–60% of cases with ankylosing spondylitis, while only relatively few of these patients develop a clinically manifest inflammatory bowel disease, whether Crohn’s disease or ulcerative colitis. Up to 10% of patients suffer from psoriasis (Stolwijk et al. 2015; El Maghraoui 2011). An additional psoriasis predisposes to higher frequency of peripheral joint involvement and a more severe course of the ankylosing spondylitis (Pérez Alamino et al. 2011).

Clinical Scenario and Indications for Imaging

Lower back pain caused by sacroiliitis is often one of the first symptoms, those may already occur in childhood or early adulthood, usually prior to 45 years of age. In most cases, involvement of the spine occurs in later stages of the disease. As there exists no single, specific diagnostic test, the diagnosis of ankylosing spondylitis, and the other seronegative spondylarthropathies, is based on the combination of clinical symptoms, physical exam, laboratory tests (HLA-B27 serostatus and C-reactive protein), and imaging. Thus, when spondylarthropathy is suspected, imaging is indicated (Schueller-Weidekamm et al. 2014; Sudoł-Szopińska et al. 2015).

MRI

MRI is the method of choice in early stages of the disease (Fig. 1). MRI is able to diagnose bone marrow edema of the sacrum and ilium adjacent to the sacroiliac joints (Rudwaleit et al. 2009). Synovial gadolinium enhancement correlates with disease activity, as measured by laboratory inflammation markers. Erosions, with loss and irregular margins of subchondral bone of the joints of the sacroiliac joints, can also be diagnosed in MRI, but occur in later stages of the disease. In spinal MRI, edema in the anterior corners of the vertebral bodies (termed “Romanus lesions”) might be seen, indicating inflammation of the entheses that might result in formation of bridging syndesmophytes (Fig. 2). The facet joints may also exhibit edema. As syndesmophytes are thin and have low signal intensity on all pulse sequences, ankylosis of the spine may only be poorly visualized on MRI. Disc-related signal intensity abnormalities, in particular hyperintensities on STIR-weighted images, of the central portion and sparing of the anterior and posterior part of the endplates, and later corresponding erosions and irregularities are referred to as “Andersson lesions.”
Fig. 1

In early stage of sacroiliitis, MRI is the most sensitive imaging modality, and thus method of choice. Adjacent to the joint, the sacrum and ilium show bone marrow edema (arrow in a) and contrast-enhancement (arrow in c). In later stages of the disease, erosions (e), sclerosis (d and e), and finally fusion of the joints (f) occur, which can be detected on plain radiography (d) and more sensitively on CT (e and f) as well. (a) Axial fat-suppressed proton density-weighted MRI. (b) Axial unenhanced T1-weighted MRI. (c) Axial gadolinium-enhanced T1-weighted MRI. (d) a.p. radiograph. (d and e) Axial CT bone window

Fig. 2

In early stage of HLA-B27 associated spondylarthritis, MRI (a, b and c) is the method of choice, able to detect enthesitis at the ventral corners of the vertebral bodies. These often subtle findings, referred to as “Romanus lesions,” are hypointense on T1-weighted (arrows in a), and hyperintense on T2-weighted (arrows in b) images, with fluid-sensitive sequences (arrows in c) being most sensitive. In MRI, “Romanus lesions” may be visible, before syndesmophytes can be detected in CT. With progression of the disease, syndesmophytes, calcification of the syndesmophytes, the interspinous ligaments and finally fusion of facet, and uncovertebral joints occur. These changes subsequently result in ankylosis of the spine, visible on radiographs (f), and more sensitively on CT (d, e, and g). As syndesmophytes and calcifications are hypointense, they can be difficult to diagnose in MRI. Severe ankylosis of the spine results in bamboo-like appearance (e to g) and predisposes to highly instable fractures, even caused by minor trauma (g). (a) Sagittal T1-weighted MRI. (b) Sagittal T2-weighted MRI. (c) Sagittal fat-saturated T2-weighted MRI. (d, e, and g) Sagittal CT reformats bone window. (f) Lateral radiograph.

Radiography and CT

Usually radiographs are recommended as the first imaging modality. Especially in young patients without relevant degeneration of the joints, radiography and even CT may not show pathological findings when initial symptoms of the sacroiliac joints occur. While radiography and even CT are still normal, MRI might already reveal pathologic findings. Radiography and CT show pathological findings in later stages of the disease. Compared to radiography, the sensitivity of CT is much higher, because of its higher spatial resolution, even small structural changes of cortical and spongious bone can be detected. Sacroiliac joints may show bilateral erosions (Fig. 3), then sclerosis occurs, which may result in the endstage of complete fusion of the joints (Fig. 4). A widespread radiographic scoring system consists of the modified New York criteria (Table 1), although already proposed in 1984 (Van der Linden et al. 1984). In relatively early stages of the disease, erosions and later adjacent sclerosis of the anterior corners of the vertebral bodies occur, designated the “shiny corners” sign on radiographs. Squaring of the vertebral bodies is relatively characteristic for ankylosing spondylitis, most frequently seen in the lumbar spine. With progression of the disease, ossified syndesmophytes, bridging adjacent vertebral bodies, may result in extended ankylosis of the spine, in the endstage with involvement of the whole spine (“bamboo spine”). Involvement of the facet and uncovertebral joints can lead to erosions and later fusion as well. During the course of the disease, osteopenia occurs and can be seen on radiographs and CT. Localized vertebral or discovertebral lesions of the spine, with erosions, and irregularities of the central portion, and sparing of the anterior and posterior aspect of the endplates, are referred to as “Andersson lesions” that regularly occur in patients with ankylosing spondylitis. On radiographs and CT these osseous erosions and irregularities of the endplates can be depicted, while MRI is more sensitive for earlier stages by detecting disc-related signal intensity abnormalities, in particular hyperintensities on STIR-weighted images.
Fig. 3

Ankylosing spondylitis. Case 1: 79-year-old woman with ankylosing spondylitis. Lateral radiograph (a) and sagittal CT (b) show squaring of the lumbar vertebral bodies. While the lumbar spine does at least show subtle syndesmophytes (a and b), the thoracic spine shows bridging syndesmophytes and hyperkyphosis (c). Furthermore, CT of the sacroiliac joints shows erosions with subtle adjacent sclerosis (c). (a) Lateral radiograph of the lumbar spine. (b) Sagittal CT reformat bone window of the lumbar spine. (c) Lateral radiograph of the thoracic spine. (d) Axial CT bone window of the sacroiliac joints

Fig. 4

Ankylosing spondylitis. Case 2: 68-year-old woman with severe ankylosing spondylitis. Spinal CT shows bridging syndesmophytes (a, b and c), fusion of the uncovertebral und facet joints (d and e), and thoracic hyperkyphosis (b). Squaring of the vertebral bodies, accentuated on the lumbar spine (c). The sacroiliac joints are fused as well (e). Note osteopenia of the whole axial skeleton (ae). (a) Sagittally reformatted postmyelography CT of the cervical spine. (b) Sagittal CT reformat of the thoracic spine. (c) Sagittal CT reformat of the lumbar spine. (d) Coronal reformat of postmyelography CT of the cervical spine. (e) Coronal CT reformat of the lumbar spine and sacroiliac joints

Table 1

Modified New York criteria for scoring of sacroiliitis

Grade

Radiographic findings

0

No abnormalities

1

Suspicious changes

2

Minimal abnormalities: Small localized areas with erosions or sclerosis without alteration in the joint width

3

Unequivocal abnormality, with one or more of the following: Erosions, evidence of sclerosis, widening, narrowing or partial ankylosis

4

Severe abnormality: Complete ankylosis

In patients with severe ankylosing of the spine, even minor trauma may cause highly instable fractures (Figs. 5 and 6). MRI can be helpful to diagnose occult fractures, which may be hard to diagnose using CT or radiography, due to osteopenia. Even if no fracture is directly seen, MRI may reveal the bone marrow edema. Furthermore, MRI is optimal for evaluation of the spinal canal and neural structures.
Fig. 5

Ankylosing spondylitis. Case 3: 50-year-old woman with severe ankylosing spondylitis and instable fracture of T9 and T10 caused by minor fall. Besides typical findings of progressed ankylosing spondylitis, like bridging syndesmophytes, fusion of facet and uncovertebral joints, osteopenia, and “bamboo spine,” sagittal CT (a) shows instable, displaced fracture of T9 and T10. This instable fracture was stabilized by internal fixation from T8 to T12, displayed on sagittal CT (b), a.p. (c) and lateral (d) radiographs. (a) Sagittal CT reformat after trauma. (b) Sagittal CT reformat bone window after surgery and dorsal stabilization with internal fixation. (c) Postoperative a.p. radiograph. (d) Postoperative lateral radiograph

Fig. 6

Ankylosing spondylitis. Case 4: 68-year-old woman with severe ankylosing spondylitis. Radiography (a) of the cervical spine prior to trauma shows progressed ankylosing spondylitis, with bridging syndesmophytes and fusion of uncovertebral and facet joints. After a minor fall, sagittal CT (b) revealed instable, displaced fracture of the cervical spine at C5/C6, besides bamboo-like appearance of the whole spine. This fracture was initially stabilized by ventral and dorsal fixation of only this segment (c and d), leading to increased biomechanical stress to the adjacent segments and after another minor trauma, finally instable fracture directly cranially adjacent to this spondylodesis (c and d). Thus, dorsal stabilization extending from craniocervical junction to mid-thoracic level was performed (e). (a) Lateral radiograph of the cervical spine. (b) Sagittal CT reformat of the whole spine bone window. (c and d) Sagittal CT reformats of cervical spine bone window. (e) Lateral localizer of intraoperatively acquired CT

Imaging Technique and Recommended Protocol

In early stages of the disease, MRI is the method of choice. The sequence protocol for the sacroiliac joints should include axial and or coronal fluid-sensitive sequences (e.g., STIR-weighted, fat-saturated T2-weighted, or fat-saturated proton density-weighted), axial and/or coronal nonenhanced and fat-saturated gadolinium-enhanced T1-weighted sequences (Fig. 7). An MRI of the sacroiliac joints should include scans of the lumbar spine as well. A sequence protocol for spinal MRI should include sagittal fluid-sensitive sequences (e.g., STIR-weighted, or fat-saturated T2-weighted), T2-weighted, nonenhanced, and fat-saturated gadolinium-enhanced T1-weighted, and axial T2-weighted and fat-saturated gadolinium-enhanced T1-weighted sequences (Schueller-Weidekamm et al. 2014; Sudoł-Szopińska et al. 2015). In Table 2, a suggestion for an MRI pulse sequence protocol is shown.
Fig. 7

Ankylosing spondylitis. Case 5: Sacroiliitis in a 13 year-old, HLA-B27 positive boy with lower back pain. While unenhanced T1-weighted MRI appears nearly normal, fat-suppressed proton density-weighted MRI shows bone marrow edema of the sacrum and ilium adjacent to the left sacroiliac joint (arrows in b). Gadolinium-enhanced T1-weighted images and subtraction of gadolinium-enhanced and nonenhanced T1-weighted images exhibit corresponding enhancement (arrows in c and d). (a) Axial unenhanced T1-weighted MRI. (b) Axial, fat-saturated proton density-weighted MRI. (c) Gadolinium-enhanced T1-weighted MRI. (d) subtraction of gadolinium-enhanced and nonenhanced T1-weighted MRI

Table 2

Suggested MRI pulse sequence protocol

Pulse sequence

Orientation

Slice thickness (mm)

STIRw

Sagittal or coronal

3

T1w

Sagittal

3

T2w

Sagittal and axial

3

Contrast-enhanced, fat-saturated T1w

Sagittal and axial

3

Interpretation Checklist and Structured Reporting

Always evaluate alignment and potential degeneration of the spine.

Radiography and CT

  • Sacroiliac joints:
    • Bilateral erosions, later sclerosis, and finally fusion of the sacroiliac joints, highly suggestive for ankylosing spondylitis

    • Osteopenia, often present but unspecific

  • Spine:
    • Erosions, later sclerosis (shiny corner sign) of the anterior corners of the vertebral bodies, highly suggestive for ankylosing spondylitis

    • Andersson lesions, disc-related erosions and irregularities of the central potion with sparing of the anterior and posterior part of the endplates

    • Squaring of vertebral bodies

    • First subtle, later bridging syndesmophytes, finally, and very specific for ankylosing spondylitis

    • In later stages fusion of facet and uncovertebral joints

    • Endstage of ankylosing of the spine (“bamboo spine”) with decreased lumbar and cervical lordosis and accentuated thoracic hyperkyphosis, and consecutive loss of sagittal balance

    • Osteopenia, often present but unspecific

MRI

  • Sacroiliac Joints:
    • Bilateral sacroiliitis highly suggestive for ankylosing spondylitis: First bone marrow edema, later erosions, and finally fusion of the joints

  • Spine:
    • If present, synovial gadolinium enhancement indicates on-going inflammatory activity

    • Contrary to CT and radiography, syndesmophytes, and fusion of facet and uncovertebral joints are difficult to diagnose in MRI

    • Andersson lesions

    • Romanus lesions

Treatment Monitoring: Follow-up Scheme and Findings/Pitfalls

In young patients with back pain, and without history of malignancy or trauma, MRI is the method of choice when ankylosing spondylitis or any other spondylarthropathy is suspected. MRI may detect pathologic findings in early stages of the disease, when radiography and even CT may still be normal. Radiography is the most important imaging technique for follow-up of patients with already confirmed ankylosing spondylitis. After trauma, CT and in certain cases additional MRI is indicated, as patients with severe ankylosing spondylitis have an increased risk for instable fracture.

Therapy

Therapeutic options for ankylosing spondylosis comprise physical therapy, medication, and surgery. There is no cure for ankylosing spondylitis, the aim of the therapy is to stop, or at least decelerate progression of the disease. Physical therapy includes exercises that are supposed to preserve the flexibility. Nonsteroidal anti-inflammatory drugs (NSAIDs) reduce pain and inflammation activity. Disease modifying antirheumatic drugs (DMARDs) are beneficial for patients with peripheral arthritis in particular, but less in axial manifestation. Tumor necrosis factor (TNF)-alpha blockers (antagonists) have been demonstrated to reduce clinical and laboratory disease activity. In patients with inadequate response to TNF alpha antagonists, Interleukin-17A inhibitors are an option.

Noninflammatory Ankylosis of the Spine

Noninflammatory conditions with ossifications of the paravertebral ligaments may result in ankylosis of the spine as well. The most frequent entities of noninflammatory ankylosis are diffuse idiopathic skeletal hyperostosis (DISH), and ossification of the posterior longitudinal ligament (OPLL), but exclusive ossification of the flava ligaments (OFL) may occur as well. Involvement of different paraspinal ligaments may be associated with each other, or with other ossifications of entheses, tendons, and ligaments. Usually elderly people are affected, and some ethnical predispositions were shown. For instance, the highest incidence of OPLL is observed in Japan. The pathophysiology remains still undetermined. Although DISH and OPLL are relatively frequent incidental findings in imaging studies, and are often asymptomatic, these conditions inherit the potential for causing symptoms. When OPLL leads to relevant spinal canal stenosis, and consecutive myelopathy, surgical decompression of the spinal cord is indicated. DISH with extended prevertebral hyperostosis may cause unspecific symptoms as well, like dysphagia.

Diffuse Idiopathic Skeletal Hyperostosis (DISH), and Ossification of the Posterior Longitudinal Ligament (OPLL)

Synonyms

Asymmetrical skeletal hyperostosis; Forestier disease; Senile ankylosing hyperostosis

Definition of Entity and Clinical Highlights

Spinal manifestation of diffuse idiopathic skeletal hyperostosis are flowing anterior vertebral ossifications of more than four contiguous levels, lack of significant degenerative disc disease, and lack of arthritis of facet and sacroiliac joints (Resnick and Niwayama 1976). The cervical spine is most often affected, followed by thoracic spine. These spinal hyperostoses are usually accentuated on the right side. DISH is most often an incidental finding in radiologic studies, but extended hyperostosis may become symptomatic.

In OPLL, flowing ossifications are localized posteriorly to multiple vertebral bodies, caused by ossification of the posterior longitudinal ligament, with relatively minimal degeneration of the vertebral discs, and without ankylosis of the facet joints. The formation of ectopic new bone can result in significant narrowing of the a.p.-diameter of the spinal canal, and thus may cause compression of the spinal cord. Most often, the cervical spine is affected, especially on mid-cervical level. Less frequent, OPLL may also occur in mid-thoracic and lumbar spine. Cervical myelopathy with spastic paresis and further progression to paralysis may occur.

Basic Epidemiology and Demographics

DISH usually occurs in middle-aged and older patients, and rarely before the age of 50. The male-female-ratio is about 2:1. The incidence differs among ethnic groups, with Caucasians being most often affected. The reported incidence varies notably.

The highest prevalence of OPLL is observed in Japan, and other Asian countries ranging from 2% to 4%, everywhere else being much lower. The male-female-ratio is about 2:1. OPLL usually affects patients older than 50 years and rarely occurs under the age of 30.

Pathophysiology

The definitive causes of DISH-related exaggerated bone proliferation are still undetermined. Associations with diabetes mellitus, dyslipidemia, hyperuricemia, osteoarthritis, rheumatoid arthritis, gout, or calcium pyrophosphate deposition disease have been postulated, and several involved genes have been identified.

The etiology of OPLL is still undetermined, as well, but mechanical stress to ligaments cells seems to be among the contributing factors, up-regulating signaling pathways to promote osteoblastic differentiation (Chen et al. 2017). Linkage of several genes has been identified (He et al. 2014; Furushima et al. 2002; Koga et al. 1998).

Clinical Scenario and Indications for Imaging

DISH often is an incidental finding on imaging studies, but extended hyperostosis may become symptomatic. For instance, large ossifications ventrally to the cervical spine may cause dysphagia, and after trauma may lead to instable fractures. If asymptomatic, no imaging studies are indicated.

In cases without relevant stenosis of the spinal canal, OPLL often is an incidental finding in asymptomatic patients. With increasing narrowing of the spinal canal, and consecutive compression of the spinal cord, OPLL may become symptomatic, due to progressive myelopathy with neurologic symptoms of tetra- or paraparesis. When OPLL is already known or in patients with clinical symptoms suggestive for spinal cord compression, MRI is the method of choice for evaluation of compressive myelopathy. When MRI is contraindicated, postmyelography CT with multiplanar reformats is an alternative. For determining the extent of ossifications and prior to surgical decompression, CT with multiplanar reformats is indicated (Figs. 8, 9, and 10).
Fig. 8

Lateral radiographs are sufficient for diagnosing DISH, with extended bridging ossifications ventrally located to the spine. DISH usually affects the cervical spine (a), while thoracic (b) and especially lumbar spine show no or at least less ossifications. On MRI, in particular on sagittal images these ossifications can easily be overlooked, as they usually have low signal intensity in all pulse sequence. Note the relatively minimal degeneration of the discs that is typical for DISH. (a) Lateral radiograph of the cervical spine. (b) Lateral radiograph of the thoracic spine. (c) Sagittal CT reformat bone window. (d) Axial CT bone window. (e) Sagittal T1-weighted MRI. (f) Axial T1-weighted MRI. (g) Sagittal T2-weighted MRI. (h) Axial T2-weighted MRI

Fig. 9

DISH. Case 1: Spinal DISH in an 81-year-old woman. Lateral radiograph reveal extended ventrally located spondylophytes, bridging more than four cervical vertebral bodies (a), without significant decrease of disc space height. On thoracic spine, these changes are typically less severe and thus may only be subtle on radiographs (b) but obvious on CT (c). On MRI (dh), the signal intensity of these flowing spondylophytes are hypointense or if present equivalent to bone marrow. Thus, on MRI DISH can be easily overlooked, especially if a saturator is positioned ventrally to the vertebral bodies. (a and b) Lateral radiograph. (c) Sagittal CT reformat bone window. (d) Sagittal T1-weighted MRI. (e) Sagittal T2-weighted MRI. (f) Sagittal STIR-weighted MRI. (g) Axial T1-weighted MRI. (h) Axial T2-weighted MRI

Fig. 10

DISH. Case 2: Spinal DISH in an 86-year-old man. CT shows flowing ossification of the anterior longitudinal ligament, bridging more than four vertebral bodies. The extended spondylophytes ventrally to C3 to C6 cause moderate displacement of the larynx. (a) Sagittal CT reformat bone window. (b and c) Axial CT bone window

Imaging Technique and Recommended Protocol

DISH

Radiography or CT is sufficient for diagnosing DISH.

OPLL
Due to its localization ventrally inside the spinal canal, OPLL is superimposed over facet joints on lateral radiographs and thus findings may be subtle and easily overlooked using radiography. Therefore, CT with multiplanar reformats is the best imaging tool for detecting the full extent of ossifications and spinal stenosis (Fig. 11). Axial CT images reveal characteristic ossifications midline or laterally deviated along the posterior surface of the vertebral bodies, so called “upside-down T” or “bowtie.” The compression of the spinal cord can be visualized by MRI and postmyelography CT, but in case of myelopathy, the intramedullary edema can only be diagnosed by MRI. OPLL typically shows low signal intensity on all pulse sequences, but extended ossifications may exhibit central bone marrow with signal-hyperintensities on T1- and T2-weighted sequences. On T2∗-weighted gradient echo sequences, the extent of ossifications and spinal stenosis can be exaggerated by susceptibility effects (Figs. 12 and 13). Although MRI reliably shows relevant spinal stenosis and is the best imaging tool to evaluate the spinal cord, CT is usually recommended prior to surgical decompression (Fig. 14).
Fig. 11

OPLL may easily be overlooked on radiography, as the flowing multilevel ossification of the posterior longitudinal ligament is superimposed by the facet joints. CT (ad) is the best imaging tool for visualizing the extent of the ossifications, but narrowing of the spinal canal and of course cord compression is best seen on MRI (eh). Cord compression and potential intramedullary edema can be detected using T2-weighted sequences. Gradient echo sequences are the most sensitive pulse sequences for detecting calcifications, but may exaggerate the extent of spinal canal stenosis. Note the “upside-down T” or “bowtie” on axial CT (b and d) and gradient echo MRI (h). (a and c) Sagittal CT reformats bone window. (b and d) Axial CT bone window. (e) Sagittal T1-weighted MRI. (f) Axial T2-weighted MRI. (g) Sagittal T2-weighted MRI. (h) Axial T2∗-weighted sequence

Fig. 12

OPLL Case 1: OPLL in a 42-year-old male. In MRI (ac) narrowing of the spinal canal and compression of the spinal cord is clearly displayed, but the flowing multilevel ossification of the posterior longitudinal ligament has low signal intensity on all pulse sequences and thus is better visualized on CT (dg). CT is usually needed for surgical planning. As the extent of OPLL caused symptomatic myelopathy, surgical decompression by laminectomy C3 to C6 and implantation of dorsal spondylodesis was performed. (a) Sagittal T1-weighted MRI. (b) Sagittal T2-weighted MRI. (c) Axial T2-weighted MRI. (d) Sagittal CT reformat bone window. (e) Axial CT bone window. (f) Sagittal CT reformat bone window. (g) Axial CT bone window. (h) Lateral radiograph. (ae) prior and (fh) postsurgery

Fig. 13

OPLL Case 2: OPLL in a 65-year-old male. In MRI (ae) the flowing multilevel ossification of the posterior longitudinal ligament shows low signal intensity on all pulse sequences. Compression of the spinal cord and potential intramedullary edema are best displayed on T2-weighted images. Gradient echo sequences are the most sensitive MRI pulse sequences for detection of calcifications. However, when using MRI, gradient echo sequences (e) can visualize the ossifications best, but may exaggerate the extent of spinal canal stenosis. CT better shows the exact extent of ossifications than MRI. (a) Sagittal T1-weighted MRI. (b) Sagittal T2-weighted MRI. (c) Axial T1-weighted MRI. (d) Axial T2-weighted MRI. (e) Axial gradient echo sequence. (f) Sagittal CT reformat bone window. (g) Axial CT bone window

Fig. 14

OPLL Case 3: OPLL in a 69-year-old male. OPLL usually shows low signal intensity on all MRI pulse sequences. Spinal canal stenosis, cord compression, and potential intramedullary edema. As CT shows the exact extent of the ossifications, it is recommended prior to surgical decompression of symptomatic OPLL. Due to OPLL with high grade narrowing of the spinal canal and symptomatic myelopathy, dorsal decompression surgery was performed, with stabilization by implantation of an internal fixation. (a) Sagittal T1-weighted MRI. (b) Sagittal T2-weighted MRI. (c) Sagittal STIR-weighted MRI. (d) Axial T2-weighted MRI. (e) Axial gradient echo sequence. (f) Sagittal CT reformat bone window. (g) Axial CT bone window. (h) Sagittal CT reformat bone window. (i) Axial CT bone window. (j) Lateral radiograph. (ag) prior to and (hj) after dorsal decompression surgery

Interpretation Checklist and Structured Reporting

DISH
  • Radiography:
    • Sufficient for diagnosis

    • Flowing ossifications ventrally located at the vertebral bodies

    • Relatively minimal degeneration of the discs

  • CT:
    • Flowing ossifications ventrally located at the vertebral bodies

    • Axial CT shows right-lateral accentuation of these ossifications

    • Indicated for complications, for example, after trauma

  • MRI:
    • Not necessary for diagnosis, but indicated for imaging of complications or trauma

    • Early, small ossifications have low signal intensity in all pulse sequences

    • Extended, bulky ossifications may contain bone marrow, with bone marrow-equivalent signal intensity (increased signal intensity on T1- and T2-weighted sequences)

OPLL
  • Radiography:
    • Flowing multilevel ossification posterior to vertebral bodies: may easily be overlooked, as these OPLL is superimposed by facet joints on lateral radiographs, and vertebral bodies, laminae, and spinal processes on a.p. radiographs.

    • No ankylosis of facet joints.

    • Relatively minimal degeneration of intervertebral discs.

  • CT:
    • Axial images: “upside down T” or “bowtie”

    • Sagittal reformats: flowing multilevel ossifications of the posterior longitudinal ligament

  • Postmyelography CT:
    • Compression of the spinal cord can be better visualized than on plain CT scans

  • MRI:
    • Flowing multilevel ossification posterior to vertebral bodies

    • “upside down T” or “bowtie” on axial image

    • Ossifications have low signal intensity on all pulse sequences, but extended ossifications may contain bone marrow, displaying fat equivalent high signal intensity on T1- and T2-weighted sequences

    • Intramedullary edema on T2-weighted images

    • CAVE: the extent of spinal canal stenosis may be exaggerated on T2∗-weighted gradient echo sequences

Treatment Monitoring: Follow-up Scheme and Findings/Pitfalls

Incidental OPLL can be observed, as long as still asymptomatic. In case of high grade spinal canal stenosis or symptomatic compressive myelopathy, decompressive surgery is indicated. Two surgical strategies are possible, anterior decompression with fusion and posterior decompression with laminectomy.

Spondylodiscitis (Pyogenic Vertebral Osteomyelitis)

The risk for spondylodiscitis is increased in elderly, immobile, and immune-suppressed patients, in particular after trauma or surgery. As clinical symptoms and laboratory tests are unspecific, imaging plays a pivotal role in its diagnosis. In certain cases, image-guided acquisition of tissue for microbiological tests may be useful for diagnosis, targeted antibiotic therapy, or evaluation of suspicious imaging findings. In early stages of the disease, radiography and CT may be normal, or at least unspecific, in case of preexisting degenerative conditions. MRI is the imaging method of choice, as it shows pathological findings even in early stages of the disease and potential involvement of the paraspinal tissue (Fig. 15). CT and radiography are helpful for evaluating spinal stability. Thus, clinical neuroradiology is essential in the diagnostic workup of suspected or confirmed spondylodiscitis.
Fig. 15

Spondylodiscitis. Radiographs (a and c) show decreased height of disc space and erosions of adjacent vertebral endplates. Besides osseous structures, the paraspinal soft tissue has to be regarded carefully as well, as for example abscess with gas inclusions in the psoas muscle may be detected on radiographs as well (c). In MRI the affected discs show hyperintense signal intensity on fat-saturated T2-weighted (b and g), and STIR-weighted (d) images. CT (e and f) shows reduced disc space height, and erosions of the adjacent vertebral endplates (e), and contrast-enhanced CT may detect inflammation of paraspinal and epidural soft-tissue (f) as well. Inflammation of the paraspinal soft tissue is best visualized on gadolinium-enhanced fat-saturated T1-weighted images (i). (a and c) a.p. radiograph. (b) Sagittal T2-weighted fat-saturated MRI. (d) Coronal STIR-weighted MRI. (e) Sagittal CT reformat bone window. (f) Sagittal CT reformat soft tissue window. (g) Sagittal T2-weighted fat-saturated MRI. (h) Sagittal T1-weighted MRI. (i) Axial gadolinium-enhanced T1-weighted fat-saturated MRI

Definition of Entity and Clinical Highlights

Spondylodiscitis (Synonym: pyogenic vertebral osteomyelitis) comprises pyogenic infection of the vertebral disc (discitis) and adjacent vertebral bodies (spondylitis).

Basic Epidemiology and Demographics

The overall incidence is about 2.4 cases per 100,000. The incidence increases with age. In persons older than 70 years, the incidence is about 6.5 per 100,000, in persons younger than 20 years only about 0.3 per 100,000 (Zimmerli 2010). A predominance of males is observed, with an approximate ratio of 2:1.

Pathophysiology

Most often it is caused by hematogenous seeding, followed by direct inoculation in spinal surgery, or contiguous spread of an infection of adjacent tissue. Staphylococcus aureus is by far the most common microorganism being responsible for about 50% of cases, followed by Escherichia coli, Pseudomonas aeruginosa, and Candida albicans.

In adults, the vertebral discs are rather avascular, in children they are relatively well vascularized (Ratcliffe 1985). Thus, primary hematogenous discitis only occurs in children, while in adults spondylitis leads to secondary infection of adjacent disc.

Clinical Scenario and Indications for Imaging

Clinical signs are various and unspecific. Local pain is the most common initial sign (86%). Severe, lancinating pain may indicate an epidural abscess. While the most common location of vertebral osteomyelitis is lumbar (58%), followed by thoracic (30%) and cervical (11%) spine, the most common locations of epidural abscesses are inverse, with cervical (28%) spine followed by thoracic (22%) and lumbar (12%) spine. Fever might be present, but is an infrequent sign (35–60%), probably due to antipyretic effect of commonly used nonsteroidal anti-inflammatory drugs (NSAIDs). Neurologic deficits like hypaesthesia, paresis, radiculopathy occur in about one third of cases. Spinal tenderness on percussion is reported in less than one fifth of cases (Zimmerli 2010).

Imaging Technique and Recommended Protocol

Due to its superior soft-tissue-contrast, the imaging method of choice is MRI, being able to detect early stages of the disease. MRI is very sensitive in detecting inflammation of the vertebral disc and body, epi- and paraspinal tissue, when radiography and even CT may still be without pathological finding (Fig. 16). Affected vertebral discs and bodies show hyperintensity on T2-weighted and hypointensities on T1-weighted images (Fig. 17). A suggestion for an MRI pulse sequence protocol is shown in Table 3, including sagittal T1 weighted, sagittal or coronal STIR-weighted, sagittal T2 weighted, sagittal and axial gadolinium-enhanced T1 weighted fat-saturated sequences (Prodi et al. 2016).
Fig. 16

Spondylodiscitis. Case 1. 80-year-old male with spondylodiscitis C6/7. Initial CT already shows erosions of adjacent vertebral endplates (a) and epidural empyema (b). In MRI the disc and adjacent vertebral bodies have hyperintense signal intensity on T2-weighted images (c and d). The epidural empyema is visible on T2-weighted images (c and d), but the extent of paravertebral soft tissue involvement is best seen on gadolinium-enhanced, fat-saturated T1-weighted images (f). Despite initiation of antibiotic therapy, the clinical status of the patient worsened, and MRI performed 8 days later (gj) showed progression of both, osseous and soft tissue involvement as well. (a) Sagittal CT reformat bone window. (b) Sagittal contrast-enhanced CT reformat soft tissue window. (c) Sagittal T2-weighted fat-saturated MRI. (d) Axial T2-weighted MRI. (e) Sagittal T1-weighted MRI. (f) Sagittal gadolinium-enhanced T1-weighted fat-saturated MRI. (g) Sagittal T2-weighted fat-saturated MRI. (h) Axial T2-weighted MRI. (i) Sagittal T1-weighted MRI. (j) Sagittal gadolinium-enhanced T1-weighted fat-saturated MRI. (cf) Initial MRI. (gj) Second MRI after clinical worsening despite antibiotic therapy

Fig. 17

Spondylodiscitis. Case 2. 37-year-old male with thoracic pain and elevated inflammatory markers. Initially performed radiography showed decreased height of the disc space and subtle erosions T8/9 (arrows in a). MRI revealed the whole extent of the inflammatory process, involving disc space, vertebral bodies, and surrounding epidural and paravertebral tissue. The destruction of the adjacent vertebral endplates is best detected with CT. (a) a.p. radiograph. (b) Sagittal fat-saturated T2-weighted MRI. (c) Sagittal T2-weighted MRI. (d) Axial T2-weighted MRI. (e) Sagittal T1-weighted MRI. (f) Sagittal gadolinium-enhanced T1-weighted MRI. (g) Axial gadolinium-enhanced fat-saturated T1-weighted MRI. (h) Sagittal CT reformat

Table 3

Suggested MRI pulse sequence protocol

Pulse sequence

Orientation

Slice thickness (mm)

STIRw

Sagittal or coronal

3

T1w

Sagittal

3

T2w

Sagittal and axial

3

Contrast-enhanced, fat-saturated T1w

Sagittal and axial

3

MRI delivers criteria with very high and high sensitivity for pyogenic osteomyelitis, including the presence of paraspinal or epidural inflammation (97.7% sensitivity), disc enhancement (95.4% sensitivity), hyperintensity or fluid-equivalent disc signal intensity on T2-weighted MR images (93.2% sensitivity), erosion or destruction of vertebral endplates (84.1% sensitivity), and effacement of the nuclear cleft (83.3% sensitivity) (Fig. 18). Criteria with low sensitivity are decreased height of the intervertebral space (52.3% sensitivity) and disc hypointensity on T1-weighted MRI (29.5% sensitivity) (Ledermann et al. 2003).
Fig. 18

Spondylodiscitis. Case 3. 81-year-old male with lumbar pain and elevated inflammatory markers. Besides spondylodiscitis T12/L1, with T2-hyerintense liquefaction of the disc space (b and c), and destruction of the vertebral bodies, MRI revealed bilateral abscesses in the psoas muscle (c and d). The destruction of the vertebral bodies is demonstrated by CT (e and f). (a) Sagittal T1-weighted MRI. (b) Sagittal T2-weighted MRI. (c) Coronal STIR-weighted MRI. (d) Axial T2-weighted MRI. (e) Sagittal CT reformat bone window. (f) Axial CT bone window. (g) Sagittal CT reformat soft tissue window. (h) Axial CT soft tissue window

As in the first 2 weeks, osseous signs are negative on radiographs, radiography is inappropriate as initial imaging method. In early stages only abnormalities of paravertebral soft tissue might be detected, like changes in paraspinal soft tissue density, or loss of fat planes. Reduced height of the vertebral disc, and erosions of vertebral endplates, destruction of the vertebral body, fusion of vertebral body, sclerosis, and deformity only occur in later stages and cannot be detected in the first 2 weeks (Byrd et al. 1983; Pineda et al. 2009).

CT inherits a higher spatial resolution compared to radiography, and thus even subtle erosions of the vertebral endplates can be detected, which might not be seen on plain radiography, but might be difficult to differentiate from degenerative conditions in early stages of the disease. Inflammation of the epidural and paraspinal soft tissue can be detected, with increased sensitivity in contrast-enhanced CT.

Magnetic resonance imaging is most sensitive for diagnosing vertebral osteomyelitis in early stages of the disease, thus being the imaging modality of choice (Carragee 1997; Prodi et al. 2016). In Table 3, a suggestion for an MRI pulse sequence protocol is shown.

Interpretation Checklist and Structured Reporting

Always evaluate spinal alignment, height of vertebral bodies and disc space, lining of endplates, integrity of vertebral bodies, and of course paraspinal soft tissue.

MRI

  • Method of choice, due to its superior soft tissue contrast

  • Disc hyperintense on T2- and STIR-weighted images

  • Narrowing of disc space

  • Diffuse gadolinium-enhancement of vertebral disc

  • Edema, later osseous erosions and destruction of adjacent vertebral bodies/ endplates

  • Diffuse gadolinium-enhancement of adjacent vertebral bodies

  • Paraspinal and epidural abscess or phlegmon (Fig. 19)

  • Compression of spinal cord possible

Fig. 19

Spondylodiscitis. Case 4. 82-year-old female sepsis caused by Staphylococcus aureus. As lancinating pain of the spine and progressive tetraparesis occurred, spinal MRI was performed. Spondylodiscitis L4/5 and L5/S1 (arrows in b and d) was diagnosed, accompanied by epidural empyema extending from cervical to lumbar level. (a and b) Sagittal T2-weighted MRI. (c and d) Sagittal gadolinium-enhanced fat-saturated T1-weighted MRI. (e and f) Axial gadolinium-enhanced fat-saturated T1-weighted MRI

CT

  • Iso- to hypodense swelling of paraspinal soft tissue

  • Gas inclusions in paraspinal soft tissue

  • Contrast-enhancement of affected disc, vertebral body, and adjacent paravertebral soft tissue

  • Reduced disc height

  • Erosions or sclerosis of vertebral endplates, and later destruction of vertebral bodies, potential bony sequestra

  • Potential spinal deformity in late course of the disease

Radiography

  • No pathologic findings within the first weeks

  • Osteolysis of endplates and later vertebral bodies, osteosclerosis

  • Changes of paraspinal soft tissue density, for example, by edema or gas inclusions

  • Late stage of the disease: spinal deformity

Treatment Monitoring: Follow-up Scheme and Findings/Pitfalls

Pathological MRI findings in spondylodiscitis remain present for months or may even increase despite sufficient antibiotic therapy. In particular, compared with baseline MRI, follow-up MRI more frequently show decreased vertebral body height. Less frequently epidural enhancement, epidural canal empyema, and epidural canal compromise are observed in routine follow-up MRI. But gadolinium-enhancement of vertebral body and disc space and bone marrow edema often appear equivocal or even worse compared with baseline MRI (Kowalski et al. 2007; Baxi et al. 2012). Thus, a lack of correlation between clinical follow-up status and MRI can be observed (Baxi et al. 2012). At least, clinical worsening is never associated with improvement of MRI. Consequently, routine follow-up is of limited value, but repetition of MRI is always indicated, if the clinical status deteriorates.

Differential Diagnosis

Tuberculous Spondylodiscitis

The differentiation of pyogenic from tuberculous spondylodiscitis is often difficult by clinical, as well as radiological means, but is crucial for choosing the appropriate therapy, as tuberculosis requires antituberculous medication to reduce morbidity. Although pyogenic and tuberculous spondylodiscitis often show differences in clinical presentation, laboratory, and imaging studies (Table 4), the specificity of all these parameters may be relatively low. The clinical onset of symptoms usually is acute in pyogenic, versus chronic, or prolongated in tuberculous spondylodiscitis. While clinical laboratory markers, particularly C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and leucocyte count, are highly elevated in pyogenic spondylodiscitis, they are relatively low, or only moderately elevated in most cases of tuberculous spondylodiscitis. Frequent differences between both entities can be observed, concerning number of involved segments, involvement of intervertebral discs, and paravertebral soft tissue. In pyogenic spondylodiscitis, usually only one segment is affected, the intervertebral disc is involved early in the temporal course, and paravertebral involvement reveals rather small epidural abscesses, in contrary to tuberculous spondylodiscitis, often affecting several segments, late involvement of the intervertebral disc, and extended involvement of the paravertebral soft tissue, with often large, and frequently calcified paraspinal abscesses (Fig. 20). In cases, when clinical symptoms, inflammation markers, blood cultures, and radiological studies are inconclusive, image-guided biopsy may be necessary to gain tissue for histological, or even molecular examinations.
Table 4

Clinical, laboratory marker, and imaging differences of pyogenic versus tuberculous spondylodiscitis

 

Pyogenic

Tuberculous

Symptoms

Acute

Chronic

Clinical laboratory markers of inflammation

Highly elevated

Relatively low, or only moderately elevated

Involved segments

Usually monosegmental

Often polysegmental

Involvement of intervertebral disc

Early

Late

Paravertebral involvement

Rather small epidural abscesses

Often large paraspinal abscesses, frequently calcified

Fig. 20

Tuberculous spondylodiscitis with large psoas abscesses. 36-year-old African male with paraplegia. Polysegmental involvement, (ac), large abscesses in the paravertebral soft tissue and the psoas muscles (df) already were suggestive for tuberculous spondylodiscitis. CT-guided drainage of the psoas abscess (g) confirmed the diagnosis. (a) Sagittal T2-weighted MRI. (b) Sagittal T1 weighted MRI. (c) Sagittal Gadolinium-enhanced fat-saturated T1-weighted MRI. (d) Axial T2-weighted MRI. (e) Axial Gadolinium-enhanced T1-weighted MRI. (f) Coronal Gadolinium-enhanced fat-saturated T1-weighted MRI. (g) CT-guided drainage of psoas abscess

Differential Diagnostic Considerations in Imaging of Inflammatory Diseases of the Spine

The two major categories of inflammatory disorders of the spine are infectious and noninfectious. Infectious disease can be located in the spine itself, or the surrounding structures. Hematogenous seeding is the most common cause of pyogenic and nonpyogenic infection, but direct inoculation is possible as well. Noninfectious inflammatory conditions of the spine are systemic diseases that can be divided into two main categories, rheumatoid disorders including rheumatoid arthritis, and seronegative spondylarthropathies.

Clinical, laboratory, and also imaging findings in infectious and noninfectious, and even noninflammatory, degenerative diseases of the spine may overlap. Thus, clinical neuroradiology may not lead to a definitive diagnosis, but to certain imaging patterns, and possible differential diagnoses.

Seronegative spondylarthropathies can be classified in ankylosing spondylitis (Bechterew disease), the most frequent seronegative spondylarthropathy, or as psoriatic spondylarthritis, reactive arthritis (formerly termed Reiter syndrome), enteropathic spondylarthritis (associated with inflammatory bowel disease), and undifferentiated spondylarthritis. Imaging findings of different types of seronegative spondylarthropathies are similar and often nonspecific. Erosions, later sclerosis of the anterior corners of the vertebral bodies, known as “shiny corner” sign, and T2-hyperintensities of the anterior corners of the vertebral bodies, so-called “Romanus lesions” are highly suggestive for ankylosing spondylitis. “Andersson lesions,” disc-related erosions and irregularities of the central potion with sparing of the anterior and posterior part of the endplates, frequently occur in other seronegative spondylarthropathies as well. Squaring of vertebral bodies can be observed frequently, but is also unspecific. First subtle, later bridging syndesmophytes, are relatively specific for ankylosing spondylitis. In later stages of ankylosing spondylitis, fusion of facet and uncovertebral joints occurs, with the endstage of ankylosing of the whole spine (“bamboo spine”) with decreased lumbar and cervical lordosis and accentuated thoracic hyperkyphosis, and consecutive loss of sagittal balance. Osteopenia is often present in ankylosing spondylitis, but is very unspecific.

Ossifications of the paravertebral ligaments, resulting in ankylosis of the spine, may occur in noninflammatory conditions as well, like diffuse idiopathic skeletal hyperostosis (DISH), ossification of the posterior longitudinal ligament (OPLL), or ossification of the flava ligament (OFL).

Imaging differential diagnoses of spondylodiscitis/pyogenic vertebral osteomyelitis include activated osteochondrosis, and tuberculous spondylitis. In spondylodiscitis, MRI is the method of choice, due to its superior soft tissue contrast. While narrowing of disc space, edema, later osseous erosions and destruction of adjacent vertebral bodies/ endplates, and diffuse gadolinium-enhancement of adjacent vertebral bodies may occur in degenerative conditions, like activated osteochondrosis as well, disc hyperintensities on T2- and STIR-weighted images, diffuse gadolinium-enhancement of vertebral discs, and paraspinal and epidural abscess or phlegmon are highly sensitive for spondylodiscitis.

In pyogenic vertebral osteomyelitis, usually only one segment is affected, the intervertebral disc is involved early in the temporal course, and paravertebral involvement reveals rather small epidural abscesses, in contrary to tuberculous spondylodiscitis, often affecting several segments, late involvement of the intervertebral disc, and extended involvement of the paravertebral soft tissue, with often large, and frequently calcified paraspinal abscesses.

Imaging findings in spondylodiscitis remain positive even months after sufficient antibiotic therapy, but clinical worsening is never associated with improvement of MRI. Consequently, routine follow-up in pyogenic vertebral osteomyelitis is of limited value, but repetition of MRI is always indicated, if the clinical status deteriorates.

Thus, clinical neuroradiology alone may not be capable to lead to a definitive diagnosis, but certain imaging patterns may contribute to possible differential diagnoses. Therefore, regarding clinical and laboratory findings, as well, is crucial for appropriate interpretation of imaging studies.

References

  1. Baxi S, Malani PN, Gomez-Hassan D, Cinti SK. Association between follow-up magnetic resonance imaging and clinical status among patients with spinal infections. Infect Dis Clin Pract (Baltim Md). 2012;20(5):326–9.CrossRefGoogle Scholar
  2. Brown MA, Laval SH, Brophy S, Calin A. Recurrence risk modelling of the genetic susceptibility to ankylosing spondylitis. Ann Rheum Dis. 2000;59(11):883–6.CrossRefGoogle Scholar
  3. Byrd SE, Biggers SL, Locke GE. The radiographic evaluation of infections of the spine. J Natl Med Assoc. 1983;75(10):969–77.PubMedPubMedCentralGoogle Scholar
  4. Carragee EJ. The clinical use of magnetic resonance imaging in pyogenic vertebral osteomyelitis. Spine (Phila Pa 1976). 1997;22(7):780–5.CrossRefGoogle Scholar
  5. Chen D, Chen Y, Li T, Shi L, Pan M, Chen D. The role of Cx43-mediated NFкB signaling pathway in ossification of posterior longitudinal ligament: an in vivo and in vitro study. Spine (Phila Pa 1976). 2017;  https://doi.org/10.1097/BRS.0000000000002165. [Epub ahead of print]CrossRefGoogle Scholar
  6. Dean LE, Jones GT, MacDonald AG, Downham C, Sturrock RD, Macfarlane GJ. Global prevalence of ankylosing spondylitis. Rheumatology (Oxford). 2014;53(4):650–7.  https://doi.org/10.1093/rheumatology/ket387.. Epub 2013 Dec 9CrossRefGoogle Scholar
  7. El Maghraoui A. Extra-articular manifestations of ankylosing spondylitis: prevalence, characteristics and therapeutic implications. Eur J Intern Med. 2011;22:554.CrossRefGoogle Scholar
  8. Furushima K, Shimo-Onoda K, Maeda S, Nobukuni T, Ikari K, Koga H, Komiya S, Nakajima T, Harata S, Inoue I. Large-scale screening for candidate genes of ossification of the posterior longitudinal ligament of the spine. J Bone Miner Res. 2002;17(1):128–37.CrossRefGoogle Scholar
  9. He H, Mao L, Xu P, Xi Y, Xu N, Xue M, Yu J, Ye X. Ossification of the posterior longitudinal ligament related genes identification using microarray gene expression profiling and bioinformatics analysis. Gene. 2014;533(2):515–9.  https://doi.org/10.1016/j.gene.2013.09.001.. Epub 2013 Sep 18CrossRefPubMedGoogle Scholar
  10. Koga H, Sakou T, Taketomi E, Hayashi K, Numasawa T, Harata S, Yone K, Matsunaga S, Otterud B, Inoue I, Leppert M. Genetic mapping of ossification of the posterior longitudinal ligament of the spine. Am J Hum Genet. 1998;62(6):1460–7.CrossRefGoogle Scholar
  11. Kowalski TJ, Layton KF, Berbari EF, Steckelberg JM, Huddleston PM, Wald JT, Osmon DR. Follow-up MR imaging in patients with pyogenic spine infections: lack of correlation with clinical features. AJNR Am J Neuroradiol. 2007;28(4):693–9.PubMedGoogle Scholar
  12. Ledermann HP, Schweitzer ME, Morrison WB, Carrino JA. MR imaging findings in spinal infections: rules or myths? Radiology. 2003;228(2):506–14.. Epub 2003 Jun 11CrossRefGoogle Scholar
  13. Pérez Alamino R, Maldonado Cocco JA, Citera G, et al. Differential features between primary ankylosing spondylitis and spondylitis associated with psoriasis and inflammatory bowel disease. J Rheumatol. 2011;38:1656.CrossRefGoogle Scholar
  14. Pineda C, Espinosa R, Pena A. Radiographic imaging in osteomyelitis: the role of plain radiography, computed tomography, ultrasonography, magnetic resonance imaging, and scintigraphy. Semin Plast Surg. 2009;23(2):80–9.  https://doi.org/10.1055/s-0029-1214160.CrossRefPubMedPubMedCentralGoogle Scholar
  15. Ratcliffe JF. Anatomic basis for the pathogenesis and radiologic features of vertebral osteomyelitis and its differentiation from childhood discitis. A microarteriographic investigation. Acta Radiol Diagn (Stockh). 1985;26(2):137–43.CrossRefGoogle Scholar
  16. Resnick D, Niwayama G. Radiographic and pathologic features of spinal involvement in diffuse idiopathic skeletal hyperostosis (DISH). Radiology. 1976;119(3):559–68.CrossRefGoogle Scholar
  17. Rudwaleit M, Jurik AG, Hermann KG, Landewé R, van der Heijde D, Baraliakos X, Marzo-Ortega H, Ostergaard M, Braun J, Sieper J. Defining active sacroiliitis on magnetic resonance imaging (MRI) for classification of axial spondyloarthritis: a consensual approach by the ASAS/OMERACT MRI group. Ann Rheum Dis. 2009;68(10):1520–7.  https://doi.org/10.1136/ard.2009.110767.. Epub 2009 May 18CrossRefPubMedGoogle Scholar
  18. Schueller-Weidekamm C, Mascarenhas VV, Sudol-Szopinska I, Boutry N, Plagou A, Klauser A, Wick M, Platzgummer H, Jans L, Mester A, Kainberger F, Aström G, Guglielmi G, Eshed I. Imaging and interpretation of axial spondylarthritis: the radiologist’s perspective–consensus of the arthritis subcommittee of the ESSR. Semin Musculoskelet Radiol. 2014;18(3):265–79.  https://doi.org/10.1055/s-0034-1375569.. Epub 2014 Jun 4CrossRefPubMedGoogle Scholar
  19. Stolwijk C, van Tubergen A, Castillo-Ortiz JD, Boonen A. Prevalence of extra-articular manifestations in patients with ankylosing spondylitis: a systematic review and meta-analysis. Ann Rheum Dis. 2015;74(1):65–73.  https://doi.org/10.1136/annrheumdis-2013-203582.. Epub 2013 Sep 2CrossRefPubMedGoogle Scholar
  20. Sudoł-Szopińska I, Jurik AG, Eshed I, Lennart J, Grainger A, Østergaard M, Klauser A, Cotten A, Wick MC, Maas M, Miese F, Egund N, Boutry N, Rupreht M, Reijnierse M, Oei EH, Meier R, O'Connor P, Feydy A, Mascarenhas V, Plagou A, Simoni P, Platzgummer H, Rennie WJ, Mester A, Teh J, Robinson P, Guglielmi G, Åström G, Schueller-Weiderkamm C. Recommendations of the ESSR arthritis subcommittee for the use of magnetic resonance imaging in musculoskeletal rheumatic diseases. Semin Musculoskelet Radiol. 2015;19(4):396–411.  https://doi.org/10.1055/s-0035-1564696.. Epub 2015 Nov 19CrossRefPubMedGoogle Scholar
  21. van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum. 1984;27(4):361–8.CrossRefGoogle Scholar
  22. Zimmerli W. Clinical practice. Vertebral osteomyelitis. N Engl J Med. 2010;362(11):1022–9.  https://doi.org/10.1056/NEJMcp0910753.. ReviewCrossRefPubMedGoogle Scholar

Further Readings

  1. Braun J, Bollow M, Sieper J. Radiologic diagnosis and pathology of the spondyloarthropathies. Rheum Dis Clin N Am. 1998;24(4):697–735.. ReviewCrossRefGoogle Scholar
  2. Ehara S, Shimamura T, Nakamura R, Yamazaki K. Paravertebral ligamentous ossification: DISH, OPLL and OLF. Eur J Radiol. 1998;27(3):196–205.CrossRefGoogle Scholar
  3. Grigoryan M, Roemer FW, Mohr A, Genant HK. Imaging in spondyloarthropathies. Curr Rheumatol Rep. 2004;6(2):102–9.CrossRefGoogle Scholar
  4. Hong SH, Choi JY, Lee JW, Kim NR, Choi JA, Kang HS. MR imaging assessment of the spine: infection or an imitation? Radiographics. 2009;29(2):599–612.  https://doi.org/10.1148/rg.292085137.CrossRefPubMedGoogle Scholar
  5. Nascimento FA, Gatto LA, Lages RO, Neto HM, Demartini Z, Koppe GL. Diffuse idiopathic skeletal hyperostosis: a review. Surg Neurol Int. 2014;5(Suppl 3):S122–5.  https://doi.org/10.4103/2152-7806.130675. eCollection 2014CrossRefPubMedPubMedCentralGoogle Scholar
  6. Paparo F, Revelli M, Semprini A, Camellino D, Garlaschi A, Cimmino MA, Rollandi GA, Leone A. Seronegative spondyloarthropathies: what radiologists should know. Radiol Med. 2014;119(3):156–63.  https://doi.org/10.1007/s11547-013-0316-5.. Epub 2013 Nov 22CrossRefPubMedGoogle Scholar
  7. Prodi E, Grassi R, Iacobellis F, Cianfoni A. Imaging in Spondylodiskitis. Magn Reson Imaging Clin N Am. 2016;24(3):581–600.  https://doi.org/10.1016/j.mric.2016.04.005.CrossRefPubMedGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of NeuroradiologyUniversity of HeidelbergHeidelbergGermany
  2. 2.Institute of Diagnostic and Interventional RadiologyUniversity Medical Center RostockRostockGermany

Section editors and affiliations

  • Mario Muto
    • 1
  1. 1.Neuroradiology DepartmentCardarelli HospitalNaplesItaly

Personalised recommendations