High-resolution multislice CT is more sensitive for demonstrating a vacuum phenomenon, disk calcification, subchondral osteosclerosis, and associated bony changes, particularly on sagittal, para-axial, and coronal reformations (Fig. 7.6). Herniations including diffuse bulging disk are clearly demonstrated on axial scans and sagittal reformations (Fig. 7.7).

Anterior and lateral osteophytes do not compress nerve roots in the lumbar spine; large lateral osteophytes may be associated with lumbar scoliosis (Fig. 7.6). Axial reformats demonstrate accurately facet joint osteoarthrosis with disk space narrowing, sclerosis, and osteophyte formation (Fig. 7.8).

Spondylosis deformans with degenerative lumbar scoliosis leads to transverse vertebral displacement, spondylolisthesis, retrolisthesis, and anterior and anterolateral protrusion of disk material and osteophytes (Fig. 7.9).

Disk calcifications of degenerative origin are mainly located within the anulus fibrosus. Lumbar intervertebral disk calcifications are noted in 50 % of elderly patients [14].

Schmorl nodes appear as a round radiolucent lesion, sometimes containing gas, with a rim of bony sclerosis on CT (Fig. 7.10).

MRI is the modality of choice for the evaluation of the degenerative spine, as it allows an analysis of the disk, bone marrow, and facet changes as well as structures that might be injured secondary to degenerative changes such as nerve roots and muscles. Routine MRI of the spine includes sagittal spin-echo T1 and fast-spin-echo T2-weighted imaging. T2 (sagittal and/or coronal) sequences with fat saturation (STIR, fatsat) are useful for the visualization of bone marrow edema. Axial T1- and T2-weighted images are useful for visualization of nerve root compression in patients with radicular pain. The use of intravenous gadolinium may demonstrate enhancement related to neovascularization involving the intervertebral disk and/or subchondral bone.

On T1- and T2-weighted MRI, the signal of the normal intervertebral disk is, respectively, lower and higher than that of the vertebral body; the high signal is related to bounded water by the proteoglycans of the nucleus pulposus. According to Pfirrmann et al. [15], five grades can be described for lumbar disk degeneration on T2-weighted MRI (Fig. 7.11). In normal young patients, the high signal on T2-weighted images appears homogeneous in the central area of the disk; the peripheral anulus may demonstrate a low signal (grade 1) (Figs. 7.11, 7.12, 7.13, and 7.14). Loss of signal intensity in the nucleus pulposus on T2-weighted MRI closely correlates with disk dehydration related to alteration of proteoglycans. During the second decade of life, a horizontal linear hypointense band appears within the nucleus pulposus on T2-weighted images as a result of the development of collagen fibers within the nucleus pulposus (grade 2). Later, a diffuse signal loss is noted on T2-weighted images and is associated with mild narrowing of the intervertebral space (grade 3) (Figs. 7.11, 7.12, 7.13, and 7.14). At this stage, posterior radial tears may be detected as an area of high signal intensity on T2-weighted images (sometimes described as a HIZ or high-intensity zone lesion) in the posterior and peripheral anulus; enhancement is possible after intravenous administration of gadolinium (Fig. 7.13). A black disk with significant narrowing of the space (grade 4) or with a collapsed disk space (grade 5) corresponds to severe disk degeneration.

Neovascularization is often observed within the degenerative intervertebral disk and leads to a band-like enhancement parallel to the end plates or, less commonly, in the center of the disk. Such findings are associated with local back pain [16] (Fig. 7.14).

Intradiscal gas appears hypointense on all sequences, and the vacuum phenomenon is therefore more effectively detected on radiographs and CT. Intradiscal calcifications can have various presentations on MRI: they may appear hypointense on T1- and T2-weighted images or hyperintense on T1-weighted images due to the presence of fatty marrow within ossification of the disk [17, 18].

End plate bone marrow is frequently abnormal on MRI in the setting of DDD. According to Modic et al. [19, 20], two types of signal intensity changes involving the bone marrow of the adjacent vertebral body may be associated with DDD. Type 1 changes are visualized as low signal intensity on T1-weighted images and high signal intensity on T2-weighted images with enhancement after administration of gadolinium (Fig. 7.15). This results from the replacement of normal bone marrow with fibrovascular marrow with an increase in free water and hypervascularity responsible for enhancement after gadolinium injection [19]. Modic type 1 changes are noted in 4 % of patients with back pain and are closely correlated with painful disk derangement [15, 21, 22]. A positive pain provocation test is not clearly correlated with Modic type 1 end plate changes [21, 22]. Similar changes are observed after surgery or percutaneous treatment of a disk herniation.

Modic type 1 end plate changes may at times simulate infectious diskitis and should be correlated with patients’ symptoms, clinical history, and laboratory data. In difficult cases, differentiation of degenerative and infectious end plate abnormalities may require a biopsy, MRI follow-up, or fluorodeoxyglucose (FDG) positron emission tomography (PET) [23]. Modic type 1 end plate changes may also simulate erosive diskitis associated with rheumatoid arthritis, gout, or chronic hemodialysis [24–27].

Modic type 2 end plate changes are visualized as high-signal-intensity lesions on T1- and T2-weighted images without enhancement; type 2 changes represent fatty marrow and are observed in 16 % of patients presenting with back pain [28] (Figs. 7.16 and 7.17).

Modic type 1 changes represent a dynamic process that often converts to Modic type 2 changes over time (Fig. 7.13). Modic type 2 end plate changes rarely progress [29]. Conversion of Modic type 1 changes to Modic type 2 changes is a very slow process (only 50 % conversion over an observation interval ranging from 1 to 6 years); regression of back pain is noted in two thirds of patients that fully convert from Modic type 1 changes to Modic type 2 changes; associated factors (apophyseal joint osteoarthritis, instability) may explain the absence or partial regression of symptoms [29]. Rapid conversion from Modic 1 to 2 changes is seen after a lumbar posterior arthrodesis [30]. Reverse transformation of Modic type 2 changes to Modic type 1 changes occurs rarely and is probably related to superimposed disease [31, 32].

Modic type 3 changes represent sclerosis of the end plates at the end stage of DDD and appear as a low signal intensity on T1- and T2-weighted images without enhancement (Fig. 7.18).

The signal appearance of recent cartilaginous Schmorl node formation is similar to that of the corresponding intervertebral disk on T1-weighted images and appears slightly hyperintense on T2-weighted images; enhancement is possible and may remain in the chronic stages. During acute and subacute stages, edema and inflammatory changes involving the surrounding bone marrow are noted as a low signal intensity on T1-weighted images, high signal intensity on T2-weighted images, and postcontrast enhancement and may be correlated with acute or subacute back pain (Fig. 7.19).