The spine consists of a total of 33 vertebrae (7 cervical, 12 thoracic, 5 lumbar, 5 sacral, and 4 coccygeal) (Fig. 21-1A). Although vertebrae in the sacrum and coccyx are fused, vertebrae in the cervical, thoracic, and lumbar spine articulate through a series of diarthrodial (i.e., synovial) joints that are stabilized by several ligaments and muscular attachments. The vertebra serves as a “building block” for the spine, and the 24 presacral vertebrae consist of a series of functional spinal units consisting of two vertebral bodies, the facet joint that forms the articulation between the two vertebrae and the intervertebral disk. Most vertebrae in the spine share a similar anatomy. Every vertebra consists of a vertebral body, pedicles, articular processes, pars interarticularis, transverse processes, lamina, and spinous process (Fig. 21-1B).

The vertebral body is a cylindrical mass of trabecular, cancellous bone that is found in the anterior aspect of the vertebral body. Vertebral bodies vary in size, generally increasing in size when moving more caudal in the spine. In the thoracic spine, the vertebral bodies have articulations for the ribs. In the cervical spine, vertebral bodies articulate with each other through saddle-shaped joints referred to as the “joints of Luschka” or uncovertebral joints. The intervertebral disk lies between the vertebral bodies of two adjacent vertebrae.

Just posterior to the vertebral body is the dorsal arch. In contrast to the vertebral body, whose anatomy is relatively simple, the dorsal arch consists of a series of processes that enclose the spinal canal, allow for articulation with the neighboring vertebral bodies, and provide attachment sites for ligaments and muscles. The dorsal arch of the vertebra is connected to the vertebral body by a pair of two stout pillars referred to as the pedicles. The dorsal “arch” is composed of a pair of flat surfaces called the lamina. These form the roof of the spinal canal. Where the lamina meet in the midline, a large process projects dorsally. This is called the spinous process.

The remaining portions of the dorsal arch (transverse process, articular processes, and pars interarticularis) are found at the junction of the lamina and the pedicles. At this point, the transverse processes extend to either side of the vertebral arch. In the thoracic spine, these articulate with the rib. The articular process extends superiorly and inferiorly from the junction of the lamina and the pedicles to form the superior articular process and inferior articular process, respectively. The area between the articular processes at the confluence of these various processes is referred to as the pars interarticularis. The superior articular process of a given vertebra articulates with the complementary inferior articular process of the level above. These processes have cartilage and form a diarthrodial joint called the facet joint. A vertebra typically has four facet joints (two with the vertebrae above and two with the vertebrae below). Typically, the superior articulating process is directed dorsally, whereas the inferior

articular process is directed ventrally (toward the belly button). The specific orientation of the facet joints varies between the cervical, thoracic, and lumbar spine and is suited both to the types of loads and type of motion experienced by the specific spinal segments (e.g., rotation and lateral bending in the cervical spine and flexion/extension in the lumbar spine). The spinal nerve root exits the spinal cord from a space just anterior to the lateral aspect of the facet joint. This space, referred to as the intervertebral foramen, is bordered anteriorly by the intervertebral disk and posteriorly by the lateral aspect of the facet joint and ligamentum flavum, and the superior and inferior boundaries are formed by the pedicles of the levels above and below (Fig. 21-1B).

The discussion above highlights important structures that may contribute to compression of neural elements and lead to lower extremity symptoms. In addition to the bony structures mentioned earlier, soft tissue structures that might contribute to impingement include the posterior longitudinal ligament (PLL) and the ligamentum flavum (Fig. 21-1C). The PLL runs along the posterior surfaces of the vertebral bodies from the vertebral body of C2 to the sacrum and ventral to the spinal cord. The ligamentum flavum is an elastic structure that serves to help the vertebral column maintain a normal posture and runs dorsal to the spinal cord. The ligamentum flavum is actually a series of small ligaments that serve to connect adjacent vertebral lamina.

Similar to the spine, the spinal cord is an anatomically segmented structure. There are 31 segments in the spinal cord: 8 cervical, 12 thoracic, 5 lumbar, 5 sacral, and 1 coccygeal. Each segment of the spinal cord innervates a single somite during development; this translates to each spinal nerve retaining its relationship with the characteristic areas of skin and muscles. The area of skin innervated by a single spinal segment is referred to as a dermatome. Although the relationship between spinal segment and dermatome in the trunk is quite simple and reflects the segmental nature of the spine, the relationship in the limbs is more complex because of the outgrowth of the limb buds during development.³ These are illustrated in Figure 21-2.

Given the segmental nature of the cord, it is typically described in cross section. The cord consists of a roughly H-shaped area of gray matter that is surrounded by myelinated white matter. Each limb of the “H” shape of the gray matter can be divided into horns (posterior and anterior), whereas the white matter surrounding it is described as funiculi. Afferent fibers (i.e., sensory inputs from the limb) enter the cord via the dorsal roots and end on the ipsilateral side. Here, these inputs synapse on neurons in the ipsilateral gray matter in the posterolateral horn and feed into a complex system of sensory inputs. A portion of these inputs gives rise to the sensory pathways that ascend in the dorsal portion of the cord, whereas others feed into local reflex circuits. α motor neurons that innervate skeletal muscles can be found in the anterior horns. There are characteristic enlargements of the anterior horns in the cervical and lumbar regions (more α motor neurons) to account for motor innervation of the extremities. Anterior horn cells are arranged in cigar-shaped columns such that multiple levels may contribute to the function of a given muscle (i.e., the quadriceps is innervated by anterior horn cells corresponding to the L2 and L3 segments).³ Output from the α motor neurons leaves the cord via the ventral roots. The remainder of the spinal cord has a fairly characteristic organization; fibers are generally organized by the type of information they carry. A complete understanding of this schema is beyond the scope of this chapter, but can be found in Figure 21-3.

The motor neuron cells in the anterior horn are modulated by a complex system of inputs from the corticospinal tracts that descend from the cerebral cortex as well as brain stem and diencephalic nuclei. The anterior horn cells are involved in the reflex response. A reflex is defined as an involuntary response to a sensory input with circuitry that is contained entirely within the spinal cord. The deep tendon reflexes commonly tested in the clinical setting represent an example of a simple reflex loop (Fig. 21-4).³ The reflex at the knee, for example, involves tapping the patellar tendon, which causes a small stretch of the quadriceps muscle. This stretch is detected by sensory neurons within the muscle and carried to the spine by the afferent pathway. These sensory neurons synapse onto the α motor neurons in the anterior horn of the spinal cord as well as an inhibitory interneuron. The α motor neuron receives a positive stimulus and acts to initiate quadriceps contraction, whereas the inhibitory interneuron suppresses the α motor neurons of the antagonist muscle group (hamstrings). These stimuli then exit the spinal cord via the ventral nerve roots, traveling via the femoral nerve to the quadriceps and via the tibial nerve to the hamstrings. Contraction of the quadriceps and relaxation of the hamstrings lead to knee extension, or the knee-jerk commonly seen on examination. From this example, it can be deduced that damage to the exiting nerve root (e.g., due to compression in the neural foramen) can diminish or eliminate the reflex response. Because of the segmental nature of the spine, reflexes are typically tied to a principal cord segment (in this case, L2). In this way, reflexes are important tools because they can be easily tested and can be used to localize lesions. The reflexes at the biceps, brachioradialis, triceps, and ankle (Achilles) are governed by a similar loop; the localization of these reflexes is shown in Table 21-1.

In truth, the monosynaptic reflex described above is a simplification.³ The firing of the α motor neuron is modulated
by several inputs from the descending lateral cortical tracts. In general, the descending tracts serve to modulate the firing of the α motor neurons and allow volitional control of the upper and lower extremities. Injury to the lateral descending tracts, as can be seen in certain neurologic disorders or in compression of the cervical spine, can lead to increased muscle tone, hyperreflexia, and pathologic reflexes. It is important to understand the distinction between upper and lower motor neuron symptoms, and these are summarized in Table 21-2.

Finally, it is important to note that the human spinal cord reaches adult size faster than the vertebral column. As a result, it is typically shorter than the vertebral column, extending from the cervical spine to approximately the level of L1/L2.³ Below this region, the spinal canal contains the spinal nerves for the lumbar vertebral levels, but these nerve roots are actually given off at higher levels and travel in the cord as a structure called the cauda equina (horse’s tail) (Fig. 21-5).