Neural Repair and Regeneration after Spinal Cord Injury and Spine Trauma

By Alexander R. Vaccaro

Neural Repair and Regeneration after Spinal Cord Injury and Spine Trauma provides readers with a comprehensive overview on the most up-to-date strategies to repair and regenerate the injured spinal cord following SCI and spine trauma. With contributions by international authors, chapters put regenerative approaches in context, allowing the reader to understand the challenges and future directions of regenerative therapies. Recent clinical trial advancements are thoroughly discussed, with the impact of trial findings addressed. Additionally, major ongoing clinical trials are included with thoughts from experts in the field. Recent clinical practice guidelines for the management of traumatic spinal cord injury are featured throughout.

These guidelines are quickly being adopted as the standard of care worldwide, and the comprehensive information found within this book will place these recommendations in context with current knowledge surrounding spinal cord injury and spine trauma.

• Contains contributions by international authors
• Covers recent clinical trial advancements and findings and updates on ongoing trials
• Presents an overview of clinical practice guidelines for the management of traumatic spinal cord injury featured
• Provides the reader with insights regarding the translation of research from bench to bedside and the skills needed to understand the translational pathway using real-life examples

• Language: English
• Publisher: Academic Press
• Release date: Feb 17, 2022
• ISBN: 9780128198360

Book preview

Introduction

Neural Repair and Regeneration after Spinal Cord Injury and Spine Trauma was written to provide the current state-of-the-art on spinal cord injury and spine trauma in one location. The book combines both research and clinical insights into one reference volume geared toward anyone studying or caring for those with a spinal cord injury. This includes clinical spine fellows, residents, neurosurgeons, clinical and basic researchers, orthopaedic surgeons, critical care specialists, medical students, and rehabilitation specialists.

Globally, approximately half a million people suffer a spinal cord injury or spine trauma every year. As a result, the lives of these individuals are altered dramatically and they experience significant impairment, reduced quality of life, and daunting financial costs. Currently, there are no cures for spinal cord injury. Researchers have invested significant resources examining approaches to repair and regenerate the injured spinal cord, including cellular strategies, rehabilitation, surgical techniques, and neuroprotective drugs. This textbook will cover the latest and most promising advances in the management of spine trauma and spinal cord injury.

The book provides an overview of spine anatomy before covering the pathophysiology and epidemiology of spinal cord injury, as well as diagnostics and injury classification systems. Management issues based on the level of injury and specific patient populations are also discussed. From there, the book addresses the use of surgical decompression, immunosuppressants, and rehabilitation to improve patient outcomes. An in-depth examination of the latest spinal cord injury research is then undertaken, touching on promising neuroprotective and neuroregenerative approaches. The book ends with expert commentary on what the future of spinal cord injury research and patient management may look like.

Chapter 1: Anatomy

Laureen D. Hachem ¹ , Ali Moghaddamjou ¹ , and Michael G. Fehlings ¹ , ²       ¹ Division of Neurosurgery, Department of Surgery, University of Toronto, Toronto, ON, Canada      ² Division of Neurosurgery, Toronto Western Hospital, University Health Network, Toronto, ON, Canada

Abstract

Understanding the functional anatomy and biomechanics of the spine is essential in the management of spinal trauma and spinal cord injury. Moreover, insight into the complexity of spinal tracts and level-specific differences in spinal cord anatomy is the foundation of understanding disease processes and developing novel therapies for spinal cord injury. This chapter provides an overview of spine and spinal cord anatomy and highlights relevance to spinal trauma.

Keywords

Anatomy; Autonomic nervous system; Gray matter; Spinal cord; Spinal trauma; Vertebrae; White matter tracts

Understanding the functional anatomy and biomechanics of the spine is essential in the management of spinal trauma and spinal cord injury. Moreover, insight into the complexity of spinal tracts and level-specific differences in spinal cord anatomy is the foundation of understanding disease processes and developing novel therapies for spinal cord injury. This chapter provides an overview of spine and spinal cord anatomy and highlights relevance to spinal trauma.

Spinal column

The spinal column is composed of several bony and ligamentous structures which surround and protect the spinal cord. As the primary support of the body, trauma to the spinal column can lead to significant deformities and unstable injures, which may compromise the underlying neural elements. Fig. 1.1 shows the anatomical relationship between spinal cord level, vertebral level, and corresponding nerve roots.

Vertebrae

The general structure of spinal vertebrae is largely preserved across spinal levels. Each vertebra is composed of the ventral body and a dorsal arch (made up of the pedicles and laminae). At the junction of the pedicle and lamina arise the superior and inferior articular processes and the transverse processes. The superior and inferior articular processes of adjacent vertebrae form the facet joint. This is surrounded by a capsule which is continuous medially with the ligamentum flavum. At the dorsal aspect of the lamina arises the midline spinous process. The space between pedicles of adjacent vertebrae forms the intervertebral foramen, which houses the exiting nerve root along with associated neurovascular structures. Transforaminal ligaments may be seen particularly in the lumbar levels and may serve to protect structures which traverse the foramen from injury. ¹ The intervertebral foramen is bounded by the body and disc anteriorly, facet joint posteriorly, and pedicles superiorly and inferiorly. The lateral recess is an important anatomical space bounded laterally by the pedicle, medially by the thecal sac, anteriorly by the vertebral body and disc, and posteriorly by the facet and ligamentum flavum. This region is a common site of stenosis in lumbar degenerative disc disease. ²

Variations in vertebral structure throughout the spine allow for the unique biomechanics at each level. The first vertebrae, C1 (the atlas), lacks a vertebral body and instead consists of two lateral masses bounded by anterior and posterior arches. The facets of C1, which articulate with the occipital condyles, are concave and face medially, thus providing close to 50% of flexion–extension of the neck but limited lateral displacement. ³ C2 (the axis) comprises an odontoid process extending from the vertebral body, which articulates with the posterior aspect of the C1 anterior arch forming the atlanto-axial joint, which provides 50% of rotation of the cervical spine. ³

Figure 1.1  Relationship between spinal cord level, vertebral level, and nerve roots.

The remaining cervical vertebrae (C3–7) are termed the subaxial spine. All cervical vertebrae have transverse foramina, which house the bilateral vertebral arteries at levels C1–6. Between C3–7, the posterolateral lip of each vertebra (termed the uncinate process) articulates with the side of the adjacent rostral vertebrae forming the uncovertebral joint. This joint is a common site of bony spur formation and degeneration. ⁴ The vertebral body is smallest within the subaxial cervical spine and wider in the lateral dimension with transverse processes projecting out laterally. Typically, spinous processes of C2–6 are short and bifid. The orientation of facet joints at the cervical level allows for free motion in all directions and facilitates the range of motion in the neck. ⁵ Thoracic vertebrae have heart-shaped bodies with transverse processes directed posteriorly and an articular surface with the corresponding rib. Spinous processes of thoracic vertebrae point inferiorly. At thoracic levels, facet joints allow for lateral flexion and rotation but limited flexion and extension. Lumbar vertebrae have large kidney-shaped bodies with posterior pointing transverse processes and spinous processes projecting horizontally. The orientation of facets in the lumbar region allows for primarily flexion and extension of the spinal column. ⁵

Intervertebral discs

Intervertebral discs separate the bodies of vertebrae forming a secondary cartilaginous joint to provide mobility and act as shock absorbers. The central portion of the disc is made up of the nucleus pulposus, a gelatinous substance derived from the embryonic notochord. The nucleus pulposus is composed of 70% water with the remaining component derived from type II collagen fibers and proteoglycans. ⁶ With age, the water and proteoglycan content of the disc decreases. This biochemical change leads to loss of disc height imparting abnormal loading on adjacent facet joints precipitating degenerative changes. ⁷ The nucleus pulposus is surrounded by an outer annulus fibrosus composed of multiple sheets of collagen with differing orientations providing increased strength. The superior and inferior aspects of the disc are composed of cartilaginous endplates, which anchor them to the adjacent vertebral bodies. Discs are thicker at the anterior portion in the cervical and lumbar spine contributing to the lordotic curve of the spine at these levels. ⁸

Ligaments

Ligaments of the spinal column are essential to maintaining stability. The anterior longitudinal ligament (ALL) runs along the anterior surface of vertebral bodies connecting from the occiput of the skull to the sacrum. Rostrally, the connection between the occiput and C1 is called the anterior occipitoatlantal membrane and between C1 and C2 the anterior atlantoaxial membrane. The posterior longitudinal ligament (PLL) runs along the posterior surface of vertebral bodies and extends rostrally as the tectorial membrane attaching to the basion.

Ligaments of the posterior arch include the ligamentum flavum, which originates midway of the inferior portion of the lamina and inserts onto the superior surface of the caudal lamina. Rostrally it becomes the posterior atlantoaxial membrane and posterior occipitoatlantal membrane. Interspinous ligaments connect the spinous processes of adjacent vertebrae and fuse with the supraspinous ligaments, which connect the tips of spinous processes. Intertransverse ligaments are thin connections between adjacent transverse processes. Assessment of posterior ligamentous integrity is critical in the evaluation of spinal fractures and a key component in treatment decision-making. ⁹ Disruption of the posterior ligamentous complex (composed of the supraspinous ligament, interspinous ligament, facet capsule, and ligamentum flavum) indicates a potentially unstable injury, which in many cases may necessitate surgical intervention. ¹⁰ , ¹¹  Fig. 1.2 shows the appearance of spinal ligaments on MRI. Additional ligaments are seen at the craniocervical junction. ¹² The apical ligament attaches the dens to the basion with two alar ligaments running lateral to the apical ligament connecting the dens to the occipital condyles. The transverse ligament (TAL) attaches along the anterior arch of the atlas around the odontoid process. Longitudinal bands extend rostral and caudal from the TAL together termed the cruciate ligament.

Ossification of spinal ligaments can often predispose to injuries and fractures in the setting of even low impact traumas. Ossification of the posterior longitudinal ligament (OPLL) is seen in 0.16%–2.4% of the population, most commonly within the cervical spine. ¹³ Direct ventral compression on the spinal cord leads to symptoms of myelopathy, which are exacerbated in the setting of minor traumas. ¹⁴ Diffuse idiopathic skeletal hyperostosis (DISH) is defined as ossification along the anterolateral aspect of at least four contiguous vertebrae. ¹⁵ This produces regions of long lever arms putting stress on adjacent vertebrae and thus increasing susceptibility to fractures. ¹⁶

Meninges

The spinal cord is surrounded by the meninges, which is composed of the outer dura mater, intermediate arachnoid, and inner pia. The arachnoid and pia together form the leptomeninges. The epidural space lies between the vertebral column and the enclosed dura matter. This space contains adipose tissue, venous plexuses along with meningovertebral ligaments that tether the dura to the surrounding supportive ligaments. The subdural space is considered a potential space between the arachnoid and dura matter. The subarachnoid space between the pia and arachnoid contains the cerebrospinal fluid bathing the spinal cord.

Figure 1.2  Spinal ligaments on MRI.(A) Sagittal T2-weighted MRI. 1, apical ligament; 2, anterior occipitoatlantal membrane; 3, anterior atlantoaxial membrane; 4, anterior longitudinal ligament; 5, tectorial membrane; 6, posterior longitudinal ligament; 7, posterior occipitoatlantal membrane; 8, posterior atlantoaxial membrane; 9, ligamentum flavum; 10, interspinous ligament; 11, supraspinous ligament. (B) Coronal T2 STIR sequence MRI displaying alar and transverse ligaments.

Spinal nerves

There are 31 spinal nerves: 8 cervical, 12 thoracic, 5 lumbar, 5 sacral, and 1 coccygeal. C1–7 nerves exit above their respective vertebrae, and all other nerves exit below their corresponding vertebrae. Spinal nerves are made of a dorsal (sensory) and ventral (motor) root, which are in turn composed of multiple rootlets. Both roots are covered by the leptomeninges along with a layer of dura, which continues as the epineurium once it exits the spinal column. Fig. 1.3 outlines the dermatomes and myotomes corresponding to each spinal nerve root.

Dorsal roots relay somatosensory information and are associated with a dorsal root ganglion, which houses the neuronal cell bodies of primary sensory neurons. These pseudo-unipolar neurons relay sensory information from the periphery into the central gray matter. Dorsal rootlets arise from the dorsolateral sulcus in the dorsal root entry zone (DREZ). Ventral roots carry motor information from efferent somatic motor neurons. Between T1 and L2, these fibers are also joined by autonomic preganglionic axons arising from the interomediolateral column of the sympathetic nervous system.

Figure 1.3  Myotomes and dermatomes. Adapted from the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI).

Dorsal and ventral roots combine to form a single spinal nerve. As each spinal nerve exits the intervertebral foramen, it gives off a branch termed the recurrent meningeal branch that enters the vertebral canal and supplies the PLL, portions of the annulus, dura matter, facet joints, and dorsal vertebral periosteum. The continuing spinal nerve then gives off a dorsal and ventral ramus, which each carries mixed motor and sensory information. The ventral rami of spinal nerves supply the anterolateral portions of the trunk and extremities. At the thoracic levels, these form the intercostal nerves, while at other levels, they form various nerve plexuses: cervical plexus (C1–C4), brachial plexus (C5–T1), lumbar plexus (L1–4), and sacral plexus (L4–S4). The dorsal rami mainly supply muscles and dermatomes of the back.

Rami communicantes are additional connections between spinal nerves and the sympathetic chain. White rami communicantes exit at levels T1–L2 and carry preganglionic axons from the spinal cord intermediolateral cell column to ganglia in the sympathetic chain. These fibers either synapse at the ganglion of the respective level or course along the sympathetic chain to distal levels. The postganglionic axons then exit the sympathetic chain and join spinal nerves via gray rami communicantes, which are present at every spinal level.

Gray matter: nuclei and rexed lamina

Gray matter within the spinal cord is located within the central portion of the cord and consists of both projection neurons and interneurons. Grossly, the gray matter can be divided into the dorsal horns, which house neurons receiving somatosensory information from the body, and the ventral horn consisting of motor neurons, which project to skeletal muscles. The cervical and lumbosacral enlargements represent increased populations of motor neurons that provide innervation to the upper and lower extremities, respectively. The gray matter of the spinal cord can be further divided into functional nuclei and cellular architecture, which will be discussed in this section (Fig. 1.4).

Figure 1.4  Spinal cord gray matter organization.

Rexed lamina

In the early 1950s, Bror Rexed described a system of 10 layers of gray matter (I-X) within the spinal cord, termed the Rexed laminae, corresponding to different portions of the gray matter based on cellular architecture. ¹⁷ While often defined as distinct zones, the borders of these laminae are not discrete. Moreover, distinct functional nuclei have been identified within these various laminae. ¹⁸

Lamina I comprises the most dorsal portion of the spinal cord gray matter. Early work in rodents identified four morphological neuronal types, which have subsequently been shown in various other species. Fusiform neurons are the most abundant and typically found in the lateral third of this lamina. Multipolar neurons, pyramidal neurons, and flattened neurons are found in the medial portion, dorsal border, and middle third of lamina I, respectively. ¹⁹ , ²⁰ These neurons receive information from A delta and C fibers carrying pain and temperature sensation from the body along with mechanical stimuli from A-beta fibers. This layer is thought to contribute to nearly half of the spinothalamic outflow to the brain. ²¹ Moreover, it also receives input from the hypothalamus and brainstem to modulate pain.

Lamina II contains densely packed neurons with rich dendrites and is often termed the substantia gelatinosa. It was initially thought to be composed primarily of stalked (outer portion) and islet (central portion) neurons; however, additional cells have been identified in humans including filamentous and stellate cells. ²² Neurons in this layer serve primarily as interneurons projecting to ventral lamina in the gray matter (III, IV, V). The lateral portion of this lamina receives input from C- and A-delta fibers. Neurons in lamina III and IV have similar functions relaying light touch and proprioception inputs and are modulated by lamina II interneurons. ¹⁸

Lamina V neurons relay nociceptive information to the brain and receive inputs from corticospinal and rubrospinal tracts. Lamina VI contains interneurons involved in spinal reflexes and relays information from muscle spindles to the brain via spinocerebellar tracts. Lamina VII is the largest region of the gray matter composed primarily of interneurons that modulate motoneurons. Lamina VIII contains both interneurons and projection neurons that aid in coordination of motor activity. Lamina IX is composed of groups of motor neurons that mediate skeletal muscle function. Lamina X surrounds the central canal region, termed the gray commissure, and includes axonal decussations of various pathways.

Gray matter nuclei

Populations of neurons may be further subdivided into specific nuclei based on function. The posteromarginal nucleus is found in lamina I of the spinal cord gray matter. Neurons within this nucleus receive information on pain and temperature. The substantia gelatinosa is found in rexed lamina II. It receives input from afferents of the dorsal root ganglia, primarily pain, temperature, and mechanoreception. In addition, it receives input from periaqueductal gray matter, locus coeruleus, gigantocellular reticular nucleus, and nucleus raphe magnus to modulate pain responses. The substantia gelatinosa plays an important role in the modulation of pain responses in the CNS. ²³ The nucleus proprius is found in laminae III and IV of the spinal cord gray matter and relays temperature and mechanical inputs via the spinocerebellar and spinothalamic tracts. Clarke's nucleus is composed of interneurons important for proprioception. This population of neurons is located in lamina VII at T1–L2 spinal levels. This nucleus is the origin of the dorsal spinocerebellar tract, and its fibers project ipsilaterally to the cerebellum. Lamina IX of the gray matter is composed of various groups of specialized motor nuclei innervating skeletal and visceral functions. The phrenic motor nucleus is found at segments C3–7 and innervates muscles of the diaphragm. Onuf's nucleus is found at the lower lumbar and sacral levels and controls micturition. Motor neurons from this nucleus project to perineal muscles and the anal and urethral sphincters.

White matter tracts

The spinal cord white matter is composed of a series of ascending and descending tracts, which relay sensorimotor signals between the brain and peripheral structures (Fig. 1.5).

Descending pathways

Corticospinal tracts: The corticospinal tract is the central motor relay pathway. Layer 5 pyramidal cells in the cortex project through the corona radiata and internal capsule into the cerebral peduncles. Approximately 85% of fibers cross at the pyramidal decussation forming the lateral corticospinal tract. These fibers ultimately synapse onto the anterior horn cells, which project out to the target muscle controlling movement. Classically, there was thought to be a somatotopic representation within the lateral corticospinal tract with upper extremities medial to lower extremities. However, to date, there has been little evidence in humans to support this organization. The remaining 15% of fibers that do not cross at the pyramidal decussation continue down in the anterior corticospinal tract.

Figure 1.5  Spinal cord white matter organization.

Rubrospinal tract: The rubrospinal tract is involved in voluntary movement. Neuronal cell bodies reside in the magnocellular red nucleus of the midbrain with axons that cross over at this level and descend in the lateral funiculus of the spinal cord. Most of these projections terminate in the cervical and lumbar enlargements, and thus, the rubrospinal tract is believed to be important in the modulation of fine motor control. Rubrospinal neurons synapse primarily onto neurons within Rexed laminae V and VI. ²⁴

Reticulospinal tracts: The reticulospinal tract is involved in the preparation of movements, postural control, and regulation of sensory and autonomic functions. The medial reticulospinal tract originates from neurons in the gigantocellular reticular nucleus and pontine caudal reticular nucleus. These neurons project ipsilaterally dispersed within the ventral and lateral columns and synapse onto laminae VI–IX in the ventral horn increasing muscle tone and voluntary movements. The lateral reticulospinal tract projects from neurons in the gigantocellular reticular nucleus traveling in the ventrolateral funiculus synapsing onto laminae V and VI and reduces muscle tone and voluntary movements. ²⁵

Vestibulospinal tracts: The vestibulospinal tracts regulate postural extensor activity of the limbs and trunk. The medial vestibulospinal tract originates in the medial vestibular nuclei (with inputs primarily from the semicircular canals). It travels in the ventral funiculus with both ipsilateral and contralateral projections onto laminae VII and VIII to coordinate head position and regulate the vestibulocollic reflex. The lateral vestibulospinal tract runs from the lateral vestibular nucleus (with inputs primarily from the otoliths) in the ipsilateral ventrolateral funiculus and controls extensor tone in the extremities. ²⁴ , ²⁶ , ²⁷

Tectospinal tract: The tectospinal tract is important for the coordination of head and eye movements. Output from the superior colliculus travels in the ventral white matter to upper cervical levels in rexed laminae VI, VII, VIII. The majority of these fibers run contralaterally.

Ascending pathways

Spinothalamic tract: The spinothalamic tract relays pain, temperature, and crude touch from the body. Nociceptive input is sent to layers I, II, and V of the spinal cord gray matter, whereas visceral input is sent to layers I and V. Second-order neurons within these laminae decussate and ascend rostrally in either the ventral or lateral spinothalamic tract. The ventral spinothalamic tract carries crude touch and pressure sensation, running in the ventral funiculus to join the medial lemniscus at the medulla and pons. The lateral spinothalamic tract transmits pain and temperature sensation and runs in the ventral portion of the lateral funiculus continuing as the spinal lemniscus. Second-order neurons in the spinothalamic tract ultimately synapse onto third-order neurons in the thalamus.

Dorsal columns: The dorsal columns pathway is responsible for proprioception, vibration sense, and fine touch. Primary sensory neurons in the dorsal root ganglion relay information that ascends in the dorsal funiculus. Information from lower extremities arising below T6 ascends more medially in the fasciculus gracilis, and sensory input from upper extremities above T6 ascends in the more lateral fasciculus cuneatus. In the caudal medulla, these neurons synapse onto second-order neurons of the nucleus gracilis and nucleus cuneatus, respectively. The second-order neurons decussate and ascend contralaterally to synapse onto third-order neurons of the thalamus.

Spinocerebellar tracts: Spinocerebellar pathways are composed of a dorsal and ventral component. The dorsal spinocerebellar tract relays afferent proprioceptive input from ipsilateral muscle spindles and Golgi tendons to the cerebellum. First-order neurons enter the dorsal horn and synapse onto second-order neurons within Clarke's nucleus. These second-order neurons then ascend ipsilaterally forming the dorsal spinocerebellar tract within the lateral funiculus entering the cerebellum via the inferior cerebellar peduncle. Inputs from upper thoracic and cervical levels ascend within fasciculus cuneatus and synapse onto second-order neurons within an analogous nucleus in the medulla termed the accessory cuneate nucleus. These neurons then project rostrally forming the cuneocerebellar tract similarly entering the cerebellum via the inferior cerebellar peduncle.

The ventral spinocerebellar tract relays inputs primarily from lower thoracic and lumbar segments. Here, second-order neurons are located within spinal border cells in the lumbar anterior horn (T12–L5), cross at the level of the spinal cord, and ascend in the contralateral lateral funiculus. Fibers enter the cerebellum via the superior cerebellar peduncle and then recross the midline to terminate ipsilateral to their site of origin.

Spinotectal tract: The spinotectal tract is thought to be responsible for nociceptive sensory information particularly with the facilitation of reflexive head movements toward noxious stimuli. The tract ascends through the anterolateral white matter toward the midbrain terminating in the contralateral superior colliculus.

Spinoolivary tract: The spinoolivary tract is responsible for the transmission of unconscious proprioception from tendons and muscles along with cutaneous information from the olivary bodies. Functionally, this tract is important in balance control. The ascending information travels through the dorsal root ganglia synapsing at the second-order neurons in the posterior gray matter. The axons then cross the midline and travel through the intersection between the anterior and lateral white matter columns. The axons synapse at the inferior olivary nuclei at the level of the medulla prior to entering the cerebellum via the inferior cerebellar peduncle.

Vascular supply

Arterial supply

Primary blood supply to the spinal cord is derived from the anterior spinal artery and the paired posterior spinal arteries, which typically arise from the vertebral arteries. The anterior spinal artery runs in the ventral median fissure supplying the anterior two thirds of the spinal cord, and the posterior arteries supply the dorsal columns. Loss of blood supply to these vessels thus produces a classic pattern of deficits whereby anterior spinal artery infarcts lead to leg weakness and loss of pain below the level of injury, and posterior spinal artery infarcts cause loss of proprioception and vibration sense below the level of injury.

At each level, the dorsal ramus of segmental arteries enters the intervertebral foramen as the segmental spinal artery, which in turn gives rise to various branches to supplement blood supply to the cord: (1) branch to the vertebral body and dura, (2) radicular branches (anterior or posterior radicular arteries), which supply the dorsal and ventral nerve roots, and (3) spinal medullary arteries, which augment flow to the anterior and posterior spinal arteries. During the third stage of fetal development, most of the medullary branches involute. However, a large anterior medullary branch remains typically arising from a left posterior intercostal artery between T8 and L2 called the artery of Adamkiewicz. Blood supply from this artery is critical in supplementing cord perfusion. Indeed, injury to this artery can lead to anterior spinal cord infarctions. ²⁸

Impaired spinal cord perfusion is one of the hallmarks of spinal trauma. Maintaining adequate blood flow to the cord is of utmost importance, and as such, current guidelines recommend maintaining a mean arterial pressure over 85mmHg for 7 days following injury. ²⁹ , ³⁰

Venous drainage

Intrinsic veins (sulcal and radial veins) drain the parenchyma of the spinal cord into the extrinsic venous system. The extrinsic venous system is composed of a network of pial veins, anterior and posterior spinal veins, and radicular veins. ³¹ The anterior median spinal vein runs adjacent to the anterior spinal artery coursing along the entire length of the cord. ³² The dorsal median spinal vein runs in the posterior median sulcus, and two dorsolateral spinal veins run adjacent to the posterior spinal arteries. Radicular veins drain the anterior and dorsal median veins into the extradural vertebral venous plexus (Batson's plexus). This network is composed of the internal venous plexus (posterior to the vertebral body and anterior to the vertebral arch) and the external venous plexus (anterior to the vertebral body and posterior to the vertebral arch). ³³ Basivertebral veins within the vertebral body connect internal and external systems. The veins that comprise Batson's plexus are valveless and thus allow for retrograde flow. This therefore serves a potential route for the spread of urinary infections or metastatic tumors from pelvic organs to the spine.

Autonomic nervous system

The autonomic nervous system (ANS) is a major component in maintaining homeostasis and is responsible for the control of vital functions. The ANS can be divided into the sympathetic and the parasympathetic systems, which have synergistic roles. The general structure of these divisions is made up of a two-neuron pathway composed of a preganglionic and postganglionic neuron. The preganglionic neurons traverse through the ventral roots and synapse with postganglionic neurons near the spinal cord or at the target organ as is the case with the parasympathetic system. The sympathetic preganglionic neurons are located at T1–L2. Surrounding the sympathetic system, the parasympathetic system is located at the cranial nerves and the sacral spine segments.

Cardiovascular function

The autonomic nervous system plays a vital role in the control of the cardiovascular system. The parasympathetic system decreases cardiac activity by reducing cardiac contractility and heart rate via the vagus nerve. The sympathetic system increases cardiac activity by increasing heart rate and cardiac contractility through T1–T5 thoracic sympathetic output. Spinal cord injury can lead to significant disruptions in the cardiovascular function. Traumatic spinal cord injuries above T6 result in loss of sympathetic tone and unopposed parasympathetic outflow, leading to the phenomenon of neurogenic shock, which is characterized by hypotension and bradycardia. In addition, impaired modulation of the sympathetic activity following trauma can result in autonomic dysreflexia. This phenomenon is characterized by autonomic overactivity in response to noxious stimuli below the neurological level of injury. ³⁴

Respiratory function

Understanding spinal cord anatomy is essential in the study of respiratory drive. The phrenic nerve originating from C3, C4, and C5 spinal nerves innervates the diaphragm. Accessory muscles including the posterior thoracic muscles (T1–5), pectoralis muscles (C4–T1), intercostal muscles (T2–T11), abdominal muscles (T7–L1), and the trapezius muscle (C2–C6) further aid in respiration. Control of breathing is also mediated through the spinal cord via the phrenic motor neurons, prephrenic interneurons, and white matter tracts. Spinal interneurons play a major role in generating respiratory drive, independent of the medulla. ³⁵–³⁷ Furthermore, there is animal evidence that cervical excitatory interneurons contribute to restoring respiratory function in chronic nontraumatic spinal cord injury. ³⁸

Conclusion

The anatomy and function of the spinal cord are closely related. The gray matter of the spinal cord organized by the Rexed lamina classification is responsible for the modulation of signals. The white matter tracts of the cord serve as a relay of ascending and descending information. Through the autonomic nervous system, the spinal cord also plays an important role in maintaining vital respiratory and cardiac functions. Knowledge of the structure of the spinal cord gray matter, autonomic nervous system, and white matter tracts will allow for the localization of pathological lesions in injuries.

References

1. Golub B.S, Silverman B. Transforaminal ligaments of the lumbar spine.  J Bone Jt Surg Am Vol . 1969;51(5):947–956.

2. Ciric I, Mikhael M.A, Tarkington J.A, Vick N.A. The lateral recess syndrome. A variant of spinal stenosis.  J Neurosurg . 1980;53(4):433–443.

3. Bogduk N, Mercer S. Biomechanics of the cervical spine. I: normal kinematics.  Clin Biomech . 2000;15(9):633–648.

4. Nagamoto Y, Ishii T, Iwasaki M, et al. Three-dimensional motion of the uncovertebral joint during head rotation.  J Neurosurg Spine . 2012;17(4):327–333.

5. Jaumard N.V, Welch W.C, Winkelstein B.A. Spinal facet joint biomechanics and mechanotransduction in normal, injury and degenerative conditions.  J Biomech Eng . 2011;133(7):071010.

6. Aladin D.M, Cheung K.M, Ngan A.H, et al. Nanostructure of collagen fibrils in human nucleus pulposus and its correlation with macroscale tissue mechanics.  J Orthop Res . 2010;28(4):497–502.

7. Adams M.A, Dolan P, Hutton W.C, Porter R.W. Diurnal changes in spinal mechanics and their clinical significance.  J Bone & Joint Surg Br . 1990;72(2):266–270.

8. Been E, Li L, Hunter D.J, Kalichman L. Geometry of the vertebral bodies and the intervertebral discs in lumbar segments adjacent to spondylolysis and spondylolisthesis: pilot study.  Eur Spine J . 2011;20(7):1159–1165.

9. Vaccaro A.R, Lee J.Y, Schweitzer Jr. K.M, et al. Assessment of injury to the posterior ligamentous complex in thoracolumbar spine trauma.  Spine J Off J North Am Spine Soc . 2006;6(5):524–528.

10. Vaccaro A.R, Lehman Jr. R.A, Hurlbert R.J, et al. A new classification of thoracolumbar injuries: the importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status.  Spine . 2005;30(20):2325–2333.

11. Vaccaro A.R, Oner C, Kepler C.K, et al. AOSpine thoracolumbar spine injury classification system: fracture description, neurological status, and key modifiers.  Spine . 2013;38(23):2028–2037.

12. Tubbs R.S, Hallock J.D, Radcliff V, et al. Ligaments of the craniocervical junction.  J Neurosurg Spine . 2011;14(6):697–709. .

13. Smith Z.A, Buchanan C.C, Raphael D, Khoo L.T. Ossification of the posterior longitudinal ligament: pathogenesis, management, and current surgical approaches. A review.  Neurosurg Focus . 2011;30(3):E10.

14. Song K.J, Park C.I, Kim D.Y, Jung Y.R, Lee K.B. Is OPLL-induced canal stenosis a risk factor of cord injury in cervical trauma?  Acta Orthop Belg . 2014;80(4):567–574.

15. Mader R, Verlaan J.J, Buskila D. Diffuse idiopathic skeletal hyperostosis: clinical features and pathogenic mechanisms.  Nat Rev Rheumatol . 2013;9(12):741–750.

16. Taljanovic M.S, Hunter T.B, Wisneski R.J, et al. Imaging characteristics of diffuse idiopathic skeletal hyperostosis with an emphasis on acute spinal fractures: review.  AJR Am J Roentgenol . 2009;193(3 Suppl. l):S10–S19 Quiz S20-14.

17. Rexed B. The cytoarchitectonic organization of the spinal cord in the cat.  J Comp Neurol . 1952;96(3):414–495.

18. Heise C, Kayalioglu G. Cytoarchitecture of the spinal cord. In: Watson C, Paxinos G, Kayalioglu G, eds.  The spinal cord: a Christopher and Dana Reeve foundation text and atlas . vol. 1. Amsterdam; Boston: Elsevier/Academic Press; 2009:64–93.

19. Galhardo V, Lima D. Structural characterization of marginal (lamina I) spinal cord neurons in the cat: a Golgi study.  J Comp Neurol . 1999;414(3):315–333.

20. Lima D, Coimbra A. A Golgi study of the neuronal population of the marginal zone (lamina I) of the rat spinal cord.  J Comp Neurol . 1986;244(1):53–71.

21. Willis W.D, Kenshalo Jr. D.R, Leonard R.B. The cells of origin of the primate spinothalamic tract.  J Comp Neurol . 1979;188(4):543–573.

22. Gobel S. Golgi studies of the neurons in layer II of the dorsal horn of the medulla (trigeminal nucleus caudalis).  J Comp Neurol . 1978;180(2):395–413.

23. Wall P.D. The role of substantia gelatinosa as a gate control.  Res Publ Assoc Res Nerv Ment Dis . 1980;58:205–231.

24. Watson C, Harvey A.R. Projections from the brain to the spinal cord. In: Watson C, Paxinos G, Kayalioglu G, eds.  The spinal cord: a Christopher and Dana Reeve foundation text and atlas . vol. 1. Amsterdam; Boston: Elsevier/Academic Press; 2009:168–179.

25. Nyberg-Hansen R. Sites and mode of termination of reticulo-spinal fibers in the cat. an experimental study with silver impregnation methods.  J Comp Neurol . 1965;124:71–99.

26. Pompeiano O. Spinovestibular relations: anatomical and physiological aspects.  Prog Brain Res . 1972;37:263–296.

27. Hayes N.L, Rustioni A. Descending projections from brainstem and sensorimotor cortex to spinal enlargements in the cat. Single and double retrograde tracer studies.  Exp Brain Res . 1981;41(2):89–107.

28. Phillips D, Dhall S.S, Uzelac A, Talbott J.F. Gunshot wound causing anterior spinal cord infarction due to injury to the artery of Adamkiewicz.  Spine J Off J North Am Spine Soc . 2016;16(9):e603–604.

29. Hawryluk G, Whetstone W, Saigal R, et al. Mean arterial blood pressure correlates with neurological recovery after human spinal cord injury: analysis of high frequency physiologic data.  J Neurotrauma . 2015;32(24):1958–1967.

30. Squair J.W, Bélanger L.M, Tsang A, et al. Spinal cord perfusion pressure predicts neurologic recovery in acute spinal cord injury.  Neurology . 2017;89(16):1660–1667.

31. Krings T. Vascular malformations of the spine and spinal cord∗: anatomy, classification, treatment.  Clin Neuroradiol . 2010;20(1):5–24.

32. Green K, Reddy V, Hogg J.P. Neuroanatomy, spinal cord veins. [Updated 2020 Jul 27]. In:  StatPearls [internet]. Treasure island (FL) . StatPearls Publishing; January 2021 Available from:. https://www.ncbi.nlm.nih.gov/books/NBK542182/.

33. Griessenauer C.J, Raborn J, Foreman P, Shoja M.M, Loukas M, Tubbs R.S. Venous drainage of the spine and spinal cord: a comprehensive review of its history, embryology, anatomy, physiology, and pathology.  Clin Anat . 2015;28(1):75–87.

34. Furlan J.C, Fehlings M.G. Cardiovascular complications after acute spinal cord injury: pathophysiology, diagnosis, and management.  Neurosurg Focus . 2008;25(5):E13.

35. Ghali M.G, Marchenko V. Patterns of phrenic nerve discharge after complete high cervical spinal cord injury in the decerebrate rat.  J Neurotrauma . 2016;33(12):1115–1127.

36. Palisses R, Perségol L, Viala D. Evidence for respiratory interneurones in the C3-C5 cervical spinal cord in the decorticate rabbit.  Exp Brain Res . 1989;78(3):624–632.

37. Lane M.A, White T.E, Coutts M.A, et al. Cervical prephrenic interneurons in the normal and lesioned spinal cord of the adult rat.  J Comp Neurol . 2008;511(5):692–709.

38. Satkunendrarajah K, Karadimas S.K, Laliberte A.M, Montandon G, Fehlings M.G.Cervical excitatory neurons sustain breathing after spinal cord injury.  Nature . 2018;562(7727):419–422.

Chapter 2: Epidemiology

Thorsten Jentzsch ¹ , ² , ³ , Anoushka Singh ² , and Michael G. Fehlings ¹ , ²       ¹ Division of Neurosurgery, Department of Surgery, University of Toronto, Toronto, ON, Canada      ² Division of Neurosurgery, Toronto Western Hospital, University Health Network, Toronto, ON, Canada      ³ Department of Orthopaedics, Balgrist University Hospital, University of Zurich, Zurich, Switzerland

Abstract

Spinal cord injury (SCI) is a devastating, life-changing neurotrauma with a high-disease burden. It leads to instant and permanent neurological impairment below the injury level due to disruption of afferent and efferent neural pathways. SCI is often associated with increased morbidity and premature mortality, leading to strain of the physical, emotional, and social well-being of each individual. It also represents a costly diagnosis for the healthcare system with direct and indirect costs in the millions of dollars, particularly if the cervical spine of a young person is affected. This chapter will examine the prevalence, incidence, and costs of SCI.

Keywords

Costs; Epidemiology; Incidence; Prevalence; Spinal cord injury; Treatment

Overview

Spinal cord injury

Spinal cord injury (SCI) is a devastating, life-changing neurotrauma with a high disease burden. ¹ It leads to instant and permanent neurological impairment below the injury level due to disruption of afferent and efferent neural pathways. SCI is often associated with increased morbidity and premature mortality leading to physical, emotional, and social well-being strain for each individual. It also represents a costly diagnosis for the healthcare system with direct and indirect costs in the millions of dollars, particularly if the cervical spine of a young person is affected. Clinical outcome can in part be improved by urgent surgical decompression and rehabilitation. ² , ³ Neuroprotective and neuroregenerative interventions continue to be studied to improve motor, sensory, and sphincter (bladder/bowel) functions in acute and chronic stages. ⁴

Epidemiology

Epidemiology is derived from the Greek words epi (=upon), demos (=people), and logos (=study) with a literal meaning of studying what is upon people. ⁵ It is defined as the study of the occurrence and distribution of events or states related to health in specified populations. This includes investigation of the influencing factors and measures to effectively control these health problems. ⁶ Measures of occurrence include prevalence and incidence. Prevalence describes the existing cases at a designated time and measures disease burden. Point prevalence depicts cases with an outcome at a designated time, while period prevalence defines cases with an outcome over a period of time. The latter is less useful because there is no distinction between existing (prevalent) and new (incident) cases. Incidence describes the frequency of new cases during a time period. Incidence risk depicts the proportion of new cases, while incidence rate defines the frequency of new cases (i.e., new cases divided by total person-time at risk). While the incidence risk is useful for rare diseases in a static population, the incidence rate is helpful for common diseases in a dynamic population. ⁷

Prevalence, incidence, and costs of spinal cord injury

Understanding the prevalence and incidence of SCI enables healthcare providers to improve prevention and management programs as well as to estimate costs. SCI has a reported prevalence of 906 per million population and an incidence of 40.1 per million population in the United States. ⁸ This means that over 280,000 people were affected by SCI, with an annual incidence of over 17,000 in 2016. ⁹ It is important to recognize that these numbers likely represent an underestimate because only patients at the extreme spectrum of disease are usually accounted for. There are several milder clinical conditions, such as degenerative cervical myelopathy, which are usually not taken into consideration in these calculations. ¹⁰ The most common causes of traumatic SCI (TSCI) are traffic accidents and falls, while a common cause for nontraumatic SCI (NTSCI) is degenerative cervical myelopathy (DCM). ⁸ In low- and middle-income developing countries, traffic accidents are the most common cause of SCI with a reported proportion of 35%–53.8% (followed by falls [22.6%–37%]), while falls are the number one cause in high-income developed countries with a proportion of 37.9%–63%. ¹¹ In low- to middle-income countries, low velocity falls are usually associated with carrying heavy loads on the head or crush injuries from collapsing ceilings in younger individuals, while in high-income countries, low velocity falls are usually associated with slipping and tripping in the elderly. ¹² In developed countries, there is a demographic shift in TSCI toward the elderly, low velocity falls, and the cervical spine region. A recent study has shown that the most common cause of new TSCIs in Finland was low velocity falls in 36.2%, while high velocity falls and motor vehicle accidents were the cause in only 25.5% and 19.2%, respectively. ¹³ Cervical spine SCI increased sixfold from 1970 to 2011, from 59 to 372 cases in Finland, ¹⁴ with a projected increase in SCI of the cervical spine from low velocity falls by 50% from 2011 to 2030. With an aging population across the globe, similar findings are reported worldwide. ¹⁵

It is estimated that the SCI costs for society are higher than $4 billion per year. ¹⁶ Previous studies have reported varying costs with mean charges during the first year of $523,089, followed by charges of $79,759 per year. ¹⁷ These charges can rise even higher if there is a high-level cervical injury with charges during the first year of $1,064,716, followed by $184,891 per year. ¹⁸ Therefore, multifactorial strategies for the investigation of better and novel treatment options for SCI remain a very important research topic. ¹⁹

History

Highlights about the history of SCI have been summarized before by New et al. ²⁰ Literature suggests that knowledge about spina bifida, a birth defect with incomplete spinal closure, has been present since the beginning of humanity. ²¹ Later on, Nicholaas Tulp actually used the term spina bifida in 1641. ²² Hippocrates, who lived 460–377 before common era, assumed that moisture and cold were linked to paraplegia, and it was not until the 16th century that this belief changed. ²³ He was also the first to describe SCI due to infection, which may have been tuberculosis. ²⁴ Poliomyelitis was named by Jakob von Heine in 1840 and infected several hundred thousands of patients before an inactivated poliovirus vaccine was developed by Jonas Salk in 1952. ²⁵ , ²⁶

Knowledge about SCI due to circulation alterations and inflammation was initiated in 1866 and later amended thanks to an enhanced understanding of the spinal cord's blood supply in the 19th century. ²⁷ , ²⁸ The first successful intradural and intramedullary spinal cord tumor resections were undertaken by Sir Victor Horsley in 1887 and Christian Fenger in 1890. ²⁹ , ³⁰ While complication rates were high initially, surgical techniques improved in the second half of the 20th century. ³¹ Radiotherapeutic options for spinal tumors were described in the 1950s by Wood. ³² Differences between myelomalacia and transverse myelitis were described in 1921. ³³ Transverse myelitis was recognized as an inflammatory, autoimmune, and infectious disorder in the middle of the 20th century. ³⁴ Categorization of neurological disorders into trauma, infections, degeneration, demyelination, inflammation, vascular, nutritional, and hereditary was undertaken in the 20th century. ³⁵ Forty-five years ago, Kraus et al. defined SCI as an acute, traumatic spinal cord lesion (including nerve roots) that results in motor and/or sensory deficits. ³⁶ Over the years, the Centers for Disease Control and Prevention refined this definition to a lesion of neural elements in the spinal canal (including not only the spinal cord, but also the cauda equina) and expanded the deficit to bladder and bowel dysfunction. ⁸

Pediatrics

The literature on SCI in the pediatric population is more sparse than in adults. This is complicated by the fact that varying terms have been used for nonNTSCI (i.e., spinal cord disease [preferred term], damage, dysfunction, lesion, myelopathy, and paraplegia). ³⁷ Therefore, identification of studies is more difficult, and inclusion criteria often vary limiting the generalizability of results. The International Spinal Cord Injury Data Sets for NTSCI has a basic and extended data set that discriminates congenital–genetic versus (vs.) acquired, which can be expanded to neoplastic, malignant, neural, astrocytoma, and malignant. ³⁸

It is important to notice that pediatric SCIs have different characteristics than adult SCIs. The incidence of cervical spine injuries was 5%. C2 lesions and violent etiologies were more commonly found in the preteen group (<12 years), while C4–5 lesions were more common in the teen/adult group. ³⁹ The incidence of SCI without radiological anomalies (SCIWORA) has been reported to be as high as 6%. ⁴⁰ In SCI from sports injuries and child abuse, the incidence may even rise to 75%. This emphasizes the importance of the clinical assessment. ⁴¹ A systematic review by Parent et al. pointed out that teenagers are at higher risk for scoliosis but appear to have better neurological recovery than in adults. ⁴² Apple and associates reported that scoliosis was more common in preteens than adults (23% vs. 5%). ³⁹ Another study showed that 97% of patients that sustained an SCI before their growth spurt developed scoliosis, in contrast to 52% of patients with an SCI after their growth spurt. ⁴³ Wang et al. reviewed 30 patients with SCI. Of 20 patients with complete SCI, 35% died, 35% had no neurological recovery, and 30% improved. Of the patients who improved, 83% became ambulatory. Of 10 patients with incomplete SCI, 80% improved. ⁴⁴

Another systematic review identified 862 relevant abstracts and included 25 articles from 14 countries. The age cut-off was 15 years (occasionally 19 years due to paucity of data). It differentiated between SCI and spinal cord disease. The prevalence rates for SCI were highest for the global regions Australia (51.5 cases per million population in 2011), followed by Ireland (12.1 cases per million population in 2015) and Canada (10.1 cases per million population in 2010). The prevalence rates for spinal cord disease were highest in Ireland (19.6 cases per million population in 2015), followed by Australasia (6 per million population in 2010) and Canada (2.5 cases per million population in 2010). The median SCI incidence rates were highest for the global regions North America (high income, 13.2 million per population per year), followed by Australasia (9.9 per million per year), Asia (east; 5.4 per million population per year), and Western Europe (3.3 per million population per year). The median spinal cord disease incidence rates were highest for Australasia (6.5 per million population per year) and Western Europe (6.2 per million population per year) followed by North America (high income; 2.1 per million population per year). SCI was mostly caused by traffic accidents (46%–74%), followed by falls (12%–35%) and sports/recreation (10%–25%). Spinal cord disease was most commonly caused by tumors (30%–63%) and autoimmune/inflammatory disease (28%–35%). ⁴⁵

Studies about the survival time were even more rare and had small sample sizes. It was reported that children <16 years at the time of initial injury had their yearly odds of death increased by one-third compared with elderly patients. ⁴⁵

Resources

Data collection

Sophisticated surveillance is needed to precisely identify all SCI cases. This may entail obtaining information not only from acute care institutions (especially emergency, orthopedic, and neurosurgery departments) but also from rehabilitation centers as well as state medical examiners, death certificates, and even surveillance systems in neighboring states, as previously done in Oklahoma and Utah in the United States. ⁴⁶ , ⁴⁷

In the United States, data from the hospital discharge surveys from the National Center for Health Statistics may also be used. ⁴⁸ In Canada, for example, Ontario and Alberta have these registries. ⁴⁹ Furthermore, Alberta operates a Health and Wellness database with medical records of all provincial hospitals. The Offices of the Chief Medical Examiners that investigate unexplained deaths can also be consulted. ⁵⁰ In provinces where there is only one center for SCI, such as in British Columbia, medical records from that institution can be studied. ⁵¹ A review by Singh et al. from 2014 identified nine studies with reports on prevalence and 44 studies commenting on the incidence of acute SCI. ⁸ Other potential resources can include trauma registries.

Coding

To identify cases of SCI, studies often use ICD codes. ICD-9 codes often include codes 805 (fracture of the vertebral column without mention of spinal cord lesion), 806 (fracture of vertebral column with spinal cord lesion), and 952 (spinal cord lesion without spinal bone injury). ⁸

As of October 2015, ICD-10 codes are typically used. There is a long list of codes that can be used for SCI. The ones that mention the spinal cord directly are S14.0 (concussion and edema of cervical spinal cord), S14.1, S24.0 (concussion and edema of thoracic spinal cord), S24.1, S34.0 (concussions and edema of lumbar spinal cord), S34.1, T06.0 (injuries of nerves and spinal cord involving other multiple body regions), T06.1, T09.3 (injury of spinal cord, level unspecified), and T91.3 (sequelae of injuries, of poisoning and of other consequences of external causes—sequelae of injuries of neck and trunk—sequelae of injury of spinal cord). Other codes that could potentially be accompanied by SCI are G82 (paraplegia and tetraplegia), S12.0 (fracture of the first vertebrae), S12.2, S12.7, S13.0 (traumatic rupture of cervical intervertebral disc), S13.2, S13.4, S22.0 (fracture of thoracic vertebrae), S23.0 (traumatic rupture of thoracic intervertebral disc), S23.1, S32.0 (fracture of lumbar vertebrae), S33.0 (traumatic vertebrae of lumbar intervertebral disc), S33.1, S34.3, and T91.1 (sequelae of injuries, of poisoning and of other consequences of external causes—sequelae of injuries of neck and trunk—sequelae of fracture of spine). ⁸

Prevalence

Global

Prevalence at a certain time point is defined as the proportion of people with a disease in a population. Globally, the prevalence of SCI was estimated between 236 and 1009 per million in 2011. ⁵² This is similar to the previous estimation of 110–1120 per million in 1995. ⁵³ These numbers are likely to underestimate the real prevalence due to high mortality rates at the scene of injury. They may also not be universally representative since they are from developed countries.

North America

A systematic review by Singh et al. reviewed 5874 articles, of which 48 were included. ⁸ They reported that the highest prevalence was found in the United States (906 per million), ⁵⁴ while the lowest was found in France (250 per million). ⁵⁵ A study by DeVivo et al. calculated a 30-year mean life expectancy after SCI using an estimated incidence rate of 30 cases per million. They estimated the prevalence of 906 per million, which would indicate the need of nine beds per million. ⁵⁴ An older prevalence study in the United States multiplied the average life duration of 18 years with an incidence of 30 per million to report a prevalence of 525 per million. ⁵⁶ Another study by Harvey et al. from the United States in 1990 selected area segments in 120 representative primary sampling units to survey households and nursing as well as long-term healthcare facilities to identify patients with SCI. The authors reported that the total prevalence was 721 per million. ⁵⁷ Canada's SCI prevalence was estimated to be 85,556 persons in 2012 (considering a population of 34.7 million). ⁵⁸ The prevalence of SCI has been increasing over the years as described in a study by Griffin et al. in 1985. They counted all patients with residual neurological deficits after SCI in Minnesota to report point prevalence of 197 per million population in 1950 and 473 in 1980. ⁵⁹ , ⁶⁰

Europe, Asia, and Australia

There are also other countries that have reported on the prevalence of SCI. Iceland stated a prevalence of 526 per million in 2009. ⁶¹ Finland used International Statistical Classification of Diseases and Related Health Problems (ICD)-9 and ICD-10 codes from rehabilitation centers, department of Orthopedic Surgery at Helsinki University Central Hospital, local organization of the disabled, local health centers, residential service houses, and announcements in patient magazines (to find cases not included in the previously mentioned sources) to estimate a prevalence of 280 per million in 2005. ⁶² In Norway, hospital records from eight different hospitals were reviewed, and the prevalence was found to be 365 per million in 2002. ⁶³ In France, as previously mentioned in brief, the prevalence was estimated using a 20-year mean life expectancy after SCI and coined at 250 per million. ⁵⁵ In Iran, random cluster sampling with 100 addresses as starting points and 25 households as each cluster with actual verification by a nurse was used to estimate a prevalence of 440 per million in 2009. ⁶⁴ Lastly, Australia used the Spinal Cord Injury Register to estimate a prevalence of 681 per million in 1997. ⁶⁵

Low- and middle-income countries

Developing countries comprise more than 80% of the world's population, but information about the epidemiology of SCI in these countries is underrepresented in the literature. ⁶⁶ It is often pointed out that epidemiological data is lacking (e.g., as in a report from Nepal in 2014 ⁶⁷ or India in 2013 ⁶⁸ ). Over 90% of deaths resulting from injury are located in low- and middle-income countries. ⁶⁸ The number of patients affected by an SCI per year in low- to middle-income countries was estimated as 843,316 (95% CI 393,629–1,292,385) in a recent comprehensive review (considering a population of 6.2 billion). ⁶⁹

Traumatic spinal cord injury and nontraumatic spinal cord injury

In Canada, 85,556 people were living with an SCI in 2012. Of those, 51% were TSCI and 49% were NTSCI. ⁵⁸ Another study found that the prevalence of NTSCI was 1120 per million population in Canada and 2310 per million population in India. ⁷⁰ The number of publications of NTSC has substantially increased in the past fourdecades with 1825 publications between 1974 and 1983 to 11,887 between 2004 and 13. This has been accompanied by a trend toward studies with better methodological design including larger sample sizes, multicenter, randomized controlled studies. ⁷¹

Incidence

Incidence is the frequency of new cases during a time period. The United States has the highest reported incidence of 17,700 (54 per million) cases per year, followed by New Zealand (49 per million), ⁷² Estonia (40 per million), ⁷³ and Japan (39 per million). ⁷⁴ Within the United States, Alaska had the highest incidence (83 per million), while Alabama had the lowest (29 per million). ⁸ A study suggested that the United States may be the leader in incidence due to driving behavior (seat belts) and road conditions, as well as violence. ⁴⁸ Another study suggested that Estonia may show higher numbers because of road safety, partially due to flat landscape, driving speeds, and alcohol consumption. ⁷⁵ High numbers in Japan were attributed to a generally increased risk for SCI due to ossification of the posterior longitudinal ligament (OPLL) and congenital stenosis. ⁷⁶ The lowest rates in Europe were found in Spain (8 per million), Denmark (9 per million), ⁷⁷ and the Netherlands (12 per million). ⁷⁸ Divanoglou et al. compared the incidence and causes of SCI between two different cities in Europe. In low- and middle-income developing countries, the incidence of SCI was 25.5 per million per year with ranges from 2.1 to 130.7 per million per year. ⁶⁶ In India for example, it is estimated that there are around 20,000 cases of SCI per year. ⁷⁹ The incidence of TSCI in the Middle East and North Africa was 23.2 per year with a general lack of evidence in this region. Fig. 2.1 ⁸ depicts the relative annual incidences of countries, states/provinces, and regions.

A decrease in the incidence of SCI has been observed over time in Ontario and Alberta. While the age-standardized incidence in Ontario was 46 per million in 1994/1995, it decreased to 37 per million in 1998/1999. ⁴⁹ Similarly, in Alberta, it was 57 per million in 1997/1998 but decreased to 48 per million in 1999/2000. ⁵⁰ However, there are also reports about increasing incidences of SCI in London, Ontario (Canada) from 21 per million in 1997 to 49 per million in 2000, ⁸⁰ parts of Minnesota (United States) from 22 per million in 1935–44 to 71 per million in 1975–81, ⁵⁹ parts of Western Norway from 6 per million in 1952–56 to 26 per million in 1997–2001, ⁶³ and New Zealand from 43 per million in 1979–8 to 49 per million in 1988. In Spain, the incidence has been quite stable with 8 per million in 1972–80, 14 per million in 1981–90, 13 per million in 1991–2000, and 13 per million in 2001–08. ⁸¹

The incidence was found to be higher in Thessaloniki in Greece (34 per million) than in Stockholm in Sweden (20 per million). The most common cause of injury was motor vehicle accidents (51%), followed by falls (37%) in Thessaloniki compared with falls (47%) and motor vehicle accidents (23%) in Stockholm. ⁸²

Figure 2.1  Relative annual incidences of SCIs in countries, states/provinces, and regions.The red color scheme illustrates incidences of SCI in countries. The blue color scheme highlights incidences in states/provinces and regions. mil, million. Reproduced from Singh ⁸ with permission (Dove Medical Press was the original publisher of this work).

Traumatic spinal cord injury and nontraumatic spinal cord injury

According to a recent review of 102 studies with a metaanalysis of 19 studies, the global incidence of TSCI was 10.5 cases per 100,000 persons. This corresponds to an estimated 768,473 new cases of TSCI per year worldwide. It was higher in low- and middle-income countries than high-income countries (8.7 vs. 13.7 per 100,000 persons). Traffic accidents were more common than falls. ⁶⁹ In Canada, the incidence of TSCI was 1785 cases per year. The discharge incidence of TSCI was 1389 (41 per million) cases per year. ⁵⁸

The incidence rates were highest for North America (76 per million population per year), followed by Australasia (26 per million population per year), Asia Pacific (20 per million population per year), Oceania (9 per million population per year), and Western Europe (6 per million population per year). The discharge incidence of NTSCI was 2286 (68 per million) per year. ⁵⁸

Costs

Healthcare costs are generated by various factors during the intensive acute care and chronic management of the disease state as well as complications. Krueger et al. estimated that the overall economic burden per year was 2.7 billion Canadian dollars (CAD) in 2013. Direct costs (1.6 billion CAD) were higher than indirect costs (1.1 billion CAD). The direct costs were more pronounced in tetraplegia (56%–66%) compared with paraplegia (44%–54%). They were mostly driven by attendant care (33%), home modification (12%), healthcare practitioner visits (7%), and hospitalizations (7%). Indirect costs were mainly attributed to loss of productivity, which can be calculated by quality-adjusted life years (QALYs) using the mean annual salary of 47,834 CAD in 2011 and a utility of 0.45, which is typical for patients with SCI. By losing 0.55 QALYs, 26,309 CAD (0.55×47,834 CAD) are added to the indirect costs each year after SCI. If the utility was adapted to the type of SCI, i.e., 0.55 for incomplete paraplegia, 0.45 for incomplete tetraplegia, 0.35 for complete paraplegia, and 0.25 for complete tetraplegia, indirect costs would increase by 18%–23%. Costs are also driven by the age of a patient at the time of the injury. On the one hand, for example, being 25 years instead of 35 years would increase costs by 10%–14%. On the other hand, being 45 years instead of 35 years would decrease costs by 14%–18%. On an individual basis, the lifetime burden is 1.5 million CAD for incomplete paraplegia and twice as high for complete tetraplegia. ⁸³

Clinical

Classification

SCI and spinal cord disease are classified according to their etiology and neurology using the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI). This classification is only valid for patients ≥6 years. ⁸⁴ Paraplegia indicates the loss of motor and/or sensory function in the legs and pelvis organs, while tetraplegia indicates a loss of motor and/or sensory function in the arms, trunk, legs, and pelvic organs. The majority of cases are incomplete tetraplegia (tetraparesis; 34%), followed by complete paraplegia (25%), complete tetraplegia (22%), and incomplete paraplegia (paraparesis; 17%).

McKinley et al. reported that 21% of patients present with a clinical syndrome after SCI. ⁸⁵ Central cord syndrome is the most common form (44%) of incomplete tetraplegia. It is often due to hyperextension injury in patients with cervical spondylosis. It affects the upper extremities more than the lower extremities. Cauda equina syndrome is the second most common form (25%). It is an injury to the lower motor neurons resulting in flaccid paralysis of the limbs and areflexic bladder/bowel. The primarily affected anatomical region is the cervical spine (44%–62%). ⁸ In Brown-Séquard syndrome, which is the third most common form (17%), there is a spinal cord hemisection resulting in ipsilateral proprioception/vibration/motor loss and contralateral pain/temperature loss below the lesion (as well as sensation loss at the level of the lesion). Conus medullaris syndrome is quite rare (8%). It can resemble cauda equina syndrome, but it may include upper motor neuron signs. Anterior cord syndrome is very rare (5%) and affects the corticospinal and spinothalamic tracts sparing the dorsal columns resulting in motor/pain/temperature loss with sparing of light touch/position sense. Posterior cord syndrome is the rarest