42 

How Well Can Patients Be Expected to Recover After Spinal Cord Injury?

Charles H. Tator

BRIEF ANSWER

Background

Data about neurologic recovery as part of the natural history of SCI have been collected over the past several decades in many countries. These data have accrued as a result of the development of accurate measures for classifying, grading, and serially assessing the neurologic deficits associated with the acute and subsequent rehabilitative phases of care of SCI, including long-term follow-up. The inherent slowness of recovery of the central nervous system makes accurate long-term assessment an essential part of any outcome study.

The effort to accurately classify and record the clinical neurologic features of SCI to document the natural history began in the 1940s in England at Stoke-Mandeville Hospital under Guttman.¹ These efforts then spread to the United States, Canada, Australia, and several European countries, including France and Switzerland. In 1969, the author developed a new system of grading neurologic function to assess neurologic recovery in SCI patients and applied it retrospectively to several hundred SCI patients during the spectrum of care from the acute phase to long-term follow-up. Considerable neurologic recovery was evident in some patients (class III data).^2,3 Later, this system was applied to prospectively gathered data about SCI patients, and again, considerable neurologic recovery was seen in some patients (class III data).^4,5 In 1969, Frankel et al⁶ published a similar grading method (class III data), and subsequently, many features of the two systems were combined to form the current American Spinal Injury Association (ASIA) grading and scoring system⁷ that is now used in most centers (class III data).

Since the late 1980s, the ASIA system has been used in several studies that have included long-term follow-up. These reports have confirmed that a remarkable propensity for neurologic recovery exists in SCI patients. Indeed, in the author’s own studies, the majority of SCI patients showed some degree of neurologic recovery (class III data).4,8 Also, knowledge of the natural history of acute SCI has been markedly enhanced since the advent in 1979 of multicenter, randomized, prospective clinical trials of the treatment of SCI. These trials have generally utilized the ASIA system and have produced additional data about recovery. Thus, reliable information about the natural history of SCI is now available.

Literature Review

Survival after Spinal Cord Injury

Survival after SCI has improved greatly over the past several decades due to several improvements in management. Prior to 1950, SCI was characterized by mortality rates as high as 50%, whereas recent studies have shown mortality rates in the 5 to 10% range (class II data).⁹ This marked decline in mortality can be attributed to many improvements in both the acute and rehabilitation phases of care. In the acute phase the important factors have included better first aid and more rapid referral to regional centers with trained staff and appropriate facilities for hemodynamic and respiratory resuscitation (class III data).⁸ In the rehabilitation phase the important factor has been prevention of complications, especially urinary tract infections, decubiti, deep venous thrombosis, and pulmonary embolism. In both phases of injury, mortality is related to patient age, severity of injury, accompanying injuries, and other factors. Multiple regression analysis has demonstrated an association between higher mortality and increasing age, more rostral vertebral levels of injury, and higher injury severity scores (class III data).⁸

Recovery of Spinal Stability

Methods for reducing spinal deformity and restoring spinal stability improved dramatically during the latter half of the 20th century. Previously, such methods were limited to traction and crude forms of fusion. Currently, operative reduction and internal fixation of fractures and dislocations can be successfully performed with a variety of surgical approaches and instrumentation. Thus, the majority of SCI patients recover stability and near-normal alignment of the spine. It is likely, although not proven, that restoration of alignment and stability has a positive effect on neurologic recovery, although the magnitude of such an effect is probably small. It is also likely that restoration of alignment and stability improves quality of life and reduces costs of care because of decreased pain, increased patient mobility and independence, and decreased length of hospitalization.

Recovery of Neurologic Function

Mechanisms of Neurologic Recovery

Recovery of neurologic function after SCI is the result of recovery of various anatomic structures, including the nerve roots and the gray and white matter of the spinal cord. The propensity for neurologic recovery differs among these structures: the roots show the best ability to recover, followed by the local gray matter, and then the white matter. In addition, motor roots and tracts have increased vulnerability to injury and decreased propensity for recovery compared with sensory roots and tracts. This hierarchy is probably based on an inherent combination of increasing vulnerability to injury and decreasing ability to recover. Nerve roots at or below the site of injury are the structures that recover most frequently because the peripheral nervous system is more resistant to injury and has a greater ability to recover than the central nervous system. Indeed, local nerve root recovery probably occurs to some degree in every patient after SCI. This recovery may appear as restoration of function in muscles, dermatomes, or organs innervated by the nerve roots at the immediate vertebral level of injury or by roots traversing the level of injury to innervate structures below the vertebral level of injury. For example, a C5 burst fracture may directly compress both the C5 and C6 nerve roots because the C6 nerve roots emanate from the C6 cord level, which is slightly rostral to the site of compression at C5. Due to the increasing obliquity of the roots in the more caudal segments of the spinal cord, more caudal injuries are associated with a greater extent of possible caudal neurologic recovery due to root recovery. Such recovery of nerve roots explains why conus medullaris or cauda equina injuries demonstrate the greatest caudal extent of neurologic recovery. For example, a patient with a T12 burst fracture can obtain neurologic recovery in structures innervated by the L1-L5 roots because all of these roots emanate from the lumbar cord, which is located almost entirely opposite the T12 vertebral body.